From lethal@linux-sh.org Mon Mar 1 03:33:52 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Mon, 01 Mar 2010 03:33:55 +0100 (CET) Received: from 124x34x33x190.ap124.ftth.ucom.ne.jp ([124.34.33.190]:56179 "EHLO master.linux-sh.org" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492103Ab0CACdw (ORCPT ); Mon, 1 Mar 2010 03:33:52 +0100 Received: from localhost (unknown [127.0.0.1]) by master.linux-sh.org (Postfix) with ESMTP id EE7CE63758; Mon, 1 Mar 2010 02:33:16 +0000 (UTC) X-Virus-Scanned: amavisd-new at linux-sh.org Received: from master.linux-sh.org ([127.0.0.1]) by localhost (master.linux-sh.org [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id mxxupwluW9ax; Mon, 1 Mar 2010 11:33:16 +0900 (JST) Received: by master.linux-sh.org (Postfix, from userid 500) id AAA8D6375A; Mon, 1 Mar 2010 11:33:16 +0900 (JST) Date: Mon, 1 Mar 2010 11:33:16 +0900 From: Paul Mundt To: Andrea Gelmini Cc: linux-kernel@vger.kernel.org, Ralf Baechle , linux-mips@linux-mips.org, linux-sh@vger.kernel.org Subject: Re: [PATCH 42/66] arch/mips/lib/libgcc.h: Checkpatch cleanup Message-ID: <20100301023316.GB23086@linux-sh.org> References: <1267289508-31031-1-git-send-email-andrea.gelmini@gelma.net> <1267289508-31031-43-git-send-email-andrea.gelmini@gelma.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1267289508-31031-43-git-send-email-andrea.gelmini@gelma.net> User-Agent: Mutt/1.5.13 (2006-08-11) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26074 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: lethal@linux-sh.org Precedence: bulk X-list: linux-mips On Sat, Feb 27, 2010 at 05:51:23PM +0100, Andrea Gelmini wrote: > arch/mips/lib/libgcc.h:21: ERROR: open brace '{' following union go on the same line > > Signed-off-by: Andrea Gelmini > --- > arch/mips/lib/libgcc.h | 3 +-- > arch/sh/lib/libgcc.h | 3 +-- > 2 files changed, 2 insertions(+), 4 deletions(-) > I'll apply the sh part by itself, and I suppose Ralf will do the same for MIPS. I'm a bit confused as to why these were lumped together when 65 out of 66 oneliner patches made no use of ad-hoc grouping. From davem@davemloft.net Mon Mar 1 03:34:13 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Mon, 01 Mar 2010 03:34:19 +0100 (CET) Received: from 74-93-104-97-Washington.hfc.comcastbusiness.net ([74.93.104.97]:51052 "EHLO sunset.davemloft.net" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492306Ab0CACeB (ORCPT ); Mon, 1 Mar 2010 03:34:01 +0100 Received: from localhost (localhost [127.0.0.1]) by sunset.davemloft.net (Postfix) with ESMTP id 1329F24C097; Sun, 28 Feb 2010 18:34:18 -0800 (PST) Date: Sun, 28 Feb 2010 18:34:17 -0800 (PST) Message-Id: <20100228.183417.52179576.davem@davemloft.net> To: sebastian@breakpoint.cc Cc: ralf@linux-mips.org, linux-mips@linux-mips.org, linux-ide@vger.kernel.org Subject: Re: [PATCH 3/3] ide: move dcache flushing to generic ide code From: David Miller In-Reply-To: <1267371341-16684-4-git-send-email-sebastian@breakpoint.cc> References: <1267371341-16684-1-git-send-email-sebastian@breakpoint.cc> <1267371341-16684-4-git-send-email-sebastian@breakpoint.cc> X-Mailer: Mew version 6.3 on Emacs 23.1 / Mule 6.0 (HANACHIRUSATO) Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26075 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: davem@davemloft.net Precedence: bulk X-list: linux-mips From: Sebastian Andrzej Siewior Date: Sun, 28 Feb 2010 16:35:41 +0100 > the pio callbacks are called with different kind of buffers. It could be > a straight kernel addr, kernel stack or a kmaped highmem page. > Some of this break the virt_to_page() assumptions. > This patch moves the dcache flush from architecture code into generic > ide code. ide_pio_bytes() is the only place where user pages might be > written as far as I can see. > The dcache flush is avoided in two cases: > - data which is written to the device (i.e. they are comming from the > userland) This needs a flush on sparc, otherwise an alias now exists in the kernel side copy of the buffer. The D-cache flush is intentionally unconditional for PIO mode. I definitely don't want to take the same risks you guys seem to be willing to take for this optimization which is of questionable value. I also, intrinsically, really don't like these changes. For one thing, you're optimizing PIO mode. Secondly, IDE is in deep maintainence mode, if you want to optimize cache flushing do it in the ATA layer. Thanks. From manuel.lauss@googlemail.com Mon Mar 1 17:57:46 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Mon, 01 Mar 2010 17:57:50 +0100 (CET) Received: from mail-bw0-f226.google.com ([209.85.218.226]:46587 "EHLO mail-bw0-f226.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491182Ab0CAQ5q (ORCPT ); Mon, 1 Mar 2010 17:57:46 +0100 Received: by bwz26 with SMTP id 26so2059174bwz.27 for ; Mon, 01 Mar 2010 08:57:39 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=googlemail.com; s=gamma; h=domainkey-signature:received:received:from:to:cc:subject:date :message-id:x-mailer; bh=ZhrEtnGrw5lNGsbHf5cjn6M+g2LxCITIL1NSHneOyU0=; b=mQQjhAOvggnwaQtjdkGWWrY/sV+Da5jZDAS+ghEjQ0eXU3xHjJ/fKu56CpzFdemUlJ lKjEVM7jx0tiE52BQ9WUv9NV+Eqy3xUnYHfsRRR7IpXeCZ+0om3hOiiE1p/b8LQ78SuB IULsrBcIH/aSRT2oSyDzZfflI4CdfkxAMC7fw= DomainKey-Signature: a=rsa-sha1; c=nofws; d=googlemail.com; s=gamma; h=from:to:cc:subject:date:message-id:x-mailer; b=l0ZweTv9UemvDeGXfjc2d1t8StdvtWVWCLQt3J7eXAv9DeZldXMxlshC6PMb07+b5P oi71/nB3Xph3bw6G0OCRg4ZCfLZDU3SPrHFLLlDxvsDw8ElthB70j/l+OFwXIqLs3SQO A0RE6n0/j02SIWr7aEnQWPswohb+OGs9rmgjU= Received: by 10.204.33.210 with SMTP id i18mr3142630bkd.119.1267462658554; Mon, 01 Mar 2010 08:57:38 -0800 (PST) Received: from localhost.localdomain (p5496CB22.dip.t-dialin.net [84.150.203.34]) by mx.google.com with ESMTPS id 14sm2144589bwz.14.2010.03.01.08.57.36 (version=TLSv1/SSLv3 cipher=RC4-MD5); Mon, 01 Mar 2010 08:57:37 -0800 (PST) From: Manuel Lauss To: Linux-PCMCIA Cc: Linux-MIPS , Manuel Lauss Subject: [PATCH] pcmcia: alchemy: fixup wrong comments Date: Mon, 1 Mar 2010 17:58:38 +0100 Message-Id: <1267462718-17383-1-git-send-email-manuel.lauss@gmail.com> X-Mailer: git-send-email 1.7.0 Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26076 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: manuel.lauss@googlemail.com Precedence: bulk X-list: linux-mips Commit 11b897cf84c37e6522db914793677e933ef311fb changed expected pcmcia area addresses from the 32bit pseudo to the real 36bit addresses, but did not update the comments. --- Applies to current linus-git. drivers/pcmcia/db1xxx_ss.c | 19 ++++--------------- drivers/pcmcia/xxs1500_ss.c | 16 +++------------- 2 files changed, 7 insertions(+), 28 deletions(-) diff --git a/drivers/pcmcia/db1xxx_ss.c b/drivers/pcmcia/db1xxx_ss.c index 3889cf0..9254ab0 100644 --- a/drivers/pcmcia/db1xxx_ss.c +++ b/drivers/pcmcia/db1xxx_ss.c @@ -42,7 +42,6 @@ struct db1x_pcmcia_sock { int nr; /* socket number */ void *virt_io; - /* the "pseudo" addresses of the PCMCIA space. */ phys_addr_t phys_io; phys_addr_t phys_attr; phys_addr_t phys_mem; @@ -437,7 +436,7 @@ static int __devinit db1x_pcmcia_socket_probe(struct platform_device *pdev) * This includes IRQs for Carddetection/ejection, the card * itself and optional status change detection. * Also, the memory areas covered by a socket. For these - * we require the 32bit "pseudo" addresses (see the au1000.h + * we require the real 36bit addresses (see the au1000.h * header for more information). */ @@ -459,11 +458,7 @@ static int __devinit db1x_pcmcia_socket_probe(struct platform_device *pdev) ret = -ENODEV; - /* - * pseudo-attr: The 32bit address of the PCMCIA attribute space - * for this socket (usually the 36bit address shifted 4 to the - * right). - */ + /* 36bit PCMCIA Attribute area address */ r = platform_get_resource_byname(pdev, IORESOURCE_MEM, "pcmcia-attr"); if (!r) { printk(KERN_ERR "pcmcia%d has no 'pseudo-attr' resource!\n", @@ -472,10 +467,7 @@ static int __devinit db1x_pcmcia_socket_probe(struct platform_device *pdev) } sock->phys_attr = r->start; - /* - * pseudo-mem: The 32bit address of the PCMCIA memory space for - * this socket (usually the 36bit address shifted 4 to the right) - */ + /* 36bit PCMCIA Memory area address */ r = platform_get_resource_byname(pdev, IORESOURCE_MEM, "pcmcia-mem"); if (!r) { printk(KERN_ERR "pcmcia%d has no 'pseudo-mem' resource!\n", @@ -484,10 +476,7 @@ static int __devinit db1x_pcmcia_socket_probe(struct platform_device *pdev) } sock->phys_mem = r->start; - /* - * pseudo-io: The 32bit address of the PCMCIA IO space for this - * socket (usually the 36bit address shifted 4 to the right). - */ + /* 36bit PCMCIA IO area address */ r = platform_get_resource_byname(pdev, IORESOURCE_MEM, "pcmcia-io"); if (!r) { printk(KERN_ERR "pcmcia%d has no 'pseudo-io' resource!\n", diff --git a/drivers/pcmcia/xxs1500_ss.c b/drivers/pcmcia/xxs1500_ss.c index 61560cd..f9009d3 100644 --- a/drivers/pcmcia/xxs1500_ss.c +++ b/drivers/pcmcia/xxs1500_ss.c @@ -218,11 +218,7 @@ static int __devinit xxs1500_pcmcia_probe(struct platform_device *pdev) ret = -ENODEV; - /* - * pseudo-attr: The 32bit address of the PCMCIA attribute space - * for this socket (usually the 36bit address shifted 4 to the - * right). - */ + /* 36bit PCMCIA Attribute area address */ r = platform_get_resource_byname(pdev, IORESOURCE_MEM, "pcmcia-attr"); if (!r) { dev_err(&pdev->dev, "missing 'pcmcia-attr' resource!\n"); @@ -230,10 +226,7 @@ static int __devinit xxs1500_pcmcia_probe(struct platform_device *pdev) } sock->phys_attr = r->start; - /* - * pseudo-mem: The 32bit address of the PCMCIA memory space for - * this socket (usually the 36bit address shifted 4 to the right) - */ + /* 36bit PCMCIA Memory area address */ r = platform_get_resource_byname(pdev, IORESOURCE_MEM, "pcmcia-mem"); if (!r) { dev_err(&pdev->dev, "missing 'pcmcia-mem' resource!\n"); @@ -241,10 +234,7 @@ static int __devinit xxs1500_pcmcia_probe(struct platform_device *pdev) } sock->phys_mem = r->start; - /* - * pseudo-io: The 32bit address of the PCMCIA IO space for this - * socket (usually the 36bit address shifted 4 to the right). - */ + /* 36bit PCMCIA IO area address */ r = platform_get_resource_byname(pdev, IORESOURCE_MEM, "pcmcia-io"); if (!r) { dev_err(&pdev->dev, "missing 'pcmcia-io' resource!\n"); -- 1.7.0 From sebastian@breakpoint.cc Mon Mar 1 20:59:00 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Mon, 01 Mar 2010 20:59:07 +0100 (CET) Received: from Chamillionaire.breakpoint.cc ([85.10.199.196]:42692 "EHLO Chamillionaire.breakpoint.cc" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492324Ab0CAT7A (ORCPT ); Mon, 1 Mar 2010 20:59:00 +0100 Received: id: bigeasy by Chamillionaire.breakpoint.cc with local (easymta 1.00 BETA 1) id 1NmBla-0007Ks-2h; Mon, 01 Mar 2010 20:58:58 +0100 Date: Mon, 1 Mar 2010 20:58:58 +0100 From: Sebastian Andrzej Siewior To: David Miller Cc: ralf@linux-mips.org, linux-mips@linux-mips.org, linux-ide@vger.kernel.org Subject: Re: [PATCH 3/3] ide: move dcache flushing to generic ide code Message-ID: <20100301195858.GA27906@Chamillionaire.breakpoint.cc> References: <1267371341-16684-1-git-send-email-sebastian@breakpoint.cc> <1267371341-16684-4-git-send-email-sebastian@breakpoint.cc> <20100228.183417.52179576.davem@davemloft.net> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-15 Content-Disposition: inline In-Reply-To: <20100228.183417.52179576.davem@davemloft.net> X-Key-Id: FE3F4706 X-Key-Fingerprint: FFDA BBBB 3563 1B27 75C9 925B 98D5 5C1C FE3F 4706 User-Agent: Mutt/1.5.20 (2009-06-14) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26077 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: sebastian@breakpoint.cc Precedence: bulk X-list: linux-mips * David Miller | 2010-02-28 18:34:17 [-0800]: >From: Sebastian Andrzej Siewior >Date: Sun, 28 Feb 2010 16:35:41 +0100 > >> the pio callbacks are called with different kind of buffers. It could be >> a straight kernel addr, kernel stack or a kmaped highmem page. >> Some of this break the virt_to_page() assumptions. >> This patch moves the dcache flush from architecture code into generic >> ide code. ide_pio_bytes() is the only place where user pages might be >> written as far as I can see. >> The dcache flush is avoided in two cases: >> - data which is written to the device (i.e. they are comming from the >> userland) > >This needs a flush on sparc, otherwise an alias now exists in the >kernel side copy of the buffer. The D-cache flush is intentionally >unconditional for PIO mode. I definitely don't want to take the same >risks you guys seem to be willing to take for this optimization which >is of questionable value. It is not us guys it is just me. And if it is a stupid thing to do then I get probably punched by Ralf as well once he gets back. The part I don't get is why you have to flush dcache after the copy process. I would understand a flush before reading. The dcache alias in kernel shouldn't hurt since it is not used anymore. Unless we read twice from the same page. Is this the trick? >I also, intrinsically, really don't like these changes. > >For one thing, you're optimizing PIO mode. > >Secondly, IDE is in deep maintainence mode, if you want to optimize >cache flushing do it in the ATA layer. This patch patch was mostly driven by the fact that the buffer can be a normal kernel mapping or a virtual address. The latter doesn't work with virt_to_page(). Anyway I should probably spent more time getting ATA layer wokring than on the IDE layer since it is somehow working since patch#1. > >Thanks. Sebastian From linux@dominikbrodowski.net Mon Mar 1 21:30:19 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Mon, 01 Mar 2010 21:30:22 +0100 (CET) Received: from isilmar.linta.de ([213.133.102.198]:53480 "EHLO linta.de" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1491862Ab0CAUaS (ORCPT ); Mon, 1 Mar 2010 21:30:18 +0100 Received: (qmail 22477 invoked from network); 1 Mar 2010 20:29:36 -0000 Received: from unknown (HELO comet.dominikbrodowski.net) (brodo@82.113.106.134) by isilmar.linta.de with (DHE-RSA-AES256-SHA encrypted) SMTP; 1 Mar 2010 20:29:36 -0000 Received: by comet.dominikbrodowski.net (Postfix, from userid 1000) id 7A1C480093; Mon, 1 Mar 2010 20:39:17 +0100 (CET) Date: Mon, 1 Mar 2010 20:39:17 +0100 From: Dominik Brodowski To: Manuel Lauss Cc: Linux-PCMCIA , Linux-MIPS , Manuel Lauss Subject: Re: [PATCH] pcmcia: alchemy: fixup wrong comments Message-ID: <20100301193917.GA28486@comet.dominikbrodowski.net> Mail-Followup-To: Manuel Lauss , Linux-PCMCIA , Linux-MIPS , Manuel Lauss References: <1267462718-17383-1-git-send-email-manuel.lauss@gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1267462718-17383-1-git-send-email-manuel.lauss@gmail.com> User-Agent: Mutt/1.5.20 (2009-06-14) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26078 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: linux@dominikbrodowski.net Precedence: bulk X-list: linux-mips Hey Manuel, On Mon, Mar 01, 2010 at 05:58:38PM +0100, Manuel Lauss wrote: > Commit 11b897cf84c37e6522db914793677e933ef311fb changed expected > pcmcia area addresses from the 32bit pseudo to the real 36bit > addresses, but did not update the comments. thanks for the patch. Could you add your Signed-off-by (as per Documentation/SubmittingPatches), please? Thanks Dominik From f.fainelli@gmail.com Mon Mar 1 23:36:18 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Mon, 01 Mar 2010 23:36:21 +0100 (CET) Received: from mail-bw0-f226.google.com ([209.85.218.226]:39727 "EHLO mail-bw0-f226.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492006Ab0CAWgS (ORCPT ); Mon, 1 Mar 2010 23:36:18 +0100 Received: by bwz26 with SMTP id 26so2428417bwz.27 for ; Mon, 01 Mar 2010 14:36:11 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:sender:from:date:subject :x-uid:x-length:to:reply-to:mime-version:content-type :content-transfer-encoding:message-id; bh=ck3Pkj6IVakE0a6mFqo8Mar82YUcpwOZErBdkRFSbsY=; b=TPPdcdtt2UpwDxmmYAi2IZizAIurDdkQqEdaJsV285j0SBrpVGQes5lmaVsJvTb2/I 4dPnyjPKPhHII8bj2+qYuAnhgzux2jxEGuhPmIMVWACEuU69PTIxtV4x4cojQYdnm7u7 QjnZ70kNLvkjrBSM/DeEMpm8qhKmWAG4kMC+Y= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=sender:from:date:subject:x-uid:x-length:to:reply-to:mime-version :content-type:content-transfer-encoding:message-id; b=vAvFmrzuk/DaF/pAqrQStWoPHUS6XVo4Z+QoS/NvRFTGQ+hIzFrZnNTU55kCvSG9YZ X6BbXFZp7FhbvQ2yTk5/P6g8cNrGSbeymkqueDXj84uphCSIcVDAw49LhIJeaS+s+How pQ3fqyK6yqHXZgJ5RInRX4TDw9SlqvqF/99bA= Received: by 10.204.32.146 with SMTP id c18mr3553766bkd.185.1267482970851; Mon, 01 Mar 2010 14:36:10 -0800 (PST) Received: from lenovo.localnet (153.44.69-86.rev.gaoland.net [86.69.44.153]) by mx.google.com with ESMTPS id e18sm1094063bkd.5.2010.03.01.14.36.09 (version=TLSv1/SSLv3 cipher=RC4-MD5); Mon, 01 Mar 2010 14:36:10 -0800 (PST) From: Florian Fainelli Date: Mon, 1 Mar 2010 23:36:07 +0100 Subject: [PATCH 0/4] bcm63xx: misc fixes and new boards X-UID: 171 X-Length: 516 To: ralf@linux-mips.org, linux-mips@linux-mips.org Reply-To: Florian Fainelli MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Message-Id: <201003012336.08213.florian@openwrt.org> Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26079 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: florian@openwrt.org Precedence: bulk X-list: linux-mips Hi Ralf, These are some misc fixes and two new board definitions applying on top of Maxime's series: bcm63xx updates for 2.6.34 and bcm63xx_uart updates for 2.6.34 Florian Fainelli (4): bcm63xx: fix tabs vs spaces in board_bcm963xx.c bcm63xx: add rta1025w_16 board bcm63xx: add DWVS0 board bcm63xx: fix BCM6338 and BCM6345 gpio count Thanks! From f.fainelli@gmail.com Mon Mar 1 23:36:40 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Mon, 01 Mar 2010 23:36:46 +0100 (CET) Received: from mail-bw0-f226.google.com ([209.85.218.226]:38502 "EHLO mail-bw0-f226.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492355Ab0CAWgW (ORCPT ); Mon, 1 Mar 2010 23:36:22 +0100 Received: by bwz26 with SMTP id 26so2428504bwz.27 for ; Mon, 01 Mar 2010 14:36:17 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:sender:from:date:subject :mime-version:x-uid:x-length:to:reply-to:content-type :content-transfer-encoding:message-id; bh=v4kNBoTbH27PTU++wnf23JL9aSfxk6nOL/kyLTA7R1c=; b=uM2E/PDpB79d66jI76s7BuE82T8gn1/L+7BVT2K4Oty30XVgF39AuNStRM2ZqSeyuL hjVwaxzgjxIV6hSr+atarumSDgQsVDetsryM/ChxStfazPu5GSoO1K4IkqH1ZanQwjb4 EYhJ3WwtZeIH9GCRvsD4pD6CwYQIamIKEN0QI= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=sender:from:date:subject:mime-version:x-uid:x-length:to:reply-to :content-type:content-transfer-encoding:message-id; b=edbcEtddfOAg5OUfB0nKWFvP/fXm19Do+9kSi5dHLKncKX6in95Abkt5K8A4rpLUd2 uk7HcJjfldtnSRXmv4caI4dkXfm+ISozpPXMz2tyrTwMltCoTFA4hGREga7FOOPLo5WA SyGGoZbhr9V0TXB7oVq7uK4uV5//DlL11hay8= Received: by 10.204.5.216 with SMTP id 24mr3550155bkw.141.1267482977126; Mon, 01 Mar 2010 14:36:17 -0800 (PST) Received: from lenovo.localnet (153.44.69-86.rev.gaoland.net [86.69.44.153]) by mx.google.com with ESMTPS id s17sm1094478bkd.6.2010.03.01.14.36.16 (version=TLSv1/SSLv3 cipher=RC4-MD5); Mon, 01 Mar 2010 14:36:16 -0800 (PST) From: Florian Fainelli Date: Mon, 1 Mar 2010 23:36:15 +0100 Subject: [PATCH 1/4] bcm63xx: fix tabs vs spaces in board_bcm963xx.c MIME-Version: 1.0 X-UID: 177 X-Length: 1342 To: ralf@linux-mips.org, linux-mips@linux-mips.org Reply-To: Florian Fainelli Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit Message-Id: <201003012336.15400.florian@openwrt.org> Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26080 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: florian@openwrt.org Precedence: bulk X-list: linux-mips Signed-off-by: Florian Fainelli --- diff --git a/arch/mips/bcm63xx/boards/board_bcm963xx.c b/arch/mips/bcm63xx/boards/board_bcm963xx.c index 90faa37..00fa96b 100644 --- a/arch/mips/bcm63xx/boards/board_bcm963xx.c +++ b/arch/mips/bcm63xx/boards/board_bcm963xx.c @@ -361,9 +361,9 @@ static struct board_info __initdata board_FAST2404 = { .expected_cpu_id = 0x6348, .has_uart0 = 1, - .has_enet0 = 1, - .has_enet1 = 1, - .has_pci = 1, + .has_enet0 = 1, + .has_enet1 = 1, + .has_pci = 1, .enet0 = { .has_phy = 1, From f.fainelli@gmail.com Mon Mar 1 23:37:04 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Mon, 01 Mar 2010 23:37:10 +0100 (CET) Received: from mail-bw0-f226.google.com ([209.85.218.226]:39727 "EHLO mail-bw0-f226.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492357Ab0CAWgY (ORCPT ); Mon, 1 Mar 2010 23:36:24 +0100 Received: by mail-bw0-f226.google.com with SMTP id 26so2428417bwz.27 for ; Mon, 01 Mar 2010 14:36:24 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:sender:from:date:subject :mime-version:x-uid:x-length:to:reply-to:content-type :content-transfer-encoding:message-id; bh=yOJHgz4oZ6lVxPr5zG8jLM+sNB+d+AM9ujFe9dpnsk8=; b=UNnhPJmXL+TtC1fCZoB+jKlzmjpsR+jLtzdMyhIUM2gx1/mfKJaxtQ2V/i6Brvcqzy P57G7vKL+xzxANvMAt/MMvDUaxOopoE2gXNRPIUexdYLNJxDLXMFKVfY8nMfecyNScLp 4BXoMSObhedsK3tWP392io9gZ3r5lP1iJd8IA= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=sender:from:date:subject:mime-version:x-uid:x-length:to:reply-to :content-type:content-transfer-encoding:message-id; b=JIJUMjgaFVoNy/gW/MQZaPNe1dDEOMxzg5BFPRgqjQJexHJppMTqsZAYXc88kXubWv OjKgNIM0b+QZLY9bw6CH6bRzOBmwKoTj66uiC4j8HLFI1viz7XyZWFBbVlOJ+LMqcot8 +8Wf6P/l18LH+6TGlcab2R70FB/dKvjQ2OMTA= Received: by 10.204.6.70 with SMTP id 6mr3672920bky.6.1267482984256; Mon, 01 Mar 2010 14:36:24 -0800 (PST) Received: from lenovo.localnet (153.44.69-86.rev.gaoland.net [86.69.44.153]) by mx.google.com with ESMTPS id s17sm1094853bkd.6.2010.03.01.14.36.23 (version=TLSv1/SSLv3 cipher=RC4-MD5); Mon, 01 Mar 2010 14:36:23 -0800 (PST) From: Florian Fainelli Date: Mon, 1 Mar 2010 23:36:22 +0100 Subject: [PATCH 2/4] bcm63xx: add rta1025w_16 board MIME-Version: 1.0 X-UID: 178 X-Length: 1823 To: ralf@linux-mips.org, linux-mips@linux-mips.org Reply-To: Florian Fainelli Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit Message-Id: <201003012336.22578.florian@openwrt.org> Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26081 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: florian@openwrt.org Precedence: bulk X-list: linux-mips Add the RTA1025W-16 BCM6348-based board to the list of suppported boards. Signed-off-by: Florian Fainelli --- diff --git a/arch/mips/bcm63xx/boards/board_bcm963xx.c b/arch/mips/bcm63xx/boards/board_bcm963xx.c index 00fa96b..0b1d60f 100644 --- a/arch/mips/bcm63xx/boards/board_bcm963xx.c +++ b/arch/mips/bcm63xx/boards/board_bcm963xx.c @@ -380,6 +380,25 @@ static struct board_info __initdata board_FAST2404 = { .has_ehci0 = 1, }; +static struct board_info __initdata board_rta1025w_16 = { + .name = "RTA1025W_16", + .expected_cpu_id = 0x6348, + + .has_enet0 = 1, + .has_enet1 = 1, + .has_pci = 1, + + .enet0 = { + .has_phy = 1, + .use_internal_phy = 1, + }, + .enet1 = { + .force_speed_100 = 1, + .force_duplex_full = 1, + }, +}; + + static struct board_info __initdata board_DV201AMR = { .name = "DV201AMR", .expected_cpu_id = 0x6348, @@ -569,6 +588,7 @@ static const struct board_info __initdata *bcm963xx_boards[] = { &board_FAST2404, &board_DV201AMR, &board_96348gw_a, + &board_rta1025w_16, #endif #ifdef CONFIG_BCM63XX_CPU_6358 From f.fainelli@gmail.com Mon Mar 1 23:37:28 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Mon, 01 Mar 2010 23:37:33 +0100 (CET) Received: from mail-bw0-f226.google.com ([209.85.218.226]:38502 "EHLO mail-bw0-f226.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492349Ab0CAWga (ORCPT ); Mon, 1 Mar 2010 23:36:30 +0100 Received: by mail-bw0-f226.google.com with SMTP id 26so2428504bwz.27 for ; Mon, 01 Mar 2010 14:36:30 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:sender:from:date:subject :mime-version:x-uid:x-length:to:reply-to:content-type :content-transfer-encoding:message-id; bh=cxpTMaEW94/4aqeJAKh7Jzx0BRNnmpm4tWu2KoqchoY=; b=vOJ5J41m9eDctw0w+QBnzUHVMRIRoihSqCYJIFNKYCIwsqBe4tPl4pYjzJorpgs/qT PtpwTxxySmeWAL5SyzP1pyLkf2nhDfrEDp9Aw9aseheVLmeqJ5obuUlCOmZG4W0r9wQb kGGB8rsrC3VwFwkK0uDX5p8KP4OIpvqVZzIYA= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=sender:from:date:subject:mime-version:x-uid:x-length:to:reply-to :content-type:content-transfer-encoding:message-id; b=VJkrRQPmgPdRW0G4NPqFH1qVddq6YRWpO2TXP0jSzzS9tmsft3MjoeIbQiRZ9me2Pi RVyyJz/4DKnHb4Y4F2irl/vz1kmcLDX0xI+s/c6UMI7EtJoc7OLzesQ97KDqEB9GlztJ JaU3wxBk8G9hnFefuBlzqtW/O01euLEmh+qA4= Received: by 10.204.20.137 with SMTP id f9mr3587707bkb.130.1267482990399; Mon, 01 Mar 2010 14:36:30 -0800 (PST) Received: from lenovo.localnet (153.44.69-86.rev.gaoland.net [86.69.44.153]) by mx.google.com with ESMTPS id p13sm1095155bkp.9.2010.03.01.14.36.29 (version=TLSv1/SSLv3 cipher=RC4-MD5); Mon, 01 Mar 2010 14:36:29 -0800 (PST) From: Florian Fainelli Date: Mon, 1 Mar 2010 23:36:27 +0100 Subject: [PATCH 3/4] bcm63xx: add DWVS0 board MIME-Version: 1.0 X-UID: 179 X-Length: 1693 To: ralf@linux-mips.org, linux-mips@linux-mips.org Reply-To: Florian Fainelli Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit Message-Id: <201003012336.28071.florian@openwrt.org> Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26082 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: florian@openwrt.org Precedence: bulk X-list: linux-mips The DWVS0 board is a BCM6358-based board with an on-board OHCI controler. Signed-off-by: Florian Fainelli --- diff --git a/arch/mips/bcm63xx/boards/board_bcm963xx.c b/arch/mips/bcm63xx/boards/board_bcm963xx.c index 0b1d60f..8bf2c01 100644 --- a/arch/mips/bcm63xx/boards/board_bcm963xx.c +++ b/arch/mips/bcm63xx/boards/board_bcm963xx.c @@ -567,6 +567,27 @@ static struct board_info __initdata board_AGPFS0 = { .has_ohci0 = 1, .has_ehci0 = 1, }; + +static struct board_info __initdata board_DWVS0 = { + .name = "DWV-S0", + .expected_cpu_id = 0x6358, + + .has_enet0 = 1, + .has_enet1 = 1, + .has_pci = 1, + + .enet0 = { + .has_phy = 1, + .use_internal_phy = 1, + }, + + .enet1 = { + .force_speed_100 = 1, + .force_duplex_full = 1, + }, + + .has_ohci0 = 1, +}; #endif /* @@ -595,6 +616,7 @@ static const struct board_info __initdata *bcm963xx_boards[] = { &board_96358vw, &board_96358vw2, &board_AGPFS0, + &board_DWVS0, #endif }; From f.fainelli@gmail.com Mon Mar 1 23:37:51 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Mon, 01 Mar 2010 23:37:56 +0100 (CET) Received: from mail-bw0-f226.google.com ([209.85.218.226]:39727 "EHLO mail-bw0-f226.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492362Ab0CAWge (ORCPT ); Mon, 1 Mar 2010 23:36:34 +0100 Received: by mail-bw0-f226.google.com with SMTP id 26so2428417bwz.27 for ; Mon, 01 Mar 2010 14:36:34 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:sender:from:date:subject :mime-version:x-uid:x-length:to:reply-to:content-type :content-transfer-encoding:message-id; bh=yFpFrflWQiWE+hFwO8KN6XfCxd8+qeREgASUxSqoJsE=; b=RXm/d9LAFbsAQG+ngS9MIx22gV9bsbpTV+TRfFFJWHU8TKaZ7K5EaGmMveXfY1I5R9 HruDOWmnyQLnCk8MjePLAh8u/FeaZlmaSB/pHpiStS6bQyPm8OFIxND7AM0/HEHS0s0+ /MIJd2t8iIci/GoDfWg7S+urkUE32B/rQueVA= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=sender:from:date:subject:mime-version:x-uid:x-length:to:reply-to :content-type:content-transfer-encoding:message-id; b=cAvmu3Knz4tQIUvnKKD7CXtRr/unPK/Amzllk2S0TtREUb3lmJImVgTeFA+I9WdMFg iseTiT57BM9cOlNLWyx8kGeQpg+FpfHAJHJYrSRm5ptbsmt32d9QeLJuWoqdWRU7wxjq eOgCAUZ4EuI02ffhvlcALj7RT0Pe6GCi1KMW4= Received: by 10.204.138.81 with SMTP id z17mr3508964bkt.49.1267482994577; Mon, 01 Mar 2010 14:36:34 -0800 (PST) Received: from lenovo.localnet (153.44.69-86.rev.gaoland.net [86.69.44.153]) by mx.google.com with ESMTPS id e18sm1095439bkd.2.2010.03.01.14.36.33 (version=TLSv1/SSLv3 cipher=RC4-MD5); Mon, 01 Mar 2010 14:36:34 -0800 (PST) From: Florian Fainelli Date: Mon, 1 Mar 2010 23:36:32 +0100 Subject: [PATCH 4/4] bcm63xx: fix BCM6338 and BCM6345 gpio count MIME-Version: 1.0 X-UID: 180 X-Length: 1415 To: ralf@linux-mips.org, linux-mips@linux-mips.org Reply-To: Florian Fainelli Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit Message-Id: <201003012336.32430.florian@openwrt.org> Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26083 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: florian@openwrt.org Precedence: bulk X-list: linux-mips The number of GPIOs on BCM6338 is 8, while BCM6345 has only 16 GPIOs available. Signed-off-by: Florian Fainelli --- diff --git a/arch/mips/include/asm/mach-bcm63xx/bcm63xx_gpio.h b/arch/mips/include/asm/mach-bcm63xx/bcm63xx_gpio.h index 76a0b72..43d4da0 100644 --- a/arch/mips/include/asm/mach-bcm63xx/bcm63xx_gpio.h +++ b/arch/mips/include/asm/mach-bcm63xx/bcm63xx_gpio.h @@ -10,6 +10,10 @@ static inline unsigned long bcm63xx_gpio_count(void) switch (bcm63xx_get_cpu_id()) { case BCM6358_CPU_ID: return 40; + case BCM6338_CPU_ID: + return 8; + case BCM6345_CPU_ID: + return 16; case BCM6348_CPU_ID: default: return 37; From davem@davemloft.net Tue Mar 2 01:39:42 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Tue, 02 Mar 2010 01:39:47 +0100 (CET) Received: from 74-93-104-97-Washington.hfc.comcastbusiness.net ([74.93.104.97]:45273 "EHLO sunset.davemloft.net" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491139Ab0CBAjm (ORCPT ); Tue, 2 Mar 2010 01:39:42 +0100 Received: from localhost (localhost [127.0.0.1]) by sunset.davemloft.net (Postfix) with ESMTP id 75C7624C094; Mon, 1 Mar 2010 16:39:59 -0800 (PST) Date: Mon, 01 Mar 2010 16:39:59 -0800 (PST) Message-Id: <20100301.163959.177031088.davem@davemloft.net> To: sebastian@breakpoint.cc Cc: ralf@linux-mips.org, linux-mips@linux-mips.org, linux-ide@vger.kernel.org Subject: Re: [PATCH 3/3] ide: move dcache flushing to generic ide code From: David Miller In-Reply-To: <20100301195858.GA27906@Chamillionaire.breakpoint.cc> References: <1267371341-16684-4-git-send-email-sebastian@breakpoint.cc> <20100228.183417.52179576.davem@davemloft.net> <20100301195858.GA27906@Chamillionaire.breakpoint.cc> X-Mailer: Mew version 6.3 on Emacs 23.1 / Mule 6.0 (HANACHIRUSATO) Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26084 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: davem@davemloft.net Precedence: bulk X-list: linux-mips From: Sebastian Andrzej Siewior Date: Mon, 1 Mar 2010 20:58:58 +0100 > The part I don't get is why you have to flush dcache after the copy > process. I would understand a flush before reading. The dcache alias in > kernel shouldn't hurt since it is not used anymore. Unless we read twice > from the same page. Is this the trick? Anything that puts the data into the cache on the kernel side is a problem. The page is still potentially in user space, as a result there will be thus two active mappings in the cache, one in the kernel side and one in the user side. The user can then do various operations which can access either mapping. Writing to it via write() system call, writing to it via mmap(), making the kernel write to it by doing a read() with the buffer pointing into the mmap() area. All we need is a modification on either side for the other one to potentially become stale. >>Secondly, IDE is in deep maintainence mode, if you want to optimize >>cache flushing do it in the ATA layer. > This patch patch was mostly driven by the fact that the buffer can be a > normal kernel mapping or a virtual address. The latter doesn't work with > virt_to_page(). > Anyway I should probably spent more time getting ATA layer wokring than > on the IDE layer since it is somehow working since patch#1. All buffers passed to the architecture IDE ops should be PAGE_OFFSET relative kernel virtual addresses for which __pa() works. Are kmap()'d things ending up here? It all works out on sparc64 because we don't have highmem and kernel stacks are just in normal kernel pages. From sshtylyov@mvista.com Tue Mar 2 11:34:23 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Tue, 02 Mar 2010 11:34:27 +0100 (CET) Received: from mail-bw0-f215.google.com ([209.85.218.215]:46963 "EHLO mail-bw0-f215.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491000Ab0CBKeW (ORCPT ); Tue, 2 Mar 2010 11:34:22 +0100 Received: by bwz7 with SMTP id 7so71660bwz.24 for ; Tue, 02 Mar 2010 02:34:17 -0800 (PST) Received: by 10.204.4.139 with SMTP id 11mr4395227bkr.27.1267526057143; Tue, 02 Mar 2010 02:34:17 -0800 (PST) Received: from ?192.168.2.2? (ppp91-77-213-207.pppoe.mtu-net.ru [91.77.213.207]) by mx.google.com with ESMTPS id 15sm2688896bwz.8.2010.03.02.02.34.15 (version=TLSv1/SSLv3 cipher=RC4-MD5); Tue, 02 Mar 2010 02:34:16 -0800 (PST) Message-ID: <4B8CE98D.406@ru.mvista.com> Date: Tue, 02 Mar 2010 13:33:49 +0300 From: Sergei Shtylyov User-Agent: Thunderbird 2.0.0.23 (Windows/20090812) MIME-Version: 1.0 To: Florian Fainelli CC: ralf@linux-mips.org, linux-mips@linux-mips.org Subject: Re: [PATCH 3/4] bcm63xx: add DWVS0 board References: <201003012336.28071.florian@openwrt.org> In-Reply-To: <201003012336.28071.florian@openwrt.org> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26085 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: sshtylyov@mvista.com Precedence: bulk X-list: linux-mips Hello. Florian Fainelli wrote: > The DWVS0 board is a BCM6358-based board with an on-board OHCI controler. > > Signed-off-by: Florian Fainelli > --- > diff --git a/arch/mips/bcm63xx/boards/board_bcm963xx.c b/arch/mips/bcm63xx/boards/board_bcm963xx.c > index 0b1d60f..8bf2c01 100644 > --- a/arch/mips/bcm63xx/boards/board_bcm963xx.c > +++ b/arch/mips/bcm63xx/boards/board_bcm963xx.c > @@ -567,6 +567,27 @@ static struct board_info __initdata board_AGPFS0 = { > .has_ohci0 = 1, > .has_ehci0 = 1, > }; > + > +static struct board_info __initdata board_DWVS0 = { > + .name = "DWV-S0", > + .expected_cpu_id = 0x6358, > + > + .has_enet0 = 1, > + .has_enet1 = 1, > + .has_pci = 1, > + > + .enet0 = { > + .has_phy = 1, > + .use_internal_phy = 1, > + }, > + > + .enet1 = { > + .force_speed_100 = 1, > + .force_duplex_full = 1, > + }, > + > + .has_ohci0 = 1, > Why it's not aligned with the rest of initializers? WBR, Sergei From ffainelli@freebox.fr Tue Mar 2 14:39:01 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Tue, 02 Mar 2010 14:39:04 +0100 (CET) Received: from zmc.proxad.net ([212.27.53.206]:35056 "EHLO zmc.proxad.net" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491138Ab0CBNjA (ORCPT ); Tue, 2 Mar 2010 14:39:00 +0100 Received: from localhost (localhost [127.0.0.1]) by zmc.proxad.net (Postfix) with ESMTP id 20B413418153; Tue, 2 Mar 2010 14:38:56 +0100 (CET) X-Virus-Scanned: amavisd-new at Received: from zmc.proxad.net ([127.0.0.1]) by localhost (zmc.proxad.net [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id CZNKzExolfNX; Tue, 2 Mar 2010 14:38:55 +0100 (CET) Received: from flexo.localnet (bobafett.staff.proxad.net [213.228.1.121]) by zmc.proxad.net (Postfix) with ESMTPSA id 8562F341814A; Tue, 2 Mar 2010 14:38:55 +0100 (CET) From: Florian Fainelli Organization: Freebox To: Maxime Bizon Subject: Re: [PATCH 6/7] MIPS: bcm63xx: call board_register_device from device_initcall() Date: Tue, 2 Mar 2010 14:38:47 +0100 User-Agent: KMail/1.12.2 (Linux/2.6.31-17-server; KDE/4.3.2; x86_64; ; ) Cc: linux-mips@linux-mips.org, ralf@linux-mips.org References: <1264872898-28149-1-git-send-email-mbizon@freebox.fr> <1264872898-28149-7-git-send-email-mbizon@freebox.fr> In-Reply-To: <1264872898-28149-7-git-send-email-mbizon@freebox.fr> MIME-Version: 1.0 Content-Type: Text/Plain; charset="utf-8" Content-Transfer-Encoding: 7bit Message-Id: <201003021438.47105.ffainelli@freebox.fr> Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26086 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ffainelli@freebox.fr Precedence: bulk X-list: linux-mips Hi Maxime, On Saturday 30 January 2010 18:34:57 Maxime Bizon wrote: > Some device registration (eg leds), expect subsystem initcall to be > run first, so move board device registration to device_initcall(). > > Signed-off-by: Maxime Bizon > --- > arch/mips/bcm63xx/setup.c | 2 +- > 1 files changed, 1 insertions(+), 1 deletions(-) > > diff --git a/arch/mips/bcm63xx/setup.c b/arch/mips/bcm63xx/setup.c > index d005659..04a3499 100644 > --- a/arch/mips/bcm63xx/setup.c > +++ b/arch/mips/bcm63xx/setup.c > @@ -124,4 +124,4 @@ int __init bcm63xx_register_devices(void) > return board_register_devices(); > } > > -arch_initcall(bcm63xx_register_devices); > +device_initcall(bcm63xx_register_devices); This breaks the fallback SPROM registration, that one needs to be set before the PCI subsystem is intialized, otherwise b43 gets an empty WLAN MAC address. This was the reason to move bcm63xx_register_devices to the arch_initcall level in the first place. Let's use the first stage board callback which also has to be called prior to PCI initialization. From: Florian Fainelli Subject: [PATCH] bcm63xx: register SSB SPROM fallback in board's first stage callback Signed-off-by: Florian Fainelli --- diff --git a/arch/mips/bcm63xx/boards/board_bcm963xx.c b/arch/mips/bcm63xx/boards/board_bcm963xx.c index 90faa37..f7e0be1 100644 --- a/arch/mips/bcm63xx/boards/board_bcm963xx.c +++ b/arch/mips/bcm63xx/boards/board_bcm963xx.c @@ -676,6 +676,17 @@ void __init board_prom_init(void) } bcm_gpio_writel(val, GPIO_MODE_REG); + + /* Generate MAC address for WLAN and + * register our SPROM */ +#ifdef CONFIG_SSB_PCIHOST + if (!board_get_mac_address(bcm63xx_sprom.il0mac)) { + memcpy(bcm63xx_sprom.et0mac, bcm63xx_sprom.il0mac, ETH_ALEN); + memcpy(bcm63xx_sprom.et1mac, bcm63xx_sprom.il0mac, ETH_ALEN); + if (ssb_arch_set_fallback_sprom(&bcm63xx_sprom) < 0) + printk(KERN_ERR "failed to register fallback SPROM\n"); + } +#endif } /* @@ -835,17 +846,6 @@ int __init board_register_devices(void) if (board.has_dsp) bcm63xx_dsp_register(&board.dsp); - /* Generate MAC address for WLAN and - * register our SPROM */ -#ifdef CONFIG_SSB_PCIHOST - if (!board_get_mac_address(bcm63xx_sprom.il0mac)) { - memcpy(bcm63xx_sprom.et0mac, bcm63xx_sprom.il0mac, ETH_ALEN); - memcpy(bcm63xx_sprom.et1mac, bcm63xx_sprom.il0mac, ETH_ALEN); - if (ssb_arch_set_fallback_sprom(&bcm63xx_sprom) < 0) - printk(KERN_ERR "failed to register fallback SPROM\n"); - } -#endif - /* read base address of boot chip select (0) */ if (BCMCPU_IS_6345()) val = 0x1fc00000; From yad.naveen@gmail.com Tue Mar 2 14:54:23 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Tue, 02 Mar 2010 14:54:26 +0100 (CET) Received: from mail-pw0-f49.google.com ([209.85.160.49]:34107 "EHLO mail-pw0-f49.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491868Ab0CBNyX (ORCPT ); Tue, 2 Mar 2010 14:54:23 +0100 Received: by pwj2 with SMTP id 2so123886pwj.36 for ; Tue, 02 Mar 2010 05:54:15 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:date:message-id:subject :from:to:content-type; bh=e+7hUrAoTeTQ89bmtRz5Rs9hFOIBbo0LJxgw9wtlCDI=; b=xeki5kE1u0M7QtVk24PoRLSm8knZAZVNvvHadlw34AIMHvMUGAbEUIP6/XBlkYApE9 UsZwb8sb1lMH+S5ZUivVtVZJSd/b5PwFy7dCyh47u+ij0IWxTEAMpHEr/Gxft9auAJp3 YlCHP4ybScR5tl/zx9MwBKFPZs7TLR5tm4Bl8= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:date:message-id:subject:from:to:content-type; b=sGlrxz39XJ0nvMH5WbPDJOgxOcfqKnvWCHHfWxlAyDjO4m4Gn4wGMufvlbnp8+FCTy Fb9HeKkkh5Fiol7XA2JwyCQ84K1a/nYXas+NCtOenL/zqPbduLhMD/RWIGafnKAxVHYP lKS8Z5fWXX5QPPGY6hgl+KXq2+TZoQQhccd+A= MIME-Version: 1.0 Received: by 10.114.22.12 with SMTP id 12mr252391wav.205.1267538055383; Tue, 02 Mar 2010 05:54:15 -0800 (PST) Date: Tue, 2 Mar 2010 19:24:15 +0530 Message-ID: <9bde694e1003020554p7c8ff3c2o4ae7cb5d501d1ab9@mail.gmail.com> Subject: kmemleak for MIPS From: naveen yadav To: linux-mips@linux-mips.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org Content-Type: text/plain; charset=ISO-8859-1 Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26087 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: yad.naveen@gmail.com Precedence: bulk X-list: linux-mips Hi all, I want to check kmemleak for both ARM/MIPS. i am able to find kernel patch for ARM at http://linux.derkeiler.com/Mailing-Lists/Kernel/2009-04/msg11830.html. But I could not able to trace patch for MIPS. Kind Regards Naveen From andrea.gelmini@gmail.com Tue Mar 2 16:13:12 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Tue, 02 Mar 2010 16:13:15 +0100 (CET) Received: from fg-out-1718.google.com ([72.14.220.156]:55846 "EHLO fg-out-1718.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491868Ab0CBPNM (ORCPT ); Tue, 2 Mar 2010 16:13:12 +0100 Received: by fg-out-1718.google.com with SMTP id d23so166329fga.6 for ; Tue, 02 Mar 2010 07:13:11 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:sender:received:in-reply-to :references:date:x-google-sender-auth:message-id:subject:from:to:cc :content-type; bh=y6SA0qJy0uuBvYZMW5YE7cDsjIlO3VOpEV7Mo/tWzzM=; b=XMUemizOrxbN71ljOCr/tsmJML2wyNJuK07dZ4sammeqvchBu/7K1m4x7jdl8oC1U9 i05oT0Hpe/lg7eatGgSmZ5o9obyemYG6KQrtFGC5Ocm1RztRlKvOtGDIV3kd/1UdMs8W 6Is4+Y5LQyDin52Lx7/RfT4fDfJSYjyAAuek8= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:cc:content-type; b=PPQ0r/9CHIylOGUv7vbVV1Siz6rw8KKsLvTaNLwrLZQhVai7Zk/5yg/Lp1p1AEmyEm 3QCTaFsXR56e0LGblfj5joYGZMmuVG4IGQTupJRAdnjJSfaZdzAkyW6NUHY8xuA6lemT X4eY1bqaOoZroXQIJOWFnsCucpDutspT+QXnY= MIME-Version: 1.0 Received: by 10.87.65.9 with SMTP id s9mr9877496fgk.48.1267542791131; Tue, 02 Mar 2010 07:13:11 -0800 (PST) In-Reply-To: <20100301023316.GB23086@linux-sh.org> References: <1267289508-31031-1-git-send-email-andrea.gelmini@gelma.net> <1267289508-31031-43-git-send-email-andrea.gelmini@gelma.net> <20100301023316.GB23086@linux-sh.org> Date: Tue, 2 Mar 2010 16:13:11 +0100 X-Google-Sender-Auth: 718d2b274871f915 Message-ID: <9cdbb57f1003020713x4e897af9ka87e348bf782f380@mail.gmail.com> Subject: Re: [PATCH 42/66] arch/mips/lib/libgcc.h: Checkpatch cleanup From: Andrea Gelmini To: Paul Mundt Cc: linux-kernel@vger.kernel.org, Ralf Baechle , linux-mips@linux-mips.org, linux-sh@vger.kernel.org Content-Type: text/plain; charset=UTF-8 Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26088 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: andrea.gelmini@gelma.net Precedence: bulk X-list: linux-mips 2010/3/1 Paul Mundt : Hi Paul, thanks for your reply. > I'll apply the sh part by itself, and I suppose Ralf will do the same for > MIPS. I'm a bit confused as to why these were lumped together when 65 out > of 66 oneliner patches made no use of ad-hoc grouping. After the experience with GregKH, I was looking for others maintainer interested in code cleanup. So I sent some patches over random files. If you care, I can checkpatch.pl all mips subtree and send you patches (if any). Thanks a lot for your work, Andrea From sebastian@breakpoint.cc Tue Mar 2 22:14:18 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Tue, 02 Mar 2010 22:14:21 +0100 (CET) Received: from Chamillionaire.breakpoint.cc ([85.10.199.196]:37840 "EHLO Chamillionaire.breakpoint.cc" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492382Ab0CBVOS (ORCPT ); Tue, 2 Mar 2010 22:14:18 +0100 Received: id: bigeasy by Chamillionaire.breakpoint.cc with local (easymta 1.00 BETA 1) id 1NmZPz-0001IG-2B; Tue, 02 Mar 2010 22:14:15 +0100 Date: Tue, 2 Mar 2010 22:14:15 +0100 From: Sebastian Andrzej Siewior To: David Miller Cc: sebastian@breakpoint.cc, ralf@linux-mips.org, linux-mips@linux-mips.org, linux-ide@vger.kernel.org Subject: Re: [PATCH 3/3] ide: move dcache flushing to generic ide code Message-ID: <20100302211414.GA4065@Chamillionaire.breakpoint.cc> References: <1267371341-16684-4-git-send-email-sebastian@breakpoint.cc> <20100228.183417.52179576.davem@davemloft.net> <20100301195858.GA27906@Chamillionaire.breakpoint.cc> <20100301.163959.177031088.davem@davemloft.net> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-15 Content-Disposition: inline In-Reply-To: <20100301.163959.177031088.davem@davemloft.net> X-Key-Id: FE3F4706 X-Key-Fingerprint: FFDA BBBB 3563 1B27 75C9 925B 98D5 5C1C FE3F 4706 User-Agent: Mutt/1.5.20 (2009-06-14) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26089 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: sebastian@breakpoint.cc Precedence: bulk X-list: linux-mips * David Miller | 2010-03-01 16:39:59 [-0800]: >All buffers passed to the architecture IDE ops should be PAGE_OFFSET >relative kernel virtual addresses for which __pa() works. > >Are kmap()'d things ending up here? Ah, bounce buffers are used. So no, there no highmem which get kmap()ed. Sebastian From yang.shi@windriver.com Wed Mar 3 08:26:27 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 03 Mar 2010 08:26:30 +0100 (CET) Received: from mail.windriver.com ([147.11.1.11]:33727 "EHLO mail.windriver.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491077Ab0CCH00 (ORCPT ); Wed, 3 Mar 2010 08:26:26 +0100 Received: from localhost.localdomain (pek-lpgbuild1.wrs.com [128.224.153.29]) by mail.windriver.com (8.14.3/8.14.3) with ESMTP id o237QD6u002456; Tue, 2 Mar 2010 23:26:14 -0800 (PST) From: Yang Shi To: ddaney@caviumnetworks.com, ralf@linux-mips.org Cc: linux-mips@linux-mips.org Subject: Octeon: Several misc fixes Date: Wed, 3 Mar 2010 15:26:09 +0800 Message-Id: <1267601172-17919-1-git-send-email-yang.shi@windriver.com> X-Mailer: git-send-email 1.6.0.4 Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26090 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: yang.shi@windriver.com Precedence: bulk X-list: linux-mips Hi David & Ralf, These are some misc build fixes for Octeon. Yang Shi (3): MIPS: Octeon: Remove superfluous on_each_cpu parameter MIPS: Octeon: Remove redundant declaration of octeon_reserve32_memory MIPS: Octeon: Add add_wired_entry decralation in header file Thanks, Yang From yang.shi@windriver.com Wed Mar 3 08:26:49 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 03 Mar 2010 08:26:55 +0100 (CET) Received: from mail.windriver.com ([147.11.1.11]:33740 "EHLO mail.windriver.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491110Ab0CCH03 (ORCPT ); Wed, 3 Mar 2010 08:26:29 +0100 Received: from localhost.localdomain (pek-lpgbuild1.wrs.com [128.224.153.29]) by mail.windriver.com (8.14.3/8.14.3) with ESMTP id o237QD6w002456; Tue, 2 Mar 2010 23:26:21 -0800 (PST) From: Yang Shi To: ddaney@caviumnetworks.com, ralf@linux-mips.org Cc: linux-mips@linux-mips.org Subject: [PATCH 2/3] MIPS: Octeon: Remove redundant declaration of octeon_reserve32_memory Date: Wed, 3 Mar 2010 15:26:11 +0800 Message-Id: <3118b3d0f3ed042df1ee2771325c3824e6fc7ba9.1267600234.git.yang.shi@windriver.com> X-Mailer: git-send-email 1.6.0.4 In-Reply-To: <6310b9cb3048ec0c2873d932778165370e5e7c7e.1267600234.git.yang.shi@windriver.com> References: <1267601172-17919-1-git-send-email-yang.shi@windriver.com> <6310b9cb3048ec0c2873d932778165370e5e7c7e.1267600234.git.yang.shi@windriver.com> In-Reply-To: <6310b9cb3048ec0c2873d932778165370e5e7c7e.1267600234.git.yang.shi@windriver.com> References: <6310b9cb3048ec0c2873d932778165370e5e7c7e.1267600234.git.yang.shi@windriver.com> Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26091 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: yang.shi@windriver.com Precedence: bulk X-list: linux-mips In Octeon's setup.c, octeon_reserve32_memory is defined, so remove the redundant extern declaration of this variable. Signed-off-by: Yang Shi --- arch/mips/cavium-octeon/setup.c | 3 --- 1 files changed, 0 insertions(+), 3 deletions(-) diff --git a/arch/mips/cavium-octeon/setup.c b/arch/mips/cavium-octeon/setup.c index 4eaa35f..8309d68 100644 --- a/arch/mips/cavium-octeon/setup.c +++ b/arch/mips/cavium-octeon/setup.c @@ -45,9 +45,6 @@ extern struct plat_smp_ops octeon_smp_ops; extern void pci_console_init(const char *arg); #endif -#ifdef CONFIG_CAVIUM_RESERVE32 -extern uint64_t octeon_reserve32_memory; -#endif static unsigned long long MAX_MEMORY = 512ull << 20; struct octeon_boot_descriptor *octeon_boot_desc_ptr; -- 1.6.3.3 From yang.shi@windriver.com Wed Mar 3 08:27:14 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 03 Mar 2010 08:27:21 +0100 (CET) Received: from mail.windriver.com ([147.11.1.11]:33746 "EHLO mail.windriver.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491194Ab0CCH0b (ORCPT ); Wed, 3 Mar 2010 08:26:31 +0100 Received: from localhost.localdomain (pek-lpgbuild1.wrs.com [128.224.153.29]) by mail.windriver.com (8.14.3/8.14.3) with ESMTP id o237QD6x002456; Tue, 2 Mar 2010 23:26:23 -0800 (PST) From: Yang Shi To: ddaney@caviumnetworks.com, ralf@linux-mips.org Cc: linux-mips@linux-mips.org Subject: [PATCH 3/3] MIPS: Octeon: Add add_wired_entry decralation in header file Date: Wed, 3 Mar 2010 15:26:12 +0800 Message-Id: <9e4e80f8edd43f8a164fe618c978c1dc8cd48a69.1267600234.git.yang.shi@windriver.com> X-Mailer: git-send-email 1.6.0.4 In-Reply-To: <3118b3d0f3ed042df1ee2771325c3824e6fc7ba9.1267600234.git.yang.shi@windriver.com> References: <1267601172-17919-1-git-send-email-yang.shi@windriver.com> <6310b9cb3048ec0c2873d932778165370e5e7c7e.1267600234.git.yang.shi@windriver.com> <3118b3d0f3ed042df1ee2771325c3824e6fc7ba9.1267600234.git.yang.shi@windriver.com> In-Reply-To: <6310b9cb3048ec0c2873d932778165370e5e7c7e.1267600234.git.yang.shi@windriver.com> References: <6310b9cb3048ec0c2873d932778165370e5e7c7e.1267600234.git.yang.shi@windriver.com> Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26092 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: yang.shi@windriver.com Precedence: bulk X-list: linux-mips Octeon's setup.c uses add_wired_entry, but it is not declared anywhere. Copy add_wired_entry decralation fomr pgtable-32.h to pgtable-64.h and include asm/pgtable.h into Octeon's setup.c. Signed-off-by: Yang Shi --- arch/mips/cavium-octeon/setup.c | 1 + arch/mips/include/asm/pgtable-64.h | 6 ++++++ 2 files changed, 7 insertions(+), 0 deletions(-) diff --git a/arch/mips/cavium-octeon/setup.c b/arch/mips/cavium-octeon/setup.c index 8309d68..f35ba16 100644 --- a/arch/mips/cavium-octeon/setup.c +++ b/arch/mips/cavium-octeon/setup.c @@ -30,6 +30,7 @@ #include #include #include +#include #include diff --git a/arch/mips/include/asm/pgtable-64.h b/arch/mips/include/asm/pgtable-64.h index 26dc69d..85ee34d 100644 --- a/arch/mips/include/asm/pgtable-64.h +++ b/arch/mips/include/asm/pgtable-64.h @@ -23,6 +23,12 @@ #endif /* + * - add_wired_entry() add a fixed TLB entry, and move wired register + */ +extern void add_wired_entry(unsigned long entrylo0, unsigned long entrylo1, + unsigned long entryhi, unsigned long pagemask); + +/* * Each address space has 2 4K pages as its page directory, giving 1024 * (== PTRS_PER_PGD) 8 byte pointers to pmd tables. Each pmd table is a * single 4K page, giving 512 (== PTRS_PER_PMD) 8 byte pointers to page -- 1.6.3.3 From yang.shi@windriver.com Wed Mar 3 08:27:39 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 03 Mar 2010 08:27:45 +0100 (CET) Received: from mail.windriver.com ([147.11.1.11]:33750 "EHLO mail.windriver.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491772Ab0CCH0c (ORCPT ); Wed, 3 Mar 2010 08:26:32 +0100 Received: from localhost.localdomain (pek-lpgbuild1.wrs.com [128.224.153.29]) by mail.windriver.com (8.14.3/8.14.3) with ESMTP id o237QD6v002456; Tue, 2 Mar 2010 23:26:19 -0800 (PST) From: Yang Shi To: ddaney@caviumnetworks.com, ralf@linux-mips.org Cc: linux-mips@linux-mips.org Subject: [PATCH 1/3] MIPS: Octeon: Remove superfluous on_each_cpu parameter Date: Wed, 3 Mar 2010 15:26:10 +0800 Message-Id: <6310b9cb3048ec0c2873d932778165370e5e7c7e.1267600234.git.yang.shi@windriver.com> X-Mailer: git-send-email 1.6.0.4 In-Reply-To: <1267601172-17919-1-git-send-email-yang.shi@windriver.com> References: <1267601172-17919-1-git-send-email-yang.shi@windriver.com> Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26093 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: yang.shi@windriver.com Precedence: bulk X-list: linux-mips Now, on_each_cpu just need three parameters, but the on_each_cpu still uses four parameters in Octeon's setup.c. So, remove the superfluous parameter. Signed-off-by: Yang Shi --- arch/mips/cavium-octeon/setup.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/arch/mips/cavium-octeon/setup.c b/arch/mips/cavium-octeon/setup.c index b321d3b..4eaa35f 100644 --- a/arch/mips/cavium-octeon/setup.c +++ b/arch/mips/cavium-octeon/setup.c @@ -230,7 +230,7 @@ static void octeon_hal_setup_per_cpu_reserved32(void *unused) void octeon_hal_setup_reserved32(void) { #ifdef CONFIG_CAVIUM_RESERVE32_USE_WIRED_TLB - on_each_cpu(octeon_hal_setup_per_cpu_reserved32, NULL, 0, 1); + on_each_cpu(octeon_hal_setup_per_cpu_reserved32, NULL, 1); #endif } -- 1.6.3.3 From f.fainelli@gmail.com Wed Mar 3 09:19:59 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 03 Mar 2010 09:20:05 +0100 (CET) Received: from mail-bw0-f219.google.com ([209.85.218.219]:48435 "EHLO mail-bw0-f219.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491020Ab0CCIT7 (ORCPT ); Wed, 3 Mar 2010 09:19:59 +0100 Received: by bwz19 with SMTP id 19so1021654bwz.26 for ; Wed, 03 Mar 2010 00:19:53 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:sender:from:organization:to :subject:date:user-agent:cc:references:in-reply-to:mime-version :content-type:content-transfer-encoding:message-id; bh=kr9uGS0MAVwzwYAwm0iw6w9wM1ehG4UlOvA6gWqUcUM=; b=FmmE9SIC7noNb+GYpANFKy0LsO2hqdRPanM6IwbhAd0wMvKJyv0JCC4nqHhd/RLPnU zHPWXAla4D6BYHJrzfgGTwaKWtVz/rHkN3UK9Im3G6SDrTButRZ6IYsXaqAHgor10JHl wYxnGSsdUtdrbhHsbw0JoiI1rE39cTSHfwqzQ= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=sender:from:organization:to:subject:date:user-agent:cc:references :in-reply-to:mime-version:content-type:content-transfer-encoding :message-id; b=x4+HVHq6xRHXn7CSgnWMkvElIuxYVl/QDQI2P0QLZimfIROrWHj2OVCxMinMfW1Vu/ eHl4saYraJnYjehW7hxAAQ3ssbiOH28auKNqJUcs7dzOqH8fx7Ef+3O0YqyPC6AGxdF+ pO7RxU2CUtKoCV45oPcgnd731WvUlbRPuBssg= Received: by 10.204.144.78 with SMTP id y14mr5653639bku.194.1267604393186; Wed, 03 Mar 2010 00:19:53 -0800 (PST) Received: from flexo.localnet (bobafett.staff.proxad.net [213.228.1.121]) by mx.google.com with ESMTPS id x16sm816522bku.17.2010.03.03.00.19.52 (version=SSLv3 cipher=RC4-MD5); Wed, 03 Mar 2010 00:19:52 -0800 (PST) From: Florian Fainelli Organization: OpenWrt To: Yang Shi Subject: Re: [PATCH 3/3] MIPS: Octeon: Add add_wired_entry decralation in header file Date: Wed, 3 Mar 2010 09:19:35 +0100 User-Agent: KMail/1.12.2 (Linux/2.6.31-17-server; KDE/4.3.2; x86_64; ; ) Cc: ddaney@caviumnetworks.com, ralf@linux-mips.org, linux-mips@linux-mips.org References: <1267601172-17919-1-git-send-email-yang.shi@windriver.com> <3118b3d0f3ed042df1ee2771325c3824e6fc7ba9.1267600234.git.yang.shi@windriver.com> <9e4e80f8edd43f8a164fe618c978c1dc8cd48a69.1267600234.git.yang.shi@windriver.com> In-Reply-To: <9e4e80f8edd43f8a164fe618c978c1dc8cd48a69.1267600234.git.yang.shi@windriver.com> MIME-Version: 1.0 Content-Type: Text/Plain; charset="utf-8" Content-Transfer-Encoding: 7bit Message-Id: <201003030919.36006.florian@openwrt.org> Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26094 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: florian@openwrt.org Precedence: bulk X-list: linux-mips Hi Yang, On Wednesday 03 March 2010 08:26:12 Yang Shi wrote: > Octeon's setup.c uses add_wired_entry, but it is not declared > anywhere. Copy add_wired_entry decralation fomr pgtable-32.h to > pgtable-64.h and include asm/pgtable.h into Octeon's setup.c. > > Signed-off-by: Yang Shi > --- > arch/mips/cavium-octeon/setup.c | 1 + > arch/mips/include/asm/pgtable-64.h | 6 ++++++ > 2 files changed, 7 insertions(+), 0 deletions(-) > > diff --git a/arch/mips/cavium-octeon/setup.c > b/arch/mips/cavium-octeon/setup.c index 8309d68..f35ba16 100644 > --- a/arch/mips/cavium-octeon/setup.c > +++ b/arch/mips/cavium-octeon/setup.c > @@ -30,6 +30,7 @@ > #include > #include > #include > +#include You probably meant to include instead. > > #include > > diff --git a/arch/mips/include/asm/pgtable-64.h > b/arch/mips/include/asm/pgtable-64.h index 26dc69d..85ee34d 100644 > --- a/arch/mips/include/asm/pgtable-64.h > +++ b/arch/mips/include/asm/pgtable-64.h > @@ -23,6 +23,12 @@ > #endif > > /* > + * - add_wired_entry() add a fixed TLB entry, and move wired register > + */ > +extern void add_wired_entry(unsigned long entrylo0, unsigned long > entrylo1, + unsigned long entryhi, unsigned long pagemask); > + > +/* > * Each address space has 2 4K pages as its page directory, giving 1024 > * (== PTRS_PER_PGD) 8 byte pointers to pmd tables. Each pmd table is a > * single 4K page, giving 512 (== PTRS_PER_PMD) 8 byte pointers to page > -- Regards, Florian From Yang.Shi@windriver.com Wed Mar 3 09:25:32 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 03 Mar 2010 09:25:36 +0100 (CET) Received: from mail.windriver.com ([147.11.1.11]:44859 "EHLO mail.windriver.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491772Ab0CCIZc (ORCPT ); Wed, 3 Mar 2010 09:25:32 +0100 Received: from ALA-MAIL03.corp.ad.wrs.com (ala-mail03 [147.11.57.144]) by mail.windriver.com (8.14.3/8.14.3) with ESMTP id o238PPPB011851; Wed, 3 Mar 2010 00:25:25 -0800 (PST) Received: from [128.224.161.163] ([128.224.161.163]) by ALA-MAIL03.corp.ad.wrs.com with Microsoft SMTPSVC(6.0.3790.1830); Wed, 3 Mar 2010 00:25:25 -0800 Message-ID: <4B8E1CF2.9070405@windriver.com> Date: Wed, 03 Mar 2010 16:25:22 +0800 From: Yang Shi User-Agent: Thunderbird 2.0.0.23 (X11/20090817) MIME-Version: 1.0 To: Florian Fainelli CC: ddaney@caviumnetworks.com, ralf@linux-mips.org, linux-mips@linux-mips.org Subject: Re: [PATCH 3/3] MIPS: Octeon: Add add_wired_entry decralation in header file References: <1267601172-17919-1-git-send-email-yang.shi@windriver.com> <3118b3d0f3ed042df1ee2771325c3824e6fc7ba9.1267600234.git.yang.shi@windriver.com> <9e4e80f8edd43f8a164fe618c978c1dc8cd48a69.1267600234.git.yang.shi@windriver.com> <201003030919.36006.florian@openwrt.org> In-Reply-To: <201003030919.36006.florian@openwrt.org> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 8bit X-OriginalArrivalTime: 03 Mar 2010 08:25:25.0349 (UTC) FILETIME=[12F1C950:01CABAAB] Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26095 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: yang.shi@windriver.com Precedence: bulk X-list: linux-mips Florian Fainelli 写é“: > Hi Yang, > > On Wednesday 03 March 2010 08:26:12 Yang Shi wrote: > >> Octeon's setup.c uses add_wired_entry, but it is not declared >> anywhere. Copy add_wired_entry decralation fomr pgtable-32.h to >> pgtable-64.h and include asm/pgtable.h into Octeon's setup.c. >> >> Signed-off-by: Yang Shi >> --- >> arch/mips/cavium-octeon/setup.c | 1 + >> arch/mips/include/asm/pgtable-64.h | 6 ++++++ >> 2 files changed, 7 insertions(+), 0 deletions(-) >> >> diff --git a/arch/mips/cavium-octeon/setup.c >> b/arch/mips/cavium-octeon/setup.c index 8309d68..f35ba16 100644 >> --- a/arch/mips/cavium-octeon/setup.c >> +++ b/arch/mips/cavium-octeon/setup.c >> @@ -30,6 +30,7 @@ >> #include >> #include >> #include >> +#include >> > > You probably meant to include instead. > Yes, thanks to point out this. V2 will be sent soon. Regards, Yang > >> #include >> >> diff --git a/arch/mips/include/asm/pgtable-64.h >> b/arch/mips/include/asm/pgtable-64.h index 26dc69d..85ee34d 100644 >> --- a/arch/mips/include/asm/pgtable-64.h >> +++ b/arch/mips/include/asm/pgtable-64.h >> @@ -23,6 +23,12 @@ >> #endif >> >> /* >> + * - add_wired_entry() add a fixed TLB entry, and move wired register >> + */ >> +extern void add_wired_entry(unsigned long entrylo0, unsigned long >> entrylo1, + unsigned long entryhi, unsigned long pagemask); >> + >> +/* >> * Each address space has 2 4K pages as its page directory, giving 1024 >> * (== PTRS_PER_PGD) 8 byte pointers to pmd tables. Each pmd table is a >> * single 4K page, giving 512 (== PTRS_PER_PMD) 8 byte pointers to page >> >> > > From yang.shi@windriver.com Wed Mar 3 09:43:35 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 03 Mar 2010 09:43:38 +0100 (CET) Received: from mail.windriver.com ([147.11.1.11]:47803 "EHLO mail.windriver.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491022Ab0CCIne (ORCPT ); Wed, 3 Mar 2010 09:43:34 +0100 Received: from localhost.localdomain (pek-lpgbuild1.wrs.com [128.224.153.29]) by mail.windriver.com (8.14.3/8.14.3) with ESMTP id o238hMYA014446; Wed, 3 Mar 2010 00:43:23 -0800 (PST) From: Yang Shi To: ddaney@caviumnetworks.com, ralf@linux-mips.org, f.fainelli@gmail.com Cc: linux-mips@linux-mips.org Subject: [V2] Octeon: Several misc fixes Date: Wed, 3 Mar 2010 16:43:18 +0800 Message-Id: <1267605801-5305-1-git-send-email-yang.shi@windriver.com> X-Mailer: git-send-email 1.6.0.4 Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26096 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: yang.shi@windriver.com Precedence: bulk X-list: linux-mips Hi David & Ralf, These are some misc build fixes for Octeon. Fixed two spelling errors against the first version. Yang Shi (3): MIPS: Octeon: Remove superfluous on_each_cpu parameter MIPS: Octeon: Remove redundant declaration of octeon_reserve32_memory MIPS: Octeon: Add add_wired_entry decralation in header file Thanks, Yang From yang.shi@windriver.com Wed Mar 3 09:43:57 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 03 Mar 2010 09:44:02 +0100 (CET) Received: from mail.windriver.com ([147.11.1.11]:47810 "EHLO mail.windriver.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491179Ab0CCInh (ORCPT ); Wed, 3 Mar 2010 09:43:37 +0100 Received: from localhost.localdomain (pek-lpgbuild1.wrs.com [128.224.153.29]) by mail.windriver.com (8.14.3/8.14.3) with ESMTP id o238hMYB014446; Wed, 3 Mar 2010 00:43:27 -0800 (PST) From: Yang Shi To: ddaney@caviumnetworks.com, ralf@linux-mips.org, f.fainelli@gmail.com Cc: linux-mips@linux-mips.org Subject: [PATCH 1/3] MIPS: Octeon: Remove superfluous on_each_cpu parameter Date: Wed, 3 Mar 2010 16:43:19 +0800 Message-Id: X-Mailer: git-send-email 1.6.0.4 In-Reply-To: <1267605801-5305-1-git-send-email-yang.shi@windriver.com> References: <1267605801-5305-1-git-send-email-yang.shi@windriver.com> Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26097 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: yang.shi@windriver.com Precedence: bulk X-list: linux-mips Now, on_each_cpu just need three parameters, but the on_each_cpu still uses four parameters in Octeon's setup.c. So, remove the superfluous parameter. Signed-off-by: Yang Shi --- arch/mips/cavium-octeon/setup.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/arch/mips/cavium-octeon/setup.c b/arch/mips/cavium-octeon/setup.c index b321d3b..4eaa35f 100644 --- a/arch/mips/cavium-octeon/setup.c +++ b/arch/mips/cavium-octeon/setup.c @@ -230,7 +230,7 @@ static void octeon_hal_setup_per_cpu_reserved32(void *unused) void octeon_hal_setup_reserved32(void) { #ifdef CONFIG_CAVIUM_RESERVE32_USE_WIRED_TLB - on_each_cpu(octeon_hal_setup_per_cpu_reserved32, NULL, 0, 1); + on_each_cpu(octeon_hal_setup_per_cpu_reserved32, NULL, 1); #endif } -- 1.6.3.3 From yang.shi@windriver.com Wed Mar 3 09:44:20 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 03 Mar 2010 09:44:25 +0100 (CET) Received: from mail.windriver.com ([147.11.1.11]:47817 "EHLO mail.windriver.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491773Ab0CCIni (ORCPT ); Wed, 3 Mar 2010 09:43:38 +0100 Received: from localhost.localdomain (pek-lpgbuild1.wrs.com [128.224.153.29]) by mail.windriver.com (8.14.3/8.14.3) with ESMTP id o238hMYC014446; Wed, 3 Mar 2010 00:43:29 -0800 (PST) From: Yang Shi To: ddaney@caviumnetworks.com, ralf@linux-mips.org, f.fainelli@gmail.com Cc: linux-mips@linux-mips.org Subject: [PATCH 2/3] MIPS: Octeon: Remove redundant declaration of octeon_reserve32_memory Date: Wed, 3 Mar 2010 16:43:20 +0800 Message-Id: <50e36e8549769a26986f99a23772d23fd039c230.1267604875.git.yang.shi@windriver.com> X-Mailer: git-send-email 1.6.0.4 In-Reply-To: References: <1267605801-5305-1-git-send-email-yang.shi@windriver.com> In-Reply-To: References: Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26098 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: yang.shi@windriver.com Precedence: bulk X-list: linux-mips In Octeon's setup.c, octeon_reserve32_memory is defined, so remove the redundant extern declaration of this variable. Signed-off-by: Yang Shi --- arch/mips/cavium-octeon/setup.c | 3 --- 1 files changed, 0 insertions(+), 3 deletions(-) diff --git a/arch/mips/cavium-octeon/setup.c b/arch/mips/cavium-octeon/setup.c index 4eaa35f..8309d68 100644 --- a/arch/mips/cavium-octeon/setup.c +++ b/arch/mips/cavium-octeon/setup.c @@ -45,9 +45,6 @@ extern struct plat_smp_ops octeon_smp_ops; extern void pci_console_init(const char *arg); #endif -#ifdef CONFIG_CAVIUM_RESERVE32 -extern uint64_t octeon_reserve32_memory; -#endif static unsigned long long MAX_MEMORY = 512ull << 20; struct octeon_boot_descriptor *octeon_boot_desc_ptr; -- 1.6.3.3 From yang.shi@windriver.com Wed Mar 3 09:44:43 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 03 Mar 2010 09:44:48 +0100 (CET) Received: from mail.windriver.com ([147.11.1.11]:47822 "EHLO mail.windriver.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491863Ab0CCInk (ORCPT ); Wed, 3 Mar 2010 09:43:40 +0100 Received: from localhost.localdomain (pek-lpgbuild1.wrs.com [128.224.153.29]) by mail.windriver.com (8.14.3/8.14.3) with ESMTP id o238hMYD014446; Wed, 3 Mar 2010 00:43:31 -0800 (PST) From: Yang Shi To: ddaney@caviumnetworks.com, ralf@linux-mips.org, f.fainelli@gmail.com Cc: linux-mips@linux-mips.org Subject: [PATCH 3/3] MIPS: Octeon: Add add_wired_entry decralation in header file Date: Wed, 3 Mar 2010 16:43:21 +0800 Message-Id: <004eb64c73b3bcec90612663598ada4cf678f236.1267604875.git.yang.shi@windriver.com> X-Mailer: git-send-email 1.6.0.4 In-Reply-To: <50e36e8549769a26986f99a23772d23fd039c230.1267604875.git.yang.shi@windriver.com> References: <1267605801-5305-1-git-send-email-yang.shi@windriver.com> <50e36e8549769a26986f99a23772d23fd039c230.1267604875.git.yang.shi@windriver.com> In-Reply-To: References: Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26099 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: yang.shi@windriver.com Precedence: bulk X-list: linux-mips Octeon's setup.c uses add_wired_entry, but it is not declared anywhere. Copy add_wired_entry decralation from pgtable-32.h to pgtable-64.h and include asm/pgtable.h into Octeon's setup.c. Signed-off-by: Yang Shi --- arch/mips/cavium-octeon/setup.c | 1 + arch/mips/include/asm/pgtable-64.h | 6 ++++++ 2 files changed, 7 insertions(+), 0 deletions(-) diff --git a/arch/mips/cavium-octeon/setup.c b/arch/mips/cavium-octeon/setup.c index 8309d68..d63b8e6 100644 --- a/arch/mips/cavium-octeon/setup.c +++ b/arch/mips/cavium-octeon/setup.c @@ -30,6 +30,7 @@ #include #include #include +#include #include diff --git a/arch/mips/include/asm/pgtable-64.h b/arch/mips/include/asm/pgtable-64.h index 26dc69d..85ee34d 100644 --- a/arch/mips/include/asm/pgtable-64.h +++ b/arch/mips/include/asm/pgtable-64.h @@ -23,6 +23,12 @@ #endif /* + * - add_wired_entry() add a fixed TLB entry, and move wired register + */ +extern void add_wired_entry(unsigned long entrylo0, unsigned long entrylo1, + unsigned long entryhi, unsigned long pagemask); + +/* * Each address space has 2 4K pages as its page directory, giving 1024 * (== PTRS_PER_PGD) 8 byte pointers to pmd tables. Each pmd table is a * single 4K page, giving 512 (== PTRS_PER_PMD) 8 byte pointers to page -- 1.6.3.3 From lists@nabble.com Wed Mar 3 11:11:35 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 03 Mar 2010 11:11:38 +0100 (CET) Received: from kuber.nabble.com ([216.139.236.158]:47126 "EHLO kuber.nabble.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491836Ab0CCKLe convert rfc822-to-8bit (ORCPT ); Wed, 3 Mar 2010 11:11:34 +0100 Received: from isper.nabble.com ([192.168.236.156]) by kuber.nabble.com with esmtp (Exim 4.63) (envelope-from ) id 1NmlYB-0000tD-V1 for linux-mips@linux-mips.org; Wed, 03 Mar 2010 02:11:31 -0800 Message-ID: <27766331.post@talk.nabble.com> Date: Wed, 3 Mar 2010 02:11:31 -0800 (PST) From: Dea_RU To: linux-mips@linux-mips.org Subject: TI AR7 7200 - no boot MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8BIT X-Nabble-From: dryukovz@mail.ru Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26100 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: dryukovz@mail.ru Precedence: bulk X-list: linux-mips I build kernel 2.6.33 for TI AR7 cpu 7200 version. boot halt on console init stage/ gcc 4.3.3 ============================== linwork:/opt/mipsel32/usr/bin# ./mipsel-linux-gcc-4.3.3 -v Using built-in specs. Target: mipsel-uclibc-linux Configured with: /Acorp/buildroot-2010.02-rc1/output/toolchain/gcc-4.3.3/configure --prefix=/usr --build=i386-cross-linux-uclibc --host=i386-cross-linux-uclibc --target=mipsel-uclibc-linux --enable-languages=c,c++ --with-sysroot=/opt/mipsel32/ --with-build-time-tools=/opt/mipsel32//usr/mipsel-uclibc-linux/bin --disable-__cxa_atexit --enable-target-optspace --with-gnu-ld --disable-libssp --enable-tls --enable-shared --with-gmp=/Acorp/buildroot-2010.02-rc1/output/toolchain/gmp --with-mpfr=/Acorp/buildroot-2010.02-rc1/output/toolchain/mpfr --disable-nls --enable-threads --disable-multilib --disable-decimal-float --with-abi=32 --with-tune=mips1 --disable-largefile --with-endian=little,big --with-pkgversion='Buildroot 2010.02-rc1' --with-bugurl=http://bugs.buildroot.net/ : (reconfigured) /Acorp/buildroot-2010.02-rc1/output/toolchain/gcc-4.3.3/configure --prefix=/usr --build=i386-cross-linux-uclibc --host=i386-cross-linux-uclibc --target=mipsel-uclibc-linux --enable-languages=c,c++ --with-sysroot=/opt/mipsel32/ --with-build-time-tools=/opt/mipsel32//usr/mipsel-uclibc-linux/bin --disable-__cxa_atexit --enable-target-optspace --with-gnu-ld --disable-libssp --enable-tls --enable-shared --with-gmp=/Acorp/buildroot-2010.02-rc1/output/toolchain/gmp --with-mpfr=/Acorp/buildroot-2010.02-rc1/output/toolchain/mpfr --disable-nls --enable-threads --disable-multilib --disable-decimal-float --with-abi=32 --with-tune=mips1 --disable-largefile --with-pkgversion='Buildroot 2010.02-rc1' --with-bugurl=http://bugs.buildroot.net/ Thread model: posix gcc version 4.3.3 (Buildroot 2010.02-rc1) =========== kernel cfg CONFIG_MIPS=y CONFIG_AR7=y CONFIG_LOONGSON_UART_BASE=y CONFIG_RWSEM_GENERIC_SPINLOCK=y CONFIG_ARCH_SUPPORTS_OPROFILE=y CONFIG_GENERIC_FIND_NEXT_BIT=y CONFIG_GENERIC_HWEIGHT=y CONFIG_GENERIC_CALIBRATE_DELAY=y CONFIG_GENERIC_CLOCKEVENTS=y CONFIG_GENERIC_TIME=y CONFIG_GENERIC_CMOS_UPDATE=y CONFIG_SCHED_OMIT_FRAME_POINTER=y CONFIG_GENERIC_HARDIRQS_NO__DO_IRQ=y CONFIG_BOOT_RAW=y CONFIG_CEVT_R4K_LIB=y CONFIG_CEVT_R4K=y CONFIG_CSRC_R4K_LIB=y CONFIG_CSRC_R4K=y CONFIG_DMA_NONCOHERENT=y CONFIG_DMA_NEED_PCI_MAP_STATE=y CONFIG_SYS_HAS_EARLY_PRINTK=y CONFIG_GENERIC_GPIO=y CONFIG_CPU_LITTLE_ENDIAN=y CONFIG_SYS_SUPPORTS_LITTLE_ENDIAN=y CONFIG_IRQ_CPU=y CONFIG_NO_EXCEPT_FILL=y CONFIG_SWAP_IO_SPACE=y CONFIG_MIPS_L1_CACHE_SHIFT=5 CONFIG_CPU_MIPS32_R1=y CONFIG_SYS_SUPPORTS_ZBOOT=y CONFIG_SYS_SUPPORTS_ZBOOT_UART16550=y CONFIG_SYS_HAS_CPU_MIPS32_R1=y CONFIG_CPU_MIPS32=y CONFIG_CPU_MIPSR1=y CONFIG_SYS_SUPPORTS_32BIT_KERNEL=y CONFIG_CPU_SUPPORTS_32BIT_KERNEL=y CONFIG_HARDWARE_WATCHPOINTS=y CONFIG_32BIT=y CONFIG_PAGE_SIZE_4KB=y CONFIG_CPU_HAS_PREFETCH=y CONFIG_MIPS_MT_DISABLED=y CONFIG_CPU_HAS_SYNC=y CONFIG_GENERIC_HARDIRQS=y CONFIG_GENERIC_IRQ_PROBE=y CONFIG_CPU_SUPPORTS_HIGHMEM=y CONFIG_ARCH_FLATMEM_ENABLE=y CONFIG_ARCH_POPULATES_NODE_MAP=y CONFIG_SELECT_MEMORY_MODEL=y CONFIG_FLATMEM_MANUAL=y CONFIG_FLATMEM=y CONFIG_FLAT_NODE_MEM_MAP=y CONFIG_PAGEFLAGS_EXTENDED=y CONFIG_SPLIT_PTLOCK_CPUS=4 CONFIG_ZONE_DMA_FLAG=0 CONFIG_VIRT_TO_BUS=y CONFIG_DEFAULT_MMAP_MIN_ADDR=4096 CONFIG_GENERIC_CLOCKEVENTS_BUILD=y CONFIG_HZ_250=y CONFIG_SYS_SUPPORTS_ARBIT_HZ=y CONFIG_HZ=250 CONFIG_PREEMPT_NONE=y CONFIG_SECCOMP=y CONFIG_LOCKDEP_SUPPORT=y CONFIG_STACKTRACE_SUPPORT=y CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config" CONFIG_CONSTRUCTORS=y # General setup CONFIG_EXPERIMENTAL=y CONFIG_BROKEN_ON_SMP=y CONFIG_INIT_ENV_ARG_LIMIT=32 CONFIG_LOCALVERSION="" CONFIG_HAVE_KERNEL_GZIP=y CONFIG_HAVE_KERNEL_BZIP2=y CONFIG_HAVE_KERNEL_LZMA=y CONFIG_HAVE_KERNEL_LZO=y CONFIG_KERNEL_LZMA=y CONFIG_SYSVIPC=y CONFIG_SYSVIPC_SYSCTL=y # RCU Subsystem CONFIG_TINY_RCU=y CONFIG_LOG_BUF_SHIFT=17 CONFIG_GROUP_SCHED=y CONFIG_FAIR_GROUP_SCHED=y CONFIG_USER_SCHED=y CONFIG_CC_OPTIMIZE_FOR_SIZE=y CONFIG_SYSCTL=y CONFIG_ANON_INODES=y CONFIG_EMBEDDED=y CONFIG_SYSCTL_SYSCALL=y CONFIG_KALLSYMS=y CONFIG_HOTPLUG=y CONFIG_PRINTK=y CONFIG_BASE_FULL=y CONFIG_FUTEX=y CONFIG_EPOLL=y CONFIG_SIGNALFD=y CONFIG_TIMERFD=y CONFIG_EVENTFD=y CONFIG_SHMEM=y CONFIG_AIO=y # Kernel Performance Events And Counters CONFIG_SLAB=y CONFIG_HAVE_OPROFILE=y # GCOV-based kernel profiling CONFIG_SLOW_WORK=y CONFIG_HAVE_GENERIC_DMA_COHERENT=y CONFIG_SLABINFO=y CONFIG_RT_MUTEXES=y CONFIG_BASE_SMALL=0 CONFIG_MODULES=y CONFIG_MODULE_FORCE_LOAD=y CONFIG_INLINE_SPIN_UNLOCK=y CONFIG_INLINE_SPIN_UNLOCK_IRQ=y CONFIG_INLINE_READ_UNLOCK=y CONFIG_INLINE_READ_UNLOCK_IRQ=y CONFIG_INLINE_WRITE_UNLOCK=y CONFIG_INLINE_WRITE_UNLOCK_IRQ=y # Bus options (PCI, PCMCIA, EISA, ISA, TC) CONFIG_MMU=y # Executable file formats CONFIG_BINFMT_ELF=y CONFIG_BINFMT_MISC=y CONFIG_TRAD_SIGNALS=y # Power management options CONFIG_ARCH_HIBERNATION_POSSIBLE=y CONFIG_ARCH_SUSPEND_POSSIBLE=y # Device Drivers # Generic Driver Options CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug" CONFIG_MTD=y CONFIG_MTD_DEBUG=y CONFIG_MTD_DEBUG_VERBOSE=0 # CONFIG_MTD_TESTS is not set # CONFIG_MTD_CONCAT is not set CONFIG_MTD_PARTITIONS=y CONFIG_MTD_AR7_PARTS=y # RAM/ROM/Flash chip drivers CONFIG_MTD_MAP_BANK_WIDTH_1=y CONFIG_MTD_MAP_BANK_WIDTH_2=y CONFIG_MTD_MAP_BANK_WIDTH_4=y CONFIG_MTD_CFI_I1=y CONFIG_MTD_CFI_I2=y # Input device support CONFIG_INPUT=y # Character devices CONFIG_VT=y CONFIG_HW_CONSOLE=y # Serial drivers CONFIG_SERIAL_8250=y CONFIG_SERIAL_8250_CONSOLE=y CONFIG_SERIAL_8250_NR_UARTS=2 CONFIG_SERIAL_8250_RUNTIME_UARTS=2 # Non-8250 serial port support CONFIG_SERIAL_CORE=y CONFIG_SERIAL_CORE_CONSOLE=y CONFIG_UNIX98_PTYS=y CONFIG_DEVPTS_MULTIPLE_INSTANCES=y CONFIG_LEGACY_PTYS=y CONFIG_LEGACY_PTY_COUNT=256 # PPS support CONFIG_PPS=y CONFIG_PPS_DEBUG=y CONFIG_WATCHDOG=y CONFIG_WATCHDOG_NOWAYOUT=y # Watchdog Device Drivers CONFIG_AR7_WDT=y CONFIG_SSB_POSSIBLE=y # Console display driver support CONFIG_DUMMY_CONSOLE=y CONFIG_ACCESSIBILITY=y CONFIG_RTC_LIB=y CONFIG_RTC_CLASS=y CONFIG_RTC_DEBUG=y # RTC interfaces CONFIG_RTC_INTF_SYSFS=y CONFIG_RTC_INTF_PROC=y CONFIG_RTC_INTF_DEV=y CONFIG_RTC_INTF_DEV_UIE_EMUL=y CONFIG_RTC_DRV_TEST=y # on-CPU RTC drivers CONFIG_AUXDISPLAY=y CONFIG_UIO=y # TI VLYNQ CONFIG_VLYNQ=y CONFIG_STAGING=y CONFIG_STAGING_EXCLUDE_BUILD=y # File systems CONFIG_FILE_LOCKING=y CONFIG_FSNOTIFY=y CONFIG_DNOTIFY=y CONFIG_INOTIFY=y CONFIG_INOTIFY_USER=y CONFIG_AUTOFS_FS=y CONFIG_AUTOFS4_FS=y # Caches CONFIG_FSCACHE=y CONFIG_FSCACHE_DEBUG=y # Pseudo filesystems CONFIG_PROC_FS=y CONFIG_PROC_SYSCTL=y CONFIG_PROC_PAGE_MONITOR=y CONFIG_SYSFS=y # Kernel hacking CONFIG_TRACE_IRQFLAGS_SUPPORT=y CONFIG_ENABLE_WARN_DEPRECATED=y CONFIG_ENABLE_MUST_CHECK=y CONFIG_FRAME_WARN=1024 CONFIG_UNUSED_SYMBOLS=y CONFIG_DEBUG_MEMORY_INIT=y CONFIG_HAVE_FUNCTION_TRACER=y CONFIG_HAVE_FUNCTION_GRAPH_TRACER=y CONFIG_HAVE_FUNCTION_TRACE_MCOUNT_TEST=y CONFIG_HAVE_DYNAMIC_FTRACE=y CONFIG_HAVE_FTRACE_MCOUNT_RECORD=y CONFIG_TRACING_SUPPORT=y CONFIG_HAVE_ARCH_KGDB=y CONFIG_EARLY_PRINTK=y CONFIG_CMDLINE_BOOL=y CONFIG_CMDLINE="earlyprintk=ttyS0,9600n,keep" # Security options CONFIG_DEFAULT_SECURITY_DAC=y CONFIG_DEFAULT_SECURITY="" CONFIG_CRYPTO=y # Crypto core or helper CONFIG_CRYPTO_ALGAPI=y CONFIG_CRYPTO_ALGAPI2=y CONFIG_CRYPTO_AEAD2=y CONFIG_CRYPTO_BLKCIPHER=y CONFIG_CRYPTO_BLKCIPHER2=y CONFIG_CRYPTO_HASH=y CONFIG_CRYPTO_HASH2=y CONFIG_CRYPTO_RNG2=y CONFIG_CRYPTO_PCOMP=y CONFIG_CRYPTO_MANAGER=y CONFIG_CRYPTO_MANAGER2=y CONFIG_CRYPTO_NULL=y CONFIG_CRYPTO_WORKQUEUE=y # Hash modes CONFIG_CRYPTO_HMAC=y # Digest CONFIG_CRYPTO_CRC32C=y CONFIG_CRYPTO_MD4=y CONFIG_CRYPTO_MD5=y CONFIG_CRYPTO_SHA1=y # Ciphers CONFIG_CRYPTO_AES=y # Compression CONFIG_CRYPTO_DEFLATE=y CONFIG_CRYPTO_ZLIB=y CONFIG_CRYPTO_LZO=y # Library routines CONFIG_BITREVERSE=y CONFIG_GENERIC_FIND_LAST_BIT=y CONFIG_CRC_CCITT=y CONFIG_CRC16=y CONFIG_CRC_T10DIF=y CONFIG_CRC_ITU_T=y CONFIG_CRC32=y CONFIG_CRC7=y CONFIG_LIBCRC32C=y CONFIG_ZLIB_INFLATE=y CONFIG_ZLIB_DEFLATE=y CONFIG_LZO_COMPRESS=y CONFIG_LZO_DECOMPRESS=y CONFIG_HAS_IOMEM=y CONFIG_HAS_IOPORT=y CONFIG_HAS_DMA=y CONFIG_NLATTR=y =========== booting process: =========== (psbl) Booting... Linux version 2.6.33 (root@linwork) (gcc version 4.3.3 (Buildroot 2010.02-rc1) ) #1 Wed Mar 3 15:31:18 OMST 2010 bootconsole [early0] enabled CPU revision is: 00018448 (MIPS 4KEc) TI AR7 (TNETD7200), ID: 0x002b, Revision: 0x11 Determined physical RAM map: memory: 01000000 @ 14000000 (usable) Zone PFN ranges: Normal 0x00014000 -> 0x00015000 Movable zone start PFN for each node early_node_map[1] active PFN ranges 0: 0x00014000 -> 0x00015000 Built 1 zonelists in Zone order, mobility grouping off. Total pages: 4064 Kernel command line: console=ttyS0,9600n8 earlyprintk=ttyS0,9600 PID hash table entries: 64 (order: -4, 256 bytes) Dentry cache hash table entries: 2048 (order: 1, 8192 bytes) Inode-cache hash table entries: 1024 (order: 0, 4096 bytes) Primary instruction cache 16kB, VIPT, 4-way, linesize 16 bytes. Primary data cache 8kB, 4-way, VIPT, no aliases, linesize 16 bytes Memory: 13164k/16384k available (1325k kernel code, 3220k reserved, 408k data, 120k init, 0k highmem) NR_IRQS:256 Console: colour dummy device 80x25 ============== stop booting... what wrong ???? kernels from open-wrt booting with log: ============================== (psbl) Booting... Linux version 2.6.26.8 (agb@arrakis) (gcc version 4.1.2) #1 Wed Dec 2 15:14:35 UTC 2009 console [early0] enabled CPU revision is: 00018448 (MIPS 4KEc) Clocks: Sync 2:1 mode Clocks: Setting CPU clock Adjusted requested frequency 211000000 to 211968000 Clocks: base = 35328000, frequency = 211968000, prediv = 1, postdiv = 1, postdiv2 = -1, mul = 6 Clocks: Setting DSP clock Clocks: base = 25000000, frequency = 105984000, prediv = 1, postdiv = 2, postdiv2 = 1, mul = 10 Clocks: Setting USB clock Adjusted requested frequency 48000000 to 47863741 Clocks: base = 105984000, frequency = 48000000, prediv = 1, postdiv = 31, postdiv2 = -1, mul = 14 TI AR7 (TNETD7200), ID: 0x002b, Revision: 0x11 Determined physical RAM map: memory: 01000000 @ 14000000 (usable) Initrd not found or empty - disabling initrd Zone PFN ranges: Normal 81920 -> 86016 Movable zone start PFN for each node early_node_map[1] active PFN ranges 0: 81920 -> 86016 Built 1 zonelists in Zone order, mobility grouping off. Total pages: 4064 Kernel command line: init=/etc/preinit rootfstype=squashfs,jffs2, console=ttyS0,9600n8 Primary instruction cache 16kB, VIPT, 4-way, linesize 16 bytes. Primary data cache 8kB, 4-way, VIPT, no aliases, linesize 16 bytes PID hash table entries: 64 (order: 6, 256 bytes) Dentry cache hash table entries: 2048 (order: 1, 8192 bytes) Inode-cache hash table entries: 1024 (order: 0, 4096 bytes) Memory: 12560k/16384k available (2001k kernel code, 3824k reserved, 417k data, 124k init, 0k highmem) SLUB: Genslabs=6, HWalign=32, Order=0-3, MinObjects=0, CPUs=1, Nodes=1 Mount-cache hash table entries: 512 net_namespace: 644 bytes NET: Registered protocol family 16 NET: Registered protocol family 2 IP route cache hash table entries: 1024 (order: 0, 4096 bytes) TCP established hash table entries: 512 (order: 0, 4096 bytes) TCP bind hash table entries: 512 (order: -1, 2048 bytes) TCP: Hash tables configured (established 512 bind 512) TCP reno registered NET: Registered protocol family 1 squashfs: version 3.0 (2006/03/15) Phillip Lougher Registering mini_fo version $Id$ JFFS2 version 2.2. (NAND) (SUMMARY) © 2001-2006 Red Hat, Inc. msgmni has been set to 24 io scheduler noop registered io scheduler deadline registered (default) Serial: 8250/16550 driver $Revision: 1.90 $ 2 ports, IRQ sharing disabled serial8250: ttyS0 at MMIO 0x8610e00 (irq = 15) is a TI-AR7 -- View this message in context: http://old.nabble.com/TI-AR7-7200---no-boot-tp27766331p27766331.html Sent from the linux-mips main mailing list archive at Nabble.com. From f.fainelli@gmail.com Wed Mar 3 11:25:11 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 03 Mar 2010 11:25:14 +0100 (CET) Received: from mail-ew0-f223.google.com ([209.85.219.223]:35170 "EHLO mail-ew0-f223.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491048Ab0CCKZL convert rfc822-to-8bit (ORCPT ); Wed, 3 Mar 2010 11:25:11 +0100 Received: by ewy23 with SMTP id 23so745755ewy.24 for ; Wed, 03 Mar 2010 02:25:05 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:sender:from:organization:to :subject:date:user-agent:cc:references:in-reply-to:mime-version :content-type:content-transfer-encoding:message-id; bh=iNRHM8LCHKiOX3VyIVr1pxcx6UvEXFhdnBSF4BjBNvc=; b=HTAGd7LZlBn2fW//2f39MbzK8VV7oyYYkrjqNx3JaG5qFXuUH9lq73VCI7m1nIXNm8 BAEe4AqQ2SK9yzxXc9rivdvoR/HUMtHr2sGqVQ4ipBufRJwCVME1QMJr2lj0M/7W++ma g1dhVl8G1sBZBqvxIPMRZKKTWNY6RU/Q2KrHk= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=sender:from:organization:to:subject:date:user-agent:cc:references :in-reply-to:mime-version:content-type:content-transfer-encoding :message-id; b=DPXPvQUxJrwVXf1WCiqry0/9GZuH620zL0xJ0cZOtZQut67kvbl9PQmbkd8HsAzVb4 SO/UuYE4DOG5EG0yDF8LOIjYWSeVyhnejiRzBkEtZ0fj4V6b1KBLAm22rD+LJYQyPNR3 Oe0ygitq1Suwy80ALNnfx9lKoP0CvopXTHg2M= Received: by 10.213.104.95 with SMTP id n31mr125720ebo.27.1267611905257; Wed, 03 Mar 2010 02:25:05 -0800 (PST) Received: from flexo.localnet (bobafett.staff.proxad.net [213.228.1.121]) by mx.google.com with ESMTPS id 24sm1785410eyx.6.2010.03.03.02.25.02 (version=SSLv3 cipher=RC4-MD5); Wed, 03 Mar 2010 02:25:02 -0800 (PST) From: Florian Fainelli Organization: OpenWrt To: "Dea_RU" Subject: Re: TI AR7 7200 - no boot Date: Wed, 3 Mar 2010 11:24:45 +0100 User-Agent: KMail/1.12.2 (Linux/2.6.31-17-server; KDE/4.3.2; x86_64; ; ) Cc: linux-mips@linux-mips.org References: <27766331.post@talk.nabble.com> In-Reply-To: <27766331.post@talk.nabble.com> MIME-Version: 1.0 Content-Type: Text/Plain; charset="utf-8" Content-Transfer-Encoding: 8BIT Message-Id: <201003031124.46336.florian@openwrt.org> Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26101 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: florian@openwrt.org Precedence: bulk X-list: linux-mips Hi, On Wednesday 03 March 2010 11:11:31 Dea_RU wrote: > I build kernel 2.6.33 for TI AR7 cpu 7200 version. > > boot halt on console init stage/ [snip] > > CONFIG_MIPS=y > > CONFIG_AR7=y > CONFIG_LOONGSON_UART_BASE=y > CONFIG_RWSEM_GENERIC_SPINLOCK=y > CONFIG_ARCH_SUPPORTS_OPROFILE=y > CONFIG_GENERIC_FIND_NEXT_BIT=y > CONFIG_GENERIC_HWEIGHT=y > CONFIG_GENERIC_CALIBRATE_DELAY=y > CONFIG_GENERIC_CLOCKEVENTS=y > CONFIG_GENERIC_TIME=y > CONFIG_GENERIC_CMOS_UPDATE=y > CONFIG_SCHED_OMIT_FRAME_POINTER=y > CONFIG_GENERIC_HARDIRQS_NO__DO_IRQ=y > CONFIG_BOOT_RAW=y [snip] > > what wrong ???? > > kernels from open-wrt booting with log: > ============================== > (psbl) > > Booting... > Linux version 2.6.26.8 (agb@arrakis) (gcc version 4.1.2) #1 Wed Dec 2 > 15:14:35 UTC 2009 > console [early0] enabled > CPU revision is: 00018448 (MIPS 4KEc) > Clocks: Sync 2:1 mode > Clocks: Setting CPU clock > Adjusted requested frequency 211000000 to 211968000 > Clocks: base = 35328000, frequency = 211968000, prediv = 1, postdiv = 1, > postdiv2 = -1, mul = 6 > Clocks: Setting DSP clock > Clocks: base = 25000000, frequency = 105984000, prediv = 1, postdiv = 2, > postdiv2 = 1, mul = 10 > Clocks: Setting USB clock > Adjusted requested frequency 48000000 to 47863741 > Clocks: base = 105984000, frequency = 48000000, prediv = 1, postdiv = 31, > postdiv2 = -1, mul = 14 > TI AR7 (TNETD7200), ID: 0x002b, Revision: 0x11 > Determined physical RAM map: > memory: 01000000 @ 14000000 (usable) > Initrd not found or empty - disabling initrd > Zone PFN ranges: > Normal 81920 -> 86016 > Movable zone start PFN for each node > early_node_map[1] active PFN ranges > 0: 81920 -> 86016 > Built 1 zonelists in Zone order, mobility grouping off. Total pages: 4064 > Kernel command line: init=/etc/preinit rootfstype=squashfs,jffs2, > console=ttyS0,9600n8 > Primary instruction cache 16kB, VIPT, 4-way, linesize 16 bytes. > Primary data cache 8kB, 4-way, VIPT, no aliases, linesize 16 bytes > PID hash table entries: 64 (order: 6, 256 bytes) > Dentry cache hash table entries: 2048 (order: 1, 8192 bytes) > Inode-cache hash table entries: 1024 (order: 0, 4096 bytes) > Memory: 12560k/16384k available (2001k kernel code, 3824k reserved, 417k > data, 124k init, 0k highmem) > SLUB: Genslabs=6, HWalign=32, Order=0-3, MinObjects=0, CPUs=1, Nodes=1 > Mount-cache hash table entries: 512 > net_namespace: 644 bytes > NET: Registered protocol family 16 > NET: Registered protocol family 2 > IP route cache hash table entries: 1024 (order: 0, 4096 bytes) > TCP established hash table entries: 512 (order: 0, 4096 bytes) > TCP bind hash table entries: 512 (order: -1, 2048 bytes) > TCP: Hash tables configured (established 512 bind 512) > TCP reno registered > NET: Registered protocol family 1 > squashfs: version 3.0 (2006/03/15) Phillip Lougher > Registering mini_fo version $Id$ > JFFS2 version 2.2. (NAND) (SUMMARY) © 2001-2006 Red Hat, Inc. > msgmni has been set to 24 > io scheduler noop registered > io scheduler deadline registered (default) > Serial: 8250/16550 driver $Revision: 1.90 $ 2 ports, IRQ sharing disabled > serial8250: ttyS0 at MMIO 0x8610e00 (irq = 15) is a TI-AR7 > You have one of these devices with bogus UART, (there are quite some out there) can you try with the following patch and tell me if that gives you an usable console: diff --git a/arch/mips/ar7/platform.c b/arch/mips/ar7/platform.c index 246df7a..15cbeeb 100644 --- a/arch/mips/ar7/platform.c +++ b/arch/mips/ar7/platform.c @@ -551,6 +551,7 @@ static int __init ar7_register_uarts(void) uart_port.irq = AR7_IRQ_UART0; uart_port.mapbase = AR7_REGS_UART0; uart_port.membase = ioremap(uart_port.mapbase, 256); + uart_port.flags = UPF_IOREMAP | UPF_BOOT_AUTOCONF; res = early_serial_setup(&uart_port); if (res) @@ -562,6 +563,7 @@ static int __init ar7_register_uarts(void) uart_port.irq = AR7_IRQ_UART1; uart_port.mapbase = UR8_REGS_UART1; uart_port.membase = ioremap(uart_port.mapbase, 256); + uart_port.flags = UPF_IOREMAP | UPF_BOOT_AUTOCONF; res = early_serial_setup(&uart_port); if (res) Thanks. -- Regards, Florian From lists@nabble.com Wed Mar 3 11:59:51 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 03 Mar 2010 11:59:54 +0100 (CET) Received: from kuber.nabble.com ([216.139.236.158]:45768 "EHLO kuber.nabble.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1490990Ab0CCK7v (ORCPT ); Wed, 3 Mar 2010 11:59:51 +0100 Received: from isper.nabble.com ([192.168.236.156]) by kuber.nabble.com with esmtp (Exim 4.63) (envelope-from ) id 1NmmIv-0003xf-Gn for linux-mips@linux-mips.org; Wed, 03 Mar 2010 02:59:49 -0800 Message-ID: <27766728.post@talk.nabble.com> Date: Wed, 3 Mar 2010 02:59:49 -0800 (PST) From: Dea_RU To: linux-mips@linux-mips.org Subject: Re: TI AR7 7200 - no boot In-Reply-To: <201003031124.46336.florian@openwrt.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Nabble-From: dryukovz@mail.ru References: <27766331.post@talk.nabble.com> <201003031124.46336.florian@openwrt.org> Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26102 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: dryukovz@mail.ru Precedence: bulk X-list: linux-mips Florian Fainelli-4 wrote: > > Hi, > > You have one of these devices with bogus UART, (there are quite some out > there) > can you try with the following patch and tell me if that gives you an > usable console: > > diff --git a/arch/mips/ar7/platform.c b/arch/mips/ar7/platform.c > index 246df7a..15cbeeb 100644 > --- a/arch/mips/ar7/platform.c > +++ b/arch/mips/ar7/platform.c > @@ -551,6 +551,7 @@ static int __init ar7_register_uarts(void) > uart_port.irq = AR7_IRQ_UART0; > uart_port.mapbase = AR7_REGS_UART0; > uart_port.membase = ioremap(uart_port.mapbase, 256); > + uart_port.flags = UPF_IOREMAP | UPF_BOOT_AUTOCONF; > > res = early_serial_setup(&uart_port); > if (res) > @@ -562,6 +563,7 @@ static int __init ar7_register_uarts(void) > uart_port.irq = AR7_IRQ_UART1; > uart_port.mapbase = UR8_REGS_UART1; > uart_port.membase = ioremap(uart_port.mapbase, 256); > + uart_port.flags = UPF_IOREMAP | UPF_BOOT_AUTOCONF; > > res = early_serial_setup(&uart_port); > if (res) > > Thanks. > -- > Regards, Florian > after remark printk with message console color line 2927 /drivers/char/vt.c boot process stop with: ==================== .... Primary data cache 8kB, 4-way, VIPT, no aliases, linesize 16 bytes Memory: 13172k/16384k available (1314k kernel code, 3212k reserved, 411k data, 120k init, 0k highmem) NR_IRQS:256 Calibrating delay loop... =============== next freeze on delay loop clculation !? IRQ not init correctly ??? maby after send byte over USART cpu set flag "USART TX buffer cleared"? and kernel go to interput vector ??? ========= in kernel src 2.6.33 from mips.org patch based code not present ! /arch/mips/ar7/platform.c i added port number flags for version 2.6.33 .... uart_port[0].flags = UPF_IOREMAP | UPF_BOOT_AUTOCONF; uart_port[1].flags = UPF_IOREMAP | UPF_BOOT_AUTOCONF; ============== after added flags and remark console collor booting stop with: ...... Primary data cache 8kB, 4-way, VIPT, no aliases, linesize 16 bytes Memory: 13172k/16384k available (1314k kernel code, 3212k reserved, 411k data, 120k init, 0k highmem) NR_IRQS:256 Calibrating delay loop... -- View this message in context: http://old.nabble.com/TI-AR7-7200---no-boot-tp27766331p27766728.html Sent from the linux-mips main mailing list archive at Nabble.com. From muvarov@gmail.com Wed Mar 3 12:05:39 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 03 Mar 2010 12:05:46 +0100 (CET) Received: from mail-bw0-f226.google.com ([209.85.218.226]:60457 "EHLO mail-bw0-f226.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1490990Ab0CCLFj (ORCPT ); Wed, 3 Mar 2010 12:05:39 +0100 Received: by bwz26 with SMTP id 26so1188708bwz.27 for ; Wed, 03 Mar 2010 03:05:31 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:to:cc:from:subject:date :message-id:user-agent:mime-version:content-type :content-transfer-encoding; bh=1nXrqKmR9J2bgDxLGEfEErXxraaXlYosngJRbGtdYXs=; b=FQeuz4iwXwH8gvwki/KzrKynhqxs4pKRaQ+BL6CGJvRFUhQGPUGUdPzArBAY0aJFEm t2kBz9miywoXrTVwomkqjUj+gWuhWCcMnfn3zcicJhhwTjZYY/bU7ZVbrDV9qosNXao9 XIMXWi9zbt9qI+sOdikJMSbJUE4tskOETGgdo= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=to:cc:from:subject:date:message-id:user-agent:mime-version :content-type:content-transfer-encoding; b=oPht3QpLNdWtRbl2d+v8OrZVtH0hBhvBp1UkKTVFK7nvHp4XK/GWABWeEEzTkE6Z45 D4LPSc34nR+zMgHGzjbyt2mr6KzkfHWBKMZSdcJf5pRbhAqpiCKkG4f3FPZfeyg4iCH6 YdNU7lUGBQhPM/hwWeTi0Ct/nTkJeiV5OuHV8= Received: by 10.204.20.137 with SMTP id f9mr5952648bkb.130.1267614331024; Wed, 03 Mar 2010 03:05:31 -0800 (PST) Received: from ?127.0.1.1? (mail.dev.rtsoft.ru [213.79.90.226]) by mx.google.com with ESMTPS id 14sm3569954bwz.10.2010.03.03.03.05.29 (version=TLSv1/SSLv3 cipher=RC4-MD5); Wed, 03 Mar 2010 03:05:30 -0800 (PST) To: linux-mips@linux-mips.org, kexec@lists.infradead.org Cc: horms@verge.net.au, ralf@linux-mips.org From: Maxim Uvarov Subject: [PATCH 1/2] MIPS kexec,kdump support Date: Wed, 03 Mar 2010 14:05:27 +0300 Message-ID: <20100303110527.11233.20400.stgit@muvarov> User-Agent: StGIT/0.14.2 MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26103 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: muvarov@gmail.com Precedence: bulk X-list: linux-mips Hello folks, Please find here MIPS crash and kdump patches. This is patch set of 3 patches: 1. generic MIPS changes (kernel); 2. MIPS Cavium Octeon board kexec/kdump code (kernel); 3. Kexec user space MIPS changes. Patches were tested on the latest linux-mips@ git kernel and the latest kexec-tools git on Cavium Octeon 50xx board. I also made the same code working on RMI XLR/XLS boards for both mips32 and mips64 kernels. Best regards, Maxim Uvarov. --- arch/mips/Kconfig | 24 ++++++++++ arch/mips/Makefile | 4 ++ arch/mips/include/asm/kexec.h | 31 +++++++++++-- arch/mips/include/asm/smp.h | 7 +++ arch/mips/kernel/Makefile | 3 + arch/mips/kernel/crash.c | 75 +++++++++++++++++++++++++++++++ arch/mips/kernel/crash_dump.c | 86 +++++++++++++++++++++++++++++++++++ arch/mips/kernel/machine_kexec.c | 30 +++++++++++- arch/mips/kernel/relocate_kernel.S | 88 ++++++++++++++++++++++++++++++++++++ arch/mips/kernel/setup.c | 55 +++++++++++++++++++++++ arch/mips/kernel/smp.c | 18 +++++++ 11 files changed, 413 insertions(+), 8 deletions(-) create mode 100644 arch/mips/kernel/crash.c create mode 100644 arch/mips/kernel/crash_dump.c diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig index 591ca0c..8079260 100644 --- a/arch/mips/Kconfig +++ b/arch/mips/Kconfig @@ -2035,6 +2035,30 @@ config KEXEC support. As of this writing the exact hardware interface is strongly in flux, so no good recommendation can be made. +config CRASH_DUMP + bool "kernel crash dumps (EXPERIMENTAL)" + depends on EXPERIMENTAL + help + Generate crash dump after being started by kexec. + This should be normally only set in special crash dump kernels + which are loaded in the main kernel with kexec-tools into + a specially reserved region and then later executed after + a crash by kdump/kexec. The crash dump kernel must be compiled + to a memory address not used by the main kernel or BIOS using + PHYSICAL_START. + +config PHYSICAL_START + hex "Physical address where the kernel is loaded" + default "0xffffffff84000000" if 64BIT + default "0x84000000" if 32BIT + depends on CRASH_DUMP + help + This gives the CKSEG0 or KSEG0 address where the kernel is loaded. + If you plan to use kernel for capturing the crash dump change + this value to start of the reserved region (the "X" value as + specified in the "crashkernel=YM@XM" command line boot parameter + passed to the panic-ed kernel). + config SECCOMP bool "Enable seccomp to safely compute untrusted bytecode" depends on PROC_FS diff --git a/arch/mips/Makefile b/arch/mips/Makefile index 2f2eac2..431283d 100644 --- a/arch/mips/Makefile +++ b/arch/mips/Makefile @@ -646,6 +646,10 @@ else load-$(CONFIG_CPU_CAVIUM_OCTEON) += 0xffffffff81100000 endif +ifdef CONFIG_PHYSICAL_START +load-y = $(CONFIG_PHYSICAL_START) +endif + cflags-y += -I$(srctree)/arch/mips/include/asm/mach-generic drivers-$(CONFIG_PCI) += arch/mips/pci/ diff --git a/arch/mips/include/asm/kexec.h b/arch/mips/include/asm/kexec.h index 4314892..e9eba1a 100644 --- a/arch/mips/include/asm/kexec.h +++ b/arch/mips/include/asm/kexec.h @@ -9,22 +9,45 @@ #ifndef _MIPS_KEXEC # define _MIPS_KEXEC +#include + +extern unsigned long long elfcorehdr_addr; + /* Maximum physical address we can use pages from */ #define KEXEC_SOURCE_MEMORY_LIMIT (0x20000000) /* Maximum address we can reach in physical address mode */ #define KEXEC_DESTINATION_MEMORY_LIMIT (0x20000000) /* Maximum address we can use for the control code buffer */ #define KEXEC_CONTROL_MEMORY_LIMIT (0x20000000) - -#define KEXEC_CONTROL_PAGE_SIZE 4096 +/* Reserve 3*4096 bytes for board-specific info */ +#define KEXEC_CONTROL_PAGE_SIZE (4096 + 3*4096) /* The native architecture */ #define KEXEC_ARCH KEXEC_ARCH_MIPS +#define MAX_NOTE_BYTES 1024 static inline void crash_setup_regs(struct pt_regs *newregs, - struct pt_regs *oldregs) + struct pt_regs *oldregs) { - /* Dummy implementation for now */ + if (oldregs) + memcpy(newregs, oldregs, sizeof(*newregs)); + else + prepare_frametrace(newregs); } +#ifdef CONFIG_KEXEC +struct kimage; +extern unsigned long kexec_args[4]; +extern int (*_machine_kexec_prepare)(struct kimage *); +extern void (*_machine_kexec_shutdown)(void); +extern void (*_machine_crash_shutdown)(struct pt_regs *regs); +extern void default_machine_crash_shutdown(struct pt_regs *regs); +#ifdef CONFIG_SMP +extern const unsigned char kexec_smp_wait[]; +extern unsigned long secondary_kexec_args[4]; +extern void (*relocated_kexec_smp_wait) (void *); +extern atomic_t kexec_ready_to_reboot; +#endif +#endif + #endif /* !_MIPS_KEXEC */ diff --git a/arch/mips/include/asm/smp.h b/arch/mips/include/asm/smp.h index af42385..9b50048 100644 --- a/arch/mips/include/asm/smp.h +++ b/arch/mips/include/asm/smp.h @@ -40,6 +40,8 @@ extern int __cpu_logical_map[NR_CPUS]; #define SMP_CALL_FUNCTION 0x2 /* Octeon - Tell another core to flush its icache */ #define SMP_ICACHE_FLUSH 0x4 +/* Used by kexec crashdump to save all cpu's state */ +#define SMP_DUMP 0x8 extern volatile cpumask_t cpu_callin_map; @@ -91,4 +93,9 @@ static inline void arch_send_call_function_ipi_mask(const struct cpumask *mask) mp_ops->send_ipi_mask(mask, SMP_CALL_FUNCTION); } +extern void core_send_ipi(int cpu, unsigned int action); +#if defined(CONFIG_KEXEC) +extern void (*dump_ipi_function_ptr)(void *); +void dump_send_ipi(void (*dump_ipi_callback)(void *)); +#endif #endif /* __ASM_SMP_H */ diff --git a/arch/mips/kernel/Makefile b/arch/mips/kernel/Makefile index ef20957..7ae634d 100644 --- a/arch/mips/kernel/Makefile +++ b/arch/mips/kernel/Makefile @@ -91,7 +91,8 @@ obj-$(CONFIG_I8253) += i8253.o obj-$(CONFIG_GPIO_TXX9) += gpio_txx9.o -obj-$(CONFIG_KEXEC) += machine_kexec.o relocate_kernel.o +obj-$(CONFIG_KEXEC) += machine_kexec.o relocate_kernel.o crash.o +obj-$(CONFIG_CRASH_DUMP) += crash_dump.o obj-$(CONFIG_EARLY_PRINTK) += early_printk.o obj-$(CONFIG_SPINLOCK_TEST) += spinlock_test.o diff --git a/arch/mips/kernel/crash.c b/arch/mips/kernel/crash.c new file mode 100644 index 0000000..24247ea --- /dev/null +++ b/arch/mips/kernel/crash.c @@ -0,0 +1,75 @@ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#ifdef CONFIG_CRASH_DUMP +unsigned long long elfcorehdr_addr = ELFCORE_ADDR_MAX; +#endif + +/* This keeps a track of which one is crashing cpu. */ +int crashing_cpu = -1; +static cpumask_t cpus_in_crash = CPU_MASK_NONE; + +#ifdef CONFIG_SMP +void crash_shutdown_secondary(void *ignore) +{ + struct pt_regs *regs; + int cpu = smp_processor_id(); + + regs = task_pt_regs(current); + + if (!cpu_online(cpu)) + return; + + local_irq_disable(); + if (!cpu_isset(cpu, cpus_in_crash)) + crash_save_cpu(regs, cpu); + cpu_set(cpu, cpus_in_crash); + + while (!atomic_read(&kexec_ready_to_reboot)) + cpu_relax(); + relocated_kexec_smp_wait(NULL); + /* NOTREACHED */ +} + +static void crash_kexec_prepare_cpus(void) +{ + unsigned int msecs; + + unsigned int ncpus = num_online_cpus() - 1;/* Excluding the panic cpu */ + + dump_send_ipi(crash_shutdown_secondary); + smp_wmb(); + + /* + * The crash CPU sends an IPI and wait for other CPUs to + * respond. Delay of at least 10 seconds. + */ + printk(KERN_EMERG "Sending IPI to other cpus...\n"); + msecs = 10000; + while ((cpus_weight(cpus_in_crash) < ncpus) && (--msecs > 0)) { + cpu_relax(); + mdelay(1); + } +} + +#else +static void crash_kexec_prepare_cpus() {} +#endif + +void default_machine_crash_shutdown(struct pt_regs *regs) +{ + local_irq_disable(); + crashing_cpu = smp_processor_id(); + crash_save_cpu(regs, crashing_cpu); + crash_kexec_prepare_cpus(); + cpu_set(crashing_cpu, cpus_in_crash); +} diff --git a/arch/mips/kernel/crash_dump.c b/arch/mips/kernel/crash_dump.c new file mode 100644 index 0000000..49e5efa --- /dev/null +++ b/arch/mips/kernel/crash_dump.c @@ -0,0 +1,86 @@ +#include +#include +#include +#include + +#ifdef CONFIG_PROC_VMCORE +static int __init parse_elfcorehdr(char *p) +{ + if (p) + elfcorehdr_addr = memparse(p, &p); + return 1; +} +__setup("elfcorehdr=", parse_elfcorehdr); +#endif + +static int __init parse_savemaxmem(char *p) +{ + if (p) + saved_max_pfn = (memparse(p, &p) >> PAGE_SHIFT) - 1; + + return 1; +} +__setup("savemaxmem=", parse_savemaxmem); + + +static void *kdump_buf_page; + +/** + * copy_oldmem_page - copy one page from "oldmem" + * @pfn: page frame number to be copied + * @buf: target memory address for the copy; this can be in kernel address + * space or user address space (see @userbuf) + * @csize: number of bytes to copy + * @offset: offset in bytes into the page (based on pfn) to begin the copy + * @userbuf: if set, @buf is in user address space, use copy_to_user(), + * otherwise @buf is in kernel address space, use memcpy(). + * + * Copy a page from "oldmem". For this page, there is no pte mapped + * in the current kernel. + * + * Calling copy_to_user() in atomic context is not desirable. Hence first + * copying the data to a pre-allocated kernel page and then copying to user + * space in non-atomic context. + */ +ssize_t copy_oldmem_page(unsigned long pfn, char *buf, + size_t csize, unsigned long offset, int userbuf) +{ + void *vaddr; + + if (!csize) + return 0; + + vaddr = kmap_atomic_pfn(pfn, KM_PTE0); + + if (!userbuf) { + memcpy(buf, (vaddr + offset), csize); + kunmap_atomic(vaddr, KM_PTE0); + } else { + if (!kdump_buf_page) { + printk(KERN_WARNING "Kdump: Kdump buffer page not" + " allocated\n"); + return -EFAULT; + } + copy_page(kdump_buf_page, vaddr); + kunmap_atomic(vaddr, KM_PTE0); + if (copy_to_user(buf, (kdump_buf_page + offset), csize)) + return -EFAULT; + } + + return csize; +} + +static int __init kdump_buf_page_init(void) +{ + int ret = 0; + + kdump_buf_page = kmalloc(PAGE_SIZE, GFP_KERNEL); + if (!kdump_buf_page) { + printk(KERN_WARNING "Kdump: Failed to allocate kdump buffer" + " page\n"); + ret = -ENOMEM; + } + + return ret; +} +arch_initcall(kdump_buf_page_init); diff --git a/arch/mips/kernel/machine_kexec.c b/arch/mips/kernel/machine_kexec.c index 85beb9b..4d25843 100644 --- a/arch/mips/kernel/machine_kexec.c +++ b/arch/mips/kernel/machine_kexec.c @@ -19,9 +19,19 @@ extern const size_t relocate_new_kernel_size; extern unsigned long kexec_start_address; extern unsigned long kexec_indirection_page; +int (*_machine_kexec_prepare)(struct kimage *) = NULL; +void (*_machine_kexec_shutdown)(void) = NULL; +void (*_machine_crash_shutdown)(struct pt_regs *regs) = NULL; +#ifdef CONFIG_SMP +void (*relocated_kexec_smp_wait) (void *); +atomic_t kexec_ready_to_reboot = ATOMIC_INIT(0); +#endif + int machine_kexec_prepare(struct kimage *kimage) { + if (_machine_kexec_prepare) + return _machine_kexec_prepare(kimage); return 0; } @@ -33,11 +43,17 @@ machine_kexec_cleanup(struct kimage *kimage) void machine_shutdown(void) { + if (_machine_kexec_shutdown) + _machine_kexec_shutdown(); } void machine_crash_shutdown(struct pt_regs *regs) { + if (_machine_crash_shutdown) + _machine_crash_shutdown(regs); + else + default_machine_crash_shutdown(regs); } typedef void (*noretfun_t)(void) __attribute__((noreturn)); @@ -52,7 +68,9 @@ machine_kexec(struct kimage *image) reboot_code_buffer = (unsigned long)page_address(image->control_code_page); - kexec_start_address = image->start; + kexec_start_address = + (unsigned long) phys_to_virt(image->start); + kexec_indirection_page = (unsigned long) phys_to_virt(image->head & PAGE_MASK); @@ -63,7 +81,7 @@ machine_kexec(struct kimage *image) * The generic kexec code builds a page list with physical * addresses. they are directly accessible through KSEG0 (or * CKSEG0 or XPHYS if on 64bit system), hence the - * pys_to_virt() call. + * phys_to_virt() call. */ for (ptr = &image->head; (entry = *ptr) && !(entry &IND_DONE); ptr = (entry & IND_INDIRECTION) ? @@ -81,5 +99,13 @@ machine_kexec(struct kimage *image) printk("Will call new kernel at %08lx\n", image->start); printk("Bye ...\n"); __flush_cache_all(); +#ifdef CONFIG_SMP + /* All secondary cpus now may jump to kexec_wait cycle */ + relocated_kexec_smp_wait = reboot_code_buffer + + (void *)(kexec_smp_wait - relocate_new_kernel); + smp_wmb(); + atomic_set(&kexec_ready_to_reboot, 1); +#endif ((noretfun_t) reboot_code_buffer)(); } + diff --git a/arch/mips/kernel/relocate_kernel.S b/arch/mips/kernel/relocate_kernel.S index 87481f9..0abaf7a 100644 --- a/arch/mips/kernel/relocate_kernel.S +++ b/arch/mips/kernel/relocate_kernel.S @@ -15,6 +15,11 @@ #include LEAF(relocate_new_kernel) + PTR_L a0, arg0 + PTR_L a1, arg1 + PTR_L a2, arg2 + PTR_L a3, arg3 + PTR_L s0, kexec_indirection_page PTR_L s1, kexec_start_address @@ -26,7 +31,6 @@ process_entry: and s3, s2, 0x1 beq s3, zero, 1f and s4, s2, ~0x1 /* store destination addr in s4 */ - move a0, s4 b process_entry 1: @@ -60,10 +64,92 @@ copy_word: b process_entry done: +#ifdef CONFIG_SMP + /* kexec_flag reset is signal to other CPUs what kernel + was moved to it's location. Note - we need relocated address + of kexec_flag. */ + + bal 1f + 1: move t1,ra; + PTR_LA t2,1b + PTR_LA t0,kexec_flag + PTR_SUB t0,t0,t2; + PTR_ADD t0,t1,t0; + LONG_S zero,(t0) +#endif + + sync /* jump to kexec_start_address */ j s1 END(relocate_new_kernel) +#ifdef CONFIG_SMP +/* + * Other CPUs should wait until code is relocated and + * then start at entry (?) point. + */ +LEAF(kexec_smp_wait) + PTR_L a0, s_arg0 + PTR_L a1, s_arg1 + PTR_L a2, s_arg2 + PTR_L a3, s_arg3 + PTR_L s1, kexec_start_address + + /* Non-relocated address works for args and kexec_start_address ( old + * kernel is not overwritten). But we need relocated address of + * kexec_flag. + */ + + bal 1f +1: move t1,ra; + PTR_LA t2,1b + PTR_LA t0,kexec_flag + PTR_SUB t0,t0,t2; + PTR_ADD t0,t1,t0; + +1: LONG_L s0, (t0) + bne s0, zero,1b + + sync + j s1 + END(kexec_smp_wait) +#endif + +#ifdef __mips64 + /* all PTR's must be aligned to 8 byte in 64-bit mode */ + .align 3 +#endif + +/* All parameters to new kernel are passed in registers a0-a3. + * kexec_args[0..3] are uses to prepare register values. + */ + +kexec_args: + EXPORT(kexec_args) +arg0: PTR 0x0 +arg1: PTR 0x0 +arg2: PTR 0x0 +arg3: PTR 0x0 + .size kexec_args,PTRSIZE*4 + +#ifdef CONFIG_SMP +/* + * Secondary CPUs may have different kernel parameters in + * their registers a0-a3. secondary_kexec_args[0..3] are used + * to prepare register values. + */ +secondary_kexec_args: + EXPORT(secondary_kexec_args) +s_arg0: PTR 0x0 +s_arg1: PTR 0x0 +s_arg2: PTR 0x0 +s_arg3: PTR 0x0 + .size secondary_kexec_args,PTRSIZE*4 +kexec_flag: + LONG 0x1 + +#endif + kexec_start_address: EXPORT(kexec_start_address) PTR 0x0 diff --git a/arch/mips/kernel/setup.c b/arch/mips/kernel/setup.c index f9513f9..d706bd9 100644 --- a/arch/mips/kernel/setup.c +++ b/arch/mips/kernel/setup.c @@ -21,6 +21,7 @@ #include #include #include +#include #include #include @@ -487,10 +488,61 @@ static void __init arch_mem_init(char **cmdline_p) } bootmem_init(); +#ifdef CONFIG_KEXEC + if (crashk_res.start != crashk_res.end) + reserve_bootmem(crashk_res.start, + crashk_res.end - crashk_res.start + 1, + BOOTMEM_DEFAULT); +#endif sparse_init(); paging_init(); } +#ifdef CONFIG_KEXEC +static inline unsigned long long get_total_mem(void) +{ + unsigned long long total; + + total = max_pfn - min_low_pfn; + return total << PAGE_SHIFT; +} + +static void __init mips_parse_crashkernel(void) +{ + unsigned long long total_mem; + unsigned long long crash_size, crash_base; + int ret; + + total_mem = get_total_mem(); + ret = parse_crashkernel(boot_command_line, total_mem, + &crash_size, &crash_base); + if (ret != 0 || crash_size <= 0) + return; + + crashk_res.start = crash_base; + crashk_res.end = crash_base + crash_size - 1; +} +static void __init request_crashkernel(struct resource *res) +{ + int ret; + + ret = request_resource(res, &crashk_res); + if (!ret) + printk(KERN_INFO "Reserving %ldMB of memory at %ldMB " + "for crashkernel\n", + (unsigned long)((crashk_res.end - + crashk_res.start + 1) >> 20), + (unsigned long)(crashk_res.start >> 20)); +} +#else +static void __init mips_parse_crashkernel(void) +{ +} +static void __init request_crashkernel(struct resource *res) +{ +} +#endif + static void __init resource_init(void) { int i; @@ -506,6 +558,8 @@ static void __init resource_init(void) /* * Request address space for all standard RAM. */ + mips_parse_crashkernel(); + for (i = 0; i < boot_mem_map.nr_map; i++) { struct resource *res; unsigned long start, end; @@ -541,6 +595,7 @@ static void __init resource_init(void) */ request_resource(res, &code_resource); request_resource(res, &data_resource); + request_crashkernel(res); } } diff --git a/arch/mips/kernel/smp.c b/arch/mips/kernel/smp.c index 6cdca19..e2f4d53 100644 --- a/arch/mips/kernel/smp.c +++ b/arch/mips/kernel/smp.c @@ -402,3 +402,21 @@ void flush_tlb_one(unsigned long vaddr) EXPORT_SYMBOL(flush_tlb_page); EXPORT_SYMBOL(flush_tlb_one); + +#if defined(CONFIG_KEXEC) +void (*dump_ipi_function_ptr)(void *) = NULL; +void dump_send_ipi(void (*dump_ipi_callback)(void *)) +{ + int i; + int cpu = smp_processor_id(); + + dump_ipi_function_ptr = dump_ipi_callback; + smp_mb(); + for_each_online_cpu(i) + if (i != cpu) + core_send_ipi(i, SMP_DUMP); + +} +EXPORT_SYMBOL(dump_send_ipi); +#endif + Signed-off-by: Maxim Uvarov From muvarov@gmail.com Wed Mar 3 12:06:04 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 03 Mar 2010 12:06:12 +0100 (CET) Received: from mail-bw0-f226.google.com ([209.85.218.226]:60457 "EHLO mail-bw0-f226.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491120Ab0CCLFj (ORCPT ); Wed, 3 Mar 2010 12:05:39 +0100 Received: by mail-bw0-f226.google.com with SMTP id 26so1188708bwz.27 for ; Wed, 03 Mar 2010 03:05:39 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:to:cc:from:subject:date :message-id:in-reply-to:references:user-agent:mime-version :content-type:content-transfer-encoding; bh=3aF9N4T+CRJwoloTH9XjL7mXI+oRP4EmAh/Xcf/x6Hs=; b=a6qhc1auwRtYV0US9OrMYGXzrSEKi8AClKM9OmWsBPOw5ZXCPkeBsHZHhq/DX5Kly7 PRjb9faIvPR4oeX32DsM5EDDmWPDL78W8x/h8H+qFsKb3nCNadESVLSLUCDJj6VtVR36 ADNm7YkaxiLXnOBVFnVuBSTOXrqdmQKF2ssPs= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=to:cc:from:subject:date:message-id:in-reply-to:references :user-agent:mime-version:content-type:content-transfer-encoding; b=NK3gWSjM+HndhsZ1ZMF9D9qfKkeoBlQ/j2T4vH50T/63eX/9KAis4MWVdiwGm7W3hM YkM5JnwXH3IEL5lpWEsGnbYTpPpb2PyRjlR0Uyy/0VfBx6yeAw9Gjocgh86vX9rAwM0h HUPwY3fYEeFEwiIT8S4v38PewOkxPOmaPdKr4= Received: by 10.204.25.194 with SMTP id a2mr5826561bkc.191.1267614339283; Wed, 03 Mar 2010 03:05:39 -0800 (PST) Received: from ?127.0.1.1? (mail.dev.rtsoft.ru [213.79.90.226]) by mx.google.com with ESMTPS id 14sm3626984bwz.14.2010.03.03.03.05.37 (version=TLSv1/SSLv3 cipher=RC4-MD5); Wed, 03 Mar 2010 03:05:38 -0800 (PST) To: linux-mips@linux-mips.org, kexec@lists.infradead.org Cc: horms@verge.net.au, ralf@linux-mips.org From: Maxim Uvarov Subject: [PATCH 2/2] MIPS Cavium Octeon board kexec,kdump support Date: Wed, 03 Mar 2010 14:05:36 +0300 Message-ID: <20100303110536.11233.99481.stgit@muvarov> In-Reply-To: <20100303110527.11233.20400.stgit@muvarov> References: <20100303110527.11233.20400.stgit@muvarov> User-Agent: StGIT/0.14.2 MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26104 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: muvarov@gmail.com Precedence: bulk X-list: linux-mips Hello folks, Please find here MIPS crash and kdump patches. This is patch set of 3 patches: 1. generic MIPS changes (kernel); 2. MIPS Cavium Octeon board kexec/kdump code (kernel); 3. Kexec user space MIPS changes. Patches were tested on the latest linux-mips@ git kernel and the latest kexec-tools git on Cavium Octeon 50xx board. I also made the same code working on RMI XLR/XLS boards for both mips32 and mips64 kernels. Best regards, Maxim Uvarov. --- arch/mips/cavium-octeon/dma-octeon.c | 6 arch/mips/cavium-octeon/executive/cvmx-bootmem.c | 5 arch/mips/cavium-octeon/setup.c | 309 +++++++++++++++++++++- arch/mips/cavium-octeon/smp.c | 5 arch/mips/include/asm/octeon/cvmx-bootmem.h | 2 arch/mips/kernel/relocate_kernel.S | 19 + 6 files changed, 326 insertions(+), 20 deletions(-) diff --git a/arch/mips/cavium-octeon/dma-octeon.c b/arch/mips/cavium-octeon/dma-octeon.c index be531ec..679d7b5 100644 --- a/arch/mips/cavium-octeon/dma-octeon.c +++ b/arch/mips/cavium-octeon/dma-octeon.c @@ -105,7 +105,8 @@ dma_addr_t octeon_map_dma_mem(struct device *dev, void *ptr, size_t size) pr_warning("dma_map_single: Warning: " "Mapping memory address that might " "conflict with devices 0x%llx-0x%llx\n", - physical, physical+size-1); + (unsigned long long) physical, + (unsigned long long) (physical + size - 1)); /* The 2nd 256MB is mapped at 256<<20 instead of 0x410000000 */ if ((physical >= 0x410000000ull) && physical < 0x420000000ull) result = physical - 0x400000000ull; @@ -115,7 +116,8 @@ dma_addr_t octeon_map_dma_mem(struct device *dev, void *ptr, size_t size) panic("dma_map_single: Attempt to map address " "0x%llx-0x%llx, which can't be accessed " "according to the dma mask 0x%llx\n", - physical, physical+size-1, dma_mask); + physical, (unsigned long long)(physical+size-1), + dma_mask); goto done; case OCTEON_DMA_BAR_TYPE_BIG: diff --git a/arch/mips/cavium-octeon/executive/cvmx-bootmem.c b/arch/mips/cavium-octeon/executive/cvmx-bootmem.c index fdf5f19..1ae35ec 100644 --- a/arch/mips/cavium-octeon/executive/cvmx-bootmem.c +++ b/arch/mips/cavium-octeon/executive/cvmx-bootmem.c @@ -688,3 +688,8 @@ int64_t cvmx_bootmem_phy_named_block_alloc(uint64_t size, uint64_t min_addr, cvmx_spinlock_unlock((cvmx_spinlock_t *)&(cvmx_bootmem_desc->lock)); return addr_allocated; } + +void *cvmx_bootmem_get_desc(void) +{ + return cvmx_bootmem_desc; +} diff --git a/arch/mips/cavium-octeon/setup.c b/arch/mips/cavium-octeon/setup.c index b321d3b..2f246d2 100644 --- a/arch/mips/cavium-octeon/setup.c +++ b/arch/mips/cavium-octeon/setup.c @@ -20,6 +20,7 @@ #include #include #include +#include #include #include @@ -55,11 +56,205 @@ struct octeon_boot_descriptor *octeon_boot_desc_ptr; struct cvmx_bootinfo *octeon_bootinfo; EXPORT_SYMBOL(octeon_bootinfo); +static unsigned long long RESERVE_LOW_MEM = 0ull; +#ifdef CONFIG_KEXEC +#ifdef CONFIG_SMP +/* + * Wait for relocation code is prepared and send + * secondary CPUs to spin until kernel is relocated. + */ +void octeon_kexec_smp_down(void *ignored) +{ + int cpu = smp_processor_id(); + local_irq_disable(); + cpu_clear(cpu, cpu_online_map); + while (!atomic_read(&kexec_ready_to_reboot)) + cpu_relax(); + asm volatile ("synci 0($0)\n"); + relocated_kexec_smp_wait(NULL); +} +#endif + +#define OCTEON_DDR0_BASE (0x0ULL) +#define OCTEON_DDR0_SIZE (0x010000000ULL) +#define OCTEON_DDR1_BASE (0x410000000ULL) +#define OCTEON_DDR1_SIZE (0x010000000ULL) +#define OCTEON_DDR2_BASE (0x020000000ULL) +#define OCTEON_DDR2_SIZE (0x3e0000000ULL) +#define OCTEON_MAX_PHY_MEM_SIZE (16*1024*1024*1024ULL) + +static struct kimage *kimage_ptr; + +static void kexec_bootmem_init(uint64_t mem_size, uint32_t low_reserved_bytes) +{ + int64_t addr; + struct cvmx_bootmem_desc *bootmem_desc; + + bootmem_desc = + (struct cvmx_bootmem_desc *)cvmx_bootmem_get_desc(); + + if (mem_size > OCTEON_MAX_PHY_MEM_SIZE) { + mem_size = OCTEON_MAX_PHY_MEM_SIZE; + printk(KERN_ERR "ERROR: requested memory size too large," + "truncating to maximum size\n"); + } + + bootmem_desc->major_version = CVMX_BOOTMEM_DESC_MAJ_VER; + bootmem_desc->minor_version = CVMX_BOOTMEM_DESC_MIN_VER; + + addr = (OCTEON_DDR0_BASE + RESERVE_LOW_MEM + low_reserved_bytes); + bootmem_desc->head_addr = 0; + + if (mem_size <= OCTEON_DDR0_SIZE) { + __cvmx_bootmem_phy_free(addr, + mem_size - RESERVE_LOW_MEM - + low_reserved_bytes, 0); + return; + } + + __cvmx_bootmem_phy_free(addr, + OCTEON_DDR0_SIZE - RESERVE_LOW_MEM - + low_reserved_bytes, 0); + + mem_size -= OCTEON_DDR0_SIZE; + + if (mem_size > OCTEON_DDR1_SIZE) { + __cvmx_bootmem_phy_free(OCTEON_DDR1_BASE, OCTEON_DDR1_SIZE, 0); + __cvmx_bootmem_phy_free(OCTEON_DDR2_BASE, + mem_size - OCTEON_DDR1_SIZE, 0); + } else + __cvmx_bootmem_phy_free(OCTEON_DDR1_BASE, mem_size, 0); +} + +int octeon_kexec_prepare(struct kimage *image) +{ + int i; + char *bootloader = "kexec"; + + octeon_boot_desc_ptr->argc = 0; + for (i = 0; i < image->nr_segments; i++) { + if (!strncmp(bootloader, (char *)image->segment[i].buf, + strlen(bootloader))) { + /* + * convert command line string to array + * of parameters (as bootloader does). + */ + int argc = 0, offt; + char *str = (char *)image->segment[i].buf; + char *ptr = strchr(str, ' '); + while (ptr && (OCTEON_ARGV_MAX_ARGS > argc)) { + *ptr = '\0'; + if (ptr[1] != ' ') { + offt = (int)(ptr - str + 1); + octeon_boot_desc_ptr->argv[argc] = + image->segment[i].mem + offt; + argc++; + } + ptr = strchr(ptr + 1, ' '); + } + octeon_boot_desc_ptr->argc = argc; + break; + } + } + + /* + * Information about segments will be needed during pre-boot memory + * initialization. + */ + kimage_ptr = image; + return 0; +} + +static void octeon_generic_shutdown(void) +{ + int cpu, i; + struct cvmx_bootmem_desc *bootmem_desc; + void *named_block_array_ptr; + + bootmem_desc = + (struct cvmx_bootmem_desc *)cvmx_bootmem_get_desc(); + named_block_array_ptr = + cvmx_phys_to_ptr(bootmem_desc->named_block_array_addr); + +#ifdef CONFIG_SMP + /* disable watchdogs */ + for_each_online_cpu(cpu) + cvmx_write_csr(CVMX_CIU_WDOGX(cpu_logical_map(cpu)), 0); +#else + cvmx_write_csr(CVMX_CIU_WDOGX(cvmx_get_core_num()), 0); +#endif + if (kimage_ptr != kexec_crash_image) { + memset(named_block_array_ptr, + 0x0, + CVMX_BOOTMEM_NUM_NAMED_BLOCKS * + sizeof(struct cvmx_bootmem_named_block_desc)); + /* + * Mark all memory (except low 0x100000 bytes) as free. + * It is the same thing that bootloader does. + */ + kexec_bootmem_init(octeon_bootinfo->dram_size*1024ULL*1024ULL, + 0x100000); + /* + * Allocate all segments to avoid their corruption during boot. + */ + for (i = 0; i < kimage_ptr->nr_segments; i++) + cvmx_bootmem_alloc_address( + kimage_ptr->segment[i].memsz + 2*PAGE_SIZE, + kimage_ptr->segment[i].mem - PAGE_SIZE, + PAGE_SIZE); + } else { + /* + * Do not mark all memory as free. Free only named sections + * leaving the rest of memory unchanged. + */ + struct cvmx_bootmem_named_block_desc *ptr = + (struct cvmx_bootmem_named_block_desc *) + named_block_array_ptr; + + for (i = 0; i < bootmem_desc->named_block_num_blocks; i++) + if (ptr[i].size) + cvmx_bootmem_free_named(ptr[i].name); + } + kexec_args[2] = 1UL; /* running on octeon_main_processor */ + kexec_args[3] = (unsigned long)octeon_boot_desc_ptr; +#ifdef CONFIG_SMP + secondary_kexec_args[2] = 0UL; /* running on secondary cpu */ + secondary_kexec_args[3] = (unsigned long)octeon_boot_desc_ptr; +#endif +} + +void octeon_shutdown(void) +{ + octeon_generic_shutdown(); +#ifdef CONFIG_SMP + smp_call_function(octeon_kexec_smp_down, NULL, 0); + smp_wmb(); + while (num_online_cpus() > 1) { + cpu_relax(); + mdelay(1); + } +#endif +} + +void octeon_crash_shutdown(struct pt_regs *regs) +{ + octeon_generic_shutdown(); + default_machine_crash_shutdown(regs); +} + +#endif + #ifdef CONFIG_CAVIUM_RESERVE32 uint64_t octeon_reserve32_memory; EXPORT_SYMBOL(octeon_reserve32_memory); #endif +#ifdef CONFIG_KEXEC +/* crashkernel cmdline parameter is parsed _after_ memory setup + * we also parse it here (workaround for EHB5200) */ +static uint64_t crashk_size, crashk_base; +#endif + static int octeon_uart; extern asmlinkage void handle_int(void); @@ -467,6 +662,8 @@ void __init prom_init(void) int i; int argc; struct uart_port octeon_port; + char *p; + const char *arg; #ifdef CONFIG_CAVIUM_RESERVE32 int64_t addr = -1; #endif @@ -629,6 +826,15 @@ void __init prom_init(void) if (octeon_is_simulation()) MAX_MEMORY = 64ull << 20; + arg = strstr(arcs_cmdline, "mem="); + if (arg) { + MAX_MEMORY = memparse(arg + 4, &p); + if (MAX_MEMORY == 0) + MAX_MEMORY = 32ull << 30; + if (*p == '@') + RESERVE_LOW_MEM = memparse(p + 1, &p); + } + arcs_cmdline[0] = 0; argc = octeon_boot_desc_ptr->argc; for (i = 0; i < argc; i++) { @@ -636,16 +842,25 @@ void __init prom_init(void) cvmx_phys_to_ptr(octeon_boot_desc_ptr->argv[i]); if ((strncmp(arg, "MEM=", 4) == 0) || (strncmp(arg, "mem=", 4) == 0)) { - sscanf(arg + 4, "%llu", &MAX_MEMORY); - MAX_MEMORY <<= 20; + MAX_MEMORY = memparse(arg + 4, &p); if (MAX_MEMORY == 0) MAX_MEMORY = 32ull << 30; + if (*p == '@') + RESERVE_LOW_MEM = memparse(p + 1, &p); } else if (strcmp(arg, "ecc_verbose") == 0) { #ifdef CONFIG_CAVIUM_REPORT_SINGLE_BIT_ECC __cvmx_interrupt_ecc_report_single_bit_errors = 1; pr_notice("Reporting of single bit ECC errors is " "turned on\n"); #endif +#ifdef CONFIG_KEXEC + } else if (strncmp(arg, "crashkernel=", 12) == 0) { + crashk_size = memparse(arg+12, &p); + if (*p == '@') + crashk_base = memparse(p+1, &p); + strcat(arcs_cmdline, " "); + strcat(arcs_cmdline, arg); +#endif } else if (strlen(arcs_cmdline) + strlen(arg) + 1 < sizeof(arcs_cmdline) - 1) { strcat(arcs_cmdline, " "); @@ -685,6 +900,12 @@ void __init prom_init(void) _machine_restart = octeon_restart; _machine_halt = octeon_halt; +#ifdef CONFIG_KEXEC + _machine_kexec_shutdown = octeon_shutdown; + _machine_crash_shutdown = octeon_crash_shutdown; + _machine_kexec_prepare = octeon_kexec_prepare; +#endif + memset(&octeon_port, 0, sizeof(octeon_port)); /* * For early_serial_setup we don't set the port type or @@ -717,17 +938,15 @@ void __init plat_mem_setup(void) { uint64_t mem_alloc_size; uint64_t total; + uint64_t crashk_end; +#ifndef CONFIG_CRASH_DUMP int64_t memory; + uint64_t kernel_start; + uint64_t kernel_size; +#endif total = 0; - - /* First add the init memory we will be returning. */ - memory = __pa_symbol(&__init_begin) & PAGE_MASK; - mem_alloc_size = (__pa_symbol(&__init_end) & PAGE_MASK) - memory; - if (mem_alloc_size > 0) { - add_memory_region(memory, mem_alloc_size, BOOT_MEM_RAM); - total += mem_alloc_size; - } + crashk_end = 0; /* * The Mips memory init uses the first memory location for @@ -740,6 +959,17 @@ void __init plat_mem_setup(void) if (mem_alloc_size > MAX_MEMORY) mem_alloc_size = MAX_MEMORY; +/* Crashkernel ignores bootmem list. It relies on mem=X@Y option */ +#ifdef CONFIG_CRASH_DUMP + add_memory_region(RESERVE_LOW_MEM, MAX_MEMORY, BOOT_MEM_RAM); + total += MAX_MEMORY; +#else +#ifdef CONFIG_KEXEC + if (crashk_size > 0) { + add_memory_region(crashk_base, crashk_size, BOOT_MEM_RAM); + crashk_end = crashk_base + crashk_size; + } +#endif /* * When allocating memory, we want incrementing addresses from * bootmem_alloc so the code in add_memory_region can merge @@ -750,9 +980,10 @@ void __init plat_mem_setup(void) && (total < MAX_MEMORY)) { #if defined(CONFIG_64BIT) || defined(CONFIG_64BIT_PHYS_ADDR) memory = cvmx_bootmem_phy_alloc(mem_alloc_size, - __pa_symbol(&__init_end), -1, - 0x100000, + RESERVE_LOW_MEM, + /*-1*/0, 0x100000, CVMX_BOOTMEM_FLAG_NO_LOCKING); + #elif defined(CONFIG_HIGHMEM) memory = cvmx_bootmem_phy_alloc(mem_alloc_size, 0, 1ull << 31, 0x100000, @@ -763,19 +994,61 @@ void __init plat_mem_setup(void) CVMX_BOOTMEM_FLAG_NO_LOCKING); #endif if (memory >= 0) { - /* - * This function automatically merges address - * regions next to each other if they are - * received in incrementing order. - */ - add_memory_region(memory, mem_alloc_size, BOOT_MEM_RAM); +#ifdef CONFIG_KEXEC + uint64_t end = memory + mem_alloc_size; + /* This function automatically merges address + regions next to each other if they are + received in incrementing order */ + if (memory < crashk_base && end > crashk_end) { + /* region is fully in */ + add_memory_region(memory, + crashk_base - memory, + BOOT_MEM_RAM); + total += crashk_base - memory; + add_memory_region(crashk_end, + end - crashk_end, + BOOT_MEM_RAM); + total += end - crashk_end; + continue; + } + + if (memory >= crashk_base && end <= crashk_end) + /* Entire memory region is within the + * new kernel's memory, ignore it. */ + continue; + + if (memory > crashk_base && memory < crashk_end && + end > crashk_end) { + /* Overlap with the beginning of the + * region, reserve the beginning. */ + mem_alloc_size -= crashk_end - memory; + memory = crashk_end; + } else if (memory < crashk_base && end > crashk_base && + end < crashk_end) + /* Overlap with the beginning of the + * region, chop of end. */ + mem_alloc_size -= end - crashk_base; +#endif + add_memory_region(memory, mem_alloc_size, + BOOT_MEM_RAM); total += mem_alloc_size; + + /* Recovering mem_alloc_size */ + mem_alloc_size = 4 << 20; } else { break; } } cvmx_bootmem_unlock(); + /* Add the memory region for the kernel. */ + kernel_start = (unsigned long) _text; + kernel_size = ALIGN(_end - _text, 0x100000); + + /* Adjust for physical offset. */ + kernel_start &= ~0xffffffff80000000ULL; + add_memory_region(kernel_start, kernel_size, BOOT_MEM_RAM); +#endif /* CONFIG_CRASH_DUMP */ #ifdef CONFIG_CAVIUM_RESERVE32 /* * Now that we've allocated the kernel memory it is safe to diff --git a/arch/mips/cavium-octeon/smp.c b/arch/mips/cavium-octeon/smp.c index 51e9802..06fe291 100644 --- a/arch/mips/cavium-octeon/smp.c +++ b/arch/mips/cavium-octeon/smp.c @@ -65,6 +65,11 @@ void octeon_send_ipi_single(int cpu, unsigned int action) cvmx_write_csr(CVMX_CIU_MBOX_SETX(coreid), action); } +void core_send_ipi(int cpu, unsigned int action) +{ + octeon_send_ipi_single(cpu, action); +} + static inline void octeon_send_ipi_mask(const struct cpumask *mask, unsigned int action) { diff --git a/arch/mips/include/asm/octeon/cvmx-bootmem.h b/arch/mips/include/asm/octeon/cvmx-bootmem.h index 8e708bd..2f3cb9f 100644 --- a/arch/mips/include/asm/octeon/cvmx-bootmem.h +++ b/arch/mips/include/asm/octeon/cvmx-bootmem.h @@ -370,4 +370,6 @@ void cvmx_bootmem_lock(void); */ void cvmx_bootmem_unlock(void); +void *cvmx_bootmem_get_desc(void); + #endif /* __CVMX_BOOTMEM_H__ */ diff --git a/arch/mips/kernel/relocate_kernel.S b/arch/mips/kernel/relocate_kernel.S index 0abaf7a..961c10b 100644 --- a/arch/mips/kernel/relocate_kernel.S +++ b/arch/mips/kernel/relocate_kernel.S @@ -78,7 +78,19 @@ done: LONG_S zero,(t0) #endif +#ifdef CONFIG_CPU_CAVIUM_OCTEON + /* We need to flush I-cache before jumping to new kernel. + * Unfortunatelly, this code is cpu-specific. + */ + .set push + .set noreorder + syncw + syncw + synci 0($0) + .set pop +#else sync +#endif /* jump to kexec_start_address */ j s1 END(relocate_new_kernel) @@ -110,7 +122,14 @@ LEAF(kexec_smp_wait) 1: LONG_L s0, (t0) bne s0, zero,1b +#ifdef CONFIG_CPU_CAVIUM_OCTEON + .set push + .set noreorder + synci 0($0) + .set pop +#else sync +#endif j s1 END(kexec_smp_wait) #endif Signed-off-by: Maxim Uvarov From muvarov@gmail.com Wed Mar 3 12:06:31 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 03 Mar 2010 12:06:38 +0100 (CET) Received: from mail-bw0-f226.google.com ([209.85.218.226]:60457 "EHLO mail-bw0-f226.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491836Ab0CCLF5 (ORCPT ); Wed, 3 Mar 2010 12:05:57 +0100 Received: by mail-bw0-f226.google.com with SMTP id 26so1188708bwz.27 for ; Wed, 03 Mar 2010 03:05:57 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:to:cc:from:subject:date :message-id:user-agent:mime-version:content-type :content-transfer-encoding; bh=m0IyZWZTzercOFRXDrkf1WqgT4+BeLwlSi1bSdHdKI8=; b=riqcc9gK1tN08sEggj2ci68Y8hjBAlEfTESrN70yAVdTKh8mlYinYokcnJ5f6CLg7M sEeOMvh77aSYQpgXQz8MuHqVMX5KAL29YEfuOCwpaeWUDbb6NgzSVeskojVRj2YTRfG8 +J5oQXZa/GJl+PmyIuimhivX+b8/RwupxbC2s= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=to:cc:from:subject:date:message-id:user-agent:mime-version :content-type:content-transfer-encoding; b=sfrIA5QPBooxVfYVQza524hw/PjTrNV5BQ/xOsv4CP/A7HhwqksazbYR048la5ViTm l2ms3VIQhj6Kg7rr9PW5d+9av5MaxCwnBmGV6TtDmFBSbCqy8YzB0EtcrbjugUu6RpnW cwMJV2obyLnbcR2R7Bpw7NbuYq04EpRFTBkJA= Received: by 10.204.11.11 with SMTP id r11mr5965425bkr.12.1267614357078; Wed, 03 Mar 2010 03:05:57 -0800 (PST) Received: from ?127.0.1.1? (mail.dev.rtsoft.ru [213.79.90.226]) by mx.google.com with ESMTPS id 14sm3627184bwz.14.2010.03.03.03.05.55 (version=TLSv1/SSLv3 cipher=RC4-MD5); Wed, 03 Mar 2010 03:05:56 -0800 (PST) To: linux-mips@linux-mips.org, kexec@lists.infradead.org Cc: horms@verge.net.au, ralf@linux-mips.org From: Maxim Uvarov Subject: [PATCH] some kexec MIPS improvements Date: Wed, 03 Mar 2010 14:05:53 +0300 Message-ID: <20100303110553.11274.9038.stgit@muvarov> User-Agent: StGIT/0.14.2 MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26105 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: muvarov@gmail.com Precedence: bulk X-list: linux-mips Hello folks, Please find here MIPS crash and kdump patches. This is patch set of 3 patches: 1. generic MIPS changes (kernel); 2. MIPS Cavium Octeon board kexec/kdump code (kernel); 3. Kexec user space MIPS changes. Patches were tested on the latest linux-mips@ git kernel and the latest kexec-tools git on Cavium Octeon 50xx board. I also made the same code working on RMI XLR/XLS boards for both mips32 and mips64 kernels. Best regards, Maxim Uvarov. From: Maxim Uvarov Changes are: - using simple mips* ) in configure.ac to make it compilable on mips2 and mips64 - remove kexec/arch/mips/mips-setup-simple.S which prepares cmdline for new kernel, it is better to move this work to kernel code. BTW this code was compilable only on o32 because of t4 is not defined on 64-64 or n32 MIPS ABIs. - simple put cmdline as string, kernel code should catch cmdline like this int board_kexec_prepare(struct kimage *image) { int i; char *bootloader = "kexec"; board_boot_desc_ptr->argc = 0; for(i=0;inr_segments;i++) { printk("segment %d if (!strncmp(bootloader, (char*)image->segment[i].buf, strlen(bootloader))) { /* * convert command line string to array * of parameters (as bootloader does). */ int argc = 0, offt; char *str = (char *)image->segment[i].buf; char *ptr = strchr(str, ' '); while (ptr && (ARGV_MAX_ARGS > argc)) { *ptr = '\0'; if (ptr[1] != ' ') { offt = (int)(ptr - str + 1); boot_desc_ptr->argv[argc] = image->segment[i].mem + offt; argc++; } ptr = strchr(ptr + 1, ' '); } boot_desc_ptr->argc = argc; break; } } Keep it as string make code simple and more readable. - add crashdump support - do not redefine syscalls numbers if they defined in system remove fixups for /proc/iomem. If your board provides wrong /proc/iomem please fix kernel, or at least you local version of kexec. No need to support it in main line. At least add option --fake-iomem - some minor fixes Signed-off-by: Maxim Uvarov --- configure.ac | 2 kexec/arch/mips/Makefile | 4 kexec/arch/mips/crashdump-mips.c | 377 +++++++++++++++++++++++++++++++++++ kexec/arch/mips/crashdump-mips.h | 27 +++ kexec/arch/mips/kexec-elf-mips.c | 141 +++++++------ kexec/arch/mips/kexec-mips.c | 51 +---- kexec/arch/mips/kexec-mips.h | 9 + kexec/arch/mips/mips-setup-simple.S | 110 ---------- kexec/crashdump.h | 2 kexec/kexec-syscall.h | 2 10 files changed, 510 insertions(+), 215 deletions(-) create mode 100644 kexec/arch/mips/crashdump-mips.c create mode 100644 kexec/arch/mips/crashdump-mips.h delete mode 100644 kexec/arch/mips/mips-setup-simple.S diff --git a/configure.ac b/configure.ac index a911f8a..88d2e0b 100644 --- a/configure.ac +++ b/configure.ac @@ -39,7 +39,7 @@ case $target_cpu in sh4|sh4a|sh3|sh ) ARCH="sh" ;; - mips|mipsel ) + mips* ) ARCH="mips" ;; cris|crisv32 ) diff --git a/kexec/arch/mips/Makefile b/kexec/arch/mips/Makefile index f94bd89..831b263 100644 --- a/kexec/arch/mips/Makefile +++ b/kexec/arch/mips/Makefile @@ -4,7 +4,7 @@ mips_KEXEC_SRCS = kexec/arch/mips/kexec-mips.c mips_KEXEC_SRCS += kexec/arch/mips/kexec-elf-mips.c mips_KEXEC_SRCS += kexec/arch/mips/kexec-elf-rel-mips.c -mips_KEXEC_SRCS += kexec/arch/mips/mips-setup-simple.S +mips_KEXEC_SRCS += kexec/arch/mips/crashdump-mips.c mips_ADD_BUFFER = mips_ADD_SEGMENT = @@ -12,5 +12,7 @@ mips_VIRT_TO_PHYS = dist += kexec/arch/mips/Makefile $(mips_KEXEC_SRCS) \ kexec/arch/mips/kexec-mips.h \ + kexec/arch/mips/crashdump-mips.h \ kexec/arch/mips/include/arch/options.h +CFLAGS +=-Wall -Werror diff --git a/kexec/arch/mips/crashdump-mips.c b/kexec/arch/mips/crashdump-mips.c new file mode 100644 index 0000000..17cf52d --- /dev/null +++ b/kexec/arch/mips/crashdump-mips.c @@ -0,0 +1,377 @@ +/* + * kexec: Linux boots Linux + * + * 2005 (C) IBM Corporation. + * 2008 (C) MontaVista Software, Inc. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation (version 2 of the License). + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "../../kexec.h" +#include "../../kexec-elf.h" +#include "../../kexec-syscall.h" +#include "../../crashdump.h" +#include "kexec-mips.h" +#include "crashdump-mips.h" +#include "unused.h" + +/* Stores a sorted list of RAM memory ranges for which to create elf headers. + * A separate program header is created for backup region */ +static struct memory_range crash_memory_range[CRASH_MAX_MEMORY_RANGES]; + +/* Memory region reserved for storing panic kernel and other data. */ +static struct memory_range crash_reserved_mem; + +/* + * To store the memory size of the first kernel and this value will be + * passed to the second kernel as command line (savemaxmem=xM). + * The second kernel will be calculated saved_max_pfn based on this + * variable. + */ +unsigned long long saved_max_mem; + +/* Removes crash reserve region from list of memory chunks for whom elf program + * headers have to be created. Assuming crash reserve region to be a single + * continuous area fully contained inside one of the memory chunks */ +static int exclude_crash_reserve_region(int *nr_ranges) +{ + int i, j, tidx = -1; + unsigned long long cstart, cend; + struct memory_range temp_region = { + .start = 0, + .end = 0 + }; + + /* Crash reserved region. */ + cstart = crash_reserved_mem.start; + cend = crash_reserved_mem.end; + + for (i = 0; i < (*nr_ranges); i++) { + unsigned long long mstart, mend; + mstart = crash_memory_range[i].start; + mend = crash_memory_range[i].end; + if (cstart < mend && cend > mstart) { + if (cstart != mstart && cend != mend) { + /* Split memory region */ + crash_memory_range[i].end = cstart - 1; + temp_region.start = cend + 1; + temp_region.end = mend; + temp_region.type = RANGE_RAM; + tidx = i+1; + } else if (cstart != mstart) + crash_memory_range[i].end = cstart - 1; + else + crash_memory_range[i].start = cend + 1; + } + } + /* Insert split memory region, if any. */ + if (tidx >= 0) { + if (*nr_ranges == CRASH_MAX_MEMORY_RANGES) { + /* No space to insert another element. */ + fprintf(stderr, "Error: Number of crash memory ranges" + " excedeed the max limit\n"); + return -1; + } + for (j = (*nr_ranges - 1); j >= tidx; j--) + crash_memory_range[j+1] = crash_memory_range[j]; + crash_memory_range[tidx].start = temp_region.start; + crash_memory_range[tidx].end = temp_region.end; + crash_memory_range[tidx].type = temp_region.type; + (*nr_ranges)++; + } + return 0; +} +/* Reads the appropriate file and retrieves the SYSTEM RAM regions for whom to + * create Elf headers. Keeping it separate from get_memory_ranges() as + * requirements are different in the case of normal kexec and crashdumps. + * + * Normal kexec needs to look at all of available physical memory irrespective + * of the fact how much of it is being used by currently running kernel. + * Crashdumps need to have access to memory regions actually being used by + * running kernel. Expecting a different file/data structure than /proc/iomem + * to look into down the line. May be something like /proc/kernelmem or may + * be zone data structures exported from kernel. + */ +static int get_crash_memory_ranges(struct memory_range **range, int *ranges) +{ + const char iomem[] = "/proc/iomem"; + int i, memory_ranges = 0; + char line[MAX_LINE]; + FILE *fp; + unsigned long long start, end; + + fp = fopen(iomem, "r"); + if (!fp) { + fprintf(stderr, "Cannot open %s: %s\n", + iomem, strerror(errno)); + return -1; + } + /* Separate segment for backup region */ + crash_memory_range[0].start = BACKUP_SRC_START; + crash_memory_range[0].end = BACKUP_SRC_END; + crash_memory_range[0].type = RANGE_RAM; + memory_ranges++; + + while (fgets(line, sizeof(line), fp) != 0) { + char *str; + int type, consumed, count; + if (memory_ranges >= CRASH_MAX_MEMORY_RANGES) + break; + count = sscanf(line, "%Lx-%Lx : %n", + &start, &end, &consumed); + if (count != 2) + continue; + str = line + consumed; + + /* Only Dumping memory of type System RAM. */ + if (memcmp(str, "System RAM\n", 11) == 0) { + type = RANGE_RAM; + } else if (memcmp(str, "Crash kernel\n", 13) == 0) { + /* Reserved memory region. New kernel can + * use this region to boot into. */ + crash_reserved_mem.start = start; + crash_reserved_mem.end = end; + crash_reserved_mem.type = RANGE_RAM; + continue; + } else + continue; + + if (start == BACKUP_SRC_START && end >= (BACKUP_SRC_END + 1)) + start = BACKUP_SRC_END + 1; + + crash_memory_range[memory_ranges].start = start; + crash_memory_range[memory_ranges].end = end; + crash_memory_range[memory_ranges].type = type; + memory_ranges++; + + /* Segregate linearly mapped region. */ + if ((MAXMEM - 1) >= start && (MAXMEM - 1) <= end) { + crash_memory_range[memory_ranges - 1].end = MAXMEM - 1; + + /* Add segregated region. */ + crash_memory_range[memory_ranges].start = MAXMEM; + crash_memory_range[memory_ranges].end = end; + crash_memory_range[memory_ranges].type = type; + memory_ranges++; + } + } + fclose(fp); + + if (exclude_crash_reserve_region(&memory_ranges) < 0) + return -1; + + for (i = 0; i < memory_ranges; i++) + if (saved_max_mem < crash_memory_range[i].end) + saved_max_mem = crash_memory_range[i].end + 1; + + *range = crash_memory_range; + *ranges = memory_ranges; + return 0; +} + +/* Converts unsigned long to ascii string. */ +static void ultoa(unsigned long i, char *str) +{ + int j = 0, k; + char tmp; + + do { + str[j++] = i % 10 + '0'; + } while ((i /= 10) > 0); + str[j] = '\0'; + + /* Reverse the string. */ + for (j = 0, k = strlen(str) - 1; j < k; j++, k--) { + tmp = str[k]; + str[k] = str[j]; + str[j] = tmp; + } +} + +/* Adds the appropriate mem= options to command line, indicating the + * memory region the new kernel can use to boot into. */ +static int cmdline_add_mem(char *cmdline, unsigned long addr, + unsigned long size) +{ + int cmdlen, len; + char str[50], *ptr; + + addr = addr/1024; + size = size/1024; + ptr = str; + strcpy(str, " mem="); + ptr += strlen(str); + ultoa(size, ptr); + strcat(str, "K@"); + ptr = str + strlen(str); + ultoa(addr, ptr); + strcat(str, "K"); + len = strlen(str); + cmdlen = strlen(cmdline) + len; + if (cmdlen > (COMMAND_LINE_SIZE - 1)) + die("Command line overflow\n"); + strcat(cmdline, str); + + return 0; +} + +/* Adds the elfcorehdr= command line parameter to command line. */ +static int cmdline_add_elfcorehdr(char *cmdline, unsigned long addr) +{ + int cmdlen, len, align = 1024; + char str[30], *ptr; + + /* Passing in elfcorehdr=xxxK format. Saves space required in cmdline. + * Ensure 1K alignment*/ + if (addr%align) + return -1; + addr = addr/align; + ptr = str; + strcpy(str, " elfcorehdr="); + ptr += strlen(str); + ultoa(addr, ptr); + strcat(str, "K"); + len = strlen(str); + cmdlen = strlen(cmdline) + len; + if (cmdlen > (COMMAND_LINE_SIZE - 1)) + die("Command line overflow\n"); + strcat(cmdline, str); + return 0; +} + +/* Adds the savemaxmem= command line parameter to command line. */ +static int cmdline_add_savemaxmem(char *cmdline, unsigned long long addr) +{ + int cmdlen, len, align = 1024; + char str[30], *ptr; + + /* Passing in savemaxmem=xxxM format. Saves space required in cmdline.*/ + addr = addr/(align*align); + ptr = str; + strcpy(str, " savemaxmem="); + ptr += strlen(str); + ultoa(addr, ptr); + strcat(str, "M"); + len = strlen(str); + cmdlen = strlen(cmdline) + len; + if (cmdlen > (COMMAND_LINE_SIZE - 1)) + die("Command line overflow\n"); + strcat(cmdline, str); + return 0; +} + +#ifdef __mips64 +static struct crash_elf_info elf_info64 = { + class: ELFCLASS64, + data : ELFDATA2MSB, + machine : EM_MIPS, + backup_src_start : BACKUP_SRC_START, + backup_src_end : BACKUP_SRC_END, + page_offset : PAGE_OFFSET, + lowmem_limit : MAXMEM, +}; +#endif +static struct crash_elf_info elf_info32 = { + class: ELFCLASS32, + data : ELFDATA2MSB, + machine : EM_MIPS, + backup_src_start : BACKUP_SRC_START, + backup_src_end : BACKUP_SRC_END, + page_offset : PAGE_OFFSET, + lowmem_limit : MAXMEM, +}; + +/* Loads additional segments in case of a panic kernel is being loaded. + * One segment for backup region, another segment for storing elf headers + * for crash memory image. + */ +int load_crashdump_segments(struct kexec_info *info, char* mod_cmdline, + unsigned long UNUSED(max_addr), + unsigned long UNUSED(min_base)) +{ + void *tmp; + unsigned long sz, elfcorehdr; + int nr_ranges, align = 1024; + struct memory_range *mem_range; + + if (get_crash_memory_ranges(&mem_range, &nr_ranges) < 0) + return -1; + + /* Create a backup region segment to store backup data*/ + sz = (BACKUP_SRC_SIZE + align - 1) & ~(align - 1); + tmp = xmalloc(sz); + memset(tmp, 0, sz); + info->backup_start = add_buffer(info, tmp, sz, sz, align, + crash_reserved_mem.start, + crash_reserved_mem.end, -1); + +#ifdef __mips64 + /* Create elf header segment and store crash image data. */ + if (arch_options.core_header_type == CORE_TYPE_ELF64) { + if (crash_create_elf64_headers(info, &elf_info64, + crash_memory_range, nr_ranges, + &tmp, &sz, + ELF_CORE_HEADER_ALIGN) < 0) + return -1; + } else { + if (crash_create_elf32_headers(info, &elf_info32, + crash_memory_range, nr_ranges, + &tmp, &sz, + ELF_CORE_HEADER_ALIGN) < 0) + return -1; + } +#else + if (crash_create_elf32_headers(info, &elf_info32, + crash_memory_range, nr_ranges, + &tmp, &sz, + ELF_CORE_HEADER_ALIGN) < 0) + return -1; +#endif + elfcorehdr = add_buffer(info, tmp, sz, sz, align, + crash_reserved_mem.start, + crash_reserved_mem.end, -1); + + /* + * backup segment is after elfcorehdr, so use elfcorehdr as top of + * kernel's available memory + */ + cmdline_add_mem(mod_cmdline, crash_reserved_mem.start, + elfcorehdr - crash_reserved_mem.start); + cmdline_add_elfcorehdr(mod_cmdline, elfcorehdr); + cmdline_add_savemaxmem(mod_cmdline, saved_max_mem); + +#ifdef DEBUG + printf("CRASH MEMORY RANGES:\n"); + printf("%016Lx-%016Lx\n", crash_reserved_mem.start, + crash_reserved_mem.end); +#endif + return 0; +} + +int is_crashkernel_mem_reserved(void) +{ + uint64_t start, end; + + return parse_iomem_single("Crash kernel\n", &start, &end) == 0 ? + (start != end) : 0; +} + diff --git a/kexec/arch/mips/crashdump-mips.h b/kexec/arch/mips/crashdump-mips.h new file mode 100644 index 0000000..c986835 --- /dev/null +++ b/kexec/arch/mips/crashdump-mips.h @@ -0,0 +1,27 @@ +#ifndef CRASHDUMP_MIPS_H +#define CRASHDUMP_MIPS_H + +struct kexec_info; +int load_crashdump_segments(struct kexec_info *info, char *mod_cmdline, + unsigned long max_addr, unsigned long min_base); +#ifdef __mips64 +#define PAGE_OFFSET 0xa800000000000000ULL +#else +#define PAGE_OFFSET 0x80000000 +#endif +#define __pa(x) ((unsigned long)(X) & 0x7fffffff) + +#define MAXMEM 0x80000000 + +#define CRASH_MAX_MEMMAP_NR (KEXEC_MAX_SEGMENTS + 1) +#define CRASH_MAX_MEMORY_RANGES (MAX_MEMORY_RANGES + 2) + +#define COMMAND_LINE_SIZE 512 + +/* Backup Region, First 1M of System RAM. */ +#define BACKUP_SRC_START 0x00000000 +#define BACKUP_SRC_END 0x000fffff +#define BACKUP_SRC_SIZE (BACKUP_SRC_END - BACKUP_SRC_START + 1) + +extern struct arch_options_t arch_options; +#endif /* CRASHDUMP_MIPS_H */ diff --git a/kexec/arch/mips/kexec-elf-mips.c b/kexec/arch/mips/kexec-elf-mips.c index 95695b6..ce6bf0c 100644 --- a/kexec/arch/mips/kexec-elf-mips.c +++ b/kexec/arch/mips/kexec-elf-mips.c @@ -25,55 +25,21 @@ #include #include "../../kexec.h" #include "../../kexec-elf.h" +#include "../../kexec-syscall.h" #include "kexec-mips.h" +#include "crashdump-mips.h" #include static const int probe_debug = 0; #define BOOTLOADER "kexec" #define MAX_COMMAND_LINE 256 - #define UPSZ(X) ((sizeof(X) + 3) & ~3) -static struct boot_notes { - Elf_Bhdr hdr; - Elf_Nhdr bl_hdr; - unsigned char bl_desc[UPSZ(BOOTLOADER)]; - Elf_Nhdr blv_hdr; - unsigned char blv_desc[UPSZ(BOOTLOADER_VERSION)]; - Elf_Nhdr cmd_hdr; - unsigned char command_line[0]; -} elf_boot_notes = { - .hdr = { - .b_signature = 0x0E1FB007, - .b_size = sizeof(elf_boot_notes), - .b_checksum = 0, - .b_records = 3, - }, - .bl_hdr = { - .n_namesz = 0, - .n_descsz = sizeof(BOOTLOADER), - .n_type = EBN_BOOTLOADER_NAME, - }, - .bl_desc = BOOTLOADER, - .blv_hdr = { - .n_namesz = 0, - .n_descsz = sizeof(BOOTLOADER_VERSION), - .n_type = EBN_BOOTLOADER_VERSION, - }, - .blv_desc = BOOTLOADER_VERSION, - .cmd_hdr = { - .n_namesz = 0, - .n_descsz = 0, - .n_type = EBN_COMMAND_LINE, - }, -}; - - #define OPT_APPEND (OPT_ARCH_MAX+0) +static char cmdline_buf[256] = "kexec "; int elf_mips_probe(const char *buf, off_t len) { - struct mem_ehdr ehdr; int result; result = build_elf_exec_info(buf, len, &ehdr, 0); @@ -108,16 +74,14 @@ int elf_mips_load(int argc, char **argv, const char *buf, off_t len, struct kexec_info *info) { struct mem_ehdr ehdr; - char *arg_buf; - size_t arg_bytes; - unsigned long arg_base; - struct boot_notes *notes; - size_t note_bytes; const char *command_line; int command_line_len; - unsigned char *setup_start; - uint32_t setup_size; + char *crash_cmdline; int opt; + int result; + unsigned long cmdline_addr; + size_t i; + unsigned long bss_start = 0, bss_size = 0; static const struct option options[] = { KEXEC_ARCH_OPTIONS {"command-line", 1, 0, OPT_APPEND}, @@ -144,38 +108,83 @@ int elf_mips_load(int argc, char **argv, const char *buf, off_t len, break; } } + command_line_len = 0; - setup_simple_regs.spr9 = 0; - if (command_line) { - command_line_len = strlen(command_line) + 1; - setup_simple_regs.spr9 = 2; + + /* Need to append some command line parameters internally in case of + * taking crash dumps. + */ + if (info->kexec_flags & KEXEC_ON_CRASH) { + crash_cmdline = xmalloc(COMMAND_LINE_SIZE); + memset((void *)crash_cmdline, 0, COMMAND_LINE_SIZE); + } else + crash_cmdline = NULL; + + result = build_elf_exec_info(buf, len, &ehdr, 0); + if (result < 0) + die("ELF exec parse failed\n"); + + /* Read in the PT_LOAD segments and remove CKSEG0 mask from address*/ + for (i = 0; i < ehdr.e_phnum; i++) { + struct mem_phdr *phdr; + phdr = &ehdr.e_phdr[i]; + if (phdr->p_type == PT_LOAD) + phdr->p_paddr = virt_to_phys(phdr->p_paddr); } - /* Load the ELF executable */ - elf_exec_build_load(info, &ehdr, buf, len, 0); + for (i = 0; i < ehdr.e_shnum; i++) { + struct mem_shdr *shdr; + unsigned char *strtab; + strtab = (unsigned char *)ehdr.e_shdr[ehdr.e_shstrndx].sh_data; + + shdr = &ehdr.e_shdr[i]; + if (shdr->sh_size && + strcmp((char *)&strtab[shdr->sh_name], + ".bss") == 0) { + bss_start = virt_to_phys(shdr->sh_addr); + bss_size = shdr->sh_size; + break; + } - setup_start = setup_simple_start; - setup_size = setup_simple_size; - setup_simple_regs.spr8 = ehdr.e_entry; + } - note_bytes = sizeof(elf_boot_notes) + ((command_line_len + 3) & ~3); - arg_bytes = note_bytes + ((setup_size + 3) & ~3); + /* Load the Elf data */ + result = elf_exec_load(&ehdr, info); + if (result < 0) + die("ELF exec load failed\n"); - arg_buf = xmalloc(arg_bytes); - arg_base = add_buffer_virt(info, - arg_buf, arg_bytes, arg_bytes, 4, 0, elf_max_addr(&ehdr), 1); + info->entry = (void *)virt_to_phys(ehdr.e_entry); - notes = (struct boot_notes *)(arg_buf + ((setup_size + 3) & ~3)); + /* Put cmdline right after bss for crash*/ + if (info->kexec_flags & KEXEC_ON_CRASH) + cmdline_addr = bss_start + bss_size; + else + cmdline_addr = 0; - memcpy(arg_buf, setup_start, setup_size); - memcpy(notes, &elf_boot_notes, sizeof(elf_boot_notes)); - memcpy(notes->command_line, command_line, command_line_len); + if (!bss_size) + die("No .bss segment present\n"); - notes->hdr.b_size = note_bytes; - notes->cmd_hdr.n_descsz = command_line_len; - notes->hdr.b_checksum = compute_ip_checksum(notes, note_bytes); + if (command_line) + command_line_len = strlen(command_line) + 1; + + if (info->kexec_flags & KEXEC_ON_CRASH) { + result = load_crashdump_segments(info, crash_cmdline, + 0, 0); + if (result < 0) { + free(crash_cmdline); + return -1; + } + } - info->entry = (void *)arg_base; + if (command_line) + strncat(cmdline_buf, command_line, command_line_len); + if (crash_cmdline) + strncat(cmdline_buf, crash_cmdline, + sizeof(crash_cmdline) - + strlen(crash_cmdline) - 1); + add_buffer(info, cmdline_buf, sizeof(cmdline_buf), + sizeof(cmdline_buf), sizeof(void *), + cmdline_addr, 0x0fffffff, 1); return 0; } diff --git a/kexec/arch/mips/kexec-mips.c b/kexec/arch/mips/kexec-mips.c index cb75a2d..bd12bb3 100644 --- a/kexec/arch/mips/kexec-mips.c +++ b/kexec/arch/mips/kexec-mips.c @@ -21,8 +21,6 @@ #include "kexec-mips.h" #include -#define MAX_MEMORY_RANGES 64 -#define MAX_LINE 160 static struct memory_range memory_range[MAX_MEMORY_RANGES]; /* Return a sorted list of memory ranges. */ @@ -31,16 +29,6 @@ int get_memory_ranges(struct memory_range **range, int *ranges, { int memory_ranges = 0; -#if 1 - /* this is valid for gemini2 platform based on tx4938 - * in our case, /proc/iomem doesn't report ram space - */ - memory_range[memory_ranges].start = 0x00000000; - memory_range[memory_ranges].end = 0x04000000; - memory_range[memory_ranges].type = RANGE_RAM; - memory_ranges++; -#else -#error Please, fix this for your platform const char iomem[] = "/proc/iomem"; char line[MAX_LINE]; FILE *fp; @@ -61,31 +49,19 @@ int get_memory_ranges(struct memory_range **range, int *ranges, continue; str = line + consumed; end = end + 1; -#if 0 - printf("%016Lx-%016Lx : %s\n", start, end, str); -#endif if (memcmp(str, "System RAM\n", 11) == 0) { type = RANGE_RAM; } else if (memcmp(str, "reserved\n", 9) == 0) { type = RANGE_RESERVED; - } else if (memcmp(str, "ACPI Tables\n", 12) == 0) { - type = RANGE_ACPI; - } else if (memcmp(str, "ACPI Non-volatile Storage\n", 26) == 0) { - type = RANGE_ACPI_NVS; } else { continue; } memory_range[memory_ranges].start = start; memory_range[memory_ranges].end = end; memory_range[memory_ranges].type = type; -#if 0 - printf("%016Lx-%016Lx : %x\n", start, end, type); -#endif memory_ranges++; } fclose(fp); -#endif - *range = memory_range; *ranges = memory_ranges; return 0; @@ -98,8 +74,18 @@ int file_types = sizeof(file_type) / sizeof(file_type[0]); void arch_usage(void) { +#ifdef __mips64 + fprintf(stderr, " --elf32-core-headers Prepare core headers in " + "ELF32 format\n"); +#endif } +#ifdef __mips64 +struct arch_options_t arch_options = { + .core_header_type = CORE_TYPE_ELF64 +}; +#endif + int arch_process_options(int argc, char **argv) { static const struct option options[] = { @@ -126,12 +112,14 @@ const struct arch_map_entry arches[] = { /* For compatibility with older patches * use KEXEC_ARCH_DEFAULT instead of KEXEC_ARCH_MIPS here. */ - { "mips", KEXEC_ARCH_DEFAULT }, + { "mips", KEXEC_ARCH_MIPS }, + { "mips64", KEXEC_ARCH_MIPS }, { NULL, 0 }, }; int arch_compat_trampoline(struct kexec_info *UNUSED(info)) { + return 0; } @@ -139,18 +127,9 @@ void arch_update_purgatory(struct kexec_info *UNUSED(info)) { } -/* - * Adding a dummy function, so that build on mips will not break. - * Need to implement the actual checking code - */ -int is_crashkernel_mem_reserved(void) -{ - return 1; -} - unsigned long virt_to_phys(unsigned long addr) { - return addr - 0x80000000; + return addr & 0x7fffffff; } /* @@ -159,7 +138,7 @@ unsigned long virt_to_phys(unsigned long addr) void add_segment(struct kexec_info *info, const void *buf, size_t bufsz, unsigned long base, size_t memsz) { - add_segment_phys_virt(info, buf, bufsz, base, memsz, 1); + add_segment_phys_virt(info, buf, bufsz, virt_to_phys(base), memsz, 1); } /* diff --git a/kexec/arch/mips/kexec-mips.h b/kexec/arch/mips/kexec-mips.h index 13f82be..6d062fc 100644 --- a/kexec/arch/mips/kexec-mips.h +++ b/kexec/arch/mips/kexec-mips.h @@ -1,6 +1,11 @@ #ifndef KEXEC_MIPS_H #define KEXEC_MIPS_H +#define MAX_MEMORY_RANGES 64 +#define MAX_LINE 160 + +#define CORE_TYPE_ELF32 1 +#define CORE_TYPE_ELF64 2 extern unsigned char setup_simple_start[]; extern uint32_t setup_simple_size; @@ -14,4 +19,8 @@ int elf_mips_load(int argc, char **argv, const char *buf, off_t len, struct kexec_info *info); void elf_mips_usage(void); +struct arch_options_t { + int core_header_type; +}; + #endif /* KEXEC_MIPS_H */ diff --git a/kexec/arch/mips/mips-setup-simple.S b/kexec/arch/mips/mips-setup-simple.S deleted file mode 100644 index 1acdee3..0000000 --- a/kexec/arch/mips/mips-setup-simple.S +++ /dev/null @@ -1,110 +0,0 @@ -/* - * mips-setup-simple.S - code to execute before kernel to handle command line - * Copyright (C) 2007 Francesco Chiechi, Alessandro Rubini - * Copyright (C) 2007 Tvblob s.r.l. - * - * derived from Albert Herranz idea (ppc) adding command line support - * (boot_notes structure) - * - * This source code is licensed under the GNU General Public License, - * Version 2. See the file COPYING for more details. - */ - -/* - * Only suitable for platforms booting with MMU turned off. - * -- Albert Herranz - */ -#include "regdef.h" - -/* returns t0 = relocated address of sym */ -/* modifies t1 t2 */ -/* sym must not be global or this will not work (at least AFAIK up to now) */ -#define RELOC_SYM(sym) \ - move t0,ra; /* save ra */ \ - bal 1f; \ -1: \ - move t1,ra; /* now t1 is 1b (where we are now) */ \ - move ra,t0; /* restore ra */ \ - lui t2,%hi(1b); \ - ori t2,t2,%lo(1b); \ - lui t0,%hi(sym); \ - ori t0,t0,%lo(sym); \ - sub t0,t0,t2; /* t0 = offset between sym and 1b */ \ - add t0,t1,t0; /* t0 = actual address in memory */ - - .data - .globl setup_simple_start -setup_simple_start: - - /* should perform here any required setup */ - - /* Initialize GOT pointer (verify if needed) */ - bal 1f - nop - .word _GLOBAL_OFFSET_TABLE_ - 1: - move gp, ra - lw t1, 0(ra) - move gp, t1 - - /* spr8 relocation */ - RELOC_SYM(spr8) - - move t4,t0 // save pointer to kernel start addr - lw t3,0(t0) // save kernel start address - - /* spr9 relocation */ - RELOC_SYM(spr9) - lw a0,0(t0) // load argc - - // this code is to be changed if boot_notes struct changes - lw t2,12(t4) // t2 is size of boot_notes struct - addi t2,t2,3 - srl t2,t2,2 - sll v1,t2,2 // v1 = size of boot_notes struct - // aligned to word boundary - - addi t0,t4,0x20 // t0 contains the address of "kexec" string - add v0,t4,v1 // v0 points to last word of boot_notes - addi v0,v0,8 // v0 points to address after boot_notes - sw t0,0(v0) // store pointer to "kexec" string there - - lw t2,-8(t0) // t2 is size of "kexec" string in bytes - addi t2,t2,3 - srl t2,t2,2 - sll v1,t2,2 // v1 = size of "kexec" string - // aligned to word boundary - add t2,t0,v1 - addi t0,t2,4 // t0 points to size of version string - - lw t2,0(t0) // t2 is size of version string in bytes - addi t2,t2,3 - srl t2,t2,2 - sll v1,t2,2 // v1 = size of version string - // aligned to word boundary - - addi t0,t0,8 // t0 points to version string - add t0,t0,v1 // t0 points to start of command_line record - addi t0,t0,12 // t0 points command line - - sw t0,4(v0) // store pointer to command line - - move a1,v0 // load argv - li a2,0 - li a3,0 - - jr t3 - nop - - .balign 4 - .globl setup_simple_regs -setup_simple_regs: -spr8: .long 0x00000000 -spr9: .long 0x00000000 - -setup_simple_end: - - .globl setup_simple_size -setup_simple_size: - .long setup_simple_end - setup_simple_start - diff --git a/kexec/crashdump.h b/kexec/crashdump.h index 1789b53..30d6f29 100644 --- a/kexec/crashdump.h +++ b/kexec/crashdump.h @@ -26,7 +26,7 @@ struct crash_elf_info { unsigned long backup_src_start; unsigned long backup_src_end; - unsigned long page_offset; + unsigned long long page_offset; unsigned long lowmem_limit; int (*get_note_info)(int cpu, uint64_t *addr, uint64_t *len); diff --git a/kexec/kexec-syscall.h b/kexec/kexec-syscall.h index 69c9686..7f41a1b 100644 --- a/kexec/kexec-syscall.h +++ b/kexec/kexec-syscall.h @@ -22,6 +22,7 @@ #define LINUX_REBOOT_CMD_KEXEC_OLD2 0x18263645 #define LINUX_REBOOT_CMD_KEXEC 0x45584543 +#ifndef __NR_kexec_load #ifdef __i386__ #define __NR_kexec_load 283 #endif @@ -60,6 +61,7 @@ #ifndef __NR_kexec_load #error Unknown processor architecture. Needs a kexec_load syscall number. #endif +#endif /*ifndef __NR_kexec_load*/ struct kexec_segment; Signed-off-by: Maxim Uvarov From lists@nabble.com Wed Mar 3 13:58:53 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 03 Mar 2010 13:58:57 +0100 (CET) Received: from kuber.nabble.com ([216.139.236.158]:40801 "EHLO kuber.nabble.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491065Ab0CCM6w (ORCPT ); Wed, 3 Mar 2010 13:58:52 +0100 Received: from isper.nabble.com ([192.168.236.156]) by kuber.nabble.com with esmtp (Exim 4.63) (envelope-from ) id 1NmoA4-0003yM-G9 for linux-mips@linux-mips.org; Wed, 03 Mar 2010 04:58:48 -0800 Message-ID: <27767861.post@talk.nabble.com> Date: Wed, 3 Mar 2010 04:58:48 -0800 (PST) From: Dea_RU To: linux-mips@linux-mips.org Subject: Re: TI AR7 7200 - no boot In-Reply-To: <27766728.post@talk.nabble.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Nabble-From: dryukovz@mail.ru References: <27766331.post@talk.nabble.com> <201003031124.46336.florian@openwrt.org> <27766728.post@talk.nabble.com> Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26106 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: dryukovz@mail.ru Precedence: bulk X-list: linux-mips i adder printk(...) before all function in kernel init funktion /init/main.c log show last message before called fundtion local_irq_enable(). If delete my debug printk-s kernel stop after first console output "Calibrating delay loop... " I do not MIPS arch interput specific ..... Can be MIPS need init internel timer and link overflow flag to interput ?? ------ log with added printk ---- ...... Memory: 13172k/16384k available (1314k kernel code, 3212k reserved, 411k data, 120k init, 0k highmem) early_irq_init NR_IRQS:256 <- kernel init_IRQ prio_tree_init init_timers hrtimers_init softirq_init timerkepping_init time init profile_init early_boot_irqs_on local_irq_enable <- last my debug printk message before call to local_irq_enable() ------------------------------- -- View this message in context: http://old.nabble.com/TI-AR7-7200---no-boot-tp27766331p27767861.html Sent from the linux-mips main mailing list archive at Nabble.com. From f.fainelli@gmail.com Wed Mar 3 14:05:17 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 03 Mar 2010 14:05:22 +0100 (CET) Received: from mail-ew0-f223.google.com ([209.85.219.223]:59076 "EHLO mail-ew0-f223.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491839Ab0CCNFR (ORCPT ); Wed, 3 Mar 2010 14:05:17 +0100 Received: by ewy23 with SMTP id 23so817987ewy.24 for ; Wed, 03 Mar 2010 05:05:11 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:sender:from:organization:to :subject:date:user-agent:cc:references:in-reply-to:mime-version :content-type:content-transfer-encoding:message-id; bh=GO8laIMURquqoWTWYTvsWaGIGytqK2JxPTyUZ7gmqwo=; b=wvkhY3+ubCMuYF4alqbHwOjSihVxIV730lOK2BTbbtlmrpnqfmQaRn44T8dWEvItbv 7B9Y/WUEhLEpepbB/A4IIWSEW/0JmxdskRcjsypGb7Mx9NNdQiZUNLFLtymt8AkCc/wK tropTKOJEw4t5XpU4mm85YG55okxFevxnev/s= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=sender:from:organization:to:subject:date:user-agent:cc:references :in-reply-to:mime-version:content-type:content-transfer-encoding :message-id; b=ks4I7y6hzOCPK8towkV0WCciTpZ412wUQc0EUMMlEgsUt3fhUaCvfKxh0zbYEtQ0dJ o7pGVC/Ask3F5gPqOIL2M9pwCY20dTBU+YrJZ1Bw/gdQFsJsCf7rLQd6S5e+lRG5NVcA tvXHnrAuwdTgfj2gCye48ByxW80NKmRQnpnXg= Received: by 10.213.96.198 with SMTP id i6mr5974126ebn.45.1267621511113; Wed, 03 Mar 2010 05:05:11 -0800 (PST) Received: from flexo.localnet (bobafett.staff.proxad.net [213.228.1.121]) by mx.google.com with ESMTPS id 7sm16795662eyb.33.2010.03.03.05.05.09 (version=SSLv3 cipher=RC4-MD5); Wed, 03 Mar 2010 05:05:10 -0800 (PST) From: Florian Fainelli Organization: OpenWrt To: "Dea_RU" Subject: Re: TI AR7 7200 - no boot Date: Wed, 3 Mar 2010 14:04:52 +0100 User-Agent: KMail/1.12.2 (Linux/2.6.31-17-server; KDE/4.3.2; x86_64; ; ) Cc: linux-mips@linux-mips.org References: <27766331.post@talk.nabble.com> <27766728.post@talk.nabble.com> <27767861.post@talk.nabble.com> In-Reply-To: <27767861.post@talk.nabble.com> MIME-Version: 1.0 Content-Type: Text/Plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Message-Id: <201003031404.53039.florian@openwrt.org> Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26107 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: florian@openwrt.org Precedence: bulk X-list: linux-mips On Wednesday 03 March 2010 13:58:48 Dea_RU wrote: > i adder printk(...) before all function in kernel init funktion > /init/main.c > > log show last message before called fundtion local_irq_enable(). > > If delete my debug printk-s kernel stop after first console output > "Calibrating delay loop... " > > I do not MIPS arch interput specific ..... Can be MIPS need init internel > timer and link overflow flag to interput ?? All handlers are not correctly setup, so you should include this patch in your tree: http://www.linux-mips.org/archives/linux-mips/2010-01/msg00019.html > > > ------ log with added printk ---- > ...... > Memory: 13172k/16384k available (1314k kernel code, 3212k reserved, 411k > data, 120k init, 0k highmem) > early_irq_init > NR_IRQS:256 <- kernel > init_IRQ > prio_tree_init > init_timers > hrtimers_init > softirq_init > timerkepping_init > time init > profile_init > early_boot_irqs_on > local_irq_enable <- last my debug printk message before call to > local_irq_enable() > ------------------------------- > -- Regards, Florian From lists@nabble.com Wed Mar 3 15:24:07 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 03 Mar 2010 15:24:14 +0100 (CET) Received: from kuber.nabble.com ([216.139.236.158]:48898 "EHLO kuber.nabble.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491793Ab0CCOYH convert rfc822-to-8bit (ORCPT ); Wed, 3 Mar 2010 15:24:07 +0100 Received: from isper.nabble.com ([192.168.236.156]) by kuber.nabble.com with esmtp (Exim 4.63) (envelope-from ) id 1NmpUZ-0002DK-DL for linux-mips@linux-mips.org; Wed, 03 Mar 2010 06:24:03 -0800 Message-ID: <27768844.post@talk.nabble.com> Date: Wed, 3 Mar 2010 06:24:03 -0800 (PST) From: Dea_RU To: linux-mips@linux-mips.org Subject: Re: TI AR7 7200 - no boot In-Reply-To: <201003031404.53039.florian@openwrt.org> MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8BIT X-Nabble-From: dryukovz@mail.ru References: <27766331.post@talk.nabble.com> <201003031124.46336.florian@openwrt.org> <27766728.post@talk.nabble.com> <27767861.post@talk.nabble.com> <201003031404.53039.florian@openwrt.org> Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26108 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: dryukovz@mail.ru Precedence: bulk X-list: linux-mips it work ! great work ! why this patch not present in src 2.6.33 ??? ============ log (psbl) Booting... Linux version 2.6.33 (root@linwork) (gcc version 4.3.3 (Buildroot 2010.02-rc1) ) #13 Wed Mar 3 20:04:20 OMST 2010 bootconsole [early0] enabled CPU revision is: 00018448 (MIPS 4KEc) TI AR7 (TNETD7200), ID: 0x002b, Revision: 0x11 Determined physical RAM map: memory: 01000000 @ 14000000 (usable) Zone PFN ranges: Normal 0x00014000 -> 0x00015000 Movable zone start PFN for each node early_node_map[1] active PFN ranges 0: 0x00014000 -> 0x00015000 Built 1 zonelists in Zone order, mobility grouping off. Total pages: 4064 Kernel command line: console=ttyS0,9600n8 earlyprintk=ttyS0,9600 PID hash table entries: 64 (order: -4, 256 bytes) Dentry cache hash table entries: 2048 (order: 1, 8192 bytes) Inode-cache hash table entries: 1024 (order: 0, 4096 bytes) Primary instruction cache 16kB, VIPT, 4-way, linesize 16 bytes. Primary data cache 8kB, 4-way, VIPT, no aliases, linesize 16 bytes Memory: 13172k/16384k available (1316k kernel code, 3212k reserved, 409k data, 120k init, 0k highmem) NR_IRQS:256 Console: colour dummy device 80x25 Calibrating delay loop... 149.50 BogoMIPS (lpj=299008) Mount-cache hash table entries: 512 rest_init Clocks: Sync 2:1 mode Clocks: Setting CPU clock Adjusted requested frequency 211000000 to 211968000 Clocks: base = 35328000, frequency = 211968000, prediv = 1, postdiv = 1, postdiv2 = -1, mul = 6 Clocks: Setting DSP clock Clocks: base = 25000000, frequency = 105984000, prediv = 1, postdiv = 2, postdiv2 = 1, mul = 10 Clocks: Setting USB clock Adjusted requested frequency 48000000 to 47863741 Clocks: base = 105984000, frequency = 48000000, prediv = 1, postdiv = 31, postdiv2 = -1, mul = 14 LinuxPPS API ver. 1 registered Software ver. 5.3.6 - Copyright 2005-2007 Rodolfo Giometti Switching to clocksource MIPS Slow work thread pool: Starting up Slow work thread pool: Ready FS-Cache: Loaded msgmni has been set to 25 alg: No test for cipher_null (cipher_null-generic) alg: No test for ecb(cipher_null) (ecb-cipher_null) alg: No test for digest_null (digest_null-generic) alg: No test for compress_null (compress_null-generic) alg: No test for stdrng (krng) Serial: 8250/16550 driver, 2 ports, IRQ sharing disabled serial8250: ttyS0 at MMIO 0x8610e00 (irq = 15) is a XScale Ã}qâ„¢ya)¶‰/½¾£ÂÖ ².W,XvÃ!aâ„¢,Ó¯\“× ---------- can not select USART options for show it ... -- View this message in context: http://old.nabble.com/TI-AR7-7200---no-boot-tp27766331p27768844.html Sent from the linux-mips main mailing list archive at Nabble.com. From sgi-linux-mips@m.gmane.org Wed Mar 3 17:08:59 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 03 Mar 2010 17:09:04 +0100 (CET) Received: from lo.gmane.org ([80.91.229.12]:53087 "EHLO lo.gmane.org" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492188Ab0CCQI5 (ORCPT ); Wed, 3 Mar 2010 17:08:57 +0100 Received: from list by lo.gmane.org with local (Exim 4.69) (envelope-from ) id 1Nmr81-0006jF-4S for linux-mips@linux-mips.org; Wed, 03 Mar 2010 17:08:53 +0100 Received: from chipmunk.wormnet.eu ([195.195.131.226]) by main.gmane.org with esmtp (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Wed, 03 Mar 2010 17:08:53 +0100 Received: from alex by chipmunk.wormnet.eu with local (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Wed, 03 Mar 2010 17:08:53 +0100 X-Injected-Via-Gmane: http://gmane.org/ To: linux-mips@linux-mips.org From: Alexander Clouter Subject: Re: TI AR7 7200 - no boot Date: Wed, 3 Mar 2010 15:54:39 +0000 Lines: 28 Message-ID: References: <27766331.post@talk.nabble.com> <27766728.post@talk.nabble.com> <27767861.post@talk.nabble.com> <201003031404.53039.florian@openwrt.org> X-Complaints-To: usenet@dough.gmane.org X-Gmane-NNTP-Posting-Host: chipmunk.wormnet.eu User-Agent: tin/1.9.3-20080506 ("Dalintober") (UNIX) (Linux/2.6.26-2-sparc64-smp (sparc64)) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26109 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: alex@digriz.org.uk Precedence: bulk X-list: linux-mips Florian Fainelli wrote: > > On Wednesday 03 March 2010 13:58:48 Dea_RU wrote: >> >> If delete my debug printk-s kernel stop after first console output >> "Calibrating delay loop... " >> >> I do not MIPS arch interput specific ..... Can be MIPS need init internel >> timer and link overflow flag to interput ?? > > All handlers are not correctly setup, so you should include this patch in your > tree: > > http://www.linux-mips.org/archives/linux-mips/2010-01/msg00019.html > If I could slip in with some shameless self advertising: http://www.digriz.org.uk/wag54g Same(ish) chipset, the 'Creating Firmwares' section should interest you and get you started on your buildroot journey. Cheers -- Alexander Clouter .sigmonster says: guru, n: A computer owner who can read the manual. From wilbur512@gmail.com Wed Mar 3 17:34:31 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 03 Mar 2010 17:34:34 +0100 (CET) Received: from mail-gx0-f224.google.com ([209.85.217.224]:41758 "EHLO mail-gx0-f224.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492180Ab0CCQeb (ORCPT ); Wed, 3 Mar 2010 17:34:31 +0100 Received: by gxk24 with SMTP id 24so635484gxk.7 for ; Wed, 03 Mar 2010 08:34:23 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:in-reply-to:references :date:message-id:subject:from:to:cc:content-type; bh=s3ViADiP22IoAQyguDqntN640mhHos1Ss0CiUswtWeA=; b=UqGIDDc3p6pQwaOKXwY7pRBt1wxkNAqEKjkxdTua92xUxg61mB4cjH8xB1hI2E2VRj M/DsqpijUB+aPUQnk9mzE0GaMyvPxYGFleTgY7/85Orga++zelQMZ61ehL9WSKN2F7Aq WYDsaLSDxeN41gzzNp8GmkSlf5bzB1X/19aAA= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; b=ga+qZNo/L2nTT8uk3nRdan/k7+8ONNtZ0l3RfeMEtVNyDIp1o/Irt/zjK+/oiIExIQ xQVS412HRKWOV1zc1ozoXNhP44jAC+gfze8dqCYNIOt8j16CleldGVGTIDZao0hNN0CI f963NWisN53bCwHR389VtYNjEGLYIwdki2Juw= MIME-Version: 1.0 Received: by 10.90.38.27 with SMTP id l27mr4992343agl.6.1267634063016; Wed, 03 Mar 2010 08:34:23 -0800 (PST) In-Reply-To: <20100303110527.11233.20400.stgit@muvarov> References: <20100303110527.11233.20400.stgit@muvarov> Date: Thu, 4 Mar 2010 00:34:22 +0800 Message-ID: Subject: Re: [PATCH 1/2] MIPS kexec,kdump support From: "wilbur.chan" To: Maxim Uvarov Cc: linux-mips@linux-mips.org, kexec@lists.infradead.org, horms@verge.net.au, ralf@linux-mips.org Content-Type: text/plain; charset=ISO-8859-1 Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26110 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: wilbur512@gmail.com Precedence: bulk X-list: linux-mips 2010/3/3 Maxim Uvarov : > Hello folks, > > Please find here MIPS crash and kdump patches. > This is patch set of 3 patches: > 1. generic MIPS changes (kernel); > 2. MIPS Cavium Octeon board kexec/kdump code (kernel); > 3. Kexec user space MIPS changes. > > Patches were tested on the latest linux-mips@ git kernel and the latest > kexec-tools git on Cavium Octeon 50xx board. > > I also made the same code working on RMI XLR/XLS boards for both > mips32 and mips64 kernels. > > Best regards, > Maxim Uvarov. > > Signed-off-by: Maxim Uvarov > > > Hi, Maxim In XLR series , 1)How to protect boardinfo and pass it to second kernel ? 2)If all cpus jumped to same entry point , did you change head.s ,if so , how ? Thank you! From David.Daney@caviumnetworks.com Wed Mar 3 18:37:24 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 03 Mar 2010 18:37:28 +0100 (CET) Received: from mail3.caviumnetworks.com ([12.108.191.235]:18061 "EHLO mail3.caviumnetworks.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491942Ab0CCRhX (ORCPT ); Wed, 3 Mar 2010 18:37:23 +0100 Received: from caexch01.caveonetworks.com (Not Verified[192.168.16.9]) by mail3.caviumnetworks.com with MailMarshal (v6,7,2,8378) id ; Wed, 03 Mar 2010 09:37:25 -0800 Received: from caexch01.caveonetworks.com ([192.168.16.9]) by caexch01.caveonetworks.com with Microsoft SMTPSVC(6.0.3790.3959); Wed, 3 Mar 2010 09:36:17 -0800 Received: from dd1.caveonetworks.com ([12.108.191.236]) by caexch01.caveonetworks.com over TLS secured channel with Microsoft SMTPSVC(6.0.3790.3959); Wed, 3 Mar 2010 09:36:17 -0800 Message-ID: <4B8E9E0C.4050809@caviumnetworks.com> Date: Wed, 03 Mar 2010 09:36:12 -0800 From: David Daney User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.8) Gecko/20100225 Fedora/3.0.2-1.fc12 Thunderbird/3.0.2 MIME-Version: 1.0 To: Maxim Uvarov CC: linux-mips@linux-mips.org, kexec@lists.infradead.org, horms@verge.net.au, ralf@linux-mips.org Subject: Re: [PATCH 1/2] MIPS kexec,kdump support References: <20100303110527.11233.20400.stgit@muvarov> In-Reply-To: <20100303110527.11233.20400.stgit@muvarov> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit X-OriginalArrivalTime: 03 Mar 2010 17:36:17.0245 (UTC) FILETIME=[0768E0D0:01CABAF8] Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26111 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ddaney@caviumnetworks.com Precedence: bulk X-list: linux-mips On 03/03/2010 03:05 AM, Maxim Uvarov wrote: > Hello folks, > > Please find here MIPS crash and kdump patches. > This is patch set of 3 patches: > 1. generic MIPS changes (kernel); > 2. MIPS Cavium Octeon board kexec/kdump code (kernel); > 3. Kexec user space MIPS changes. > > Patches were tested on the latest linux-mips@ git kernel and the latest > kexec-tools git on Cavium Octeon 50xx board. > > I also made the same code working on RMI XLR/XLS boards for both > mips32 and mips64 kernels. > > Best regards, > Maxim Uvarov. > > > --- > > arch/mips/Kconfig | 24 ++++++++++ [...] > > > Signed-off-by: Maxim Uvarov > That Signed-off-by: needs to be just above the '---' not at the end. David Daney From David.Daney@caviumnetworks.com Wed Mar 3 18:56:47 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 03 Mar 2010 18:56:51 +0100 (CET) Received: from smtp2.caviumnetworks.com ([209.113.159.134]:15568 "EHLO smtp2.caviumnetworks.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491194Ab0CCR4q (ORCPT ); Wed, 3 Mar 2010 18:56:46 +0100 Received: from maexch1.caveonetworks.com (Not Verified[192.168.14.20]) by smtp2.caviumnetworks.com with MailMarshal (v6,5,4,7535) id ; Wed, 03 Mar 2010 12:50:21 -0500 Received: from caexch01.caveonetworks.com ([192.168.16.9]) by maexch1.caveonetworks.com with Microsoft SMTPSVC(6.0.3790.3959); Wed, 3 Mar 2010 12:56:43 -0500 Received: from caexch01.caveonetworks.com ([192.168.16.9]) by caexch01.caveonetworks.com with Microsoft SMTPSVC(6.0.3790.3959); Wed, 3 Mar 2010 09:53:06 -0800 Received: from dd1.caveonetworks.com ([12.108.191.236]) by caexch01.caveonetworks.com over TLS secured channel with Microsoft SMTPSVC(6.0.3790.3959); Wed, 3 Mar 2010 09:53:06 -0800 Message-ID: <4B8EA1FD.5050007@caviumnetworks.com> Date: Wed, 03 Mar 2010 09:53:01 -0800 From: David Daney User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.8) Gecko/20100225 Fedora/3.0.2-1.fc12 Thunderbird/3.0.2 MIME-Version: 1.0 To: Yang Shi CC: ralf@linux-mips.org, f.fainelli@gmail.com, linux-mips@linux-mips.org Subject: Re: [PATCH 1/3] MIPS: Octeon: Remove superfluous on_each_cpu parameter References: <1267605801-5305-1-git-send-email-yang.shi@windriver.com> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-OriginalArrivalTime: 03 Mar 2010 17:53:06.0861 (UTC) FILETIME=[61300DD0:01CABAFA] Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26112 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ddaney@caviumnetworks.com Precedence: bulk X-list: linux-mips On 03/03/2010 12:43 AM, Yang Shi wrote: > Now, on_each_cpu just need three parameters, but the on_each_cpu > still uses four parameters in Octeon's setup.c. So, remove the > superfluous parameter. > > Signed-off-by: Yang Shi NAK! We are removing CONFIG_CAVIUM_RESERVE32_USE_WIRED_TLB completely. I will send the removal patch soon. David Daney > --- > arch/mips/cavium-octeon/setup.c | 2 +- > 1 files changed, 1 insertions(+), 1 deletions(-) > > diff --git a/arch/mips/cavium-octeon/setup.c b/arch/mips/cavium-octeon/setup.c > index b321d3b..4eaa35f 100644 > --- a/arch/mips/cavium-octeon/setup.c > +++ b/arch/mips/cavium-octeon/setup.c > @@ -230,7 +230,7 @@ static void octeon_hal_setup_per_cpu_reserved32(void *unused) > void octeon_hal_setup_reserved32(void) > { > #ifdef CONFIG_CAVIUM_RESERVE32_USE_WIRED_TLB > - on_each_cpu(octeon_hal_setup_per_cpu_reserved32, NULL, 0, 1); > + on_each_cpu(octeon_hal_setup_per_cpu_reserved32, NULL, 1); > #endif > } > From David.Daney@caviumnetworks.com Wed Mar 3 19:24:53 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 03 Mar 2010 19:24:56 +0100 (CET) Received: from mail3.caviumnetworks.com ([12.108.191.235]:19276 "EHLO mail3.caviumnetworks.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491194Ab0CCSYw (ORCPT ); Wed, 3 Mar 2010 19:24:52 +0100 Received: from caexch01.caveonetworks.com (Not Verified[192.168.16.9]) by mail3.caviumnetworks.com with MailMarshal (v6,7,2,8378) id ; Wed, 03 Mar 2010 10:24:59 -0800 Received: from caexch01.caveonetworks.com ([192.168.16.9]) by caexch01.caveonetworks.com with Microsoft SMTPSVC(6.0.3790.3959); Wed, 3 Mar 2010 10:24:16 -0800 Received: from dd1.caveonetworks.com ([12.108.191.236]) by caexch01.caveonetworks.com over TLS secured channel with Microsoft SMTPSVC(6.0.3790.3959); Wed, 3 Mar 2010 10:24:16 -0800 Message-ID: <4B8EA94B.1020203@caviumnetworks.com> Date: Wed, 03 Mar 2010 10:24:11 -0800 From: David Daney User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.8) Gecko/20100225 Fedora/3.0.2-1.fc12 Thunderbird/3.0.2 MIME-Version: 1.0 To: Yang Shi CC: ralf@linux-mips.org, f.fainelli@gmail.com, linux-mips@linux-mips.org Subject: Re: [PATCH 3/3] MIPS: Octeon: Add add_wired_entry decralation in header file References: <1267605801-5305-1-git-send-email-yang.shi@windriver.com> <50e36e8549769a26986f99a23772d23fd039c230.1267604875.git.yang.shi@windriver.com> <004eb64c73b3bcec90612663598ada4cf678f236.1267604875.git.yang.shi@windriver.com> In-Reply-To: <004eb64c73b3bcec90612663598ada4cf678f236.1267604875.git.yang.shi@windriver.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-OriginalArrivalTime: 03 Mar 2010 18:24:16.0388 (UTC) FILETIME=[BB833C40:01CABAFE] Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26113 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ddaney@caviumnetworks.com Precedence: bulk X-list: linux-mips On 03/03/2010 12:43 AM, Yang Shi wrote: > Octeon's setup.c uses add_wired_entry, but it is not declared > anywhere. Copy add_wired_entry decralation from pgtable-32.h to > pgtable-64.h and include asm/pgtable.h into Octeon's setup.c. > > Signed-off-by: Yang Shi NAK! We are removing the use of add_wired_entry(), so adding a declaration will not be necessary. David Daney > --- > arch/mips/cavium-octeon/setup.c | 1 + > arch/mips/include/asm/pgtable-64.h | 6 ++++++ > 2 files changed, 7 insertions(+), 0 deletions(-) > > diff --git a/arch/mips/cavium-octeon/setup.c b/arch/mips/cavium-octeon/setup.c > index 8309d68..d63b8e6 100644 > --- a/arch/mips/cavium-octeon/setup.c > +++ b/arch/mips/cavium-octeon/setup.c > @@ -30,6 +30,7 @@ > #include > #include > #include > +#include > > #include > > diff --git a/arch/mips/include/asm/pgtable-64.h b/arch/mips/include/asm/pgtable-64.h > index 26dc69d..85ee34d 100644 > --- a/arch/mips/include/asm/pgtable-64.h > +++ b/arch/mips/include/asm/pgtable-64.h > @@ -23,6 +23,12 @@ > #endif > > /* > + * - add_wired_entry() add a fixed TLB entry, and move wired register > + */ > +extern void add_wired_entry(unsigned long entrylo0, unsigned long entrylo1, > + unsigned long entryhi, unsigned long pagemask); > + > +/* > * Each address space has 2 4K pages as its page directory, giving 1024 > * (== PTRS_PER_PGD) 8 byte pointers to pmd tables. Each pmd table is a > * single 4K page, giving 512 (== PTRS_PER_PMD) 8 byte pointers to page From David.Daney@caviumnetworks.com Wed Mar 3 19:25:52 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 03 Mar 2010 19:25:56 +0100 (CET) Received: from mail3.caviumnetworks.com ([12.108.191.235]:19299 "EHLO mail3.caviumnetworks.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491194Ab0CCSZw (ORCPT ); Wed, 3 Mar 2010 19:25:52 +0100 Received: from caexch01.caveonetworks.com (Not Verified[192.168.16.9]) by mail3.caviumnetworks.com with MailMarshal (v6,7,2,8378) id ; Wed, 03 Mar 2010 10:25:59 -0800 Received: from caexch01.caveonetworks.com ([192.168.16.9]) by caexch01.caveonetworks.com with Microsoft SMTPSVC(6.0.3790.3959); Wed, 3 Mar 2010 10:25:50 -0800 Received: from dd1.caveonetworks.com ([12.108.191.236]) by caexch01.caveonetworks.com over TLS secured channel with Microsoft SMTPSVC(6.0.3790.3959); Wed, 3 Mar 2010 10:25:50 -0800 Message-ID: <4B8EA9AD.8070106@caviumnetworks.com> Date: Wed, 03 Mar 2010 10:25:49 -0800 From: David Daney User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.8) Gecko/20100225 Fedora/3.0.2-1.fc12 Thunderbird/3.0.2 MIME-Version: 1.0 To: Yang Shi , ralf@linux-mips.org CC: f.fainelli@gmail.com, linux-mips@linux-mips.org Subject: Re: [PATCH 2/3] MIPS: Octeon: Remove redundant declaration of octeon_reserve32_memory References: <1267605801-5305-1-git-send-email-yang.shi@windriver.com> <50e36e8549769a26986f99a23772d23fd039c230.1267604875.git.yang.shi@windriver.com> In-Reply-To: <50e36e8549769a26986f99a23772d23fd039c230.1267604875.git.yang.shi@windriver.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-OriginalArrivalTime: 03 Mar 2010 18:25:50.0029 (UTC) FILETIME=[F353B7D0:01CABAFE] Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26114 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ddaney@caviumnetworks.com Precedence: bulk X-list: linux-mips On 03/03/2010 12:43 AM, Yang Shi wrote: > In Octeon's setup.c, octeon_reserve32_memory is defined, so remove the > redundant extern declaration of this variable. > > Signed-off-by: Yang Shi Acked-by: David Daney This looks good to me. Thanks, David Daney > --- > arch/mips/cavium-octeon/setup.c | 3 --- > 1 files changed, 0 insertions(+), 3 deletions(-) > > diff --git a/arch/mips/cavium-octeon/setup.c b/arch/mips/cavium-octeon/setup.c > index 4eaa35f..8309d68 100644 > --- a/arch/mips/cavium-octeon/setup.c > +++ b/arch/mips/cavium-octeon/setup.c > @@ -45,9 +45,6 @@ extern struct plat_smp_ops octeon_smp_ops; > extern void pci_console_init(const char *arg); > #endif > > -#ifdef CONFIG_CAVIUM_RESERVE32 > -extern uint64_t octeon_reserve32_memory; > -#endif > static unsigned long long MAX_MEMORY = 512ull<< 20; > > struct octeon_boot_descriptor *octeon_boot_desc_ptr; From David.Daney@caviumnetworks.com Wed Mar 3 20:07:58 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 03 Mar 2010 20:08:01 +0100 (CET) Received: from mail3.caviumnetworks.com ([12.108.191.235]:1331 "EHLO mail3.caviumnetworks.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492297Ab0CCTH6 (ORCPT ); Wed, 3 Mar 2010 20:07:58 +0100 Received: from caexch01.caveonetworks.com (Not Verified[192.168.16.9]) by mail3.caviumnetworks.com with MailMarshal (v6,7,2,8378) id ; Wed, 03 Mar 2010 11:08:04 -0800 Received: from caexch01.caveonetworks.com ([192.168.16.9]) by caexch01.caveonetworks.com with Microsoft SMTPSVC(6.0.3790.3959); Wed, 3 Mar 2010 11:07:16 -0800 Received: from dd1.caveonetworks.com ([12.108.191.236]) by caexch01.caveonetworks.com over TLS secured channel with Microsoft SMTPSVC(6.0.3790.3959); Wed, 3 Mar 2010 11:07:16 -0800 Received: from dd1.caveonetworks.com (localhost.localdomain [127.0.0.1]) by dd1.caveonetworks.com (8.14.3/8.14.2) with ESMTP id o23J7BTU027745; Wed, 3 Mar 2010 11:07:11 -0800 Received: (from ddaney@localhost) by dd1.caveonetworks.com (8.14.3/8.14.3/Submit) id o23J799r027744; Wed, 3 Mar 2010 11:07:09 -0800 From: David Daney To: linux-mips@linux-mips.org, ralf@linux-mips.org Cc: David Daney Subject: [PATCH] MIPS: Octeon: Remove vestiges of CONFIG_CAVIUM_RESERVE32_USE_WIRED_TLB Date: Wed, 3 Mar 2010 11:07:07 -0800 Message-Id: <1267643227-27710-1-git-send-email-ddaney@caviumnetworks.com> X-Mailer: git-send-email 1.6.6.1 X-OriginalArrivalTime: 03 Mar 2010 19:07:16.0061 (UTC) FILETIME=[BD1E18D0:01CABB04] Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26115 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ddaney@caviumnetworks.com Precedence: bulk X-list: linux-mips The config option CAVIUM_RESERVE32_USE_WIRED_TLB is not supported. Remove the dead code controlled by it. Signed-off-by: David Daney --- arch/mips/cavium-octeon/setup.c | 67 +-------------------------------------- 1 files changed, 1 insertions(+), 66 deletions(-) diff --git a/arch/mips/cavium-octeon/setup.c b/arch/mips/cavium-octeon/setup.c index f58eed6..4ba3277 100644 --- a/arch/mips/cavium-octeon/setup.c +++ b/arch/mips/cavium-octeon/setup.c @@ -186,54 +186,6 @@ void octeon_check_cpu_bist(void) write_octeon_c0_dcacheerr(0); } -#ifdef CONFIG_CAVIUM_RESERVE32_USE_WIRED_TLB -/** - * Called on every core to setup the wired tlb entry needed - * if CONFIG_CAVIUM_RESERVE32_USE_WIRED_TLB is set. - * - */ -static void octeon_hal_setup_per_cpu_reserved32(void *unused) -{ - /* - * The config has selected to wire the reserve32 memory for all - * userspace applications. We need to put a wired TLB entry in for each - * 512MB of reserve32 memory. We only handle double 256MB pages here, - * so reserve32 must be multiple of 512MB. - */ - uint32_t size = CONFIG_CAVIUM_RESERVE32; - uint32_t entrylo0 = - 0x7 | ((octeon_reserve32_memory & ((1ul << 40) - 1)) >> 6); - uint32_t entrylo1 = entrylo0 + (256 << 14); - uint32_t entryhi = (0x80000000UL - (CONFIG_CAVIUM_RESERVE32 << 20)); - while (size >= 512) { -#if 0 - pr_info("CPU%d: Adding double wired TLB entry for 0x%lx\n", - smp_processor_id(), entryhi); -#endif - add_wired_entry(entrylo0, entrylo1, entryhi, PM_256M); - entrylo0 += 512 << 14; - entrylo1 += 512 << 14; - entryhi += 512 << 20; - size -= 512; - } -} -#endif /* CONFIG_CAVIUM_RESERVE32_USE_WIRED_TLB */ - -/** - * Called to release the named block which was used to made sure - * that nobody used the memory for something else during - * init. Now we'll free it so userspace apps can use this - * memory region with bootmem_alloc. - * - * This function is called only once from prom_free_prom_memory(). - */ -void octeon_hal_setup_reserved32(void) -{ -#ifdef CONFIG_CAVIUM_RESERVE32_USE_WIRED_TLB - on_each_cpu(octeon_hal_setup_per_cpu_reserved32, NULL, 0, 1); -#endif -} - /** * Reboot Octeon * @@ -502,25 +454,13 @@ void __init prom_init(void) * memory when it is getting memory from the * bootloader. Later, after the memory allocations are * complete, the reserve32 will be freed. - */ -#ifdef CONFIG_CAVIUM_RESERVE32_USE_WIRED_TLB - if (CONFIG_CAVIUM_RESERVE32 & 0x1ff) - pr_err("CAVIUM_RESERVE32 isn't a multiple of 512MB. " - "This is required if CAVIUM_RESERVE32_USE_WIRED_TLB " - "is set\n"); - else - addr = cvmx_bootmem_phy_named_block_alloc(CONFIG_CAVIUM_RESERVE32 << 20, - 0, 0, 512 << 20, - "CAVIUM_RESERVE32", 0); -#else - /* + * * Allocate memory for RESERVED32 aligned on 2MB boundary. This * is in case we later use hugetlb entries with it. */ addr = cvmx_bootmem_phy_named_block_alloc(CONFIG_CAVIUM_RESERVE32 << 20, 0, 0, 2 << 20, "CAVIUM_RESERVE32", 0); -#endif if (addr < 0) pr_err("Failed to allocate CAVIUM_RESERVE32 memory area\n"); else @@ -822,9 +762,4 @@ void prom_free_prom_memory(void) panic("Unable to request_irq(OCTEON_IRQ_RML)\n"); } #endif - - /* This call is here so that it is performed after any TLB - initializations. It needs to be after these in case the - CONFIG_CAVIUM_RESERVE32_USE_WIRED_TLB option is set */ - octeon_hal_setup_reserved32(); } -- 1.6.6.1 From David.Daney@caviumnetworks.com Wed Mar 3 20:08:20 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 03 Mar 2010 20:08:25 +0100 (CET) Received: from mail3.caviumnetworks.com ([12.108.191.235]:1324 "EHLO mail3.caviumnetworks.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492294Ab0CCTH6 (ORCPT ); Wed, 3 Mar 2010 20:07:58 +0100 Received: from caexch01.caveonetworks.com (Not Verified[192.168.16.9]) by mail3.caviumnetworks.com with MailMarshal (v6,7,2,8378) id ; Wed, 03 Mar 2010 11:08:04 -0800 Received: from caexch01.caveonetworks.com ([192.168.16.9]) by caexch01.caveonetworks.com with Microsoft SMTPSVC(6.0.3790.3959); Wed, 3 Mar 2010 11:07:49 -0800 Received: from dd1.caveonetworks.com ([12.108.191.236]) by caexch01.caveonetworks.com over TLS secured channel with Microsoft SMTPSVC(6.0.3790.3959); Wed, 3 Mar 2010 11:07:48 -0800 Received: from dd1.caveonetworks.com (localhost.localdomain [127.0.0.1]) by dd1.caveonetworks.com (8.14.3/8.14.2) with ESMTP id o23J7iAd027838; Wed, 3 Mar 2010 11:07:44 -0800 Received: (from ddaney@localhost) by dd1.caveonetworks.com (8.14.3/8.14.3/Submit) id o23J7ig2027837; Wed, 3 Mar 2010 11:07:44 -0800 From: David Daney To: linux-mips@linux-mips.org, ralf@linux-mips.org Cc: David Daney Subject: [PATCH] MIPS: Octeon: Remove #if 0 code. Date: Wed, 3 Mar 2010 11:07:43 -0800 Message-Id: <1267643263-27803-1-git-send-email-ddaney@caviumnetworks.com> X-Mailer: git-send-email 1.6.6.1 X-OriginalArrivalTime: 03 Mar 2010 19:07:49.0030 (UTC) FILETIME=[D0C4C460:01CABB04] Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26116 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ddaney@caviumnetworks.com Precedence: bulk X-list: linux-mips Signed-off-by: David Daney --- arch/mips/cavium-octeon/setup.c | 12 ------------ 1 files changed, 0 insertions(+), 12 deletions(-) diff --git a/arch/mips/cavium-octeon/setup.c b/arch/mips/cavium-octeon/setup.c index 4ba3277..a7f8c9d 100644 --- a/arch/mips/cavium-octeon/setup.c +++ b/arch/mips/cavium-octeon/setup.c @@ -246,18 +246,6 @@ static void octeon_halt(void) octeon_kill_core(NULL); } -#if 0 -/** - * Platform time init specifics. - * Returns - */ -void __init plat_time_init(void) -{ - /* Nothing special here, but we are required to have one */ -} - -#endif - /** * Handle all the error condition interrupts that might occur. * -- 1.6.6.1 From lists@nabble.com Wed Mar 3 20:20:23 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 03 Mar 2010 20:20:27 +0100 (CET) Received: from kuber.nabble.com ([216.139.236.158]:41090 "EHLO kuber.nabble.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492379Ab0CCTUX convert rfc822-to-8bit (ORCPT ); Wed, 3 Mar 2010 20:20:23 +0100 Received: from isper.nabble.com ([192.168.236.156]) by kuber.nabble.com with esmtp (Exim 4.63) (envelope-from ) id 1Nmu7K-00086N-75 for linux-mips@linux-mips.org; Wed, 03 Mar 2010 11:20:22 -0800 Message-ID: <27772495.post@talk.nabble.com> Date: Wed, 3 Mar 2010 11:20:22 -0800 (PST) From: Dea_RU To: linux-mips@linux-mips.org Subject: Re: TI AR7 7200 - no boot In-Reply-To: <27768844.post@talk.nabble.com> MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8BIT X-Nabble-From: dryukovz@mail.ru References: <27766331.post@talk.nabble.com> <201003031124.46336.florian@openwrt.org> <27766728.post@talk.nabble.com> <27767861.post@talk.nabble.com> <201003031404.53039.florian@openwrt.org> <27768844.post@talk.nabble.com> Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26117 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: dryukovz@mail.ru Precedence: bulk X-list: linux-mips I show this rtunk left 10 day.... and patch src 2.6.33 from dir root/trunk/target/linux/ar7/patches-2.6.32 to day i see : florian: [ar7] add missing patch to arch/mips/kernel/traps.c to allow ar7 to setup … :confused: ------------------ for 2.6.33 need this modify ?? or not ? arch/mips/include/asm/page.h --------- #define UNCAC_ADDR(addr) ((addr) - PAGE_OFFSET + UNCAC_BASE + PHYS_OFFSET) #define CAC_ADDR(addr) ((addr) - UNCAC_BASE + PAGE_OFFSET - PHYS_OFFSET) --------- ---------------- my console not work correctly - baudrate no standart ?! :( i change port type ---- i change port type in ar7platform.c to: uart_port[0].type = PORT_16550A to uart_port[0].type = PORT_AR7 boot process ttyS0 detected as AR7. Console baudrate not work correct again ....&-( ======= log ================ ............ alg: No test for stdrng (krng) Serial: 8250/16550 driver, 1 ports, IRQ sharing disabled serial8250: ttyS0 at MMIO 0x8610e00 (irq = 15) is a AR7 IdÃIdÃIdÃIdÃIdÃIdÃYdÃYdÃYdÃYdÃYdÃYdÃYdÃYdÃYdÃYdÃYdÃYd ...... ========================== if remart set_termios() in serial_core.c i see all messages ... -- View this message in context: http://old.nabble.com/TI-AR7-7200---no-boot-tp27766331p27772495.html Sent from the linux-mips main mailing list archive at Nabble.com. From f.fainelli@gmail.com Wed Mar 3 21:05:53 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 03 Mar 2010 21:05:56 +0100 (CET) Received: from mail-qy0-f202.google.com ([209.85.221.202]:52360 "EHLO mail-qy0-f202.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492026Ab0CCUFx (ORCPT ); Wed, 3 Mar 2010 21:05:53 +0100 Received: by qyk40 with SMTP id 40so1273979qyk.23 for ; Wed, 03 Mar 2010 12:05:47 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:sender:from:reply-to:to :subject:date:user-agent:cc:references:in-reply-to:mime-version :content-type:content-transfer-encoding:message-id; bh=l+ZVcVJtJhqi9xfps4F4iDRSC9PCLietqkIEA1oGCyE=; b=TiTc6CtzL4fgkVdVKQju1FGBjIiFZUHW9S/jJpjq4uJnOyhyA/c3kqZJaTrMJdF3pc nU2JPLe7ngBJcco2yejF9LR0cmshScBT3VeU0yZmKKLN6oKj9hypHEnUYpu6rqIsujSK OtcQXeUXwlEbb7Qp+BH9+sN/siIcO3WzhRVNM= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=sender:from:reply-to:to:subject:date:user-agent:cc:references :in-reply-to:mime-version:content-type:content-transfer-encoding :message-id; b=TyInW96tBybeizjv6ahcTNUmvKs3UHnn6ZQsH7yrJZBOtWRFapfIqU71DDw4p8kWdE qcck1W3p2IS0/dTVvog+hRfdzJGHLC5VYso8L+5hony9wrXAwUuHq3DekWQ7jjaltOwo tUTPNKLe2krO5HXlJIUjc8fRRv8MFIuIWenb4= Received: by 10.224.81.85 with SMTP id w21mr12302qak.129.1267646747080; Wed, 03 Mar 2010 12:05:47 -0800 (PST) Received: from lenovo.localnet (153.44.69-86.rev.gaoland.net [86.69.44.153]) by mx.google.com with ESMTPS id y67sm632326iby.21.2010.03.03.12.05.44 (version=TLSv1/SSLv3 cipher=RC4-MD5); Wed, 03 Mar 2010 12:05:44 -0800 (PST) From: Florian Fainelli Reply-To: Florian Fainelli To: "Dea_RU" Subject: Re: TI AR7 7200 - no boot Date: Wed, 3 Mar 2010 21:05:29 +0100 User-Agent: KMail/1.12.4 (Linux/2.6.32-trunk-686; KDE/4.3.4; i686; ; ) Cc: linux-mips@linux-mips.org References: <27766331.post@talk.nabble.com> <27768844.post@talk.nabble.com> <27772495.post@talk.nabble.com> In-Reply-To: <27772495.post@talk.nabble.com> MIME-Version: 1.0 Content-Type: multipart/signed; boundary="nextPart4853521.4Xh0pBEFZh"; protocol="application/pgp-signature"; micalg=pgp-sha1 Content-Transfer-Encoding: 7bit Message-Id: <201003032105.43176.florian@openwrt.org> Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26118 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: florian@openwrt.org Precedence: bulk X-list: linux-mips --nextPart4853521.4Xh0pBEFZh Content-Type: Text/Plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable Hi, Le mercredi 3 mars 2010 20:20:22, Dea_RU a =C3=A9crit : > I show this rtunk left 10 day.... >=20 > and patch src 2.6.33 from dir root/trunk/target/linux/ar7/patches-2.6.32 Ok, then why do not you simply use OpenWrt trunk? >=20 > to day i see : >=20 > florian: [ar7] add missing patch to arch/mips/kernel/traps.c to allow ar7 > to setup =E2=80=A6 :confused: >=20 > ------------------ > for 2.6.33 need this modify ?? or not ? >=20 > arch/mips/include/asm/page.h > --------- > #define UNCAC_ADDR(addr) ((addr) - PAGE_OFFSET + UNCAC_BASE + > PHYS_OFFSET) > #define CAC_ADDR(addr) ((addr) - UNCAC_BASE + PAGE_OFFSET - > PHYS_OFFSET) > --------- This is not needed. >=20 >=20 > ---------------- > my console not work correctly - baudrate no standart ?! :( i change port > type >=20 > ---- i change port type in ar7platform.c to: >=20 > uart_port[0].type =3D PORT_16550A > to > uart_port[0].type =3D PORT_AR7 >=20 > boot process ttyS0 detected as AR7. >=20 > Console baudrate not work correct again ....&-( You will also need this patch:=20 https://dev.openwrt.org/browser/trunk/target/linux/ar7/patches-2.6.32/500- serial_kludge.patch >=20 > =3D=3D=3D=3D=3D=3D=3D log =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D > ............ > alg: No test for stdrng (krng) > Serial: 8250/16550 driver, 1 ports, IRQ sharing disabled > serial8250: ttyS0 at MMIO 0x8610e00 (irq =3D 15) is a AR7 > Id=10=C3=81=05=16Id=10=C3=81=05=12Id=10=C3=81=05=12Id=10=C3=81=05=12Id=10= =C3=81=05=12Id=10=C3=81=05=12Yd=10=C3=81=05=12Yd=10=C3=81=05=12Yd=10=C3=81= =05=12Yd=10=C3=81=05=12Yd=10=C3=81=05=12Yd=10=C3=81=05=12Yd=10 > =C3=81=05=12Yd=10=C3=81=05=12Yd=10=C3=81=05=16Yd=10=C3=81=05=16Yd=10=C3= =81=05=16Yd=10 ...... >=20 > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D >=20 > if remart set_termios() in serial_core.c i see all messages ... >=20 =2D-=20 Best regards, Florian Fainelli Email: florian@openwrt.org Web: http://openwrt.org IRC: [florian] on irc.freenode.net =2D------------------------------ --nextPart4853521.4Xh0pBEFZh Content-Type: application/pgp-signature; name=signature.asc Content-Description: This is a digitally signed message part. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.10 (GNU/Linux) iEYEABECAAYFAkuOwQsACgkQlyvkmBGtjybwnwCfdXkE6np8kjIjoklMPhxd9Q5c gdEAn0BWhLgAsicfjAdE/MHELuicepXf =Qww9 -----END PGP SIGNATURE----- --nextPart4853521.4Xh0pBEFZh-- From Yang.Shi@windriver.com Thu Mar 4 03:08:51 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Thu, 04 Mar 2010 03:08:55 +0100 (CET) Received: from mail.windriver.com ([147.11.1.11]:56584 "EHLO mail.windriver.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492133Ab0CDCIv (ORCPT ); Thu, 4 Mar 2010 03:08:51 +0100 Received: from ALA-MAIL03.corp.ad.wrs.com (ala-mail03 [147.11.57.144]) by mail.windriver.com (8.14.3/8.14.3) with ESMTP id o2428dhY026535; Wed, 3 Mar 2010 18:08:39 -0800 (PST) Received: from [128.224.162.222] ([128.224.162.222]) by ALA-MAIL03.corp.ad.wrs.com with Microsoft SMTPSVC(6.0.3790.1830); Wed, 3 Mar 2010 18:08:38 -0800 Message-ID: <4B8F1625.5060709@windriver.com> Date: Thu, 04 Mar 2010 10:08:37 +0800 From: Yang Shi User-Agent: Thunderbird 2.0.0.23 (X11/20090817) MIME-Version: 1.0 To: David Daney CC: ralf@linux-mips.org, f.fainelli@gmail.com, linux-mips@linux-mips.org Subject: Re: [PATCH 1/3] MIPS: Octeon: Remove superfluous on_each_cpu parameter References: <1267605801-5305-1-git-send-email-yang.shi@windriver.com> <4B8EA1FD.5050007@caviumnetworks.com> In-Reply-To: <4B8EA1FD.5050007@caviumnetworks.com> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 8bit X-OriginalArrivalTime: 04 Mar 2010 02:08:38.0868 (UTC) FILETIME=[9AD84940:01CABB3F] Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26119 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: yang.shi@windriver.com Precedence: bulk X-list: linux-mips David Daney 写é“: > On 03/03/2010 12:43 AM, Yang Shi wrote: > >> Now, on_each_cpu just need three parameters, but the on_each_cpu >> still uses four parameters in Octeon's setup.c. So, remove the >> superfluous parameter. >> >> Signed-off-by: Yang Shi >> > > > NAK! We are removing CONFIG_CAVIUM_RESERVE32_USE_WIRED_TLB completely. > > I will send the removal patch soon. > Thanks David. Will apply your patch to remove WIRED_TLB. Regards, Yang > David Daney > > >> --- >> arch/mips/cavium-octeon/setup.c | 2 +- >> 1 files changed, 1 insertions(+), 1 deletions(-) >> >> diff --git a/arch/mips/cavium-octeon/setup.c b/arch/mips/cavium-octeon/setup.c >> index b321d3b..4eaa35f 100644 >> --- a/arch/mips/cavium-octeon/setup.c >> +++ b/arch/mips/cavium-octeon/setup.c >> @@ -230,7 +230,7 @@ static void octeon_hal_setup_per_cpu_reserved32(void *unused) >> void octeon_hal_setup_reserved32(void) >> { >> #ifdef CONFIG_CAVIUM_RESERVE32_USE_WIRED_TLB >> - on_each_cpu(octeon_hal_setup_per_cpu_reserved32, NULL, 0, 1); >> + on_each_cpu(octeon_hal_setup_per_cpu_reserved32, NULL, 1); >> #endif >> } >> >> > > > From r0bertz@gentoo.org Thu Mar 4 04:44:21 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Thu, 04 Mar 2010 04:44:26 +0100 (CET) Received: from smtp.gentoo.org ([140.211.166.183]:46747 "EHLO smtp.gentoo.org" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1490970Ab0CDDoV (ORCPT ); Thu, 4 Mar 2010 04:44:21 +0100 Received: by smtp.gentoo.org (Postfix, from userid 2181) id C5A971B4001; Thu, 4 Mar 2010 03:44:09 +0000 (UTC) Date: Thu, 4 Mar 2010 03:44:09 +0000 From: Zhang Le To: Wu Zhangjin Cc: Ralf Baechle , Dmitry Torokhov , Pavel Machek , "Rafael J. Wysocki" , linux-mips@linux-mips.org, zhangfx@lemote.com Subject: Re: [PATCH v11 0/9] Loongson: YeeLoong: add platform drivers Message-ID: <20100304034409.GA11523@woodpecker.gentoo.org> Mail-Followup-To: Wu Zhangjin , Ralf Baechle , Dmitry Torokhov , Pavel Machek , "Rafael J. Wysocki" , linux-mips@linux-mips.org, zhangfx@lemote.com References: <1264940423.21259.2.camel@falcon> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1264940423.21259.2.camel@falcon> User-Agent: Mutt/1.5.16 (2007-06-09) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26120 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: r0bertz@gentoo.org Precedence: bulk X-list: linux-mips On 20:20 Sun 31 Jan , Wu Zhangjin wrote: > On Sun, 2010-01-31 at 20:15 +0800, Wu Zhangjin wrote: > > From: Wu Zhangjin > > (ooh, just re-generate the patchset with "git format-patch" but forgot > to put the comments, here it is!) > > This patchset adds several platform specific drivers for the YeeLoong > netbook > made by Lemote. > > It is completely based on the v10 revision and with the following > changes: > > O Rebased on the latest linux-queue git tree of Ralf. > O Split the EC revision related handling out of the "input/hotkey" > driver. > O Rewrite the AC & Battery driver with the new power_supply interface > as "Pavel Machek" suggested. > > All of them have been tested again. > > Best Regards, > Wu Zhangjin > > > > > Wu Zhangjin (9): > > MIPS: add subdirectory for platform extension drivers > > Loongson: YeeLoong: add platform driver > > Loongson: YeeLoong: add backlight driver > > Loongson: YeeLoong: add hardware monitoring driver > > Loongson: YeeLoong: add video output driver > > Loongson: YeeLoong: add suspend support > > Loongson: YeeLoong: add input/hotkey driver > > Loongson: YeeLoong: Co-operate with the revisions of EC > > Loongson: YeeLoong: add power_supply based battery driver > > > > arch/mips/include/asm/mach-loongson/ec_kb3310b.h | 188 ++++ > > arch/mips/include/asm/mach-loongson/loongson.h | 6 + > > arch/mips/loongson/common/cmdline.c | 8 + > > arch/mips/loongson/lemote-2f/Makefile | 2 +- > > arch/mips/loongson/lemote-2f/ec_kb3310b.c | 12 +- > > arch/mips/loongson/lemote-2f/ec_kb3310b.h | 188 ---- > > arch/mips/loongson/lemote-2f/platform.c | 39 + > > arch/mips/loongson/lemote-2f/pm.c | 4 +- > > arch/mips/loongson/lemote-2f/reset.c | 2 +- > > drivers/platform/Kconfig | 4 + > > drivers/platform/Makefile | 1 + > > drivers/platform/mips/Kconfig | 32 + > > drivers/platform/mips/Makefile | 5 + > > drivers/platform/mips/yeeloong_laptop.c | 1192 ++++++++++++++++++++++ > > 14 files changed, 1483 insertions(+), 200 deletions(-) > > create mode 100644 arch/mips/include/asm/mach-loongson/ec_kb3310b.h > > delete mode 100644 arch/mips/loongson/lemote-2f/ec_kb3310b.h > > create mode 100644 arch/mips/loongson/lemote-2f/platform.c > > create mode 100644 drivers/platform/mips/Kconfig > > create mode 100644 drivers/platform/mips/Makefile > > create mode 100644 drivers/platform/mips/yeeloong_laptop.c > > What's the status of these patches? Ralf, any chance for 2.6.34? Thanks! Zhang, Le From yang.shi@windriver.com Thu Mar 4 10:39:44 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Thu, 04 Mar 2010 10:39:47 +0100 (CET) Received: from mail.windriver.com ([147.11.1.11]:49564 "EHLO mail.windriver.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491187Ab0CDJjo (ORCPT ); Thu, 4 Mar 2010 10:39:44 +0100 Received: from localhost.localdomain (pek-lpgbuild1.wrs.com [128.224.153.29]) by mail.windriver.com (8.14.3/8.14.3) with ESMTP id o249dXVZ024419; Thu, 4 Mar 2010 01:39:36 -0800 (PST) From: Yang Shi To: ralf@linux-mips.org Cc: linux-mips@linux-mips.org Subject: [PATCH] MIPS: Protect current_cpu_data with preempt disable in delay() Date: Thu, 4 Mar 2010 17:39:33 +0800 Message-Id: <1267695573-27360-1-git-send-email-yang.shi@windriver.com> X-Mailer: git-send-email 1.6.0.4 Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26121 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: yang.shi@windriver.com Precedence: bulk X-list: linux-mips During machine restart with reboot command, get the following bug info: BUG: using smp_processor_id() in preemptible [00000000] code: reboot/1989 caller is __udelay+0x14/0x70 Call Trace: [] dump_stack+0x8/0x34 [] debug_smp_processor_id+0xf4/0x110 [] __udelay+0x14/0x70 [] md_notify_reboot+0x12c/0x148 [] notifier_call_chain+0x64/0xc8 [] __blocking_notifier_call_chain+0x64/0xc0 [] kernel_restart_prepare+0x1c/0x38 [] kernel_restart+0x14/0x50 [] SyS_reboot+0x10c/0x1f0 [] handle_sysn32+0x44/0x84 The root cause is that current_cpu_data is accessed in preemptible context, so protect it with preempt_disable/preempt_enable pair in delay(). Signed-off-by: Yang Shi --- arch/mips/lib/delay.c | 6 +++++- 1 files changed, 5 insertions(+), 1 deletions(-) diff --git a/arch/mips/lib/delay.c b/arch/mips/lib/delay.c index 6b3b1de..dc38064 100644 --- a/arch/mips/lib/delay.c +++ b/arch/mips/lib/delay.c @@ -41,7 +41,11 @@ EXPORT_SYMBOL(__delay); void __udelay(unsigned long us) { - unsigned int lpj = current_cpu_data.udelay_val; + unsigned int lpj; + + preempt_disable(); + lpj = current_cpu_data.udelay_val; + preempt_enable(); __delay((us * 0x000010c7ull * HZ * lpj) >> 32); } -- 1.6.3.3 From muvarov@gmail.com Thu Mar 4 12:35:44 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Thu, 04 Mar 2010 12:35:47 +0100 (CET) Received: from mail-gw0-f49.google.com ([74.125.83.49]:46412 "EHLO mail-gw0-f49.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1490999Ab0CDLfo convert rfc822-to-8bit (ORCPT ); Thu, 4 Mar 2010 12:35:44 +0100 Received: by gwj21 with SMTP id 21so1207305gwj.36 for ; Thu, 04 Mar 2010 03:35:37 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:in-reply-to:references :date:message-id:subject:from:to:cc:content-type :content-transfer-encoding; bh=n6BWJSZcHse9VR/zabGgrG6JXxxZ2rfAv/ox4xolTNQ=; b=oanIZCHFDUpKSCP2vr/w8qa7RAN4BPtH9/AvFHzoS7FPfDFmm/MpacOBBoKqWVnlnQ MW6tYN5gRv6sWV4iQ1Stpxb62D+Q333yK3jXqr+4C8P5gVKUYMeiq99ZL9TNMHq2Z8sT X7JFX+Pqa+IvcgkKVxSN7JyGlB1PIfTLY1Zqg= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type:content-transfer-encoding; b=aZqQ++K1j9vp/CgSqZumIId19i6V835JE9t7f2oKjh7eyjKe6pVWwIEyDB8iPYb8GD kqzSZapna0o7m3QdnUex2qS6yuYtVOwOSP3jI6V4J1kkHdx3Mx1jufnyEGKC+hRpsEaE u2OWTnsD4hAI58c5o0sfwgUXEFVHFKPQ0vPyA= MIME-Version: 1.0 Received: by 10.150.81.5 with SMTP id e5mr2390216ybb.158.1267702537387; Thu, 04 Mar 2010 03:35:37 -0800 (PST) In-Reply-To: <4B8E9E0C.4050809@caviumnetworks.com> References: <20100303110527.11233.20400.stgit@muvarov> <4B8E9E0C.4050809@caviumnetworks.com> Date: Thu, 4 Mar 2010 14:35:37 +0300 Message-ID: <572af9171003040335w16434350y60a9f39676dbfa3d@mail.gmail.com> Subject: Re: [PATCH 1/2] MIPS kexec,kdump support From: Maxim Uvarov To: David Daney Cc: linux-mips@linux-mips.org, kexec@lists.infradead.org, horms@verge.net.au, ralf@linux-mips.org Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8BIT Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26122 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: muvarov@gmail.com Precedence: bulk X-list: linux-mips 2010/3/3 David Daney : > On 03/03/2010 03:05 AM, Maxim Uvarov wrote: >> >> Hello folks, >> >> Please find here MIPS crash and kdump patches. >> This is patch set of 3 patches: >> 1. generic MIPS changes (kernel); >> 2. MIPS Cavium Octeon board kexec/kdump code (kernel); >> 3. Kexec user space MIPS changes. >> >> Patches were tested on the latest linux-mips@ git kernel and the latest >> kexec-tools git on Cavium Octeon 50xx board. >> >> I also made the same code working on RMI XLR/XLS boards for both >> mips32 and mips64 kernels. >> >> Best regards, >> Maxim Uvarov. >> >> >> --- >> >>  arch/mips/Kconfig                  |   24 ++++++++++ > > [...] >> >> >> Signed-off-by: Maxim Uvarov >> > > That Signed-off-by: needs to be just above the '---' not at the end. > > David Daney > Thanks David, I corrected stgit email template. I think I don't need resend patches with only this change. -- Best regards, Maxim Uvarov From muvarov@gmail.com Thu Mar 4 12:58:42 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Thu, 04 Mar 2010 12:58:45 +0100 (CET) Received: from mail-yx0-f179.google.com ([209.85.210.179]:33539 "EHLO mail-yx0-f179.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491851Ab0CDL6m convert rfc822-to-8bit (ORCPT ); Thu, 4 Mar 2010 12:58:42 +0100 Received: by yxe9 with SMTP id 9so1225765yxe.22 for ; Thu, 04 Mar 2010 03:58:35 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:in-reply-to:references :date:message-id:subject:from:to:cc:content-type :content-transfer-encoding; bh=1tRr09/lk/etBJnnzP87tj6O0jjThngLIZg2F+TntPs=; b=rbQiAUTRSdC7763c7f1eZbx0KqeIWtiYpPSn3QZ23oSrsxwcTSH8t/3DrXyBtFEca0 EYoVscgQbdZFEMX8PHJH5Yc6hRGPmhwikoUUg+JWeftLqMSBV3ZuYqPRuVtOx6y2juQL eA3yUEuqrVl8gmnb3+ZfbEzPpE+wh8bXbz0BE= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type:content-transfer-encoding; b=E6idz5oS3qf1TlsbSlkUE0HGoDr5gX/YwLnMm3w4HweVWX7a1g5vAYbCy1L2T3Y3cl /4BeNxMOI2SHL7N8Ythj/7vpzu5hVNdB7sKarjvlGnNdvY06dCN+9SSply2JQwNzx0yR bcScu6cRlwFHaAMOSA6DKbuIEKzVCRyICaMiw= MIME-Version: 1.0 Received: by 10.150.166.15 with SMTP id o15mr2357347ybe.306.1267703914872; Thu, 04 Mar 2010 03:58:34 -0800 (PST) In-Reply-To: References: <20100303110527.11233.20400.stgit@muvarov> Date: Thu, 4 Mar 2010 14:58:34 +0300 Message-ID: <572af9171003040358n25bc2462yc8c16a9acdac70db@mail.gmail.com> Subject: Re: [PATCH 1/2] MIPS kexec,kdump support From: Maxim Uvarov To: "wilbur.chan" Cc: linux-mips@linux-mips.org, kexec@lists.infradead.org, horms@verge.net.au, ralf@linux-mips.org Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8BIT Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26123 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: muvarov@gmail.com Precedence: bulk X-list: linux-mips 2010/3/3 wilbur.chan : > 2010/3/3 Maxim Uvarov : >> Hello folks, >> >> Please find here MIPS crash and kdump patches. >> This is patch set of 3 patches: >> 1. generic MIPS changes (kernel); >> 2. MIPS Cavium Octeon board kexec/kdump code (kernel); >> 3. Kexec user space MIPS changes. >> >> Patches were tested on the latest linux-mips@ git kernel and the latest >> kexec-tools git on Cavium Octeon 50xx board. >> >> I also made the same code working on RMI XLR/XLS boards for both >> mips32 and mips64 kernels. >> >> Best regards, >> Maxim Uvarov. >> > >> Signed-off-by: Maxim Uvarov >> >> >> > > Hi, Maxim > > In XLR series , > > 1)How to protect  boardinfo and pass it to second kernel ? > It is very simple. I just looked at physical addresses where boardinfo is and protect this region. For xls simple add additional exclude region to setup.c static struct boot_mem_map_exclude_region static_exclude_regions[] = { + [1] = {0xc000000, 0xd000000 }, /*Bootloader stuctures*/ To pass it to second kernel copy pointer to original psb_info to static variable in beginning of prom_init() and provide this pointer as kexec_args[3]. kexec_args[0] is argc on XLS kexec_args[1] is argv on XLS > 2)If all cpus jumped to same entry point , did you change head.s, if so , how ? > Please take a look how other platforms do this. You need write .macro kernel_entry_setup which should be located in somewhere is include/asm-mips/mach-rmi. Then you write this macro it will be executed before kernel_entry. So after kexec all cpus jump to this entry point and after they you can do all things what you want: - make cpu0 boot new kernel and other cpus go to boot slaves procedure; - take first cpu and make it boot, and all others go to boot slaves procedure; - boot only cpu0 and infinite loop others cpus (you might want to do this for kdump) > > > Thank you! > -- Best regards, Maxim Uvarov From David.Daney@caviumnetworks.com Thu Mar 4 19:24:52 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Thu, 04 Mar 2010 19:24:55 +0100 (CET) Received: from mail3.caviumnetworks.com ([12.108.191.235]:5396 "EHLO mail3.caviumnetworks.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492267Ab0CDSYw (ORCPT ); Thu, 4 Mar 2010 19:24:52 +0100 Received: from caexch01.caveonetworks.com (Not Verified[192.168.16.9]) by mail3.caviumnetworks.com with MailMarshal (v6,7,2,8378) id ; Thu, 04 Mar 2010 10:24:58 -0800 Received: from caexch01.caveonetworks.com ([192.168.16.9]) by caexch01.caveonetworks.com with Microsoft SMTPSVC(6.0.3790.3959); Thu, 4 Mar 2010 10:23:53 -0800 Received: from dd1.caveonetworks.com ([12.108.191.236]) by caexch01.caveonetworks.com over TLS secured channel with Microsoft SMTPSVC(6.0.3790.3959); Thu, 4 Mar 2010 10:23:53 -0800 Message-ID: <4B8FFAB3.1090409@caviumnetworks.com> Date: Thu, 04 Mar 2010 10:23:47 -0800 From: David Daney User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.8) Gecko/20100301 Fedora/3.0.3-1.fc12 Thunderbird/3.0.3 MIME-Version: 1.0 To: Yang Shi , ralf@linux-mips.org CC: linux-mips@linux-mips.org Subject: Re: [PATCH] MIPS: Protect current_cpu_data with preempt disable in delay() References: <1267695573-27360-1-git-send-email-yang.shi@windriver.com> In-Reply-To: <1267695573-27360-1-git-send-email-yang.shi@windriver.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-OriginalArrivalTime: 04 Mar 2010 18:23:53.0236 (UTC) FILETIME=[D8204540:01CABBC7] Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26124 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ddaney@caviumnetworks.com Precedence: bulk X-list: linux-mips On 03/04/2010 01:39 AM, Yang Shi wrote: > During machine restart with reboot command, get the following > bug info: > > BUG: using smp_processor_id() in preemptible [00000000] code: reboot/1989 > caller is __udelay+0x14/0x70 > Call Trace: > [] dump_stack+0x8/0x34 > [] debug_smp_processor_id+0xf4/0x110 > [] __udelay+0x14/0x70 > [] md_notify_reboot+0x12c/0x148 > [] notifier_call_chain+0x64/0xc8 > [] __blocking_notifier_call_chain+0x64/0xc0 > [] kernel_restart_prepare+0x1c/0x38 > [] kernel_restart+0x14/0x50 > [] SyS_reboot+0x10c/0x1f0 > [] handle_sysn32+0x44/0x84 > > The root cause is that current_cpu_data is accessed in preemptible > context, so protect it with preempt_disable/preempt_enable pair > in delay(). > > Signed-off-by: Yang Shi > --- > arch/mips/lib/delay.c | 6 +++++- > 1 files changed, 5 insertions(+), 1 deletions(-) > > diff --git a/arch/mips/lib/delay.c b/arch/mips/lib/delay.c > index 6b3b1de..dc38064 100644 > --- a/arch/mips/lib/delay.c > +++ b/arch/mips/lib/delay.c > @@ -41,7 +41,11 @@ EXPORT_SYMBOL(__delay); > > void __udelay(unsigned long us) > { > - unsigned int lpj = current_cpu_data.udelay_val; > + unsigned int lpj; > + > + preempt_disable(); > + lpj = current_cpu_data.udelay_val; > + preempt_enable(); > > __delay((us * 0x000010c7ull * HZ * lpj)>> 32); > } This doesn't seem like the best approach. Perhaps we should either use raw_current_cpu_data and no preempt_disable(), or if we are concerned about migrating to a CPU with a different lpj value, move the preempt_enable after the call to __delay(). David Daney From Yang.Shi@windriver.com Fri Mar 5 02:10:34 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Fri, 05 Mar 2010 02:10:42 +0100 (CET) Received: from mail.windriver.com ([147.11.1.11]:42921 "EHLO mail.windriver.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492015Ab0CEBKe (ORCPT ); Fri, 5 Mar 2010 02:10:34 +0100 Received: from ALA-MAIL03.corp.ad.wrs.com (ala-mail03 [147.11.57.144]) by mail.windriver.com (8.14.3/8.14.3) with ESMTP id o251ANXM020018; Thu, 4 Mar 2010 17:10:23 -0800 (PST) Received: from [128.224.162.222] ([128.224.162.222]) by ALA-MAIL03.corp.ad.wrs.com with Microsoft SMTPSVC(6.0.3790.1830); Thu, 4 Mar 2010 17:10:22 -0800 Message-ID: <4B9059FE.3090801@windriver.com> Date: Fri, 05 Mar 2010 09:10:22 +0800 From: Yang Shi User-Agent: Thunderbird 2.0.0.23 (X11/20090817) MIME-Version: 1.0 To: David Daney CC: ralf@linux-mips.org, linux-mips@linux-mips.org Subject: Re: [PATCH] MIPS: Protect current_cpu_data with preempt disable in delay() References: <1267695573-27360-1-git-send-email-yang.shi@windriver.com> <4B8FFAB3.1090409@caviumnetworks.com> In-Reply-To: <4B8FFAB3.1090409@caviumnetworks.com> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 8bit X-OriginalArrivalTime: 05 Mar 2010 01:10:22.0945 (UTC) FILETIME=[A1868510:01CABC00] Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26125 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: yang.shi@windriver.com Precedence: bulk X-list: linux-mips David Daney 写é“: > On 03/04/2010 01:39 AM, Yang Shi wrote: > >> During machine restart with reboot command, get the following >> bug info: >> >> BUG: using smp_processor_id() in preemptible [00000000] code: reboot/1989 >> caller is __udelay+0x14/0x70 >> Call Trace: >> [] dump_stack+0x8/0x34 >> [] debug_smp_processor_id+0xf4/0x110 >> [] __udelay+0x14/0x70 >> [] md_notify_reboot+0x12c/0x148 >> [] notifier_call_chain+0x64/0xc8 >> [] __blocking_notifier_call_chain+0x64/0xc0 >> [] kernel_restart_prepare+0x1c/0x38 >> [] kernel_restart+0x14/0x50 >> [] SyS_reboot+0x10c/0x1f0 >> [] handle_sysn32+0x44/0x84 >> >> The root cause is that current_cpu_data is accessed in preemptible >> context, so protect it with preempt_disable/preempt_enable pair >> in delay(). >> >> Signed-off-by: Yang Shi >> --- >> arch/mips/lib/delay.c | 6 +++++- >> 1 files changed, 5 insertions(+), 1 deletions(-) >> >> diff --git a/arch/mips/lib/delay.c b/arch/mips/lib/delay.c >> index 6b3b1de..dc38064 100644 >> --- a/arch/mips/lib/delay.c >> +++ b/arch/mips/lib/delay.c >> @@ -41,7 +41,11 @@ EXPORT_SYMBOL(__delay); >> >> void __udelay(unsigned long us) >> { >> - unsigned int lpj = current_cpu_data.udelay_val; >> + unsigned int lpj; >> + >> + preempt_disable(); >> + lpj = current_cpu_data.udelay_val; >> + preempt_enable(); >> >> __delay((us * 0x000010c7ull * HZ * lpj)>> 32); >> } >> > > This doesn't seem like the best approach. > > Perhaps we should either use raw_current_cpu_data and no > preempt_disable(), or if we are concerned about migrating to a CPU with > a different lpj value, move the preempt_enable after the call to __delay(). > Thanks David. Yes, actually I also has this concern, so this patch is just a rough fix. And I tried raw_current_cpu_data as well, but I'm not sure if it's safe or not. Another proposal is to change cpu_data and current_cpu_data to per CPU variables, of course this is a big change. Regards, Yang > David Daney > > From yang.shi@windriver.com Fri Mar 5 08:09:00 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Fri, 05 Mar 2010 08:09:08 +0100 (CET) Received: from mail.windriver.com ([147.11.1.11]:54870 "EHLO mail.windriver.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491051Ab0CEHJA (ORCPT ); Fri, 5 Mar 2010 08:09:00 +0100 Received: from localhost.localdomain (pek-lpgbuild1.wrs.com [128.224.153.29]) by mail.windriver.com (8.14.3/8.14.3) with ESMTP id o2578GrL004033; Thu, 4 Mar 2010 23:08:17 -0800 (PST) From: Yang Shi To: ddaney@caviumnetworks.com, ben-linux@fluff.org, khali@linux-fr.org, ralf@linux-mips.org Cc: linux-mips@linux-mips.org, linux-i2c@vger.kernel.org Subject: [PATCH] MIPS: Octeon: Register EEPROM device on the I2C bus Date: Fri, 5 Mar 2010 15:08:15 +0800 Message-Id: <1267772895-25409-1-git-send-email-yang.shi@windriver.com> X-Mailer: git-send-email 1.6.0.4 Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26126 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: yang.shi@windriver.com Precedence: bulk X-list: linux-mips An EEPROM resides on 0x50 of I2C bus on CN56xx/57xx board, register this device. Signed-off-by: Yang Shi --- arch/mips/cavium-octeon/octeon-platform.c | 3 +++ 1 files changed, 3 insertions(+), 0 deletions(-) diff --git a/arch/mips/cavium-octeon/octeon-platform.c b/arch/mips/cavium-octeon/octeon-platform.c index 62ac30e..5bfa513 100644 --- a/arch/mips/cavium-octeon/octeon-platform.c +++ b/arch/mips/cavium-octeon/octeon-platform.c @@ -164,6 +164,9 @@ static struct i2c_board_info __initdata octeon_i2c_devices[] = { { I2C_BOARD_INFO("ds1337", 0x68), }, + { + I2C_BOARD_INFO("eeprom", 0x50), + }, }; static int __init octeon_i2c_devices_init(void) -- 1.6.3.3 From wsa@pengutronix.de Fri Mar 5 08:11:59 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Fri, 05 Mar 2010 08:12:07 +0100 (CET) Received: from metis.ext.pengutronix.de ([92.198.50.35]:55187 "EHLO metis.ext.pengutronix.de" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491948Ab0CEHL7 (ORCPT ); Fri, 5 Mar 2010 08:11:59 +0100 Received: from octopus.hi.pengutronix.de ([2001:6f8:1178:2:215:17ff:fe12:23b0]) by metis.ext.pengutronix.de with esmtp (Exim 4.69) (envelope-from ) id 1NnRhD-00081d-NA; Fri, 05 Mar 2010 08:11:39 +0100 Received: from wsa by octopus.hi.pengutronix.de with local (Exim 4.69) (envelope-from ) id 1NnRh4-0007bz-3U; Fri, 05 Mar 2010 08:11:30 +0100 Date: Fri, 5 Mar 2010 08:11:30 +0100 From: Wolfram Sang To: Yang Shi Cc: ddaney@caviumnetworks.com, ben-linux@fluff.org, khali@linux-fr.org, ralf@linux-mips.org, linux-mips@linux-mips.org, linux-i2c@vger.kernel.org Subject: Re: [PATCH] MIPS: Octeon: Register EEPROM device on the I2C bus Message-ID: <20100305071130.GB21925@pengutronix.de> References: <1267772895-25409-1-git-send-email-yang.shi@windriver.com> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="UHN/qo2QbUvPLonB" Content-Disposition: inline In-Reply-To: <1267772895-25409-1-git-send-email-yang.shi@windriver.com> User-Agent: Mutt/1.5.18 (2008-05-17) X-SA-Exim-Connect-IP: 2001:6f8:1178:2:215:17ff:fe12:23b0 X-SA-Exim-Mail-From: wsa@pengutronix.de X-SA-Exim-Scanned: No (on metis.ext.pengutronix.de); SAEximRunCond expanded to false X-PTX-Original-Recipient: linux-mips@linux-mips.org Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26127 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: w.sang@pengutronix.de Precedence: bulk X-list: linux-mips --UHN/qo2QbUvPLonB Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable > + { > + I2C_BOARD_INFO("eeprom", 0x50), > + }, Is the use of 'eeprom' instead of 'at24' intentional? Regards --=20 Pengutronix e.K. | Wolfram Sang | Industrial Linux Solutions | http://www.pengutronix.de/ | --UHN/qo2QbUvPLonB Content-Type: application/pgp-signature; name="signature.asc" Content-Description: Digital signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) iEYEARECAAYFAkuQrqIACgkQD27XaX1/VRvOxACbB3tjlz8LFvaToAhx4vOJp0La +PYAoMMcDlXWyahLNZ7BEnriAIdm5MrH =vrmz -----END PGP SIGNATURE----- --UHN/qo2QbUvPLonB-- From Yang.Shi@windriver.com Fri Mar 5 08:31:42 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Fri, 05 Mar 2010 08:32:07 +0100 (CET) Received: from mail.windriver.com ([147.11.1.11]:58003 "EHLO mail.windriver.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491948Ab0CEHbm (ORCPT ); Fri, 5 Mar 2010 08:31:42 +0100 Received: from ALA-MAIL03.corp.ad.wrs.com (ala-mail03 [147.11.57.144]) by mail.windriver.com (8.14.3/8.14.3) with ESMTP id o257VFeI007061; Thu, 4 Mar 2010 23:31:15 -0800 (PST) Received: from [128.224.161.163] ([128.224.161.163]) by ALA-MAIL03.corp.ad.wrs.com with Microsoft SMTPSVC(6.0.3790.1830); Thu, 4 Mar 2010 23:31:14 -0800 Message-ID: <4B90B341.9000601@windriver.com> Date: Fri, 05 Mar 2010 15:31:13 +0800 From: Yang Shi User-Agent: Thunderbird 2.0.0.23 (X11/20090817) MIME-Version: 1.0 To: Wolfram Sang CC: ddaney@caviumnetworks.com, ben-linux@fluff.org, khali@linux-fr.org, ralf@linux-mips.org, linux-mips@linux-mips.org, linux-i2c@vger.kernel.org Subject: Re: [PATCH] MIPS: Octeon: Register EEPROM device on the I2C bus References: <1267772895-25409-1-git-send-email-yang.shi@windriver.com> <20100305071130.GB21925@pengutronix.de> In-Reply-To: <20100305071130.GB21925@pengutronix.de> Content-Type: text/plain; charset=GB2312 Content-Transfer-Encoding: 8bit X-OriginalArrivalTime: 05 Mar 2010 07:31:14.0899 (UTC) FILETIME=[D659FA30:01CABC35] Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26128 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: yang.shi@windriver.com Precedence: bulk X-list: linux-mips Wolfram Sang дµÀ: >> + { >> + I2C_BOARD_INFO("eeprom", 0x50), >> + }, >> > > Is the use of 'eeprom' instead of 'at24' intentional? > Unfortunately, at24 driver can't work on this board, I must use legacy eeprom. My config: #CONFIG_EEPROM_AT24 is not set CONFIG_EEPROM_LEGACY=y Regards, Yang > Regards > > From wsa@pengutronix.de Fri Mar 5 08:42:14 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Fri, 05 Mar 2010 08:42:21 +0100 (CET) Received: from metis.ext.pengutronix.de ([92.198.50.35]:43923 "EHLO metis.ext.pengutronix.de" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491996Ab0CEHmO (ORCPT ); Fri, 5 Mar 2010 08:42:14 +0100 Received: from octopus.hi.pengutronix.de ([2001:6f8:1178:2:215:17ff:fe12:23b0]) by metis.ext.pengutronix.de with esmtp (Exim 4.69) (envelope-from ) id 1NnSAW-0000yI-Jh; Fri, 05 Mar 2010 08:41:56 +0100 Received: from wsa by octopus.hi.pengutronix.de with local (Exim 4.69) (envelope-from ) id 1NnSAV-0001lv-VB; Fri, 05 Mar 2010 08:41:55 +0100 Date: Fri, 5 Mar 2010 08:41:55 +0100 From: Wolfram Sang To: Yang Shi Cc: ddaney@caviumnetworks.com, ben-linux@fluff.org, khali@linux-fr.org, ralf@linux-mips.org, linux-mips@linux-mips.org, linux-i2c@vger.kernel.org Subject: Re: [PATCH] MIPS: Octeon: Register EEPROM device on the I2C bus Message-ID: <20100305074155.GD21925@pengutronix.de> References: <1267772895-25409-1-git-send-email-yang.shi@windriver.com> <20100305071130.GB21925@pengutronix.de> <4B90B341.9000601@windriver.com> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="rqzD5py0kzyFAOWN" Content-Disposition: inline In-Reply-To: <4B90B341.9000601@windriver.com> User-Agent: Mutt/1.5.18 (2008-05-17) X-SA-Exim-Connect-IP: 2001:6f8:1178:2:215:17ff:fe12:23b0 X-SA-Exim-Mail-From: wsa@pengutronix.de X-SA-Exim-Scanned: No (on metis.ext.pengutronix.de); SAEximRunCond expanded to false X-PTX-Original-Recipient: linux-mips@linux-mips.org Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26129 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: w.sang@pengutronix.de Precedence: bulk X-list: linux-mips --rqzD5py0kzyFAOWN Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable > > Is the use of 'eeprom' instead of 'at24' intentional? > > =20 >=20 > Unfortunately, at24 driver can't work on this board, I must use legacy > eeprom. Well, you are of course free to choose here :) I'd just be interested if there is a software limitation which prevents you= from using AT24. Because, it _should_ work with all kind of eeproms the legacy d= river deals with. Otherwise it is probably a bug which needs to be fixed. Regards, Wolfram --=20 Pengutronix e.K. | Wolfram Sang | Industrial Linux Solutions | http://www.pengutronix.de/ | --rqzD5py0kzyFAOWN Content-Type: application/pgp-signature; name="signature.asc" Content-Description: Digital signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) iEYEARECAAYFAkuQtcMACgkQD27XaX1/VRvudwCcDh5SKi1uF8McyFVJphGbTzrw DyQAn3VWHIc3yopV++1OvlM+2Ixe2e1+ =YLUF -----END PGP SIGNATURE----- --rqzD5py0kzyFAOWN-- From Yang.Shi@windriver.com Fri Mar 5 08:54:14 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Fri, 05 Mar 2010 08:54:21 +0100 (CET) Received: from mail.windriver.com ([147.11.1.11]:60608 "EHLO mail.windriver.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491932Ab0CEHyO (ORCPT ); Fri, 5 Mar 2010 08:54:14 +0100 Received: from ALA-MAIL03.corp.ad.wrs.com (ala-mail03 [147.11.57.144]) by mail.windriver.com (8.14.3/8.14.3) with ESMTP id o257rkUQ009521; Thu, 4 Mar 2010 23:53:46 -0800 (PST) Received: from [128.224.161.163] ([128.224.161.163]) by ALA-MAIL03.corp.ad.wrs.com with Microsoft SMTPSVC(6.0.3790.1830); Thu, 4 Mar 2010 23:53:45 -0800 Message-ID: <4B90B888.6060005@windriver.com> Date: Fri, 05 Mar 2010 15:53:44 +0800 From: Yang Shi User-Agent: Thunderbird 2.0.0.23 (X11/20090817) MIME-Version: 1.0 To: Wolfram Sang CC: ddaney@caviumnetworks.com, ben-linux@fluff.org, khali@linux-fr.org, ralf@linux-mips.org, linux-mips@linux-mips.org, linux-i2c@vger.kernel.org Subject: Re: [PATCH] MIPS: Octeon: Register EEPROM device on the I2C bus References: <1267772895-25409-1-git-send-email-yang.shi@windriver.com> <20100305071130.GB21925@pengutronix.de> <4B90B341.9000601@windriver.com> <20100305074155.GD21925@pengutronix.de> In-Reply-To: <20100305074155.GD21925@pengutronix.de> Content-Type: text/plain; charset=GB2312 Content-Transfer-Encoding: 8bit X-OriginalArrivalTime: 05 Mar 2010 07:53:45.0916 (UTC) FILETIME=[FB9ECFC0:01CABC38] Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26130 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: yang.shi@windriver.com Precedence: bulk X-list: linux-mips Wolfram Sang дµÀ: >>> Is the use of 'eeprom' instead of 'at24' intentional? >>> >>> >> Unfortunately, at24 driver can't work on this board, I must use legacy >> eeprom. >> > > Well, you are of course free to choose here :) > > I'd just be interested if there is a software limitation which prevents you from > using AT24. Because, it _should_ work with all kind of eeproms the legacy driver > deals with. Otherwise it is probably a bug which needs to be fixed. > Thanks to point out this. Let me take a look at this. Regards, Yang > Regards, > > Wolfram > > From khali@linux-fr.org Fri Mar 5 09:50:43 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Fri, 05 Mar 2010 09:50:46 +0100 (CET) Received: from bamako.nerim.net ([62.4.17.28]:61502 "EHLO bamako.nerim.net" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492059Ab0CEIun convert rfc822-to-8bit (ORCPT ); Fri, 5 Mar 2010 09:50:43 +0100 Received: from localhost (localhost [127.0.0.1]) by bamako.nerim.net (Postfix) with ESMTP id 40AFE39DC6B; Fri, 5 Mar 2010 09:50:41 +0100 (CET) X-Virus-Scanned: amavisd-new at nerim.net Received: from bamako.nerim.net ([127.0.0.1]) by localhost (bamako.nerim.net [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id tGUr8aZPQJ2S; Fri, 5 Mar 2010 09:50:40 +0100 (CET) Received: from hyperion.delvare (jdelvare.pck.nerim.net [62.212.121.182]) by bamako.nerim.net (Postfix) with ESMTP id ED17139DE4E; Fri, 5 Mar 2010 09:50:39 +0100 (CET) Date: Fri, 5 Mar 2010 09:50:40 +0100 From: Jean Delvare To: Yang Shi Cc: Wolfram Sang , ddaney@caviumnetworks.com, ben-linux@fluff.org, ralf@linux-mips.org, linux-mips@linux-mips.org, linux-i2c@vger.kernel.org, Konstantin Lazarev Subject: Re: [PATCH] MIPS: Octeon: Register EEPROM device on the I2C bus Message-ID: <20100305095040.6ab4612c@hyperion.delvare> In-Reply-To: <4B90B888.6060005@windriver.com> References: <1267772895-25409-1-git-send-email-yang.shi@windriver.com> <20100305071130.GB21925@pengutronix.de> <4B90B341.9000601@windriver.com> <20100305074155.GD21925@pengutronix.de> <4B90B888.6060005@windriver.com> X-Mailer: Claws Mail 3.5.0 (GTK+ 2.14.4; i586-suse-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8BIT Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26131 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: khali@linux-fr.org Precedence: bulk X-list: linux-mips Hi Yang, Wolfram, On Fri, 05 Mar 2010 15:53:44 +0800, Yang Shi wrote: > Wolfram Sang 写é“: > >>> Is the use of 'eeprom' instead of 'at24' intentional? > >>> > >>> > >> Unfortunately, at24 driver can't work on this board, I must use legacy > >> eeprom. > >> > > > > Well, you are of course free to choose here :) > > > > I'd just be interested if there is a software limitation which prevents you from > > using AT24. Because, it _should_ work with all kind of eeproms the legacy driver > > deals with. Otherwise it is probably a bug which needs to be fixed. > > > > Thanks to point out this. Let me take a look at this. One limitation of the at24 driver is that it needs the underlying controller to support either raw I2C access or at least I2C block transactions. Konstantin Lazarev complained about that one month ago already. I am currently working on improving the at24 driver so that it falls back to byte transactions when block transactions are not available. I might also add word transaction support (as the eeprom driver has) as it is often the best performance/compatibility trade-off. I'll post the patch when I'm done. I'm not yet sure what will happen to the legacy eeprom driver in the long run, but I would prefer new designs to not rely on it. -- Jean Delvare From ml.lawnick@gmx.de Fri Mar 5 11:09:45 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Fri, 05 Mar 2010 11:09:49 +0100 (CET) Received: from mail.gmx.net ([213.165.64.20]:49858 "HELO mail.gmx.net" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with SMTP id S1491083Ab0CEKJo (ORCPT ); Fri, 5 Mar 2010 11:09:44 +0100 Received: (qmail invoked by alias); 05 Mar 2010 10:09:38 -0000 Received: from unknown (EHLO [192.100.130.239]) [192.100.130.239] by mail.gmx.net (mp005) with SMTP; 05 Mar 2010 11:09:38 +0100 X-Authenticated: #54578410 X-Provags-ID: V01U2FsdGVkX18vz2lGKR0vGyqROMmgbGagitm/AS23IlHh/bHnVG m+fbdYmS9wX8/r Message-ID: <4B90D85E.6040308@gmx.de> Date: Fri, 05 Mar 2010 11:09:34 +0100 From: Michael Lawnick User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; de; rv:1.8.1.23) Gecko/20090812 Lightning/0.9 Thunderbird/2.0.0.23 Mnenhy/0.7.5.0 MIME-Version: 1.0 To: Jean Delvare CC: Yang Shi , Wolfram Sang , ddaney@caviumnetworks.com, ben-linux@fluff.org, ralf@linux-mips.org, linux-mips@linux-mips.org, linux-i2c@vger.kernel.org, Konstantin Lazarev Subject: Re: [PATCH] MIPS: Octeon: Register EEPROM device on the I2C bus References: <1267772895-25409-1-git-send-email-yang.shi@windriver.com> <20100305071130.GB21925@pengutronix.de> <4B90B341.9000601@windriver.com> <20100305074155.GD21925@pengutronix.de> <4B90B888.6060005@windriver.com> <20100305095040.6ab4612c@hyperion.delvare> In-Reply-To: <20100305095040.6ab4612c@hyperion.delvare> Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-Y-GMX-Trusted: 0 X-FuHaFi: 0.58999999999999997 Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26132 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ml.lawnick@gmx.de Precedence: bulk X-list: linux-mips Jean Delvare said the following: > Hi Yang, Wolfram, > > On Fri, 05 Mar 2010 15:53:44 +0800, Yang Shi wrote: >> Wolfram Sang 写é“: >> >>> Is the use of 'eeprom' instead of 'at24' intentional? >> >>> >> >>> >> >> Unfortunately, at24 driver can't work on this board, I must use legacy >> >> eeprom. >> >> >> > >> > Well, you are of course free to choose here :) >> > >> > I'd just be interested if there is a software limitation which prevents you from >> > using AT24. Because, it _should_ work with all kind of eeproms the legacy driver >> > deals with. Otherwise it is probably a bug which needs to be fixed. >> > >> >> Thanks to point out this. Let me take a look at this. > > One limitation of the at24 driver is that it needs the underlying > controller to support either raw I2C access or at least I2C block > transactions. Konstantin Lazarev complained about that one month ago > already. > > I am currently working on improving the at24 driver so that it falls > back to byte transactions when block transactions are not available. I > might also add word transaction support (as the eeprom driver has) as > it is often the best performance/compatibility trade-off. I'll post the > patch when I'm done. > > I'm not yet sure what will happen to the legacy eeprom driver in the > long run, but I would prefer new designs to not rely on it. > If I don't get all wrong, my 2 Cents: i2c-octeon will/should be based on raw i2c from kernel version .34 on. (my patch :-) ) So it should support all transfer modes that i2c can. Currently it is limited to 8 bytes per transaction. If I misunderstood something, please ignore the noise. -- KR Michael From Yang.Shi@windriver.com Fri Mar 5 11:39:47 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Fri, 05 Mar 2010 11:39:50 +0100 (CET) Received: from mail.windriver.com ([147.11.1.11]:57323 "EHLO mail.windriver.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492021Ab0CEKjr (ORCPT ); Fri, 5 Mar 2010 11:39:47 +0100 Received: from ALA-MAIL03.corp.ad.wrs.com (ala-mail03 [147.11.57.144]) by mail.windriver.com (8.14.3/8.14.3) with ESMTP id o25Ad6gu006108; Fri, 5 Mar 2010 02:39:06 -0800 (PST) Received: from [128.224.161.163] ([128.224.161.163]) by ALA-MAIL03.corp.ad.wrs.com with Microsoft SMTPSVC(6.0.3790.1830); Fri, 5 Mar 2010 02:39:06 -0800 Message-ID: <4B90DF48.50005@windriver.com> Date: Fri, 05 Mar 2010 18:39:04 +0800 From: Yang Shi User-Agent: Thunderbird 2.0.0.23 (X11/20090817) MIME-Version: 1.0 To: Michael Lawnick CC: Jean Delvare , Wolfram Sang , ddaney@caviumnetworks.com, ben-linux@fluff.org, ralf@linux-mips.org, linux-mips@linux-mips.org, linux-i2c@vger.kernel.org, Konstantin Lazarev Subject: Re: [PATCH] MIPS: Octeon: Register EEPROM device on the I2C bus References: <1267772895-25409-1-git-send-email-yang.shi@windriver.com> <20100305071130.GB21925@pengutronix.de> <4B90B341.9000601@windriver.com> <20100305074155.GD21925@pengutronix.de> <4B90B888.6060005@windriver.com> <20100305095040.6ab4612c@hyperion.delvare> <4B90D85E.6040308@gmx.de> In-Reply-To: <4B90D85E.6040308@gmx.de> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 8bit X-OriginalArrivalTime: 05 Mar 2010 10:39:06.0347 (UTC) FILETIME=[14A873B0:01CABC50] Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26133 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: yang.shi@windriver.com Precedence: bulk X-list: linux-mips Hi guys, Thanks for you guys kind advice. I think I find the cause. I coded a wrong eeprom type, "at24" can't work here, it should be "24c64". It works with AT24 eeprom driver, but I'm not sure if this is the right type. So, a possible correct to the patch is that: +{ + I2C_BOARD_INFO("24c64",·0x50), +}, Regards, Yang Michael Lawnick 写é“: > Jean Delvare said the following: > >> Hi Yang, Wolfram, >> >> On Fri, 05 Mar 2010 15:53:44 +0800, Yang Shi wrote: >> >>> Wolfram Sang 写é“: >>> >>>>>> Is the use of 'eeprom' instead of 'at24' intentional? >>>>>> >>>>>> >>>>>> >>>>> Unfortunately, at24 driver can't work on this board, I must use legacy >>>>> eeprom. >>>>> >>>>> >>>> Well, you are of course free to choose here :) >>>> >>>> I'd just be interested if there is a software limitation which prevents you from >>>> using AT24. Because, it _should_ work with all kind of eeproms the legacy driver >>>> deals with. Otherwise it is probably a bug which needs to be fixed. >>>> >>>> >>> Thanks to point out this. Let me take a look at this. >>> >> One limitation of the at24 driver is that it needs the underlying >> controller to support either raw I2C access or at least I2C block >> transactions. Konstantin Lazarev complained about that one month ago >> already. >> >> I am currently working on improving the at24 driver so that it falls >> back to byte transactions when block transactions are not available. I >> might also add word transaction support (as the eeprom driver has) as >> it is often the best performance/compatibility trade-off. I'll post the >> patch when I'm done. >> >> I'm not yet sure what will happen to the legacy eeprom driver in the >> long run, but I would prefer new designs to not rely on it. >> >> > > If I don't get all wrong, my 2 Cents: > i2c-octeon will/should be based on raw i2c from kernel version .34 on. > (my patch :-) ) So it should support all transfer modes that i2c can. > Currently it is limited to 8 bytes per transaction. > > If I misunderstood something, please ignore the noise. > > From khali@linux-fr.org Fri Mar 5 11:52:15 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Fri, 05 Mar 2010 11:52:19 +0100 (CET) Received: from poutre.nerim.net ([62.4.16.124]:52853 "EHLO poutre.nerim.net" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492021Ab0CEKwP convert rfc822-to-8bit (ORCPT ); Fri, 5 Mar 2010 11:52:15 +0100 Received: from localhost (localhost [127.0.0.1]) by poutre.nerim.net (Postfix) with ESMTP id A185239DE5D; Fri, 5 Mar 2010 11:52:13 +0100 (CET) X-Virus-Scanned: amavisd-new at nerim.net Received: from poutre.nerim.net ([127.0.0.1]) by localhost (poutre.nerim.net [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id K9jcy-4BTAHe; Fri, 5 Mar 2010 11:52:12 +0100 (CET) Received: from hyperion.delvare (jdelvare.pck.nerim.net [62.212.121.182]) by poutre.nerim.net (Postfix) with ESMTP id 5917E39DE52; Fri, 5 Mar 2010 11:52:12 +0100 (CET) Date: Fri, 5 Mar 2010 11:52:13 +0100 From: Jean Delvare To: Yang Shi Cc: Michael Lawnick , Wolfram Sang , ddaney@caviumnetworks.com, ben-linux@fluff.org, ralf@linux-mips.org, linux-mips@linux-mips.org, linux-i2c@vger.kernel.org, Konstantin Lazarev Subject: Re: [PATCH] MIPS: Octeon: Register EEPROM device on the I2C bus Message-ID: <20100305115213.4b504710@hyperion.delvare> In-Reply-To: <4B90DF48.50005@windriver.com> References: <1267772895-25409-1-git-send-email-yang.shi@windriver.com> <20100305071130.GB21925@pengutronix.de> <4B90B341.9000601@windriver.com> <20100305074155.GD21925@pengutronix.de> <4B90B888.6060005@windriver.com> <20100305095040.6ab4612c@hyperion.delvare> <4B90D85E.6040308@gmx.de> <4B90DF48.50005@windriver.com> X-Mailer: Claws Mail 3.5.0 (GTK+ 2.14.4; i586-suse-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8BIT Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26134 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: khali@linux-fr.org Precedence: bulk X-list: linux-mips On Fri, 05 Mar 2010 18:39:04 +0800, Yang Shi wrote: > Hi guys, > > Thanks for you guys kind advice. I think I find the cause. I coded a > wrong eeprom type, "at24" can't work here, it should be "24c64". It > works with AT24 eeprom driver, but I'm not sure if this is the right type. Actually, "at24" can work but you have to provide extra parameters detailing the size, page size etc. of the EEPROM. All the "24cXX" names are shortcuts with predefined sizes. > > So, a possible correct to the patch is that: > > +{ > + I2C_BOARD_INFO("24c64",·0x50), > +}, Well, what EEPROM type do you have exactly? 24c64 is for 64 kbit (8 kByte) EEPROMs using 16-bit addressing. You must use the correct type, otherwise the at24 driver will misbehave. I am a little surprised because originally you went for "eeprom" which is not compatible with "24c64" (8-bit vs. 16-bit addressing). Also note that you may want to provide specific page size if you have tight control over what hardware is used and you intend to write to the EEPROM on a regular basis. The driver defaults to safe but slow settings. OTOH, if you do _not_ want to write to the EEPROM, you want to provide the AT24_FLAG_READONLY flag. -- Jean Delvare From p2@psychaos.be Fri Mar 5 15:11:15 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Fri, 05 Mar 2010 15:11:18 +0100 (CET) Received: from apfelkorn.psychaos.be ([195.144.77.38]:41678 "EHLO apfelkorn.psychaos.be" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492352Ab0CEOLP (ORCPT ); Fri, 5 Mar 2010 15:11:15 +0100 Received: from p2 by apfelkorn.psychaos.be with local (Exim 4.69) (envelope-from ) id 1NnYFG-0001ap-39 for linux-mips@linux-mips.org; Fri, 05 Mar 2010 16:11:14 +0200 Date: Fri, 5 Mar 2010 16:11:14 +0200 From: Peter 'p2' De Schrijver To: linux-mips@linux-mips.org Subject: linux 2.6.33 on movidis x16 Message-ID: <20100305141113.GD2437@apfelkorn> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline X-Unexpected-Header: The spanish inquisition ! X-mate: Mate, mann gewohnt sich an alles X-Paddo: Munch, Munch User-Agent: Mutt/1.5.18 (2008-05-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26135 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: p2@debian.org Precedence: bulk X-list: linux-mips Hi, We are trying to get linux 2.6.33 to work on the movidis x16 board. Main thing missing is the addresses of the PHYs. Does anyone know those ? Unfortunately I don't have physical access to the board so I can't just look for the PHY components being used. Thanks, Peter. From mchehab@linuxtv.org Sat Mar 6 15:26:15 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Sat, 06 Mar 2010 15:26:20 +0100 (CET) Received: from www.linuxtv.org ([130.149.80.248]:41990 "EHLO www.linuxtv.org" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492151Ab0CFO0P (ORCPT ); Sat, 6 Mar 2010 15:26:15 +0100 Received: from mchehab by www.linuxtv.org with local (Exim 4.69) (envelope-from ) id 1NnuxI-0000ir-3G; Sat, 06 Mar 2010 15:26:12 +0100 Subject: [git:v4l-dvb/master] MIPS: Fixup of the r4k timer From: Patch from Wu Zhangjin To: linuxtv-commits@linuxtv.org Data: Mon, 1 Feb 2010 17:10:55 +0800 Cc: Wu Zhangjin , Shane McDonald , mbizon@freebox.fr, linux-mips@linux-mips.org, David VomLehn , Ralf Baechle Mail-followup-to: linux-media@vger.kernel.org Forward-to: linux-media@vger.kernel.org Reply-to: linux-media@vger.kernel.org Message-Id: Date: Sat, 06 Mar 2010 15:26:12 +0100 Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26136 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: linuxtv-commits-bounces@linuxtv.org Precedence: bulk X-list: linux-mips From: Wu Zhangjin As reported by Maxime Bizon, the commit "MIPS: PowerTV: Fix support for timer interrupts with > 64 external IRQs" have broken the r4k timer since it didn't initialize the cp0_compare_irq_shift variable used in c0_compare_int_pending() on the architectures whose cpu_has_mips_r2 is false. This patch fixes it via initializing the cp0_compare_irq_shift as the cp0_compare_irq used in the old c0_compare_int_pending(). Reported-by: Maxime Bizon Signed-off-by: Wu Zhangjin Cc: David VomLehn Cc: mbizon@freebox.fr Cc: linux-mips@linux-mips.org Patchwork: http://patchwork.linux-mips.org/patch/922/ Tested-by: Shane McDonald Signed-off-by: Ralf Baechle arch/mips/kernel/traps.c | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) --- http://git.linuxtv.org/v4l-dvb.git?a=commitdiff;h=f4fc580bec5fb76560329c8c537b9b71d8d032b6 diff --git a/arch/mips/kernel/traps.c b/arch/mips/kernel/traps.c index 338dfe8..31b204b 100644 --- a/arch/mips/kernel/traps.c +++ b/arch/mips/kernel/traps.c @@ -1501,6 +1501,7 @@ void __cpuinit per_cpu_trap_init(void) cp0_perfcount_irq = -1; } else { cp0_compare_irq = CP0_LEGACY_COMPARE_IRQ; + cp0_compare_irq_shift = cp0_compare_irq; cp0_perfcount_irq = -1; } From mchehab@linuxtv.org Sat Mar 6 15:28:04 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Sat, 06 Mar 2010 15:28:12 +0100 (CET) Received: from www.linuxtv.org ([130.149.80.248]:33640 "EHLO www.linuxtv.org" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492108Ab0CFO2E (ORCPT ); Sat, 6 Mar 2010 15:28:04 +0100 Received: from mchehab by www.linuxtv.org with local (Exim 4.69) (envelope-from ) id 1Nnuz4-0001b4-KN; Sat, 06 Mar 2010 15:28:02 +0100 Subject: [git:v4l-dvb/master] MIPS: Add support of LZO-compressed kernels From: Patch from Wu Zhangjin To: linuxtv-commits@linuxtv.org Data: Fri, 15 Jan 2010 20:34:46 +0800 Cc: Wu Zhangjin , linux-mips@linux-mips.org, Sergei Shtylyov , Ralf Baechle Mail-followup-to: linux-media@vger.kernel.org Forward-to: linux-media@vger.kernel.org Reply-to: linux-media@vger.kernel.org Message-Id: Date: Sat, 06 Mar 2010 15:28:02 +0100 Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26137 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: linuxtv-commits-bounces@linuxtv.org Precedence: bulk X-list: linux-mips From: Wu Zhangjin The necessary changes to the x86 Kconfig and boot/compressed to allow the use of this new compression method. Signed-off-by: Wu Zhangjin Cc: linux-mips@linux-mips.org Cc: Sergei Shtylyov Patchwork: http://patchwork.linux-mips.org/patch/857/ Signed-off-by: Ralf Baechle arch/mips/Kconfig | 1 + arch/mips/boot/compressed/Makefile | 2 ++ arch/mips/boot/compressed/decompress.c | 4 ++++ 3 files changed, 7 insertions(+), 0 deletions(-) --- http://git.linuxtv.org/v4l-dvb.git?a=commitdiff;h=fe1d45e08650213ec83a72d3499c3dd703243792 diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig index 9541171..8b5d174 100644 --- a/arch/mips/Kconfig +++ b/arch/mips/Kconfig @@ -1311,6 +1311,7 @@ config SYS_SUPPORTS_ZBOOT select HAVE_KERNEL_GZIP select HAVE_KERNEL_BZIP2 select HAVE_KERNEL_LZMA + select HAVE_KERNEL_LZO config SYS_SUPPORTS_ZBOOT_UART16550 bool diff --git a/arch/mips/boot/compressed/Makefile b/arch/mips/boot/compressed/Makefile index 671d344..bdcfd49 100644 --- a/arch/mips/boot/compressed/Makefile +++ b/arch/mips/boot/compressed/Makefile @@ -41,9 +41,11 @@ $(obj)/vmlinux.bin: $(KBUILD_IMAGE) suffix_$(CONFIG_KERNEL_GZIP) = gz suffix_$(CONFIG_KERNEL_BZIP2) = bz2 suffix_$(CONFIG_KERNEL_LZMA) = lzma +suffix_$(CONFIG_KERNEL_LZO) = lzo tool_$(CONFIG_KERNEL_GZIP) = gzip tool_$(CONFIG_KERNEL_BZIP2) = bzip2 tool_$(CONFIG_KERNEL_LZMA) = lzma +tool_$(CONFIG_KERNEL_LZO) = lzo $(obj)/vmlinux.$(suffix_y): $(obj)/vmlinux.bin $(call if_changed,$(tool_y)) diff --git a/arch/mips/boot/compressed/decompress.c b/arch/mips/boot/compressed/decompress.c index e48fd72..55d02b3 100644 --- a/arch/mips/boot/compressed/decompress.c +++ b/arch/mips/boot/compressed/decompress.c @@ -77,6 +77,10 @@ void *memset(void *s, int c, size_t n) #include "../../../../lib/decompress_unlzma.c" #endif +#ifdef CONFIG_KERNEL_LZO +#include "../../../../lib/decompress_unlzo.c" +#endif + void decompress_kernel(unsigned long boot_heap_start) { int zimage_size; From mchehab@linuxtv.org Sat Mar 6 15:35:55 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Sat, 06 Mar 2010 15:35:58 +0100 (CET) Received: from www.linuxtv.org ([130.149.80.248]:34448 "EHLO www.linuxtv.org" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492103Ab0CFOfz (ORCPT ); Sat, 6 Mar 2010 15:35:55 +0100 Received: from mchehab by www.linuxtv.org with local (Exim 4.69) (envelope-from ) id 1Nnuwe-0000Ll-KD; Sat, 06 Mar 2010 15:25:32 +0100 Subject: [git:v4l-dvb/master] MIPS: Highmem: Fix build error From: Patch from Yoichi Yuasa To: linuxtv-commits@linuxtv.org Data: Sat, 20 Feb 2010 21:23:22 +0900 Cc: linux-mips , Yoichi Yuasa , Ralf Baechle Mail-followup-to: linux-media@vger.kernel.org Forward-to: linux-media@vger.kernel.org Reply-to: linux-media@vger.kernel.org Message-Id: Date: Sat, 06 Mar 2010 15:25:32 +0100 Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26138 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: linuxtv-commits-bounces@linuxtv.org Precedence: bulk X-list: linux-mips From: Yoichi Yuasa arch/mips/mm/highmem.c: In function 'kmap_init': arch/mips/mm/highmem.c:130: error: 'init_mm' undeclared (first use in this function) arch/mips/mm/highmem.c:130: error: (Each undeclared identifier is reported only once arch/mips/mm/highmem.c:130: error: for each function it appears in.) Signed-off-by: Yoichi Yuasa Cc: linux-mips Patchwork: http://patchwork.linux-mips.org/patch/980/ Signed-off-by: Ralf Baechle arch/mips/mm/highmem.c | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) --- http://git.linuxtv.org/v4l-dvb.git?a=commitdiff;h=52ab320ac560af3333191a473e56615fb48fff95 diff --git a/arch/mips/mm/highmem.c b/arch/mips/mm/highmem.c index e274fda..127d732 100644 --- a/arch/mips/mm/highmem.c +++ b/arch/mips/mm/highmem.c @@ -1,5 +1,6 @@ #include #include +#include #include #include #include From mchehab@linuxtv.org Sat Mar 6 15:48:26 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Sat, 06 Mar 2010 15:48:30 +0100 (CET) Received: from www.linuxtv.org ([130.149.80.248]:34832 "EHLO www.linuxtv.org" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492103Ab0CFOs0 (ORCPT ); Sat, 6 Mar 2010 15:48:26 +0100 Received: from mchehab by www.linuxtv.org with local (Exim 4.69) (envelope-from ) id 1Nnuwe-0000Ld-Ch; Sat, 06 Mar 2010 15:25:32 +0100 Subject: [git:v4l-dvb/master] MIPS: BCM47xx: Fix 128MB RAM support From: Patch from Hauke Mehrtens To: linuxtv-commits@linuxtv.org Data: Sat, 20 Feb 2010 19:51:20 +0100 Cc: linux-mips@linux-mips.org, Hauke Mehrtens , Ralf Baechle Mail-followup-to: linux-media@vger.kernel.org Forward-to: linux-media@vger.kernel.org Reply-to: linux-media@vger.kernel.org Message-Id: Date: Sat, 06 Mar 2010 15:25:32 +0100 Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26139 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: linuxtv-commits-bounces@linuxtv.org Precedence: bulk X-list: linux-mips From: Hauke Mehrtens Ignoring the last page when ddr size is 128M. Cached accesses to last page is causing the processor to prefetch using address above 128M stepping out of the DDR address space. Signed-off-by: Hauke Mehrtens Cc: linux-mips@linux-mips.org Patchwork: http://patchwork.linux-mips.org/patch/981/ Signed-off-by: Ralf Baechle arch/mips/bcm47xx/prom.c | 8 ++++++++ 1 files changed, 8 insertions(+), 0 deletions(-) --- http://git.linuxtv.org/v4l-dvb.git?a=commitdiff;h=84a6fcb368a080620d12fc4d79e07902dbee7335 diff --git a/arch/mips/bcm47xx/prom.c b/arch/mips/bcm47xx/prom.c index c51405e..29d3cbf 100644 --- a/arch/mips/bcm47xx/prom.c +++ b/arch/mips/bcm47xx/prom.c @@ -141,6 +141,14 @@ static __init void prom_init_mem(void) break; } + /* Ignoring the last page when ddr size is 128M. Cached + * accesses to last page is causing the processor to prefetch + * using address above 128M stepping out of the ddr address + * space. + */ + if (mem == 0x8000000) + mem -= 0x1000; + add_memory_region(0, mem, BOOT_MEM_RAM); } From Yang.Shi@windriver.com Mon Mar 8 05:44:09 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Mon, 08 Mar 2010 05:44:12 +0100 (CET) Received: from mail.windriver.com ([147.11.1.11]:55982 "EHLO mail.windriver.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1490960Ab0CHEoJ (ORCPT ); Mon, 8 Mar 2010 05:44:09 +0100 Received: from ALA-MAIL03.corp.ad.wrs.com (ala-mail03 [147.11.57.144]) by mail.windriver.com (8.14.3/8.14.3) with ESMTP id o284hnjt015773; Sun, 7 Mar 2010 20:43:49 -0800 (PST) Received: from [128.224.162.222] ([128.224.162.222]) by ALA-MAIL03.corp.ad.wrs.com with Microsoft SMTPSVC(6.0.3790.1830); Sun, 7 Mar 2010 20:43:48 -0800 Message-ID: <4B948083.6000703@windriver.com> Date: Mon, 08 Mar 2010 12:43:47 +0800 From: Yang Shi User-Agent: Thunderbird 2.0.0.23 (X11/20090817) MIME-Version: 1.0 To: Jean Delvare CC: Michael Lawnick , Linux I2C , linux-mips@linux-mips.org Subject: Re: [PATCH] MIPS: Octeon: Register EEPROM device on the I2C bus References: <1267772895-25409-1-git-send-email-yang.shi@windriver.com> <20100305071130.GB21925@pengutronix.de> <4B90B341.9000601@windriver.com> <20100305074155.GD21925@pengutronix.de> <4B90B888.6060005@windriver.com> <20100305095040.6ab4612c@hyperion.delvare> <4B90D85E.6040308@gmx.de> <4B90DF48.50005@windriver.com> <20100305115213.4b504710@hyperion.delvare> <4B90E83A.5020106@gmx.de> <20100305124200.6f6eccfc@hyperion.delvare> In-Reply-To: <20100305124200.6f6eccfc@hyperion.delvare> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 8bit X-OriginalArrivalTime: 08 Mar 2010 04:43:49.0127 (UTC) FILETIME=[F1D81170:01CABE79] Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26140 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: yang.shi@windriver.com Precedence: bulk X-list: linux-mips Jean Delvare 写é“: > Hi Michael, > > On Fri, 05 Mar 2010 12:17:14 +0100, Michael Lawnick wrote: > >> Jean Delvare said the following: >> >>> Well, what EEPROM type do you have exactly? 24c64 is for 64 kbit (8 >>> kByte) EEPROMs using 16-bit addressing. You must use the correct type, >>> otherwise the at24 driver will misbehave. I am a little surprised >>> because originally you went for "eeprom" which is not compatible with >>> "24c64" (8-bit vs. 16-bit addressing). >>> >> Furthermore this brings up another issue: >> 0x50 typically is SPD-eeprom (DDR initialisation). Corrupting the >> contents might make your board unbootable - and using a 16bit driver >> instead of an 8-bit one can corrupt your contents already on >> (positioned) reading! >> > > This is totally correct, but better said loud to the list and the > original poster than only privately to me ;) > Thanks a lot to point out this. I double checked the manual, the eeprom is SPD of the DIMMs. I will rework my patch, then send V2 soon. Regards, Yang From yang.shi@windriver.com Mon Mar 8 06:30:28 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Mon, 08 Mar 2010 06:30:31 +0100 (CET) Received: from mail.windriver.com ([147.11.1.11]:61578 "EHLO mail.windriver.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1490965Ab0CHFa2 (ORCPT ); Mon, 8 Mar 2010 06:30:28 +0100 Received: from localhost.localdomain (pek-vmbuild11.wrs.com [128.224.153.31]) by mail.windriver.com (8.14.3/8.14.3) with ESMTP id o285Tpto021143; Sun, 7 Mar 2010 21:29:52 -0800 (PST) From: Yang Shi To: ddaney@caviumnetworks.com, ben-linux@fluff.org, khali@linux-fr.org, ralf@linux-mips.org Cc: linux-mips@linux-mips.org, linux-i2c@vger.kernel.org Subject: [PATCH V2] MIPS: Octeon: Register EEPROM device on the I2C bus Date: Mon, 8 Mar 2010 13:29:50 +0800 Message-Id: <1268026190-18300-1-git-send-email-yang.shi@windriver.com> X-Mailer: git-send-email 1.6.0.3 Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26141 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: yang.shi@windriver.com Precedence: bulk X-list: linux-mips An SPD resides on 0x50 of the I2C bus on CN56xx/57xx board, register this device. Signed-off-by: Yang Shi --- arch/mips/cavium-octeon/octeon-platform.c | 3 +++ 1 files changed, 3 insertions(+), 0 deletions(-) diff --git a/arch/mips/cavium-octeon/octeon-platform.c b/arch/mips/cavium-octeon/octeon-platform.c index 62ac30e..3138982 100644 --- a/arch/mips/cavium-octeon/octeon-platform.c +++ b/arch/mips/cavium-octeon/octeon-platform.c @@ -164,6 +164,9 @@ static struct i2c_board_info __initdata octeon_i2c_devices[] = { { I2C_BOARD_INFO("ds1337", 0x68), }, + { + I2C_BOARD_INFO("spd", 0x50), + }, }; static int __init octeon_i2c_devices_init(void) -- 1.6.3.3 From wsa@pengutronix.de Mon Mar 8 07:02:24 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Mon, 08 Mar 2010 07:02:28 +0100 (CET) Received: from metis.ext.pengutronix.de ([92.198.50.35]:51823 "EHLO metis.ext.pengutronix.de" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491143Ab0CHGCY (ORCPT ); Mon, 8 Mar 2010 07:02:24 +0100 Received: from octopus.hi.pengutronix.de ([2001:6f8:1178:2:215:17ff:fe12:23b0]) by metis.ext.pengutronix.de with esmtp (Exim 4.69) (envelope-from ) id 1NoW2Z-0004uI-In; Mon, 08 Mar 2010 07:02:07 +0100 Received: from wsa by octopus.hi.pengutronix.de with local (Exim 4.69) (envelope-from ) id 1NoW2R-0008Nj-Ar; Mon, 08 Mar 2010 07:01:59 +0100 Date: Mon, 8 Mar 2010 07:01:59 +0100 From: Wolfram Sang To: Yang Shi Cc: ddaney@caviumnetworks.com, ben-linux@fluff.org, khali@linux-fr.org, ralf@linux-mips.org, linux-mips@linux-mips.org, linux-i2c@vger.kernel.org Subject: Re: [PATCH V2] MIPS: Octeon: Register EEPROM device on the I2C bus Message-ID: <20100308060159.GA31209@pengutronix.de> References: <1268026190-18300-1-git-send-email-yang.shi@windriver.com> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="EVF5PPMfhYS0aIcm" Content-Disposition: inline In-Reply-To: <1268026190-18300-1-git-send-email-yang.shi@windriver.com> User-Agent: Mutt/1.5.18 (2008-05-17) X-SA-Exim-Connect-IP: 2001:6f8:1178:2:215:17ff:fe12:23b0 X-SA-Exim-Mail-From: wsa@pengutronix.de X-SA-Exim-Scanned: No (on metis.ext.pengutronix.de); SAEximRunCond expanded to false X-PTX-Original-Recipient: linux-mips@linux-mips.org Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26142 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: w.sang@pengutronix.de Precedence: bulk X-list: linux-mips --EVF5PPMfhYS0aIcm Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Mon, Mar 08, 2010 at 01:29:50PM +0800, Yang Shi wrote: > An SPD resides on 0x50 of the I2C bus on CN56xx/57xx board, > register this device. >=20 > Signed-off-by: Yang Shi Glad it worked with at24, after all. Acked-by: Wolfram Sang --=20 Pengutronix e.K. | Wolfram Sang | Industrial Linux Solutions | http://www.pengutronix.de/ | --EVF5PPMfhYS0aIcm Content-Type: application/pgp-signature; name="signature.asc" Content-Description: Digital signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) iEYEARECAAYFAkuUktcACgkQD27XaX1/VRv8vACeLbxiqll0Z1KOU+vt/poudlcv 6y4AnREPFYXDEnuY7Sa/Z1m63DA8Tx6e =mFOa -----END PGP SIGNATURE----- --EVF5PPMfhYS0aIcm-- From Yang.Shi@windriver.com Mon Mar 8 10:42:48 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Mon, 08 Mar 2010 10:42:53 +0100 (CET) Received: from mail.windriver.com ([147.11.1.11]:63179 "EHLO mail.windriver.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491022Ab0CHJms (ORCPT ); Mon, 8 Mar 2010 10:42:48 +0100 Received: from ALA-MAIL03.corp.ad.wrs.com (ala-mail03 [147.11.57.144]) by mail.windriver.com (8.14.3/8.14.3) with ESMTP id o289gaLe024132; Mon, 8 Mar 2010 01:42:36 -0800 (PST) Received: from [128.224.162.222] ([128.224.162.222]) by ALA-MAIL03.corp.ad.wrs.com with Microsoft SMTPSVC(6.0.3790.1830); Mon, 8 Mar 2010 01:42:35 -0800 Message-ID: <4B94C68A.4000002@windriver.com> Date: Mon, 08 Mar 2010 17:42:34 +0800 From: Yang Shi User-Agent: Thunderbird 2.0.0.23 (X11/20090817) MIME-Version: 1.0 To: Yang Shi CC: David Daney , ralf@linux-mips.org, linux-mips@linux-mips.org Subject: Re: [PATCH] MIPS: Protect current_cpu_data with preempt disable in delay() References: <1267695573-27360-1-git-send-email-yang.shi@windriver.com> <4B8FFAB3.1090409@caviumnetworks.com> <4B9059FE.3090801@windriver.com> In-Reply-To: <4B9059FE.3090801@windriver.com> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 8bit X-OriginalArrivalTime: 08 Mar 2010 09:42:35.0862 (UTC) FILETIME=[AF02BF60:01CABEA3] Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26143 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: yang.shi@windriver.com Precedence: bulk X-list: linux-mips Hi folks, Any advice on this topic? Thanks, Yang Yang Shi 写é“: > David Daney 写é“: > >> On 03/04/2010 01:39 AM, Yang Shi wrote: >> >> >>> During machine restart with reboot command, get the following >>> bug info: >>> >>> BUG: using smp_processor_id() in preemptible [00000000] code: reboot/1989 >>> caller is __udelay+0x14/0x70 >>> Call Trace: >>> [] dump_stack+0x8/0x34 >>> [] debug_smp_processor_id+0xf4/0x110 >>> [] __udelay+0x14/0x70 >>> [] md_notify_reboot+0x12c/0x148 >>> [] notifier_call_chain+0x64/0xc8 >>> [] __blocking_notifier_call_chain+0x64/0xc0 >>> [] kernel_restart_prepare+0x1c/0x38 >>> [] kernel_restart+0x14/0x50 >>> [] SyS_reboot+0x10c/0x1f0 >>> [] handle_sysn32+0x44/0x84 >>> >>> The root cause is that current_cpu_data is accessed in preemptible >>> context, so protect it with preempt_disable/preempt_enable pair >>> in delay(). >>> >>> Signed-off-by: Yang Shi >>> --- >>> arch/mips/lib/delay.c | 6 +++++- >>> 1 files changed, 5 insertions(+), 1 deletions(-) >>> >>> diff --git a/arch/mips/lib/delay.c b/arch/mips/lib/delay.c >>> index 6b3b1de..dc38064 100644 >>> --- a/arch/mips/lib/delay.c >>> +++ b/arch/mips/lib/delay.c >>> @@ -41,7 +41,11 @@ EXPORT_SYMBOL(__delay); >>> >>> void __udelay(unsigned long us) >>> { >>> - unsigned int lpj = current_cpu_data.udelay_val; >>> + unsigned int lpj; >>> + >>> + preempt_disable(); >>> + lpj = current_cpu_data.udelay_val; >>> + preempt_enable(); >>> >>> __delay((us * 0x000010c7ull * HZ * lpj)>> 32); >>> } >>> >>> >> This doesn't seem like the best approach. >> >> Perhaps we should either use raw_current_cpu_data and no >> preempt_disable(), or if we are concerned about migrating to a CPU with >> a different lpj value, move the preempt_enable after the call to __delay(). >> >> > > Thanks David. > > Yes, actually I also has this concern, so this patch is just a rough > fix. And I tried raw_current_cpu_data as well, but I'm not sure if it's > safe or not. Another proposal is to change cpu_data and current_cpu_data > to per CPU variables, of course this is a big change. > > Regards, > Yang > > >> David Daney >> >> >> > > > > From David.Daney@caviumnetworks.com Mon Mar 8 20:04:40 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Mon, 08 Mar 2010 20:04:43 +0100 (CET) Received: from mail3.caviumnetworks.com ([12.108.191.235]:3130 "EHLO mail3.caviumnetworks.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492476Ab0CHTEk (ORCPT ); Mon, 8 Mar 2010 20:04:40 +0100 Received: from caexch01.caveonetworks.com (Not Verified[192.168.16.9]) by mail3.caviumnetworks.com with MailMarshal (v6,7,2,8378) id ; Mon, 08 Mar 2010 11:04:44 -0800 Received: from caexch01.caveonetworks.com ([192.168.16.9]) by caexch01.caveonetworks.com with Microsoft SMTPSVC(6.0.3790.3959); Mon, 8 Mar 2010 11:04:34 -0800 Received: from dd1.caveonetworks.com ([12.108.191.236]) by caexch01.caveonetworks.com over TLS secured channel with Microsoft SMTPSVC(6.0.3790.3959); Mon, 8 Mar 2010 11:04:34 -0800 Received: from dd1.caveonetworks.com (localhost.localdomain [127.0.0.1]) by dd1.caveonetworks.com (8.14.3/8.14.3) with ESMTP id o28J4U4H031792; Mon, 8 Mar 2010 11:04:30 -0800 Received: (from ddaney@localhost) by dd1.caveonetworks.com (8.14.3/8.14.3/Submit) id o28J4Rqc031791; Mon, 8 Mar 2010 11:04:27 -0800 From: David Daney To: linux-i2c@vger.kernel.org, ben-linux@fluff.org, khali@linux-fr.org Cc: linux-mips@linux-mips.org, ralf@linux-mips.org, rade.bozic.ext@nsn.com, David Daney Subject: [PATCH] i2c: Fix section mismatch errors in i2c-octeon.c Date: Mon, 8 Mar 2010 11:04:21 -0800 Message-Id: <1268075061-31748-1-git-send-email-ddaney@caviumnetworks.com> X-Mailer: git-send-email 1.6.6.1 X-OriginalArrivalTime: 08 Mar 2010 19:04:34.0359 (UTC) FILETIME=[30CD1470:01CABEF2] Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26144 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ddaney@caviumnetworks.com Precedence: bulk X-list: linux-mips Signed-off-by: David Daney --- drivers/i2c/busses/i2c-octeon.c | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/i2c/busses/i2c-octeon.c b/drivers/i2c/busses/i2c-octeon.c index 6037550..be88d5b 100644 --- a/drivers/i2c/busses/i2c-octeon.c +++ b/drivers/i2c/busses/i2c-octeon.c @@ -446,7 +446,7 @@ static struct i2c_adapter octeon_i2c_ops = { /** * octeon_i2c_setclock - Calculate and set clock divisors. */ -static int __init octeon_i2c_setclock(struct octeon_i2c *i2c) +static int __devinit octeon_i2c_setclock(struct octeon_i2c *i2c) { int tclk, thp_base, inc, thp_idx, mdiv_idx, ndiv_idx, foscl, diff; int thp = 0x18, mdiv = 2, ndiv = 0, delta_hz = 1000000; @@ -489,7 +489,7 @@ static int __init octeon_i2c_setclock(struct octeon_i2c *i2c) return 0; } -static int __init octeon_i2c_initlowlevel(struct octeon_i2c *i2c) +static int __devinit octeon_i2c_initlowlevel(struct octeon_i2c *i2c) { u8 status; int tries; -- 1.6.6.1 From manuel.lauss@googlemail.com Mon Mar 8 20:22:01 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Mon, 08 Mar 2010 20:22:05 +0100 (CET) Received: from mail-fx0-f217.google.com ([209.85.220.217]:53615 "EHLO mail-fx0-f217.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492351Ab0CHTWB (ORCPT ); Mon, 8 Mar 2010 20:22:01 +0100 Received: by fxm9 with SMTP id 9so1891951fxm.24 for ; Mon, 08 Mar 2010 11:21:55 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=googlemail.com; s=gamma; h=domainkey-signature:received:received:from:to:cc:subject:date :message-id:x-mailer; bh=1Ob8hnhua9ueuzarEOXmgKvP69HpkgXeSeu/hjfWqgY=; b=Js5WIf5YDx7grhIjxxuHGQJ4dhE5M5QdEcWFZXhxxHaDJ2DWWuhnrVPaboyoDhMy/k 8H3GVTL9hn2em4/RfWexASvRylHXfP9+Xb8TiaCFGEeF12zT/wIP+y+8H1tsejSocKR7 HCeblbxPfVIvotTQKna0cngx32//jNt8pKRdk= DomainKey-Signature: a=rsa-sha1; c=nofws; d=googlemail.com; s=gamma; h=from:to:cc:subject:date:message-id:x-mailer; b=Jxjg6++PzJIT6qmgKaNPlHRrHlA4alsojF0OGNNtLkZ2N++HVInYRIoGR/oZgnAmYu 6RtXLzjRUTo9yxXLw6SI5zj4Osin/PQykR/KMhQMaDTkPSHcicioyZa919NdxzzJtSL/ t98Rge5D6kJn0ao0ks6IpQVnG9bjh9EGl8WHI= Received: by 10.223.97.220 with SMTP id m28mr201198fan.36.1268076114743; Mon, 08 Mar 2010 11:21:54 -0800 (PST) Received: from localhost.localdomain (p5496CE16.dip.t-dialin.net [84.150.206.22]) by mx.google.com with ESMTPS id b17sm9039919fka.43.2010.03.08.11.21.53 (version=TLSv1/SSLv3 cipher=RC4-MD5); Mon, 08 Mar 2010 11:21:54 -0800 (PST) From: Manuel Lauss To: Linux-MIPS Cc: Manuel Lauss Subject: [PATCH] MIPS: Alchemy: sleepcode without compile-time cputype dependencies Date: Mon, 8 Mar 2010 20:22:59 +0100 Message-Id: <1268076181-29642-1-git-send-email-manuel.lauss@gmail.com> X-Mailer: git-send-email 1.7.0.2 Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26145 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: manuel.lauss@googlemail.com Precedence: bulk X-list: linux-mips Split the low-level sleepcode into per-memory-controller-generation functions and figure out which one to call at runtime instead of relying on compile-time-defined cpu types. Signed-off-by: Manuel Lauss --- Works on the DB1200. arch/mips/alchemy/common/power.c | 12 +++- arch/mips/alchemy/common/sleeper.S | 81 ++++++++++++++++++---------- arch/mips/include/asm/mach-au1x00/au1000.h | 3 +- 3 files changed, 64 insertions(+), 32 deletions(-) diff --git a/arch/mips/alchemy/common/power.c b/arch/mips/alchemy/common/power.c index 6ab7b42..c07101c 100644 --- a/arch/mips/alchemy/common/power.c +++ b/arch/mips/alchemy/common/power.c @@ -209,9 +209,15 @@ static void restore_core_regs(void) void au_sleep(void) { - save_core_regs(); - au1xxx_save_and_sleep(); - restore_core_regs(); + int cpuid = alchemy_get_cputype(); + if (cpuid != ALCHEMY_CPU_UNKNOWN) { + save_core_regs(); + if (cpuid <= ALCHEMY_CPU_AU1500) + alchemy_sleep_au1000(); + else if (cpuid <= ALCHEMY_CPU_AU1200) + alchemy_sleep_au1550(); + restore_core_regs(); + } } #endif /* CONFIG_PM */ diff --git a/arch/mips/alchemy/common/sleeper.S b/arch/mips/alchemy/common/sleeper.S index 4f4b167..77f3c74 100644 --- a/arch/mips/alchemy/common/sleeper.S +++ b/arch/mips/alchemy/common/sleeper.S @@ -22,10 +22,9 @@ .set noat .align 5 -/* Save all of the processor general registers and go to sleep. - * A wakeup condition will get us back here to restore the registers. - */ -LEAF(au1xxx_save_and_sleep) + +/* preparatory stuff */ +.macro SETUP_SLEEP subu sp, PT_SIZE sw $1, PT_R1(sp) sw $2, PT_R2(sp) @@ -69,12 +68,32 @@ LEAF(au1xxx_save_and_sleep) */ lui t3, 0xb190 /* sys_xxx */ sw sp, 0x0018(t3) - la k0, 3f /* resume path */ + la k0, alchemy_sleep_wakeup /* resume path */ sw k0, 0x001c(t3) +.endm - /* Put SDRAM into self refresh: Preload instructions into cache, - * issue a precharge, auto/self refresh, then sleep commands to it. - */ +.macro DO_SLEEP + /* put power supply and processor to sleep */ + sw zero, 0x0078(t3) /* sys_slppwr */ + sync + sw zero, 0x007c(t3) /* sys_sleep */ + sync + nop + nop + nop + nop + nop + nop + nop + nop +.endm + +/* sleep code for Au1000/Au1100/Au1500 memory controller type */ +LEAF(alchemy_sleep_au1000) + + SETUP_SLEEP + + /* cache following instructions, as memory gets put to sleep */ la t0, 1f .set mips3 cache 0x14, 0(t0) @@ -84,17 +103,32 @@ LEAF(au1xxx_save_and_sleep) .set mips0 1: lui a0, 0xb400 /* mem_xxx */ -#if defined(CONFIG_SOC_AU1000) || defined(CONFIG_SOC_AU1100) || \ - defined(CONFIG_SOC_AU1500) sw zero, 0x001c(a0) /* Precharge */ sync sw zero, 0x0020(a0) /* Auto Refresh */ sync sw zero, 0x0030(a0) /* Sleep */ sync -#endif -#if defined(CONFIG_SOC_AU1550) || defined(CONFIG_SOC_AU1200) + DO_SLEEP + +END(alchemy_sleep_au1000) + +/* sleep code for Au1550/Au1200 memory controller type */ +LEAF(alchemy_sleep_au1550) + + SETUP_SLEEP + + /* cache following instructions, as memory gets put to sleep */ + la t0, 1f + .set mips3 + cache 0x14, 0(t0) + cache 0x14, 32(t0) + cache 0x14, 64(t0) + cache 0x14, 96(t0) + .set mips0 + +1: lui a0, 0xb400 /* mem_xxx */ sw zero, 0x08c0(a0) /* Precharge */ sync sw zero, 0x08d0(a0) /* Self Refresh */ @@ -114,26 +148,17 @@ LEAF(au1xxx_save_and_sleep) and t1, t0, t1 /* clear CE[1:0] */ sw t1, 0x0840(a0) /* mem_sdconfiga */ sync -#endif - /* put power supply and processor to sleep */ - sw zero, 0x0078(t3) /* sys_slppwr */ - sync - sw zero, 0x007c(t3) /* sys_sleep */ - sync - nop - nop - nop - nop - nop - nop - nop - nop + DO_SLEEP + +END(alchemy_sleep_au1550) + /* This is where we return upon wakeup. * Reload all of the registers and return. */ -3: lw k0, 0x20(sp) +LEAF(alchemy_sleep_wakeup) + lw k0, 0x20(sp) mtc0 k0, CP0_STATUS lw k0, 0x1c(sp) mtc0 k0, CP0_CONTEXT @@ -169,4 +194,4 @@ LEAF(au1xxx_save_and_sleep) lw $31, PT_R31(sp) jr ra addiu sp, PT_SIZE -END(au1xxx_save_and_sleep) +END(alchemy_sleep_wakeup) diff --git a/arch/mips/include/asm/mach-au1x00/au1000.h b/arch/mips/include/asm/mach-au1x00/au1000.h index ae07423..cb91714 100644 --- a/arch/mips/include/asm/mach-au1x00/au1000.h +++ b/arch/mips/include/asm/mach-au1x00/au1000.h @@ -188,7 +188,8 @@ extern unsigned long get_au1x00_uart_baud_base(void); extern unsigned long au1xxx_calc_clock(void); /* PM: arch/mips/alchemy/common/sleeper.S, power.c, irq.c */ -void au1xxx_save_and_sleep(void); +void alchemy_sleep_au1000(void); +void alchemy_sleep_au1550(void); void au_sleep(void); void save_au1xxx_intctl(void); void restore_au1xxx_intctl(void); -- 1.7.0.2 From manuel.lauss@googlemail.com Mon Mar 8 20:22:23 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Mon, 08 Mar 2010 20:22:30 +0100 (CET) Received: from mail-fx0-f217.google.com ([209.85.220.217]:62704 "EHLO mail-fx0-f217.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492480Ab0CHTWC (ORCPT ); Mon, 8 Mar 2010 20:22:02 +0100 Received: by fxm9 with SMTP id 9so1891969fxm.24 for ; Mon, 08 Mar 2010 11:21:57 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=googlemail.com; s=gamma; h=domainkey-signature:received:received:from:to:cc:subject:date :message-id:x-mailer:in-reply-to:references; bh=BIo0YI5tjujgAcptVJEDPdiJ3779bXEtJMFpnn7i+RM=; b=XBD7TA4mrgIIH7aIVsGqErB+fEiRiFkuTiFPRLY18euqTa6HwhKCpLBP5im3XRR605 sgMKYTZ7SD00oXZd/G0inkxBvLsdNixVeSN6L1QOUgexGlgwkMhlG2xJd68eLA15+vmg p7rDKysbrf1DOvF86Ov2UH07Cy2nuLAO/mzvo= DomainKey-Signature: a=rsa-sha1; c=nofws; d=googlemail.com; s=gamma; h=from:to:cc:subject:date:message-id:x-mailer:in-reply-to:references; b=llmk5YwIRZTrmgF8FkcUcXTBYBhi2cmqg5Hy/rAAEndJ46adgpvcJURH61qYTPDAVr O7ShqMMA7l40eXvgqZYxQ1CeglsKnWEtAGBrMlpAyukepITaaZ4vifa3HfuBCogMPOUa RyAWuMBCs74Qb/LaaKrclNwck3txZLQ1q5CNY= Received: by 10.223.4.211 with SMTP id 19mr6247718fas.72.1268076115750; Mon, 08 Mar 2010 11:21:55 -0800 (PST) Received: from localhost.localdomain (p5496CE16.dip.t-dialin.net [84.150.206.22]) by mx.google.com with ESMTPS id b17sm9039919fka.43.2010.03.08.11.21.54 (version=TLSv1/SSLv3 cipher=RC4-MD5); Mon, 08 Mar 2010 11:21:55 -0800 (PST) From: Manuel Lauss To: Linux-MIPS Cc: Manuel Lauss , Manuel Lauss Subject: [RFC PATCH] MIPS: Alchemy: add sysdev for both irq controllers Date: Mon, 8 Mar 2010 20:23:00 +0100 Message-Id: <1268076181-29642-2-git-send-email-manuel.lauss@gmail.com> X-Mailer: git-send-email 1.7.0.2 In-Reply-To: <1268076181-29642-1-git-send-email-manuel.lauss@gmail.com> References: <1268076181-29642-1-git-send-email-manuel.lauss@gmail.com> Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26146 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: manuel.lauss@googlemail.com Precedence: bulk X-list: linux-mips From: Manuel Lauss Use a sysdev to implement PM methods for the Au1000 interrupt controllers. Signed-off-by: Manuel Lauss --- Works on DB1200, not sure though whether this is the correct approach. Applies cleanly only on top of "sleepcode-without-compile-time-cputype" patch. arch/mips/alchemy/common/irq.c | 196 ++++++++++++++++------------ arch/mips/alchemy/common/power.c | 5 - arch/mips/include/asm/mach-au1x00/au1000.h | 2 - 3 files changed, 112 insertions(+), 91 deletions(-) diff --git a/arch/mips/alchemy/common/irq.c b/arch/mips/alchemy/common/irq.c index b2821ac..fd0e577 100644 --- a/arch/mips/alchemy/common/irq.c +++ b/arch/mips/alchemy/common/irq.c @@ -29,6 +29,7 @@ #include #include #include +#include #include #include @@ -216,90 +217,6 @@ struct au1xxx_irqmap au1200_irqmap[] __initdata = { }; -#ifdef CONFIG_PM - -/* - * Save/restore the interrupt controller state. - * Called from the save/restore core registers as part of the - * au_sleep function in power.c.....maybe I should just pm_register() - * them instead? - */ -static unsigned int sleep_intctl_config0[2]; -static unsigned int sleep_intctl_config1[2]; -static unsigned int sleep_intctl_config2[2]; -static unsigned int sleep_intctl_src[2]; -static unsigned int sleep_intctl_assign[2]; -static unsigned int sleep_intctl_wake[2]; -static unsigned int sleep_intctl_mask[2]; - -void save_au1xxx_intctl(void) -{ - sleep_intctl_config0[0] = au_readl(IC0_CFG0RD); - sleep_intctl_config1[0] = au_readl(IC0_CFG1RD); - sleep_intctl_config2[0] = au_readl(IC0_CFG2RD); - sleep_intctl_src[0] = au_readl(IC0_SRCRD); - sleep_intctl_assign[0] = au_readl(IC0_ASSIGNRD); - sleep_intctl_wake[0] = au_readl(IC0_WAKERD); - sleep_intctl_mask[0] = au_readl(IC0_MASKRD); - - sleep_intctl_config0[1] = au_readl(IC1_CFG0RD); - sleep_intctl_config1[1] = au_readl(IC1_CFG1RD); - sleep_intctl_config2[1] = au_readl(IC1_CFG2RD); - sleep_intctl_src[1] = au_readl(IC1_SRCRD); - sleep_intctl_assign[1] = au_readl(IC1_ASSIGNRD); - sleep_intctl_wake[1] = au_readl(IC1_WAKERD); - sleep_intctl_mask[1] = au_readl(IC1_MASKRD); -} - -/* - * For most restore operations, we clear the entire register and - * then set the bits we found during the save. - */ -void restore_au1xxx_intctl(void) -{ - au_writel(0xffffffff, IC0_MASKCLR); au_sync(); - - au_writel(0xffffffff, IC0_CFG0CLR); au_sync(); - au_writel(sleep_intctl_config0[0], IC0_CFG0SET); au_sync(); - au_writel(0xffffffff, IC0_CFG1CLR); au_sync(); - au_writel(sleep_intctl_config1[0], IC0_CFG1SET); au_sync(); - au_writel(0xffffffff, IC0_CFG2CLR); au_sync(); - au_writel(sleep_intctl_config2[0], IC0_CFG2SET); au_sync(); - au_writel(0xffffffff, IC0_SRCCLR); au_sync(); - au_writel(sleep_intctl_src[0], IC0_SRCSET); au_sync(); - au_writel(0xffffffff, IC0_ASSIGNCLR); au_sync(); - au_writel(sleep_intctl_assign[0], IC0_ASSIGNSET); au_sync(); - au_writel(0xffffffff, IC0_WAKECLR); au_sync(); - au_writel(sleep_intctl_wake[0], IC0_WAKESET); au_sync(); - au_writel(0xffffffff, IC0_RISINGCLR); au_sync(); - au_writel(0xffffffff, IC0_FALLINGCLR); au_sync(); - au_writel(0x00000000, IC0_TESTBIT); au_sync(); - - au_writel(0xffffffff, IC1_MASKCLR); au_sync(); - - au_writel(0xffffffff, IC1_CFG0CLR); au_sync(); - au_writel(sleep_intctl_config0[1], IC1_CFG0SET); au_sync(); - au_writel(0xffffffff, IC1_CFG1CLR); au_sync(); - au_writel(sleep_intctl_config1[1], IC1_CFG1SET); au_sync(); - au_writel(0xffffffff, IC1_CFG2CLR); au_sync(); - au_writel(sleep_intctl_config2[1], IC1_CFG2SET); au_sync(); - au_writel(0xffffffff, IC1_SRCCLR); au_sync(); - au_writel(sleep_intctl_src[1], IC1_SRCSET); au_sync(); - au_writel(0xffffffff, IC1_ASSIGNCLR); au_sync(); - au_writel(sleep_intctl_assign[1], IC1_ASSIGNSET); au_sync(); - au_writel(0xffffffff, IC1_WAKECLR); au_sync(); - au_writel(sleep_intctl_wake[1], IC1_WAKESET); au_sync(); - au_writel(0xffffffff, IC1_RISINGCLR); au_sync(); - au_writel(0xffffffff, IC1_FALLINGCLR); au_sync(); - au_writel(0x00000000, IC1_TESTBIT); au_sync(); - - au_writel(sleep_intctl_mask[1], IC1_MASKSET); au_sync(); - - au_writel(sleep_intctl_mask[0], IC0_MASKSET); au_sync(); -} -#endif /* CONFIG_PM */ - - static void au1x_ic0_unmask(unsigned int irq_nr) { unsigned int bit = irq_nr - AU1000_INTC0_INT_BASE; @@ -635,3 +552,114 @@ void __init arch_init_irq(void) break; } } + +struct alchemy_ic_sysdev { + struct sys_device sysdev; + unsigned long pmdata[7 * 2]; +}; + +static int alchemy_ic_suspend(struct sys_device *dev, pm_message_t state) +{ + struct alchemy_ic_sysdev *icdev = + container_of(dev, struct alchemy_ic_sysdev, sysdev); + + icdev->pmdata[ 0] = au_readl(IC0_CFG0RD); + icdev->pmdata[ 1] = au_readl(IC0_CFG1RD); + icdev->pmdata[ 2] = au_readl(IC0_CFG2RD); + icdev->pmdata[ 3] = au_readl(IC0_SRCRD); + icdev->pmdata[ 4] = au_readl(IC0_ASSIGNRD); + icdev->pmdata[ 5] = au_readl(IC0_WAKERD); + icdev->pmdata[ 6] = au_readl(IC0_MASKRD); + + icdev->pmdata[ 7] = au_readl(IC1_CFG0RD); + icdev->pmdata[ 8] = au_readl(IC1_CFG1RD); + icdev->pmdata[ 9] = au_readl(IC1_CFG2RD); + icdev->pmdata[10] = au_readl(IC1_SRCRD); + icdev->pmdata[11] = au_readl(IC1_ASSIGNRD); + icdev->pmdata[12] = au_readl(IC1_WAKERD); + icdev->pmdata[13] = au_readl(IC1_MASKRD); + + return 0; +} + +static int alchemy_ic_resume(struct sys_device *dev) +{ + struct alchemy_ic_sysdev *icdev = + container_of(dev, struct alchemy_ic_sysdev, sysdev); + + au_writel(0xffffffff, IC0_MASKCLR); + au_writel(0xffffffff, IC0_CFG0CLR); + au_writel(0xffffffff, IC0_CFG1CLR); + au_writel(0xffffffff, IC0_CFG2CLR); + au_writel(0xffffffff, IC0_SRCCLR); + au_writel(0xffffffff, IC0_ASSIGNCLR); + au_writel(0xffffffff, IC0_WAKECLR); + au_writel(0xffffffff, IC0_RISINGCLR); + au_writel(0xffffffff, IC0_FALLINGCLR); + au_writel(0x00000000, IC0_TESTBIT); + au_sync(); + au_writel(icdev->pmdata[ 0], IC0_CFG0SET); + au_writel(icdev->pmdata[ 1], IC0_CFG1SET); + au_writel(icdev->pmdata[ 2], IC0_CFG2SET); + au_writel(icdev->pmdata[ 3], IC0_SRCSET); + au_writel(icdev->pmdata[ 4], IC0_ASSIGNSET); + au_writel(icdev->pmdata[ 5], IC0_WAKESET); + au_sync(); + + au_writel(0xffffffff, IC1_MASKCLR); + au_writel(0xffffffff, IC1_CFG0CLR); + au_writel(0xffffffff, IC1_CFG1CLR); + au_writel(0xffffffff, IC1_CFG2CLR); + au_writel(0xffffffff, IC1_SRCCLR); + au_writel(0xffffffff, IC1_ASSIGNCLR); + au_writel(0xffffffff, IC1_WAKECLR); + au_writel(0xffffffff, IC1_RISINGCLR); + au_writel(0xffffffff, IC1_FALLINGCLR); + au_writel(0x00000000, IC1_TESTBIT); + au_sync(); + au_writel(icdev->pmdata[ 7], IC1_CFG0SET); + au_writel(icdev->pmdata[ 8], IC1_CFG1SET); + au_writel(icdev->pmdata[ 9], IC1_CFG2SET); + au_writel(icdev->pmdata[10], IC1_SRCSET); + au_writel(icdev->pmdata[11], IC1_ASSIGNSET); + au_writel(icdev->pmdata[12], IC1_WAKESET); + au_sync(); + + au_writel(icdev->pmdata[13], IC1_MASKSET); + au_sync(); + au_writel(icdev->pmdata[ 6], IC0_MASKSET); + au_sync(); + + return 0; +} + +static struct sysdev_class alchemy_ic_sysdev_class = { + .name = "ic", + .suspend = alchemy_ic_suspend, + .resume = alchemy_ic_resume, +}; + +static int __init alchemy_ic_sysdev_init(void) +{ + struct alchemy_ic_sysdev *icdev; + int err; + + icdev = kzalloc(sizeof(struct alchemy_ic_sysdev), GFP_KERNEL); + if (!icdev) + return -ENOMEM; + + err = sysdev_class_register(&alchemy_ic_sysdev_class); + if (err) { + kfree(icdev); + return err; + } + + icdev->sysdev.id = 0; + icdev->sysdev.cls = &alchemy_ic_sysdev_class; + err = sysdev_register(&icdev->sysdev); + if (err) + kfree(icdev); + + return err; +} +device_initcall(alchemy_ic_sysdev_init); diff --git a/arch/mips/alchemy/common/power.c b/arch/mips/alchemy/common/power.c index c07101c..e097094 100644 --- a/arch/mips/alchemy/common/power.c +++ b/arch/mips/alchemy/common/power.c @@ -106,9 +106,6 @@ static void save_core_regs(void) sleep_usb[1] = au_readl(0xb4020024); /* OTG_MUX */ #endif - /* Save interrupt controller state. */ - save_au1xxx_intctl(); - /* Clocks and PLLs. */ sleep_sys_clocks[0] = au_readl(SYS_FREQCTRL0); sleep_sys_clocks[1] = au_readl(SYS_FREQCTRL1); @@ -200,8 +197,6 @@ static void restore_core_regs(void) au_writel(sleep_uart0_clkdiv, UART0_ADDR + UART_CLK); au_sync(); } - restore_au1xxx_intctl(); - #if defined(CONFIG_SOC_AU1550) || defined(CONFIG_SOC_AU1200) au1xxx_dbdma_resume(); #endif diff --git a/arch/mips/include/asm/mach-au1x00/au1000.h b/arch/mips/include/asm/mach-au1x00/au1000.h index cb91714..c84fcd6 100644 --- a/arch/mips/include/asm/mach-au1x00/au1000.h +++ b/arch/mips/include/asm/mach-au1x00/au1000.h @@ -191,8 +191,6 @@ extern unsigned long au1xxx_calc_clock(void); void alchemy_sleep_au1000(void); void alchemy_sleep_au1550(void); void au_sleep(void); -void save_au1xxx_intctl(void); -void restore_au1xxx_intctl(void); /* SOC Interrupt numbers */ -- 1.7.0.2 From manuel.lauss@googlemail.com Mon Mar 8 20:22:50 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Mon, 08 Mar 2010 20:22:58 +0100 (CET) Received: from mail-fx0-f217.google.com ([209.85.220.217]:63613 "EHLO mail-fx0-f217.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492481Ab0CHTWC (ORCPT ); Mon, 8 Mar 2010 20:22:02 +0100 Received: by fxm9 with SMTP id 9so1891984fxm.24 for ; Mon, 08 Mar 2010 11:21:57 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=googlemail.com; s=gamma; h=domainkey-signature:received:received:from:to:cc:subject:date :message-id:x-mailer:in-reply-to:references; bh=fCQjeZS/3WdHUkjxDMDleRega9y9gUm1oM2nZS1cASI=; b=YlgRNoLra2dENwr0J2/MdolrLMwH1iYYR+QRPXOdjvyf92Sog/zufR+Oq9VSSItjLU aXIhCtoFH76gAv989lrLiHFwuvCsW7ufvsgFvApq8oN/09JGyFoWUU0vRL+C3jHKbUSc duAYJopPG9qUXf7TZiJYC71dBmKPr3RdDgAfw= DomainKey-Signature: a=rsa-sha1; c=nofws; d=googlemail.com; s=gamma; h=from:to:cc:subject:date:message-id:x-mailer:in-reply-to:references; b=cQDSGrVTeUjZujxx6KBWfnX0Cw99ak75RZS0HhDICoPtC1ulw3EoHiiX1ZU4ucHD1X U98q1ij6HAotzFZE60vBtqA0IfYZsbgJ7HJSMtkLE3pSMNHZZ0d3wOrQl50Z8iaMOzwR d7T3QdoeXH8sB48JpTuoOjPbmRhkHAo1WxYY0= Received: by 10.223.5.70 with SMTP id 6mr2622482fau.18.1268076116956; Mon, 08 Mar 2010 11:21:56 -0800 (PST) Received: from localhost.localdomain (p5496CE16.dip.t-dialin.net [84.150.206.22]) by mx.google.com with ESMTPS id b17sm9039919fka.43.2010.03.08.11.21.55 (version=TLSv1/SSLv3 cipher=RC4-MD5); Mon, 08 Mar 2010 11:21:56 -0800 (PST) From: Manuel Lauss To: Linux-MIPS Cc: Manuel Lauss Subject: [PATCH] MIPS: Alchemy: move MMC driver registration to board code. Date: Mon, 8 Mar 2010 20:23:01 +0100 Message-Id: <1268076181-29642-3-git-send-email-manuel.lauss@gmail.com> X-Mailer: git-send-email 1.7.0.2 In-Reply-To: <1268076181-29642-1-git-send-email-manuel.lauss@gmail.com> References: <1268076181-29642-1-git-send-email-manuel.lauss@gmail.com> Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26147 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: manuel.lauss@googlemail.com Precedence: bulk X-list: linux-mips Where it really belongs to. Signed-off-by: Manuel Lauss --- arch/mips/alchemy/common/platform.c | 81 ------------------------- arch/mips/alchemy/devboards/db1200/platform.c | 48 +++++++++++++-- arch/mips/alchemy/devboards/pb1200/platform.c | 79 +++++++++++++++++++++++- 3 files changed, 119 insertions(+), 89 deletions(-) diff --git a/arch/mips/alchemy/common/platform.c b/arch/mips/alchemy/common/platform.c index 2580e77..e1ed1ad 100644 --- a/arch/mips/alchemy/common/platform.c +++ b/arch/mips/alchemy/common/platform.c @@ -18,7 +18,6 @@ #include #include -#include #include #define PORT(_base, _irq) \ @@ -231,82 +230,6 @@ static struct platform_device au1200_lcd_device = { .num_resources = ARRAY_SIZE(au1200_lcd_resources), .resource = au1200_lcd_resources, }; - -static u64 au1xxx_mmc_dmamask = DMA_BIT_MASK(32); - -extern struct au1xmmc_platform_data au1xmmc_platdata[2]; - -static struct resource au1200_mmc0_resources[] = { - [0] = { - .start = SD0_PHYS_ADDR, - .end = SD0_PHYS_ADDR + 0x7ffff, - .flags = IORESOURCE_MEM, - }, - [1] = { - .start = AU1200_SD_INT, - .end = AU1200_SD_INT, - .flags = IORESOURCE_IRQ, - }, - [2] = { - .start = DSCR_CMD0_SDMS_TX0, - .end = DSCR_CMD0_SDMS_TX0, - .flags = IORESOURCE_DMA, - }, - [3] = { - .start = DSCR_CMD0_SDMS_RX0, - .end = DSCR_CMD0_SDMS_RX0, - .flags = IORESOURCE_DMA, - } -}; - -static struct platform_device au1200_mmc0_device = { - .name = "au1xxx-mmc", - .id = 0, - .dev = { - .dma_mask = &au1xxx_mmc_dmamask, - .coherent_dma_mask = DMA_BIT_MASK(32), - .platform_data = &au1xmmc_platdata[0], - }, - .num_resources = ARRAY_SIZE(au1200_mmc0_resources), - .resource = au1200_mmc0_resources, -}; - -#ifndef CONFIG_MIPS_DB1200 -static struct resource au1200_mmc1_resources[] = { - [0] = { - .start = SD1_PHYS_ADDR, - .end = SD1_PHYS_ADDR + 0x7ffff, - .flags = IORESOURCE_MEM, - }, - [1] = { - .start = AU1200_SD_INT, - .end = AU1200_SD_INT, - .flags = IORESOURCE_IRQ, - }, - [2] = { - .start = DSCR_CMD0_SDMS_TX1, - .end = DSCR_CMD0_SDMS_TX1, - .flags = IORESOURCE_DMA, - }, - [3] = { - .start = DSCR_CMD0_SDMS_RX1, - .end = DSCR_CMD0_SDMS_RX1, - .flags = IORESOURCE_DMA, - } -}; - -static struct platform_device au1200_mmc1_device = { - .name = "au1xxx-mmc", - .id = 1, - .dev = { - .dma_mask = &au1xxx_mmc_dmamask, - .coherent_dma_mask = DMA_BIT_MASK(32), - .platform_data = &au1xmmc_platdata[1], - }, - .num_resources = ARRAY_SIZE(au1200_mmc1_resources), - .resource = au1200_mmc1_resources, -}; -#endif /* #ifndef CONFIG_MIPS_DB1200 */ #endif /* #ifdef CONFIG_SOC_AU1200 */ /* All Alchemy demoboards with I2C have this #define in their headers */ @@ -421,10 +344,6 @@ static struct platform_device *au1xxx_platform_devices[] __initdata = { &au1xxx_usb_gdt_device, &au1xxx_usb_otg_device, &au1200_lcd_device, - &au1200_mmc0_device, -#ifndef CONFIG_MIPS_DB1200 - &au1200_mmc1_device, -#endif #endif #ifdef SMBUS_PSC_BASE &pbdb_smbus_device, diff --git a/arch/mips/alchemy/devboards/db1200/platform.c b/arch/mips/alchemy/devboards/db1200/platform.c index 3cb95a9..9afb287 100644 --- a/arch/mips/alchemy/devboards/db1200/platform.c +++ b/arch/mips/alchemy/devboards/db1200/platform.c @@ -329,14 +329,49 @@ static struct led_classdev db1200_mmc_led = { }; /* needed by arch/mips/alchemy/common/platform.c */ -struct au1xmmc_platform_data au1xmmc_platdata[] = { +static struct au1xmmc_platform_data db1200_mmc_platdata = { + .cd_setup = db1200_mmc_cd_setup, + .set_power = db1200_mmc_set_power, + .card_inserted = db1200_mmc_card_inserted, + .card_readonly = db1200_mmc_card_readonly, + .led = &db1200_mmc_led, +}; + +static u64 au1xxx_mmc_dmamask = DMA_BIT_MASK(32); + +static struct resource au1200_mmc0_resources[] = { [0] = { - .cd_setup = db1200_mmc_cd_setup, - .set_power = db1200_mmc_set_power, - .card_inserted = db1200_mmc_card_inserted, - .card_readonly = db1200_mmc_card_readonly, - .led = &db1200_mmc_led, + .start = SD0_PHYS_ADDR, + .end = SD0_PHYS_ADDR + 0x7ffff, + .flags = IORESOURCE_MEM, + }, + [1] = { + .start = AU1200_SD_INT, + .end = AU1200_SD_INT, + .flags = IORESOURCE_IRQ, + }, + [2] = { + .start = DSCR_CMD0_SDMS_TX0, + .end = DSCR_CMD0_SDMS_TX0, + .flags = IORESOURCE_DMA, + }, + [3] = { + .start = DSCR_CMD0_SDMS_RX0, + .end = DSCR_CMD0_SDMS_RX0, + .flags = IORESOURCE_DMA, + } +}; + +static struct platform_device db1200_mmc0_dev = { + .name = "au1xxx-mmc", + .id = 0, + .dev = { + .dma_mask = &au1xxx_mmc_dmamask, + .coherent_dma_mask = DMA_BIT_MASK(32), + .platform_data = &db1200_mmc_platdata, }, + .num_resources = ARRAY_SIZE(au1200_mmc0_resources), + .resource = au1200_mmc0_resources, }; /**********************************************************************/ @@ -436,6 +471,7 @@ static struct platform_device *db1200_devs[] __initdata = { &db1200_rtc_dev, &db1200_nand_dev, &db1200_audio_dev, + &db1200_mmc0_dev, }; static int __init db1200_dev_init(void) diff --git a/arch/mips/alchemy/devboards/pb1200/platform.c b/arch/mips/alchemy/devboards/pb1200/platform.c index 3ef2dce..89e6f1d 100644 --- a/arch/mips/alchemy/devboards/pb1200/platform.c +++ b/arch/mips/alchemy/devboards/pb1200/platform.c @@ -25,6 +25,7 @@ #include #include +#include #include #include @@ -88,7 +89,7 @@ static int pb1200mmc1_card_inserted(void *mmc_host) return (bcsr_read(BCSR_SIGSTAT) & BCSR_INT_SD1INSERT) ? 1 : 0; } -const struct au1xmmc_platform_data au1xmmc_platdata[2] = { +static struct au1xmmc_platform_data pb1200_mmc_platdata[2] = { [0] = { .set_power = pb1200mmc0_set_power, .card_inserted = pb1200mmc0_card_inserted, @@ -105,6 +106,78 @@ const struct au1xmmc_platform_data au1xmmc_platdata[2] = { }, }; +static u64 au1xxx_mmc_dmamask = DMA_BIT_MASK(32); + +static struct resource au1200_mmc0_resources[] = { + [0] = { + .start = SD0_PHYS_ADDR, + .end = SD0_PHYS_ADDR + 0x7ffff, + .flags = IORESOURCE_MEM, + }, + [1] = { + .start = AU1200_SD_INT, + .end = AU1200_SD_INT, + .flags = IORESOURCE_IRQ, + }, + [2] = { + .start = DSCR_CMD0_SDMS_TX0, + .end = DSCR_CMD0_SDMS_TX0, + .flags = IORESOURCE_DMA, + }, + [3] = { + .start = DSCR_CMD0_SDMS_RX0, + .end = DSCR_CMD0_SDMS_RX0, + .flags = IORESOURCE_DMA, + } +}; + +static struct platform_device pb1200_mmc0_device = { + .name = "au1xxx-mmc", + .id = 0, + .dev = { + .dma_mask = &au1xxx_mmc_dmamask, + .coherent_dma_mask = DMA_BIT_MASK(32), + .platform_data = &pb1200_mmc_platdata[0], + }, + .num_resources = ARRAY_SIZE(au1200_mmc0_resources), + .resource = au1200_mmc0_resources, +}; + +static struct resource au1200_mmc1_resources[] = { + [0] = { + .start = SD1_PHYS_ADDR, + .end = SD1_PHYS_ADDR + 0x7ffff, + .flags = IORESOURCE_MEM, + }, + [1] = { + .start = AU1200_SD_INT, + .end = AU1200_SD_INT, + .flags = IORESOURCE_IRQ, + }, + [2] = { + .start = DSCR_CMD0_SDMS_TX1, + .end = DSCR_CMD0_SDMS_TX1, + .flags = IORESOURCE_DMA, + }, + [3] = { + .start = DSCR_CMD0_SDMS_RX1, + .end = DSCR_CMD0_SDMS_RX1, + .flags = IORESOURCE_DMA, + } +}; + +static struct platform_device pb1200_mmc1_device = { + .name = "au1xxx-mmc", + .id = 1, + .dev = { + .dma_mask = &au1xxx_mmc_dmamask, + .coherent_dma_mask = DMA_BIT_MASK(32), + .platform_data = &pb1200_mmc_platdata[1], + }, + .num_resources = ARRAY_SIZE(au1200_mmc1_resources), + .resource = au1200_mmc1_resources, +}; + static struct resource ide_resources[] = { [0] = { .start = IDE_PHYS_ADDR, @@ -163,7 +236,9 @@ static struct platform_device smc91c111_device = { static struct platform_device *board_platform_devices[] __initdata = { &ide_device, - &smc91c111_device + &smc91c111_device, + &pb1200_mmc0_device, + &pb1200_mmc1_device }; static int __init board_register_devices(void) -- 1.7.0.2 From manuel.lauss@googlemail.com Mon Mar 8 20:42:33 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Mon, 08 Mar 2010 20:42:37 +0100 (CET) Received: from mail-fx0-f217.google.com ([209.85.220.217]:47010 "EHLO mail-fx0-f217.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492485Ab0CHTmd (ORCPT ); Mon, 8 Mar 2010 20:42:33 +0100 Received: by fxm9 with SMTP id 9so1912814fxm.24 for ; Mon, 08 Mar 2010 11:42:27 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=googlemail.com; s=gamma; h=domainkey-signature:received:received:from:to:cc:subject:date :message-id:x-mailer; bh=p3kMDNbK4c798AvhRYFxaKkauU7l6S2I/XmUt6pJ71U=; b=BNAfPScKpN2RUuBGyBJXyZw1ASxPWN6Rz6f/+3dohcUUMemq/QXwVCu08yHXqdkur7 K/b8POAImVaYYewHF1lviTIr4QyIMZuTBw7cyBH6dENZsNBm0eMr4x91zNXE9xVOyEdq eEmR5H8rHv8SBfs6wTYL8EyGV9gwLxlvZwmVQ= DomainKey-Signature: a=rsa-sha1; c=nofws; d=googlemail.com; s=gamma; h=from:to:cc:subject:date:message-id:x-mailer; b=VG7hrc9sjA6vuUc0hcjCsG5Kv8KurRefjzeD1HeVapP/N3xqECeMykwHQG1IC+Crt9 t09ZuPCnK38H3T/piDEWepair/UG4aNg2WRDg2833NCdqLKhGSBKaj+Xkr6vj4jgtiZv MD+f7TzwXLL6zF0aamw5FGX7ZDmJOceAcgDjw= Received: by 10.87.63.8 with SMTP id q8mr3661232fgk.3.1268077347576; Mon, 08 Mar 2010 11:42:27 -0800 (PST) Received: from localhost.localdomain (p5496CE16.dip.t-dialin.net [84.150.206.22]) by mx.google.com with ESMTPS id p9sm9071894fkb.33.2010.03.08.11.42.24 (version=TLSv1/SSLv3 cipher=RC4-MD5); Mon, 08 Mar 2010 11:42:26 -0800 (PST) From: Manuel Lauss To: linux-usb@vger.kernel.org Cc: linux-mips@linux-mips.org, Manuel Lauss Subject: [PATCH] USB: ehci-au1xxx does not need EHCI IO watchdog Date: Mon, 8 Mar 2010 20:43:32 +0100 Message-Id: <1268077412-29804-1-git-send-email-manuel.lauss@gmail.com> X-Mailer: git-send-email 1.7.0.2 Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26148 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: manuel.lauss@googlemail.com Precedence: bulk X-list: linux-mips I've been running variations of this patch for well over a year now; my usual zoo of test devices didn't trigger any ill effects even under heavy load. As a nice sideeffect idle-wakeups are reduced from 20/s to about 2/s (EHCI hub with mouse and kbd). Signed-off-by: Manuel Lauss --- drivers/usb/host/ehci-au1xxx.c | 11 ++++++++++- 1 files changed, 10 insertions(+), 1 deletions(-) diff --git a/drivers/usb/host/ehci-au1xxx.c b/drivers/usb/host/ehci-au1xxx.c index e3a74e7..7a27b7c 100644 --- a/drivers/usb/host/ehci-au1xxx.c +++ b/drivers/usb/host/ehci-au1xxx.c @@ -69,6 +69,15 @@ static void au1xxx_stop_ehc(void) au_sync(); } +static int au1xxx_ehci_setup(struct usb_hcd *hcd) +{ + struct ehci_hcd *ehci = hcd_to_ehci(hcd); + int ret = ehci_init(hcd); + + ehci->need_io_watchdog = 0; + return ret; +} + static const struct hc_driver ehci_au1xxx_hc_driver = { .description = hcd_name, .product_desc = "Au1xxx EHCI", @@ -86,7 +95,7 @@ static const struct hc_driver ehci_au1xxx_hc_driver = { * FIXME -- ehci_init() doesn't do enough here. * See ehci-ppc-soc for a complete implementation. */ - .reset = ehci_init, + .reset = au1xxx_ehci_setup, .start = ehci_run, .stop = ehci_stop, .shutdown = ehci_shutdown, -- 1.7.0.2 From yang.shi@windriver.com Tue Mar 9 07:24:36 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Tue, 09 Mar 2010 07:24:41 +0100 (CET) Received: from mail.windriver.com ([147.11.1.11]:32864 "EHLO mail.windriver.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491206Ab0CIGYg (ORCPT ); Tue, 9 Mar 2010 07:24:36 +0100 Received: from localhost.localdomain (pek-lpgbuild1.wrs.com [128.224.153.29]) by mail.windriver.com (8.14.3/8.14.3) with ESMTP id o296ONxo028846; Mon, 8 Mar 2010 22:24:24 -0800 (PST) From: Yang Shi To: ddaney@caviumnetworks.com, ralf@linux-mips.org Cc: linux-mips@linux-mips.org Subject: [PATCH] MIPS: Octeon: Fix wrong variable type in smp.c Date: Tue, 9 Mar 2010 14:24:22 +0800 Message-Id: <1268115862-25976-1-git-send-email-yang.shi@windriver.com> X-Mailer: git-send-email 1.6.0.4 Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26149 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: yang.shi@windriver.com Precedence: bulk X-list: linux-mips Change cvmx_ciu_wdogx_t type to "union cvmx_ciu_wdogx". Signed-off-by: Yang Shi --- arch/mips/cavium-octeon/smp.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/arch/mips/cavium-octeon/smp.c b/arch/mips/cavium-octeon/smp.c index 51e9802..52d61ba 100644 --- a/arch/mips/cavium-octeon/smp.c +++ b/arch/mips/cavium-octeon/smp.c @@ -281,7 +281,7 @@ static void octeon_cpu_die(unsigned int cpu) #ifdef CONFIG_CAVIUM_OCTEON_WATCHDOG /* Disable the watchdog */ - cvmx_ciu_wdogx_t ciu_wdog; + union cvmx_ciu_wdogx ciu_wdog; ciu_wdog.u64 = cvmx_read_csr(CVMX_CIU_WDOGX(cpu)); ciu_wdog.s.mode = 0; cvmx_write_csr(CVMX_CIU_WDOGX(cpu), ciu_wdog.u64); -- 1.6.3.3 From sshtylyov@mvista.com Tue Mar 9 12:34:13 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Tue, 09 Mar 2010 12:34:16 +0100 (CET) Received: from mail-ew0-f224.google.com ([209.85.219.224]:45975 "EHLO mail-ew0-f224.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492507Ab0CILeN (ORCPT ); Tue, 9 Mar 2010 12:34:13 +0100 Received: by ewy24 with SMTP id 24so1006882ewy.27 for ; Tue, 09 Mar 2010 03:34:06 -0800 (PST) Received: by 10.213.37.82 with SMTP id w18mr4210152ebd.97.1268134446140; Tue, 09 Mar 2010 03:34:06 -0800 (PST) Received: from [192.168.2.2] (ppp91-77-52-186.pppoe.mtu-net.ru [91.77.52.186]) by mx.google.com with ESMTPS id 15sm3054865ewy.0.2010.03.09.03.34.04 (version=TLSv1/SSLv3 cipher=RC4-MD5); Tue, 09 Mar 2010 03:34:05 -0800 (PST) Message-ID: <4B963210.7030906@ru.mvista.com> Date: Tue, 09 Mar 2010 14:33:36 +0300 From: Sergei Shtylyov User-Agent: Thunderbird 2.0.0.23 (Windows/20090812) MIME-Version: 1.0 To: Manuel Lauss CC: Linux-MIPS , Manuel Lauss Subject: Re: [PATCH] MIPS: Alchemy: move MMC driver registration to board code. References: <1268076181-29642-1-git-send-email-manuel.lauss@gmail.com> <1268076181-29642-3-git-send-email-manuel.lauss@gmail.com> In-Reply-To: <1268076181-29642-3-git-send-email-manuel.lauss@gmail.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26150 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: sshtylyov@mvista.com Precedence: bulk X-list: linux-mips Hello. Manuel Lauss wrote: > Where it really belongs to. > I disagree (again). SoC platform devices dont belong with the board code. WBR, Sergei From manuel.lauss@googlemail.com Tue Mar 9 12:45:46 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Tue, 09 Mar 2010 12:45:49 +0100 (CET) Received: from mail-fx0-f227.google.com ([209.85.220.227]:49284 "EHLO mail-fx0-f227.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492507Ab0CILpq convert rfc822-to-8bit (ORCPT ); Tue, 9 Mar 2010 12:45:46 +0100 Received: by fxm27 with SMTP id 27so2641296fxm.28 for ; Tue, 09 Mar 2010 03:45:40 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=googlemail.com; s=gamma; h=domainkey-signature:mime-version:received:in-reply-to:references :date:message-id:subject:from:to:cc:content-type :content-transfer-encoding; bh=0/N+Q0B8tVce6yQ+0StSKlNCRx/z3hPiNFtXgXL4BWI=; b=TtvbrAVsiZVH0jqm7pPojwDUn9jBu8Ml9IPH5d/ez8klTXPFiU+h6B0so5jPimx3pT 2Cd34CULaVredJWeaPB4PS+qfOGC2mavjN7Wv6+zicKhqcAcCpJiISjoYdxOOHykh+2R 5/My8lfORRzt7bGYl7XLyBLPQW0KDVzCXxM7Y= DomainKey-Signature: a=rsa-sha1; c=nofws; d=googlemail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type:content-transfer-encoding; b=q5Dlj4uGN6i0mDML2vJlUCkwcqsewVgaslPJBBzz/zoAxMdC+HMBtVErzuRNDPsU3W z7/UjkOrUHmysL4CyN55IPQ3yXpJERqaoPdygrWMVewaQIDPVunq/JaBXGKjEW9nBazu C1LBC5TUaNHdCLl8hPpJW0EtG2Ux+bvKIcGrg= MIME-Version: 1.0 Received: by 10.223.63.208 with SMTP id c16mr5274178fai.29.1268135140086; Tue, 09 Mar 2010 03:45:40 -0800 (PST) In-Reply-To: <4B963210.7030906@ru.mvista.com> References: <1268076181-29642-1-git-send-email-manuel.lauss@gmail.com> <1268076181-29642-3-git-send-email-manuel.lauss@gmail.com> <4B963210.7030906@ru.mvista.com> Date: Tue, 9 Mar 2010 12:45:39 +0100 Message-ID: Subject: Re: [PATCH] MIPS: Alchemy: move MMC driver registration to board code. From: Manuel Lauss To: Sergei Shtylyov Cc: Linux-MIPS , Manuel Lauss Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8BIT Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26151 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: manuel.lauss@googlemail.com Precedence: bulk X-list: linux-mips On Tue, Mar 9, 2010 at 12:33 PM, Sergei Shtylyov wrote: > Hello. > > Manuel Lauss wrote: > >> Where it really belongs to. >> > >  I disagree (again). SoC platform devices dont belong with the board code. Figured as much. However with additional boards the #ifdef mess in common/platform.c is only going to get worse. MUCH worse. Just look at the au1000-eth platform data situation! I have these platform devices on Au1200/Au1300 even thought they don't have a built-in MAC. The board which uses the device should register it. The UARTs are kind of a special case since they need to be fixed up with the correct busclock, but that too could be exported to an initialization wrapper. But, consider the patch withdrawn. Manuel From sshtylyov@mvista.com Tue Mar 9 12:52:18 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Tue, 09 Mar 2010 12:52:22 +0100 (CET) Received: from mail-ew0-f224.google.com ([209.85.219.224]:62431 "EHLO mail-ew0-f224.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492516Ab0CILwS (ORCPT ); Tue, 9 Mar 2010 12:52:18 +0100 Received: by ewy24 with SMTP id 24so1012567ewy.27 for ; Tue, 09 Mar 2010 03:52:12 -0800 (PST) Received: by 10.213.100.151 with SMTP id y23mr4124128ebn.78.1268135532382; Tue, 09 Mar 2010 03:52:12 -0800 (PST) Received: from [192.168.2.2] (ppp91-77-52-186.pppoe.mtu-net.ru [91.77.52.186]) by mx.google.com with ESMTPS id 16sm3051478ewy.11.2010.03.09.03.52.11 (version=TLSv1/SSLv3 cipher=RC4-MD5); Tue, 09 Mar 2010 03:52:11 -0800 (PST) Message-ID: <4B96364E.5050202@mvista.com> Date: Tue, 09 Mar 2010 14:51:42 +0300 From: Sergei Shtylyov User-Agent: Thunderbird 2.0.0.23 (Windows/20090812) MIME-Version: 1.0 To: Manuel Lauss CC: Linux-MIPS , Manuel Lauss Subject: Re: [PATCH] MIPS: Alchemy: move MMC driver registration to board code. References: <1268076181-29642-1-git-send-email-manuel.lauss@gmail.com> <1268076181-29642-3-git-send-email-manuel.lauss@gmail.com> <4B963210.7030906@ru.mvista.com> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26152 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: sshtylyov@mvista.com Precedence: bulk X-list: linux-mips Manuel Lauss wrote: >>> Where it really belongs to. >> I disagree (again). SoC platform devices dont belong with the board code. >> > > Figured as much. However with additional boards the #ifdef mess in > common/platform.c > is only going to get worse. MUCH worse. We could probably eliminate the board #ifdef in platfrom.c by not supplying the platfrom data for MMC1. > Just look at the au1000-eth platform data situation! > I have these platform devices on Au1200/Au1300 even thought they don't have > a built-in MAC. > Need to add the SoC type checks then when registering the devices. Or at least the #ifdef's. :-) > The board which uses the device should register it. Contrarywise, the SoC that has the devices, should register them. > But, consider the patch withdrawn. > Thanks. > Manuel > WBR, Sergei From manuel.lauss@googlemail.com Tue Mar 9 13:03:59 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Tue, 09 Mar 2010 13:04:03 +0100 (CET) Received: from mail-fx0-f227.google.com ([209.85.220.227]:43812 "EHLO mail-fx0-f227.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492509Ab0CIMD7 convert rfc822-to-8bit (ORCPT ); Tue, 9 Mar 2010 13:03:59 +0100 Received: by fxm27 with SMTP id 27so2658572fxm.28 for ; Tue, 09 Mar 2010 04:03:52 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=googlemail.com; s=gamma; h=domainkey-signature:mime-version:received:in-reply-to:references :date:message-id:subject:from:to:cc:content-type :content-transfer-encoding; bh=T0VB6hVTHbfq+mRpJL3Rxliu2gQKKI4p2HjGruflRDg=; b=inWPgeFgi0kGefZK/OGcIg4BYfBXEWsw65bNHp2IoZEPItHBXEYX3C87IPk51siUBb hWxPJjZh8csE90/9Wd68liaSg/GQ82Xv0w6AaRkwYrpd1dneWAP6nCHLlYCY/YwOoj/v uafwRsHq9K87AH4hGvhwN7BFGSCBIaSvFBusY= DomainKey-Signature: a=rsa-sha1; c=nofws; d=googlemail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type:content-transfer-encoding; b=GFn+WAmdJv34fjCKarUGEUSpEq09Ow52I3Kxn6LbLXpsWJoxCcO/WWBwwed7VSlSHZ wrT5EVHuaL7LmUrT+xrIQs8ExGYlvASn1CIlUrVkDDm20R+gYGBCsLeXtW5buYlnV72u J9e+la0O0td2i/5WDrnTULh3sMWwL1XGbui5g= MIME-Version: 1.0 Received: by 10.223.7.90 with SMTP id c26mr1392086fac.19.1268136232029; Tue, 09 Mar 2010 04:03:52 -0800 (PST) In-Reply-To: <4B96364E.5050202@mvista.com> References: <1268076181-29642-1-git-send-email-manuel.lauss@gmail.com> <1268076181-29642-3-git-send-email-manuel.lauss@gmail.com> <4B963210.7030906@ru.mvista.com> <4B96364E.5050202@mvista.com> Date: Tue, 9 Mar 2010 13:03:51 +0100 Message-ID: Subject: Re: [PATCH] MIPS: Alchemy: move MMC driver registration to board code. From: Manuel Lauss To: Sergei Shtylyov Cc: Linux-MIPS , Manuel Lauss Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8BIT Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26153 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: manuel.lauss@googlemail.com Precedence: bulk X-list: linux-mips On Tue, Mar 9, 2010 at 12:51 PM, Sergei Shtylyov wrote: > Manuel Lauss wrote: > >>>> Where it really belongs to. >>> >>> I disagree (again). SoC platform devices dont belong with the board code. >>> >> >> Figured as much.  However with additional boards the #ifdef mess in >> common/platform.c >> is only going to get worse. MUCH worse. > >  We could probably eliminate the board #ifdef in platfrom.c by not supplying > the platfrom data for MMC1. What if I wanted to build a kernel which supports multiple different Au1200-based systems, like the SH-port does with its mach vector? >> Just look at the au1000-eth platform data situation! >> I have these platform devices on Au1200/Au1300 even thought they don't >> have >> a built-in MAC. >> > >  Need to add the SoC type checks then when registering the devices. Or at > least the #ifdef's. :-) I'd like to get rid of the ifdefs, not encourage them to mate and multiply ;-) >> The board which uses the device should register it. > >  Contrarywise, the SoC that has the devices, should register them. My point is that most drivers require additional information from the board, and maybe due to hardware design the ids may need to be swapped. Rather than #ifdeffing these cases for every board in a central file I'd let the board using the devices sort this out. In my case, I don't need UART0 of the Au1200, but need UART1 to be ttyS0. And on a personal note, that file just bothers me. It's messy, can cause merge conflicts, it references structures defined inside board-specific code. In short, it just plain annoys my sense of aesthetics. Manuel From ralf@linux-mips.org Tue Mar 9 15:16:22 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Tue, 09 Mar 2010 15:16:26 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:56180 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492147Ab0CIOQW (ORCPT ); Tue, 9 Mar 2010 15:16:22 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o29EGC1g030848; Tue, 9 Mar 2010 15:16:15 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o29EG9A8030695; Tue, 9 Mar 2010 15:16:09 +0100 Date: Tue, 9 Mar 2010 15:16:06 +0100 From: Ralf Baechle To: David Daney Cc: Yang Shi , f.fainelli@gmail.com, linux-mips@linux-mips.org Subject: Re: [PATCH 3/3] MIPS: Octeon: Add add_wired_entry decralation in header file Message-ID: <20100309141605.GA30829@linux-mips.org> References: <1267605801-5305-1-git-send-email-yang.shi@windriver.com> <50e36e8549769a26986f99a23772d23fd039c230.1267604875.git.yang.shi@windriver.com> <004eb64c73b3bcec90612663598ada4cf678f236.1267604875.git.yang.shi@windriver.com> <4B8EA94B.1020203@caviumnetworks.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <4B8EA94B.1020203@caviumnetworks.com> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26154 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Wed, Mar 03, 2010 at 10:24:11AM -0800, David Daney wrote: > On 03/03/2010 12:43 AM, Yang Shi wrote: > >Octeon's setup.c uses add_wired_entry, but it is not declared > >anywhere. Copy add_wired_entry decralation from pgtable-32.h to > >pgtable-64.h and include asm/pgtable.h into Octeon's setup.c. > > > >Signed-off-by: Yang Shi > > NAK! > > We are removing the use of add_wired_entry(), so adding a > declaration will not be necessary. For explanation - add_wired_entry is a horrible API; it requires the user to have knowledge about the TLB structure, code differently for 32-bit and 64-bit kernels. It's just the API of terror. Ralf From f.fainelli@gmail.com Tue Mar 9 15:47:07 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Tue, 09 Mar 2010 15:47:13 +0100 (CET) Received: from fg-out-1718.google.com ([72.14.220.155]:17526 "EHLO fg-out-1718.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492055Ab0CIOrH (ORCPT ); Tue, 9 Mar 2010 15:47:07 +0100 Received: by fg-out-1718.google.com with SMTP id d23so63206fga.6 for ; Tue, 09 Mar 2010 06:47:06 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:sender:from:date:subject :mime-version:x-uid:x-length:organization:to:cc:content-type :content-transfer-encoding:message-id; bh=mq78ZMY0IcvTwZ1rgy0FOKXeB+5Lyyf9f5daJF79cGQ=; b=kM2POItohemHxMT6bhfiK3U2O9myO/i8xoC5OuRDjOeay1T+1A+sKD13iq11+daOye HdfuLz7/vWWY1UhNxjtkIy8DcD9erkYSE9ychCHl5x1/FTLSBOl0wT4smVPZ/7bePMVG ddxUxVcGShP3sLpwiixsWFHWaBR9yzrAV7Iec= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=sender:from:date:subject:mime-version:x-uid:x-length:organization :to:cc:content-type:content-transfer-encoding:message-id; b=De6YrtQhjAlVE19T/f/c4oFiQPHiv2PF/ccCny3nqXkOTM1XFsRl2nygndqesjyIxd wiI1X+rtu1tOgT8H8ExS3r3dc6I4z75QBg2Uyox0cwzXQ/ri2kqArAa9I2lSYFwJNokO bTjjPfsNnl05O09wxRFnwDadc4nPxBczgim8I= Received: by 10.87.72.22 with SMTP id z22mr272651fgk.35.1268146026597; Tue, 09 Mar 2010 06:47:06 -0800 (PST) Received: from flexo.localnet (bobafett.staff.proxad.net [213.228.1.121]) by mx.google.com with ESMTPS id l19sm6220744fgb.1.2010.03.09.06.47.05 (version=SSLv3 cipher=RC4-MD5); Tue, 09 Mar 2010 06:47:05 -0800 (PST) From: Florian Fainelli Date: Tue, 9 Mar 2010 15:46:01 +0100 Subject: [PATCH] MIPS: make CAC_ADDR and UNCAC_ADDR account for PHYS_OFFSET MIME-Version: 1.0 X-UID: 24871 X-Length: 3378 Organization: OpenWrt To: linux-mips@linux-mips.org Cc: ralf@linux-mips.org Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit Message-Id: <201003091546.01281.florian@openwrt.org> Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26155 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: florian@openwrt.org Precedence: bulk X-list: linux-mips On AR7, we already redefine PHYS_OFFSET to match the system specifities, it is however not sufficient when unsing dma_{map,unmap}_single, specifically in the Ethernet driver, we must also adjust CAC_ADDR and UNCAC_ADDR for DMA to work correctly. This patch fixes the following issue, seen in cpmac_open: ops[#1]: Cpu 0 $ 0 : 00000000 10008400 a0f5b120 00000000 $ 4 : 94c59000 94270f64 00000020 00000010 $ 8 : 00000010 94103ce0 0000000a 94c03400 $12 : ffffffff 94c03408 94c03410 00000001 $16 : a0f5ba20 00000041 94c592c0 94c59200 $20 : 94c59000 000005ee 00002000 9438c8f0 $24 : 00000010 00000000 $28 : 94fac000 94fadd58 94390000 942724a8 Hi : 00000000 Lo : 00000001 epc : 94272518 cpmac_open+0x208/0x3f8 Not tainted ra : 942724a8 cpmac_open+0x198/0x3f8 Status: 10008403 KERNEL EXL IE Cause : 3080000c BadVA : 00000000 PrId : 00018448 (MIPS 4KEc) Modules linked in: Process ifconfig (pid: 278, threadinfo=94fac000, task=94e79590, tls=00000000) Stack : 7f8da120 2ab05cb0 94c59000 943356f0 00000000 943d0000 94c59000 943356f0 94c59030 943d0000 943c27c0 94fade10 00000000 94fade20 94c59000 9428e5a4 00000000 94c59000 00000041 94289768 94c59000 00000041 00001002 00001043 00000000 9428d810 00000000 94fade10 7f8da4e8 9428e6b8 00000000 7f8da4a8 7f8da4e8 00008914 00000000 942f7f2c 00000000 00000008 00408000 00008913 ... Call Trace: [<94272518>] cpmac_open+0x208/0x3f8 [<9428e5a4>] dev_open+0x164/0x264 [<9428d810>] dev_change_flags+0xd0/0x1bc [<942f7f2c>] devinet_ioctl+0x2d8/0x908 [<942771f8>] sock_ioctl+0x29c/0x2fc [<941a0fb4>] vfs_ioctl+0x2c/0x7c [<941a16ec>] do_vfs_ioctl+0x5dc/0x630 [<941a1790>] sys_ioctl+0x50/0x88 [<94101e10>] stack_done+0x20/0x3c Signed-off-by: Regards, Florian Fainelli --- diff --git a/arch/mips/include/asm/page.h b/arch/mips/include/asm/page.h index ac32572..7b11df5 100644 --- a/arch/mips/include/asm/page.h +++ b/arch/mips/include/asm/page.h @@ -188,8 +188,10 @@ typedef struct { unsigned long pgprot; } pgprot_t; #define VM_DATA_DEFAULT_FLAGS (VM_READ | VM_WRITE | VM_EXEC | \ VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC) -#define UNCAC_ADDR(addr) ((addr) - PAGE_OFFSET + UNCAC_BASE) -#define CAC_ADDR(addr) ((addr) - UNCAC_BASE + PAGE_OFFSET) +#define UNCAC_ADDR(addr) ((addr) - PAGE_OFFSET + UNCAC_BASE + \ + PHYS_OFFSET) +#define CAC_ADDR(addr) ((addr) - UNCAC_BASE + PAGE_OFFSET + \ + PHYS_OFFSET) #include #include From wuzhangjin@gmail.com Tue Mar 9 17:34:06 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Tue, 09 Mar 2010 17:34:11 +0100 (CET) Received: from mail-fx0-f227.google.com ([209.85.220.227]:34863 "EHLO mail-fx0-f227.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491194Ab0CIQeG (ORCPT ); Tue, 9 Mar 2010 17:34:06 +0100 Received: by fxm27 with SMTP id 27so2968813fxm.28 for ; Tue, 09 Mar 2010 08:34:01 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:from:to:cc:subject:date :message-id:x-mailer; bh=rYOHgFXmHgPDw/W+yk0UbExd6NQrL5qkhiDnUvuwqjk=; b=XPjgWVMPU4FYt4zHHTKf4LliGPz6OZmz6LJhKwXmoA/zzNjGMVQIdQx/mOmKyQzPjo 4gQ9Ig26bAcWkn0/flErUUP2gMMNuP1K/LRAi/IAxxblIzZRViqJbsZTT6AAn3HZRtta /oYnuQqiRuu0vmm97BuDNlf+dZqPHEmhA2xJQ= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=from:to:cc:subject:date:message-id:x-mailer; b=IE4stG8S8mG1QzR+ANhUaCZE091yVZB4bxJcsGrCCEjm01gniZi0gsvfaAF9gwFFPY NqmCysQzrmXlyyGi3WvHHkOFpsrG+f0YMfD5x+P4Hd8q7B5pxigJDqMIKNoNQu/vnIzV jRMfG7xLfSAcGXu8viu/FxL0KRNpJyo0vBKpg= Received: by 10.223.77.85 with SMTP id f21mr29236fak.40.1268152441443; Tue, 09 Mar 2010 08:34:01 -0800 (PST) Received: from localhost.localdomain ([202.201.12.142]) by mx.google.com with ESMTPS id p9sm11297164fkb.33.2010.03.09.08.33.57 (version=TLSv1/SSLv3 cipher=RC4-MD5); Tue, 09 Mar 2010 08:34:00 -0800 (PST) From: Wu Zhangjin To: Ralf Baechle Cc: linux-mips@linux-mips.org, Steven Rostedt , Frederic Weisbecker , Ingo Molnar , Wu Zhangjin Subject: [PATCH] MIPS: Don't trace irqsoff for idle Date: Wed, 10 Mar 2010 00:27:28 +0800 Message-Id: <1268152048-30522-1-git-send-email-wuzhangin@gmail.com> X-Mailer: git-send-email 1.7.0.1 Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26156 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: wuzhangjin@gmail.com Precedence: bulk X-list: linux-mips From: Wu Zhangjin As the X86 platform did in arch/x86/kernel/{process_32.c,process_64.c}, we also don't trace irqsoff for idle. If "There's no useful work to be done", we don't care about the irqsoff duration. If we trace for idle, the max duration of irqsoff will be always the idle time and eventually make the irqsoff tracer out of action. Signed-off-by: Wu Zhangjin --- arch/mips/kernel/process.c | 7 ++++++- 1 files changed, 6 insertions(+), 1 deletions(-) diff --git a/arch/mips/kernel/process.c b/arch/mips/kernel/process.c index f3d73e1..87d19dd 100644 --- a/arch/mips/kernel/process.c +++ b/arch/mips/kernel/process.c @@ -64,8 +64,13 @@ void __noreturn cpu_idle(void) smtc_idle_loop_hook(); #endif - if (cpu_wait) + + if (cpu_wait) { + /* Don't trace irqs off for idle */ + stop_critical_timings(); (*cpu_wait)(); + start_critical_timings(); + } } #ifdef CONFIG_HOTPLUG_CPU if (!cpu_online(cpu) && !cpu_isset(cpu, cpu_callin_map) && -- 1.7.0.1 From wuzhangjin@gmail.com Tue Mar 9 18:00:01 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Tue, 09 Mar 2010 18:00:07 +0100 (CET) Received: from mail-fx0-f227.google.com ([209.85.220.227]:40734 "EHLO mail-fx0-f227.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492259Ab0CIRAB (ORCPT ); Tue, 9 Mar 2010 18:00:01 +0100 Received: by fxm27 with SMTP id 27so2998744fxm.28 for ; Tue, 09 Mar 2010 08:59:54 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:from:to:cc:subject:date :message-id:x-mailer; bh=bPQ808PTuOzOs9H1lNEBXTRB+1OBivpqMYhXWPYuFNo=; b=JVwge/m92trriRcUHLhO83vZipa7EH5NIbTNWrg642drTZJ+wekxHEBLOP5BFrY6F+ 15J70vpNl7uOP675revfRSk37KwQtpsYkIafHfYDxl5op97mEsVOcJUFt7CzzTQ5iuzU MnhamojHxjojbYXsqjqHE2av6pnY04rFAmses= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=from:to:cc:subject:date:message-id:x-mailer; b=A4vVcdfd1nUBugEnoOWNnkPzY8KXZzorA5aBdwTr/5OYQCf14apBuNaUFtJRChhug/ CK6XJuPD3oN+4+1MXiXK6K8BMHWxFk6CbjGR1iCKtlAKRWPge4r8mqSZYMHZmEediSe9 Kzee6TloiTfWVbUqV/taPdRYHCc04WTNBesY0= Received: by 10.223.81.90 with SMTP id w26mr101400fak.9.1268153994336; Tue, 09 Mar 2010 08:59:54 -0800 (PST) Received: from localhost.localdomain ([202.201.12.142]) by mx.google.com with ESMTPS id z15sm11394784fkz.21.2010.03.09.08.59.51 (version=TLSv1/SSLv3 cipher=RC4-MD5); Tue, 09 Mar 2010 08:59:53 -0800 (PST) From: Wu Zhangjin To: Ralf Baechle Cc: linux-mips@linux-mips.org, Wu Zhangjin Subject: [PATCH] Loongson: Add module info to the loongson2_clock driver Date: Wed, 10 Mar 2010 00:53:21 +0800 Message-Id: <1268153601-4396-1-git-send-email-wuzhangin@gmail.com> X-Mailer: git-send-email 1.7.0.1 Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26157 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: wuzhangjin@gmail.com Precedence: bulk X-list: linux-mips From: Wu Zhangjin This patch fixes the warning of the tool which checks the module info of the loongson2_clock driver when inserting it into the kernel: "Feb 25 23:42:27 localhost kernel: [ 4.965000] loongson2_clock: module license 'unspecified' taints kernel." Reported-by: Liu Shiwei Signed-off-by: Wu Zhangjin --- arch/mips/kernel/cpufreq/loongson2_clock.c | 4 ++++ 1 files changed, 4 insertions(+), 0 deletions(-) diff --git a/arch/mips/kernel/cpufreq/loongson2_clock.c b/arch/mips/kernel/cpufreq/loongson2_clock.c index d7ca256..3eaae1d 100644 --- a/arch/mips/kernel/cpufreq/loongson2_clock.c +++ b/arch/mips/kernel/cpufreq/loongson2_clock.c @@ -164,3 +164,7 @@ void loongson2_cpu_wait(void) spin_unlock_irqrestore(&loongson2_wait_lock, flags); } EXPORT_SYMBOL_GPL(loongson2_cpu_wait); + +MODULE_AUTHOR("Yanhua "); +MODULE_DESCRIPTION("cpuclock driver of Loongson2F"); +MODULE_LICENSE("GPL"); -- 1.7.0.1 From wuzhangjin@gmail.com Tue Mar 9 18:19:15 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Tue, 09 Mar 2010 18:19:23 +0100 (CET) Received: from mail-fx0-f227.google.com ([209.85.220.227]:63765 "EHLO mail-fx0-f227.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492254Ab0CIRTO (ORCPT ); Tue, 9 Mar 2010 18:19:14 +0100 Received: by fxm27 with SMTP id 27so3020549fxm.28 for ; Tue, 09 Mar 2010 09:19:08 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:from:to:cc:subject:date :message-id:x-mailer:mime-version:content-type :content-transfer-encoding; bh=mupSUUzEgodFDhcU6NPOe2Qucdsh8bnmY1TEvcbx5+8=; b=Pvk6Yd8+oUM+FT4zPn1p1Pekg0xLS6rUo7cNoS7/+sSooVzRzIAftJWBOThVkGyv3Q mB05A5H4nocIZDbxJR1vb/p2OKaskF1LNvrn51Wl9VnI8vJW4fmdT4Tgqr9J54DHZ3ha CcZ4yXMQ0S+tMlnuRtrKPKaFb8CZtlPYcPBes= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=from:to:cc:subject:date:message-id:x-mailer:mime-version :content-type:content-transfer-encoding; b=NPrO1epbVuuoF6XIpLKq/2lAWvUAcCODQuO159lSYBZX5GnrMrjC9Sw/0FayTrqwrf rrMkprdzqwe4MwtMggdjeW+lg6iftBLUldzHl1zUN0W0X00wzcaZwrydcEeG2OT4KuNd lJCWiA+1f4vm0akvK/cKZsyngq+5t330QcAHI= Received: by 10.223.36.92 with SMTP id s28mr109255fad.28.1268155146307; Tue, 09 Mar 2010 09:19:06 -0800 (PST) Received: from localhost.localdomain ([202.201.12.142]) by mx.google.com with ESMTPS id 31sm11418183fkt.47.2010.03.09.09.19.02 (version=TLSv1/SSLv3 cipher=RC4-MD5); Tue, 09 Mar 2010 09:19:05 -0800 (PST) From: Wu Zhangjin To: Ralf Baechle , Greg KH Cc: linux-mips@linux-mips.org, linux-kernel@vger.kernel.org, Wu Zhangjin Subject: [PATCH 0/3] Workaround the Out-of-order Issue of Loongson-2F Date: Wed, 10 Mar 2010 01:12:30 +0800 Message-Id: X-Mailer: git-send-email 1.7.0.1 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26158 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: wuzhangjin@gmail.com Precedence: bulk X-list: linux-mips From: Wu Zhangjin As the Chapter 15: "Errata: Issue of Out-of-order in loongson"[1] shows, to workaround the Issue of Loongson-2F，We need to do: o When switching from user model to kernel model, you should flush the branch target history such as BTB and RAS. o Doing some tricks to the indirect branch target to make sure that the indirect branch target can not be in the I/O region. This patchset applied the above methods and for the binutils patch[3] have been merged into binutils 2.20.1, so, it's time to upstream this patchset now. without this patch, the machines will hang when the instruction sequence hit the Out-of-order Issue of Loongson-2F, therefore, this patchset is very urgent for both 2.6.33 and 2.6.34. [1] Chinese Version: http://www.loongson.cn/uploadfile/file/20080821113149.pdf [2] English Version of Chapter 15: http://groups.google.com.hk/group/loongson-dev/msg/e0d2e220958f10a6?dmode=source [3] http://sourceware.org/ml/binutils/2009-11/msg00387.html Regards, Wu Zhangjin Wu Zhangjin (3): Loongson-2F: Flush the branch target history such as BTB and RAS Loongson-2F: Enable fixups of binutils 2.20.1 Loongson-2F: Fixup of problems introduced by -mfix-loongson2f-jump of binutils 2.20.1 arch/mips/Makefile | 4 +++- arch/mips/include/asm/stackframe.h | 19 +++++++++++++++++++ arch/mips/loongson/common/reset.c | 12 +++++++++++- 3 files changed, 33 insertions(+), 2 deletions(-) From wuzhangjin@gmail.com Tue Mar 9 18:19:42 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Tue, 09 Mar 2010 18:19:48 +0100 (CET) Received: from mail-fx0-f227.google.com ([209.85.220.227]:63765 "EHLO mail-fx0-f227.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492262Ab0CIRT2 (ORCPT ); Tue, 9 Mar 2010 18:19:28 +0100 Received: by mail-fx0-f227.google.com with SMTP id 27so3020549fxm.28 for ; Tue, 09 Mar 2010 09:19:28 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:from:to:cc:subject:date :message-id:x-mailer:in-reply-to:references:in-reply-to:references :mime-version:content-type:content-transfer-encoding; bh=Zhsk794ng9QQOuZI3ijPZPOhqctJa1+QjFi5slISCbo=; b=P+uFP4FrzoZFH8s75Td9RAqlWEPIOKgMmnt8OvUtPIjyo1LuH5x7V0E+Y4nydgr63c NCpdA6b8bNJC76SFvclfeHlvqa/5K4Zq0iYeQRpsAxCZN6xkuS7boiPieAR5Z/B0BSkf hU5/TP3+R4wjWipFPRfYCgLduzxyjDVPJnnOk= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=from:to:cc:subject:date:message-id:x-mailer:in-reply-to:references :mime-version:content-type:content-transfer-encoding; b=uws2FXVvp60uh7KSEz2R5A8Q8PLpEBI38zyYfcrAYMP/nFb5otPiFBhN3XFPfwxUE5 S+2fQ8gWZ1ePN9JBq2wJ8yU2HArwUqw+tg4ec5nFMPm8GuFBoL2s+3g6fxkOaJaxw+Yf K4NPDVIzS14AUh4r60Ln2VUK4W+hvSRHpnPUM= Received: by 10.223.77.85 with SMTP id f21mr100743fak.40.1268155166516; Tue, 09 Mar 2010 09:19:26 -0800 (PST) Received: from localhost.localdomain ([202.201.12.142]) by mx.google.com with ESMTPS id 31sm11418183fkt.47.2010.03.09.09.19.22 (version=TLSv1/SSLv3 cipher=RC4-MD5); Tue, 09 Mar 2010 09:19:25 -0800 (PST) From: Wu Zhangjin To: Ralf Baechle , Greg KH Cc: linux-mips@linux-mips.org, linux-kernel@vger.kernel.org, Wu Zhangjin Subject: [PATCH 1/3] Loongson-2F: Flush the branch target history such as BTB and RAS Date: Wed, 10 Mar 2010 01:12:31 +0800 Message-Id: X-Mailer: git-send-email 1.7.0.1 In-Reply-To: References: In-Reply-To: References: MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26159 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: wuzhangjin@gmail.com Precedence: bulk X-list: linux-mips From: Wu Zhangjin As the Chapter 15: "Errata: Issue of Out-of-order in loongson"[1] shows, to workaround the Issue of Loongson-2F，We need to do: "When switching from user model to kernel model, you should flush the branch target history such as BTB and RAS." This patch did clear BTB(branch target buffer), forbid RAS(row address strobe) via Loongson-2F's 64bit diagnostic register. [1] Chinese Version: http://www.loongson.cn/uploadfile/file/200808211 [2] English Version of Chapter 15: http://groups.google.com.hk/group/loongson-dev/msg/e0d2e220958f10a6?dmode=source Signed-off-by: Wu Zhangjin --- arch/mips/include/asm/stackframe.h | 19 +++++++++++++++++++ 1 files changed, 19 insertions(+), 0 deletions(-) diff --git a/arch/mips/include/asm/stackframe.h b/arch/mips/include/asm/stackframe.h index 3b6da33..b84cfda 100644 --- a/arch/mips/include/asm/stackframe.h +++ b/arch/mips/include/asm/stackframe.h @@ -121,6 +121,25 @@ .endm #else .macro get_saved_sp /* Uniprocessor variation */ +#ifdef CONFIG_CPU_LOONGSON2F + /* + * Clear BTB(branch target buffer), forbid RAS(row address + * strobe) to workaround the Out-of-oder Issue in Loongson2F + * via it's diagnostic register. + */ + move k0, ra + jal 1f + nop +1: jal 1f + nop +1: jal 1f + nop +1: jal 1f + nop +1: move ra, k0 + li k0, 3 + mtc0 k0, $22 +#endif /* CONFIG_CPU_LOONGSON2F */ #if defined(CONFIG_32BIT) || defined(KBUILD_64BIT_SYM32) lui k1, %hi(kernelsp) #else -- 1.7.0.1 From wuzhangjin@gmail.com Tue Mar 9 18:20:08 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Tue, 09 Mar 2010 18:20:13 +0100 (CET) Received: from mail-fx0-f227.google.com ([209.85.220.227]:63765 "EHLO mail-fx0-f227.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492263Ab0CIRTg (ORCPT ); Tue, 9 Mar 2010 18:19:36 +0100 Received: by mail-fx0-f227.google.com with SMTP id 27so3020549fxm.28 for ; Tue, 09 Mar 2010 09:19:36 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:from:to:cc:subject:date :message-id:x-mailer:in-reply-to:references:in-reply-to:references; bh=P2v0PzTZ8PcrOj5xc9fJWbeIzqwSM6dukdMjQ4UhZUY=; b=RTZarxeffifNZA54oC1C7AsBCg/tZUEkRuS/Hk4uPXIagB5sibDYN3FNExCzbsRV0L dWYz4zTXYYvmg5MlfYjsC1gtNFqEUBFOoabWl5ZXeDk243GJsvxHk/72gxLPDuTqEPn4 SR/61NZshyJLl98CW8BNXoMZTJ/cmsrcAyOT8= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=from:to:cc:subject:date:message-id:x-mailer:in-reply-to:references; b=O4b1L1zn0hEsqd9PqPfDUppr5b1vo/JqNCX6eKXzYwPipD8OO1BsrfhbNeGfs3L4w0 +2K5W9NcTtUj290a5fXEcfnUj1qqYHDm+46sR6U6CfVyNfM5/4JOyN8FQ8QTCvduf3mE XWGRJNBA+Voi7c7BRxQlkjbBa7FmkK18D/o9E= Received: by 10.223.5.71 with SMTP id 7mr90998fau.48.1268155175101; Tue, 09 Mar 2010 09:19:35 -0800 (PST) Received: from localhost.localdomain ([202.201.12.142]) by mx.google.com with ESMTPS id 31sm11418183fkt.47.2010.03.09.09.19.30 (version=TLSv1/SSLv3 cipher=RC4-MD5); Tue, 09 Mar 2010 09:19:34 -0800 (PST) From: Wu Zhangjin To: Ralf Baechle , Greg KH Cc: linux-mips@linux-mips.org, linux-kernel@vger.kernel.org, Wu Zhangjin , Zhang Le , Wu Zhangjin Subject: [PATCH 2/3] Loongson-2F: Enable fixups of binutils 2.20.1 Date: Wed, 10 Mar 2010 01:12:32 +0800 Message-Id: <6ca9906a5469bc5cc4aafeac6f90ab3798a5a837.1268153722.git.wuzhangjin@gmail.com> X-Mailer: git-send-email 1.7.0.1 In-Reply-To: References: In-Reply-To: References: Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26160 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: wuzhangjin@gmail.com Precedence: bulk X-list: linux-mips From: Wu Zhangjin As the "Fixups of Loongson2F" patch[1] to binutils have been applied into binutils 2.20.1. It's time to enable the options provided by the patch to compile the kernel. Without these fixups, the system will hang unexpectedly for the bug of processor. To learn more about these fixups, please refer to the following references. [1] "Fixups of Loongson2F" patch for binutils(actually for gas) http://sourceware.org/ml/binutils/2009-11/msg00387.html [2] Chapter 15 of "Loongson2F User Manual"(Chinese Version) http://www.loongson.cn/uploadfile/file/200808211 [3] English Version of the above chapter 15 http://groups.google.com.hk/group/loongson-dev/msg/e0d2e220958f10a6?dmode=source Signed-off-by: Zhang Le Signed-off-by: Wu Zhangjin --- arch/mips/Makefile | 4 +++- 1 files changed, 3 insertions(+), 1 deletions(-) diff --git a/arch/mips/Makefile b/arch/mips/Makefile index 2f2eac2..5ae342e 100644 --- a/arch/mips/Makefile +++ b/arch/mips/Makefile @@ -135,7 +135,9 @@ cflags-$(CONFIG_CPU_LOONGSON2) += -Wa,--trap cflags-$(CONFIG_CPU_LOONGSON2E) += \ $(call cc-option,-march=loongson2e,-march=r4600) cflags-$(CONFIG_CPU_LOONGSON2F) += \ - $(call cc-option,-march=loongson2f,-march=r4600) + $(call cc-option,-march=loongson2f,-march=r4600) \ + $(call as-option,-Wa$(comma)-mfix-loongson2f-nop,) \ + $(call as-option,-Wa$(comma)-mfix-loongson2f-jump,) cflags-$(CONFIG_CPU_MIPS32_R1) += $(call cc-option,-march=mips32,-mips32 -U_MIPS_ISA -D_MIPS_ISA=_MIPS_ISA_MIPS32) \ -Wa,-mips32 -Wa,--trap -- 1.7.0.1 From wuzhangjin@gmail.com Tue Mar 9 18:20:32 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Tue, 09 Mar 2010 18:20:35 +0100 (CET) Received: from mail-fx0-f227.google.com ([209.85.220.227]:63765 "EHLO mail-fx0-f227.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492254Ab0CIRTm (ORCPT ); Tue, 9 Mar 2010 18:19:42 +0100 Received: by mail-fx0-f227.google.com with SMTP id 27so3020549fxm.28 for ; Tue, 09 Mar 2010 09:19:42 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:from:to:cc:subject:date :message-id:x-mailer:in-reply-to:references:in-reply-to:references; bh=QS8NJKpJm7YwLPEKWb7eGN6Hw+7ud0Ynctl57TgyVWA=; b=uwm6cq8ubto3z5R0kImLWGQTlU8tPLMNJXY/mbicJlW4yli6oH0e2jTXK2NpPwDObc bUDpQlmFXaHULkdBoPWyJl95VUh0edydTNjTwEAVi2KgzWMQfV7FmUeLfEoVsWBY0nms mvHJYQVseTt5i0J5Ms4Y4IsdLPCypoSyiiLdc= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=from:to:cc:subject:date:message-id:x-mailer:in-reply-to:references; b=LZKJXvmYl3FiyJQRohIaxVfgAXVqnoYZJ1Cvf/Gn37ztET6qiu1BTeAovX4N2ImM6t nM7/HCgL4JAJFscHkeaUQJMbpNZ+xv9NSS9IZAB59CqZLhw0q1IuCfRe9cWHQQl2pWQm vv2O6aWM2F/ywWlbrsqLpSqfiFOa8/4kgmqn0= Received: by 10.223.57.133 with SMTP id c5mr130357fah.11.1268155180413; Tue, 09 Mar 2010 09:19:40 -0800 (PST) Received: from localhost.localdomain ([202.201.12.142]) by mx.google.com with ESMTPS id 31sm11418183fkt.47.2010.03.09.09.19.36 (version=TLSv1/SSLv3 cipher=RC4-MD5); Tue, 09 Mar 2010 09:19:39 -0800 (PST) From: Wu Zhangjin To: Ralf Baechle , Greg KH Cc: linux-mips@linux-mips.org, linux-kernel@vger.kernel.org, Wu Zhangjin Subject: [PATCH 3/3] Loongson-2F: Fixup of problems introduced by -mfix-loongson2f-jump of binutils 2.20.1 Date: Wed, 10 Mar 2010 01:12:33 +0800 Message-Id: <1f3b69fc1a72f55f736c13307c2e7e03b07eb2d1.1268153722.git.wuzhangjin@gmail.com> X-Mailer: git-send-email 1.7.0.1 In-Reply-To: References: In-Reply-To: References: Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26161 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: wuzhangjin@gmail.com Precedence: bulk X-list: linux-mips From: Wu Zhangjin The -mfix-loongson2f-jump option provided by the binutils 2.20.1 have fixed the Out-of-order Issue of Loongson-2F described in Chapter 15 of "Loongson2F User Manual"[1,2], but introduced some problems. The option changes all of the jumping target to "addr & 0xcfffffff" through the at($1) register, but for the REBOOT address of loongson-2F: 0xbfc00000, this is totally wrong, so, this patch try to avoid the problem via telling the assembler not to use at($1) register. [1] Loongson2F User Manual(Chinese Version) http://www.loongson.cn/uploadfile/file/200808211 [2] English Version of Chapter 15: http://groups.google.com.hk/group/loongson-dev/msg/e0d2e220958f10a6?dmode=source Reported-and-tested-by: Liu Shiwei Signed-off-by: Wu Zhangjin --- arch/mips/loongson/common/reset.c | 12 +++++++++++- 1 files changed, 11 insertions(+), 1 deletions(-) diff --git a/arch/mips/loongson/common/reset.c b/arch/mips/loongson/common/reset.c index 4bd9c18..d5f1a50 100644 --- a/arch/mips/loongson/common/reset.c +++ b/arch/mips/loongson/common/reset.c @@ -21,8 +21,18 @@ static void loongson_restart(char *command) /* do preparation for reboot */ mach_prepare_reboot(); - /* reboot via jumping to boot base address */ + /* reboot via jumping to boot base address + * + * ".set noat" and ".set at" are used to ensure the address not broken + * by the -mfix-loongson2f-jump option provided by binutils 2.20.1 and + * higher which try to change the jumping address to "addr & + * 0xcfffffff" via the at($1) register, this is totally wrong for + * 0xbfc00000(LOONGSON_BOOT_BASE). + */ + + __asm__ __volatile__(".set noat\n"); ((void (*)(void))ioremap_nocache(LOONGSON_BOOT_BASE, 4)) (); + __asm__ __volatile__(".set at\n"); } static void loongson_poweroff(void) -- 1.7.0.1 From ralf@linux-mips.org Tue Mar 9 20:03:13 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Tue, 09 Mar 2010 20:03:17 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:35009 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492538Ab0CITDN (ORCPT ); Tue, 9 Mar 2010 20:03:13 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o29J3BBV001099; Tue, 9 Mar 2010 20:03:11 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o29J3942001096; Tue, 9 Mar 2010 20:03:09 +0100 Date: Tue, 9 Mar 2010 20:03:09 +0100 From: Ralf Baechle To: Yang Shi Cc: ddaney@caviumnetworks.com, linux-mips@linux-mips.org Subject: Re: [PATCH] MIPS: Octeon: Fix wrong variable type in smp.c Message-ID: <20100309190309.GA301@linux-mips.org> References: <1268115862-25976-1-git-send-email-yang.shi@windriver.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1268115862-25976-1-git-send-email-yang.shi@windriver.com> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26162 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Tue, Mar 09, 2010 at 02:24:22PM +0800, Yang Shi wrote: > Change cvmx_ciu_wdogx_t type to "union cvmx_ciu_wdogx". > > Signed-off-by: Yang Shi > --- > arch/mips/cavium-octeon/smp.c | 2 +- > 1 files changed, 1 insertions(+), 1 deletions(-) > > diff --git a/arch/mips/cavium-octeon/smp.c b/arch/mips/cavium-octeon/smp.c > index 51e9802..52d61ba 100644 > --- a/arch/mips/cavium-octeon/smp.c > +++ b/arch/mips/cavium-octeon/smp.c > @@ -281,7 +281,7 @@ static void octeon_cpu_die(unsigned int cpu) > > #ifdef CONFIG_CAVIUM_OCTEON_WATCHDOG > /* Disable the watchdog */ > - cvmx_ciu_wdogx_t ciu_wdog; > + union cvmx_ciu_wdogx ciu_wdog; > ciu_wdog.u64 = cvmx_read_csr(CVMX_CIU_WDOGX(cpu)); > ciu_wdog.s.mode = 0; > cvmx_write_csr(CVMX_CIU_WDOGX(cpu), ciu_wdog.u64); David, I think this ifdef should be replaced by a notifier called from __cpu_die(). Ralf From ralf@linux-mips.org Tue Mar 9 20:17:52 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Tue, 09 Mar 2010 20:17:57 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:44010 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492542Ab0CITRw (ORCPT ); Tue, 9 Mar 2010 20:17:52 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o29JHotC002083; Tue, 9 Mar 2010 20:17:51 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o29JHoU5002080; Tue, 9 Mar 2010 20:17:50 +0100 Date: Tue, 9 Mar 2010 20:17:50 +0100 From: Ralf Baechle To: Yang Shi Cc: ddaney@caviumnetworks.com, linux-mips@linux-mips.org Subject: Re: [PATCH] MIPS: Octeon: Fix wrong variable type in smp.c Message-ID: <20100309191750.GA1960@linux-mips.org> References: <1268115862-25976-1-git-send-email-yang.shi@windriver.com> <20100309190309.GA301@linux-mips.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20100309190309.GA301@linux-mips.org> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26163 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Tue, Mar 09, 2010 at 08:03:09PM +0100, Ralf Baechle wrote: > > @@ -281,7 +281,7 @@ static void octeon_cpu_die(unsigned int cpu) > > > > #ifdef CONFIG_CAVIUM_OCTEON_WATCHDOG > > /* Disable the watchdog */ > > - cvmx_ciu_wdogx_t ciu_wdog; > > + union cvmx_ciu_wdogx ciu_wdog; > > ciu_wdog.u64 = cvmx_read_csr(CVMX_CIU_WDOGX(cpu)); > > ciu_wdog.s.mode = 0; > > cvmx_write_csr(CVMX_CIU_WDOGX(cpu), ciu_wdog.u64); > > David, > > I think this ifdef should be replaced by a notifier called from > __cpu_die(). Since this is unused I'll just remove it for now. Ralf From post@pfrst.de Tue Mar 9 22:03:46 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Tue, 09 Mar 2010 22:03:51 +0100 (CET) Received: from mail1.pearl-online.net ([62.159.194.147]:42774 "EHLO mail1.pearl-online.net" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492282Ab0CIVDq (ORCPT ); Tue, 9 Mar 2010 22:03:46 +0100 Received: from Mobile0.Peter (unknown [95.157.15.72]) by mail1.pearl-online.net (Postfix) with ESMTPA id 1A7B82012A; Tue, 9 Mar 2010 22:04:44 +0100 (CET) Received: from Opal.Peter (Opal.Peter [192.168.1.1]) by Mobile0.Peter (8.12.6/8.12.6/Sendmail/Linux 2.2.13) with ESMTP id o29MLZoJ001217; Tue, 9 Mar 2010 22:21:35 GMT Received: from Opal.Peter (localhost [127.0.0.1]) by Opal.Peter (8.12.11.Beta0/8.12.11.Beta0/Sendmail/Linux 2.4.24-1-386) with ESMTP id o29L3S5s001017; Tue, 9 Mar 2010 22:03:28 +0100 Received: from localhost (pf@localhost) by Opal.Peter (8.12.11.Beta0/8.12.11.Beta0/Debian-1) with ESMTP id o29L3SqJ001013; Tue, 9 Mar 2010 22:03:28 +0100 X-Authentication-Warning: Opal.Peter: pf owned process doing -bs Date: Tue, 9 Mar 2010 22:03:27 +0100 (CET) From: peter fuerst X-Sender: pf@Opal.Peter To: Florian Fainelli cc: linux-mips@linux-mips.org, ralf@linux-mips.org Subject: Re: [PATCH] MIPS: make CAC_ADDR and UNCAC_ADDR account for PHYS_OFFSET In-Reply-To: <201003091546.01281.florian@openwrt.org> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26164 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: post@pfrst.de Precedence: bulk X-list: linux-mips Hi Florian, thats funny! On Tue, 9 Mar 2010, Florian Fainelli wrote: > Date: Tue, 9 Mar 2010 15:46:01 +0100 > From: Florian Fainelli > To: linux-mips@linux-mips.org > Cc: ralf@linux-mips.org > Subject: [PATCH] MIPS: make CAC_ADDR and UNCAC_ADDR account for > PHYS_OFFSET > > On AR7, ... > > Signed-off-by: Regards, Florian Fainelli > --- > diff --git a/arch/mips/include/asm/page.h b/arch/mips/include/asm/page.h > index ac32572..7b11df5 100644 > --- a/arch/mips/include/asm/page.h > +++ b/arch/mips/include/asm/page.h > @@ -188,8 +188,10 @@ typedef struct { unsigned long pgprot; } pgprot_t; > #define VM_DATA_DEFAULT_FLAGS (VM_READ | VM_WRITE | VM_EXEC | \ > VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC) > > -#define UNCAC_ADDR(addr) ((addr) - PAGE_OFFSET + UNCAC_BASE) > -#define CAC_ADDR(addr) ((addr) - UNCAC_BASE + PAGE_OFFSET) > +#define UNCAC_ADDR(addr) ((addr) - PAGE_OFFSET + UNCAC_BASE + \ > + PHYS_OFFSET) > +#define CAC_ADDR(addr) ((addr) - UNCAC_BASE + PAGE_OFFSET + \ > + PHYS_OFFSET) > > #include > #include > I assume, you don't want "+" PHYS_OFFSET in both defines. Two years and a month ago almost the same patch (which used to work on the machine that needed it :) was submitted: --- a/linux-2.6.24/include/asm-mips/page.h Fri Jan 25 12:23:51 2008 +++ b/linux-2.6.24/include/asm-mips/page.h Wed Feb 6 23:26:31 2008 @@ -184,8 +184,8 @@ #define VM_DATA_DEFAULT_FLAGS (VM_READ | VM_WRITE | VM_EXEC | \ VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC) -#define UNCAC_ADDR(addr) ((addr) - PAGE_OFFSET + UNCAC_BASE) -#define CAC_ADDR(addr) ((addr) - UNCAC_BASE + PAGE_OFFSET) +#define UNCAC_ADDR(addr) ((addr) - PAGE_OFFSET + PHYS_OFFSET + UNCAC_BASE) +#define CAC_ADDR(addr) ((addr) - UNCAC_BASE + PAGE_OFFSET - PHYS_OFFSET) #include #include But correct versions of these macros seem to be essential for very "exotic" systems only ;-) kind regards peter From f.fainelli@gmail.com Tue Mar 9 22:24:01 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Tue, 09 Mar 2010 22:24:04 +0100 (CET) Received: from mail-bw0-f215.google.com ([209.85.218.215]:34519 "EHLO mail-bw0-f215.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492547Ab0CIVYB (ORCPT ); Tue, 9 Mar 2010 22:24:01 +0100 Received: by bwz7 with SMTP id 7so5508152bwz.24 for ; Tue, 09 Mar 2010 13:23:55 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:sender:from:reply-to:to :subject:date:user-agent:cc:references:in-reply-to:mime-version :content-type:content-transfer-encoding:message-id; bh=JhQcs8UoidxjVy1BQocjX8WNxK2VH69w0DYMGYBf/FI=; b=lzClh7O5ZTCTIUX4BLDTixSaVXwbeBE/18nT9SSrfywwLq5iYUP5wYOi2veSVT5LXr P5Ilu0SL0l2j8pDQ/u5QLCXbOPUbEAFfnJ0JjLUJJSPBrZlm4hL62eC1C54wSdnky9t6 pip6OwLth44Kq1kLaX71cwgTrbRXB93w8D2nU= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=sender:from:reply-to:to:subject:date:user-agent:cc:references :in-reply-to:mime-version:content-type:content-transfer-encoding :message-id; b=KBBdCWf+T8cWUjujzX0WKzZeAEI+N14Copq9SScYjDGjBfuDlCatwb9stU1imHmkoc 4euPkX7vTErHFUS/5brVrL4E1TOAiEzNTMpomwIV4kqVMn2rBBWGO1nnVKIPuP+8HvRF ehnaqbsZWMqtve6csdziVpxca+p0n04Fzl94A= Received: by 10.204.26.135 with SMTP id e7mr404223bkc.202.1268169834331; Tue, 09 Mar 2010 13:23:54 -0800 (PST) Received: from lenovo.localnet (153.44.69-86.rev.gaoland.net [86.69.44.153]) by mx.google.com with ESMTPS id a11sm23972730bkc.21.2010.03.09.13.23.52 (version=TLSv1/SSLv3 cipher=RC4-MD5); Tue, 09 Mar 2010 13:23:53 -0800 (PST) From: Florian Fainelli Reply-To: Florian Fainelli To: peter fuerst Subject: Re: [PATCH] MIPS: make CAC_ADDR and UNCAC_ADDR account for PHYS_OFFSET Date: Tue, 9 Mar 2010 22:23:46 +0100 User-Agent: KMail/1.12.4 (Linux/2.6.32-trunk-686; KDE/4.3.4; i686; ; ) Cc: linux-mips@linux-mips.org, ralf@linux-mips.org References: In-Reply-To: MIME-Version: 1.0 Content-Type: multipart/signed; boundary="nextPart1591353.nO66AmWWmk"; protocol="application/pgp-signature"; micalg=pgp-sha1 Content-Transfer-Encoding: 7bit Message-Id: <201003092223.50837.florian@openwrt.org> Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26165 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: florian@openwrt.org Precedence: bulk X-list: linux-mips --nextPart1591353.nO66AmWWmk Content-Type: Text/Plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Le mardi 9 mars 2010 22:03:27, peter fuerst a =E9crit : > Hi Florian, thats funny! >=20 > On Tue, 9 Mar 2010, Florian Fainelli wrote: > > Date: Tue, 9 Mar 2010 15:46:01 +0100 > > From: Florian Fainelli > > To: linux-mips@linux-mips.org > > Cc: ralf@linux-mips.org > > Subject: [PATCH] MIPS: make CAC_ADDR and UNCAC_ADDR account for > > PHYS_OFFSET > > > > On AR7, ... > > > > Signed-off-by: Regards, Florian Fainelli > > --- > > diff --git a/arch/mips/include/asm/page.h > > b/arch/mips/include/asm/page.h index ac32572..7b11df5 100644 > > --- a/arch/mips/include/asm/page.h > > +++ b/arch/mips/include/asm/page.h > > @@ -188,8 +188,10 @@ typedef struct { unsigned long pgprot; } pgprot_= t; > > #define VM_DATA_DEFAULT_FLAGS (VM_READ | VM_WRITE | VM_EXEC | \ > > VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC) > > > > -#define UNCAC_ADDR(addr) ((addr) - PAGE_OFFSET + UNCAC_BASE) > > -#define CAC_ADDR(addr) ((addr) - UNCAC_BASE + PAGE_OFFSET) > > +#define UNCAC_ADDR(addr) ((addr) - PAGE_OFFSET + UNCAC_BASE + \ > > + PHYS_OFFSET) > > +#define CAC_ADDR(addr) ((addr) - UNCAC_BASE + PAGE_OFFSET + \ > > + PHYS_OFFSET) > > > > #include > > #include >=20 > I assume, you don't want "+" PHYS_OFFSET in both defines. >=20 > Two years and a month ago almost the same patch (which used to work on > the machine that needed it :) was submitted: Oh I actually did even search for that one. >=20 > --- a/linux-2.6.24/include/asm-mips/page.h Fri Jan 25 12:23:51 2008 > +++ b/linux-2.6.24/include/asm-mips/page.h Wed Feb 6 23:26:31 2008 > @@ -184,8 +184,8 @@ > #define VM_DATA_DEFAULT_FLAGS (VM_READ | VM_WRITE | VM_EXEC | \ > VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC) >=20 > -#define UNCAC_ADDR(addr) ((addr) - PAGE_OFFSET + UNCAC_BASE) > -#define CAC_ADDR(addr) ((addr) - UNCAC_BASE + PAGE_OFFSET) > +#define UNCAC_ADDR(addr) ((addr) - PAGE_OFFSET + PHYS_OFFSET + > UNCAC_BASE) +#define CAC_ADDR(addr) ((addr) - UNCAC_BASE + PAGE_OFFSET - > PHYS_OFFSET) >=20 > #include > #include >=20 > But correct versions of these macros seem to be essential for very > "exotic" systems only ;-) You are right. This is actually needed when people do weird designs, and th= at=20 happens ;) Will resubmit with the proper fixing. =2D- =20 Best regards, Florian Fainelli Email: florian@openwrt.org Web: http://openwrt.org IRC: [florian] on irc.freenode.net =2D------------------------------ --nextPart1591353.nO66AmWWmk Content-Type: application/pgp-signature; name=signature.asc Content-Description: This is a digitally signed message part. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.10 (GNU/Linux) iEYEABECAAYFAkuWvGIACgkQlyvkmBGtjyaHiwCfXuKVHV63x0jLKD9NrAmYLFoE dS0AoJNMN3fet6y4+pZ3gF8U2u6uxIyH =sHSC -----END PGP SIGNATURE----- --nextPart1591353.nO66AmWWmk-- From ralf@linux-mips.org Wed Mar 10 00:29:38 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 10 Mar 2010 00:29:41 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:38510 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492561Ab0CIX3i (ORCPT ); Wed, 10 Mar 2010 00:29:38 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o29NTZKg020037; Wed, 10 Mar 2010 00:29:35 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o29NTYWS020035; Wed, 10 Mar 2010 00:29:34 +0100 Date: Wed, 10 Mar 2010 00:29:33 +0100 From: Ralf Baechle To: David Daney Cc: linux-mips@linux-mips.org Subject: Re: [PATCH] MIPS: Octeon: Remove vestiges of CONFIG_CAVIUM_RESERVE32_USE_WIRED_TLB Message-ID: <20100309232933.GA2848@linux-mips.org> References: <1267643227-27710-1-git-send-email-ddaney@caviumnetworks.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1267643227-27710-1-git-send-email-ddaney@caviumnetworks.com> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26166 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Wed, Mar 03, 2010 at 11:07:07AM -0800, David Daney wrote: > The config option CAVIUM_RESERVE32_USE_WIRED_TLB is not supported. > Remove the dead code controlled by it. Thanks, applied. Ralf From ralf@linux-mips.org Wed Mar 10 00:50:06 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 10 Mar 2010 00:50:09 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:47349 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492555Ab0CIXuG (ORCPT ); Wed, 10 Mar 2010 00:50:06 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o29No4tK021665; Wed, 10 Mar 2010 00:50:05 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o29No3ua021663; Wed, 10 Mar 2010 00:50:04 +0100 Date: Wed, 10 Mar 2010 00:50:02 +0100 From: Ralf Baechle To: David Daney Cc: linux-mips@linux-mips.org Subject: Re: [PATCH] MIPS: Octeon: Remove #if 0 code. Message-ID: <20100309235002.GA21051@linux-mips.org> References: <1267643263-27803-1-git-send-email-ddaney@caviumnetworks.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1267643263-27803-1-git-send-email-ddaney@caviumnetworks.com> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26167 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Wed, Mar 03, 2010 at 11:07:43AM -0800, David Daney wrote: > Signed-off-by: David Daney Applied. Thanks, Ralf From yuasa.linux@gmail.com Wed Mar 10 07:58:09 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 10 Mar 2010 07:58:14 +0100 (CET) Received: from mail-yw0-f186.google.com ([209.85.211.186]:49713 "EHLO mail-yw0-f186.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491803Ab0CJG6J (ORCPT ); Wed, 10 Mar 2010 07:58:09 +0100 Received: by ywh16 with SMTP id 16so4554754ywh.0 for ; Tue, 09 Mar 2010 22:58:01 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:sender:date:from:to:cc :subject:message-id:x-mailer:mime-version:content-type :content-transfer-encoding; bh=pJYmVtfvtWBiMBqPM7ZKJBJ+05zUgjG4Mm6xy8XMIcs=; b=VuCRCHrOGiZ8mLAmzkGjdLS1dxHg53gEDUzqH8RP85xqdanlsjvPCIUsMSenCmdU/C X/7PVhV5Y6B512TtGBDdKYZ39UO/zIcowVQekRUy7iM1cyxMjA0Ubkk8Hk4LMxcdls77 0x6V6oKh2mcugGj/1tFjbMPKROTLvWFMJ3lSw= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=sender:date:from:to:cc:subject:message-id:x-mailer:mime-version :content-type:content-transfer-encoding; b=ZeFAWJ7Ns0TLYqrjlOyI1GuzV/sjX/iuZzufhSryh+jOk6L9qZhKZxxLpEwyhPIpOh /u3ubrfZ/JgJKJEZC5ynb9Wnuf978l20ZWOoXDRPO6BBhEd0FzuwRx5VZfgXjvoYhPF0 tvoyd4XrMnf6UCEttXAmmIZy3VeSKu4GkAMrs= Received: by 10.101.21.15 with SMTP id y15mr770039ani.190.1268204281712; Tue, 09 Mar 2010 22:58:01 -0800 (PST) Received: from stratos.skybright.jp (sannin29006.nirai.ne.jp [203.160.29.6]) by mx.google.com with ESMTPS id 15sm4974598gxk.2.2010.03.09.22.57.58 (version=TLSv1/SSLv3 cipher=RC4-MD5); Tue, 09 Mar 2010 22:58:01 -0800 (PST) Date: Wed, 10 Mar 2010 15:57:56 +0900 From: Yoichi Yuasa To: Dominik Brodowski Cc: yuasa@linux-mips.org, linux-pcmcia@lists.infradead.org, Ralf Baechle , linux-mips Subject: [PATCH] pcmcia/vrc4171: use local spinlock for device local lock. Message-Id: <20100310155756.496e2bbf.yuasa@linux-mips.org> X-Mailer: Sylpheed 3.0.0 (GTK+ 2.16.6; x86_64-unknown-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26168 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: yuasa@linux-mips.org Precedence: bulk X-list: linux-mips struct pcmcia_socket lock had been used before. Signed-off-by: Yoichi Yuasa --- drivers/pcmcia/vrc4171_card.c | 5 +++-- 1 files changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/pcmcia/vrc4171_card.c b/drivers/pcmcia/vrc4171_card.c index c9fcbdc..aaccdb9 100644 --- a/drivers/pcmcia/vrc4171_card.c +++ b/drivers/pcmcia/vrc4171_card.c @@ -105,6 +105,7 @@ typedef struct vrc4171_socket { char name[24]; int csc_irq; int io_irq; + spinlock_t lock; } vrc4171_socket_t; static vrc4171_socket_t vrc4171_sockets[CARD_MAX_SLOTS]; @@ -327,7 +328,7 @@ static int pccard_set_socket(struct pcmcia_socket *sock, socket_state_t *state) slot = sock->sock; socket = &vrc4171_sockets[slot]; - spin_lock_irq(&sock->lock); + spin_lock_irq(&socket->lock); voltage = set_Vcc_value(state->Vcc); exca_write_byte(slot, CARD_VOLTAGE_SELECT, voltage); @@ -370,7 +371,7 @@ static int pccard_set_socket(struct pcmcia_socket *sock, socket_state_t *state) cscint |= I365_CSC_DETECT; exca_write_byte(slot, I365_CSCINT, cscint); - spin_unlock_irq(&sock->lock); + spin_unlock_irq(&socket->lock); return 0; } -- 1.7.0.2 From f.fainelli@gmail.com Wed Mar 10 09:52:27 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 10 Mar 2010 09:52:30 +0100 (CET) Received: from mail-fx0-f217.google.com ([209.85.220.217]:36115 "EHLO mail-fx0-f217.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491820Ab0CJIw1 (ORCPT ); Wed, 10 Mar 2010 09:52:27 +0100 Received: by fxm9 with SMTP id 9so3602007fxm.24 for ; Wed, 10 Mar 2010 00:52:21 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:sender:from:date:subject :mime-version:x-uid:x-length:organization:to:cc:content-type :content-transfer-encoding:message-id; bh=PV7uJ48vU8yCv7q18E5A/ROfennsztzN3/r6Pejhmoc=; b=wS/8yxZ+yI0j/IKHpz9cMUxtUOV+MCKEPzIOEkNZEDbGI6tT5lewaMUUMnN84SJ4Uo 7CKSUuLmuTAEO0w7+UQRlb2+LUGnPiiIpORvyvIbOpImuwv/Bl22lh/sF8bKv2JmGkKe mhROJUq+DHn5d0w87dUzKHBXrydCJe/aMl01Y= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=sender:from:date:subject:mime-version:x-uid:x-length:organization :to:cc:content-type:content-transfer-encoding:message-id; b=Bh813b7BXNgb9O7koiRYV2nfXM/TkXk7IYYDlv5wNgaExne+STA2A17INKwDZEU5AC JCX5EUfnbf0+4YOPu/XQ9X7SdYtH0Ai+kSrKlc/GvT+TA1n4++FKqWrZ49Knl7JyKnz5 6fXEYrx8bix6PPEwNiDNY3PAy5SVaMT3qE7OQ= Received: by 10.223.143.82 with SMTP id t18mr1121388fau.52.1268211140636; Wed, 10 Mar 2010 00:52:20 -0800 (PST) Received: from flexo.localnet (bobafett.staff.proxad.net [213.228.1.121]) by mx.google.com with ESMTPS id 26sm12863247fks.22.2010.03.10.00.52.19 (version=SSLv3 cipher=RC4-MD5); Wed, 10 Mar 2010 00:52:19 -0800 (PST) From: Florian Fainelli Date: Wed, 10 Mar 2010 09:51:09 +0100 Subject: [PATCH v2] MIPS: make CAC_ADDR and UNCAC_ADDR account for PHYS_OFFSET MIME-Version: 1.0 X-UID: 24995 X-Length: 3378 Organization: OpenWrt To: linux-mips@linux-mips.org Cc: ralf@linux-mips.org, peter fuerst Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit Message-Id: <201003100951.10028.florian@openwrt.org> Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26169 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: florian@openwrt.org Precedence: bulk X-list: linux-mips On AR7, we already redefine PHYS_OFFSET to match the system specifities, it is however not sufficient when unsing dma_{map,unmap}_single, specifically in the Ethernet driver, we must also adjust CAC_ADDR and UNCAC_ADDR for DMA to work correctly. This patch fixes the following issue, seen in cpmac_open: ops[#1]: Cpu 0 $ 0 : 00000000 10008400 a0f5b120 00000000 $ 4 : 94c59000 94270f64 00000020 00000010 $ 8 : 00000010 94103ce0 0000000a 94c03400 $12 : ffffffff 94c03408 94c03410 00000001 $16 : a0f5ba20 00000041 94c592c0 94c59200 $20 : 94c59000 000005ee 00002000 9438c8f0 $24 : 00000010 00000000 $28 : 94fac000 94fadd58 94390000 942724a8 Hi : 00000000 Lo : 00000001 epc : 94272518 cpmac_open+0x208/0x3f8 Not tainted ra : 942724a8 cpmac_open+0x198/0x3f8 Status: 10008403 KERNEL EXL IE Cause : 3080000c BadVA : 00000000 PrId : 00018448 (MIPS 4KEc) Modules linked in: Process ifconfig (pid: 278, threadinfo=94fac000, task=94e79590, tls=00000000) Stack : 7f8da120 2ab05cb0 94c59000 943356f0 00000000 943d0000 94c59000 943356f0 94c59030 943d0000 943c27c0 94fade10 00000000 94fade20 94c59000 9428e5a4 00000000 94c59000 00000041 94289768 94c59000 00000041 00001002 00001043 00000000 9428d810 00000000 94fade10 7f8da4e8 9428e6b8 00000000 7f8da4a8 7f8da4e8 00008914 00000000 942f7f2c 00000000 00000008 00408000 00008913 ... Call Trace: [<94272518>] cpmac_open+0x208/0x3f8 [<9428e5a4>] dev_open+0x164/0x264 [<9428d810>] dev_change_flags+0xd0/0x1bc [<942f7f2c>] devinet_ioctl+0x2d8/0x908 [<942771f8>] sock_ioctl+0x29c/0x2fc [<941a0fb4>] vfs_ioctl+0x2c/0x7c [<941a16ec>] do_vfs_ioctl+0x5dc/0x630 [<941a1790>] sys_ioctl+0x50/0x88 [<94101e10>] stack_done+0x20/0x3c Signed-off-by: peter fuerst Signed-off-by: Regards, Florian Fainelli --- diff --git a/arch/mips/include/asm/page.h b/arch/mips/include/asm/page.h index ac32572..a16beaf 100644 --- a/arch/mips/include/asm/page.h +++ b/arch/mips/include/asm/page.h @@ -188,8 +188,10 @@ typedef struct { unsigned long pgprot; } pgprot_t; #define VM_DATA_DEFAULT_FLAGS (VM_READ | VM_WRITE | VM_EXEC | \ VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC) -#define UNCAC_ADDR(addr) ((addr) - PAGE_OFFSET + UNCAC_BASE) -#define CAC_ADDR(addr) ((addr) - UNCAC_BASE + PAGE_OFFSET) +#define UNCAC_ADDR(addr) ((addr) - PAGE_OFFSET + UNCAC_BASE + \ + PHYS_OFFSET) +#define CAC_ADDR(addr) ((addr) - UNCAC_BASE + PAGE_OFFSET - \ + PHYS_OFFSET) #include #include From sshtylyov@mvista.com Wed Mar 10 11:45:12 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 10 Mar 2010 11:45:18 +0100 (CET) Received: from ey-out-1920.google.com ([74.125.78.146]:18729 "EHLO ey-out-1920.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492161Ab0CJKpM (ORCPT ); Wed, 10 Mar 2010 11:45:12 +0100 Received: by ey-out-1920.google.com with SMTP id 13so826430eye.52 for ; Wed, 10 Mar 2010 02:45:11 -0800 (PST) Received: by 10.213.104.89 with SMTP id n25mr4887336ebo.78.1268217908122; Wed, 10 Mar 2010 02:45:08 -0800 (PST) Received: from [192.168.2.2] (ppp85-140-114-94.pppoe.mtu-net.ru [85.140.114.94]) by mx.google.com with ESMTPS id 14sm3753633ewy.10.2010.03.10.02.45.06 (version=TLSv1/SSLv3 cipher=RC4-MD5); Wed, 10 Mar 2010 02:45:06 -0800 (PST) Message-ID: <4B977815.4000703@ru.mvista.com> Date: Wed, 10 Mar 2010 13:44:37 +0300 From: Sergei Shtylyov User-Agent: Thunderbird 2.0.0.23 (Windows/20090812) MIME-Version: 1.0 To: Wu Zhangjin CC: Ralf Baechle , Greg KH , linux-mips@linux-mips.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH 1/3] Loongson-2F: Flush the branch target history such as BTB and RAS References: In-Reply-To: Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 8bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26170 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: sshtylyov@mvista.com Precedence: bulk X-list: linux-mips Hello. Wu Zhangjin wrote: > From: Wu Zhangjin > > As the Chapter 15: "Errata: Issue of Out-of-order in loongson"[1] shows, to > workaround the Issue of Loongson-2F，We need to do: > > "When switching from user model to kernel model, you should flush the branch > target history such as BTB and RAS." > > This patch did clear BTB(branch target buffer), forbid RAS(row address strobe) > via Loongson-2F's 64bit diagnostic register. > > [1] Chinese Version: http://www.loongson.cn/uploadfile/file/200808211 > [2] English Version of Chapter 15: > http://groups.google.com.hk/group/loongson-dev/msg/e0d2e220958f10a6?dmode=source > > Signed-off-by: Wu Zhangjin > --- > arch/mips/include/asm/stackframe.h | 19 +++++++++++++++++++ > 1 files changed, 19 insertions(+), 0 deletions(-) > > diff --git a/arch/mips/include/asm/stackframe.h b/arch/mips/include/asm/stackframe.h > index 3b6da33..b84cfda 100644 > --- a/arch/mips/include/asm/stackframe.h > +++ b/arch/mips/include/asm/stackframe.h > @@ -121,6 +121,25 @@ > .endm > #else > .macro get_saved_sp /* Uniprocessor variation */ > +#ifdef CONFIG_CPU_LOONGSON2F > + /* > + * Clear BTB(branch target buffer), forbid RAS(row address > + * strobe) No spaces before the left paren... > to workaround the Out-of-oder Issue in Loongson2F > + * via it's diagnostic register. > Only "its". > + */ > + move k0, ra > Operands misalined... WBR, Sergei From wuzhangjin@gmail.com Wed Mar 10 16:30:43 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 10 Mar 2010 16:30:47 +0100 (CET) Received: from mail-pz0-f185.google.com ([209.85.222.185]:33055 "EHLO mail-pz0-f185.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492334Ab0CJPan (ORCPT ); Wed, 10 Mar 2010 16:30:43 +0100 Received: by pzk15 with SMTP id 15so4676588pzk.21 for ; Wed, 10 Mar 2010 07:30:35 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:subject:from:reply-to:to:cc :in-reply-to:references:content-type:organization:date:message-id :mime-version:x-mailer:content-transfer-encoding; bh=SQ+lGJkXfLsBokfGNBa8HG9EaBvcmeC0REL3zP7AGOo=; b=ieqvcBhObZwTt8k5tQmeQ5H2s2/U8Hm50YZOGVB1jf8t28Vis3wra9c4CBYtpnXR1G Pq0OkeEOpiPSAYrA+ILbaNZw8aJI9lvzDtJOT6z7i3nh0DtKsE7zPHQUSUW289GCWu7H hOBR9ZN6mEN1U3RGg3hdXjc+v1jQs7ZcqXtdc= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=subject:from:reply-to:to:cc:in-reply-to:references:content-type :organization:date:message-id:mime-version:x-mailer :content-transfer-encoding; b=Em+J66uA+sEgOKw0BIOxRK+riWl+r0CR3tm1ooXjzinzJbFQqwLEjs/U4zTEuXLeXy DcJUcYjTP9xLu1Hwz/3wiySfvsTovVE68PZrWY0vDl6hMG2UYMB/daa1AYlqYqWPAnfd qGwUPFK8Qzol+wQQjSf3AGce9YLXeAv/5RP0E= Received: by 10.142.250.19 with SMTP id x19mr522202wfh.23.1268235035282; Wed, 10 Mar 2010 07:30:35 -0800 (PST) Received: from [202.201.12.142] ([202.201.12.142]) by mx.google.com with ESMTPS id 21sm7513057pzk.4.2010.03.10.07.30.31 (version=TLSv1/SSLv3 cipher=RC4-MD5); Wed, 10 Mar 2010 07:30:34 -0800 (PST) Subject: Re: [PATCH 1/3] Loongson-2F: Flush the branch target history such as BTB and RAS From: Wu Zhangjin Reply-To: wuzhangjin@gmail.com To: Sergei Shtylyov Cc: Ralf Baechle , Greg KH , linux-mips@linux-mips.org, linux-kernel@vger.kernel.org In-Reply-To: <4B977815.4000703@ru.mvista.com> References: <4B977815.4000703@ru.mvista.com> Content-Type: text/plain; charset="UTF-8" Organization: DSLab, Lanzhou University, China Date: Wed, 10 Mar 2010 23:24:04 +0800 Message-ID: <1268234645.19976.1.camel@falcon> Mime-Version: 1.0 X-Mailer: Evolution 2.28.2 Content-Transfer-Encoding: 7bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26171 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: wuzhangjin@gmail.com Precedence: bulk X-list: linux-mips Hi, Thanks very much for your careful review, will resend a new one later. Regards, Wu Zhangjin On Wed, 2010-03-10 at 13:44 +0300, Sergei Shtylyov wrote: [...] > > diff --git a/arch/mips/include/asm/stackframe.h b/arch/mips/include/asm/stackframe.h > > index 3b6da33..b84cfda 100644 > > --- a/arch/mips/include/asm/stackframe.h > > +++ b/arch/mips/include/asm/stackframe.h > > @@ -121,6 +121,25 @@ > > .endm > > #else > > .macro get_saved_sp /* Uniprocessor variation */ > > +#ifdef CONFIG_CPU_LOONGSON2F > > + /* > > + * Clear BTB(branch target buffer), forbid RAS(row address > > + * strobe) > > No spaces before the left paren... > > > to workaround the Out-of-oder Issue in Loongson2F > > + * via it's diagnostic register. > > > > Only "its". > > > + */ > > + move k0, ra > > > > Operands misalined... > > WBR, Sergei > > From ralf@linux-mips.org Wed Mar 10 16:33:23 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 10 Mar 2010 16:33:27 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:54650 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492497Ab0CJPdW (ORCPT ); Wed, 10 Mar 2010 16:33:22 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o2AFXIpW013636; Wed, 10 Mar 2010 16:33:19 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o2AFXGkS013633; Wed, 10 Mar 2010 16:33:16 +0100 Date: Wed, 10 Mar 2010 16:33:15 +0100 From: Ralf Baechle To: David Daney Cc: Yang Shi , linux-mips@linux-mips.org Subject: Re: [PATCH] MIPS: Protect current_cpu_data with preempt disable in delay() Message-ID: <20100310153315.GA12476@linux-mips.org> References: <1267695573-27360-1-git-send-email-yang.shi@windriver.com> <4B8FFAB3.1090409@caviumnetworks.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <4B8FFAB3.1090409@caviumnetworks.com> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26172 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Thu, Mar 04, 2010 at 10:23:47AM -0800, David Daney wrote: > On 03/04/2010 01:39 AM, Yang Shi wrote: > >During machine restart with reboot command, get the following > >bug info: > > > >BUG: using smp_processor_id() in preemptible [00000000] code: reboot/1989 > >caller is __udelay+0x14/0x70 > >Call Trace: > >[] dump_stack+0x8/0x34 > >[] debug_smp_processor_id+0xf4/0x110 > >[] __udelay+0x14/0x70 > >[] md_notify_reboot+0x12c/0x148 > >[] notifier_call_chain+0x64/0xc8 > >[] __blocking_notifier_call_chain+0x64/0xc0 > >[] kernel_restart_prepare+0x1c/0x38 > >[] kernel_restart+0x14/0x50 > >[] SyS_reboot+0x10c/0x1f0 > >[] handle_sysn32+0x44/0x84 > > > >The root cause is that current_cpu_data is accessed in preemptible > >context, so protect it with preempt_disable/preempt_enable pair > >in delay(). > > > >Signed-off-by: Yang Shi > >--- > > arch/mips/lib/delay.c | 6 +++++- > > 1 files changed, 5 insertions(+), 1 deletions(-) > > > >diff --git a/arch/mips/lib/delay.c b/arch/mips/lib/delay.c > >index 6b3b1de..dc38064 100644 > >--- a/arch/mips/lib/delay.c > >+++ b/arch/mips/lib/delay.c > >@@ -41,7 +41,11 @@ EXPORT_SYMBOL(__delay); > > > > void __udelay(unsigned long us) > > { > >- unsigned int lpj = current_cpu_data.udelay_val; > >+ unsigned int lpj; > >+ > >+ preempt_disable(); > >+ lpj = current_cpu_data.udelay_val; > >+ preempt_enable(); > > > > __delay((us * 0x000010c7ull * HZ * lpj)>> 32); > > } > > This doesn't seem like the best approach. > > Perhaps we should either use raw_current_cpu_data and no > preempt_disable(), or if we are concerned about migrating to a CPU > with a different lpj value, move the preempt_enable after the call > to __delay(). Udelay() is supposed to guarantee a minimum delay and when being migrated to another CPU with higher bogomips this guarantee might be violated. So it'd even have to be something like: void __udelay(unsigned long us) { unsigned int lpj = current_cpu_data.udelay_val; unsigned int lpj; preempt_disable(); lpj = current_cpu_data.udelay_val; __delay((us * 0x000010c7ull * HZ * lpj)>> 32); preempt_enable(); } But preempt_disable() itself is not atomic, so using it from bh or irq context could result in a corrupted preemption counter. So the raw_ version will have to do. I doubt it's much of a problem but at some point we will have to revisit the delay by c0_count patch submitted a while ago. The patch wasn't right but the problem it was addressing is real. Ralf From wuzhangjin@gmail.com Wed Mar 10 16:48:13 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 10 Mar 2010 16:48:18 +0100 (CET) Received: from mail-pv0-f177.google.com ([74.125.83.177]:65486 "EHLO mail-pv0-f177.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492106Ab0CJPsN (ORCPT ); Wed, 10 Mar 2010 16:48:13 +0100 Received: by pvd12 with SMTP id 12so607605pvd.36 for ; Wed, 10 Mar 2010 07:48:06 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:from:to:cc:subject:date :message-id:x-mailer; bh=JWLzpnpYPckrebmogDGle3A8wRCbjnPKmL8yFWzzO+Q=; b=T6ddnDT2LCVdrSfo+FSXHkvFTse7/zipYcYHxyxwQi4RdzM1DrTb0fRQcBmJWf3Vok fiEPXxfoSI8u5CWahkvQCx2DtuhfMVdJTtu/jzY2nmOwB6JEB+29na9pnP4GvSC6Z+cf pKaV4hhGNheutzg39cwo1hVHxt69o8XLiBCDU= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=from:to:cc:subject:date:message-id:x-mailer; b=Z67tBXXXXPRmqVP+G25BGfcDMb2mzSeToAag6i1z8M0TvENi9343F6F9B2MaPuJCJm 6p4W9CUjAFoZ743/DTaHScI8vkzxO8tfPuhiuL5ZYrRqNNT9JBltFHv3NT0OW81vBOuF jHhOkrHrFuCtD0y6WSZhmbf0vSx4WPZj5T/Rk= Received: by 10.115.100.18 with SMTP id c18mr912427wam.62.1268236086647; Wed, 10 Mar 2010 07:48:06 -0800 (PST) Received: from localhost.localdomain ([202.201.12.142]) by mx.google.com with ESMTPS id 23sm7483770pzk.10.2010.03.10.07.48.03 (version=TLSv1/SSLv3 cipher=RC4-MD5); Wed, 10 Mar 2010 07:48:05 -0800 (PST) From: Wu Zhangjin To: Ralf Baechle Cc: linux-mips@linux-mips.org, Wu Zhangjin Subject: [PATCH] Loongson: Lemote-2F: Fixup of _rdmsr and _wrmsr Date: Wed, 10 Mar 2010 23:41:34 +0800 Message-Id: <7c2dec50764082fafa83895b740f644fc592afa4.1268235182.git.wuzhangjin@gmail.com> X-Mailer: git-send-email 1.7.0.1 Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26173 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: wuzhangjin@gmail.com Precedence: bulk X-list: linux-mips From: Wu Zhangjin The _rdmsr, _wrmsr operation must be atomic to ensure the accessing to msr address is what we want. Without this patch, in some cases, the reboot operation(fs2f_reboot) defined in arch/mips/loongson/lemote-2f/reset.c may fail for it called _rdmsr, _wrmsr but may be interrupted/preempted by the other related operations and make the _rdmsr get the wrong value or make the _wrmsr write a wrong value to an unexpected target. Signed-off-by: Wu Zhangjin --- arch/mips/pci/ops-loongson2.c | 9 +++++++++ 1 files changed, 9 insertions(+), 0 deletions(-) diff --git a/arch/mips/pci/ops-loongson2.c b/arch/mips/pci/ops-loongson2.c index 2bb4057..1f93dfb 100644 --- a/arch/mips/pci/ops-loongson2.c +++ b/arch/mips/pci/ops-loongson2.c @@ -180,15 +180,20 @@ struct pci_ops loongson_pci_ops = { }; #ifdef CONFIG_CS5536 +DEFINE_SPINLOCK(msr_lock); void _rdmsr(u32 msr, u32 *hi, u32 *lo) { struct pci_bus bus = { .number = PCI_BUS_CS5536 }; u32 devfn = PCI_DEVFN(PCI_IDSEL_CS5536, 0); + unsigned long flags; + + spin_lock_irqsave(&msr_lock, flags); loongson_pcibios_write(&bus, devfn, PCI_MSR_ADDR, 4, msr); loongson_pcibios_read(&bus, devfn, PCI_MSR_DATA_LO, 4, lo); loongson_pcibios_read(&bus, devfn, PCI_MSR_DATA_HI, 4, hi); + spin_unlock_irqrestore(&msr_lock, flags); } EXPORT_SYMBOL(_rdmsr); @@ -198,9 +203,13 @@ void _wrmsr(u32 msr, u32 hi, u32 lo) .number = PCI_BUS_CS5536 }; u32 devfn = PCI_DEVFN(PCI_IDSEL_CS5536, 0); + unsigned long flags; + + spin_lock_irqsave(&msr_lock, flags); loongson_pcibios_write(&bus, devfn, PCI_MSR_ADDR, 4, msr); loongson_pcibios_write(&bus, devfn, PCI_MSR_DATA_LO, 4, lo); loongson_pcibios_write(&bus, devfn, PCI_MSR_DATA_HI, 4, hi); + spin_unlock_irqrestore(&msr_lock, flags); } EXPORT_SYMBOL(_wrmsr); #endif -- 1.7.0.1 From ralf@linux-mips.org Wed Mar 10 17:04:47 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 10 Mar 2010 17:04:51 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:42109 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492497Ab0CJQEr (ORCPT ); Wed, 10 Mar 2010 17:04:47 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o2AG4jCC015563; Wed, 10 Mar 2010 17:04:45 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o2AG4iiY015560; Wed, 10 Mar 2010 17:04:44 +0100 Date: Wed, 10 Mar 2010 17:04:42 +0100 From: Ralf Baechle To: Manuel Lauss Cc: Linux-MIPS , Manuel Lauss Subject: Re: [PATCH] MIPS: Alchemy: sleepcode without compile-time cputype dependencies Message-ID: <20100310160442.GA15118@linux-mips.org> References: <1268076181-29642-1-git-send-email-manuel.lauss@gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1268076181-29642-1-git-send-email-manuel.lauss@gmail.com> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26174 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Mon, Mar 08, 2010 at 08:22:59PM +0100, Manuel Lauss wrote: > Split the low-level sleepcode into per-memory-controller-generation > functions and figure out which one to call at runtime instead of > relying on compile-time-defined cpu types. > > Signed-off-by: Manuel Lauss Thanks, queued for 2.6.35. Ralf From ralf@linux-mips.org Wed Mar 10 17:13:15 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 10 Mar 2010 17:13:19 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:49867 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492609Ab0CJQNP (ORCPT ); Wed, 10 Mar 2010 17:13:15 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o2AGDDJk016130; Wed, 10 Mar 2010 17:13:13 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o2AGDCow016128; Wed, 10 Mar 2010 17:13:12 +0100 Date: Wed, 10 Mar 2010 17:13:12 +0100 From: Ralf Baechle To: Manuel Lauss Cc: Linux-MIPS , Manuel Lauss , Manuel Lauss Subject: Re: [RFC PATCH] MIPS: Alchemy: add sysdev for both irq controllers Message-ID: <20100310161312.GB15118@linux-mips.org> References: <1268076181-29642-1-git-send-email-manuel.lauss@gmail.com> <1268076181-29642-2-git-send-email-manuel.lauss@gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1268076181-29642-2-git-send-email-manuel.lauss@gmail.com> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26175 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Mon, Mar 08, 2010 at 08:23:00PM +0100, Manuel Lauss wrote: > From: Manuel Lauss > > Use a sysdev to implement PM methods for the Au1000 interrupt controllers. > > Signed-off-by: Manuel Lauss > --- > Works on DB1200, not sure though whether this is the correct approach. > Applies cleanly only on top of "sleepcode-without-compile-time-cputype" > patch. At a quick glance this looks good but since you marked your patch as RFC I'll do the same in patchwork (http://patchwork.linux-mips.org/patch/1039/) until you ask me to queue it or resubmit. Thanks, Ralf From ralf@linux-mips.org Wed Mar 10 17:48:29 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 10 Mar 2010 17:48:34 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:60683 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492624Ab0CJQs3 (ORCPT ); Wed, 10 Mar 2010 17:48:29 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o2AGmQ71018013; Wed, 10 Mar 2010 17:48:27 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o2AGmOF9018009; Wed, 10 Mar 2010 17:48:24 +0100 Date: Wed, 10 Mar 2010 17:48:24 +0100 From: Ralf Baechle To: Manuel Lauss Cc: Sergei Shtylyov , Linux-MIPS , Manuel Lauss Subject: Re: [PATCH] MIPS: Alchemy: move MMC driver registration to board code. Message-ID: <20100310164824.GC15118@linux-mips.org> References: <1268076181-29642-1-git-send-email-manuel.lauss@gmail.com> <1268076181-29642-3-git-send-email-manuel.lauss@gmail.com> <4B963210.7030906@ru.mvista.com> <4B96364E.5050202@mvista.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26176 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Tue, Mar 09, 2010 at 01:03:51PM +0100, Manuel Lauss wrote: > And on a personal note, that file just bothers me. It's messy, can > cause merge conflicts, Eye cancer. > it references structures defined inside board-specific code. In short, > it just plain annoys > my sense of aesthetics. Indeed - and I don't think Sergej disagrees with that. I agree with him that device registration code should primarily be done in the SOC code - but you'll need to somehow get that code to communicate with the platform code about what really needs to be done then register the remainder of the truely platform-specific platform devices. Something like that. Ralf From rollercow@sucs.org Wed Mar 10 18:50:32 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 10 Mar 2010 18:50:35 +0100 (CET) Received: from silver.sucs.swan.ac.uk ([137.44.10.1]:35880 "EHLO silver.sucs.swan.ac.uk" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492389Ab0CJRuc (ORCPT ); Wed, 10 Mar 2010 18:50:32 +0100 Received: from [137.44.141.35] (helo=[192.168.10.193]) by silver.sucs.swan.ac.uk with esmtpsa (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.69) (envelope-from ) id 1NpQ0p-00076r-T5; Wed, 10 Mar 2010 17:48:04 +0000 Message-ID: <4B97DB49.2010009@sucs.org> Date: Wed, 10 Mar 2010 17:47:53 +0000 From: Chris Jones User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.7) Gecko/20100204 Lightning/1.0b1 Icedove/3.0.1 MIME-Version: 1.0 To: debian-mips@lists.debian.org, linux-mips@linux-mips.org CC: SUCS Staff , Steve Whitehouse Subject: SGI Onyx4 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26177 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: rollercow@sucs.org Precedence: bulk X-list: linux-mips Dear all, The Swansea University Computer Society has been given an SGI Onxy4. We would very much like to get Linux running on it but don't have any local people with both the skills and time. Would anyone be interested in helping to port Linux to it? If so please let me know off list! Many thanks! -- Chris Jones, SUCS Admin http://sucs.org From manuel.lauss@googlemail.com Wed Mar 10 19:34:08 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 10 Mar 2010 19:34:15 +0100 (CET) Received: from mail-fx0-f217.google.com ([209.85.220.217]:44138 "EHLO mail-fx0-f217.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492620Ab0CJSeH convert rfc822-to-8bit (ORCPT ); Wed, 10 Mar 2010 19:34:07 +0100 Received: by fxm9 with SMTP id 9so4148700fxm.24 for ; Wed, 10 Mar 2010 10:34:00 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=googlemail.com; s=gamma; h=domainkey-signature:mime-version:received:in-reply-to:references :date:message-id:subject:from:to:cc:content-type :content-transfer-encoding; bh=1+NTcxS3mtuZBtS5Q7frQkxrLTLE2HzmfAEvFOu4jKI=; b=lK11ahXweQmABnK+aIyhWRxditsNPt8A1LB1louXyYTxZlRvVhqVKyD/0b8Ttv40ER txSt3L/NxRr1afJDRaEmvV0AXWHO/FLoI1AtFnEFne/imvP/OvzTYVHFCHZ5kbn1ObvA Grw5Say6HjD5FPU1/Io47gwbAp5pt3/Ouoc5w= DomainKey-Signature: a=rsa-sha1; c=nofws; d=googlemail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type:content-transfer-encoding; b=S5piBG/+BwGFOEH+GFTtaHLwIQMeHhzmZwYQsRHJO4h0LyMYl67gEYQKDk+IXgljsj UYcMcKEiHdRKm/ZR3DzhRAqvFGISaryRvUFDozMZIFcs22WvDXZmFwkjhMcHVryVbn+n xnGioH74YO7yhMe7lV0zexNQUqj98ghEEW5aQ= MIME-Version: 1.0 Received: by 10.223.4.197 with SMTP id 5mr2167111fas.1.1268246040666; Wed, 10 Mar 2010 10:34:00 -0800 (PST) In-Reply-To: <20100310164824.GC15118@linux-mips.org> References: <1268076181-29642-1-git-send-email-manuel.lauss@gmail.com> <1268076181-29642-3-git-send-email-manuel.lauss@gmail.com> <4B963210.7030906@ru.mvista.com> <4B96364E.5050202@mvista.com> <20100310164824.GC15118@linux-mips.org> Date: Wed, 10 Mar 2010 19:34:00 +0100 Message-ID: Subject: Re: [PATCH] MIPS: Alchemy: move MMC driver registration to board code. From: Manuel Lauss To: Ralf Baechle Cc: Sergei Shtylyov , Linux-MIPS , Manuel Lauss Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8BIT Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26178 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: manuel.lauss@googlemail.com Precedence: bulk X-list: linux-mips On Wed, Mar 10, 2010 at 5:48 PM, Ralf Baechle wrote: > On Tue, Mar 09, 2010 at 01:03:51PM +0100, Manuel Lauss wrote: > >> And on a personal note, that file just bothers me.  It's messy, can >> cause merge conflicts, > > Eye cancer. > >> it references structures defined inside board-specific code. In short, >> it just plain annoys >> my sense of aesthetics. > > Indeed - and I don't think Sergej disagrees with that.  I agree with him > that device registration code should primarily be done in the SOC code - For things like fixed internal interrupts, sure. But for pieces that depend on where the (and which) chip is used, not so much. By this logic there should be tons of ifdefs in the interrupt tables for every in-tree board. > but you'll need to somehow get that code to communicate with the platform > code about what really needs to be done then register the remainder of > the truely platform-specific platform devices.  Something like that. I prefer a simpler solution. Manuel From manuel.lauss@googlemail.com Wed Mar 10 20:03:36 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 10 Mar 2010 20:03:42 +0100 (CET) Received: from mail-fx0-f217.google.com ([209.85.220.217]:58189 "EHLO mail-fx0-f217.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491940Ab0CJTDg (ORCPT ); Wed, 10 Mar 2010 20:03:36 +0100 Received: by fxm9 with SMTP id 9so4178073fxm.24 for ; Wed, 10 Mar 2010 11:03:29 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=googlemail.com; s=gamma; h=domainkey-signature:mime-version:received:in-reply-to:references :date:message-id:subject:from:to:cc:content-type; bh=tf9EaUrKYoMk4jXBEAiNDG2slOj0Dmw6Ip+tXcn3krE=; b=KWipMntl3xauDGSPiIsJvvAaKBlNyxEkjWBoAGcyR76AFFYKHeShSHi6WT22L+fAqc BjzoZFweiilGGzmBMyyIUf0VUfVit1NdlaHaLhui/zfO2g1I7+NwpdZ3VQKMLdgHsJwJ CoVB3tXekurzPEG8oCB2EWAZE+W3Qmo0gFArs= DomainKey-Signature: a=rsa-sha1; c=nofws; d=googlemail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; b=C2v/drgYWR2nvHXzP2hY/1S5wHG2nnZX6iwlogzHHu1Y+rUUQUwKJfY7vw2+i7J7IB eOD+dZhigIkzS1aUu2NjqmguH+q3cMzU/5zVA1XKv8grJnYgAXKPluCg6Q92+Aum+jUL hvZhG2XPgLkF/pFVEttRurkouvM1XiRrlU4W4= MIME-Version: 1.0 Received: by 10.223.76.74 with SMTP id b10mr2146451fak.55.1268247809521; Wed, 10 Mar 2010 11:03:29 -0800 (PST) In-Reply-To: <20100310161312.GB15118@linux-mips.org> References: <1268076181-29642-1-git-send-email-manuel.lauss@gmail.com> <1268076181-29642-2-git-send-email-manuel.lauss@gmail.com> <20100310161312.GB15118@linux-mips.org> Date: Wed, 10 Mar 2010 20:03:29 +0100 Message-ID: Subject: Re: [RFC PATCH] MIPS: Alchemy: add sysdev for both irq controllers From: Manuel Lauss To: Ralf Baechle Cc: Linux-MIPS , Manuel Lauss Content-Type: text/plain; charset=ISO-8859-1 Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26179 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: manuel.lauss@googlemail.com Precedence: bulk X-list: linux-mips On Wed, Mar 10, 2010 at 5:13 PM, Ralf Baechle wrote: > On Mon, Mar 08, 2010 at 08:23:00PM +0100, Manuel Lauss wrote: > >> From: Manuel Lauss >> >> Use a sysdev to implement PM methods for the Au1000 interrupt controllers. >> >> Signed-off-by: Manuel Lauss >> --- >> Works on DB1200, not sure though whether this is the correct approach. >> Applies cleanly only on top of "sleepcode-without-compile-time-cputype" >> patch. > > At a quick glance this looks good but since you marked your patch as RFC > I'll do the same in patchwork (http://patchwork.linux-mips.org/patch/1039/) > until you ask me to queue it or resubmit. I'll resend a refined version. Manuel From linux@dominikbrodowski.net Wed Mar 10 21:44:38 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 10 Mar 2010 21:44:45 +0100 (CET) Received: from isilmar.linta.de ([213.133.102.198]:52828 "EHLO linta.de" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1491076Ab0CJUoi (ORCPT ); Wed, 10 Mar 2010 21:44:38 +0100 Received: (qmail 15308 invoked from network); 10 Mar 2010 20:44:25 -0000 Received: from p54a06b14.dip.t-dialin.net (HELO comet.dominikbrodowski.net) (brodo@84.160.107.20) by isilmar.linta.de with (DHE-RSA-AES256-SHA encrypted) SMTP; 10 Mar 2010 20:44:25 -0000 Received: by comet.dominikbrodowski.net (Postfix, from userid 1000) id 63CC78009B; Wed, 10 Mar 2010 21:35:34 +0100 (CET) Date: Wed, 10 Mar 2010 21:35:34 +0100 From: Dominik Brodowski To: Yoichi Yuasa Cc: linux-pcmcia@lists.infradead.org, Ralf Baechle , linux-mips Subject: Re: [PATCH] pcmcia/vrc4171: use local spinlock for device local lock. Message-ID: <20100310203534.GA5922@comet.dominikbrodowski.net> Mail-Followup-To: Yoichi Yuasa , linux-pcmcia@lists.infradead.org, Ralf Baechle , linux-mips References: <20100310155756.496e2bbf.yuasa@linux-mips.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20100310155756.496e2bbf.yuasa@linux-mips.org> User-Agent: Mutt/1.5.20 (2009-06-14) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26180 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: linux@dominikbrodowski.net Precedence: bulk X-list: linux-mips Hey, On Wed, Mar 10, 2010 at 03:57:56PM +0900, Yoichi Yuasa wrote: > struct pcmcia_socket lock had been used before. Applied, thanks. Dominik From David.Daney@caviumnetworks.com Thu Mar 11 01:18:05 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Thu, 11 Mar 2010 01:18:08 +0100 (CET) Received: from mail3.caviumnetworks.com ([12.108.191.235]:19430 "EHLO mail3.caviumnetworks.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492412Ab0CKASE (ORCPT ); Thu, 11 Mar 2010 01:18:04 +0100 Received: from caexch01.caveonetworks.com (Not Verified[192.168.16.9]) by mail3.caviumnetworks.com with MailMarshal (v6,7,2,8378) id ; Wed, 10 Mar 2010 16:18:10 -0800 Received: from caexch01.caveonetworks.com ([192.168.16.9]) by caexch01.caveonetworks.com with Microsoft SMTPSVC(6.0.3790.3959); Wed, 10 Mar 2010 16:17:40 -0800 Received: from dd1.caveonetworks.com ([12.108.191.236]) by caexch01.caveonetworks.com over TLS secured channel with Microsoft SMTPSVC(6.0.3790.3959); Wed, 10 Mar 2010 16:17:40 -0800 Message-ID: <4B98369F.2060907@caviumnetworks.com> Date: Wed, 10 Mar 2010 16:17:35 -0800 From: David Daney User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.8) Gecko/20100301 Fedora/3.0.3-1.fc12 Thunderbird/3.0.3 MIME-Version: 1.0 To: Wu Zhangjin , Ralf Baechle CC: linux-mips@linux-mips.org Subject: Re: [PATCH] Loongson: Lemote-2F: Fixup of _rdmsr and _wrmsr References: <7c2dec50764082fafa83895b740f644fc592afa4.1268235182.git.wuzhangjin@gmail.com> In-Reply-To: <7c2dec50764082fafa83895b740f644fc592afa4.1268235182.git.wuzhangjin@gmail.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-OriginalArrivalTime: 11 Mar 2010 00:17:40.0679 (UTC) FILETIME=[43258570:01CAC0B0] Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26181 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ddaney@caviumnetworks.com Precedence: bulk X-list: linux-mips On 03/10/2010 07:41 AM, Wu Zhangjin wrote: > From: Wu Zhangjin > > The _rdmsr, _wrmsr operation must be atomic to ensure the accessing to msr > address is what we want. > > Without this patch, in some cases, the reboot operation(fs2f_reboot) defined in > arch/mips/loongson/lemote-2f/reset.c may fail for it called _rdmsr, _wrmsr but > may be interrupted/preempted by the other related operations and make the > _rdmsr get the wrong value or make the _wrmsr write a wrong value to an > unexpected target. > > Signed-off-by: Wu Zhangjin > --- > arch/mips/pci/ops-loongson2.c | 9 +++++++++ > 1 files changed, 9 insertions(+), 0 deletions(-) > > diff --git a/arch/mips/pci/ops-loongson2.c b/arch/mips/pci/ops-loongson2.c > index 2bb4057..1f93dfb 100644 > --- a/arch/mips/pci/ops-loongson2.c > +++ b/arch/mips/pci/ops-loongson2.c > @@ -180,15 +180,20 @@ struct pci_ops loongson_pci_ops = { > }; > > #ifdef CONFIG_CS5536 > +DEFINE_SPINLOCK(msr_lock); Should this be DEFINE_RAW_SPINLOCK instead? David Daney > void _rdmsr(u32 msr, u32 *hi, u32 *lo) > { > struct pci_bus bus = { > .number = PCI_BUS_CS5536 > }; > u32 devfn = PCI_DEVFN(PCI_IDSEL_CS5536, 0); > + unsigned long flags; > + > + spin_lock_irqsave(&msr_lock, flags); > loongson_pcibios_write(&bus, devfn, PCI_MSR_ADDR, 4, msr); > loongson_pcibios_read(&bus, devfn, PCI_MSR_DATA_LO, 4, lo); > loongson_pcibios_read(&bus, devfn, PCI_MSR_DATA_HI, 4, hi); > + spin_unlock_irqrestore(&msr_lock, flags); > } > EXPORT_SYMBOL(_rdmsr); > > @@ -198,9 +203,13 @@ void _wrmsr(u32 msr, u32 hi, u32 lo) > .number = PCI_BUS_CS5536 > }; > u32 devfn = PCI_DEVFN(PCI_IDSEL_CS5536, 0); > + unsigned long flags; > + > + spin_lock_irqsave(&msr_lock, flags); > loongson_pcibios_write(&bus, devfn, PCI_MSR_ADDR, 4, msr); > loongson_pcibios_write(&bus, devfn, PCI_MSR_DATA_LO, 4, lo); > loongson_pcibios_write(&bus, devfn, PCI_MSR_DATA_HI, 4, hi); > + spin_unlock_irqrestore(&msr_lock, flags); > } > EXPORT_SYMBOL(_wrmsr); > #endif From ralf@linux-mips.org Thu Mar 11 01:53:05 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Thu, 11 Mar 2010 01:53:09 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:42483 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492640Ab0CKAxF (ORCPT ); Thu, 11 Mar 2010 01:53:05 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o2B0quqt028896; Thu, 11 Mar 2010 01:52:57 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o2B0qr6B028888; Thu, 11 Mar 2010 01:52:53 +0100 Date: Thu, 11 Mar 2010 01:52:49 +0100 From: Ralf Baechle To: David Daney Cc: Yang Shi , f.fainelli@gmail.com, linux-mips@linux-mips.org Subject: Re: [PATCH 2/3] MIPS: Octeon: Remove redundant declaration of octeon_reserve32_memory Message-ID: <20100311005247.GA18065@linux-mips.org> References: <1267605801-5305-1-git-send-email-yang.shi@windriver.com> <50e36e8549769a26986f99a23772d23fd039c230.1267604875.git.yang.shi@windriver.com> <4B8EA9AD.8070106@caviumnetworks.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <4B8EA9AD.8070106@caviumnetworks.com> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26182 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Wed, Mar 03, 2010 at 10:25:49AM -0800, David Daney wrote: > On 03/03/2010 12:43 AM, Yang Shi wrote: > >In Octeon's setup.c, octeon_reserve32_memory is defined, so remove the > >redundant extern declaration of this variable. > > > >Signed-off-by: Yang Shi > > Acked-by: David Daney > > This looks good to me. Thanks, > David Daney Applied, thanks. Ralf From wuzhangjin@gmail.com Thu Mar 11 02:22:18 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Thu, 11 Mar 2010 02:22:21 +0100 (CET) Received: from mail-ew0-f224.google.com ([209.85.219.224]:42891 "EHLO mail-ew0-f224.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492637Ab0CKBWS (ORCPT ); Thu, 11 Mar 2010 02:22:18 +0100 Received: by ewy24 with SMTP id 24so1823947ewy.27 for ; Wed, 10 Mar 2010 17:22:12 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:subject:from:reply-to:to:cc :in-reply-to:references:content-type:organization:date:message-id :mime-version:x-mailer:content-transfer-encoding; bh=2XSk2EvzL9AEe5yNzUbOSYU+LP28vFmq+b97BY+/mZg=; b=Arj8BoDdUi2/5TCqW8DbxVpgkik1962ZgjoxGKPUfVJLH/+qkDBfAWi/sYQW40X6jG pfvxqufslHdYSSGJM7pL2DRQuMd8TdKZa36wR9hFG9SF+SQ0/sjuDFEVzHnaZDCj+q95 SGfJ7wJhnhIyy98pZ2fxhDUReRg0rlEznPYdM= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=subject:from:reply-to:to:cc:in-reply-to:references:content-type :organization:date:message-id:mime-version:x-mailer :content-transfer-encoding; b=KWQHwAx6kYzcCEjddZ50sYw9vLPl2MgXeVVkq6T/PGtq9DPKK6sj0vvmIlqvQsAVmN Zy4QKqAiDvgFuaM9b01ok6MSbmvEP4PwOyv5IqbSmHK7EBJw9BEBsX2OkZMQ57vKTFzc FjbybnUsnUv/g5/iucn4cljC0oAnPiPgUXiVw= Received: by 10.213.51.195 with SMTP id e3mr4787836ebg.96.1268270531994; Wed, 10 Mar 2010 17:22:11 -0800 (PST) Received: from [202.201.12.142] ([202.201.12.142]) by mx.google.com with ESMTPS id 14sm4333853ewy.10.2010.03.10.17.22.07 (version=TLSv1/SSLv3 cipher=RC4-MD5); Wed, 10 Mar 2010 17:22:10 -0800 (PST) Subject: Re: [PATCH] Loongson: Lemote-2F: Fixup of _rdmsr and _wrmsr From: Wu Zhangjin Reply-To: wuzhangjin@gmail.com To: David Daney Cc: Ralf Baechle , linux-mips@linux-mips.org In-Reply-To: <4B98369F.2060907@caviumnetworks.com> References: <7c2dec50764082fafa83895b740f644fc592afa4.1268235182.git.wuzhangjin@gmail.com> <4B98369F.2060907@caviumnetworks.com> Content-Type: text/plain; charset="UTF-8" Organization: DSLab, Lanzhou University, China Date: Thu, 11 Mar 2010 09:15:41 +0800 Message-ID: <1268270141.12056.7.camel@falcon> Mime-Version: 1.0 X-Mailer: Evolution 2.28.2 Content-Transfer-Encoding: 7bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26183 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: wuzhangjin@gmail.com Precedence: bulk X-list: linux-mips On Wed, 2010-03-10 at 16:17 -0800, David Daney wrote: > On 03/10/2010 07:41 AM, Wu Zhangjin wrote: > > From: Wu Zhangjin > > > > The _rdmsr, _wrmsr operation must be atomic to ensure the accessing to msr > > address is what we want. > > > > Without this patch, in some cases, the reboot operation(fs2f_reboot) defined in > > arch/mips/loongson/lemote-2f/reset.c may fail for it called _rdmsr, _wrmsr but > > may be interrupted/preempted by the other related operations and make the > > _rdmsr get the wrong value or make the _wrmsr write a wrong value to an > > unexpected target. > > > > Signed-off-by: Wu Zhangjin > > --- > > arch/mips/pci/ops-loongson2.c | 9 +++++++++ > > 1 files changed, 9 insertions(+), 0 deletions(-) > > > > diff --git a/arch/mips/pci/ops-loongson2.c b/arch/mips/pci/ops-loongson2.c > > index 2bb4057..1f93dfb 100644 > > --- a/arch/mips/pci/ops-loongson2.c > > +++ b/arch/mips/pci/ops-loongson2.c > > @@ -180,15 +180,20 @@ struct pci_ops loongson_pci_ops = { > > }; > > > > #ifdef CONFIG_CS5536 > > +DEFINE_SPINLOCK(msr_lock); > > > Should this be DEFINE_RAW_SPINLOCK instead? > It should be, thanks! Regards, Wu Zhangjin From Yang.Shi@windriver.com Thu Mar 11 03:53:51 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Thu, 11 Mar 2010 03:53:54 +0100 (CET) Received: from mail.windriver.com ([147.11.1.11]:63161 "EHLO mail.windriver.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492230Ab0CKCxv (ORCPT ); Thu, 11 Mar 2010 03:53:51 +0100 Received: from ALA-MAIL03.corp.ad.wrs.com (ala-mail03 [147.11.57.144]) by mail.windriver.com (8.14.3/8.14.3) with ESMTP id o2B2rhM4021231; Wed, 10 Mar 2010 18:53:43 -0800 (PST) Received: from [128.224.162.222] ([128.224.162.222]) by ALA-MAIL03.corp.ad.wrs.com with Microsoft SMTPSVC(6.0.3790.1830); Wed, 10 Mar 2010 18:53:42 -0800 Message-ID: <4B985B38.1050105@windriver.com> Date: Thu, 11 Mar 2010 10:53:44 +0800 From: Yang Shi User-Agent: Thunderbird 2.0.0.23 (X11/20090817) MIME-Version: 1.0 To: Ralf Baechle CC: ddaney@caviumnetworks.com, linux-mips@linux-mips.org Subject: Re: [PATCH] MIPS: Octeon: Fix wrong variable type in smp.c References: <1268115862-25976-1-git-send-email-yang.shi@windriver.com> <20100309190309.GA301@linux-mips.org> <20100309191750.GA1960@linux-mips.org> In-Reply-To: <20100309191750.GA1960@linux-mips.org> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 8bit X-OriginalArrivalTime: 11 Mar 2010 02:53:42.0925 (UTC) FILETIME=[0F7AE3D0:01CAC0C6] Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26184 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: yang.shi@windriver.com Precedence: bulk X-list: linux-mips Ralf Baechle 写é“: > On Tue, Mar 09, 2010 at 08:03:09PM +0100, Ralf Baechle wrote: > > >>> @@ -281,7 +281,7 @@ static void octeon_cpu_die(unsigned int cpu) >>> >>> #ifdef CONFIG_CAVIUM_OCTEON_WATCHDOG >>> /* Disable the watchdog */ >>> - cvmx_ciu_wdogx_t ciu_wdog; >>> + union cvmx_ciu_wdogx ciu_wdog; >>> ciu_wdog.u64 = cvmx_read_csr(CVMX_CIU_WDOGX(cpu)); >>> ciu_wdog.s.mode = 0; >>> cvmx_write_csr(CVMX_CIU_WDOGX(cpu), ciu_wdog.u64); >>> >> David, >> >> I think this ifdef should be replaced by a notifier called from >> __cpu_die(). >> > > Since this is unused I'll just remove it for now. > This breaks kernel build when HOTPLUG_CPU is enabled. Regards, Yang > Ralf > > From wuzhangjin@gmail.com Thu Mar 11 04:04:22 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Thu, 11 Mar 2010 04:04:26 +0100 (CET) Received: from mail-ew0-f224.google.com ([209.85.219.224]:33978 "EHLO mail-ew0-f224.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491867Ab0CKDEW (ORCPT ); Thu, 11 Mar 2010 04:04:22 +0100 Received: by ewy24 with SMTP id 24so1842821ewy.27 for ; Wed, 10 Mar 2010 19:04:15 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:from:to:cc:subject:date :message-id:x-mailer; bh=jK+qhN8lHzICC6Jx7pteJ4qsmQU7yV8FGIQ2lYyUWAo=; b=AkgiG8Brv01xXc16ynycx3X4AKQrfEt8eKeAWKsItwNZeijNbg/CdTteWl/hb354t3 +NQ+1q+7b/MFypmunDpdjI42fMw2+Il37MoutCj0h9aKIoO3yaS0m8k3Ahj+97aLKQz0 pXvb0JjkAOYKgP8vN7EgL9IKYUgKc5zI65g6U= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=from:to:cc:subject:date:message-id:x-mailer; b=hYMtEQs8/r5S8S04SVq1tmxK+ngz4ZK8s7S9PrKqPx5jd1IaCkrc0z9aaw9PPh/lbL tRxmhTRD6ToDWCQmRbk22V36GPUc6Q40Srb5ggk0XAuowmwjHeXKfT4v55/6+E5xZewk EyCZWRodkQHMjosxZm+v6KeFuEQLAy2mOGUwE= Received: by 10.213.97.28 with SMTP id j28mr5823117ebn.44.1268276653827; Wed, 10 Mar 2010 19:04:13 -0800 (PST) Received: from localhost.localdomain ([202.201.12.142]) by mx.google.com with ESMTPS id 14sm4394844ewy.10.2010.03.10.19.04.08 (version=TLSv1/SSLv3 cipher=RC4-MD5); Wed, 10 Mar 2010 19:04:12 -0800 (PST) From: Wu Zhangjin To: Ralf Baechle Cc: linux-mips@linux-mips.org, David Daney , Wu Zhangjin Subject: [PATCH v2] Loongson: Lemote-2F: Fixup of _rdmsr and _wrmsr Date: Thu, 11 Mar 2010 10:57:33 +0800 Message-Id: <5c63967ed3f891da1f6bc1fc78c272d0407d5d25.1268276186.git.wuzhangjin@gmail.com> X-Mailer: git-send-email 1.7.0.1 Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26185 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: wuzhangjin@gmail.com Precedence: bulk X-list: linux-mips From: Wu Zhangjin Changes from the old version: o Using raw spinlock instead of spinlock as suggested by David Daney. The _rdmsr, _wrmsr operation must be atomic to ensure the accessing to msr address is what we want. Without this patch, in some cases, the reboot operation(fs2f_reboot) defined in arch/mips/loongson/lemote-2f/reset.c may fail for it called _rdmsr, _wrmsr but may be interrupted/preempted by the other related operations and make the _rdmsr get the wrong value or make the _wrmsr write a wrong value to an unexpected target. Signed-off-by: Wu Zhangjin --- arch/mips/pci/ops-loongson2.c | 10 ++++++++++ 1 files changed, 10 insertions(+), 0 deletions(-) diff --git a/arch/mips/pci/ops-loongson2.c b/arch/mips/pci/ops-loongson2.c index 2bb4057..dc434ec 100644 --- a/arch/mips/pci/ops-loongson2.c +++ b/arch/mips/pci/ops-loongson2.c @@ -180,15 +180,21 @@ struct pci_ops loongson_pci_ops = { }; #ifdef CONFIG_CS5536 +DEFINE_RAW_SPINLOCK(msr_lock); + void _rdmsr(u32 msr, u32 *hi, u32 *lo) { struct pci_bus bus = { .number = PCI_BUS_CS5536 }; u32 devfn = PCI_DEVFN(PCI_IDSEL_CS5536, 0); + unsigned long flags; + + spin_lock_irqsave(&msr_lock, flags); loongson_pcibios_write(&bus, devfn, PCI_MSR_ADDR, 4, msr); loongson_pcibios_read(&bus, devfn, PCI_MSR_DATA_LO, 4, lo); loongson_pcibios_read(&bus, devfn, PCI_MSR_DATA_HI, 4, hi); + spin_unlock_irqrestore(&msr_lock, flags); } EXPORT_SYMBOL(_rdmsr); @@ -198,9 +204,13 @@ void _wrmsr(u32 msr, u32 hi, u32 lo) .number = PCI_BUS_CS5536 }; u32 devfn = PCI_DEVFN(PCI_IDSEL_CS5536, 0); + unsigned long flags; + + spin_lock_irqsave(&msr_lock, flags); loongson_pcibios_write(&bus, devfn, PCI_MSR_ADDR, 4, msr); loongson_pcibios_write(&bus, devfn, PCI_MSR_DATA_LO, 4, lo); loongson_pcibios_write(&bus, devfn, PCI_MSR_DATA_HI, 4, hi); + spin_unlock_irqrestore(&msr_lock, flags); } EXPORT_SYMBOL(_wrmsr); #endif -- 1.7.0.1 From wuzhangjin@gmail.com Thu Mar 11 04:11:52 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Thu, 11 Mar 2010 04:11:56 +0100 (CET) Received: from mail-bw0-f215.google.com ([209.85.218.215]:44214 "EHLO mail-bw0-f215.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1490963Ab0CKDLw (ORCPT ); Thu, 11 Mar 2010 04:11:52 +0100 Received: by bwz7 with SMTP id 7so6743579bwz.24 for ; Wed, 10 Mar 2010 19:11:45 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:from:to:cc:subject:date :message-id:x-mailer:mime-version:content-type :content-transfer-encoding; bh=M+/1Nf6//C1Mhbr3obKmGo5Tb9r0JGT/4+1EJGzfUzM=; b=gdE2Hu8IlccXs0N8YhLoLiB+dE0aULgO0se/TcECOWenLfyTFO1SfCd/GuVo/kZKdn dVlQ8V05lHKvXP2izIzLZhNXA+98VOOxxaoJbKL2yTwvyvbMz5dUDWWEJmDc4kNo55Td 2vkPTbrLCQK0f1vtUdePA8d+VxbIA9t7q7qrM= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=from:to:cc:subject:date:message-id:x-mailer:mime-version :content-type:content-transfer-encoding; b=cPuE2CLki/9FS0Mn2cQs5Q+hIrD2EuZT1JR4hJEMxSXyl5KQImtG2/EuNb3MFAn/xi zilN3gL3/euCbri3scmBwa5WcUfQufa3/eD++arGgcugAS3hRw1jyamh4NqevDFUiEZW 729AZ5F4bLT4UQI6FlICTlO6otIdQpEesGvrg= Received: by 10.204.7.139 with SMTP id d11mr215201bkd.162.1268277102328; Wed, 10 Mar 2010 19:11:42 -0800 (PST) Received: from localhost.localdomain ([202.201.12.142]) by mx.google.com with ESMTPS id g18sm30688020bkw.13.2010.03.10.19.11.37 (version=TLSv1/SSLv3 cipher=RC4-MD5); Wed, 10 Mar 2010 19:11:41 -0800 (PST) From: Wu Zhangjin To: Ralf Baechle Cc: linux-mips@linux-mips.org, Sergei Shtylyov , Wu Zhangjin Subject: [PATCH v2 0/3] Workaround the Out-of-order Issue of Loongson-2F Date: Thu, 11 Mar 2010 11:05:01 +0800 Message-Id: X-Mailer: git-send-email 1.7.0.1 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26186 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: wuzhangjin@gmail.com Precedence: bulk X-list: linux-mips From: Wu Zhangjin Changes from old version: o Cleanup some comments and align some instructions. ---- As the Chapter 15: "Errata: Issue of Out-of-order in loongson"[1] shows, to workaround the Issue of Loongson-2F，We need to do: o When switching from user model to kernel model, you should flush the branch target history such as BTB and RAS. o Doing some tricks to the indirect branch target to make sure that the indirect branch target can not be in the I/O region. This patchset applied the above methods and for the binutils patch[3] have been merged into binutils 2.20.1, so, it's time to upstream this patchset now. without this patch, the machines will hang when the instruction sequence hit the Out-of-order Issue of Loongson-2F, therefore, this patchset is very urgent for both 2.6.33 and 2.6.34. [1] Chinese Version: http://www.loongson.cn/uploadfile/file/20080821113149.pdf [2] English Version of Chapter 15: http://groups.google.com.hk/group/loongson-dev/msg/e0d2e220958f10a6?dmode=source [3] http://sourceware.org/ml/binutils/2009-11/msg00387.html Regards, Wu Zhangjin Wu Zhangjin (3): Loongson-2F: Flush the branch target history such as BTB and RAS Loongson-2F: Enable fixups of binutils 2.20.1 Loongson-2F: Fixup of problems introduced by -mfix-loongson2f-jump of binutils 2.20.1 arch/mips/Makefile | 4 +++- arch/mips/include/asm/stackframe.h | 19 +++++++++++++++++++ arch/mips/loongson/common/reset.c | 11 ++++++++++- 3 files changed, 32 insertions(+), 2 deletions(-) From wuzhangjin@gmail.com Thu Mar 11 04:12:14 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Thu, 11 Mar 2010 04:12:20 +0100 (CET) Received: from mail-bw0-f215.google.com ([209.85.218.215]:44214 "EHLO mail-bw0-f215.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491069Ab0CKDMA (ORCPT ); Thu, 11 Mar 2010 04:12:00 +0100 Received: by mail-bw0-f215.google.com with SMTP id 7so6743579bwz.24 for ; Wed, 10 Mar 2010 19:12:00 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:from:to:cc:subject:date :message-id:x-mailer:in-reply-to:references:in-reply-to:references :mime-version:content-type:content-transfer-encoding; bh=Op90jrXvfNgeLyTOb7jEkgGfxp0RHsZl8xhaeMbPWoQ=; b=RcyuveuyVyT1psbDRkDsjq1g0IJM92exyFTKgyoN5n5xVUi16twFRhSkIwXYfpLxic F2ol9OzKeB9dM1xdezJL/y1RjeuqgKVrRaSJ1UiScqT19JJz5PovG+rE0YW4dJjgq0W+ 7ltz6rhQwZuSUPpQFS9sk+buM4DJ+7QrwZyMs= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=from:to:cc:subject:date:message-id:x-mailer:in-reply-to:references :mime-version:content-type:content-transfer-encoding; b=JTBp13OaVmPQ4c+nGX8zZjqaJOkD2q3xHg41sYWWcGWKP7RFLOVQ1hNG/KsRN9QCJy zEv/2vJZdLPqph/1TtgqSfkMzVFpMRrtN4+n09xNrZvKtC4GiZWDyj5w/5u21CXQDlW+ n8VmCg/E3+uzvIRMM+7YLeObgBpFHdMIodbsI= Received: by 10.204.9.23 with SMTP id j23mr2862185bkj.132.1268277116826; Wed, 10 Mar 2010 19:11:56 -0800 (PST) Received: from localhost.localdomain ([202.201.12.142]) by mx.google.com with ESMTPS id g18sm30688020bkw.13.2010.03.10.19.11.51 (version=TLSv1/SSLv3 cipher=RC4-MD5); Wed, 10 Mar 2010 19:11:55 -0800 (PST) From: Wu Zhangjin To: Ralf Baechle Cc: linux-mips@linux-mips.org, Sergei Shtylyov , Wu Zhangjin Subject: [PATCH v2 1/3] Loongson-2F: Flush the branch target history such as BTB and RAS Date: Thu, 11 Mar 2010 11:05:02 +0800 Message-Id: <23a3955f40b7466a3e850ae32098591b1a6a5ad4.1268276417.git.wuzhangjin@gmail.com> X-Mailer: git-send-email 1.7.0.1 In-Reply-To: References: In-Reply-To: References: MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26187 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: wuzhangjin@gmail.com Precedence: bulk X-list: linux-mips From: Wu Zhangjin As the Chapter 15: "Errata: Issue of Out-of-order in loongson"[1] shows, to workaround the Issue of Loongson-2F，We need to do: "When switching from user model to kernel model, you should flush the branch target history such as BTB and RAS." This patch did clear BTB(branch target buffer), forbid RAS(row address strobe) via Loongson-2F's 64bit diagnostic register. [1] Chinese Version: http://www.loongson.cn/uploadfile/file/200808211 [2] English Version of Chapter 15: http://groups.google.com.hk/group/loongson-dev/msg/e0d2e220958f10a6?dmode=source Signed-off-by: Wu Zhangjin --- arch/mips/include/asm/stackframe.h | 19 +++++++++++++++++++ 1 files changed, 19 insertions(+), 0 deletions(-) diff --git a/arch/mips/include/asm/stackframe.h b/arch/mips/include/asm/stackframe.h index 3b6da33..52a62f5 100644 --- a/arch/mips/include/asm/stackframe.h +++ b/arch/mips/include/asm/stackframe.h @@ -121,6 +121,25 @@ .endm #else .macro get_saved_sp /* Uniprocessor variation */ +#ifdef CONFIG_CPU_LOONGSON2F + /* + * Clear BTB (branch target buffer), forbid RAS (row address + * strobe) to workaround the Out-of-order Issue in Loongson2F + * via its diagnostic register. + */ + move k0, ra + jal 1f + nop +1: jal 1f + nop +1: jal 1f + nop +1: jal 1f + nop +1: move ra, k0 + li k0, 3 + mtc0 k0, $22 +#endif /* CONFIG_CPU_LOONGSON2F */ #if defined(CONFIG_32BIT) || defined(KBUILD_64BIT_SYM32) lui k1, %hi(kernelsp) #else -- 1.7.0.1 From wuzhangjin@gmail.com Thu Mar 11 04:12:39 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Thu, 11 Mar 2010 04:12:43 +0100 (CET) Received: from mail-bw0-f215.google.com ([209.85.218.215]:44214 "EHLO mail-bw0-f215.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491121Ab0CKDMH (ORCPT ); Thu, 11 Mar 2010 04:12:07 +0100 Received: by mail-bw0-f215.google.com with SMTP id 7so6743579bwz.24 for ; Wed, 10 Mar 2010 19:12:06 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:from:to:cc:subject:date :message-id:x-mailer:in-reply-to:references:in-reply-to:references; bh=P2v0PzTZ8PcrOj5xc9fJWbeIzqwSM6dukdMjQ4UhZUY=; b=fizzoJF/ThTAq74JA5pU6dhM++z1FkFYOlPmGoKrQHvqg2SoJiNC9guPwdZcmXVLR4 MbpMq6Fnl0EHqDa/t09p7hWWr9i0bh/JpU6egyg8COde5MPR9EFIUXKFPIgDigfSH+ER CULPHudKvklyGOwOK8FOrx2OC3w2HpHGvq7T4= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=from:to:cc:subject:date:message-id:x-mailer:in-reply-to:references; b=pw9k8nKii1N9Co8/B9o+EBqEuD8u+qONQkYOPKzW4M4uJGMSVSZM2whISthGnxDUe7 0emnu7wtJy9ZzF0SenQONppjwnkni3T2OH9aQ4NyB1MLr+uIJ/ZKB4aHPD47430Kf7io IS5zGuNknVjXiwwJskgJpR8AepJUIL36nkzF8= Received: by 10.204.129.218 with SMTP id p26mr2801141bks.145.1268277124937; Wed, 10 Mar 2010 19:12:04 -0800 (PST) Received: from localhost.localdomain ([202.201.12.142]) by mx.google.com with ESMTPS id g18sm30688020bkw.13.2010.03.10.19.11.58 (version=TLSv1/SSLv3 cipher=RC4-MD5); Wed, 10 Mar 2010 19:12:03 -0800 (PST) From: Wu Zhangjin To: Ralf Baechle Cc: linux-mips@linux-mips.org, Sergei Shtylyov , Wu Zhangjin , Zhang Le , Wu Zhangjin Subject: [PATCH v2 2/3] Loongson-2F: Enable fixups of binutils 2.20.1 Date: Thu, 11 Mar 2010 11:05:03 +0800 Message-Id: X-Mailer: git-send-email 1.7.0.1 In-Reply-To: References: In-Reply-To: References: Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26188 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: wuzhangjin@gmail.com Precedence: bulk X-list: linux-mips From: Wu Zhangjin As the "Fixups of Loongson2F" patch[1] to binutils have been applied into binutils 2.20.1. It's time to enable the options provided by the patch to compile the kernel. Without these fixups, the system will hang unexpectedly for the bug of processor. To learn more about these fixups, please refer to the following references. [1] "Fixups of Loongson2F" patch for binutils(actually for gas) http://sourceware.org/ml/binutils/2009-11/msg00387.html [2] Chapter 15 of "Loongson2F User Manual"(Chinese Version) http://www.loongson.cn/uploadfile/file/200808211 [3] English Version of the above chapter 15 http://groups.google.com.hk/group/loongson-dev/msg/e0d2e220958f10a6?dmode=source Signed-off-by: Zhang Le Signed-off-by: Wu Zhangjin --- arch/mips/Makefile | 4 +++- 1 files changed, 3 insertions(+), 1 deletions(-) diff --git a/arch/mips/Makefile b/arch/mips/Makefile index 2f2eac2..5ae342e 100644 --- a/arch/mips/Makefile +++ b/arch/mips/Makefile @@ -135,7 +135,9 @@ cflags-$(CONFIG_CPU_LOONGSON2) += -Wa,--trap cflags-$(CONFIG_CPU_LOONGSON2E) += \ $(call cc-option,-march=loongson2e,-march=r4600) cflags-$(CONFIG_CPU_LOONGSON2F) += \ - $(call cc-option,-march=loongson2f,-march=r4600) + $(call cc-option,-march=loongson2f,-march=r4600) \ + $(call as-option,-Wa$(comma)-mfix-loongson2f-nop,) \ + $(call as-option,-Wa$(comma)-mfix-loongson2f-jump,) cflags-$(CONFIG_CPU_MIPS32_R1) += $(call cc-option,-march=mips32,-mips32 -U_MIPS_ISA -D_MIPS_ISA=_MIPS_ISA_MIPS32) \ -Wa,-mips32 -Wa,--trap -- 1.7.0.1 From Yang.Shi@windriver.com Thu Mar 11 04:13:01 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Thu, 11 Mar 2010 04:13:07 +0100 (CET) Received: from mail.windriver.com ([147.11.1.11]:33088 "EHLO mail.windriver.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491124Ab0CKDMN (ORCPT ); Thu, 11 Mar 2010 04:12:13 +0100 Received: from ALA-MAIL03.corp.ad.wrs.com (ala-mail03 [147.11.57.144]) by mail.windriver.com (8.14.3/8.14.3) with ESMTP id o2B3C6qD024004; Wed, 10 Mar 2010 19:12:06 -0800 (PST) Received: from [128.224.162.222] ([128.224.162.222]) by ALA-MAIL03.corp.ad.wrs.com with Microsoft SMTPSVC(6.0.3790.1830); Wed, 10 Mar 2010 19:12:05 -0800 Message-ID: <4B985F86.8010701@windriver.com> Date: Thu, 11 Mar 2010 11:12:06 +0800 From: Yang Shi User-Agent: Thunderbird 2.0.0.23 (X11/20090817) MIME-Version: 1.0 To: Ralf Baechle CC: David Daney , linux-mips@linux-mips.org Subject: Re: [PATCH] MIPS: Protect current_cpu_data with preempt disable in delay() References: <1267695573-27360-1-git-send-email-yang.shi@windriver.com> <4B8FFAB3.1090409@caviumnetworks.com> <20100310153315.GA12476@linux-mips.org> In-Reply-To: <20100310153315.GA12476@linux-mips.org> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 8bit X-OriginalArrivalTime: 11 Mar 2010 03:12:05.0532 (UTC) FILETIME=[A0AF5DC0:01CAC0C8] Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26189 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: yang.shi@windriver.com Precedence: bulk X-list: linux-mips Ralf Baechle 写é“: > On Thu, Mar 04, 2010 at 10:23:47AM -0800, David Daney wrote: > > >> On 03/04/2010 01:39 AM, Yang Shi wrote: >> >>> During machine restart with reboot command, get the following >>> bug info: >>> >>> BUG: using smp_processor_id() in preemptible [00000000] code: reboot/1989 >>> caller is __udelay+0x14/0x70 >>> Call Trace: >>> [] dump_stack+0x8/0x34 >>> [] debug_smp_processor_id+0xf4/0x110 >>> [] __udelay+0x14/0x70 >>> [] md_notify_reboot+0x12c/0x148 >>> [] notifier_call_chain+0x64/0xc8 >>> [] __blocking_notifier_call_chain+0x64/0xc0 >>> [] kernel_restart_prepare+0x1c/0x38 >>> [] kernel_restart+0x14/0x50 >>> [] SyS_reboot+0x10c/0x1f0 >>> [] handle_sysn32+0x44/0x84 >>> >>> The root cause is that current_cpu_data is accessed in preemptible >>> context, so protect it with preempt_disable/preempt_enable pair >>> in delay(). >>> >>> Signed-off-by: Yang Shi >>> --- >>> arch/mips/lib/delay.c | 6 +++++- >>> 1 files changed, 5 insertions(+), 1 deletions(-) >>> >>> diff --git a/arch/mips/lib/delay.c b/arch/mips/lib/delay.c >>> index 6b3b1de..dc38064 100644 >>> --- a/arch/mips/lib/delay.c >>> +++ b/arch/mips/lib/delay.c >>> @@ -41,7 +41,11 @@ EXPORT_SYMBOL(__delay); >>> >>> void __udelay(unsigned long us) >>> { >>> - unsigned int lpj = current_cpu_data.udelay_val; >>> + unsigned int lpj; >>> + >>> + preempt_disable(); >>> + lpj = current_cpu_data.udelay_val; >>> + preempt_enable(); >>> >>> __delay((us * 0x000010c7ull * HZ * lpj)>> 32); >>> } >>> >> This doesn't seem like the best approach. >> >> Perhaps we should either use raw_current_cpu_data and no >> preempt_disable(), or if we are concerned about migrating to a CPU >> with a different lpj value, move the preempt_enable after the call >> to __delay(). >> > > Udelay() is supposed to guarantee a minimum delay and when being migrated > to another CPU with higher bogomips this guarantee might be violated. So > it'd even have to be something like: > > void __udelay(unsigned long us) > { > unsigned int lpj = current_cpu_data.udelay_val; > unsigned int lpj; > > preempt_disable(); > lpj = current_cpu_data.udelay_val; > > __delay((us * 0x000010c7ull * HZ * lpj)>> 32); > preempt_enable(); > } > > But preempt_disable() itself is not atomic, so using it from bh or irq > context could result in a corrupted preemption counter. So the raw_ > version will have to do. I doubt it's much of a problem but at some > point we will have to revisit the delay by c0_count patch submitted a > while ago. The patch wasn't right but the problem it was addressing > is real. > Thanks Ralf. Do we need raw_ version patch before revisiting the delay by c0_count patch although it's not an ideal fix. Regards, Yang > Ralf > > From wuzhangjin@gmail.com Thu Mar 11 04:13:24 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Thu, 11 Mar 2010 04:13:30 +0100 (CET) Received: from mail-bw0-f215.google.com ([209.85.218.215]:44214 "EHLO mail-bw0-f215.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491149Ab0CKDMN (ORCPT ); Thu, 11 Mar 2010 04:12:13 +0100 Received: by mail-bw0-f215.google.com with SMTP id 7so6743579bwz.24 for ; Wed, 10 Mar 2010 19:12:13 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:from:to:cc:subject:date :message-id:x-mailer:in-reply-to:references:in-reply-to:references; bh=zr893OWQ3vUwhyg6oJp4MD6FThtWm5mK31cbV664TB8=; b=GTyKF3r4cyVFaMyu1zZlWDtJqhUyWeaomeTRb/NDCxRXkvYq9tt1ipXQLoTWabVYHi wX7hwVrcv9rtczLttDq/a5L90uk/d7JqiXvdVjd40dD+kv7TDNWO1NXqo1wkl6zMjFMp VKPRR8RANejuKbPzaMA11XmkBOuPH8pBW8pZc= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=from:to:cc:subject:date:message-id:x-mailer:in-reply-to:references; b=LcFdt2zig7Ql34mhXaSEoL+4ywF09rNKz+/Z4bAlMYO8urO14JmJZLB7qkYWDLfex1 9UG++uU3FwOViwBf8kFmkGwqTLQ0t3NotjYA1OKTKVbUwoQAAPr0XZe/3CoqNPYybkA6 J8ZqVNHfUd2aF5eDB2CveEKZLi2fJUpjwHNhU= Received: by 10.204.21.197 with SMTP id k5mr2957704bkb.28.1268277131942; Wed, 10 Mar 2010 19:12:11 -0800 (PST) Received: from localhost.localdomain ([202.201.12.142]) by mx.google.com with ESMTPS id g18sm30688020bkw.13.2010.03.10.19.12.06 (version=TLSv1/SSLv3 cipher=RC4-MD5); Wed, 10 Mar 2010 19:12:10 -0800 (PST) From: Wu Zhangjin To: Ralf Baechle Cc: linux-mips@linux-mips.org, Sergei Shtylyov , Wu Zhangjin Subject: [PATCH v2 3/3] Loongson-2F: Fixup of problems introduced by -mfix-loongson2f-jump of binutils 2.20.1 Date: Thu, 11 Mar 2010 11:05:04 +0800 Message-Id: <1bd51326ffa900d034ee81825e803e3bdc46fdaf.1268276417.git.wuzhangjin@gmail.com> X-Mailer: git-send-email 1.7.0.1 In-Reply-To: References: In-Reply-To: References: Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26190 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: wuzhangjin@gmail.com Precedence: bulk X-list: linux-mips From: Wu Zhangjin The -mfix-loongson2f-jump option provided by the binutils 2.20.1 have fixed the Out-of-order Issue of Loongson-2F described in Chapter 15 of "Loongson2F User Manual"[1,2], but introduced some problems. The option changes all of the jumping target to "addr & 0xcfffffff" through the at($1) register, but for the REBOOT address of loongson-2F: 0xbfc00000, this is totally wrong, so, this patch try to avoid the problem via telling the assembler not to use at($1) register. [1] Loongson2F User Manual(Chinese Version) http://www.loongson.cn/uploadfile/file/200808211 [2] English Version of Chapter 15: http://groups.google.com.hk/group/loongson-dev/msg/e0d2e220958f10a6?dmode=source Reported-and-tested-by: Liu Shiwei Signed-off-by: Wu Zhangjin --- arch/mips/loongson/common/reset.c | 11 ++++++++++- 1 files changed, 10 insertions(+), 1 deletions(-) diff --git a/arch/mips/loongson/common/reset.c b/arch/mips/loongson/common/reset.c index 4bd9c18..dc311f2 100644 --- a/arch/mips/loongson/common/reset.c +++ b/arch/mips/loongson/common/reset.c @@ -21,8 +21,17 @@ static void loongson_restart(char *command) /* do preparation for reboot */ mach_prepare_reboot(); - /* reboot via jumping to boot base address */ + /* reboot via jumping to boot base address + * + * ".set noat" and ".set at" are used to ensure the address not break + * by the -mfix-loongson2f-jump option provided by binutils 2.20.1 (or + * higher version) which try to change the jumping address to "addr & + * 0xcfffffff" via the at($1) register, this is totally wrong for + * 0xbfc00000 (LOONGSON_BOOT_BASE). + */ + __asm__ __volatile__(".set noat\n"); ((void (*)(void))ioremap_nocache(LOONGSON_BOOT_BASE, 4)) (); + __asm__ __volatile__(".set at\n"); } static void loongson_poweroff(void) -- 1.7.0.1 From wuzhangjin@gmail.com Thu Mar 11 04:18:23 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Thu, 11 Mar 2010 04:18:28 +0100 (CET) Received: from mail-bw0-f222.google.com ([209.85.218.222]:38814 "EHLO mail-bw0-f222.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491124Ab0CKDSX (ORCPT ); Thu, 11 Mar 2010 04:18:23 +0100 Received: by bwz22 with SMTP id 22so2533181bwz.28 for ; Wed, 10 Mar 2010 19:18:17 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:from:to:cc:subject:date :message-id:x-mailer; bh=OABVKFmG5HfY3FTLMsWsjqgFn/jHClUB6vOaxMo+Lc0=; b=MTl2DU+S5d+6ohamHwyQ3G6/X/3wMoRMGpYxCjdWdHuasH+J1wXKXamrALwi9GQenZ o0GiTyesNuxjJOBg2UaKYsBLHNZh5L+UOAx7u6F6TeHEf/di58qxx6b9Beui5nACLISK 9QHU9K7yAglePjk7NVa6bTN4Yen3EYpQrnTi4= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=from:to:cc:subject:date:message-id:x-mailer; b=cYxQA2aon/NhldDuD9CzWtSxBTTw1NwKTFOlYLMubkyo9AGDrU201O2T/szOV5Vuum e9Af+ZeD8pQPdRuN/dnBHY0/nwTNCzUF2O45N2Ft6KQapVDyVijlkzA6DUMONd9UterU /HeaBRQ940x4xw15OTaopMSKeeMLZ70ACKNa0= Received: by 10.204.137.16 with SMTP id u16mr2825308bkt.165.1268277497435; Wed, 10 Mar 2010 19:18:17 -0800 (PST) Received: from localhost.localdomain ([202.201.12.142]) by mx.google.com with ESMTPS id 24sm30711336bkr.18.2010.03.10.19.18.10 (version=TLSv1/SSLv3 cipher=RC4-MD5); Wed, 10 Mar 2010 19:18:15 -0800 (PST) From: Wu Zhangjin To: Ralf Baechle Cc: Pavel Machek , "Rafael J. Wysocki" , linux-mips@linux-mips.org, Wu Zhangjin , Liu Shiwei Subject: [v12 9/9] Loongson: YeeLoong: add power_supply based battery driver Date: Thu, 11 Mar 2010 11:11:21 +0800 Message-Id: X-Mailer: git-send-email 1.7.0.1 Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26191 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: wuzhangjin@gmail.com Precedence: bulk X-list: linux-mips From: Wu Zhangjin (This patch is one of the "yeeloong platform driver" patchset.) Changes from old revision: o Fixes the bugs In the old driver, it didn't show the current charging status and reported the wrong full charge and voltage value. Capacity = Charge_full_last (Dynamic) / Charge_full_design (Fixed) Percentage Charge = Charge_now (Dyn-Dynamic) / Charge_full_last (Dynamic) Based on the old emulated APM battery driver and the power_supply class, this patch adds a new battery driver. References: 1. Documentation/power/power_supply_class.txt 2. drivers/power/ Signed-off-by: Liu Shiwei Signed-off-by: Wu Zhangjin --- drivers/platform/mips/Kconfig | 1 + drivers/platform/mips/yeeloong_laptop.c | 228 +++++++++++++++++++++++++++++++ 2 files changed, 229 insertions(+), 0 deletions(-) diff --git a/drivers/platform/mips/Kconfig b/drivers/platform/mips/Kconfig index 01560b0..cdfccea 100644 --- a/drivers/platform/mips/Kconfig +++ b/drivers/platform/mips/Kconfig @@ -21,6 +21,7 @@ config LEMOTE_YEELOONG2F select HWMON select VIDEO_OUTPUT_CONTROL select INPUT_SPARSEKMAP + select POWER_SUPPLY depends on INPUT help YeeLoong netbook is a mini laptop made by Lemote, which is basically diff --git a/drivers/platform/mips/yeeloong_laptop.c b/drivers/platform/mips/yeeloong_laptop.c index 877257a..ead26e0 100644 --- a/drivers/platform/mips/yeeloong_laptop.c +++ b/drivers/platform/mips/yeeloong_laptop.c @@ -19,6 +19,7 @@ #include #include #include +#include /* for AC & Battery subdriver */ #include @@ -349,6 +350,213 @@ static void yeeloong_hwmon_exit(void) } } +/* AC & Battery subdriver */ + +static struct power_supply yeeloong_ac, yeeloong_bat; + +#define AC_OFFLINE 0 +#define AC_ONLINE 1 + +static int yeeloong_get_ac_props(struct power_supply *psy, + enum power_supply_property psp, + union power_supply_propval *val) +{ + switch (psp) { + case POWER_SUPPLY_PROP_ONLINE: + val->intval = ((ec_read(REG_BAT_POWER)) & BIT_BAT_POWER_ACIN) ? + AC_ONLINE : AC_OFFLINE; + break; + default: + return -EINVAL; + } + + return 0; +} + +static enum power_supply_property yeeloong_ac_props[] = { + POWER_SUPPLY_PROP_ONLINE, +}; + +static struct power_supply yeeloong_ac = { + .name = "yeeloong-ac", + .type = POWER_SUPPLY_TYPE_MAINS, + .properties = yeeloong_ac_props, + .num_properties = ARRAY_SIZE(yeeloong_ac_props), + .get_property = yeeloong_get_ac_props, +}; + +#define BAT_CAP_CRITICAL 5 +#define BAT_CAP_HIGH 99 + +static int yeeloong_bat_get_ex_property(enum power_supply_property psp, + union power_supply_propval *val) +{ + int bat_in, curr_cap, cap_level, status, charge, health; + + status = ec_read(REG_BAT_STATUS); + bat_in = status & BIT_BAT_STATUS_IN; + curr_cap = get_bat_info(RELATIVE_CAP); + if (status & BIT_BAT_STATUS_FULL) + curr_cap = 100; + + switch (psp) { + case POWER_SUPPLY_PROP_PRESENT: + val->intval = bat_in; + break; + case POWER_SUPPLY_PROP_CAPACITY: + val->intval = curr_cap; + break; + case POWER_SUPPLY_PROP_CAPACITY_LEVEL: + cap_level = POWER_SUPPLY_CAPACITY_LEVEL_NORMAL; + if (status & BIT_BAT_STATUS_LOW) { + cap_level = POWER_SUPPLY_CAPACITY_LEVEL_LOW; + if (curr_cap <= BAT_CAP_CRITICAL) + cap_level = + POWER_SUPPLY_CAPACITY_LEVEL_CRITICAL; + } else if (status & BIT_BAT_STATUS_FULL) { + cap_level = POWER_SUPPLY_CAPACITY_LEVEL_FULL; + if (curr_cap >= BAT_CAP_HIGH) + cap_level = POWER_SUPPLY_CAPACITY_LEVEL_HIGH; + } else if (status & BIT_BAT_STATUS_DESTROY) + cap_level = POWER_SUPPLY_CAPACITY_LEVEL_UNKNOWN; + val->intval = cap_level; + break; + case POWER_SUPPLY_PROP_TIME_TO_EMPTY_NOW: + /* seconds */ + val->intval = bat_in ? (curr_cap - 3) * 54 + 142 : 0; + break; + case POWER_SUPPLY_PROP_STATUS: + if (!bat_in) + charge = POWER_SUPPLY_STATUS_UNKNOWN; + else { + if (status & BIT_BAT_STATUS_FULL) { + val->intval = POWER_SUPPLY_STATUS_FULL; + break; + } + + charge = ec_read(REG_BAT_CHARGE); + if (charge & FLAG_BAT_CHARGE_DISCHARGE) + charge = POWER_SUPPLY_STATUS_DISCHARGING; + else if (charge & FLAG_BAT_CHARGE_CHARGE) + charge = POWER_SUPPLY_STATUS_CHARGING; + else + charge = POWER_SUPPLY_STATUS_NOT_CHARGING; + } + val->intval = charge; + break; + case POWER_SUPPLY_PROP_HEALTH: + if (!bat_in) /* no battery present */ + health = POWER_SUPPLY_HEALTH_UNKNOWN; + else { /* Assume it is good */ + health = POWER_SUPPLY_HEALTH_GOOD; + if (status & + (BIT_BAT_STATUS_DESTROY | BIT_BAT_STATUS_LOW)) + health = POWER_SUPPLY_HEALTH_DEAD; + if (ec_read(REG_BAT_CHARGE_STATUS) & + BIT_BAT_CHARGE_STATUS_OVERTEMP) + health = POWER_SUPPLY_HEALTH_OVERHEAT; + } + val->intval = health; + break; + case POWER_SUPPLY_PROP_CHARGE_NOW: /* 1/100(%)*1000 µAh */ + val->intval = curr_cap * get_bat_info(FULLCHG_CAP) * 10; + break; + default: + return -EINVAL; + } + return 0; +} + +static int yeeloong_get_bat_props(struct power_supply *psy, + enum power_supply_property psp, + union power_supply_propval *val) +{ + switch (psp) { + /* Fixed information */ + case POWER_SUPPLY_PROP_VOLTAGE_MAX_DESIGN: + val->intval = get_bat_info(DESIGN_VOL) * 1000; /* mV -> µV */ + break; + case POWER_SUPPLY_PROP_CHARGE_FULL_DESIGN: + val->intval = get_bat_info(DESIGN_CAP) * 1000; /*mAh -> µAh*/ + break; + case POWER_SUPPLY_PROP_CHARGE_FULL: + val->intval = get_bat_info(FULLCHG_CAP) * 1000; /* µAh */ + break; + case POWER_SUPPLY_PROP_MANUFACTURER: + val->strval = + (ec_read(REG_BAT_VENDOR) == FLAG_BAT_VENDOR_SANYO) ? + "SANYO" : "SIMPLO"; + break; + /* Dynamic information */ + case POWER_SUPPLY_PROP_CURRENT_NOW: + val->intval = get_battery_current() * 1000; /* mA -> µA */ + break; + case POWER_SUPPLY_PROP_VOLTAGE_NOW: + val->intval = get_battery_voltage() * 1000; /* mV -> µV */ + break; + case POWER_SUPPLY_PROP_TEMP: + val->intval = get_battery_temp(); /* Celcius */ + break; + /* Dynamic but relative information */ + default: + return yeeloong_bat_get_ex_property(psp, val); + } + + return 0; +} + +static enum power_supply_property yeeloong_bat_props[] = { + POWER_SUPPLY_PROP_STATUS, + POWER_SUPPLY_PROP_PRESENT, + POWER_SUPPLY_PROP_VOLTAGE_MAX_DESIGN, + POWER_SUPPLY_PROP_CHARGE_FULL_DESIGN, + POWER_SUPPLY_PROP_CHARGE_FULL, + POWER_SUPPLY_PROP_CHARGE_NOW, + POWER_SUPPLY_PROP_CURRENT_NOW, + POWER_SUPPLY_PROP_VOLTAGE_NOW, + POWER_SUPPLY_PROP_HEALTH, + POWER_SUPPLY_PROP_TIME_TO_EMPTY_NOW, + POWER_SUPPLY_PROP_CAPACITY, + POWER_SUPPLY_PROP_CAPACITY_LEVEL, + POWER_SUPPLY_PROP_TEMP, + POWER_SUPPLY_PROP_MANUFACTURER, +}; + +static struct power_supply yeeloong_bat = { + .name = "yeeloong-bat", + .type = POWER_SUPPLY_TYPE_BATTERY, + .properties = yeeloong_bat_props, + .num_properties = ARRAY_SIZE(yeeloong_bat_props), + .get_property = yeeloong_get_bat_props, +}; + +static int ac_bat_initialized; + +static int yeeloong_bat_init(void) +{ + int ret; + + ret = power_supply_register(NULL, &yeeloong_ac); + if (ret) + return ret; + ret = power_supply_register(NULL, &yeeloong_bat); + if (ret) { + power_supply_unregister(&yeeloong_ac); + return ret; + } + ac_bat_initialized = 1; + + return 0; +} + +static void yeeloong_bat_exit(void) +{ + ac_bat_initialized = 0; + + power_supply_unregister(&yeeloong_ac); + power_supply_unregister(&yeeloong_bat); +} + /* video output subdriver */ static int lcd_video_output_get(struct output_device *od) @@ -623,6 +831,15 @@ static int usb0_handler(int status) return status; } +static int ac_bat_handler(int status) +{ + if (ac_bat_initialized) { + power_supply_changed(&yeeloong_ac); + power_supply_changed(&yeeloong_bat); + } + return status; +} + static void do_event_action(int event) { sci_handler handler; @@ -668,6 +885,9 @@ static void do_event_action(int event) case EVENT_AUDIO_VOLUME: reg = REG_AUDIO_VOLUME; break; + case EVENT_AC_BAT: + handler = ac_bat_handler; + break; default: break; } @@ -926,6 +1146,13 @@ static int __init yeeloong_init(void) return ret; } + ret = yeeloong_bat_init(); + if (ret) { + pr_err("Fail to register yeeloong battery driver.\n"); + yeeloong_bat_exit(); + return ret; + } + ret = yeeloong_hwmon_init(); if (ret) { pr_err("Fail to register yeeloong hwmon driver.\n"); @@ -955,6 +1182,7 @@ static void __exit yeeloong_exit(void) yeeloong_hotkey_exit(); yeeloong_vo_exit(); yeeloong_hwmon_exit(); + yeeloong_bat_exit(); yeeloong_backlight_exit(); platform_driver_unregister(&platform_driver); -- 1.7.0.1 From wuzhangjin@gmail.com Thu Mar 11 04:25:14 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Thu, 11 Mar 2010 04:25:19 +0100 (CET) Received: from mail-ww0-f49.google.com ([74.125.82.49]:49251 "EHLO mail-ww0-f49.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491124Ab0CKDZO (ORCPT ); Thu, 11 Mar 2010 04:25:14 +0100 Received: by wwd20 with SMTP id 20so612417wwd.36 for ; Wed, 10 Mar 2010 19:25:08 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:from:to:cc:subject:date :message-id:x-mailer; bh=l9w/7tNjOl8FO82dYNVNKgQCVlqaPT5qHU9rT0kVhmc=; b=o/sF4aDY7HX+nOgDr6q0xuNPNSd721kJ04kK/5XaJEw2FUurJ969Nmt3PpAzNNDMBn GQ6ryInpCXNXd4uvfNj0wrlA/QRbSj14MM64+8Mw0t3At6JIgPGm9HRFSTD3qs1rtn63 KZKHjsfeB3o5RdBS4pqvl8yKBgeA0NtIwkCUg= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=from:to:cc:subject:date:message-id:x-mailer; b=qETP9F5hGz3Fh3/Q5Qv0ZuryRxyoUQ6cHSOF0D/HXkylVS16wGIVJVivMfe/auJ6jN eMDnhu28Pxf1YxKG7EKwzUfv+ix5a6eMnsB4c8feK3MaVRBW9YqZxCP8soGoRzwVSu33 E0dklRFf423xlUDPqQ257xWp/9FTFWUe5/lFo= Received: by 10.216.86.16 with SMTP id v16mr1578779wee.162.1268277906591; Wed, 10 Mar 2010 19:25:06 -0800 (PST) Received: from localhost.localdomain ([202.201.12.142]) by mx.google.com with ESMTPS id j8sm24374600gvb.16.2010.03.10.19.25.01 (version=TLSv1/SSLv3 cipher=RC4-MD5); Wed, 10 Mar 2010 19:25:05 -0800 (PST) From: Wu Zhangjin To: Ralf Baechle Cc: linux-mips@linux-mips.org, David Daney , Wu Zhangjin Subject: [-queue v9 3/3] MIPS: r4k: Add a high resolution sched_clock() Date: Thu, 11 Mar 2010 11:18:20 +0800 Message-Id: X-Mailer: git-send-email 1.7.0.1 Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26192 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: wuzhangjin@gmail.com Precedence: bulk X-list: linux-mips From: Wu Zhangjin (This patch is one of the "high resolution sched_clock()" patchset.) (v8 -> v9: O Make it depends on 64BIT for the current mips_sched_clock() only support 64bit. v7 -> v8: O Make it works with the exisiting clocksource_mips.mult, clocksource_mips.shift and copes with the 64bit calculation's overflow problem with the method introduced by David Daney in "MIPS: Octeon: Use non-overflowing arithmetic in sched_clock". To reduce the duplication, I have abstracted an inline mips_sched_clock() function to arch/mips/include/asm/time.h from arch/mips/cavium-octeon/csrc-octeon.c. v6 -> v7: O Make it depends on !CPU_FREQ and CPU_HAS_FIXED_C0_COUNT This sched_clock() is only available with the processor has fixed cp0 MIPS count register or even has dynamic cp0 MIPS count register but with CPU_FREQ disabled. NOTE: If your processor has fixed c0 count, please select CPU_HAS_FIXED_C0_COUNT for it and send a related patch to Ralf. v5 -> v6: o hard-codes the cycle2ns_scale_factor as 8 for 30(cs->shift) is too big. With 30, the return value of sched_clock() will also overflow quickly. o moves the sched_clock() back into csrc-r4k.c as David and Sergei recommended. o inits c0 count as zero for PRINTK_TIME=y. o drops the HR_SCHED_CLCOK option for the current sched_clock() is stable enough to replace the jiffies based one. ) This patch adds a cnt32_to_63() and MIPS c0 count based sched_clock(), which provides high resolution. Without it, the Ftrace for MIPS will give useless timestamp information. Because cnt32_to_63() needs to be called at least once per half period to work properly, Differ from the old version, this v2 revision set up a kernel timer to ensure the requirement of some MIPSs which have short c0 count period. And also, we init the c0 count as ZERO(just as jiffies does) in time_init() before plat_time_init(), without it, PRINTK_TIME=y will get wrong timestamp information. (NOTE: some platforms have initiazlied c0 count as zero, but some not, this may introduce some duplication, perhaps a new patch is needed to remove the initialized of c0 count in the platforms later?) This is originally from arch/arm/plat-orion/time.c This revision works well for function graph tracer now, and also, PRINTK_TIME=y will get normal timestamp informatin. Signed-off-by: Wu Zhangjin --- arch/mips/Kconfig | 13 ++++++++++++ arch/mips/kernel/csrc-r4k.c | 45 +++++++++++++++++++++++++++++++++++++++++++ arch/mips/kernel/time.c | 5 ++++ 3 files changed, 63 insertions(+), 0 deletions(-) diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig index 591ca0c..791c21f 100644 --- a/arch/mips/Kconfig +++ b/arch/mips/Kconfig @@ -1936,6 +1936,19 @@ config NR_CPUS source "kernel/time/Kconfig" # +# High Resolution sched_clock() support +# + +config CPU_HAS_FIXED_C0_COUNT + bool + +config CPU_SUPPORTS_HR_SCHED_CLOCK + bool + depends on CPU_HAS_FIXED_C0_COUNT || !CPU_FREQ + depends on 64BIT + default y + +# # Timer Interrupt Frequency Configuration # diff --git a/arch/mips/kernel/csrc-r4k.c b/arch/mips/kernel/csrc-r4k.c index e95a3cd..89473ba 100644 --- a/arch/mips/kernel/csrc-r4k.c +++ b/arch/mips/kernel/csrc-r4k.c @@ -6,7 +6,9 @@ * Copyright (C) 2007 by Ralf Baechle */ #include +#include #include +#include #include @@ -22,6 +24,47 @@ static struct clocksource clocksource_mips = { .flags = CLOCK_SOURCE_IS_CONTINUOUS, }; +#ifdef CONFIG_CPU_SUPPORTS_HR_SCHED_CLOCK +/* + * MIPS sched_clock implementation. + * + * Because the hardware timer period is quite short and because cnt32_to_63() + * needs to be called at least once per half period to work properly, a kernel + * timer is set up to ensure this requirement is always met. + * + * Please refer to include/linux/cnt32_to_63.h, arch/arm/plat-orion/time.c and + * arch/mips/include/asm/time.h (mips_sched_clock) + */ +unsigned long long notrace sched_clock(void) +{ + u64 cnt = cnt32_to_63(read_c0_count()); + + if (cnt & 0x8000000000000000) + cnt &= 0x7fffffffffffffff; + + return mips_sched_clock(&clocksource_mips, cnt); +} + +static struct timer_list cnt32_to_63_keepwarm_timer; + +static void cnt32_to_63_keepwarm(unsigned long data) +{ + mod_timer(&cnt32_to_63_keepwarm_timer, round_jiffies(jiffies + data)); + sched_clock(); +} +#endif + +static inline void setup_hres_sched_clock(unsigned long clock) +{ +#ifdef CONFIG_CPU_SUPPORTS_HR_SCHED_CLOCK + unsigned long data; + + data = 0x80000000UL / clock * HZ; + setup_timer(&cnt32_to_63_keepwarm_timer, cnt32_to_63_keepwarm, data); + mod_timer(&cnt32_to_63_keepwarm_timer, round_jiffies(jiffies + data)); +#endif +} + int __init init_r4k_clocksource(void) { if (!cpu_has_counter || !mips_hpt_frequency) @@ -32,6 +75,8 @@ int __init init_r4k_clocksource(void) clocksource_set_clock(&clocksource_mips, mips_hpt_frequency); + setup_hres_sched_clock(mips_hpt_frequency); + clocksource_register(&clocksource_mips); return 0; diff --git a/arch/mips/kernel/time.c b/arch/mips/kernel/time.c index fb74974..86cf18a 100644 --- a/arch/mips/kernel/time.c +++ b/arch/mips/kernel/time.c @@ -119,6 +119,11 @@ static __init int cpu_has_mfc0_count_bug(void) void __init time_init(void) { +#ifdef CONFIG_CPU_SUPPORTS_HR_SCHED_CLOCK + if (!mips_clockevent_init() || !cpu_has_mfc0_count_bug()) + write_c0_count(0); +#endif + plat_time_init(); if (!mips_clockevent_init() || !cpu_has_mfc0_count_bug()) -- 1.7.0.1 From shinya.kuribayashi@necel.com Thu Mar 11 04:27:45 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Thu, 11 Mar 2010 04:27:49 +0100 (CET) Received: from TYO201.gate.nec.co.jp ([202.32.8.193]:38805 "EHLO tyo201.gate.nec.co.jp" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491124Ab0CKD1p (ORCPT ); Thu, 11 Mar 2010 04:27:45 +0100 Received: from relay11.aps.necel.com ([10.29.19.46]) by tyo201.gate.nec.co.jp (8.13.8/8.13.4) with ESMTP id o2B3Reo7001313; Thu, 11 Mar 2010 12:27:40 +0900 (JST) Received: from realmbox31.aps.necel.com ([10.29.19.36] [10.29.19.36]) by relay11.aps.necel.com with ESMTP; Thu, 11 Mar 2010 12:27:40 +0900 Received: from [10.114.181.193] ([10.114.181.193] [10.114.181.193]) by mbox02.aps.necel.com with ESMTP; Thu, 11 Mar 2010 12:27:40 +0900 Message-ID: <4B98632E.70806@necel.com> Date: Thu, 11 Mar 2010 12:27:42 +0900 From: Shinya Kuribayashi User-Agent: Thunderbird 2.0.0.23 (Windows/20090812) MIME-Version: 1.0 To: Wu Zhangjin CC: Ralf Baechle , Greg KH , linux-mips@linux-mips.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH 1/3] Loongson-2F: Flush the branch target history such as BTB and RAS References: In-Reply-To: Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 8bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26193 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: shinya.kuribayashi@necel.com Precedence: bulk X-list: linux-mips Wu Zhangjin wrote: > From: Wu Zhangjin > > As the Chapter 15: "Errata: Issue of Out-of-order in loongson"[1] shows, to > workaround the Issue of Loongson-2F，We need to do: > > "When switching from user model to kernel model, you should flush the branch > target history such as BTB and RAS." Just wondered, model or mode? > This patch did clear BTB(branch target buffer), forbid RAS(row address strobe) > via Loongson-2F's 64bit diagnostic register. Are you sure that RAS represents "Row Address Strobe", not "Return Address Stack?" By the way, we have a similar local workaround for vr55xx processors when switching from kernel mode to user mode. It's not necessarily related to out-of-order issues, but we need to prevent the processor from doing instruction prefetch beyond "eret" instruction. In the long term, it would be appreciated that the kernel has a set of hooks when switching KUX-modes, so that each machine could have his own, processor-specific treatmens. Shinya > [1] Chinese Version: http://www.loongson.cn/uploadfile/file/200808211 > [2] English Version of Chapter 15: > http://groups.google.com.hk/group/loongson-dev/msg/e0d2e220958f10a6?dmode=source > > Signed-off-by: Wu Zhangjin > --- > arch/mips/include/asm/stackframe.h | 19 +++++++++++++++++++ > 1 files changed, 19 insertions(+), 0 deletions(-) > > diff --git a/arch/mips/include/asm/stackframe.h b/arch/mips/include/asm/stackframe.h > index 3b6da33..b84cfda 100644 > --- a/arch/mips/include/asm/stackframe.h > +++ b/arch/mips/include/asm/stackframe.h > @@ -121,6 +121,25 @@ > .endm > #else > .macro get_saved_sp /* Uniprocessor variation */ > +#ifdef CONFIG_CPU_LOONGSON2F > + /* > + * Clear BTB(branch target buffer), forbid RAS(row address > + * strobe) to workaround the Out-of-oder Issue in Loongson2F > + * via it's diagnostic register. > + */ > + move k0, ra > + jal 1f > + nop > +1: jal 1f > + nop > +1: jal 1f > + nop > +1: jal 1f > + nop > +1: move ra, k0 > + li k0, 3 > + mtc0 k0, $22 > +#endif /* CONFIG_CPU_LOONGSON2F */ > #if defined(CONFIG_32BIT) || defined(KBUILD_64BIT_SYM32) > lui k1, %hi(kernelsp) > #else From wuzhangjin@gmail.com Thu Mar 11 04:32:03 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Thu, 11 Mar 2010 04:32:06 +0100 (CET) Received: from mail-ww0-f49.google.com ([74.125.82.49]:57169 "EHLO mail-ww0-f49.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491124Ab0CKDcD (ORCPT ); Thu, 11 Mar 2010 04:32:03 +0100 Received: by wwd20 with SMTP id 20so614877wwd.36 for ; Wed, 10 Mar 2010 19:31:57 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:subject:from:reply-to:to:cc :in-reply-to:references:content-type:organization:date:message-id :mime-version:x-mailer:content-transfer-encoding; bh=MtZfKkdDa1tHbl1XPVKTr+Qj+Qz4oZXFnPa1e35nhPw=; b=MI+tFtiPHoqaT5h80750Bp7ey/Z3kI0JViwM06CpjR6LKbP3tik3rGK0cxVHEKPzbH oevtVHVkq1ZOkqoNUcbeNnFMO9jtBOKyQ/3lEfzTeuthb5Q+2X7SNbKkM8j1a3524AOc 6127JV+ZJhH+EaGAJXFoZKKolBfxIxGvhE0qc= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=subject:from:reply-to:to:cc:in-reply-to:references:content-type :organization:date:message-id:mime-version:x-mailer :content-transfer-encoding; b=oZZRQjs+oyGoL5iMfEG/HJwt5yuMPAeiGsm+pDeZFnkXtYk5BzI/LqtWD5iwDXwbIn Sz04EprycB+iGcnf6TrKfQ97D7UVBAr5cZfkZqDazcPPW3+x7bwur8I81CPMYOP5mjgJ ZRPUqGyXBiZhkPl14xs2NLcAy0UB/Z0L0k1yw= Received: by 10.216.88.5 with SMTP id z5mr202177wee.192.1268278317084; Wed, 10 Mar 2010 19:31:57 -0800 (PST) Received: from [202.201.12.142] ([202.201.12.142]) by mx.google.com with ESMTPS id p37sm24112100gvf.25.2010.03.10.19.31.53 (version=TLSv1/SSLv3 cipher=RC4-MD5); Wed, 10 Mar 2010 19:31:56 -0800 (PST) Subject: Re: [PATCH v2] Loongson: Lemote-2F: Fixup of _rdmsr and _wrmsr From: Wu Zhangjin Reply-To: wuzhangjin@gmail.com To: Ralf Baechle Cc: linux-mips@linux-mips.org, David Daney In-Reply-To: <5c63967ed3f891da1f6bc1fc78c272d0407d5d25.1268276186.git.wuzhangjin@gmail.com> References: <5c63967ed3f891da1f6bc1fc78c272d0407d5d25.1268276186.git.wuzhangjin@gmail.com> Content-Type: text/plain; charset="UTF-8" Organization: DSLab, Lanzhou University, China Date: Thu, 11 Mar 2010 11:25:26 +0800 Message-ID: <1268277926.17798.1.camel@falcon> Mime-Version: 1.0 X-Mailer: Evolution 2.28.2 Content-Transfer-Encoding: 7bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26194 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: wuzhangjin@gmail.com Precedence: bulk X-list: linux-mips On Thu, 2010-03-11 at 10:57 +0800, Wu Zhangjin wrote: > From: Wu Zhangjin > > Changes from the old version: > o Using raw spinlock instead of spinlock as suggested by David Daney. [...] > --- a/arch/mips/pci/ops-loongson2.c > +++ b/arch/mips/pci/ops-loongson2.c > @@ -180,15 +180,21 @@ struct pci_ops loongson_pci_ops = { > }; > > #ifdef CONFIG_CS5536 > +DEFINE_RAW_SPINLOCK(msr_lock); > + > void _rdmsr(u32 msr, u32 *hi, u32 *lo) > { > struct pci_bus bus = { > .number = PCI_BUS_CS5536 > }; > u32 devfn = PCI_DEVFN(PCI_IDSEL_CS5536, 0); > + unsigned long flags; > + > + spin_lock_irqsave(&msr_lock, flags); Please ignore this patch for the operations should be raw_ too. Regards, Wu Zhangjin From wuzhangjin@gmail.com Thu Mar 11 04:38:37 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Thu, 11 Mar 2010 04:38:44 +0100 (CET) Received: from mail-ew0-f224.google.com ([209.85.219.224]:63055 "EHLO mail-ew0-f224.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491124Ab0CKDih (ORCPT ); Thu, 11 Mar 2010 04:38:37 +0100 Received: by ewy24 with SMTP id 24so1849056ewy.27 for ; Wed, 10 Mar 2010 19:38:32 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:from:to:cc:subject:date :message-id:x-mailer; bh=qjc/R9AHcZY4Jjx5KeBN8cgL43HJmwXdEVWQpnn3xeo=; b=QduLaEs4KjF1HPbcwY0cwWjb6mTTj3KOgbt9dFZ2cOYzUHzj6UXPI3vvjOBzjf6X2b 48ehndpayajz12X5YKLGCu7f7UlvXKLLEVfyWkCDQKvWEO5e3VMYsMtgqkbAPB02aIp4 SMP5Jq+9IV2Ua6B0ZC2og0mser0s6iJoNitnQ= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=from:to:cc:subject:date:message-id:x-mailer; b=P07sxbN9TcE2mIGQf/ajWG/j5xc7BqosVSRX+Cmgu0lKIAJcPg9vs3ogNCQbU1l6Cj zv52oDKI03NjVL40h/Hdmnvadbh20XFusv5MYh8EXNUHTxNCpxRRRnGUlcpqUnDWgbEB j9+nfj1sG3QwSSmnummD4Rphf9q6t1DmLTQJo= Received: by 10.213.8.26 with SMTP id f26mr5812818ebf.14.1268278712287; Wed, 10 Mar 2010 19:38:32 -0800 (PST) Received: from localhost.localdomain ([202.201.12.142]) by mx.google.com with ESMTPS id 14sm4495557ewy.6.2010.03.10.19.38.27 (version=TLSv1/SSLv3 cipher=RC4-MD5); Wed, 10 Mar 2010 19:38:31 -0800 (PST) From: Wu Zhangjin To: Ralf Baechle Cc: linux-mips@linux-mips.org, David Daney , Wu Zhangjin Subject: [PATCH v3] Loongson: Lemote-2F: Fixup of _rdmsr and _wrmsr Date: Thu, 11 Mar 2010 11:30:50 +0800 Message-Id: X-Mailer: git-send-email 1.7.0.1 Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26195 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: wuzhangjin@gmail.com Precedence: bulk X-list: linux-mips From: Wu Zhangjin v2 --> v3: o Using the right operations for the raw spinlock. v1 --> v2: o Using raw spinlock instead of spinlock as suggested by David Daney. The _rdmsr, _wrmsr operation must be atomic to ensure the accessing to msr address is what we want. Without this patch, in some cases, the reboot operation(fs2f_reboot) defined in arch/mips/loongson/lemote-2f/reset.c may fail for it called _rdmsr, _wrmsr but may be interrupted/preempted by the other related operations and make the _rdmsr get the wrong value or make the _wrmsr write a wrong value to an unexpected target. Signed-off-by: Wu Zhangjin --- arch/mips/pci/ops-loongson2.c | 10 ++++++++++ 1 files changed, 10 insertions(+), 0 deletions(-) diff --git a/arch/mips/pci/ops-loongson2.c b/arch/mips/pci/ops-loongson2.c index 2bb4057..d657ee0 100644 --- a/arch/mips/pci/ops-loongson2.c +++ b/arch/mips/pci/ops-loongson2.c @@ -180,15 +180,21 @@ struct pci_ops loongson_pci_ops = { }; #ifdef CONFIG_CS5536 +DEFINE_RAW_SPINLOCK(msr_lock); + void _rdmsr(u32 msr, u32 *hi, u32 *lo) { struct pci_bus bus = { .number = PCI_BUS_CS5536 }; u32 devfn = PCI_DEVFN(PCI_IDSEL_CS5536, 0); + unsigned long flags; + + raw_spin_lock_irqsave(&msr_lock, flags); loongson_pcibios_write(&bus, devfn, PCI_MSR_ADDR, 4, msr); loongson_pcibios_read(&bus, devfn, PCI_MSR_DATA_LO, 4, lo); loongson_pcibios_read(&bus, devfn, PCI_MSR_DATA_HI, 4, hi); + raw_spin_unlock_irqrestore(&msr_lock, flags); } EXPORT_SYMBOL(_rdmsr); @@ -198,9 +204,13 @@ void _wrmsr(u32 msr, u32 hi, u32 lo) .number = PCI_BUS_CS5536 }; u32 devfn = PCI_DEVFN(PCI_IDSEL_CS5536, 0); + unsigned long flags; + + raw_spin_lock_irqsave(&msr_lock, flags); loongson_pcibios_write(&bus, devfn, PCI_MSR_ADDR, 4, msr); loongson_pcibios_write(&bus, devfn, PCI_MSR_DATA_LO, 4, lo); loongson_pcibios_write(&bus, devfn, PCI_MSR_DATA_HI, 4, hi); + raw_spin_unlock_irqrestore(&msr_lock, flags); } EXPORT_SYMBOL(_wrmsr); #endif -- 1.7.0.1 From wuzhangjin@gmail.com Thu Mar 11 04:42:00 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Thu, 11 Mar 2010 04:42:04 +0100 (CET) Received: from mail-ew0-f224.google.com ([209.85.219.224]:34393 "EHLO mail-ew0-f224.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492047Ab0CKDmA (ORCPT ); Thu, 11 Mar 2010 04:42:00 +0100 Received: by ewy24 with SMTP id 24so1849553ewy.27 for ; Wed, 10 Mar 2010 19:41:54 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:subject:from:reply-to:to:cc :in-reply-to:references:content-type:organization:date:message-id :mime-version:x-mailer:content-transfer-encoding; bh=/QwW/Sik4jBsqpPDDxfJPhsLwABrgYfVu9jhw/o9j6Q=; b=WHd6v6xNZYzPdCgbYETsoQMghJMpmaNyiBx8PQ3cNy4IZFi2NnCwJdP5qX34C/VDNP jxLjY93WobhwjhVhAYJO09IdHeaH5Stz2w9udUcPFQ/dJjAdUhR/CI/c712fV7W5AwOk 5t7+rw6ER4L7U0ThzZlLq0y34tKSlKbRfkm24= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=subject:from:reply-to:to:cc:in-reply-to:references:content-type :organization:date:message-id:mime-version:x-mailer :content-transfer-encoding; b=qg1y9Qo1ZE+qSEb57dOD2ibzqxCAqqbZej6XAzLCatyrY+LjcBDciAPlWf+zkJVNSn 6wHYXXpiD14t2VNJM4zTLGBEw1agBgXz+BZHPCIRAlXIDHPQI9yranhsf8ItGBy82Ts6 fe3Z4dl4wCLa4hDPIg+nrvgs+Hg6g9KAmmtmM= Received: by 10.213.100.165 with SMTP id y37mr5754300ebn.71.1268278914418; Wed, 10 Mar 2010 19:41:54 -0800 (PST) Received: from [202.201.12.142] ([202.201.12.142]) by mx.google.com with ESMTPS id 16sm4430477ewy.7.2010.03.10.19.41.47 (version=TLSv1/SSLv3 cipher=RC4-MD5); Wed, 10 Mar 2010 19:41:52 -0800 (PST) Subject: Re: [PATCH 1/3] Loongson-2F: Flush the branch target history such as BTB and RAS From: Wu Zhangjin Reply-To: wuzhangjin@gmail.com To: Shinya Kuribayashi Cc: Ralf Baechle , Greg KH , linux-mips@linux-mips.org, linux-kernel@vger.kernel.org, zhangfx , yanh In-Reply-To: <4B98632E.70806@necel.com> References: <4B98632E.70806@necel.com> Content-Type: text/plain; charset="UTF-8" Organization: DSLab, Lanzhou University, China Date: Thu, 11 Mar 2010 11:35:20 +0800 Message-ID: <1268278520.17798.5.camel@falcon> Mime-Version: 1.0 X-Mailer: Evolution 2.28.2 Content-Transfer-Encoding: 8bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26196 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: wuzhangjin@gmail.com Precedence: bulk X-list: linux-mips On Thu, 2010-03-11 at 12:27 +0900, Shinya Kuribayashi wrote: > Wu Zhangjin wrote: > > From: Wu Zhangjin > > > > As the Chapter 15: "Errata: Issue of Out-of-order in loongson"[1] shows, to > > workaround the Issue of Loongson-2F，We need to do: > > > > "When switching from user model to kernel model, you should flush the branch > > target history such as BTB and RAS." > > Just wondered, model or mode? > Hmm, should be mode. > > This patch did clear BTB(branch target buffer), forbid RAS(row address strobe) > > via Loongson-2F's 64bit diagnostic register. > > Are you sure that RAS represents "Row Address Strobe", not "Return > Address Stack?" > Hi, Yanhua(from Lemote), can you help to clear this part? > By the way, we have a similar local workaround for vr55xx processors > when switching from kernel mode to user mode. It's not necessarily > related to out-of-order issues, but we need to prevent the processor > from doing instruction prefetch beyond "eret" instruction. > > In the long term, it would be appreciated that the kernel has a set > of hooks when switching KUX-modes, so that each machine could have > his own, processor-specific treatmens. > Good idea. Thanks! Regards, Wu Zhangjin From yang.shi@windriver.com Thu Mar 11 06:42:36 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Thu, 11 Mar 2010 06:42:40 +0100 (CET) Received: from mail.windriver.com ([147.11.1.11]:50512 "EHLO mail.windriver.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491775Ab0CKFmg (ORCPT ); Thu, 11 Mar 2010 06:42:36 +0100 Received: from localhost.localdomain (pek-lpgbuild1.wrs.com [128.224.153.29]) by mail.windriver.com (8.14.3/8.14.3) with ESMTP id o2B5gO9A011081; Wed, 10 Mar 2010 21:42:28 -0800 (PST) From: Yang Shi To: ralf@linux-mips.org Cc: linux-mips@linux-mips.org Subject: [PATCH] MIPS: Fix elfcore.c build warning Date: Thu, 11 Mar 2010 13:42:22 +0800 Message-Id: <1268286142-20615-1-git-send-email-yang.shi@windriver.com> X-Mailer: git-send-email 1.6.0.4 Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26197 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: yang.shi@windriver.com Precedence: bulk X-list: linux-mips kernel/elfcore.c includes elf.h which includes arch specific elf.h. In MIPS elf.h, 'struct pt_regs' is declared inside the parameter list of elf_dump_regs function. This cause kernel build warning. So, move the declaration out of the function parameter list to remove build warning. Signed-off-by: Yang Shi --- arch/mips/include/asm/elf.h | 3 ++- 1 files changed, 2 insertions(+), 1 deletions(-) diff --git a/arch/mips/include/asm/elf.h b/arch/mips/include/asm/elf.h index e53d7be..370bf50 100644 --- a/arch/mips/include/asm/elf.h +++ b/arch/mips/include/asm/elf.h @@ -311,8 +311,9 @@ do { \ #endif /* CONFIG_64BIT */ struct task_struct; +struct pt_regs; -extern void elf_dump_regs(elf_greg_t *, struct pt_regs *regs); +extern void elf_dump_regs(elf_greg_t *, struct pt_regs *); extern int dump_task_regs(struct task_struct *, elf_gregset_t *); extern int dump_task_fpu(struct task_struct *, elf_fpregset_t *); -- 1.6.3.3 From ralf@linux-mips.org Thu Mar 11 08:51:58 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Thu, 11 Mar 2010 08:52:02 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:60032 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1491885Ab0CKHv6 (ORCPT ); Thu, 11 Mar 2010 08:51:58 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o2B7ptOr009973; Thu, 11 Mar 2010 08:51:57 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o2B7prCO009971; Thu, 11 Mar 2010 08:51:53 +0100 Date: Thu, 11 Mar 2010 08:51:52 +0100 From: Ralf Baechle To: Yang Shi Cc: linux-mips@linux-mips.org Subject: Re: [PATCH] MIPS: Fix elfcore.c build warning Message-ID: <20100311075151.GB18065@linux-mips.org> References: <1268286142-20615-1-git-send-email-yang.shi@windriver.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1268286142-20615-1-git-send-email-yang.shi@windriver.com> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26198 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Thu, Mar 11, 2010 at 01:42:22PM +0800, Yang Shi wrote: > kernel/elfcore.c includes elf.h which includes arch specific elf.h. > In MIPS elf.h, 'struct pt_regs' is declared inside the parameter > list of elf_dump_regs function. This cause kernel build warning. > > So, move the declaration out of the function parameter list to > remove build warning. What fixes the warning is the forward declaration of struct pt_regs not removing the function parameter name. So I've added a patch which does just that. Thanks! Ralf From ralf@linux-mips.org Thu Mar 11 10:59:51 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Thu, 11 Mar 2010 10:59:55 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:54155 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492166Ab0CKJ7v (ORCPT ); Thu, 11 Mar 2010 10:59:51 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o2B9xmOm015325; Thu, 11 Mar 2010 10:59:49 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o2B9xjwB015316; Thu, 11 Mar 2010 10:59:45 +0100 Date: Thu, 11 Mar 2010 10:59:44 +0100 From: Ralf Baechle To: Shinya Kuribayashi Cc: Wu Zhangjin , Greg KH , linux-mips@linux-mips.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH 1/3] Loongson-2F: Flush the branch target history such as BTB and RAS Message-ID: <20100311095944.GC18065@linux-mips.org> References: <4B98632E.70806@necel.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <4B98632E.70806@necel.com> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26199 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Thu, Mar 11, 2010 at 12:27:42PM +0900, Shinya Kuribayashi wrote: > Are you sure that RAS represents "Row Address Strobe", not "Return > Address Stack?" > > By the way, we have a similar local workaround for vr55xx processors > when switching from kernel mode to user mode. It's not necessarily > related to out-of-order issues, but we need to prevent the processor > from doing instruction prefetch beyond "eret" instruction. Some R4000 revisions may do silly things in case of an NMI where c0_errorepc is pointing is pointing to an ERET instruction. There are various processors which want to save and restore core-specific registers, for example the Cavium cnMIPS core. > In the long term, it would be appreciated that the kernel has a set > of hooks when switching KUX-modes, so that each machine could have > his own, processor-specific treatmens. It seems that uasm is the tool of choice. Ralf From ralf@linux-mips.org Thu Mar 11 11:06:51 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Thu, 11 Mar 2010 11:06:54 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:36909 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492047Ab0CKKGv (ORCPT ); Thu, 11 Mar 2010 11:06:51 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o2BA6n1D015743; Thu, 11 Mar 2010 11:06:49 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o2BA6mU5015741; Thu, 11 Mar 2010 11:06:48 +0100 Date: Thu, 11 Mar 2010 11:06:48 +0100 From: Ralf Baechle To: "Robert P. J. Day" Cc: linux-mips@linux-mips.org Subject: Re: [PATCH] MIPS: Initialize an atomic_t properly with ATOMIC_INIT(0). Message-ID: <20100311100647.GD18065@linux-mips.org> References: MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26200 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Sat, Feb 27, 2010 at 12:02:51PM -0500, Robert P. J. Day wrote: > AFAIK, the technically correct way to initialize atomic variables is > with ATOMIC_INIT(n). Indeed, applied. Thanks! Ralf From mikael.starvik@axis.com Thu Mar 11 16:01:36 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Thu, 11 Mar 2010 16:01:41 +0100 (CET) Received: from krynn.se.axis.com ([193.13.178.10]:45131 "EHLO krynn.se.axis.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492521Ab0CKPBg convert rfc822-to-8bit (ORCPT ); Thu, 11 Mar 2010 16:01:36 +0100 Received: from xmail3.se.axis.com (xmail3.se.axis.com [10.0.5.75]) by krynn.se.axis.com (8.14.3/8.14.3/Debian-5) with ESMTP id o2BExqdb019855 for ; Thu, 11 Mar 2010 16:00:18 +0100 Received: from xmail3.se.axis.com ([10.0.5.75]) by xmail3.se.axis.com ([10.0.5.75]) with mapi; Thu, 11 Mar 2010 15:59:48 +0100 From: Mikael Starvik To: "linux-mips@linux-mips.org" Date: Thu, 11 Mar 2010 15:59:44 +0100 Subject: MIPS raw_local_irq_restore flags Thread-Topic: MIPS raw_local_irq_restore flags Thread-Index: AcrBK3wkk2FBNl+9TaOxNHDqJxz80A== Message-ID: <4BEA3FF3CAA35E408EA55C7BE2E61D0546AC862322@xmail3.se.axis.com> Accept-Language: sv-SE Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-cr-puzzleid: {BD95551F-236F-4DB5-9BD6-5AABD159E72A} x-cr-hashedpuzzle: gyU= Bby3 Bt5Z CS4u ChsQ Co9K D2/n D8Lp EFIC EHOd Eii6 FoZ0 HbUU Hpa+ JG0X JPZy;1;bABpAG4AdQB4AC0AbQBpAHAAcwBAAGwAaQBuAHUAeAAtAG0AaQBwAHMALgBvAHIAZwA=;Sosha1_v1;7;{BD95551F-236F-4DB5-9BD6-5AABD159E72A};bQBpAGsAYQBlAGwALgBzAHQAYQByAHYAaQBrAEAAYQB4AGkAcwAuAGMAbwBtAA==;Thu, 11 Mar 2010 14:59:44 GMT;TQBJAFAAUwAgAHIAYQB3AF8AbABvAGMAYQBsAF8AaQByAHEAXwByAGUAcwB0AG8AcgBlACAAZgBsAGEAZwBzAA== acceptlanguage: sv-SE Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 8BIT MIME-Version: 1.0 Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26201 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: mikael.starvik@axis.com Precedence: bulk X-list: linux-mips For the common case CONFIG_CPU_MIPSR2 && CONFIG_IRQ_CPU raw_local_irq_restore_flags is defined as: " beqz \\flags, 1f \n" " di \n" " ei \n" "1: \n" Doesn't this imply that you can't do recursive local_irq_save() (with different locks ofcourse)? Best Regards /Mikael From miloody@gmail.com Thu Mar 11 18:02:19 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Thu, 11 Mar 2010 18:02:22 +0100 (CET) Received: from mail-pz0-f185.google.com ([209.85.222.185]:45827 "EHLO mail-pz0-f185.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492300Ab0CKRCT (ORCPT ); Thu, 11 Mar 2010 18:02:19 +0100 Received: by pzk15 with SMTP id 15so137595pzk.21 for ; Thu, 11 Mar 2010 09:02:13 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:date:message-id:subject :from:to:content-type; bh=VmK3+9byw9jVNm3Bj3AiujA26wQCYtrD69OovXsovbo=; b=jn7F8Fp8McIjq+en3nWi0tEYJi88vimI9qx6sOzhoFRXWRMY01VVofZ8TcGThADLWn EdVZ6hkBSsT+iC37FImce5MhMKrJZW78EuZkXfXcTsYyMhudiMmQR1IlTyGZN5yfzuTU GLTGX+erKQc2/iyFneV0zT7kugfvHL1pWgQ48= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:date:message-id:subject:from:to:content-type; b=ugXT09E8Jz6571robkYEnEkjSBn74FQg9AGfDEAyd6qezK3jGL/BBle4E6lyRwfTsG BHmpPn8EpG8kRQ4NqJcWILytNeq75FQg98HDjp525yOYPjRiddBK0ai0vHKbrLtK/+6X 2UvHYURiafptXH3m3f8KUA4/FBCpxVFgmkTR8= MIME-Version: 1.0 Received: by 10.142.75.21 with SMTP id x21mr1464873wfa.212.1268326933081; Thu, 11 Mar 2010 09:02:13 -0800 (PST) Date: Fri, 12 Mar 2010 01:02:13 +0800 Message-ID: <3a665c761003110902o1da6dea9ib2723066b68f5f54@mail.gmail.com> Subject: some question about memory usage in mips From: loody To: "linux-mips@linux-mips.org" Content-Type: text/plain; charset=ISO-8859-1 Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26202 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: miloody@gmail.com Precedence: bulk X-list: linux-mips Dear all: I test usb driver under mips machine and I have some questions always bother me. 1. what is the difference between dma_pool_create and dma_alloc_coherent and where they are defined? from the result it return, dma_pool_create will return non-cache area, but dma_alloc_coherent returns catched area. why they get the dma in it? Does that have special meaning? 2. what are wmb() and rmb(). used for? do they use to flush cache? and what these 3 leterr mean? wmb, write memory back rmb, read memory back appreciate your help, miloody From David.Daney@caviumnetworks.com Thu Mar 11 18:26:24 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Thu, 11 Mar 2010 18:26:28 +0100 (CET) Received: from mail3.caviumnetworks.com ([12.108.191.235]:14933 "EHLO mail3.caviumnetworks.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492313Ab0CKR0Y (ORCPT ); Thu, 11 Mar 2010 18:26:24 +0100 Received: from caexch01.caveonetworks.com (Not Verified[192.168.16.9]) by mail3.caviumnetworks.com with MailMarshal (v6,7,2,8378) id ; Thu, 11 Mar 2010 09:26:31 -0800 Received: from caexch01.caveonetworks.com ([192.168.16.9]) by caexch01.caveonetworks.com with Microsoft SMTPSVC(6.0.3790.3959); Thu, 11 Mar 2010 09:25:49 -0800 Received: from dd1.caveonetworks.com ([12.108.191.236]) by caexch01.caveonetworks.com over TLS secured channel with Microsoft SMTPSVC(6.0.3790.3959); Thu, 11 Mar 2010 09:25:49 -0800 Message-ID: <4B992797.4070006@caviumnetworks.com> Date: Thu, 11 Mar 2010 09:25:43 -0800 From: David Daney User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.8) Gecko/20100301 Fedora/3.0.3-1.fc12 Thunderbird/3.0.3 MIME-Version: 1.0 To: Mikael Starvik CC: "linux-mips@linux-mips.org" Subject: Re: MIPS raw_local_irq_restore flags References: <4BEA3FF3CAA35E408EA55C7BE2E61D0546AC862322@xmail3.se.axis.com> In-Reply-To: <4BEA3FF3CAA35E408EA55C7BE2E61D0546AC862322@xmail3.se.axis.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-OriginalArrivalTime: 11 Mar 2010 17:25:49.0303 (UTC) FILETIME=[E46E8870:01CAC13F] Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26203 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ddaney@caviumnetworks.com Precedence: bulk X-list: linux-mips On 03/11/2010 06:59 AM, Mikael Starvik wrote: > For the common case CONFIG_CPU_MIPSR2&& CONFIG_IRQ_CPU raw_local_irq_restore_flags is defined as: > > " beqz \\flags, 1f \n" > " di \n" > " ei \n" > "1: \n" > > Doesn't this imply that you can't do recursive local_irq_save() (with different locks ofcourse)? > No. In fact local_irq_save() is intentionally designed to be used 'recursively'. David Daney From wuzhangjin@gmail.com Thu Mar 11 19:14:28 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Thu, 11 Mar 2010 19:14:34 +0100 (CET) Received: from mail-fx0-f225.google.com ([209.85.220.225]:52427 "EHLO mail-fx0-f225.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491808Ab0CKSO2 (ORCPT ); Thu, 11 Mar 2010 19:14:28 +0100 Received: by fxm25 with SMTP id 25so354721fxm.27 for ; Thu, 11 Mar 2010 10:14:21 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:from:to:cc:subject:date :message-id:x-mailer; bh=I5xjgi3IcUPYos9bMFdXsgiMIz+GDRcVkjj30DV/y5c=; b=NGUOg7jVXuO8BvEiWcYLlomGz9AvPbjnf5RjrJUy00m6SXtACMW76k+QFXu0xBOG+G tLdxqYTtrJy68DTsVYiZdykfHMecklFOLcEMimNvv/pOTmq31wqTWco+p6LVLw2b/9o7 6W+2sjOlk7ag4UFx3hWe+Qcx0TkvjQd6inwnc= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=from:to:cc:subject:date:message-id:x-mailer; b=a5tq4ksLjQfsBUIthDn3Yb9zx1AUfy7drgQ10WTiWl4PTKtow4FoawMI6+4srmMzKS RXkIxfNkyBxgHxXZ+dlBIHUcp7MQT/s+YpyT37cgALy5SWgoCeqxoAEPzTHdQ/YwwQmn jKJ6J8KOUOcPWf7QdlGJCrnHoPzVEocn22dPY= Received: by 10.223.60.142 with SMTP id p14mr596730fah.47.1268331261722; Thu, 11 Mar 2010 10:14:21 -0800 (PST) Received: from localhost.localdomain ([202.201.12.142]) by mx.google.com with ESMTPS id 13sm555734fks.0.2010.03.11.10.14.17 (version=TLSv1/SSLv3 cipher=RC4-MD5); Thu, 11 Mar 2010 10:14:20 -0800 (PST) From: Wu Zhangjin To: Ralf Baechle Cc: Steven Rostedt , linux-mips@linux-mips.org, David Daney , Wu Zhangjin Subject: [PATCH] MIPS: tracing: Optimize the implementation Date: Fri, 12 Mar 2010 02:07:37 +0800 Message-Id: <8b93c417fefa4d446f801abfd718ba94fdcb1821.1268330348.git.wuzhangjin@gmail.com> X-Mailer: git-send-email 1.7.0.1 Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26204 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: wuzhangjin@gmail.com Precedence: bulk X-list: linux-mips From: Wu Zhangjin (All of the stuff have been tested for kernel and module, including static function tracer, dynamic function tracer and function graph tracer.) Changes: + Reduce the runtime overhead o Uses macros instead of variables for the fixed instructions to reduce memory access o Moves the Initilization of the instructions which will be fixed after linking from ftrace_make_nop/ftrace_make_call to ftrace_dyn_arch_init() and encodes the instructions through uasm(arch/mips/include/asm/uasm.h). o A common macro in_module() is defined to determine which space the instruction pointer stays in and several related conditional statements are converted to conditional operator(? :) statement. + Cleanup the whole stuff Lots of comments/macros have been cleaned up to let it look better. Signed-off-by: Wu Zhangjin --- arch/mips/include/asm/ftrace.h | 10 ++- arch/mips/kernel/ftrace.c | 201 ++++++++++++++++++++++------------------ 2 files changed, 119 insertions(+), 92 deletions(-) diff --git a/arch/mips/include/asm/ftrace.h b/arch/mips/include/asm/ftrace.h index ce35c9a..e58f9ff 100644 --- a/arch/mips/include/asm/ftrace.h +++ b/arch/mips/include/asm/ftrace.h @@ -3,7 +3,7 @@ * License. See the file "COPYING" in the main directory of this archive for * more details. * - * Copyright (C) 2009 DSLab, Lanzhou University, China + * Copyright (C) 2009, 2010 DSLab, Lanzhou University, China * Author: Wu Zhangjin */ @@ -19,6 +19,12 @@ extern void _mcount(void); #define mcount _mcount +/* + * If the Instruction Pointer is in module space (0xc0000000), return ture; + * otherwise, it is in kernel space (0x80000000), return false. + */ +#define in_module(ip) (unlikely((ip) & 0x40000000)) + #define safe_load(load, src, dst, error) \ do { \ asm volatile ( \ @@ -83,8 +89,8 @@ static inline unsigned long ftrace_call_adjust(unsigned long addr) struct dyn_arch_ftrace { }; +#endif /* CONFIG_DYNAMIC_FTRACE */ -#endif /* CONFIG_DYNAMIC_FTRACE */ #endif /* __ASSEMBLY__ */ #endif /* CONFIG_FUNCTION_TRACER */ #endif /* _ASM_MIPS_FTRACE_H */ diff --git a/arch/mips/kernel/ftrace.c b/arch/mips/kernel/ftrace.c index e9e64e0..c38e3d9 100644 --- a/arch/mips/kernel/ftrace.c +++ b/arch/mips/kernel/ftrace.c @@ -2,7 +2,7 @@ * Code for replacing ftrace calls with jumps. * * Copyright (C) 2007-2008 Steven Rostedt - * Copyright (C) 2009 DSLab, Lanzhou University, China + * Copyright (C) 2009, 2010 DSLab, Lanzhou University, China * Author: Wu Zhangjin * * Thanks goes to Steven Rostedt for writing the original x86 version. @@ -15,15 +15,50 @@ #include #include #include +#include #ifdef CONFIG_DYNAMIC_FTRACE -#define JAL 0x0c000000 /* jump & link: ip --> ra, jump to target */ -#define ADDR_MASK 0x03ffffff /* op_code|addr : 31...26|25 ....0 */ -#define jump_insn_encode(op_code, addr) \ - ((unsigned int)((op_code) | (((addr) >> 2) & ADDR_MASK))) +/* Before linking, the following instructions are fixed. */ +#ifdef CONFIG_CPU_LOONGSON2F +/* insn: or at, at, zero */ +#define INSN_NOP 0x00200825 +#else +/* insn: nop */ +#define INSN_NOP 0x00000000 +#endif + +/* insn: b 1f; offset = (16 >> 2) */ +#define INSN_B_1F 0x10000004 + +/* After linking, the following instructions are fixed. */ +static unsigned int insn_jal_ftrace_caller __read_mostly; +static unsigned int insn_lui_v1_hi16_mcount __read_mostly; +static unsigned int insn_j_ftrace_graph_caller __read_mostly; -static unsigned int ftrace_nop = 0x00000000; +/* The following instructions are encoded in the run-time */ +/* insn: jal func; op_code|addr : 31...26|25 ....0 */ +#define jal(addr) \ + ((unsigned int)(0x0c000000 | (((addr) >> 2) & 0x03ffffff))) + +static inline void ftrace_dyn_arch_init_insns(void) +{ + u32 *buf; + unsigned int v1; + + /* lui v1, hi16_mcount */ + v1 = 3; + buf = (u32 *)&insn_lui_v1_hi16_mcount; + UASM_i_LA_mostly(&buf, v1, MCOUNT_ADDR); + + /* jal (ftrace_caller + 8), jump over the first two instruction */ + buf = (u32 *)&insn_jal_ftrace_caller; + uasm_i_jal(&buf, (FTRACE_ADDR + 8)); + + /* j ftrace_graph_caller */ + buf = (u32 *)&insn_j_ftrace_graph_caller; + uasm_i_j(&buf, (unsigned long)ftrace_graph_caller); +} static int ftrace_modify_code(unsigned long ip, unsigned int new_code) { @@ -31,7 +66,6 @@ static int ftrace_modify_code(unsigned long ip, unsigned int new_code) /* *(unsigned int *)ip = new_code; */ safe_store_code(new_code, ip, faulted); - if (unlikely(faulted)) return -EFAULT; @@ -40,117 +74,101 @@ static int ftrace_modify_code(unsigned long ip, unsigned int new_code) return 0; } -static int lui_v1; -static int jal_mcount; - int ftrace_make_nop(struct module *mod, struct dyn_ftrace *rec, unsigned long addr) { unsigned int new; - int faulted; unsigned long ip = rec->ip; - /* We have compiled module with -mlong-calls, but compiled the kernel - * without it, we need to cope with them respectively. */ - if (ip & 0x40000000) { - /* record it for ftrace_make_call */ - if (lui_v1 == 0) { - /* lui_v1 = *(unsigned int *)ip; */ - safe_load_code(lui_v1, ip, faulted); - - if (unlikely(faulted)) - return -EFAULT; - } - - /* lui v1, hi_16bit_of_mcount --> b 1f (0x10000004) - * addiu v1, v1, low_16bit_of_mcount - * move at, ra - * jalr v1 - * nop - * 1f: (ip + 12) - */ - new = 0x10000004; - } else { - /* record/calculate it for ftrace_make_call */ - if (jal_mcount == 0) { - /* We can record it directly like this: - * jal_mcount = *(unsigned int *)ip; - * Herein, jump over the first two nop instructions */ - jal_mcount = jump_insn_encode(JAL, (MCOUNT_ADDR + 8)); - } - - /* move at, ra - * jalr v1 --> nop - */ - new = ftrace_nop; - } + /* + * We have compiled modules with -mlong-calls, but compiled kernel + * without it, therefore, need to cope with them respectively. + * + * For module: + * + * lui v1, hi16_mcount --> b 1f + * addiu v1, v1, lo16_mcount + * move at, ra + * jalr v1 + * nop + * 1f: (ip + 16) + * For kernel: + * + * move at, ra + * jal _mcount --> nop + * + */ + new = in_module(ip) ? INSN_B_1F : INSN_NOP; + return ftrace_modify_code(ip, new); } -static int modified; /* initialized as 0 by default */ - int ftrace_make_call(struct dyn_ftrace *rec, unsigned long addr) { unsigned int new; unsigned long ip = rec->ip; - /* We just need to remove the "b ftrace_stub" at the fist time! */ - if (modified == 0) { - modified = 1; - ftrace_modify_code(addr, ftrace_nop); - } - /* ip, module: 0xc0000000, kernel: 0x80000000 */ - new = (ip & 0x40000000) ? lui_v1 : jal_mcount; + new = in_module(ip) ? insn_lui_v1_hi16_mcount : insn_jal_ftrace_caller; return ftrace_modify_code(ip, new); } -#define FTRACE_CALL_IP ((unsigned long)(&ftrace_call)) - int ftrace_update_ftrace_func(ftrace_func_t func) { unsigned int new; - new = jump_insn_encode(JAL, (unsigned long)func); + new = jal((unsigned long)func); - return ftrace_modify_code(FTRACE_CALL_IP, new); + return ftrace_modify_code((unsigned long)(&ftrace_call), new); } int __init ftrace_dyn_arch_init(void *data) { + ftrace_dyn_arch_init_insns(); + + /* + * We are safe to remove the "b ftrace_stub" for the current + * ftrace_caller() is almost empty (only the stack operations), and + * most importantly, the calling to mcount will be disabled later in + * ftrace_init(), then there is no 'big' overhead. And in the future, + * we are hoping the function_trace_stop is initialized as 1 then we + * can eventually remove that 'useless' "b ftrace_stub" and the + * following nop. Currently, although we can call ftrace_stop() to set + * function_trace_stop as 1, but it will change the behavior of the + * Function Tracer. + */ + ftrace_modify_code(MCOUNT_ADDR, INSN_NOP); + /* The return code is retured via data */ *(unsigned long *)data = 0; return 0; } -#endif /* CONFIG_DYNAMIC_FTRACE */ +#endif /* CONFIG_DYNAMIC_FTRACE */ #ifdef CONFIG_FUNCTION_GRAPH_TRACER #ifdef CONFIG_DYNAMIC_FTRACE extern void ftrace_graph_call(void); -#define JMP 0x08000000 /* jump to target directly */ -#define CALL_FTRACE_GRAPH_CALLER \ - jump_insn_encode(JMP, (unsigned long)(&ftrace_graph_caller)) #define FTRACE_GRAPH_CALL_IP ((unsigned long)(&ftrace_graph_call)) int ftrace_enable_ftrace_graph_caller(void) { return ftrace_modify_code(FTRACE_GRAPH_CALL_IP, - CALL_FTRACE_GRAPH_CALLER); + insn_j_ftrace_graph_caller); } int ftrace_disable_ftrace_graph_caller(void) { - return ftrace_modify_code(FTRACE_GRAPH_CALL_IP, ftrace_nop); + return ftrace_modify_code(FTRACE_GRAPH_CALL_IP, INSN_NOP); } -#endif /* !CONFIG_DYNAMIC_FTRACE */ +#endif /* !CONFIG_DYNAMIC_FTRACE */ #ifndef KBUILD_MCOUNT_RA_ADDRESS -#define S_RA_SP (0xafbf << 16) /* s{d,w} ra, offset(sp) */ -#define S_R_SP (0xafb0 << 16) /* s{d,w} R, offset(sp) */ +#define S_RA_SP 0xafbf0000 /* s{d,w} ra, offset(sp) */ +#define S_R_SP 0xafb00000 /* s{d,w} R, offset(sp) */ #define OFFSET_MASK 0xffff /* stack offset range: 0 ~ PT_SIZE */ unsigned long ftrace_get_parent_addr(unsigned long self_addr, @@ -162,32 +180,32 @@ unsigned long ftrace_get_parent_addr(unsigned long self_addr, unsigned int code; int faulted; - /* in module or kernel? */ - if (self_addr & 0x40000000) { - /* module: move to the instruction "lui v1, HI_16BIT_OF_MCOUNT" */ - ip = self_addr - 20; - } else { - /* kernel: move to the instruction "move ra, at" */ - ip = self_addr - 12; - } + /* + * For module, move the ip from calling site of mcount to the + * instruction "lui v1, hi_16bit_of_mcount"(offset is 20), but for + * kernel, move to the instruction "move ra, at"(offset is 12) + */ + ip = self_addr - (in_module(self_addr) ? 20 : 12); - /* search the text until finding the non-store instruction or "s{d,w} - * ra, offset(sp)" instruction */ + /* + * search the text until finding the non-store instruction or "s{d,w} + * ra, offset(sp)" instruction + */ do { ip -= 4; /* get the code at "ip": code = *(unsigned int *)ip; */ safe_load_code(code, ip, faulted); - if (unlikely(faulted)) return 0; - /* If we hit the non-store instruction before finding where the + /* + * If we hit the non-store instruction before finding where the * ra is stored, then this is a leaf function and it does not - * store the ra on the stack. */ + * store the ra on the stack. + */ if ((code & S_R_SP) != S_R_SP) return parent_addr; - } while (((code & S_RA_SP) != S_RA_SP)); sp = fp + (code & OFFSET_MASK); @@ -211,16 +229,17 @@ unsigned long ftrace_get_parent_addr(unsigned long self_addr, void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr, unsigned long fp) { + int faulted; unsigned long old; struct ftrace_graph_ent trace; unsigned long return_hooker = (unsigned long) &return_to_handler; - int faulted; if (unlikely(atomic_read(¤t->tracing_graph_pause))) return; - /* "parent" is the stack address saved the return address of the caller + /* + * "parent" is the stack address saved the return address of the caller * of _mcount. * * if the gcc < 4.5, a leaf function does not save the return address @@ -241,12 +260,14 @@ void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr, if (unlikely(faulted)) goto out; #ifndef KBUILD_MCOUNT_RA_ADDRESS - parent = (unsigned long *)ftrace_get_parent_addr(self_addr, old, - (unsigned long)parent, - fp); - /* If fails when getting the stack address of the non-leaf function's - * ra, stop function graph tracer and return */ - if (parent == 0) + parent = (unsigned long *)ftrace_get_parent_addr( + self_addr, old, (unsigned long)parent, fp); + + /* + * If fails on getting the stack address of the non-leaf function's ra, + * stop function graph tracer and return + */ + if (unlikely(parent == 0)) goto out; #endif /* *parent = return_hooker; */ @@ -272,4 +293,4 @@ out: ftrace_graph_stop(); WARN_ON(1); } -#endif /* CONFIG_FUNCTION_GRAPH_TRACER */ +#endif /* CONFIG_FUNCTION_GRAPH_TRACER */ -- 1.7.0.1 From ralf@linux-mips.org Thu Mar 11 21:14:55 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Thu, 11 Mar 2010 21:14:58 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:57172 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492660Ab0CKUOz (ORCPT ); Thu, 11 Mar 2010 21:14:55 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o2BKEpjL003833; Thu, 11 Mar 2010 21:14:52 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o2BKEobw003830; Thu, 11 Mar 2010 21:14:50 +0100 Date: Thu, 11 Mar 2010 21:14:47 +0100 From: Ralf Baechle To: Wu Zhangjin Cc: linux-mips@linux-mips.org Subject: Re: [PATCH] Loongson: Add module info to the loongson2_clock driver Message-ID: <20100311201447.GA25163@linux-mips.org> References: <1268153601-4396-1-git-send-email-wuzhangin@gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1268153601-4396-1-git-send-email-wuzhangin@gmail.com> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26206 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Wed, Mar 10, 2010 at 12:53:21AM +0800, Wu Zhangjin wrote: > This patch fixes the warning of the tool which checks the module info of > the loongson2_clock driver when inserting it into the kernel: > > "Feb 25 23:42:27 localhost kernel: [ 4.965000] loongson2_clock: module > license 'unspecified' taints kernel." > > Reported-by: Liu Shiwei > Signed-off-by: Wu Zhangjin Thanks, applied. Ralf From w.sang@pengutronix.de Fri Mar 12 08:05:36 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Fri, 12 Mar 2010 08:05:40 +0100 (CET) Received: from metis.ext.pengutronix.de ([92.198.50.35]:38654 "EHLO metis.ext.pengutronix.de" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1490950Ab0CLHFg (ORCPT ); Fri, 12 Mar 2010 08:05:36 +0100 Received: from themisto.ext.pengutronix.de ([92.198.50.58] helo=pengutronix.de) by metis.ext.pengutronix.de with esmtp (Exim 4.71) (envelope-from ) id 1Npyvy-00009C-G7; Fri, 12 Mar 2010 08:05:25 +0100 From: Wolfram Sang To: kernel-janitors@vger.kernel.org Cc: Wolfram Sang , Ralf Baechle , "Eric W. Biederman" , linux-mips@linux-mips.org, linux-kernel@vger.kernel.org Date: Fri, 12 Mar 2010 08:03:49 +0100 Message-Id: <1268377431-11671-2-git-send-email-w.sang@pengutronix.de> X-Mailer: git-send-email 1.7.0 In-Reply-To: <1268377431-11671-1-git-send-email-w.sang@pengutronix.de> References: <1268377431-11671-1-git-send-email-w.sang@pengutronix.de> X-SA-Exim-Connect-IP: 92.198.50.58 X-SA-Exim-Mail-From: w.sang@pengutronix.de Subject: [PATCH 1/3] arch/mips/txx9/generic: init dynamic bin_attribute structures X-SA-Exim-Version: 4.2.1 (built Wed, 25 Jun 2008 17:14:11 +0000) X-SA-Exim-Scanned: Yes (on metis.ext.pengutronix.de) X-PTX-Original-Recipient: linux-mips@linux-mips.org Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26207 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: w.sang@pengutronix.de Precedence: bulk X-list: linux-mips Commit 6992f5334995af474c2b58d010d08bc597f0f2fe introduced this requirement. Found with coccinelle, but fixed manually. Compile tested on X86 where possible. Signed-off-by: Wolfram Sang Cc: Ralf Baechle Cc: Eric W. Biederman --- arch/mips/txx9/generic/setup.c | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/arch/mips/txx9/generic/setup.c b/arch/mips/txx9/generic/setup.c index 7174d83..95184a0 100644 --- a/arch/mips/txx9/generic/setup.c +++ b/arch/mips/txx9/generic/setup.c @@ -956,6 +956,7 @@ void __init txx9_sramc_init(struct resource *r) if (!dev->base) goto exit; dev->dev.cls = &txx9_sramc_sysdev_class; + sysfs_bin_attr_init(&dev->bindata_attr); dev->bindata_attr.attr.name = "bindata"; dev->bindata_attr.attr.mode = S_IRUSR | S_IWUSR; dev->bindata_attr.read = txx9_sram_read; -- 1.7.0 From tsbogend@alpha.franken.de Fri Mar 12 09:59:07 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Fri, 12 Mar 2010 09:59:10 +0100 (CET) Received: from elvis.franken.de ([193.175.24.41]:46094 "EHLO elvis.franken.de" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491864Ab0CLI7G (ORCPT ); Fri, 12 Mar 2010 09:59:06 +0100 Received: from uucp (helo=solo.franken.de) by elvis.franken.de with local-bsmtp (Exim 3.36 #1) id 1Nq0i1-0005TL-01; Fri, 12 Mar 2010 09:59:05 +0100 Received: by solo.franken.de (Postfix, from userid 1000) id 6B5BBC35BE; Fri, 12 Mar 2010 09:50:53 +0100 (CET) Date: Fri, 12 Mar 2010 09:50:53 +0100 From: Thomas Bogendoerfer To: Wu Zhangjin Cc: Ralf Baechle , Steven Rostedt , linux-mips@linux-mips.org, David Daney Subject: Re: [PATCH] MIPS: tracing: Optimize the implementation Message-ID: <20100312085053.GB6364@alpha.franken.de> References: <8b93c417fefa4d446f801abfd718ba94fdcb1821.1268330348.git.wuzhangjin@gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <8b93c417fefa4d446f801abfd718ba94fdcb1821.1268330348.git.wuzhangjin@gmail.com> User-Agent: Mutt/1.5.18 (2008-05-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26208 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: tsbogend@alpha.franken.de Precedence: bulk X-list: linux-mips On Fri, Mar 12, 2010 at 02:07:37AM +0800, Wu Zhangjin wrote: > +/* > + * If the Instruction Pointer is in module space (0xc0000000), return ture; > + * otherwise, it is in kernel space (0x80000000), return false. > + */ > +#define in_module(ip) (unlikely((ip) & 0x40000000)) > + looks broken for 64bit, but maybe this is a 32bit only feature... Thomas. -- Crap can work. Given enough thrust pigs will fly, but it's not necessary a good idea. [ RFC1925, 2.3 ] From ralf@linux-mips.org Fri Mar 12 11:14:20 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Fri, 12 Mar 2010 11:14:23 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:47660 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492142Ab0CLKOT (ORCPT ); Fri, 12 Mar 2010 11:14:19 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o2CAEErk023110; Fri, 12 Mar 2010 11:14:15 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o2CAEDf0023108; Fri, 12 Mar 2010 11:14:13 +0100 Date: Fri, 12 Mar 2010 11:14:13 +0100 From: Ralf Baechle To: Thomas Bogendoerfer Cc: Wu Zhangjin , Steven Rostedt , linux-mips@linux-mips.org, David Daney Subject: Re: [PATCH] MIPS: tracing: Optimize the implementation Message-ID: <20100312101413.GA22877@linux-mips.org> References: <8b93c417fefa4d446f801abfd718ba94fdcb1821.1268330348.git.wuzhangjin@gmail.com> <20100312085053.GB6364@alpha.franken.de> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20100312085053.GB6364@alpha.franken.de> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26209 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Fri, Mar 12, 2010 at 09:50:53AM +0100, Thomas Bogendoerfer wrote: > On Fri, Mar 12, 2010 at 02:07:37AM +0800, Wu Zhangjin wrote: > > +/* > > + * If the Instruction Pointer is in module space (0xc0000000), return ture; > > + * otherwise, it is in kernel space (0x80000000), return false. > > + */ > > +#define in_module(ip) (unlikely((ip) & 0x40000000)) > > + > > looks broken for 64bit, but maybe this is a 32bit only feature... This gem did already exist in the old code, so no regression: if (ip & 0x40000000) { Ralf From wuzhangjin@gmail.com Fri Mar 12 11:18:18 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Fri, 12 Mar 2010 11:18:22 +0100 (CET) Received: from mail-pz0-f185.google.com ([209.85.222.185]:40775 "EHLO mail-pz0-f185.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491778Ab0CLKSS (ORCPT ); Fri, 12 Mar 2010 11:18:18 +0100 Received: by pzk15 with SMTP id 15so606336pzk.21 for ; Fri, 12 Mar 2010 02:18:11 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:subject:from:reply-to:to:cc :in-reply-to:references:content-type:organization:date:message-id :mime-version:x-mailer:content-transfer-encoding; bh=mZbQYMLk+4voatF7uYgQaESzSIAXzmsGpjZV3V0Z8Is=; b=iDfjp/hJxWH+uxwWh6GMSPCC0XfJPh1Kl3O9T5P4VCnAR52YGtUpKHH+wWkqs/4QuM l/z4hrCekVRzJptDjhO4JpteNaGxF68snsC9XG6ACs4On6MlEy19Si2GZTN0NsAs+6ew Rhzd8cuK0MgxhNK1P/jSmV3bTwZ+gCb0NUHTA= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=subject:from:reply-to:to:cc:in-reply-to:references:content-type :organization:date:message-id:mime-version:x-mailer :content-transfer-encoding; b=B7sJ2AyJ1Da2YPGqia2vHt2JcolkzzUMN5iLzhVSHpKNpvtPnDu8oLL8wlOb14BJtk ZmpbymzaIyU/0YoT3Ze0czfbP1kCNytuI6vVqGXlJoX4mVmYNk3BdTme03JGb7m2yVZy QQeIcBySq94Lh5ZDEVeEReHuoG4UJSgGeTXeU= Received: by 10.142.61.19 with SMTP id j19mr2358529wfa.69.1268389090936; Fri, 12 Mar 2010 02:18:10 -0800 (PST) Received: from [202.201.12.142] ([202.201.12.142]) by mx.google.com with ESMTPS id 21sm1096206pzk.12.2010.03.12.02.18.06 (version=TLSv1/SSLv3 cipher=RC4-MD5); Fri, 12 Mar 2010 02:18:10 -0800 (PST) Subject: Re: [PATCH] MIPS: tracing: Optimize the implementation From: Wu Zhangjin Reply-To: wuzhangjin@gmail.com To: Thomas Bogendoerfer Cc: Ralf Baechle , Steven Rostedt , linux-mips@linux-mips.org, David Daney In-Reply-To: <20100312085053.GB6364@alpha.franken.de> References: <8b93c417fefa4d446f801abfd718ba94fdcb1821.1268330348.git.wuzhangjin@gmail.com> <20100312085053.GB6364@alpha.franken.de> Content-Type: text/plain; charset="UTF-8" Organization: DSLab, Lanzhou University, China Date: Fri, 12 Mar 2010 18:11:35 +0800 Message-ID: <1268388695.6447.3.camel@falcon> Mime-Version: 1.0 X-Mailer: Evolution 2.28.2 Content-Transfer-Encoding: 7bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26210 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: wuzhangjin@gmail.com Precedence: bulk X-list: linux-mips On Fri, 2010-03-12 at 09:50 +0100, Thomas Bogendoerfer wrote: > On Fri, Mar 12, 2010 at 02:07:37AM +0800, Wu Zhangjin wrote: > > +/* > > + * If the Instruction Pointer is in module space (0xc0000000), return ture; > > + * otherwise, it is in kernel space (0x80000000), return false. > > + */ > > +#define in_module(ip) (unlikely((ip) & 0x40000000)) > > + > > looks broken for 64bit, but maybe this is a 32bit only feature... Actually, this works well on my 64bit YeeLoong laptop ;) But we should find another better solution. Regards, Wu Zhangjin From wuzhangjin@gmail.com Fri Mar 12 12:50:20 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Fri, 12 Mar 2010 12:50:28 +0100 (CET) Received: from mail-pz0-f185.google.com ([209.85.222.185]:63194 "EHLO mail-pz0-f185.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491148Ab0CLLuU (ORCPT ); Fri, 12 Mar 2010 12:50:20 +0100 Received: by pzk15 with SMTP id 15so644972pzk.21 for ; Fri, 12 Mar 2010 03:50:13 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:subject:from:reply-to:to:cc :in-reply-to:references:content-type:organization:date:message-id :mime-version:x-mailer:content-transfer-encoding; bh=0dqy4Ue13PuSaX+6VsF1ieX9t0MvOCcxs1TN4dFE82A=; b=t5KrgIzlj5Wpvjy7+SmU0yNOdo5stFtba9KqxZQF6sspw9VO4sVzTKl7UgPLNKQQyQ Tkci0pxgCR8uV4mZJj7qXcA7C+GhfAXHULRQtPQJkPLruGaPMSUT/yEs46602G1yKWR9 g8beuLR4mT0c+wR4+1lfzs6YPe/L4flV6N+WQ= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=subject:from:reply-to:to:cc:in-reply-to:references:content-type :organization:date:message-id:mime-version:x-mailer :content-transfer-encoding; b=H1pTaPUywPsII/3DQ2GgDp4YLN5b19+33MU7pzA/zZjjI6GuwSDGIJ173bd5pildTE /DxJ3rYCYVLI4io/PX7Lp4ibJnbArAwRkK5hSxBqK2a4Dokoc05K5By4poqi0J+eHq4R 3yNF/KmWqgSXLGIFs+wUK7CnDYt/LXeFk1pN4= Received: by 10.141.213.22 with SMTP id p22mr2868815rvq.94.1268394613387; Fri, 12 Mar 2010 03:50:13 -0800 (PST) Received: from [202.201.12.142] ([202.201.12.142]) by mx.google.com with ESMTPS id 20sm1195946pzk.11.2010.03.12.03.50.09 (version=TLSv1/SSLv3 cipher=RC4-MD5); Fri, 12 Mar 2010 03:50:12 -0800 (PST) Subject: Re: [PATCH] MIPS: tracing: Optimize the implementation From: Wu Zhangjin Reply-To: wuzhangjin@gmail.com To: David Daney Cc: Ralf Baechle , Steven Rostedt , linux-mips@linux-mips.org In-Reply-To: <4B993B32.7000006@caviumnetworks.com> References: <8b93c417fefa4d446f801abfd718ba94fdcb1821.1268330348.git.wuzhangjin@gmail.com> <4B993B32.7000006@caviumnetworks.com> Content-Type: text/plain; charset="UTF-8" Organization: DSLab, Lanzhou University, China Date: Fri, 12 Mar 2010 19:43:29 +0800 Message-ID: <1268394209.6447.94.camel@falcon> Mime-Version: 1.0 X-Mailer: Evolution 2.28.2 Content-Transfer-Encoding: 7bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26211 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: wuzhangjin@gmail.com Precedence: bulk X-list: linux-mips Hi, On Thu, 2010-03-11 at 10:49 -0800, David Daney wrote: [...] > > +/* > > + * If the Instruction Pointer is in module space (0xc0000000), return ture; > > s/ture/true/ > yeah. > > + * otherwise, it is in kernel space (0x80000000), return false. > > + */ > > +#define in_module(ip) (unlikely((ip)& 0x40000000)) > > + > > This isn't universally true, but it does hold for most configurations I > think. Although I'm not sure who is the exception, we always need an universal solution, what about this: Compare module with kernel: module: lui v1, hi16_mcount <--- ip addiu v1, v1, lo16_mcount move at, ra jalr v1 nop kernel: move at, ra jal _mcount <--- ip The above _ip_ is the address have been recorded into the __mcount_loc section of the kernel by scripts/recordmcount.pl, as we can see, for kernel, the *(ip - 4) is "move at, ra": 03e0082d, a certain instruction, but for module, there is no possibility(?) of existing a "move at, ra" at *(ip -4) but a register saving operation("s {d,w} rs, offset(sp)", prefixed by 0xffb0 for 64bit and 0xafb0 for 32bit. ), and reversly, for kernel, there is no such instruction there. And consider the new option -mmcount-ra-address of gcc, some more instructions will be inserted between "move at, ra" and the calling site to mcount, so, *(ip-4) will not always be "move at, ra", then we need to check if there is a "s {d,w} rs, offset(sp)" there, if yes, it is in module, otherwise, it should be in kernel. #define S_RS_SP 0xafb00000 /* s{d,w} rs, offset(sp) */ static inline int in_module(ip) { insn = *(ip - 4); /* need to use safe_load_code instead, what about big endian? */ return ((insn & S_RS_SP) == S_RS_SP) } > > [...] > > > + /* > > + * We have compiled modules with -mlong-calls, but compiled kernel > > + * without it, therefore, need to cope with them respectively. > > + * > > + * For module: > > + * > > + * lui v1, hi16_mcount --> b 1f > > + * addiu v1, v1, lo16_mcount > > + * move at, ra > > + * jalr v1 > > + * nop > > + * 1f: (ip + 16) > > > Have you thought about just overwriting the jalr here instead of > branching around it? In any event, I don't think you can count on a > fixed size code sequence for calling _mcount. We are passing the > address of the save location of RA to _mcount too. The size of the code > will depend on the size of the functions stack frame *and* weather or > not it is a leaf function. Although in the kernel we are unlikely to > see functions with large stack frames. So even with "b 1f", we need to use the right offset, the original version for module with -mmcount-ra-address should have bugs here for the offset should be 16 + 8 or 4 (two instructions for leaf function, one instruction for non-leaf function). but for we only recorded the position of "lui v1, hi16_mcount" in the __mcount_loc section, so we need to search the position of the real calling site of mcount(jalr v1), this will goes to what you have suggested below. > > > > + * For kernel: > > + * > > + * move at, ra > > + * jal _mcount --> nop > > + * > > + */ > > + new = in_module(ip) ? INSN_B_1F : INSN_NOP; > > > What would happen if you read the code to find the first JAL or JALR, > and then overwrote it with a NOP instead of relying on the function > address to figure out which type of prolog it has? > > The reason I suggest this is that sometimes we place the entire kernel > in CSSEG. When this is done, everything has the same (short) _mcount > calling sequence. Right, then, we can search the JAL or JALR, for kernel, will get it immediatly, for module, will only several instructions, we can do this searching in ftrace_make_nop and ftrace_make_call at run-time, but just found we can use the following function to do it in ftrace_init(), looks good. static inline int is_call_mcount(unsigned int insn) { return ((insn & JAL) == JAL) || (insn == JALR_V1); } static inline unsinged long mcount_callsite(unsigned long addr) { unsigned int insn; insn = *(unsigned int *)addr; /*need safe_load_code*/ if (is_call_mcount(insn)) return addr; do { addr += 4; /* what about big endian? */ insn = *(unsigned int *)addr; /*need safe_load_code*/ } while (!is_call_mcount(insn)); return addr; } static inline unsigned long ftrace_call_adjust(unsigned long addr) { return mcount_callsite(addr); } With the above support, we only need this new ftrace_make_nop: *(unsigned int *)ip = INSN_NOP; (But for module, this may need more overhead than "b 1f". ) and ftrace_make_call: *(unsigned int *)ip = in_module(ip) : INSN_JALR_V1 : insn_jal_mcount; (And here, for module, we need more time to determine which space we are.) Any more suggestion? Thanks & Regards, Wu Zhangjin From wuzhangjin@gmail.com Fri Mar 12 16:03:49 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Fri, 12 Mar 2010 16:03:53 +0100 (CET) Received: from mail-px0-f189.google.com ([209.85.216.189]:45353 "EHLO mail-px0-f189.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492707Ab0CLPDt (ORCPT ); Fri, 12 Mar 2010 16:03:49 +0100 Received: by pxi27 with SMTP id 27so1265400pxi.0 for ; Fri, 12 Mar 2010 07:03:42 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:subject:from:reply-to:to:cc :in-reply-to:references:content-type:organization:date:message-id :mime-version:x-mailer:content-transfer-encoding; bh=7DRRm1y41sEgGyKsYWpW4b4Cv34B6XM69lXsKM6TCpQ=; b=PXxmf9m82nf8ijFsHmjQ6E9cDa7hkUsn/+YcLNLkZYOAzl+RMsx3zqWFFnz7x7MWMA LPZ5bBZxe7NxBy3oW4i1ye8S2wESZszI15FjM7cvo8xMtISJ3MB0AZ808ssjZMaEmTeO ZuoVHRBRLzHsptBmNZ7nEmT/WRTt0tsivYnaw= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=subject:from:reply-to:to:cc:in-reply-to:references:content-type :organization:date:message-id:mime-version:x-mailer :content-transfer-encoding; b=xUboVqciTDBBzi7Bt7p+KzOuLOke5DSTxBng1r6YE0AtQYdX/znRvkdZ1lctEuSIRz TT73IFrICsafvoasESvsUvIjvbEHiQEurJs/jYnSdAzYNUnoPi7yplgGz+US5VxtIAeV bWJF+1RF3AG4iUEMZTL3kufBDa4U5SCJGa/Ec= Received: by 10.143.84.6 with SMTP id m6mr2656435wfl.162.1268406222422; Fri, 12 Mar 2010 07:03:42 -0800 (PST) Received: from [202.201.12.142] ([202.201.12.142]) by mx.google.com with ESMTPS id 23sm1415526pzk.2.2010.03.12.07.03.37 (version=TLSv1/SSLv3 cipher=RC4-MD5); Fri, 12 Mar 2010 07:03:41 -0800 (PST) Subject: Re: [PATCH] MIPS: tracing: Optimize the implementation From: Wu Zhangjin Reply-To: wuzhangjin@gmail.com To: David Daney Cc: Ralf Baechle , Steven Rostedt , linux-mips@linux-mips.org In-Reply-To: <1268394209.6447.94.camel@falcon> References: <8b93c417fefa4d446f801abfd718ba94fdcb1821.1268330348.git.wuzhangjin@gmail.com> <4B993B32.7000006@caviumnetworks.com> <1268394209.6447.94.camel@falcon> Content-Type: text/plain; charset="UTF-8" Organization: DSLab, Lanzhou University, China Date: Fri, 12 Mar 2010 22:57:08 +0800 Message-ID: <1268405828.9527.3.camel@falcon> Mime-Version: 1.0 X-Mailer: Evolution 2.28.2 Content-Transfer-Encoding: 7bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26212 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: wuzhangjin@gmail.com Precedence: bulk X-list: linux-mips On Fri, 2010-03-12 at 19:43 +0800, Wu Zhangjin wrote: [...] > Right, then, we can search the JAL or JALR, for kernel, will get it > immediatly, for module, will only several instructions, we can do this > searching in ftrace_make_nop and ftrace_make_call at run-time, but just > found we can use the following function to do it in ftrace_init(), looks > good. > > static inline int is_call_mcount(unsigned int insn) > { > return ((insn & JAL) == JAL) || (insn == JALR_V1); > } > > static inline unsinged long mcount_callsite(unsigned long addr) > { > unsigned int insn; > > insn = *(unsigned int *)addr; /*need safe_load_code*/ > if (is_call_mcount(insn)) > return addr; > > do { > addr += 4; /* what about big endian? */ > insn = *(unsigned int *)addr; /*need safe_load_code*/ > } while (!is_call_mcount(insn)); > This is not possible for modules, for currently, the modules are not loaded yet. Regards, Wu Zhangjin From David.Daney@caviumnetworks.com Fri Mar 12 17:46:09 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Fri, 12 Mar 2010 17:46:13 +0100 (CET) Received: from mail3.caviumnetworks.com ([12.108.191.235]:18528 "EHLO mail3.caviumnetworks.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492250Ab0CLQqJ (ORCPT ); Fri, 12 Mar 2010 17:46:09 +0100 Received: from caexch01.caveonetworks.com (Not Verified[192.168.16.9]) by mail3.caviumnetworks.com with MailMarshal (v6,7,2,8378) id ; Fri, 12 Mar 2010 08:46:15 -0800 Received: from caexch01.caveonetworks.com ([192.168.16.9]) by caexch01.caveonetworks.com with Microsoft SMTPSVC(6.0.3790.3959); Fri, 12 Mar 2010 08:36:15 -0800 Received: from dd1.caveonetworks.com ([12.108.191.236]) by caexch01.caveonetworks.com over TLS secured channel with Microsoft SMTPSVC(6.0.3790.3959); Fri, 12 Mar 2010 08:36:15 -0800 Message-ID: <4B9A6D79.6040306@caviumnetworks.com> Date: Fri, 12 Mar 2010 08:36:09 -0800 From: David Daney User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.8) Gecko/20100301 Fedora/3.0.3-1.fc12 Thunderbird/3.0.3 MIME-Version: 1.0 To: Thomas Bogendoerfer CC: Wu Zhangjin , Ralf Baechle , Steven Rostedt , linux-mips@linux-mips.org Subject: Re: [PATCH] MIPS: tracing: Optimize the implementation References: <8b93c417fefa4d446f801abfd718ba94fdcb1821.1268330348.git.wuzhangjin@gmail.com> <20100312085053.GB6364@alpha.franken.de> In-Reply-To: <20100312085053.GB6364@alpha.franken.de> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-OriginalArrivalTime: 12 Mar 2010 16:36:15.0105 (UTC) FILETIME=[2215AF10:01CAC202] Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26213 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ddaney@caviumnetworks.com Precedence: bulk X-list: linux-mips On 03/12/2010 12:50 AM, Thomas Bogendoerfer wrote: > On Fri, Mar 12, 2010 at 02:07:37AM +0800, Wu Zhangjin wrote: >> +/* >> + * If the Instruction Pointer is in module space (0xc0000000), return ture; >> + * otherwise, it is in kernel space (0x80000000), return false. >> + */ >> +#define in_module(ip) (unlikely((ip)& 0x40000000)) >> + > > looks broken for 64bit, but maybe this is a 32bit only feature... > I initially thought the same thing. However for 64-bit kernels linked in ckseg0 it is still true. If we use the -msym32 optimization, we are forced to be in the ckseg space, so for most cases it works. David Daney. From wuzhangjin@gmail.com Fri Mar 12 18:34:42 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Fri, 12 Mar 2010 18:34:47 +0100 (CET) Received: from mail-fx0-f217.google.com ([209.85.220.217]:49724 "EHLO mail-fx0-f217.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491131Ab0CLRem (ORCPT ); Fri, 12 Mar 2010 18:34:42 +0100 Received: by fxm9 with SMTP id 9so1414298fxm.24 for ; Fri, 12 Mar 2010 09:34:36 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:from:to:cc:subject:date :message-id:x-mailer; bh=730+lr+tayoIgQakMIiiexd0Ac8umVZw7Dij17DO6Cg=; b=s8/+4i5KN4Fm7ORn3gAa0vAthiOIN8ZVkoZVqzdmmtJSfl3xvUXYyziAnlGuNF+m+C RWgB1iKtDAJPutLjPOwLipeHTEezWedi/PCyjpK6N8qfqklOp3PGosDqDNJD2EN8OqKe VahaXz7JgPJSEwmPMHRCd/bsFBmz71hy8ZtuQ= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=from:to:cc:subject:date:message-id:x-mailer; b=fmpMY3QDFt0bCeKTnwDNKw78Pfxdc6OStqmZCP/jTnS/GtMhTG6gPjBFHnMJEnZ9BC Jf4F7k/OAGSFVopI2F/HiFgjLBannWKTZfTplPVjZgsCEEN97D+2oKW01wvHSCSDI1e9 GRJ1c/kUBrz0ZSYYKkFMy/+J+7mibZydtvuOo= Received: by 10.223.16.66 with SMTP id n2mr1447136faa.83.1268415275887; Fri, 12 Mar 2010 09:34:35 -0800 (PST) Received: from localhost.localdomain ([202.201.12.142]) by mx.google.com with ESMTPS id p17sm3401820fka.9.2010.03.12.09.34.29 (version=TLSv1/SSLv3 cipher=RC4-MD5); Fri, 12 Mar 2010 09:34:34 -0800 (PST) From: Wu Zhangjin To: Ralf Baechle Cc: Steven Rostedt , linux-mips@linux-mips.org, David Daney , Thomas Bogendoerfer , Wu Zhangjin Subject: [PATCH v2] MIPS: tracing: Optimize the implementation Date: Sat, 13 Mar 2010 01:27:46 +0800 Message-Id: <0c3597eeee4dbd69eb5a74c2b46179014b291d56.1268414797.git.wuzhangjin@gmail.com> X-Mailer: git-send-email 1.7.0.1 Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26214 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: wuzhangjin@gmail.com Precedence: bulk X-list: linux-mips From: Wu Zhangjin This revision fixes the potential bugs (reported by David) when compiling the kernel with -mmcount-ra-address of gcc 4.5, I only tested it without this option for I don't have a gcc 4.5 currently. ------------ Changes: v1 -> v2: + Fixup the support of -mmcount-ra-address The offset of "b 1f" instruction should be 20 for leaf-function and 24 for non-leaf function when -mmcount-ra-address is enabled. This patch adds a new get_insn_b_1f() function to get the different "b 1f" instruction. (only test without -mmcount-ra-address.) v1: + Reduce the runtime overhead o Uses macros instead of variables for the fixed instructions to reduce memory access o Moves the Initilization of the instructions which will be fixed after linking from ftrace_make_nop/ftrace_make_call to ftrace_dyn_arch_init() and encodes the instructions through uasm(arch/mips/include/asm/uasm.h). o A common macro in_module() is defined to determine which space the instruction pointer stays in and several related conditional statements are converted to conditional operator(? :) statement. + Cleanup the whole stuff Lots of comments/macros have been cleaned up to let it look better. Signed-off-by: Wu Zhangjin --- arch/mips/include/asm/ftrace.h | 12 ++- arch/mips/kernel/ftrace.c | 236 ++++++++++++++++++++++++---------------- 2 files changed, 153 insertions(+), 95 deletions(-) diff --git a/arch/mips/include/asm/ftrace.h b/arch/mips/include/asm/ftrace.h index ce35c9a..6dbfa1e 100644 --- a/arch/mips/include/asm/ftrace.h +++ b/arch/mips/include/asm/ftrace.h @@ -3,7 +3,7 @@ * License. See the file "COPYING" in the main directory of this archive for * more details. * - * Copyright (C) 2009 DSLab, Lanzhou University, China + * Copyright (C) 2009, 2010 DSLab, Lanzhou University, China * Author: Wu Zhangjin */ @@ -19,6 +19,14 @@ extern void _mcount(void); #define mcount _mcount +/* + * If the Instruction Pointer is in module space (0xc0000000), return true; + * otherwise, it is in kernel space (0x80000000), return false. + * + * FIXME: This may not work in some cases. + */ +#define in_module(ip) (unlikely((ip) & 0x40000000)) + #define safe_load(load, src, dst, error) \ do { \ asm volatile ( \ @@ -83,8 +91,8 @@ static inline unsigned long ftrace_call_adjust(unsigned long addr) struct dyn_arch_ftrace { }; +#endif /* CONFIG_DYNAMIC_FTRACE */ -#endif /* CONFIG_DYNAMIC_FTRACE */ #endif /* __ASSEMBLY__ */ #endif /* CONFIG_FUNCTION_TRACER */ #endif /* _ASM_MIPS_FTRACE_H */ diff --git a/arch/mips/kernel/ftrace.c b/arch/mips/kernel/ftrace.c index e9e64e0..b72107e 100644 --- a/arch/mips/kernel/ftrace.c +++ b/arch/mips/kernel/ftrace.c @@ -2,7 +2,7 @@ * Code for replacing ftrace calls with jumps. * * Copyright (C) 2007-2008 Steven Rostedt - * Copyright (C) 2009 DSLab, Lanzhou University, China + * Copyright (C) 2009, 2010 DSLab, Lanzhou University, China * Author: Wu Zhangjin * * Thanks goes to Steven Rostedt for writing the original x86 version. @@ -15,15 +15,48 @@ #include #include #include +#include #ifdef CONFIG_DYNAMIC_FTRACE -#define JAL 0x0c000000 /* jump & link: ip --> ra, jump to target */ -#define ADDR_MASK 0x03ffffff /* op_code|addr : 31...26|25 ....0 */ -#define jump_insn_encode(op_code, addr) \ - ((unsigned int)((op_code) | (((addr) >> 2) & ADDR_MASK))) +/* Before linking, the following instructions are fixed. */ +#ifdef CONFIG_CPU_LOONGSON2F +#define INSN_NOP 0x00200825 /* or at, at, zero */ +#else +#define INSN_NOP 0x00000000 /* nop */ +#endif +#define INSN_B_1F_16 0x10000004 /* b 1f; offset = (16 >> 2) */ +#define INSN_B_1F_20 0x10000005 /* b 1f; offset = (20 >> 2) */ +#define INSN_B_1F_24 0x10000006 /* b 1f; offset = (24 >> 2) */ + +/* After linking, the following instructions are fixed. */ +static unsigned int insn_jal_ftrace_caller __read_mostly; +static unsigned int insn_lui_v1_hi16_mcount __read_mostly; +static unsigned int insn_j_ftrace_graph_caller __read_mostly; -static unsigned int ftrace_nop = 0x00000000; +/* The following instructions are encoded in the run-time */ +/* insn: jal func; op_code|addr : 31...26|25 ....0 */ +#define INSN_JAL(addr) \ + ((unsigned int)(0x0c000000 | (((addr) >> 2) & 0x03ffffff))) + +static inline void ftrace_dyn_arch_init_insns(void) +{ + u32 *buf; + unsigned int v1; + + /* lui v1, hi16_mcount */ + v1 = 3; + buf = (u32 *)&insn_lui_v1_hi16_mcount; + UASM_i_LA_mostly(&buf, v1, MCOUNT_ADDR); + + /* jal (ftrace_caller + 8), jump over the first two instruction */ + buf = (u32 *)&insn_jal_ftrace_caller; + uasm_i_jal(&buf, (FTRACE_ADDR + 8)); + + /* j ftrace_graph_caller */ + buf = (u32 *)&insn_j_ftrace_graph_caller; + uasm_i_j(&buf, (unsigned long)ftrace_graph_caller); +} static int ftrace_modify_code(unsigned long ip, unsigned int new_code) { @@ -31,7 +64,6 @@ static int ftrace_modify_code(unsigned long ip, unsigned int new_code) /* *(unsigned int *)ip = new_code; */ safe_store_code(new_code, ip, faulted); - if (unlikely(faulted)) return -EFAULT; @@ -40,118 +72,133 @@ static int ftrace_modify_code(unsigned long ip, unsigned int new_code) return 0; } -static int lui_v1; -static int jal_mcount; +static inline unsigned int get_insn_b_1f(unsigned int *insn, unsigned long ip) +{ +#ifndef KBUILD_MCOUNT_RA_ADDRESS + *insn = INSN_B_1F_16; + return 0; +#else + int faulted; + unsigned int code; + + ip -= 4; + + /* code = *(unsigned int *)ip; */ + safe_load_code(code, ip, faulted); + if (unlikely(faulted)) + return -EFAULT; + + if ((code & S_R_SP) != S_R_SP) + *insn = INSN_B_1F_20; + else + *insn = INSN_B_1F_24; + + return 0; +#endif +} int ftrace_make_nop(struct module *mod, struct dyn_ftrace *rec, unsigned long addr) { unsigned int new; - int faulted; unsigned long ip = rec->ip; - /* We have compiled module with -mlong-calls, but compiled the kernel - * without it, we need to cope with them respectively. */ - if (ip & 0x40000000) { - /* record it for ftrace_make_call */ - if (lui_v1 == 0) { - /* lui_v1 = *(unsigned int *)ip; */ - safe_load_code(lui_v1, ip, faulted); - - if (unlikely(faulted)) - return -EFAULT; - } - - /* lui v1, hi_16bit_of_mcount --> b 1f (0x10000004) - * addiu v1, v1, low_16bit_of_mcount - * move at, ra - * jalr v1 - * nop - * 1f: (ip + 12) - */ - new = 0x10000004; - } else { - /* record/calculate it for ftrace_make_call */ - if (jal_mcount == 0) { - /* We can record it directly like this: - * jal_mcount = *(unsigned int *)ip; - * Herein, jump over the first two nop instructions */ - jal_mcount = jump_insn_encode(JAL, (MCOUNT_ADDR + 8)); - } - - /* move at, ra - * jalr v1 --> nop - */ - new = ftrace_nop; - } + /* + * We have compiled modules with -mlong-calls, but compiled kernel + * without it, therefore, need to cope with them respectively. + * + * For module: + * + * lui v1, hi16_mcount --> b 1f + * addiu v1, v1, lo16_mcount + * move at, ra + * jalr v1 + * nop + * 1f: (ip + 16) + * For kernel: + * + * move at, ra + * jal _mcount --> nop + * + * And with the -mmcount-ra-address option, the offset may be 20 for + * leaf fuction and 24 for non-leaf function. + */ + + if (!in_module(ip)) + new = INSN_NOP; + else if (get_insn_b_1f(&new, ip)) + return -EFAULT; + return ftrace_modify_code(ip, new); } -static int modified; /* initialized as 0 by default */ - int ftrace_make_call(struct dyn_ftrace *rec, unsigned long addr) { unsigned int new; unsigned long ip = rec->ip; - /* We just need to remove the "b ftrace_stub" at the fist time! */ - if (modified == 0) { - modified = 1; - ftrace_modify_code(addr, ftrace_nop); - } - /* ip, module: 0xc0000000, kernel: 0x80000000 */ - new = (ip & 0x40000000) ? lui_v1 : jal_mcount; + new = in_module(ip) ? insn_lui_v1_hi16_mcount : insn_jal_ftrace_caller; return ftrace_modify_code(ip, new); } -#define FTRACE_CALL_IP ((unsigned long)(&ftrace_call)) - int ftrace_update_ftrace_func(ftrace_func_t func) { unsigned int new; - new = jump_insn_encode(JAL, (unsigned long)func); + new = INSN_JAL((unsigned long)func); - return ftrace_modify_code(FTRACE_CALL_IP, new); + return ftrace_modify_code((unsigned long)(&ftrace_call), new); } int __init ftrace_dyn_arch_init(void *data) { + ftrace_dyn_arch_init_insns(); + + /* + * We are safe to remove the "b ftrace_stub" for the current + * ftrace_caller() is almost empty (only the stack operations), and + * most importantly, the calling to mcount will be disabled later in + * ftrace_init(), then there is no 'big' overhead. And in the future, + * we are hoping the function_trace_stop is initialized as 1 then we + * can eventually remove that 'useless' "b ftrace_stub" and the + * following nop. Currently, although we can call ftrace_stop() to set + * function_trace_stop as 1, but it will change the behavior of the + * Function Tracer. + */ + ftrace_modify_code(MCOUNT_ADDR, INSN_NOP); + /* The return code is retured via data */ *(unsigned long *)data = 0; return 0; } -#endif /* CONFIG_DYNAMIC_FTRACE */ +#endif /* CONFIG_DYNAMIC_FTRACE */ #ifdef CONFIG_FUNCTION_GRAPH_TRACER #ifdef CONFIG_DYNAMIC_FTRACE extern void ftrace_graph_call(void); -#define JMP 0x08000000 /* jump to target directly */ -#define CALL_FTRACE_GRAPH_CALLER \ - jump_insn_encode(JMP, (unsigned long)(&ftrace_graph_caller)) #define FTRACE_GRAPH_CALL_IP ((unsigned long)(&ftrace_graph_call)) int ftrace_enable_ftrace_graph_caller(void) { return ftrace_modify_code(FTRACE_GRAPH_CALL_IP, - CALL_FTRACE_GRAPH_CALLER); + insn_j_ftrace_graph_caller); } int ftrace_disable_ftrace_graph_caller(void) { - return ftrace_modify_code(FTRACE_GRAPH_CALL_IP, ftrace_nop); + return ftrace_modify_code(FTRACE_GRAPH_CALL_IP, INSN_NOP); } -#endif /* !CONFIG_DYNAMIC_FTRACE */ +#endif /* !CONFIG_DYNAMIC_FTRACE */ #ifndef KBUILD_MCOUNT_RA_ADDRESS -#define S_RA_SP (0xafbf << 16) /* s{d,w} ra, offset(sp) */ -#define S_R_SP (0xafb0 << 16) /* s{d,w} R, offset(sp) */ -#define OFFSET_MASK 0xffff /* stack offset range: 0 ~ PT_SIZE */ +#define INSN_S_RA_SP 0xafbf0000 /* s{d,w} ra, offset(sp) */ +#define INSN_S_R_SP 0xafb00000 /* s{d,w} R, offset(sp) */ +#define STACK_OFFSET_MASK 0xffff /* stack offset range: 0 ~ PT_SIZE */ unsigned long ftrace_get_parent_addr(unsigned long self_addr, unsigned long parent, @@ -162,35 +209,35 @@ unsigned long ftrace_get_parent_addr(unsigned long self_addr, unsigned int code; int faulted; - /* in module or kernel? */ - if (self_addr & 0x40000000) { - /* module: move to the instruction "lui v1, HI_16BIT_OF_MCOUNT" */ - ip = self_addr - 20; - } else { - /* kernel: move to the instruction "move ra, at" */ - ip = self_addr - 12; - } + /* + * For module, move the ip from calling site of mcount to the + * instruction "lui v1, hi_16bit_of_mcount"(offset is 20), but for + * kernel, move to the instruction "move ra, at"(offset is 12) + */ + ip = self_addr - (in_module(self_addr) ? 20 : 12); - /* search the text until finding the non-store instruction or "s{d,w} - * ra, offset(sp)" instruction */ + /* + * search the text until finding the non-store instruction or "s{d,w} + * ra, offset(sp)" instruction + */ do { ip -= 4; /* get the code at "ip": code = *(unsigned int *)ip; */ safe_load_code(code, ip, faulted); - if (unlikely(faulted)) return 0; - /* If we hit the non-store instruction before finding where the + /* + * If we hit the non-store instruction before finding where the * ra is stored, then this is a leaf function and it does not - * store the ra on the stack. */ - if ((code & S_R_SP) != S_R_SP) + * store the ra on the stack. + */ + if ((code & INSN_S_R_SP) != INSN_S_R_SP) return parent_addr; + } while (((code & INSN_S_RA_SP) != INSN_S_RA_SP)); - } while (((code & S_RA_SP) != S_RA_SP)); - - sp = fp + (code & OFFSET_MASK); + sp = fp + (code & STACK_OFFSET_MASK); /* ra = *(unsigned long *)sp; */ safe_load_stack(ra, sp, faulted); @@ -211,16 +258,17 @@ unsigned long ftrace_get_parent_addr(unsigned long self_addr, void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr, unsigned long fp) { + int faulted; unsigned long old; struct ftrace_graph_ent trace; unsigned long return_hooker = (unsigned long) &return_to_handler; - int faulted; if (unlikely(atomic_read(¤t->tracing_graph_pause))) return; - /* "parent" is the stack address saved the return address of the caller + /* + * "parent" is the stack address saved the return address of the caller * of _mcount. * * if the gcc < 4.5, a leaf function does not save the return address @@ -241,12 +289,14 @@ void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr, if (unlikely(faulted)) goto out; #ifndef KBUILD_MCOUNT_RA_ADDRESS - parent = (unsigned long *)ftrace_get_parent_addr(self_addr, old, - (unsigned long)parent, - fp); - /* If fails when getting the stack address of the non-leaf function's - * ra, stop function graph tracer and return */ - if (parent == 0) + parent = (unsigned long *)ftrace_get_parent_addr( + self_addr, old, (unsigned long)parent, fp); + + /* + * If fails on getting the stack address of the non-leaf function's ra, + * stop function graph tracer and return + */ + if (unlikely(parent == 0)) goto out; #endif /* *parent = return_hooker; */ @@ -272,4 +322,4 @@ out: ftrace_graph_stop(); WARN_ON(1); } -#endif /* CONFIG_FUNCTION_GRAPH_TRACER */ +#endif /* CONFIG_FUNCTION_GRAPH_TRACER */ -- 1.7.0.1 From wuzhangjin@gmail.com Fri Mar 12 18:39:25 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Fri, 12 Mar 2010 18:39:30 +0100 (CET) Received: from mail-fx0-f217.google.com ([209.85.220.217]:62085 "EHLO mail-fx0-f217.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491131Ab0CLRjZ (ORCPT ); Fri, 12 Mar 2010 18:39:25 +0100 Received: by fxm9 with SMTP id 9so1418987fxm.24 for ; Fri, 12 Mar 2010 09:39:19 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:subject:from:reply-to:to:cc :in-reply-to:references:content-type:organization:date:message-id :mime-version:x-mailer:content-transfer-encoding; bh=l3oblRJOCmNBEd5gVj+XC/fZnws2Ju+AYMUXm0Ja02k=; b=tBcU6sBB4nXtxqctrnocPbSSg58ZJQj+Lm6Ct5uVVhfinsipgJM9LIZE3a/vD0xOh6 V691FiWQkgBNLeHlNncn6IwjoWDAEZ6o3CUz3+7kvc3ndX1vXK0lzn6lQ/EB4D087CWu MBd9mHBF1D2F1WQBVTsX3jfLkmh5BaQUhNeRc= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=subject:from:reply-to:to:cc:in-reply-to:references:content-type :organization:date:message-id:mime-version:x-mailer :content-transfer-encoding; b=SztyOCoIUYhf+pYpr8EHn6rcO1tC9uD1JkThLNIo5G2T09G0sgb4RC4ckre1ZfoORZ vUAgqRNfF7TVEXqmLT3l2PZZQc3FgmHShCcjiQtiddJTSbXAgzhQGDfq0CjswfcvlpkW nw/nWDvIHj5GcXiNSq4kXLur5d2RFb/3wg1dI= Received: by 10.223.72.65 with SMTP id l1mr1420425faj.66.1268415559374; Fri, 12 Mar 2010 09:39:19 -0800 (PST) Received: from [202.201.12.142] ([202.201.12.142]) by mx.google.com with ESMTPS id p9sm3407782fkb.3.2010.03.12.09.39.13 (version=TLSv1/SSLv3 cipher=RC4-MD5); Fri, 12 Mar 2010 09:39:18 -0800 (PST) Subject: Re: [PATCH v2] MIPS: tracing: Optimize the implementation From: Wu Zhangjin Reply-To: wuzhangjin@gmail.com To: Ralf Baechle Cc: Steven Rostedt , linux-mips@linux-mips.org, David Daney , Thomas Bogendoerfer In-Reply-To: <0c3597eeee4dbd69eb5a74c2b46179014b291d56.1268414797.git.wuzhangjin@gmail.com> References: <0c3597eeee4dbd69eb5a74c2b46179014b291d56.1268414797.git.wuzhangjin@gmail.com> Content-Type: text/plain; charset="UTF-8" Organization: DSLab, Lanzhou University, China Date: Sat, 13 Mar 2010 01:32:44 +0800 Message-ID: <1268415164.30235.0.camel@falcon> Mime-Version: 1.0 X-Mailer: Evolution 2.28.2 Content-Transfer-Encoding: 7bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26215 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: wuzhangjin@gmail.com Precedence: bulk X-list: linux-mips Please ignore this one, thanks! On Sat, 2010-03-13 at 01:27 +0800, Wu Zhangjin wrote: > From: Wu Zhangjin > > This revision fixes the potential bugs (reported by David) when compiling the > kernel with -mmcount-ra-address of gcc 4.5, I only tested it without this > option for I don't have a gcc 4.5 currently. > > ------------ > > Changes: > > v1 -> v2: > > + Fixup the support of -mmcount-ra-address > > The offset of "b 1f" instruction should be 20 for leaf-function and > 24 for non-leaf function when -mmcount-ra-address is enabled. This > patch adds a new get_insn_b_1f() function to get the different "b > 1f" instruction. > > (only test without -mmcount-ra-address.) > > v1: > > + Reduce the runtime overhead > > o Uses macros instead of variables for the fixed instructions to > reduce memory access > > o Moves the Initilization of the instructions which will be fixed > after linking from ftrace_make_nop/ftrace_make_call to > ftrace_dyn_arch_init() and encodes the instructions through > uasm(arch/mips/include/asm/uasm.h). > > o A common macro in_module() is defined to determine which space the > instruction pointer stays in and several related conditional > statements are converted to conditional operator(? :) statement. > > + Cleanup the whole stuff > > Lots of comments/macros have been cleaned up to let it look better. > > Signed-off-by: Wu Zhangjin > --- > arch/mips/include/asm/ftrace.h | 12 ++- > arch/mips/kernel/ftrace.c | 236 ++++++++++++++++++++++++---------------- > 2 files changed, 153 insertions(+), 95 deletions(-) > > diff --git a/arch/mips/include/asm/ftrace.h b/arch/mips/include/asm/ftrace.h > index ce35c9a..6dbfa1e 100644 > --- a/arch/mips/include/asm/ftrace.h > +++ b/arch/mips/include/asm/ftrace.h > @@ -3,7 +3,7 @@ > * License. See the file "COPYING" in the main directory of this archive for > * more details. > * > - * Copyright (C) 2009 DSLab, Lanzhou University, China > + * Copyright (C) 2009, 2010 DSLab, Lanzhou University, China > * Author: Wu Zhangjin > */ > > @@ -19,6 +19,14 @@ > extern void _mcount(void); > #define mcount _mcount > > +/* > + * If the Instruction Pointer is in module space (0xc0000000), return true; > + * otherwise, it is in kernel space (0x80000000), return false. > + * > + * FIXME: This may not work in some cases. > + */ > +#define in_module(ip) (unlikely((ip) & 0x40000000)) > + > #define safe_load(load, src, dst, error) \ > do { \ > asm volatile ( \ > @@ -83,8 +91,8 @@ static inline unsigned long ftrace_call_adjust(unsigned long addr) > > struct dyn_arch_ftrace { > }; > +#endif /* CONFIG_DYNAMIC_FTRACE */ > > -#endif /* CONFIG_DYNAMIC_FTRACE */ > #endif /* __ASSEMBLY__ */ > #endif /* CONFIG_FUNCTION_TRACER */ > #endif /* _ASM_MIPS_FTRACE_H */ > diff --git a/arch/mips/kernel/ftrace.c b/arch/mips/kernel/ftrace.c > index e9e64e0..b72107e 100644 > --- a/arch/mips/kernel/ftrace.c > +++ b/arch/mips/kernel/ftrace.c > @@ -2,7 +2,7 @@ > * Code for replacing ftrace calls with jumps. > * > * Copyright (C) 2007-2008 Steven Rostedt > - * Copyright (C) 2009 DSLab, Lanzhou University, China > + * Copyright (C) 2009, 2010 DSLab, Lanzhou University, China > * Author: Wu Zhangjin > * > * Thanks goes to Steven Rostedt for writing the original x86 version. > @@ -15,15 +15,48 @@ > #include > #include > #include > +#include > > #ifdef CONFIG_DYNAMIC_FTRACE > > -#define JAL 0x0c000000 /* jump & link: ip --> ra, jump to target */ > -#define ADDR_MASK 0x03ffffff /* op_code|addr : 31...26|25 ....0 */ > -#define jump_insn_encode(op_code, addr) \ > - ((unsigned int)((op_code) | (((addr) >> 2) & ADDR_MASK))) > +/* Before linking, the following instructions are fixed. */ > +#ifdef CONFIG_CPU_LOONGSON2F > +#define INSN_NOP 0x00200825 /* or at, at, zero */ > +#else > +#define INSN_NOP 0x00000000 /* nop */ > +#endif > +#define INSN_B_1F_16 0x10000004 /* b 1f; offset = (16 >> 2) */ > +#define INSN_B_1F_20 0x10000005 /* b 1f; offset = (20 >> 2) */ > +#define INSN_B_1F_24 0x10000006 /* b 1f; offset = (24 >> 2) */ > + > +/* After linking, the following instructions are fixed. */ > +static unsigned int insn_jal_ftrace_caller __read_mostly; > +static unsigned int insn_lui_v1_hi16_mcount __read_mostly; > +static unsigned int insn_j_ftrace_graph_caller __read_mostly; > > -static unsigned int ftrace_nop = 0x00000000; > +/* The following instructions are encoded in the run-time */ > +/* insn: jal func; op_code|addr : 31...26|25 ....0 */ > +#define INSN_JAL(addr) \ > + ((unsigned int)(0x0c000000 | (((addr) >> 2) & 0x03ffffff))) > + > +static inline void ftrace_dyn_arch_init_insns(void) > +{ > + u32 *buf; > + unsigned int v1; > + > + /* lui v1, hi16_mcount */ > + v1 = 3; > + buf = (u32 *)&insn_lui_v1_hi16_mcount; > + UASM_i_LA_mostly(&buf, v1, MCOUNT_ADDR); > + > + /* jal (ftrace_caller + 8), jump over the first two instruction */ > + buf = (u32 *)&insn_jal_ftrace_caller; > + uasm_i_jal(&buf, (FTRACE_ADDR + 8)); > + > + /* j ftrace_graph_caller */ > + buf = (u32 *)&insn_j_ftrace_graph_caller; > + uasm_i_j(&buf, (unsigned long)ftrace_graph_caller); > +} > > static int ftrace_modify_code(unsigned long ip, unsigned int new_code) > { > @@ -31,7 +64,6 @@ static int ftrace_modify_code(unsigned long ip, unsigned int new_code) > > /* *(unsigned int *)ip = new_code; */ > safe_store_code(new_code, ip, faulted); > - > if (unlikely(faulted)) > return -EFAULT; > > @@ -40,118 +72,133 @@ static int ftrace_modify_code(unsigned long ip, unsigned int new_code) > return 0; > } > > -static int lui_v1; > -static int jal_mcount; > +static inline unsigned int get_insn_b_1f(unsigned int *insn, unsigned long ip) > +{ > +#ifndef KBUILD_MCOUNT_RA_ADDRESS > + *insn = INSN_B_1F_16; > + return 0; > +#else > + int faulted; > + unsigned int code; > + > + ip -= 4; > + > + /* code = *(unsigned int *)ip; */ > + safe_load_code(code, ip, faulted); > + if (unlikely(faulted)) > + return -EFAULT; > + > + if ((code & S_R_SP) != S_R_SP) > + *insn = INSN_B_1F_20; > + else > + *insn = INSN_B_1F_24; > + > + return 0; > +#endif > +} > > int ftrace_make_nop(struct module *mod, > struct dyn_ftrace *rec, unsigned long addr) > { > unsigned int new; > - int faulted; > unsigned long ip = rec->ip; > > - /* We have compiled module with -mlong-calls, but compiled the kernel > - * without it, we need to cope with them respectively. */ > - if (ip & 0x40000000) { > - /* record it for ftrace_make_call */ > - if (lui_v1 == 0) { > - /* lui_v1 = *(unsigned int *)ip; */ > - safe_load_code(lui_v1, ip, faulted); > - > - if (unlikely(faulted)) > - return -EFAULT; > - } > - > - /* lui v1, hi_16bit_of_mcount --> b 1f (0x10000004) > - * addiu v1, v1, low_16bit_of_mcount > - * move at, ra > - * jalr v1 > - * nop > - * 1f: (ip + 12) > - */ > - new = 0x10000004; > - } else { > - /* record/calculate it for ftrace_make_call */ > - if (jal_mcount == 0) { > - /* We can record it directly like this: > - * jal_mcount = *(unsigned int *)ip; > - * Herein, jump over the first two nop instructions */ > - jal_mcount = jump_insn_encode(JAL, (MCOUNT_ADDR + 8)); > - } > - > - /* move at, ra > - * jalr v1 --> nop > - */ > - new = ftrace_nop; > - } > + /* > + * We have compiled modules with -mlong-calls, but compiled kernel > + * without it, therefore, need to cope with them respectively. > + * > + * For module: > + * > + * lui v1, hi16_mcount --> b 1f > + * addiu v1, v1, lo16_mcount > + * move at, ra > + * jalr v1 > + * nop > + * 1f: (ip + 16) > + * For kernel: > + * > + * move at, ra > + * jal _mcount --> nop > + * > + * And with the -mmcount-ra-address option, the offset may be 20 for > + * leaf fuction and 24 for non-leaf function. > + */ > + > + if (!in_module(ip)) > + new = INSN_NOP; > + else if (get_insn_b_1f(&new, ip)) > + return -EFAULT; > + > return ftrace_modify_code(ip, new); > } > > -static int modified; /* initialized as 0 by default */ > - > int ftrace_make_call(struct dyn_ftrace *rec, unsigned long addr) > { > unsigned int new; > unsigned long ip = rec->ip; > > - /* We just need to remove the "b ftrace_stub" at the fist time! */ > - if (modified == 0) { > - modified = 1; > - ftrace_modify_code(addr, ftrace_nop); > - } > - /* ip, module: 0xc0000000, kernel: 0x80000000 */ > - new = (ip & 0x40000000) ? lui_v1 : jal_mcount; > + new = in_module(ip) ? insn_lui_v1_hi16_mcount : insn_jal_ftrace_caller; > > return ftrace_modify_code(ip, new); > } > > -#define FTRACE_CALL_IP ((unsigned long)(&ftrace_call)) > - > int ftrace_update_ftrace_func(ftrace_func_t func) > { > unsigned int new; > > - new = jump_insn_encode(JAL, (unsigned long)func); > + new = INSN_JAL((unsigned long)func); > > - return ftrace_modify_code(FTRACE_CALL_IP, new); > + return ftrace_modify_code((unsigned long)(&ftrace_call), new); > } > > int __init ftrace_dyn_arch_init(void *data) > { > + ftrace_dyn_arch_init_insns(); > + > + /* > + * We are safe to remove the "b ftrace_stub" for the current > + * ftrace_caller() is almost empty (only the stack operations), and > + * most importantly, the calling to mcount will be disabled later in > + * ftrace_init(), then there is no 'big' overhead. And in the future, > + * we are hoping the function_trace_stop is initialized as 1 then we > + * can eventually remove that 'useless' "b ftrace_stub" and the > + * following nop. Currently, although we can call ftrace_stop() to set > + * function_trace_stop as 1, but it will change the behavior of the > + * Function Tracer. > + */ > + ftrace_modify_code(MCOUNT_ADDR, INSN_NOP); > + > /* The return code is retured via data */ > *(unsigned long *)data = 0; > > return 0; > } > -#endif /* CONFIG_DYNAMIC_FTRACE */ > +#endif /* CONFIG_DYNAMIC_FTRACE */ > > #ifdef CONFIG_FUNCTION_GRAPH_TRACER > > #ifdef CONFIG_DYNAMIC_FTRACE > > extern void ftrace_graph_call(void); > -#define JMP 0x08000000 /* jump to target directly */ > -#define CALL_FTRACE_GRAPH_CALLER \ > - jump_insn_encode(JMP, (unsigned long)(&ftrace_graph_caller)) > #define FTRACE_GRAPH_CALL_IP ((unsigned long)(&ftrace_graph_call)) > > int ftrace_enable_ftrace_graph_caller(void) > { > return ftrace_modify_code(FTRACE_GRAPH_CALL_IP, > - CALL_FTRACE_GRAPH_CALLER); > + insn_j_ftrace_graph_caller); > } > > int ftrace_disable_ftrace_graph_caller(void) > { > - return ftrace_modify_code(FTRACE_GRAPH_CALL_IP, ftrace_nop); > + return ftrace_modify_code(FTRACE_GRAPH_CALL_IP, INSN_NOP); > } > > -#endif /* !CONFIG_DYNAMIC_FTRACE */ > +#endif /* !CONFIG_DYNAMIC_FTRACE */ > > #ifndef KBUILD_MCOUNT_RA_ADDRESS > -#define S_RA_SP (0xafbf << 16) /* s{d,w} ra, offset(sp) */ > -#define S_R_SP (0xafb0 << 16) /* s{d,w} R, offset(sp) */ > -#define OFFSET_MASK 0xffff /* stack offset range: 0 ~ PT_SIZE */ > +#define INSN_S_RA_SP 0xafbf0000 /* s{d,w} ra, offset(sp) */ > +#define INSN_S_R_SP 0xafb00000 /* s{d,w} R, offset(sp) */ > +#define STACK_OFFSET_MASK 0xffff /* stack offset range: 0 ~ PT_SIZE */ > > unsigned long ftrace_get_parent_addr(unsigned long self_addr, > unsigned long parent, > @@ -162,35 +209,35 @@ unsigned long ftrace_get_parent_addr(unsigned long self_addr, > unsigned int code; > int faulted; > > - /* in module or kernel? */ > - if (self_addr & 0x40000000) { > - /* module: move to the instruction "lui v1, HI_16BIT_OF_MCOUNT" */ > - ip = self_addr - 20; > - } else { > - /* kernel: move to the instruction "move ra, at" */ > - ip = self_addr - 12; > - } > + /* > + * For module, move the ip from calling site of mcount to the > + * instruction "lui v1, hi_16bit_of_mcount"(offset is 20), but for > + * kernel, move to the instruction "move ra, at"(offset is 12) > + */ > + ip = self_addr - (in_module(self_addr) ? 20 : 12); > > - /* search the text until finding the non-store instruction or "s{d,w} > - * ra, offset(sp)" instruction */ > + /* > + * search the text until finding the non-store instruction or "s{d,w} > + * ra, offset(sp)" instruction > + */ > do { > ip -= 4; > > /* get the code at "ip": code = *(unsigned int *)ip; */ > safe_load_code(code, ip, faulted); > - > if (unlikely(faulted)) > return 0; > > - /* If we hit the non-store instruction before finding where the > + /* > + * If we hit the non-store instruction before finding where the > * ra is stored, then this is a leaf function and it does not > - * store the ra on the stack. */ > - if ((code & S_R_SP) != S_R_SP) > + * store the ra on the stack. > + */ > + if ((code & INSN_S_R_SP) != INSN_S_R_SP) > return parent_addr; > + } while (((code & INSN_S_RA_SP) != INSN_S_RA_SP)); > > - } while (((code & S_RA_SP) != S_RA_SP)); > - > - sp = fp + (code & OFFSET_MASK); > + sp = fp + (code & STACK_OFFSET_MASK); > > /* ra = *(unsigned long *)sp; */ > safe_load_stack(ra, sp, faulted); > @@ -211,16 +258,17 @@ unsigned long ftrace_get_parent_addr(unsigned long self_addr, > void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr, > unsigned long fp) > { > + int faulted; > unsigned long old; > struct ftrace_graph_ent trace; > unsigned long return_hooker = (unsigned long) > &return_to_handler; > - int faulted; > > if (unlikely(atomic_read(¤t->tracing_graph_pause))) > return; > > - /* "parent" is the stack address saved the return address of the caller > + /* > + * "parent" is the stack address saved the return address of the caller > * of _mcount. > * > * if the gcc < 4.5, a leaf function does not save the return address > @@ -241,12 +289,14 @@ void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr, > if (unlikely(faulted)) > goto out; > #ifndef KBUILD_MCOUNT_RA_ADDRESS > - parent = (unsigned long *)ftrace_get_parent_addr(self_addr, old, > - (unsigned long)parent, > - fp); > - /* If fails when getting the stack address of the non-leaf function's > - * ra, stop function graph tracer and return */ > - if (parent == 0) > + parent = (unsigned long *)ftrace_get_parent_addr( > + self_addr, old, (unsigned long)parent, fp); > + > + /* > + * If fails on getting the stack address of the non-leaf function's ra, > + * stop function graph tracer and return > + */ > + if (unlikely(parent == 0)) > goto out; > #endif > /* *parent = return_hooker; */ > @@ -272,4 +322,4 @@ out: > ftrace_graph_stop(); > WARN_ON(1); > } > -#endif /* CONFIG_FUNCTION_GRAPH_TRACER */ > +#endif /* CONFIG_FUNCTION_GRAPH_TRACER */ From wuzhangjin@gmail.com Fri Mar 12 18:39:48 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Fri, 12 Mar 2010 18:39:55 +0100 (CET) Received: from mail-fx0-f217.google.com ([209.85.220.217]:62085 "EHLO mail-fx0-f217.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491871Ab0CLRjn (ORCPT ); Fri, 12 Mar 2010 18:39:43 +0100 Received: by mail-fx0-f217.google.com with SMTP id 9so1418987fxm.24 for ; Fri, 12 Mar 2010 09:39:43 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:from:to:cc:subject:date :message-id:x-mailer; bh=yFKkRFNsbGA2/7mDSE9LppsAuMaggE/vrTxhrHi3uDo=; b=dzjIiYPORjIxMQtWZuWpnFCESgpfuhihlRdGknDPUSa3Aedxc7m8Ovy0R9wwFtWBk4 NsyRj54gDPKSR9h+CD1WNlcTORuk3+qQ4mpM4ZBXUdXVe/GIxUTHZkZKSINnmFf6S1sj s68V1ezfOvJLRjyGUbjPEm9f1bR2Snc4vJnis= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=from:to:cc:subject:date:message-id:x-mailer; b=CdwtIxHUOZmEtuoB7CAItYvFfeYg2RRslQCraSxw/v4VApGX/q+qfuKdby4wHejskv foJ2ev4eXgCfm97wyr3tIyvvqBZFYNesDhhb8LGbAcX3g5Yz91+Pvh9c0qbMx9pDVTqz jsXjHqxUrzsx0nHNaTmAD9MoPIi/Dia6mZH5M= Received: by 10.223.143.82 with SMTP id t18mr1376850fau.52.1268415583505; Fri, 12 Mar 2010 09:39:43 -0800 (PST) Received: from localhost.localdomain ([202.201.12.142]) by mx.google.com with ESMTPS id 35sm3410302fkt.7.2010.03.12.09.39.37 (version=TLSv1/SSLv3 cipher=RC4-MD5); Fri, 12 Mar 2010 09:39:42 -0800 (PST) From: Wu Zhangjin To: Ralf Baechle Cc: Steven Rostedt , linux-mips@linux-mips.org, David Daney , Thomas Bogendoerfer , Wu Zhangjin Subject: [PATCH v2] MIPS: tracing: Optimize the implementation Date: Sat, 13 Mar 2010 01:32:14 +0800 Message-Id: <33db7c603507d884eb4e3a27f84deb7eb4c016e8.1268415127.git.wuzhangjin@gmail.com> X-Mailer: git-send-email 1.7.0.1 Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26216 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: wuzhangjin@gmail.com Precedence: bulk X-list: linux-mips From: Wu Zhangjin Changes: v2 -> v3: + Fixes the buidling error of the changes from v2. v1 -> v2: + Fixup the support of -mmcount-ra-address The offset of "b 1f" instruction should be 20 for leaf-function and 24 for non-leaf function when -mmcount-ra-address is enabled. This patch adds a new get_insn_b_1f() function to get the different "b 1f" instruction. (only test without -mmcount-ra-address.) v1: + Reduce the runtime overhead o Uses macros instead of variables for the fixed instructions to reduce memory access o Moves the Initilization of the instructions which will be fixed after linking from ftrace_make_nop/ftrace_make_call to ftrace_dyn_arch_init() and encodes the instructions through uasm(arch/mips/include/asm/uasm.h). o A common macro in_module() is defined to determine which space the instruction pointer stays in and several related conditional statements are converted to conditional operator(? :) statement. + Cleanup the whole stuff Lots of comments/macros have been cleaned up to let it look better. Signed-off-by: Wu Zhangjin --- arch/mips/include/asm/ftrace.h | 12 ++- arch/mips/kernel/ftrace.c | 237 ++++++++++++++++++++++++---------------- 2 files changed, 154 insertions(+), 95 deletions(-) diff --git a/arch/mips/include/asm/ftrace.h b/arch/mips/include/asm/ftrace.h index ce35c9a..6dbfa1e 100644 --- a/arch/mips/include/asm/ftrace.h +++ b/arch/mips/include/asm/ftrace.h @@ -3,7 +3,7 @@ * License. See the file "COPYING" in the main directory of this archive for * more details. * - * Copyright (C) 2009 DSLab, Lanzhou University, China + * Copyright (C) 2009, 2010 DSLab, Lanzhou University, China * Author: Wu Zhangjin */ @@ -19,6 +19,14 @@ extern void _mcount(void); #define mcount _mcount +/* + * If the Instruction Pointer is in module space (0xc0000000), return true; + * otherwise, it is in kernel space (0x80000000), return false. + * + * FIXME: This may not work in some cases. + */ +#define in_module(ip) (unlikely((ip) & 0x40000000)) + #define safe_load(load, src, dst, error) \ do { \ asm volatile ( \ @@ -83,8 +91,8 @@ static inline unsigned long ftrace_call_adjust(unsigned long addr) struct dyn_arch_ftrace { }; +#endif /* CONFIG_DYNAMIC_FTRACE */ -#endif /* CONFIG_DYNAMIC_FTRACE */ #endif /* __ASSEMBLY__ */ #endif /* CONFIG_FUNCTION_TRACER */ #endif /* _ASM_MIPS_FTRACE_H */ diff --git a/arch/mips/kernel/ftrace.c b/arch/mips/kernel/ftrace.c index e9e64e0..88e05ae 100644 --- a/arch/mips/kernel/ftrace.c +++ b/arch/mips/kernel/ftrace.c @@ -2,7 +2,7 @@ * Code for replacing ftrace calls with jumps. * * Copyright (C) 2007-2008 Steven Rostedt - * Copyright (C) 2009 DSLab, Lanzhou University, China + * Copyright (C) 2009, 2010 DSLab, Lanzhou University, China * Author: Wu Zhangjin * * Thanks goes to Steven Rostedt for writing the original x86 version. @@ -15,15 +15,50 @@ #include #include #include +#include + +#define INSN_S_R_SP 0xafb00000 /* s{d,w} R, offset(sp) */ #ifdef CONFIG_DYNAMIC_FTRACE -#define JAL 0x0c000000 /* jump & link: ip --> ra, jump to target */ -#define ADDR_MASK 0x03ffffff /* op_code|addr : 31...26|25 ....0 */ -#define jump_insn_encode(op_code, addr) \ - ((unsigned int)((op_code) | (((addr) >> 2) & ADDR_MASK))) +/* Before linking, the following instructions are fixed. */ +#ifdef CONFIG_CPU_LOONGSON2F +#define INSN_NOP 0x00200825 /* or at, at, zero */ +#else +#define INSN_NOP 0x00000000 /* nop */ +#endif +#define INSN_B_1F_16 0x10000004 /* b 1f; offset = (16 >> 2) */ +#define INSN_B_1F_20 0x10000005 /* b 1f; offset = (20 >> 2) */ +#define INSN_B_1F_24 0x10000006 /* b 1f; offset = (24 >> 2) */ + +/* After linking, the following instructions are fixed. */ +static unsigned int insn_jal_ftrace_caller __read_mostly; +static unsigned int insn_lui_v1_hi16_mcount __read_mostly; +static unsigned int insn_j_ftrace_graph_caller __read_mostly; + +/* The following instructions are encoded in the run-time */ +/* insn: jal func; op_code|addr : 31...26|25 ....0 */ +#define INSN_JAL(addr) \ + ((unsigned int)(0x0c000000 | (((addr) >> 2) & 0x03ffffff))) + +static inline void ftrace_dyn_arch_init_insns(void) +{ + u32 *buf; + unsigned int v1; + + /* lui v1, hi16_mcount */ + v1 = 3; + buf = (u32 *)&insn_lui_v1_hi16_mcount; + UASM_i_LA_mostly(&buf, v1, MCOUNT_ADDR); -static unsigned int ftrace_nop = 0x00000000; + /* jal (ftrace_caller + 8), jump over the first two instruction */ + buf = (u32 *)&insn_jal_ftrace_caller; + uasm_i_jal(&buf, (FTRACE_ADDR + 8)); + + /* j ftrace_graph_caller */ + buf = (u32 *)&insn_j_ftrace_graph_caller; + uasm_i_j(&buf, (unsigned long)ftrace_graph_caller); +} static int ftrace_modify_code(unsigned long ip, unsigned int new_code) { @@ -31,7 +66,6 @@ static int ftrace_modify_code(unsigned long ip, unsigned int new_code) /* *(unsigned int *)ip = new_code; */ safe_store_code(new_code, ip, faulted); - if (unlikely(faulted)) return -EFAULT; @@ -40,118 +74,132 @@ static int ftrace_modify_code(unsigned long ip, unsigned int new_code) return 0; } -static int lui_v1; -static int jal_mcount; +static inline unsigned int get_insn_b_1f(unsigned int *insn, unsigned long ip) +{ +#ifndef KBUILD_MCOUNT_RA_ADDRESS + *insn = INSN_B_1F_16; + return 0; +#else + int faulted; + unsigned int code; + + ip -= 4; + + /* code = *(unsigned int *)ip; */ + safe_load_code(code, ip, faulted); + if (unlikely(faulted)) + return -EFAULT; + + if ((code & INSN_S_R_SP) != INSN_S_R_SP) + *insn = INSN_B_1F_20; + else + *insn = INSN_B_1F_24; + + return 0; +#endif +} int ftrace_make_nop(struct module *mod, struct dyn_ftrace *rec, unsigned long addr) { unsigned int new; - int faulted; unsigned long ip = rec->ip; - /* We have compiled module with -mlong-calls, but compiled the kernel - * without it, we need to cope with them respectively. */ - if (ip & 0x40000000) { - /* record it for ftrace_make_call */ - if (lui_v1 == 0) { - /* lui_v1 = *(unsigned int *)ip; */ - safe_load_code(lui_v1, ip, faulted); - - if (unlikely(faulted)) - return -EFAULT; - } - - /* lui v1, hi_16bit_of_mcount --> b 1f (0x10000004) - * addiu v1, v1, low_16bit_of_mcount - * move at, ra - * jalr v1 - * nop - * 1f: (ip + 12) - */ - new = 0x10000004; - } else { - /* record/calculate it for ftrace_make_call */ - if (jal_mcount == 0) { - /* We can record it directly like this: - * jal_mcount = *(unsigned int *)ip; - * Herein, jump over the first two nop instructions */ - jal_mcount = jump_insn_encode(JAL, (MCOUNT_ADDR + 8)); - } - - /* move at, ra - * jalr v1 --> nop - */ - new = ftrace_nop; - } + /* + * We have compiled modules with -mlong-calls, but compiled kernel + * without it, therefore, need to cope with them respectively. + * + * For module: + * + * lui v1, hi16_mcount --> b 1f + * addiu v1, v1, lo16_mcount + * move at, ra + * jalr v1 + * nop + * 1f: (ip + 16) + * For kernel: + * + * move at, ra + * jal _mcount --> nop + * + * And with the -mmcount-ra-address option, the offset may be 20 for + * leaf fuction and 24 for non-leaf function. + */ + + if (!in_module(ip)) + new = INSN_NOP; + else if (get_insn_b_1f(&new, ip)) + return -EFAULT; + return ftrace_modify_code(ip, new); } -static int modified; /* initialized as 0 by default */ - int ftrace_make_call(struct dyn_ftrace *rec, unsigned long addr) { unsigned int new; unsigned long ip = rec->ip; - /* We just need to remove the "b ftrace_stub" at the fist time! */ - if (modified == 0) { - modified = 1; - ftrace_modify_code(addr, ftrace_nop); - } - /* ip, module: 0xc0000000, kernel: 0x80000000 */ - new = (ip & 0x40000000) ? lui_v1 : jal_mcount; + new = in_module(ip) ? insn_lui_v1_hi16_mcount : insn_jal_ftrace_caller; return ftrace_modify_code(ip, new); } -#define FTRACE_CALL_IP ((unsigned long)(&ftrace_call)) - int ftrace_update_ftrace_func(ftrace_func_t func) { unsigned int new; - new = jump_insn_encode(JAL, (unsigned long)func); + new = INSN_JAL((unsigned long)func); - return ftrace_modify_code(FTRACE_CALL_IP, new); + return ftrace_modify_code((unsigned long)(&ftrace_call), new); } int __init ftrace_dyn_arch_init(void *data) { + ftrace_dyn_arch_init_insns(); + + /* + * We are safe to remove the "b ftrace_stub" for the current + * ftrace_caller() is almost empty (only the stack operations), and + * most importantly, the calling to mcount will be disabled later in + * ftrace_init(), then there is no 'big' overhead. And in the future, + * we are hoping the function_trace_stop is initialized as 1 then we + * can eventually remove that 'useless' "b ftrace_stub" and the + * following nop. Currently, although we can call ftrace_stop() to set + * function_trace_stop as 1, but it will change the behavior of the + * Function Tracer. + */ + ftrace_modify_code(MCOUNT_ADDR, INSN_NOP); + /* The return code is retured via data */ *(unsigned long *)data = 0; return 0; } -#endif /* CONFIG_DYNAMIC_FTRACE */ +#endif /* CONFIG_DYNAMIC_FTRACE */ #ifdef CONFIG_FUNCTION_GRAPH_TRACER #ifdef CONFIG_DYNAMIC_FTRACE extern void ftrace_graph_call(void); -#define JMP 0x08000000 /* jump to target directly */ -#define CALL_FTRACE_GRAPH_CALLER \ - jump_insn_encode(JMP, (unsigned long)(&ftrace_graph_caller)) #define FTRACE_GRAPH_CALL_IP ((unsigned long)(&ftrace_graph_call)) int ftrace_enable_ftrace_graph_caller(void) { return ftrace_modify_code(FTRACE_GRAPH_CALL_IP, - CALL_FTRACE_GRAPH_CALLER); + insn_j_ftrace_graph_caller); } int ftrace_disable_ftrace_graph_caller(void) { - return ftrace_modify_code(FTRACE_GRAPH_CALL_IP, ftrace_nop); + return ftrace_modify_code(FTRACE_GRAPH_CALL_IP, INSN_NOP); } -#endif /* !CONFIG_DYNAMIC_FTRACE */ +#endif /* !CONFIG_DYNAMIC_FTRACE */ #ifndef KBUILD_MCOUNT_RA_ADDRESS -#define S_RA_SP (0xafbf << 16) /* s{d,w} ra, offset(sp) */ -#define S_R_SP (0xafb0 << 16) /* s{d,w} R, offset(sp) */ -#define OFFSET_MASK 0xffff /* stack offset range: 0 ~ PT_SIZE */ +#define INSN_S_RA_SP 0xafbf0000 /* s{d,w} ra, offset(sp) */ +#define STACK_OFFSET_MASK 0xffff /* stack offset range: 0 ~ PT_SIZE */ unsigned long ftrace_get_parent_addr(unsigned long self_addr, unsigned long parent, @@ -162,35 +210,35 @@ unsigned long ftrace_get_parent_addr(unsigned long self_addr, unsigned int code; int faulted; - /* in module or kernel? */ - if (self_addr & 0x40000000) { - /* module: move to the instruction "lui v1, HI_16BIT_OF_MCOUNT" */ - ip = self_addr - 20; - } else { - /* kernel: move to the instruction "move ra, at" */ - ip = self_addr - 12; - } + /* + * For module, move the ip from calling site of mcount to the + * instruction "lui v1, hi_16bit_of_mcount"(offset is 20), but for + * kernel, move to the instruction "move ra, at"(offset is 12) + */ + ip = self_addr - (in_module(self_addr) ? 20 : 12); - /* search the text until finding the non-store instruction or "s{d,w} - * ra, offset(sp)" instruction */ + /* + * search the text until finding the non-store instruction or "s{d,w} + * ra, offset(sp)" instruction + */ do { ip -= 4; /* get the code at "ip": code = *(unsigned int *)ip; */ safe_load_code(code, ip, faulted); - if (unlikely(faulted)) return 0; - /* If we hit the non-store instruction before finding where the + /* + * If we hit the non-store instruction before finding where the * ra is stored, then this is a leaf function and it does not - * store the ra on the stack. */ - if ((code & S_R_SP) != S_R_SP) + * store the ra on the stack. + */ + if ((code & INSN_S_R_SP) != INSN_S_R_SP) return parent_addr; + } while (((code & INSN_S_RA_SP) != INSN_S_RA_SP)); - } while (((code & S_RA_SP) != S_RA_SP)); - - sp = fp + (code & OFFSET_MASK); + sp = fp + (code & STACK_OFFSET_MASK); /* ra = *(unsigned long *)sp; */ safe_load_stack(ra, sp, faulted); @@ -211,16 +259,17 @@ unsigned long ftrace_get_parent_addr(unsigned long self_addr, void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr, unsigned long fp) { + int faulted; unsigned long old; struct ftrace_graph_ent trace; unsigned long return_hooker = (unsigned long) &return_to_handler; - int faulted; if (unlikely(atomic_read(¤t->tracing_graph_pause))) return; - /* "parent" is the stack address saved the return address of the caller + /* + * "parent" is the stack address saved the return address of the caller * of _mcount. * * if the gcc < 4.5, a leaf function does not save the return address @@ -241,12 +290,14 @@ void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr, if (unlikely(faulted)) goto out; #ifndef KBUILD_MCOUNT_RA_ADDRESS - parent = (unsigned long *)ftrace_get_parent_addr(self_addr, old, - (unsigned long)parent, - fp); - /* If fails when getting the stack address of the non-leaf function's - * ra, stop function graph tracer and return */ - if (parent == 0) + parent = (unsigned long *)ftrace_get_parent_addr( + self_addr, old, (unsigned long)parent, fp); + + /* + * If fails on getting the stack address of the non-leaf function's ra, + * stop function graph tracer and return + */ + if (unlikely(parent == 0)) goto out; #endif /* *parent = return_hooker; */ @@ -272,4 +323,4 @@ out: ftrace_graph_stop(); WARN_ON(1); } -#endif /* CONFIG_FUNCTION_GRAPH_TRACER */ +#endif /* CONFIG_FUNCTION_GRAPH_TRACER */ -- 1.7.0.1 From wuzhangjin@gmail.com Fri Mar 12 18:46:09 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Fri, 12 Mar 2010 18:46:17 +0100 (CET) Received: from mail-fx0-f217.google.com ([209.85.220.217]:58675 "EHLO mail-fx0-f217.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492012Ab0CLRqJ (ORCPT ); Fri, 12 Mar 2010 18:46:09 +0100 Received: by fxm9 with SMTP id 9so1425630fxm.24 for ; Fri, 12 Mar 2010 09:46:03 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:subject:from:reply-to:to:cc :in-reply-to:references:content-type:organization:date:message-id :mime-version:x-mailer:content-transfer-encoding; bh=zVEkbCobBo6yhl1Ev3guW2cFnQflrEqtrMn3YcXZ2e4=; b=DJazElKHRPaqDIA+XHPfrWFD69B/oMFjTT10uizk6wI3u4IHrkJgGbgw27yRDqdQe5 S8Pu8eUwdUiv+OKB4ZfEJZeNh7xm1DS0F3gNV+l4lFPNVmneqHOb7aSLCgOxlKEyzH8U SwvjQJS1Cvhx2cD+2JxxMbq+D19u685QW+n30= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=subject:from:reply-to:to:cc:in-reply-to:references:content-type :organization:date:message-id:mime-version:x-mailer :content-transfer-encoding; b=KVttlCbtgSQqH33+1ILZb2DYpq59h5n7WWWb9E/eGa+YvTcjI86qdKfaWD0bouFSfH 5ynBnhVIHI17JV8S0jm2pQjjpvCZKZdFB9n4di1R8hvaTSZM3VHHE+ZgTwdamY7qnlOl nNhDkdWT3y1xfe1z1DR/yI5XTAJ1lDXns9sqc= Received: by 10.223.143.82 with SMTP id t18mr1385126fau.52.1268415963117; Fri, 12 Mar 2010 09:46:03 -0800 (PST) Received: from [202.201.12.142] ([202.201.12.142]) by mx.google.com with ESMTPS id k29sm3084535fkk.15.2010.03.12.09.45.57 (version=TLSv1/SSLv3 cipher=RC4-MD5); Fri, 12 Mar 2010 09:46:01 -0800 (PST) Subject: Re: [PATCH v3] MIPS: tracing: Optimize the implementation From: Wu Zhangjin Reply-To: wuzhangjin@gmail.com To: Ralf Baechle Cc: Steven Rostedt , linux-mips@linux-mips.org, David Daney , Thomas Bogendoerfer In-Reply-To: <33db7c603507d884eb4e3a27f84deb7eb4c016e8.1268415127.git.wuzhangjin@gmail.com> References: <33db7c603507d884eb4e3a27f84deb7eb4c016e8.1268415127.git.wuzhangjin@gmail.com> Content-Type: text/plain; charset="UTF-8" Organization: DSLab, Lanzhou University, China Date: Sat, 13 Mar 2010 01:39:29 +0800 Message-ID: <1268415569.20587.3.camel@falcon> Mime-Version: 1.0 X-Mailer: Evolution 2.28.2 Content-Transfer-Encoding: 7bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26217 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: wuzhangjin@gmail.com Precedence: bulk X-list: linux-mips Hi, This one fixed the building error ;) Regards, Wu Zhangjin On Sat, 2010-03-13 at 01:32 +0800, Wu Zhangjin wrote: > From: Wu Zhangjin > > Changes: > > v2 -> v3: > > + Fixes the buidling error of the changes from v2. > > v1 -> v2: > > + Fixup the support of -mmcount-ra-address > > The offset of "b 1f" instruction should be 20 for leaf-function and > 24 for non-leaf function when -mmcount-ra-address is enabled. This > patch adds a new get_insn_b_1f() function to get the different "b > 1f" instruction. > > (only test without -mmcount-ra-address.) > > v1: > > + Reduce the runtime overhead > > o Uses macros instead of variables for the fixed instructions to > reduce memory access > > o Moves the Initilization of the instructions which will be fixed > after linking from ftrace_make_nop/ftrace_make_call to > ftrace_dyn_arch_init() and encodes the instructions through > uasm(arch/mips/include/asm/uasm.h). > > o A common macro in_module() is defined to determine which space the > instruction pointer stays in and several related conditional > statements are converted to conditional operator(? :) statement. > > + Cleanup the whole stuff > > Lots of comments/macros have been cleaned up to let it look better. > > Signed-off-by: Wu Zhangjin > --- > arch/mips/include/asm/ftrace.h | 12 ++- > arch/mips/kernel/ftrace.c | 237 ++++++++++++++++++++++++---------------- > 2 files changed, 154 insertions(+), 95 deletions(-) > > diff --git a/arch/mips/include/asm/ftrace.h b/arch/mips/include/asm/ftrace.h > index ce35c9a..6dbfa1e 100644 > --- a/arch/mips/include/asm/ftrace.h > +++ b/arch/mips/include/asm/ftrace.h > @@ -3,7 +3,7 @@ > * License. See the file "COPYING" in the main directory of this archive for > * more details. > * > - * Copyright (C) 2009 DSLab, Lanzhou University, China > + * Copyright (C) 2009, 2010 DSLab, Lanzhou University, China > * Author: Wu Zhangjin > */ > > @@ -19,6 +19,14 @@ > extern void _mcount(void); > #define mcount _mcount > > +/* > + * If the Instruction Pointer is in module space (0xc0000000), return true; > + * otherwise, it is in kernel space (0x80000000), return false. > + * > + * FIXME: This may not work in some cases. > + */ > +#define in_module(ip) (unlikely((ip) & 0x40000000)) > + > #define safe_load(load, src, dst, error) \ > do { \ > asm volatile ( \ > @@ -83,8 +91,8 @@ static inline unsigned long ftrace_call_adjust(unsigned long addr) > > struct dyn_arch_ftrace { > }; > +#endif /* CONFIG_DYNAMIC_FTRACE */ > > -#endif /* CONFIG_DYNAMIC_FTRACE */ > #endif /* __ASSEMBLY__ */ > #endif /* CONFIG_FUNCTION_TRACER */ > #endif /* _ASM_MIPS_FTRACE_H */ > diff --git a/arch/mips/kernel/ftrace.c b/arch/mips/kernel/ftrace.c > index e9e64e0..88e05ae 100644 > --- a/arch/mips/kernel/ftrace.c > +++ b/arch/mips/kernel/ftrace.c > @@ -2,7 +2,7 @@ > * Code for replacing ftrace calls with jumps. > * > * Copyright (C) 2007-2008 Steven Rostedt > - * Copyright (C) 2009 DSLab, Lanzhou University, China > + * Copyright (C) 2009, 2010 DSLab, Lanzhou University, China > * Author: Wu Zhangjin > * > * Thanks goes to Steven Rostedt for writing the original x86 version. > @@ -15,15 +15,50 @@ > #include > #include > #include > +#include > + > +#define INSN_S_R_SP 0xafb00000 /* s{d,w} R, offset(sp) */ > > #ifdef CONFIG_DYNAMIC_FTRACE > > -#define JAL 0x0c000000 /* jump & link: ip --> ra, jump to target */ > -#define ADDR_MASK 0x03ffffff /* op_code|addr : 31...26|25 ....0 */ > -#define jump_insn_encode(op_code, addr) \ > - ((unsigned int)((op_code) | (((addr) >> 2) & ADDR_MASK))) > +/* Before linking, the following instructions are fixed. */ > +#ifdef CONFIG_CPU_LOONGSON2F > +#define INSN_NOP 0x00200825 /* or at, at, zero */ > +#else > +#define INSN_NOP 0x00000000 /* nop */ > +#endif > +#define INSN_B_1F_16 0x10000004 /* b 1f; offset = (16 >> 2) */ > +#define INSN_B_1F_20 0x10000005 /* b 1f; offset = (20 >> 2) */ > +#define INSN_B_1F_24 0x10000006 /* b 1f; offset = (24 >> 2) */ > + > +/* After linking, the following instructions are fixed. */ > +static unsigned int insn_jal_ftrace_caller __read_mostly; > +static unsigned int insn_lui_v1_hi16_mcount __read_mostly; > +static unsigned int insn_j_ftrace_graph_caller __read_mostly; > + > +/* The following instructions are encoded in the run-time */ > +/* insn: jal func; op_code|addr : 31...26|25 ....0 */ > +#define INSN_JAL(addr) \ > + ((unsigned int)(0x0c000000 | (((addr) >> 2) & 0x03ffffff))) > + > +static inline void ftrace_dyn_arch_init_insns(void) > +{ > + u32 *buf; > + unsigned int v1; > + > + /* lui v1, hi16_mcount */ > + v1 = 3; > + buf = (u32 *)&insn_lui_v1_hi16_mcount; > + UASM_i_LA_mostly(&buf, v1, MCOUNT_ADDR); > > -static unsigned int ftrace_nop = 0x00000000; > + /* jal (ftrace_caller + 8), jump over the first two instruction */ > + buf = (u32 *)&insn_jal_ftrace_caller; > + uasm_i_jal(&buf, (FTRACE_ADDR + 8)); > + > + /* j ftrace_graph_caller */ > + buf = (u32 *)&insn_j_ftrace_graph_caller; > + uasm_i_j(&buf, (unsigned long)ftrace_graph_caller); > +} > > static int ftrace_modify_code(unsigned long ip, unsigned int new_code) > { > @@ -31,7 +66,6 @@ static int ftrace_modify_code(unsigned long ip, unsigned int new_code) > > /* *(unsigned int *)ip = new_code; */ > safe_store_code(new_code, ip, faulted); > - > if (unlikely(faulted)) > return -EFAULT; > > @@ -40,118 +74,132 @@ static int ftrace_modify_code(unsigned long ip, unsigned int new_code) > return 0; > } > > -static int lui_v1; > -static int jal_mcount; > +static inline unsigned int get_insn_b_1f(unsigned int *insn, unsigned long ip) > +{ > +#ifndef KBUILD_MCOUNT_RA_ADDRESS > + *insn = INSN_B_1F_16; > + return 0; > +#else > + int faulted; > + unsigned int code; > + > + ip -= 4; > + > + /* code = *(unsigned int *)ip; */ > + safe_load_code(code, ip, faulted); > + if (unlikely(faulted)) > + return -EFAULT; > + > + if ((code & INSN_S_R_SP) != INSN_S_R_SP) > + *insn = INSN_B_1F_20; > + else > + *insn = INSN_B_1F_24; > + > + return 0; > +#endif > +} > > int ftrace_make_nop(struct module *mod, > struct dyn_ftrace *rec, unsigned long addr) > { > unsigned int new; > - int faulted; > unsigned long ip = rec->ip; > > - /* We have compiled module with -mlong-calls, but compiled the kernel > - * without it, we need to cope with them respectively. */ > - if (ip & 0x40000000) { > - /* record it for ftrace_make_call */ > - if (lui_v1 == 0) { > - /* lui_v1 = *(unsigned int *)ip; */ > - safe_load_code(lui_v1, ip, faulted); > - > - if (unlikely(faulted)) > - return -EFAULT; > - } > - > - /* lui v1, hi_16bit_of_mcount --> b 1f (0x10000004) > - * addiu v1, v1, low_16bit_of_mcount > - * move at, ra > - * jalr v1 > - * nop > - * 1f: (ip + 12) > - */ > - new = 0x10000004; > - } else { > - /* record/calculate it for ftrace_make_call */ > - if (jal_mcount == 0) { > - /* We can record it directly like this: > - * jal_mcount = *(unsigned int *)ip; > - * Herein, jump over the first two nop instructions */ > - jal_mcount = jump_insn_encode(JAL, (MCOUNT_ADDR + 8)); > - } > - > - /* move at, ra > - * jalr v1 --> nop > - */ > - new = ftrace_nop; > - } > + /* > + * We have compiled modules with -mlong-calls, but compiled kernel > + * without it, therefore, need to cope with them respectively. > + * > + * For module: > + * > + * lui v1, hi16_mcount --> b 1f > + * addiu v1, v1, lo16_mcount > + * move at, ra > + * jalr v1 > + * nop > + * 1f: (ip + 16) > + * For kernel: > + * > + * move at, ra > + * jal _mcount --> nop > + * > + * And with the -mmcount-ra-address option, the offset may be 20 for > + * leaf fuction and 24 for non-leaf function. > + */ > + > + if (!in_module(ip)) > + new = INSN_NOP; > + else if (get_insn_b_1f(&new, ip)) > + return -EFAULT; > + > return ftrace_modify_code(ip, new); > } > > -static int modified; /* initialized as 0 by default */ > - > int ftrace_make_call(struct dyn_ftrace *rec, unsigned long addr) > { > unsigned int new; > unsigned long ip = rec->ip; > > - /* We just need to remove the "b ftrace_stub" at the fist time! */ > - if (modified == 0) { > - modified = 1; > - ftrace_modify_code(addr, ftrace_nop); > - } > - /* ip, module: 0xc0000000, kernel: 0x80000000 */ > - new = (ip & 0x40000000) ? lui_v1 : jal_mcount; > + new = in_module(ip) ? insn_lui_v1_hi16_mcount : insn_jal_ftrace_caller; > > return ftrace_modify_code(ip, new); > } > > -#define FTRACE_CALL_IP ((unsigned long)(&ftrace_call)) > - > int ftrace_update_ftrace_func(ftrace_func_t func) > { > unsigned int new; > > - new = jump_insn_encode(JAL, (unsigned long)func); > + new = INSN_JAL((unsigned long)func); > > - return ftrace_modify_code(FTRACE_CALL_IP, new); > + return ftrace_modify_code((unsigned long)(&ftrace_call), new); > } > > int __init ftrace_dyn_arch_init(void *data) > { > + ftrace_dyn_arch_init_insns(); > + > + /* > + * We are safe to remove the "b ftrace_stub" for the current > + * ftrace_caller() is almost empty (only the stack operations), and > + * most importantly, the calling to mcount will be disabled later in > + * ftrace_init(), then there is no 'big' overhead. And in the future, > + * we are hoping the function_trace_stop is initialized as 1 then we > + * can eventually remove that 'useless' "b ftrace_stub" and the > + * following nop. Currently, although we can call ftrace_stop() to set > + * function_trace_stop as 1, but it will change the behavior of the > + * Function Tracer. > + */ > + ftrace_modify_code(MCOUNT_ADDR, INSN_NOP); > + > /* The return code is retured via data */ > *(unsigned long *)data = 0; > > return 0; > } > -#endif /* CONFIG_DYNAMIC_FTRACE */ > +#endif /* CONFIG_DYNAMIC_FTRACE */ > > #ifdef CONFIG_FUNCTION_GRAPH_TRACER > > #ifdef CONFIG_DYNAMIC_FTRACE > > extern void ftrace_graph_call(void); > -#define JMP 0x08000000 /* jump to target directly */ > -#define CALL_FTRACE_GRAPH_CALLER \ > - jump_insn_encode(JMP, (unsigned long)(&ftrace_graph_caller)) > #define FTRACE_GRAPH_CALL_IP ((unsigned long)(&ftrace_graph_call)) > > int ftrace_enable_ftrace_graph_caller(void) > { > return ftrace_modify_code(FTRACE_GRAPH_CALL_IP, > - CALL_FTRACE_GRAPH_CALLER); > + insn_j_ftrace_graph_caller); > } > > int ftrace_disable_ftrace_graph_caller(void) > { > - return ftrace_modify_code(FTRACE_GRAPH_CALL_IP, ftrace_nop); > + return ftrace_modify_code(FTRACE_GRAPH_CALL_IP, INSN_NOP); > } > > -#endif /* !CONFIG_DYNAMIC_FTRACE */ > +#endif /* !CONFIG_DYNAMIC_FTRACE */ > > #ifndef KBUILD_MCOUNT_RA_ADDRESS > -#define S_RA_SP (0xafbf << 16) /* s{d,w} ra, offset(sp) */ > -#define S_R_SP (0xafb0 << 16) /* s{d,w} R, offset(sp) */ > -#define OFFSET_MASK 0xffff /* stack offset range: 0 ~ PT_SIZE */ > +#define INSN_S_RA_SP 0xafbf0000 /* s{d,w} ra, offset(sp) */ > +#define STACK_OFFSET_MASK 0xffff /* stack offset range: 0 ~ PT_SIZE */ > > unsigned long ftrace_get_parent_addr(unsigned long self_addr, > unsigned long parent, > @@ -162,35 +210,35 @@ unsigned long ftrace_get_parent_addr(unsigned long self_addr, > unsigned int code; > int faulted; > > - /* in module or kernel? */ > - if (self_addr & 0x40000000) { > - /* module: move to the instruction "lui v1, HI_16BIT_OF_MCOUNT" */ > - ip = self_addr - 20; > - } else { > - /* kernel: move to the instruction "move ra, at" */ > - ip = self_addr - 12; > - } > + /* > + * For module, move the ip from calling site of mcount to the > + * instruction "lui v1, hi_16bit_of_mcount"(offset is 20), but for > + * kernel, move to the instruction "move ra, at"(offset is 12) > + */ > + ip = self_addr - (in_module(self_addr) ? 20 : 12); > > - /* search the text until finding the non-store instruction or "s{d,w} > - * ra, offset(sp)" instruction */ > + /* > + * search the text until finding the non-store instruction or "s{d,w} > + * ra, offset(sp)" instruction > + */ > do { > ip -= 4; > > /* get the code at "ip": code = *(unsigned int *)ip; */ > safe_load_code(code, ip, faulted); > - > if (unlikely(faulted)) > return 0; > > - /* If we hit the non-store instruction before finding where the > + /* > + * If we hit the non-store instruction before finding where the > * ra is stored, then this is a leaf function and it does not > - * store the ra on the stack. */ > - if ((code & S_R_SP) != S_R_SP) > + * store the ra on the stack. > + */ > + if ((code & INSN_S_R_SP) != INSN_S_R_SP) > return parent_addr; > + } while (((code & INSN_S_RA_SP) != INSN_S_RA_SP)); > > - } while (((code & S_RA_SP) != S_RA_SP)); > - > - sp = fp + (code & OFFSET_MASK); > + sp = fp + (code & STACK_OFFSET_MASK); > > /* ra = *(unsigned long *)sp; */ > safe_load_stack(ra, sp, faulted); > @@ -211,16 +259,17 @@ unsigned long ftrace_get_parent_addr(unsigned long self_addr, > void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr, > unsigned long fp) > { > + int faulted; > unsigned long old; > struct ftrace_graph_ent trace; > unsigned long return_hooker = (unsigned long) > &return_to_handler; > - int faulted; > > if (unlikely(atomic_read(¤t->tracing_graph_pause))) > return; > > - /* "parent" is the stack address saved the return address of the caller > + /* > + * "parent" is the stack address saved the return address of the caller > * of _mcount. > * > * if the gcc < 4.5, a leaf function does not save the return address > @@ -241,12 +290,14 @@ void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr, > if (unlikely(faulted)) > goto out; > #ifndef KBUILD_MCOUNT_RA_ADDRESS > - parent = (unsigned long *)ftrace_get_parent_addr(self_addr, old, > - (unsigned long)parent, > - fp); > - /* If fails when getting the stack address of the non-leaf function's > - * ra, stop function graph tracer and return */ > - if (parent == 0) > + parent = (unsigned long *)ftrace_get_parent_addr( > + self_addr, old, (unsigned long)parent, fp); > + > + /* > + * If fails on getting the stack address of the non-leaf function's ra, > + * stop function graph tracer and return > + */ > + if (unlikely(parent == 0)) > goto out; > #endif > /* *parent = return_hooker; */ > @@ -272,4 +323,4 @@ out: > ftrace_graph_stop(); > WARN_ON(1); > } > -#endif /* CONFIG_FUNCTION_GRAPH_TRACER */ > +#endif /* CONFIG_FUNCTION_GRAPH_TRACER */ From dmitry.torokhov@gmail.com Fri Mar 12 19:35:02 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Fri, 12 Mar 2010 19:35:07 +0100 (CET) Received: from fg-out-1718.google.com ([72.14.220.157]:60935 "EHLO fg-out-1718.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492078Ab0CLSfC (ORCPT ); Fri, 12 Mar 2010 19:35:02 +0100 Received: by fg-out-1718.google.com with SMTP id 16so445069fgg.6 for ; Fri, 12 Mar 2010 10:35:02 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:date:from:to:cc:subject :message-id:references:mime-version:content-type:content-disposition :in-reply-to:user-agent; bh=4EQCixT/lHMIvQCQyw82WpBHcR91whUHkdAKx3Oyzso=; b=U8SuxvII3h1lnsN7mrobQIT18Xfo8imE/xFZU1Jz5diaHIBQwSkHnSQITyTDUOsdwD zIFv6W85SpCWHmWnKb0iNr9hqTpYB8gsGv/oBlkE7DWiAD3yfyIk3mOaSAws2s6ge4PK G47irrZkWcy1OA/y3l155Tv23zns6EVXtGXDY= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=date:from:to:cc:subject:message-id:references:mime-version :content-type:content-disposition:in-reply-to:user-agent; b=uuZ3XIUn2LSfnk6VIyEYkzJnmfRfUApdB0GzQVD2QUSLn8kxOJ6hhwLH1stBL+3UYT sLzy5Q4x3UPbYuEkBhatpmcMAoFuqVfBVEg9oIRmaisBXkT/LoCuHe6WYnvG2f4AQj0r 1zUM5ahj2pS3SXiMGbWxDV7ciwvCevDTuBoO4= Received: by 10.103.4.25 with SMTP id g25mr2195511mui.59.1268418899707; Fri, 12 Mar 2010 10:34:59 -0800 (PST) Received: from mailhub.coreip.homeip.net (c-24-6-153-206.hsd1.ca.comcast.net [24.6.153.206]) by mx.google.com with ESMTPS id 14sm6357941muo.32.2010.03.12.10.34.57 (version=TLSv1/SSLv3 cipher=RC4-MD5); Fri, 12 Mar 2010 10:34:58 -0800 (PST) Date: Fri, 12 Mar 2010 10:34:51 -0800 From: Dmitry Torokhov To: Wolfram Sang Cc: kernel-janitors@vger.kernel.org, Ralf Baechle , "Eric W. Biederman" , linux-mips@linux-mips.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH 1/3] arch/mips/txx9/generic: init dynamic bin_attribute structures Message-ID: <20100312183450.GC8736@core.coreip.homeip.net> References: <1268377431-11671-1-git-send-email-w.sang@pengutronix.de> <1268377431-11671-2-git-send-email-w.sang@pengutronix.de> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1268377431-11671-2-git-send-email-w.sang@pengutronix.de> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26218 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: dmitry.torokhov@gmail.com Precedence: bulk X-list: linux-mips On Fri, Mar 12, 2010 at 08:03:49AM +0100, Wolfram Sang wrote: > Commit 6992f5334995af474c2b58d010d08bc597f0f2fe introduced this requirement. > Found with coccinelle, but fixed manually. Compile tested on X86 where > possible. > Regarding all 3 - it looks like these dynamically alocated attributes could be converted to statically allocated ones. I'd recommend doing that instead (in fact, I posted patch for the firmware_class couple days ago). > Signed-off-by: Wolfram Sang > Cc: Ralf Baechle > Cc: Eric W. Biederman > --- > arch/mips/txx9/generic/setup.c | 1 + > 1 files changed, 1 insertions(+), 0 deletions(-) > > diff --git a/arch/mips/txx9/generic/setup.c b/arch/mips/txx9/generic/setup.c > index 7174d83..95184a0 100644 > --- a/arch/mips/txx9/generic/setup.c > +++ b/arch/mips/txx9/generic/setup.c > @@ -956,6 +956,7 @@ void __init txx9_sramc_init(struct resource *r) > if (!dev->base) > goto exit; > dev->dev.cls = &txx9_sramc_sysdev_class; > + sysfs_bin_attr_init(&dev->bindata_attr); > dev->bindata_attr.attr.name = "bindata"; > dev->bindata_attr.attr.mode = S_IRUSR | S_IWUSR; > dev->bindata_attr.read = txx9_sram_read; -- Dmitry From sgi-linux-mips@m.gmane.org Fri Mar 12 20:40:22 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Fri, 12 Mar 2010 20:40:25 +0100 (CET) Received: from lo.gmane.org ([80.91.229.12]:52572 "EHLO lo.gmane.org" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492294Ab0CLTkV (ORCPT ); Fri, 12 Mar 2010 20:40:21 +0100 Received: from list by lo.gmane.org with local (Exim 4.69) (envelope-from ) id 1NqAiX-0000lw-Ef for linux-mips@linux-mips.org; Fri, 12 Mar 2010 20:40:17 +0100 Received: from chipmunk.wormnet.eu ([195.195.131.226]) by main.gmane.org with esmtp (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Fri, 12 Mar 2010 20:40:17 +0100 Received: from alex by chipmunk.wormnet.eu with local (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Fri, 12 Mar 2010 20:40:17 +0100 X-Injected-Via-Gmane: http://gmane.org/ To: linux-mips@linux-mips.org From: Alexander Clouter Subject: [PATCH] MIPS: AR7: fix phat finger of reset bit in vlynq_high_data Date: Fri, 12 Mar 2010 19:39:48 +0000 Lines: 23 Message-ID: <4thq67-rjo.ln1@chipmunk.wormnet.eu> X-Complaints-To: usenet@dough.gmane.org X-Gmane-NNTP-Posting-Host: chipmunk.wormnet.eu User-Agent: tin/1.9.3-20080506 ("Dalintober") (UNIX) (Linux/2.6.26-2-sparc64-smp (sparc64)) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26219 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: alex@digriz.org.uk Precedence: bulk X-list: linux-mips Seems in my whitespace cleanup 7084338eb8eb0cc021ba86c340157bad397f3f0b caused AR7 to no longer get as far as init. Fixed my phat fingering. Signed-off-by: Alexander Clouter --- arch/mips/ar7/platform.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/arch/mips/ar7/platform.c b/arch/mips/ar7/platform.c index 246df7a..0bd5f67 100644 --- a/arch/mips/ar7/platform.c +++ b/arch/mips/ar7/platform.c @@ -168,7 +168,7 @@ static struct plat_vlynq_data vlynq_high_data = { .on = vlynq_on, .off = vlynq_off, }, - .reset_bit = 26, + .reset_bit = 16, .gpio_bit = 19, }; -- 1.7.0 From sgi-linux-mips@m.gmane.org Sat Mar 13 01:09:47 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Sat, 13 Mar 2010 01:09:50 +0100 (CET) Received: from lo.gmane.org ([80.91.229.12]:36420 "EHLO lo.gmane.org" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492419Ab0CMAJr (ORCPT ); Sat, 13 Mar 2010 01:09:47 +0100 Received: from list by lo.gmane.org with local (Exim 4.69) (envelope-from ) id 1NqEvI-0002oS-5J for linux-mips@linux-mips.org; Sat, 13 Mar 2010 01:09:44 +0100 Received: from chipmunk.wormnet.eu ([195.195.131.226]) by main.gmane.org with esmtp (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Sat, 13 Mar 2010 01:09:44 +0100 Received: from alex by chipmunk.wormnet.eu with local (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Sat, 13 Mar 2010 01:09:44 +0100 X-Injected-Via-Gmane: http://gmane.org/ To: linux-mips@linux-mips.org From: Alexander Clouter Subject: [PATCH] MIPS: AR7: fix phat finger of cpmac fixed_phy_add Date: Sat, 13 Mar 2010 00:09:15 +0000 Lines: 23 Message-ID: X-Complaints-To: usenet@dough.gmane.org X-Gmane-NNTP-Posting-Host: chipmunk.wormnet.eu User-Agent: tin/1.9.3-20080506 ("Dalintober") (UNIX) (Linux/2.6.26-2-sparc64-smp (sparc64)) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26220 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: alex@digriz.org.uk Precedence: bulk X-list: linux-mips Seems I trimmed one too many lines in 7084338eb8eb0cc021ba86c340157bad397f3f0b which led to no functioning Ethernet on my WAG54Gv2. This patch restores the AWOL line. Signed-off-by: Alexander Clouter --- arch/mips/ar7/platform.c | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/arch/mips/ar7/platform.c b/arch/mips/ar7/platform.c index 0bd5f67..2fafc78 100644 --- a/arch/mips/ar7/platform.c +++ b/arch/mips/ar7/platform.c @@ -600,6 +600,7 @@ static int __init ar7_register_devices(void) } if (ar7_has_high_cpmac()) { + res = fixed_phy_add(PHY_POLL, cpmac_high.id, &fixed_phy_status); if (!res) { cpmac_get_mac(1, cpmac_high_data.dev_addr); -- 1.7.0 From wsa@pengutronix.de Sat Mar 13 03:29:08 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Sat, 13 Mar 2010 03:29:11 +0100 (CET) Received: from metis.ext.pengutronix.de ([92.198.50.35]:57608 "EHLO metis.ext.pengutronix.de" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492747Ab0CMC3I (ORCPT ); Sat, 13 Mar 2010 03:29:08 +0100 Received: from octopus.hi.pengutronix.de ([2001:6f8:1178:2:215:17ff:fe12:23b0]) by metis.ext.pengutronix.de with esmtp (Exim 4.71) (envelope-from ) id 1NqH60-0004Zy-FS; Sat, 13 Mar 2010 03:28:56 +0100 Received: from wsa by octopus.hi.pengutronix.de with local (Exim 4.69) (envelope-from ) id 1NqH5z-0002na-Op; Sat, 13 Mar 2010 03:28:55 +0100 Date: Sat, 13 Mar 2010 03:28:55 +0100 From: Wolfram Sang To: Dmitry Torokhov Cc: kernel-janitors@vger.kernel.org, Ralf Baechle , "Eric W. Biederman" , linux-mips@linux-mips.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH 1/3] arch/mips/txx9/generic: init dynamic bin_attribute structures Message-ID: <20100313022855.GD4034@pengutronix.de> References: <1268377431-11671-1-git-send-email-w.sang@pengutronix.de> <1268377431-11671-2-git-send-email-w.sang@pengutronix.de> <20100312183450.GC8736@core.coreip.homeip.net> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="DiL7RhKs8rK9YGuF" Content-Disposition: inline In-Reply-To: <20100312183450.GC8736@core.coreip.homeip.net> User-Agent: Mutt/1.5.18 (2008-05-17) X-SA-Exim-Connect-IP: 2001:6f8:1178:2:215:17ff:fe12:23b0 X-SA-Exim-Mail-From: wsa@pengutronix.de X-SA-Exim-Scanned: No (on metis.ext.pengutronix.de); SAEximRunCond expanded to false X-PTX-Original-Recipient: linux-mips@linux-mips.org Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26221 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: w.sang@pengutronix.de Precedence: bulk X-list: linux-mips --DiL7RhKs8rK9YGuF Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Fri, Mar 12, 2010 at 10:34:51AM -0800, Dmitry Torokhov wrote: > On Fri, Mar 12, 2010 at 08:03:49AM +0100, Wolfram Sang wrote: > > Commit 6992f5334995af474c2b58d010d08bc597f0f2fe introduced this require= ment. > > Found with coccinelle, but fixed manually. Compile tested on X86 where > > possible. > >=20 >=20 > Regarding all 3 - it looks like these dynamically alocated attributes > could be converted to statically allocated ones. I'd recommend doing > that instead (in fact, I posted patch for the firmware_class couple days > ago). I agree for the firmware-patch. Regarding the MIPS one, 'size' might differ= and 'private' will differ per instance. Regarding the RTC driver, 'size' might = also differ. I don't know if somebody really wants two RTCs or the SRAM for MIPS= can be instantiated more than once. Unless somebody with actual hardware jumps = in, I'd say better safe than sorry. Thanks for the comment! Wolfram --=20 Pengutronix e.K. | Wolfram Sang | Industrial Linux Solutions | http://www.pengutronix.de/ | --DiL7RhKs8rK9YGuF Content-Type: application/pgp-signature; name="signature.asc" Content-Description: Digital signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) iEYEARECAAYFAkua+GcACgkQD27XaX1/VRsmUgCgxpG37bpkUetDGyW/fT8vhGaU 9XoAoIbuIfLRmEolrdu01v0d4cjAMYZJ =pKV6 -----END PGP SIGNATURE----- --DiL7RhKs8rK9YGuF-- From wuzhangjin@gmail.com Sat Mar 13 04:59:04 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Sat, 13 Mar 2010 04:59:08 +0100 (CET) Received: from mail-pz0-f186.google.com ([209.85.222.186]:61272 "EHLO mail-pz0-f186.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491050Ab0CMD7E (ORCPT ); Sat, 13 Mar 2010 04:59:04 +0100 Received: by pzk16 with SMTP id 16so590296pzk.22 for ; Fri, 12 Mar 2010 19:58:56 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:subject:from:reply-to:to:cc :in-reply-to:references:content-type:organization:date:message-id :mime-version:x-mailer:content-transfer-encoding; bh=7dUePuOO3Q7r1rIccmG107rqk1MhTN5hivNOh25VDKs=; b=SXrwyEjlREXcgHWw0dAVf+HFvfQ0ZPjFpaMo2NnGb+yobSs9MNubUolKj60pjUJAJr C7w8bPAPUIo0vY+b9obDRtpAWhTMIJCaQJGiBrh3EYbXa2NYdsIJ85laI+ykzJB9+Fof LKvXoVyqLr6aFnH/s5IF0koX0HGRpRlDZlglM= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=subject:from:reply-to:to:cc:in-reply-to:references:content-type :organization:date:message-id:mime-version:x-mailer :content-transfer-encoding; b=HDwxFNsVORyGSo0U3gBr46GZau3iFztYoXfNaujVLHYB6V1uYCFx2nCkD/I06JWpt9 TuaBf7ofKPc4trzBOHJ2PCj6P708gxZR7VftnDvuCR9Z72GX8frz8M3ixFPc2h0OQn97 iseBBZTvZwJg2bO17xg8k0dhITRfzIWguUAyM= Received: by 10.142.195.14 with SMTP id s14mr3327581wff.186.1268452736577; Fri, 12 Mar 2010 19:58:56 -0800 (PST) Received: from [202.201.12.142] ([202.201.12.142]) by mx.google.com with ESMTPS id 21sm2130808pzk.12.2010.03.12.19.58.51 (version=TLSv1/SSLv3 cipher=RC4-MD5); Fri, 12 Mar 2010 19:58:55 -0800 (PST) Subject: Re: [PATCH 1/3] Loongson-2F: Flush the branch target history such as BTB and RAS From: Wu Zhangjin Reply-To: wuzhangjin@gmail.com To: Ralf Baechle Cc: Shinya Kuribayashi , Greg KH , linux-mips@linux-mips.org, linux-kernel@vger.kernel.org In-Reply-To: <20100311095944.GC18065@linux-mips.org> References: <4B98632E.70806@necel.com> <20100311095944.GC18065@linux-mips.org> Content-Type: text/plain; charset="UTF-8" Organization: DSLab, Lanzhou University, China Date: Sat, 13 Mar 2010 11:52:21 +0800 Message-ID: <1268452341.21443.3.camel@falcon> Mime-Version: 1.0 X-Mailer: Evolution 2.28.2 Content-Transfer-Encoding: 7bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26222 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: wuzhangjin@gmail.com Precedence: bulk X-list: linux-mips On Thu, 2010-03-11 at 10:59 +0100, Ralf Baechle wrote: > On Thu, Mar 11, 2010 at 12:27:42PM +0900, Shinya Kuribayashi wrote: > > > Are you sure that RAS represents "Row Address Strobe", not "Return > > Address Stack?" This should be Return Address Stack. Will resend this patch later. Thanks & Regards, Wu Zhangjin From wuzhangjin@gmail.com Sat Mar 13 05:17:17 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Sat, 13 Mar 2010 05:17:20 +0100 (CET) Received: from mail-gy0-f177.google.com ([209.85.160.177]:42336 "EHLO mail-gy0-f177.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491017Ab0CMERR (ORCPT ); Sat, 13 Mar 2010 05:17:17 +0100 Received: by gyg10 with SMTP id 10so201796gyg.36 for ; Fri, 12 Mar 2010 20:17:10 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:subject:from:reply-to:to:cc :in-reply-to:references:content-type:organization:date:message-id :mime-version:x-mailer:content-transfer-encoding; bh=CF5NZPK+KByhp9+exGmuk7yn2/+85SqPK8ZdRIHrrD0=; b=g76Zqyt/0fpapKtiQve03R94WSEQz+gavTsKP7beS2kLjP6jDAzPF7NEMNxCHjX7l5 Kmp36DRTf6xn4jsvW2DQXXLkmwb5pQUCL105tQeguOc67ruFMHt+KWDRrVhbt9h39hrg zQ/fc69ToYySDc+uFJ67kZYfxyYa6Ses6sUbU= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=subject:from:reply-to:to:cc:in-reply-to:references:content-type :organization:date:message-id:mime-version:x-mailer :content-transfer-encoding; b=cSJlg/C+PKaU0+ciqKQluBJNkpRaoIWPKcIFQDZ1dYQeMK5Z9LQJ2SlgpZLzFjWrB9 MBZVT8w+0H0uyRDqADSd0jtHlXGShfrwnmx+IL5lkCS3e5y61upurdewHS5E+srVnTi0 KCNa9HyLhn9PUEc1ewOJ4+e1v7me0VuvZMxV8= Received: by 10.101.90.7 with SMTP id s7mr2591703anl.176.1268453830927; Fri, 12 Mar 2010 20:17:10 -0800 (PST) Received: from [202.201.12.142] ([202.201.12.142]) by mx.google.com with ESMTPS id 23sm1409637iwn.10.2010.03.12.20.17.06 (version=TLSv1/SSLv3 cipher=RC4-MD5); Fri, 12 Mar 2010 20:17:10 -0800 (PST) Subject: Re: [PATCH] MIPS: tracing: Optimize the implementation From: Wu Zhangjin Reply-To: wuzhangjin@gmail.com To: David Daney Cc: Ralf Baechle , Steven Rostedt , linux-mips@linux-mips.org In-Reply-To: <1268394209.6447.94.camel@falcon> References: <8b93c417fefa4d446f801abfd718ba94fdcb1821.1268330348.git.wuzhangjin@gmail.com> <4B993B32.7000006@caviumnetworks.com> <1268394209.6447.94.camel@falcon> Content-Type: text/plain; charset="UTF-8" Organization: DSLab, Lanzhou University, China Date: Sat, 13 Mar 2010 12:10:36 +0800 Message-ID: <1268453436.21443.10.camel@falcon> Mime-Version: 1.0 X-Mailer: Evolution 2.28.2 Content-Transfer-Encoding: 7bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26223 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: wuzhangjin@gmail.com Precedence: bulk X-list: linux-mips On Fri, 2010-03-12 at 19:43 +0800, Wu Zhangjin wrote: [...] > > > + * otherwise, it is in kernel space (0x80000000), return false. > > > + */ > > > +#define in_module(ip) (unlikely((ip)& 0x40000000)) > > > + > > > > This isn't universally true, but it does hold for most configurations I > > think. > > Although I'm not sure who is the exception, we always need an universal > solution, what about this: > > Compare module with kernel: > > module: > > > > lui v1, hi16_mcount <--- ip > addiu v1, v1, lo16_mcount > move at, ra > jalr v1 > nop > > kernel: > > > > move at, ra > jal _mcount <--- ip > > The above _ip_ is the address have been recorded into the __mcount_loc > section of the kernel by scripts/recordmcount.pl, as we can see, for > kernel, the *(ip - 4) is "move at, ra": 03e0082d, a certain instruction, > but for module, there is no possibility(?) of existing a "move at, ra" > at *(ip -4) but a register saving operation("s {d,w} rs, offset(sp)", > prefixed by 0xffb0 for 64bit and 0xafb0 for 32bit. ), and reversly, for > kernel, there is no such instruction there. > > And consider the new option -mmcount-ra-address of gcc, some more > instructions will be inserted between "move at, ra" and the calling site > to mcount, so, *(ip-4) will not always be "move at, ra", then we need to > check if there is a "s {d,w} rs, offset(sp)" there, if yes, it is in > module, otherwise, it should be in kernel. > > #define S_RS_SP 0xafb00000 /* s{d,w} rs, offset(sp) */ > > static inline int in_module(ip) > { > insn = *(ip - 4); /* need to use safe_load_code instead, what about big > endian? */ > > return ((insn & S_RS_SP) == S_RS_SP) > } The above method is not available for some cases, to avoid bring Ftrace with extra overhead, currently, I will keep using the original version although it may not work for some cases either. And to let ftrace_make_nop/ftrace_make_call be lightweight, I will also keep using the "b 1f" method in the old version. Regards, Wu Zhangjin From wuzhangjin@gmail.com Sat Mar 13 05:41:05 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Sat, 13 Mar 2010 05:41:13 +0100 (CET) Received: from mail-pw0-f49.google.com ([209.85.160.49]:50819 "EHLO mail-pw0-f49.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491001Ab0CMElF (ORCPT ); Sat, 13 Mar 2010 05:41:05 +0100 Received: by pwj2 with SMTP id 2so1145438pwj.36 for ; Fri, 12 Mar 2010 20:40:59 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:from:to:cc:subject:date :message-id:x-mailer:mime-version:content-type :content-transfer-encoding; bh=PmhfZVavA9GI1SxBc/5y9Qb8LUrzbS7lKl5advSSUQc=; b=J+JjKnnmUwOyGt6W5CZC1eve1YwK/vaIHFTOYs8kQjD3LD3qtEDwmbDMcyIlH4f6dH OCennhcrlmyKqNUk3DG2lHFqRVZ9WdDm1ggjp9sfLGEAbclOMLH6pQF8Da30/I3F8TSs pHKBSpp5Sr/KdGp/d7j+Lhm2dPaT0grLGD0h4= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=from:to:cc:subject:date:message-id:x-mailer:mime-version :content-type:content-transfer-encoding; b=KyFNSKwWrVKeW2ANLA/pzix6iEPQ9gOKtvUa3/TBYLECqiBAuDi50gEmQ4OxEUsI99 eFKx25eguZcqnKFRs2lJeC0kT/+jkaYHQLip+os3/4v7CeKwAuYP5M4irM7R1UlbGvT6 ucMKHRLLZkJvZfpCJgaoPRbWxtXLvpzG7zk90= Received: by 10.115.64.6 with SMTP id r6mr3416873wak.85.1268455257628; Fri, 12 Mar 2010 20:40:57 -0800 (PST) Received: from localhost.localdomain ([202.201.12.142]) by mx.google.com with ESMTPS id 21sm2172318pzk.0.2010.03.12.20.40.53 (version=TLSv1/SSLv3 cipher=RC4-MD5); Fri, 12 Mar 2010 20:40:57 -0800 (PST) From: Wu Zhangjin To: Ralf Baechle Cc: linux-mips@linux-mips.org, Shinya Kuribayashi , Wu Zhangjin Subject: [PATCH v3 0/3] Workaround the Out-of-order Issue of Loongson-2F Date: Sat, 13 Mar 2010 12:34:14 +0800 Message-Id: X-Mailer: git-send-email 1.7.0.1 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26224 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: wuzhangjin@gmail.com Precedence: bulk X-list: linux-mips From: Wu Zhangjin Hi, Ralf Is that possible to apply this patchset for both 2.6.33(the stable release) and 2.6.34? then we can get a not broken support for loongson-2f with the options(-mfix-loongson2f-nop, -mfix-loongson2f-jump) provided by the latest binutils (>=2.20.1). Thanks & Regards, Wu Zhangjin ------- Changes: v2->v3: o Herein, RAS is short for Return Address Stack not Row Address Strobe, at the same time, the "model" in the translation of Chapter 15 should be "mode". (feedback from Shinya Kuribayashi) v1->v2: o Cleanup some comments and align some instructions. ---- As the Chapter 15: "Errata: Issue of Out-of-order in loongson"[1] shows, to workaround the Issue of Loongson-2F，We need to do: o When switching from user mode to kernel mode, you should flush the branch target history such as BTB and RAS. o Doing some tricks to the indirect branch target to make sure that the indirect branch target can not be in the I/O region. This patchset applied the above methods and for the binutils patch[3] have been merged into binutils 2.20.1, so, it's time to upstream this patchset now. without this patch, the machines will hang when the instruction sequence hit the Out-of-order Issue of Loongson-2F, therefore, this patchset is very urgent for both 2.6.33 and 2.6.34. [1] Chinese Version: http://www.loongson.cn/uploadfile/file/20080821113149.pdf [2] English Version of Chapter 15: http://groups.google.com.hk/group/loongson-dev/msg/e0d2e220958f10a6?dmode=source [3] http://sourceware.org/ml/binutils/2009-11/msg00387.html Regards, Wu Zhangjin Wu Zhangjin (3): Loongson-2F: Flush the branch target history such as BTB and RAS Loongson-2F: Enable fixups of binutils 2.20.1 Loongson-2F: Fixup of problems introduced by -mfix-loongson2f-jump of binutils 2.20.1 arch/mips/Makefile | 4 +++- arch/mips/include/asm/stackframe.h | 19 +++++++++++++++++++ arch/mips/loongson/common/reset.c | 11 ++++++++++- 3 files changed, 32 insertions(+), 2 deletions(-) From wuzhangjin@gmail.com Sat Mar 13 05:45:55 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Sat, 13 Mar 2010 05:45:58 +0100 (CET) Received: from mail-px0-f189.google.com ([209.85.216.189]:58659 "EHLO mail-px0-f189.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491001Ab0CMEpz (ORCPT ); Sat, 13 Mar 2010 05:45:55 +0100 Received: by pxi27 with SMTP id 27so2110016pxi.0 for ; Fri, 12 Mar 2010 20:45:46 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:from:to:cc:subject:date :message-id:x-mailer:in-reply-to:references:in-reply-to:references :mime-version:content-type:content-transfer-encoding; bh=7ZXouYudrJmeK/+cVsboMRJmXpFppTKTJ7E8DejkODo=; b=bTJcgnAlEa2xWoKMd4NJpINHLIgzeFbEo1jWOZOnmgkLklRwM3+rsfLutVWRs28mDu JsBR3ULHfjhw22WNrAD3MOcnG+RaEU7TD5ay00uE4TQF5aiASAmtzdCFlY9n9KDg8B7H WDiutx3HGZFDf/hdP3wto71bInHEgaVr+lwA0= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=from:to:cc:subject:date:message-id:x-mailer:in-reply-to:references :mime-version:content-type:content-transfer-encoding; b=G4sPRG+5/5EkJm/8GTFY4iOJOJb9gKdCaZKmphQNXz248P6FG6OW6goGLD13lQJgcV 8VZFBDWljwfXBSzMJ4mD4S0qKtFRKjJeMSBukETCGew9hoFBqR1eyHXpKl60lBlW2/ZV ZvvqoV9We6sjWDsDP783Aocs6kGVwEJZyb9z0= Received: by 10.115.64.27 with SMTP id r27mr3381177wak.6.1268455545410; Fri, 12 Mar 2010 20:45:45 -0800 (PST) Received: from localhost.localdomain ([202.201.12.142]) by mx.google.com with ESMTPS id 21sm2172318pzk.0.2010.03.12.20.45.41 (version=TLSv1/SSLv3 cipher=RC4-MD5); Fri, 12 Mar 2010 20:45:44 -0800 (PST) From: Wu Zhangjin To: Ralf Baechle Cc: linux-mips@linux-mips.org, Shinya Kuribayashi , Wu Zhangjin Subject: [PATCH v3 1/3] Loongson-2F: Flush the branch target history such as BTB and RAS Date: Sat, 13 Mar 2010 12:34:15 +0800 Message-Id: <05e2ba2596f23fa4dda64d63ce2480504b1be4ac.1268453720.git.wuzhangjin@gmail.com> X-Mailer: git-send-email 1.7.0.1 In-Reply-To: References: In-Reply-To: References: MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26225 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: wuzhangjin@gmail.com Precedence: bulk X-list: linux-mips From: Wu Zhangjin As the Chapter 15: "Errata: Issue of Out-of-order in loongson"[1] shows, to workaround the Issue of Loongson-2F，We need to do: "When switching from user mode to kernel mode, you should flush the branch target history such as BTB and RAS." This patch did clear BTB(branch target buffer), forbid RAS(return address stack) via Loongson-2F's 64bit diagnostic register. [1] Chinese Version: http://www.loongson.cn/uploadfile/file/200808211 [2] English Version of Chapter 15: http://groups.google.com.hk/group/loongson-dev/msg/e0d2e220958f10a6?dmode=source Signed-off-by: Wu Zhangjin --- arch/mips/include/asm/stackframe.h | 19 +++++++++++++++++++ 1 files changed, 19 insertions(+), 0 deletions(-) diff --git a/arch/mips/include/asm/stackframe.h b/arch/mips/include/asm/stackframe.h index 3b6da33..c841912 100644 --- a/arch/mips/include/asm/stackframe.h +++ b/arch/mips/include/asm/stackframe.h @@ -121,6 +121,25 @@ .endm #else .macro get_saved_sp /* Uniprocessor variation */ +#ifdef CONFIG_CPU_LOONGSON2F + /* + * Clear BTB (branch target buffer), forbid RAS (return address + * stack) to workaround the Out-of-order Issue in Loongson2F + * via its diagnostic register. + */ + move k0, ra + jal 1f + nop +1: jal 1f + nop +1: jal 1f + nop +1: jal 1f + nop +1: move ra, k0 + li k0, 3 + mtc0 k0, $22 +#endif /* CONFIG_CPU_LOONGSON2F */ #if defined(CONFIG_32BIT) || defined(KBUILD_64BIT_SYM32) lui k1, %hi(kernelsp) #else -- 1.7.0.1 From wuzhangjin@gmail.com Sat Mar 13 05:46:16 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Sat, 13 Mar 2010 05:46:22 +0100 (CET) Received: from mail-pv0-f177.google.com ([74.125.83.177]:44103 "EHLO mail-pv0-f177.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491150Ab0CMEp7 (ORCPT ); Sat, 13 Mar 2010 05:45:59 +0100 Received: by pvh11 with SMTP id 11so31060pvh.36 for ; Fri, 12 Mar 2010 20:45:52 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:from:to:cc:subject:date :message-id:x-mailer:in-reply-to:references:in-reply-to:references; bh=P2v0PzTZ8PcrOj5xc9fJWbeIzqwSM6dukdMjQ4UhZUY=; b=bR/lVFy+U6gmgi5IHhWj5Onu6t8qGiXtdyNElfMez9OwXOmpR4m3/7s6ea9gu6lGR7 RZ5ecqqp9Ythv++8edNYiJbsNPoMmp5OFfRZsmoG4aBcj3SwowNlCAs5NKW9Fe2ngon5 fn55FpMcAOBZ22gCy0CRS3bEd6MJ3mcvRFnsw= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=from:to:cc:subject:date:message-id:x-mailer:in-reply-to:references; b=IlCO7IcZOAgmuLAIv5WTYK3zgUPRNAYNz2L53m/z1t8OKjAraiLJ4oFZQ9GyiXxIGI IX0PjzNuiVCyHdFf3ChIwd+pVhRiBohZUzQldQZYsVZ7MN2n2gxjcVuFffDeJub3z4xc siKXO5lbxYx+YxjygX+8Q8wyZdpgheDiGjkak= Received: by 10.142.119.26 with SMTP id r26mr961531wfc.320.1268455552048; Fri, 12 Mar 2010 20:45:52 -0800 (PST) Received: from localhost.localdomain ([202.201.12.142]) by mx.google.com with ESMTPS id 21sm2172318pzk.0.2010.03.12.20.45.47 (version=TLSv1/SSLv3 cipher=RC4-MD5); Fri, 12 Mar 2010 20:45:51 -0800 (PST) From: Wu Zhangjin To: Ralf Baechle Cc: linux-mips@linux-mips.org, Shinya Kuribayashi , Wu Zhangjin , Zhang Le , Wu Zhangjin Subject: [PATCH v3 2/3] Loongson-2F: Enable fixups of binutils 2.20.1 Date: Sat, 13 Mar 2010 12:34:16 +0800 Message-Id: X-Mailer: git-send-email 1.7.0.1 In-Reply-To: References: In-Reply-To: References: Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26226 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: wuzhangjin@gmail.com Precedence: bulk X-list: linux-mips From: Wu Zhangjin As the "Fixups of Loongson2F" patch[1] to binutils have been applied into binutils 2.20.1. It's time to enable the options provided by the patch to compile the kernel. Without these fixups, the system will hang unexpectedly for the bug of processor. To learn more about these fixups, please refer to the following references. [1] "Fixups of Loongson2F" patch for binutils(actually for gas) http://sourceware.org/ml/binutils/2009-11/msg00387.html [2] Chapter 15 of "Loongson2F User Manual"(Chinese Version) http://www.loongson.cn/uploadfile/file/200808211 [3] English Version of the above chapter 15 http://groups.google.com.hk/group/loongson-dev/msg/e0d2e220958f10a6?dmode=source Signed-off-by: Zhang Le Signed-off-by: Wu Zhangjin --- arch/mips/Makefile | 4 +++- 1 files changed, 3 insertions(+), 1 deletions(-) diff --git a/arch/mips/Makefile b/arch/mips/Makefile index 2f2eac2..5ae342e 100644 --- a/arch/mips/Makefile +++ b/arch/mips/Makefile @@ -135,7 +135,9 @@ cflags-$(CONFIG_CPU_LOONGSON2) += -Wa,--trap cflags-$(CONFIG_CPU_LOONGSON2E) += \ $(call cc-option,-march=loongson2e,-march=r4600) cflags-$(CONFIG_CPU_LOONGSON2F) += \ - $(call cc-option,-march=loongson2f,-march=r4600) + $(call cc-option,-march=loongson2f,-march=r4600) \ + $(call as-option,-Wa$(comma)-mfix-loongson2f-nop,) \ + $(call as-option,-Wa$(comma)-mfix-loongson2f-jump,) cflags-$(CONFIG_CPU_MIPS32_R1) += $(call cc-option,-march=mips32,-mips32 -U_MIPS_ISA -D_MIPS_ISA=_MIPS_ISA_MIPS32) \ -Wa,-mips32 -Wa,--trap -- 1.7.0.1 From wuzhangjin@gmail.com Sat Mar 13 05:46:40 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Sat, 13 Mar 2010 05:46:45 +0100 (CET) Received: from mail-pw0-f49.google.com ([209.85.160.49]:36803 "EHLO mail-pw0-f49.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491184Ab0CMEq2 (ORCPT ); Sat, 13 Mar 2010 05:46:28 +0100 Received: by pwj2 with SMTP id 2so1146496pwj.36 for ; Fri, 12 Mar 2010 20:46:22 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:from:to:cc:subject:date :message-id:x-mailer:in-reply-to:references:in-reply-to:references; bh=zr893OWQ3vUwhyg6oJp4MD6FThtWm5mK31cbV664TB8=; b=x/ePD0JpPJZNB+qSQ7ZvlGKy0liJBS9L+7A2lv43Rg6mW7nSP8+NdwxcCIuglFQQVJ rL94QR1y+rfUgQB1xf+6l/QaTMTT1R14V/0wq2QNDVOf16uy6IqV0sVilRp6SHDJHko0 tAA36xt1OZg6xxtR/h600fU4NAf07vuqTNIhU= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=from:to:cc:subject:date:message-id:x-mailer:in-reply-to:references; b=hpgOHkTAMVe7UWU4ZtP58dZI2aLOAs0TYoHtfqcbhwF85FhXgVNgMXPNeFWHERj8ki gsGwhBkRC8bUmdzKRhelKReINo210R126jLfDW9oLZEI/c36fcDe/eAFHBOOSoVsbKYe cs6dLzRhQOZL5CjaES+mA4pb4cNTHpD6pVut4= Received: by 10.114.188.18 with SMTP id l18mr573054waf.32.1268455580133; Fri, 12 Mar 2010 20:46:20 -0800 (PST) Received: from localhost.localdomain ([202.201.12.142]) by mx.google.com with ESMTPS id 21sm2172318pzk.0.2010.03.12.20.46.16 (version=TLSv1/SSLv3 cipher=RC4-MD5); Fri, 12 Mar 2010 20:46:19 -0800 (PST) From: Wu Zhangjin To: Ralf Baechle Cc: linux-mips@linux-mips.org, Shinya Kuribayashi , Wu Zhangjin Subject: [PATCH v3 3/3] Loongson-2F: Fixup of problems introduced by -mfix-loongson2f-jump of binutils 2.20.1 Date: Sat, 13 Mar 2010 12:34:17 +0800 Message-Id: <169f2daa3d623fe56c5b0be30feeda10bc79a478.1268453720.git.wuzhangjin@gmail.com> X-Mailer: git-send-email 1.7.0.1 In-Reply-To: References: In-Reply-To: References: Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26227 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: wuzhangjin@gmail.com Precedence: bulk X-list: linux-mips From: Wu Zhangjin The -mfix-loongson2f-jump option provided by the binutils 2.20.1 have fixed the Out-of-order Issue of Loongson-2F described in Chapter 15 of "Loongson2F User Manual"[1,2], but introduced some problems. The option changes all of the jumping target to "addr & 0xcfffffff" through the at($1) register, but for the REBOOT address of loongson-2F: 0xbfc00000, this is totally wrong, so, this patch try to avoid the problem via telling the assembler not to use at($1) register. [1] Loongson2F User Manual(Chinese Version) http://www.loongson.cn/uploadfile/file/200808211 [2] English Version of Chapter 15: http://groups.google.com.hk/group/loongson-dev/msg/e0d2e220958f10a6?dmode=source Reported-and-tested-by: Liu Shiwei Signed-off-by: Wu Zhangjin --- arch/mips/loongson/common/reset.c | 11 ++++++++++- 1 files changed, 10 insertions(+), 1 deletions(-) diff --git a/arch/mips/loongson/common/reset.c b/arch/mips/loongson/common/reset.c index 4bd9c18..dc311f2 100644 --- a/arch/mips/loongson/common/reset.c +++ b/arch/mips/loongson/common/reset.c @@ -21,8 +21,17 @@ static void loongson_restart(char *command) /* do preparation for reboot */ mach_prepare_reboot(); - /* reboot via jumping to boot base address */ + /* reboot via jumping to boot base address + * + * ".set noat" and ".set at" are used to ensure the address not break + * by the -mfix-loongson2f-jump option provided by binutils 2.20.1 (or + * higher version) which try to change the jumping address to "addr & + * 0xcfffffff" via the at($1) register, this is totally wrong for + * 0xbfc00000 (LOONGSON_BOOT_BASE). + */ + __asm__ __volatile__(".set noat\n"); ((void (*)(void))ioremap_nocache(LOONGSON_BOOT_BASE, 4)) (); + __asm__ __volatile__(".set at\n"); } static void loongson_poweroff(void) -- 1.7.0.1 From dmitry.torokhov@gmail.com Sat Mar 13 09:34:12 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Sat, 13 Mar 2010 09:34:15 +0100 (CET) Received: from mail-yw0-f201.google.com ([209.85.211.201]:43202 "EHLO mail-yw0-f201.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492094Ab0CMIeM (ORCPT ); Sat, 13 Mar 2010 09:34:12 +0100 Received: by ywh39 with SMTP id 39so807261ywh.21 for ; Sat, 13 Mar 2010 00:34:03 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:date:from:to:cc:subject :message-id:references:mime-version:content-type:content-disposition :in-reply-to:user-agent; bh=LbmcyLF83DealLGlmAywXxDmnjq2Fy1VO+w8AFn5ECU=; b=BW2crtG9BCIJoIIh/AoR4tfS89wrhCM8OKzgztBxKK2iIhTaTZ0b3j9M3C55hg8Cnp H37PZ1Lw0OvbfEnB6YwXqFiWtmvd0aO9d8P/roQ7pkQOsmrKQYBDfOS2xKzO89PGpxCe S+ZLKEAGT1IVdbrtGJtdXWwmVy8CvdQcJ0IUU= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=date:from:to:cc:subject:message-id:references:mime-version :content-type:content-disposition:in-reply-to:user-agent; b=B8eGl0B/reUWxzOdxwT2w5uTFpEQqz/uL7jYw7CUAp8+WVvkOmnQmLdNUzx91+RI/g TeVvFxWnYckVqJtCKyD1qwMRy73OeA6O/XNrznlopZrtnxilb2f3Qxsdw0MypB/3dVws 4hS0n1aMh1+Jdr54jZe4F4wgEMv9KdZLz4NZ4= Received: by 10.101.210.15 with SMTP id m15mr1337941anq.111.1268469243505; Sat, 13 Mar 2010 00:34:03 -0800 (PST) Received: from mailhub.coreip.homeip.net (c-24-6-153-206.hsd1.ca.comcast.net [24.6.153.206]) by mx.google.com with ESMTPS id 22sm747206ywh.1.2010.03.13.00.34.02 (version=TLSv1/SSLv3 cipher=RC4-MD5); Sat, 13 Mar 2010 00:34:02 -0800 (PST) Date: Sat, 13 Mar 2010 00:33:59 -0800 From: Dmitry Torokhov To: Wolfram Sang Cc: kernel-janitors@vger.kernel.org, Ralf Baechle , "Eric W. Biederman" , linux-mips@linux-mips.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH 1/3] arch/mips/txx9/generic: init dynamic bin_attribute structures Message-ID: <20100313083359.GB22494@core.coreip.homeip.net> References: <1268377431-11671-1-git-send-email-w.sang@pengutronix.de> <1268377431-11671-2-git-send-email-w.sang@pengutronix.de> <20100312183450.GC8736@core.coreip.homeip.net> <20100313022855.GD4034@pengutronix.de> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20100313022855.GD4034@pengutronix.de> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26228 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: dmitry.torokhov@gmail.com Precedence: bulk X-list: linux-mips On Sat, Mar 13, 2010 at 03:28:55AM +0100, Wolfram Sang wrote: > On Fri, Mar 12, 2010 at 10:34:51AM -0800, Dmitry Torokhov wrote: > > On Fri, Mar 12, 2010 at 08:03:49AM +0100, Wolfram Sang wrote: > > > Commit 6992f5334995af474c2b58d010d08bc597f0f2fe introduced this requirement. > > > Found with coccinelle, but fixed manually. Compile tested on X86 where > > > possible. > > > > > > > Regarding all 3 - it looks like these dynamically alocated attributes > > could be converted to statically allocated ones. I'd recommend doing > > that instead (in fact, I posted patch for the firmware_class couple days > > ago). > > I agree for the firmware-patch. Regarding the MIPS one, 'size' might differ and > 'private' will differ per instance. Regarding the RTC driver, 'size' might also > differ. I don't know if somebody really wants two RTCs or the SRAM for MIPS can > be instantiated more than once. Unless somebody with actual hardware jumps in, > I'd say better safe than sorry. > Ah, right, size... I forgot about it. You are right, making the other 2 static is not an option. -- Dmitry From shinya.kuribayashi@necel.com Mon Mar 15 00:42:02 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Mon, 15 Mar 2010 00:42:10 +0100 (CET) Received: from TYO201.gate.nec.co.jp ([202.32.8.193]:54187 "EHLO tyo201.gate.nec.co.jp" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492528Ab0CNXmC (ORCPT ); Mon, 15 Mar 2010 00:42:02 +0100 Received: from relay21.aps.necel.com ([10.29.19.50]) by tyo201.gate.nec.co.jp (8.13.8/8.13.4) with ESMTP id o2ENfpXA000829; Mon, 15 Mar 2010 08:41:56 +0900 (JST) Received: from realmbox31.aps.necel.com ([10.29.19.32] [10.29.19.32]) by relay21.aps.necel.com with ESMTP; Mon, 15 Mar 2010 08:41:55 +0900 Received: from [10.114.181.193] ([10.114.181.193] [10.114.181.193]) by mbox02.aps.necel.com with ESMTP; Mon, 15 Mar 2010 08:41:55 +0900 Message-ID: <4B9D744A.5080301@necel.com> Date: Mon, 15 Mar 2010 08:42:02 +0900 From: Shinya Kuribayashi User-Agent: Thunderbird 2.0.0.23 (Windows/20090812) MIME-Version: 1.0 To: wuzhangjin@gmail.com CC: Ralf Baechle , Greg KH , linux-mips@linux-mips.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH 1/3] Loongson-2F: Flush the branch target history such as BTB and RAS References: <4B98632E.70806@necel.com> <20100311095944.GC18065@linux-mips.org> <1268452341.21443.3.camel@falcon> In-Reply-To: <1268452341.21443.3.camel@falcon> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26229 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: shinya.kuribayashi@necel.com Precedence: bulk X-list: linux-mips Wu Zhangjin wrote: > On Thu, 2010-03-11 at 10:59 +0100, Ralf Baechle wrote: >> On Thu, Mar 11, 2010 at 12:27:42PM +0900, Shinya Kuribayashi wrote: >> >>> Are you sure that RAS represents "Row Address Strobe", not "Return >>> Address Stack?" > > This should be Return Address Stack. > > Will resend this patch later. Thanks for the head's up. -- Shinya Kuribayashi NEC Electronics From w.sang@pengutronix.de Mon Mar 15 01:33:13 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Mon, 15 Mar 2010 01:33:21 +0100 (CET) Received: from metis.ext.pengutronix.de ([92.198.50.35]:56644 "EHLO metis.ext.pengutronix.de" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492535Ab0COAdN (ORCPT ); Mon, 15 Mar 2010 01:33:13 +0100 Received: from themisto.ext.pengutronix.de ([92.198.50.58] helo=pengutronix.de) by metis.ext.pengutronix.de with esmtp (Exim 4.71) (envelope-from ) id 1NqyEg-0002Pe-1J; Mon, 15 Mar 2010 01:32:49 +0100 From: Wolfram Sang To: linux-kernel@vger.kernel.org Cc: Grant Likely , kernel-janitors@vger.kernel.org, Wolfram Sang , "Eric W. Biederman" , Linus Torvalds , Ralf Baechle , Paul Gortmaker , Alessandro Zummo , linux-mips@linux-mips.org, rtc-linux@googlegroups.com Date: Mon, 15 Mar 2010 01:29:41 +0100 Message-Id: <1268612981-9009-1-git-send-email-w.sang@pengutronix.de> X-Mailer: git-send-email 1.7.0 In-Reply-To: References: X-SA-Exim-Connect-IP: 92.198.50.58 X-SA-Exim-Mail-From: w.sang@pengutronix.de Subject: [PATCH] init dynamic bin_attribute structures X-SA-Exim-Version: 4.2.1 (built Sat, 01 Aug 2009 12:09:26 +0000) X-SA-Exim-Scanned: Yes (on metis.ext.pengutronix.de) X-PTX-Original-Recipient: linux-mips@linux-mips.org Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26230 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: w.sang@pengutronix.de Precedence: bulk X-list: linux-mips Commit 6992f5334995af474c2b58d010d08bc597f0f2fe introduced this requirement. First, at25 was fixed manually. Then, other occurences were found with coccinelle and the following semantic patch. Results were reviewed and fixed up: @ init @ identifier struct_name, bin; @@ struct struct_name { ... struct bin_attribute bin; ... }; @ main extends init @ expression E; statement S; identifier name, err; @@ ( struct struct_name *name; | - struct struct_name *name = NULL; + struct struct_name *name; ) ... ( sysfs_bin_attr_init(&name->bin); | + sysfs_bin_attr_init(&name->bin); if (sysfs_create_bin_file(E, &name->bin)) S | + sysfs_bin_attr_init(&name->bin); err = sysfs_create_bin_file(E, &name->bin); ) Signed-off-by: Wolfram Sang Cc: Eric W. Biederman Cc: Linus Torvalds --- arch/mips/txx9/generic/setup.c | 1 + drivers/misc/eeprom/at25.c | 1 + drivers/rtc/rtc-ds1742.c | 1 + 3 files changed, 3 insertions(+), 0 deletions(-) diff --git a/arch/mips/txx9/generic/setup.c b/arch/mips/txx9/generic/setup.c index 7174d83..95184a0 100644 --- a/arch/mips/txx9/generic/setup.c +++ b/arch/mips/txx9/generic/setup.c @@ -956,6 +956,7 @@ void __init txx9_sramc_init(struct resource *r) if (!dev->base) goto exit; dev->dev.cls = &txx9_sramc_sysdev_class; + sysfs_bin_attr_init(&dev->bindata_attr); dev->bindata_attr.attr.name = "bindata"; dev->bindata_attr.attr.mode = S_IRUSR | S_IWUSR; dev->bindata_attr.read = txx9_sram_read; diff --git a/drivers/misc/eeprom/at25.c b/drivers/misc/eeprom/at25.c index d902d81..d194212 100644 --- a/drivers/misc/eeprom/at25.c +++ b/drivers/misc/eeprom/at25.c @@ -347,6 +347,7 @@ static int at25_probe(struct spi_device *spi) * that's sensitive for read and/or write, like ethernet addresses, * security codes, board-specific manufacturing calibrations, etc. */ + sysfs_bin_attr_init(&at25->bin); at25->bin.attr.name = "eeprom"; at25->bin.attr.mode = S_IRUSR; at25->bin.read = at25_bin_read; diff --git a/drivers/rtc/rtc-ds1742.c b/drivers/rtc/rtc-ds1742.c index a127336..cad9ceb 100644 --- a/drivers/rtc/rtc-ds1742.c +++ b/drivers/rtc/rtc-ds1742.c @@ -184,6 +184,7 @@ static int __devinit ds1742_rtc_probe(struct platform_device *pdev) pdata->size_nvram = pdata->size - RTC_SIZE; pdata->ioaddr_rtc = ioaddr + pdata->size_nvram; + sysfs_bin_attr_init(&pdata->nvram_attr); pdata->nvram_attr.attr.name = "nvram"; pdata->nvram_attr.attr.mode = S_IRUGO | S_IWUSR; pdata->nvram_attr.read = ds1742_nvram_read; -- 1.7.0 From jkosina@suse.cz Mon Mar 15 11:00:19 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Mon, 15 Mar 2010 11:00:26 +0100 (CET) Received: from cantor2.suse.de ([195.135.220.15]:39845 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491820Ab0COKAT (ORCPT ); Mon, 15 Mar 2010 11:00:19 +0100 Received: from relay2.suse.de (charybdis-ext.suse.de [195.135.221.2]) by mx2.suse.de (Postfix) with ESMTP id D0AD6867E2; Mon, 15 Mar 2010 11:00:18 +0100 (CET) Date: Mon, 15 Mar 2010 11:00:11 +0100 (CET) From: Jiri Kosina X-X-Sender: jikos@twin.jikos.cz To: Wolfram Sang Cc: linux-kernel@vger.kernel.org, Grant Likely , kernel-janitors@vger.kernel.org, "Eric W. Biederman" , Linus Torvalds , Ralf Baechle , Paul Gortmaker , Alessandro Zummo , linux-mips@linux-mips.org, rtc-linux@googlegroups.com Subject: Re: [PATCH] init dynamic bin_attribute structures In-Reply-To: <1268612981-9009-1-git-send-email-w.sang@pengutronix.de> Message-ID: References: <1268612981-9009-1-git-send-email-w.sang@pengutronix.de> User-Agent: Alpine 2.00 (LRH 1167 2008-08-23) MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26231 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: jkosina@suse.cz Precedence: bulk X-list: linux-mips On Mon, 15 Mar 2010, Wolfram Sang wrote: > Commit 6992f5334995af474c2b58d010d08bc597f0f2fe introduced this requirement. > First, at25 was fixed manually. Then, other occurences were found with > coccinelle and the following semantic patch. Results were reviewed and fixed > up: > > @ init @ > identifier struct_name, bin; > @@ > > struct struct_name { > ... > struct bin_attribute bin; > ... > }; > > @ main extends init @ > expression E; > statement S; > identifier name, err; > @@ > > ( > struct struct_name *name; > | > - struct struct_name *name = NULL; > + struct struct_name *name; > ) > ... > ( > sysfs_bin_attr_init(&name->bin); > | > + sysfs_bin_attr_init(&name->bin); > if (sysfs_create_bin_file(E, &name->bin)) > S > | > + sysfs_bin_attr_init(&name->bin); > err = sysfs_create_bin_file(E, &name->bin); > ) > > Signed-off-by: Wolfram Sang > Cc: Eric W. Biederman > Cc: Linus Torvalds > --- > arch/mips/txx9/generic/setup.c | 1 + > drivers/misc/eeprom/at25.c | 1 + > drivers/rtc/rtc-ds1742.c | 1 + > 3 files changed, 3 insertions(+), 0 deletions(-) I don't understand how cocinelle works, but the resulting patch seems to be incomplete. The fix is needed at least in firmware loader (for which Greg has already queued patch), in ACPI thermal driver [1], etc). [1] http://lkml.indiana.edu/hypermail/linux/kernel/1003.1/02680.html -- Jiri Kosina SUSE Labs, Novell Inc. From wsa@pengutronix.de Mon Mar 15 11:12:59 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Mon, 15 Mar 2010 11:13:02 +0100 (CET) Received: from metis.ext.pengutronix.de ([92.198.50.35]:47927 "EHLO metis.ext.pengutronix.de" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491908Ab0COKM7 (ORCPT ); Mon, 15 Mar 2010 11:12:59 +0100 Received: from octopus.hi.pengutronix.de ([2001:6f8:1178:2:215:17ff:fe12:23b0]) by metis.ext.pengutronix.de with esmtp (Exim 4.71) (envelope-from ) id 1Nr7Hn-0003VS-1q; Mon, 15 Mar 2010 11:12:35 +0100 Received: from wsa by octopus.hi.pengutronix.de with local (Exim 4.69) (envelope-from ) id 1Nr7Hk-0001AZ-Kl; Mon, 15 Mar 2010 11:12:32 +0100 Date: Mon, 15 Mar 2010 11:12:32 +0100 From: Wolfram Sang To: rtc-linux@googlegroups.com Cc: linux-kernel@vger.kernel.org, Grant Likely , kernel-janitors@vger.kernel.org, "Eric W. Biederman" , Linus Torvalds , Ralf Baechle , Paul Gortmaker , Alessandro Zummo , linux-mips@linux-mips.org, Jiri Kosina Subject: Re: [rtc-linux] Re: [PATCH] init dynamic bin_attribute structures Message-ID: <20100315101232.GB20604@pengutronix.de> References: <1268612981-9009-1-git-send-email-w.sang@pengutronix.de> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="neYutvxvOLaeuPCA" Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.18 (2008-05-17) X-SA-Exim-Connect-IP: 2001:6f8:1178:2:215:17ff:fe12:23b0 X-SA-Exim-Mail-From: wsa@pengutronix.de X-SA-Exim-Scanned: No (on metis.ext.pengutronix.de); SAEximRunCond expanded to false X-PTX-Original-Recipient: linux-mips@linux-mips.org Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26232 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: w.sang@pengutronix.de Precedence: bulk X-list: linux-mips --neYutvxvOLaeuPCA Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable > I don't understand how cocinelle works, but the resulting patch seems to= =20 > be incomplete. Okay, will investigate, thanks for the pointer. > The fix is needed at least in firmware loader (for which Greg has already= =20 > queued patch), in ACPI thermal driver [1], etc). Firmware loader should be handled differently, see other mails in the originating thread, so I skipped it intentionally. at24 also needed a fix, = but that one was just applied via the i2c-tree, so I skipped it, too. Will check the ACPI-case! Regards, Wolfram --=20 Pengutronix e.K. | Wolfram Sang | Industrial Linux Solutions | http://www.pengutronix.de/ | --neYutvxvOLaeuPCA Content-Type: application/pgp-signature; name="signature.asc" Content-Description: Digital signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) iEYEARECAAYFAkueCBAACgkQD27XaX1/VRswuQCdFiF8i3Vjj2s0QPm1v4UCjE2R pJYAoI5n8xxHNNFVFuS3KCJDpN1uBst/ =9hnN -----END PGP SIGNATURE----- --neYutvxvOLaeuPCA-- From ralf@linux-mips.org Mon Mar 15 18:13:02 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Mon, 15 Mar 2010 18:13:06 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:53751 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492380Ab0CORNC (ORCPT ); Mon, 15 Mar 2010 18:13:02 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o2FHClUr020278; Mon, 15 Mar 2010 18:12:48 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o2FHCkct020276; Mon, 15 Mar 2010 18:12:46 +0100 Date: Mon, 15 Mar 2010 18:12:46 +0100 From: Ralf Baechle To: Alexander Clouter Cc: linux-mips@linux-mips.org Subject: Re: [PATCH] MIPS: AR7: fix phat finger of reset bit in vlynq_high_data Message-ID: <20100315171246.GA20055@linux-mips.org> References: <4thq67-rjo.ln1@chipmunk.wormnet.eu> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <4thq67-rjo.ln1@chipmunk.wormnet.eu> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26233 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Fri, Mar 12, 2010 at 07:39:48PM +0000, Alexander Clouter wrote: > From: Alexander Clouter > Date: Fri, 12 Mar 2010 19:39:48 +0000 > To: linux-mips@linux-mips.org > Subject: [PATCH] MIPS: AR7: fix phat finger of reset bit in vlynq_high_data > > Seems in my whitespace cleanup 7084338eb8eb0cc021ba86c340157bad397f3f0b > caused AR7 to no longer get as far as init. Fixed my phat fingering. Patch looks good but it seems you phat fingered the commit ID which is 0f2536082d01448daeced8d9e82c3ba1751fefa3 (lmo) rsp. 8c2961da46abd85a71d20f2b169bf80618e (kernel.org). Applied, Ralf From ralf@linux-mips.org Mon Mar 15 18:17:30 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Mon, 15 Mar 2010 18:17:33 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:40528 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492385Ab0CORRa (ORCPT ); Mon, 15 Mar 2010 18:17:30 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o2FHHIFr020469; Mon, 15 Mar 2010 18:17:18 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o2FHHII8020467; Mon, 15 Mar 2010 18:17:18 +0100 Date: Mon, 15 Mar 2010 18:17:18 +0100 From: Ralf Baechle To: Alexander Clouter Cc: linux-mips@linux-mips.org Subject: Re: [PATCH] MIPS: AR7: fix phat finger of cpmac fixed_phy_add Message-ID: <20100315171718.GB20055@linux-mips.org> References: MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26234 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Sat, Mar 13, 2010 at 12:09:15AM +0000, Alexander Clouter wrote: > Seems I trimmed one too many lines in > 7084338eb8eb0cc021ba86c340157bad397f3f0b which led to no functioning > Ethernet on my WAG54Gv2. This patch restores the AWOL line. Phat-freely applied. Thanks! Ralf From ralf@linux-mips.org Mon Mar 15 18:49:09 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Mon, 15 Mar 2010 18:49:12 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:33382 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492375Ab0CORtJ (ORCPT ); Mon, 15 Mar 2010 18:49:09 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o2FHn6HA021200; Mon, 15 Mar 2010 18:49:07 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o2FHn5Xu021198; Mon, 15 Mar 2010 18:49:05 +0100 Date: Mon, 15 Mar 2010 18:49:05 +0100 From: Ralf Baechle To: Wolfram Sang Cc: Dmitry Torokhov , kernel-janitors@vger.kernel.org, "Eric W. Biederman" , linux-mips@linux-mips.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH 1/3] arch/mips/txx9/generic: init dynamic bin_attribute structures Message-ID: <20100315174905.GC20055@linux-mips.org> References: <1268377431-11671-1-git-send-email-w.sang@pengutronix.de> <1268377431-11671-2-git-send-email-w.sang@pengutronix.de> <20100312183450.GC8736@core.coreip.homeip.net> <20100313022855.GD4034@pengutronix.de> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20100313022855.GD4034@pengutronix.de> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26235 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Sat, Mar 13, 2010 at 03:28:55AM +0100, Wolfram Sang wrote: > > Regarding all 3 - it looks like these dynamically alocated attributes > > could be converted to statically allocated ones. I'd recommend doing > > that instead (in fact, I posted patch for the firmware_class couple days > > ago). > > I agree for the firmware-patch. Regarding the MIPS one, 'size' might differ and > 'private' will differ per instance. Regarding the RTC driver, 'size' might also > differ. I don't know if somebody really wants two RTCs or the SRAM for MIPS can > be instantiated more than once. Unless somebody with actual hardware jumps in, > I'd say better safe than sorry. On the txx9 platform you've posted the patch for additional RTCs or SRAMs would not normally be expected. On other platforms such as IP27 there would be one per node that is potencially very many. Ralf From torvalds@linux-foundation.org Mon Mar 15 19:50:45 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Mon, 15 Mar 2010 19:50:51 +0100 (CET) Received: from smtp1.linux-foundation.org ([140.211.169.13]:36884 "EHLO smtp1.linux-foundation.org" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492530Ab0COSup (ORCPT ); Mon, 15 Mar 2010 19:50:45 +0100 Received: from imap1.linux-foundation.org (imap1.linux-foundation.org [140.211.169.55]) by smtp1.linux-foundation.org (8.14.2/8.13.5/Debian-3ubuntu1.1) with ESMTP id o2FIo5PO006541 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO); Mon, 15 Mar 2010 11:50:06 -0700 Received: from localhost (localhost [127.0.0.1]) by imap1.linux-foundation.org (8.13.5.20060308/8.13.5/Debian-3ubuntu1.1) with ESMTP id o2FIo2N4016455; Mon, 15 Mar 2010 11:50:02 -0700 Date: Mon, 15 Mar 2010 11:47:53 -0700 (PDT) From: Linus Torvalds To: Jiri Kosina cc: Wolfram Sang , linux-kernel@vger.kernel.org, Grant Likely , kernel-janitors@vger.kernel.org, "Eric W. Biederman" , Ralf Baechle , Paul Gortmaker , Alessandro Zummo , linux-mips@linux-mips.org, rtc-linux@googlegroups.com Subject: Re: [PATCH] init dynamic bin_attribute structures In-Reply-To: Message-ID: References: <1268612981-9009-1-git-send-email-w.sang@pengutronix.de> User-Agent: Alpine 2.00 (LFD 1167 2008-08-23) MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-MIMEDefang-Filter: lf$Revision: 1.188 $ X-Scanned-By: MIMEDefang 2.63 on 140.211.169.13 Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26236 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: torvalds@linux-foundation.org Precedence: bulk X-list: linux-mips On Mon, 15 Mar 2010, Jiri Kosina wrote: > > I don't understand how cocinelle works, but the resulting patch seems to > be incomplete. > > The fix is needed at least in firmware loader (for which Greg has already > queued patch), in ACPI thermal driver [1], etc). The firmware loader should be done differently: http://patchwork.kernel.org/patch/84814/ and I suspect that Wolfram may have skipped the thermal one as having been done by another patch (that I didn't merge, due to having questions about the whole issue, but that I don't have any objections to) But maybe there is also some limitation to the coccinelle rule too. Linus From janr@adax.com Mon Mar 15 23:27:56 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Mon, 15 Mar 2010 23:27:59 +0100 (CET) Received: from mail1.adax.com ([208.201.231.104]:27382 "EHLO mail1.adax.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492085Ab0COW14 (ORCPT ); Mon, 15 Mar 2010 23:27:56 +0100 Received: from static-151-204-189-187.pskn.east.verizon.net (static-151-204-189-187.pskn.east.verizon.net [151.204.189.187]) by mail1.adax.com (Postfix) with ESMTP id 42629120A09; Mon, 15 Mar 2010 15:27:53 -0700 (PDT) Received: from localhost (localhost [127.0.0.1]) by static-151-204-189-187.pskn.east.verizon.net (Postfix) with ESMTP id 4466D400589; Mon, 15 Mar 2010 18:27:52 -0400 (EDT) X-Virus-Scanned: amavisd-new at pskn.east.verizon.net Received: from static-151-204-189-187.pskn.east.verizon.net ([127.0.0.1]) by localhost (static-151-204-189-187.pskn.east.verizon.net [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 4P2EqqHx9vNe; Mon, 15 Mar 2010 18:27:26 -0400 (EDT) Received: from [192.168.1.76] (jr001327.mtl-nj.adax [192.168.1.76]) by static-151-204-189-187.pskn.east.verizon.net (Postfix) with ESMTP id 04B7C400588; Mon, 15 Mar 2010 18:27:25 -0400 (EDT) Message-ID: <4B9EB45D.8050106@adax.com> Date: Mon, 15 Mar 2010 18:27:41 -0400 From: Jan Rovins User-Agent: Thunderbird 2.0.0.23 (Windows/20090812) MIME-Version: 1.0 To: Peter 'p2' De Schrijver CC: linux-mips@linux-mips.org Subject: Re: linux 2.6.33 on movidis x16 References: <20100305141113.GD2437@apfelkorn> In-Reply-To: <20100305141113.GD2437@apfelkorn> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26237 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: janr@adax.com Precedence: bulk X-list: linux-mips Peter 'p2' De Schrijver wrote: > Hi, > > We are trying to get linux 2.6.33 to work on the movidis x16 board. Main thing > missing is the addresses of the PHYs. Does anyone know those ? > Unfortunately I don't have physical access to the board so I can't just look > for the PHY components being used. > > Thanks, > > Peter. > > Do you mean the Movidis x19 ? We have 2 of these in our lab. They are running the old 2.6.21.7 from the CnUsers 1.8.1 tool chain. They are currently being used by other developers for some application testing, but I can grab the serial console boot-up messages and send them to you, if the PHY info is in there. Jan From af@chaos-agency.de Tue Mar 16 12:35:58 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Tue, 16 Mar 2010 12:36:02 +0100 (CET) Received: from mail2.marcant.net ([217.14.160.186]:56961 "EHLO mail2.marcant.net" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492128Ab0CPLf6 (ORCPT ); Tue, 16 Mar 2010 12:35:58 +0100 Received: from localhost (mcontrol3.marcant.net [217.14.168.56]) by localhost (Postfix) with ESMTP id 72F917FA007; Tue, 16 Mar 2010 12:35:55 +0100 (CET) Received: from mail2.marcant.net ([127.0.0.1]) by localhost (mail2.marcant.net [127.0.0.1]) (amavisd-maia, port 10030) with ESMTP id 05730-01; Tue, 16 Mar 2010 12:35:51 +0100 (CET) Received: from admins.marcant.net (eth2.rossini.bi.marcant.net [217.14.160.189]) by mail2.marcant.net (Postfix) with ESMTP id 44C0C7FA003; Tue, 16 Mar 2010 12:35:51 +0100 (CET) Received: from sirius.intern.marcant.net (unknown [192.168.1.67]) by admins.marcant.net (Postfix) with ESMTP id 31B62180F4; Tue, 16 Mar 2010 12:35:51 +0100 (CET) Received: from aferber by sirius.intern.marcant.net with local (Exim 4.69) (envelope-from ) id 1NrV3v-0005WF-3J; Tue, 16 Mar 2010 12:35:51 +0100 Date: Tue, 16 Mar 2010 12:35:51 +0100 From: Andreas Ferber To: Ralf Baechle Cc: Markus Wigge , Michael Buesch , linux-mips@linux-mips.org Subject: [PATCH] mips: Fix SSB PCIcore IO resource management Message-ID: <20100316113551.GB19241@marcant.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Organization: MarcanT Internet-Services GmbH User-Agent: Mutt/1.5.18 (2008-05-17) X-Virus-Scanned: Maia Mailguard 1.0.2 Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26238 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: af@chaos-agency.de Precedence: bulk X-list: linux-mips The SSB PCIcore code reused the IO resource fixup code from the original 2.4.x Broadcom patch for BCM47xx based devices, which was a quick hack for doing PCI IO resource configuration back then (the boot loader doesn't configure PCI devices on this platform). However, this code is no longer necessary since the kernel now can do PCI resource management fine all by itself, so remove the old code. When removing the code, it becomes obvious that the mem_offset setting in the PCIcore driver was wrong, however this was masked by the fixup code before, except in a few cases involving yenta_socket. For BCM47xx, the correct offset is 0, and since this is the only device using PCIcore in host mode, the offset can simply be removed unconditionally. Signed-off-by: Andreas Ferber Signed-off-by: Michael Buesch --- The patch has been tested on the following devices: - Linksys WRT54G3G V1.0 - Asus WL500g Premium V1 - Asus WL500g Premium V2 - Netgear WGT634U Regards, Andreas Index: linux-2.6.33/drivers/ssb/driver_pcicore.c =================================================================== --- linux-2.6.33.orig/drivers/ssb/driver_pcicore.c 2010-03-15 14:52:55.000000000 +0100 +++ linux-2.6.33/drivers/ssb/driver_pcicore.c 2010-03-15 15:57:38.000000000 +0100 @@ -246,20 +246,12 @@ .pci_ops = &ssb_pcicore_pciops, .io_resource = &ssb_pcicore_io_resource, .mem_resource = &ssb_pcicore_mem_resource, - .mem_offset = 0x24000000, }; -static u32 ssb_pcicore_pcibus_iobase = 0x100; -static u32 ssb_pcicore_pcibus_membase = SSB_PCI_DMA; - /* This function is called when doing a pci_enable_device(). * We must first check if the device is a device on the PCI-core bridge. */ int ssb_pcicore_plat_dev_init(struct pci_dev *d) { - struct resource *res; - int pos, size; - u32 *base; - if (d->bus->ops != &ssb_pcicore_pciops) { /* This is not a device on the PCI-core bridge. */ return -ENODEV; @@ -268,27 +260,6 @@ ssb_printk(KERN_INFO "PCI: Fixing up device %s\n", pci_name(d)); - /* Fix up resource bases */ - for (pos = 0; pos < 6; pos++) { - res = &d->resource[pos]; - if (res->flags & IORESOURCE_IO) - base = &ssb_pcicore_pcibus_iobase; - else - base = &ssb_pcicore_pcibus_membase; - res->flags |= IORESOURCE_PCI_FIXED; - if (res->end) { - size = res->end - res->start + 1; - if (*base & (size - 1)) - *base = (*base + size) & ~(size - 1); - res->start = *base; - res->end = res->start + size - 1; - *base += size; - pci_write_config_dword(d, PCI_BASE_ADDRESS_0 + (pos << 2), res->start); - } - /* Fix up PCI bridge BAR0 only */ - if (d->bus->number == 0 && PCI_SLOT(d->devfn) == 0) - break; - } /* Fix up interrupt lines */ d->irq = ssb_mips_irq(extpci_core->dev) + 2; pci_write_config_byte(d, PCI_INTERRUPT_LINE, d->irq); From wuzhangjin@gmail.com Tue Mar 16 14:33:21 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Tue, 16 Mar 2010 14:33:29 +0100 (CET) Received: from mail-px0-f189.google.com ([209.85.216.189]:40186 "EHLO mail-px0-f189.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492148Ab0CPNdV (ORCPT ); Tue, 16 Mar 2010 14:33:21 +0100 Received: by pxi27 with SMTP id 27so5764682pxi.0 for ; Tue, 16 Mar 2010 06:33:13 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:from:to:cc:subject:date :message-id:x-mailer; bh=iXJNDgo0jqj0Q88c0SDA6bvZfoHtqcF5ff38MHsjqPQ=; b=HWtMuHeYBGdZuMKymJIQUKIZvDeOa7g+GZiwn8nV5U/0Dmo6APCLow6LYM4WyHtI2T GksUyGm7LIYWPe3iYsewq+IhIcs9x7Zd1PRVvNDhGyFqDXMh/Uz2+qSedMVCp+d9q1+a FRYx0NwMCfEOQVgM2Ob9LDMJW5XY3OFx+oaB8= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=from:to:cc:subject:date:message-id:x-mailer; b=hdwnTVW15zoM4klGbdIS0FYF25Ko9HJfcUxda1wS8M71nwaPxTOIvYXsmed/ewj+CI opAVxYj3wpHAcY5AToRgoRQngYHzf8UXURc/LSl4n2KmM0OYZzG+3yK3zqUzZ/sAy1nZ m8Njyqsu7AdJUqUmTwgzpZXWIeSSiHFbIy5hA= Received: by 10.142.248.4 with SMTP id v4mr913508wfh.258.1268746393255; Tue, 16 Mar 2010 06:33:13 -0700 (PDT) Received: from localhost.localdomain ([202.201.12.142]) by mx.google.com with ESMTPS id 22sm1141838pzk.9.2010.03.16.06.33.04 (version=TLSv1/SSLv3 cipher=RC4-MD5); Tue, 16 Mar 2010 06:33:12 -0700 (PDT) From: Wu Zhangjin To: Ralf Baechle Cc: Steven Rostedt , linux-mips@linux-mips.org, Wu Zhangjin Subject: [PATCH v4] MIPS: tracing: Optimize the implementation Date: Tue, 16 Mar 2010 21:26:25 +0800 Message-Id: X-Mailer: git-send-email 1.7.0.1 Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26239 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: wuzhangjin@gmail.com Precedence: bulk X-list: linux-mips From: Wu Zhangjin Changes: v3 -> v4: + Fixup of 'undefined ftrace_graph_caller' ftrace_graph_caller variable is only needed by CONFIG_FUNCTION_GRAPH_TRACER, so, wrap it with #ifdef. v2 -> v3: + Fixes the buidling error of the changes from v2. v1 -> v2: + Fixes the support of -mmcount-ra-address The offset of "b 1f" instruction should be 20 for leaf-function and 24 for non-leaf function when -mmcount-ra-address is enabled. This patch adds a new get_insn_b_1f() function to get the different "b 1f" instruction. (only test without -mmcount-ra-address.) v1: + Reduce the runtime overhead o Uses macros instead of variables for the fixed instructions to reduce memory access o Moves the Initilization of the instructions which will be fixed after linking from ftrace_make_nop/ftrace_make_call to ftrace_dyn_arch_init() and encodes the instructions through uasm(arch/mips/include/asm/uasm.h). o A common macro in_module() is defined to determine which space the instruction pointer stays in and several related conditional statements are converted to conditional operator(? :) statement. + Cleanup the whole stuff Lots of comments/macros have been cleaned up to let it look better. Signed-off-by: Wu Zhangjin --- arch/mips/include/asm/ftrace.h | 12 ++- arch/mips/kernel/ftrace.c | 244 ++++++++++++++++++++++++---------------- 2 files changed, 157 insertions(+), 99 deletions(-) diff --git a/arch/mips/include/asm/ftrace.h b/arch/mips/include/asm/ftrace.h index ce35c9a..6dbfa1e 100644 --- a/arch/mips/include/asm/ftrace.h +++ b/arch/mips/include/asm/ftrace.h @@ -3,7 +3,7 @@ * License. See the file "COPYING" in the main directory of this archive for * more details. * - * Copyright (C) 2009 DSLab, Lanzhou University, China + * Copyright (C) 2009, 2010 DSLab, Lanzhou University, China * Author: Wu Zhangjin */ @@ -19,6 +19,14 @@ extern void _mcount(void); #define mcount _mcount +/* + * If the Instruction Pointer is in module space (0xc0000000), return true; + * otherwise, it is in kernel space (0x80000000), return false. + * + * FIXME: This may not work in some cases. + */ +#define in_module(ip) (unlikely((ip) & 0x40000000)) + #define safe_load(load, src, dst, error) \ do { \ asm volatile ( \ @@ -83,8 +91,8 @@ static inline unsigned long ftrace_call_adjust(unsigned long addr) struct dyn_arch_ftrace { }; +#endif /* CONFIG_DYNAMIC_FTRACE */ -#endif /* CONFIG_DYNAMIC_FTRACE */ #endif /* __ASSEMBLY__ */ #endif /* CONFIG_FUNCTION_TRACER */ #endif /* _ASM_MIPS_FTRACE_H */ diff --git a/arch/mips/kernel/ftrace.c b/arch/mips/kernel/ftrace.c index e9e64e0..ae906b9 100644 --- a/arch/mips/kernel/ftrace.c +++ b/arch/mips/kernel/ftrace.c @@ -2,7 +2,7 @@ * Code for replacing ftrace calls with jumps. * * Copyright (C) 2007-2008 Steven Rostedt - * Copyright (C) 2009 DSLab, Lanzhou University, China + * Copyright (C) 2009, 2010 DSLab, Lanzhou University, China * Author: Wu Zhangjin * * Thanks goes to Steven Rostedt for writing the original x86 version. @@ -15,15 +15,52 @@ #include #include #include +#include + +#define INSN_S_R_SP 0xafb00000 /* s{d,w} R, offset(sp) */ #ifdef CONFIG_DYNAMIC_FTRACE -#define JAL 0x0c000000 /* jump & link: ip --> ra, jump to target */ -#define ADDR_MASK 0x03ffffff /* op_code|addr : 31...26|25 ....0 */ -#define jump_insn_encode(op_code, addr) \ - ((unsigned int)((op_code) | (((addr) >> 2) & ADDR_MASK))) +/* Before linking, the following instructions are fixed. */ +#ifdef CONFIG_CPU_LOONGSON2F +#define INSN_NOP 0x00200825 /* or at, at, zero */ +#else +#define INSN_NOP 0x00000000 /* nop */ +#endif +#define INSN_B_1F_16 0x10000004 /* b 1f; offset = (16 >> 2) */ +#define INSN_B_1F_20 0x10000005 /* b 1f; offset = (20 >> 2) */ +#define INSN_B_1F_24 0x10000006 /* b 1f; offset = (24 >> 2) */ + +/* After linking, the following instructions are fixed. */ +static unsigned int insn_jal_ftrace_caller __read_mostly; +static unsigned int insn_lui_v1_hi16_mcount __read_mostly; +static unsigned int insn_j_ftrace_graph_caller __maybe_unused __read_mostly; + +/* The following instructions are encoded in the run-time */ +/* insn: jal func; op_code|addr : 31...26|25 ....0 */ +#define INSN_JAL(addr) \ + ((unsigned int)(0x0c000000 | (((addr) >> 2) & 0x03ffffff))) -static unsigned int ftrace_nop = 0x00000000; +static inline void ftrace_dyn_arch_init_insns(void) +{ + u32 *buf; + unsigned int v1; + + /* lui v1, hi16_mcount */ + v1 = 3; + buf = (u32 *)&insn_lui_v1_hi16_mcount; + UASM_i_LA_mostly(&buf, v1, MCOUNT_ADDR); + + /* jal (ftrace_caller + 8), jump over the first two instruction */ + buf = (u32 *)&insn_jal_ftrace_caller; + uasm_i_jal(&buf, (FTRACE_ADDR + 8)); + +#ifdef CONFIG_FUNCTION_GRAPH_TRACER + /* j ftrace_graph_caller */ + buf = (u32 *)&insn_j_ftrace_graph_caller; + uasm_i_j(&buf, (unsigned long)ftrace_graph_caller); +#endif +} static int ftrace_modify_code(unsigned long ip, unsigned int new_code) { @@ -31,7 +68,6 @@ static int ftrace_modify_code(unsigned long ip, unsigned int new_code) /* *(unsigned int *)ip = new_code; */ safe_store_code(new_code, ip, faulted); - if (unlikely(faulted)) return -EFAULT; @@ -40,118 +76,130 @@ static int ftrace_modify_code(unsigned long ip, unsigned int new_code) return 0; } -static int lui_v1; -static int jal_mcount; +static inline unsigned int get_insn_b_1f(unsigned int *insn, unsigned long ip) +{ +#ifndef KBUILD_MCOUNT_RA_ADDRESS + *insn = INSN_B_1F_16; + return 0; +#else + int faulted; + unsigned int code; + + ip -= 4; + + /* code = *(unsigned int *)ip; */ + safe_load_code(code, ip, faulted); + if (unlikely(faulted)) + return -EFAULT; + + if ((code & INSN_S_R_SP) != INSN_S_R_SP) + *insn = INSN_B_1F_20; + else + *insn = INSN_B_1F_24; + + return 0; +#endif +} int ftrace_make_nop(struct module *mod, struct dyn_ftrace *rec, unsigned long addr) { unsigned int new; - int faulted; unsigned long ip = rec->ip; - /* We have compiled module with -mlong-calls, but compiled the kernel - * without it, we need to cope with them respectively. */ - if (ip & 0x40000000) { - /* record it for ftrace_make_call */ - if (lui_v1 == 0) { - /* lui_v1 = *(unsigned int *)ip; */ - safe_load_code(lui_v1, ip, faulted); - - if (unlikely(faulted)) - return -EFAULT; - } - - /* lui v1, hi_16bit_of_mcount --> b 1f (0x10000004) - * addiu v1, v1, low_16bit_of_mcount - * move at, ra - * jalr v1 - * nop - * 1f: (ip + 12) - */ - new = 0x10000004; - } else { - /* record/calculate it for ftrace_make_call */ - if (jal_mcount == 0) { - /* We can record it directly like this: - * jal_mcount = *(unsigned int *)ip; - * Herein, jump over the first two nop instructions */ - jal_mcount = jump_insn_encode(JAL, (MCOUNT_ADDR + 8)); - } - - /* move at, ra - * jalr v1 --> nop - */ - new = ftrace_nop; - } + /* + * We have compiled modules with -mlong-calls, but compiled kernel + * without it, therefore, need to cope with them respectively. + * + * For module: + * + * lui v1, hi16_mcount --> b 1f + * addiu v1, v1, lo16_mcount + * move at, ra + * jalr v1 + * nop + * 1f: (ip + 16) + * For kernel: + * + * move at, ra + * jal _mcount --> nop + * + * And with the -mmcount-ra-address option, the offset may be 20 for + * leaf fuction and 24 for non-leaf function. + */ + + if (!in_module(ip)) + new = INSN_NOP; + else if (get_insn_b_1f(&new, ip)) + return -EFAULT; + return ftrace_modify_code(ip, new); } -static int modified; /* initialized as 0 by default */ - int ftrace_make_call(struct dyn_ftrace *rec, unsigned long addr) { unsigned int new; unsigned long ip = rec->ip; - /* We just need to remove the "b ftrace_stub" at the fist time! */ - if (modified == 0) { - modified = 1; - ftrace_modify_code(addr, ftrace_nop); - } - /* ip, module: 0xc0000000, kernel: 0x80000000 */ - new = (ip & 0x40000000) ? lui_v1 : jal_mcount; + new = in_module(ip) ? insn_lui_v1_hi16_mcount : insn_jal_ftrace_caller; return ftrace_modify_code(ip, new); } -#define FTRACE_CALL_IP ((unsigned long)(&ftrace_call)) - int ftrace_update_ftrace_func(ftrace_func_t func) { unsigned int new; - new = jump_insn_encode(JAL, (unsigned long)func); + new = INSN_JAL((unsigned long)func); - return ftrace_modify_code(FTRACE_CALL_IP, new); + return ftrace_modify_code((unsigned long)(&ftrace_call), new); } int __init ftrace_dyn_arch_init(void *data) { + ftrace_dyn_arch_init_insns(); + + /* + * We are safe to remove the "b ftrace_stub" for the current + * ftrace_caller() is almost empty (only the stack operations), and + * most importantly, the calling to mcount will be disabled later in + * ftrace_init(), then there is no 'big' overhead. And in the future, + * we are hoping the function_trace_stop is initialized as 1 then we + * can eventually remove that 'useless' "b ftrace_stub" and the + * following nop. Currently, although we can call ftrace_stop() to set + * function_trace_stop as 1, but it will change the behavior of the + * Function Tracer. + */ + ftrace_modify_code(MCOUNT_ADDR, INSN_NOP); + /* The return code is retured via data */ *(unsigned long *)data = 0; return 0; } -#endif /* CONFIG_DYNAMIC_FTRACE */ +#endif /* CONFIG_DYNAMIC_FTRACE */ #ifdef CONFIG_FUNCTION_GRAPH_TRACER #ifdef CONFIG_DYNAMIC_FTRACE - extern void ftrace_graph_call(void); -#define JMP 0x08000000 /* jump to target directly */ -#define CALL_FTRACE_GRAPH_CALLER \ - jump_insn_encode(JMP, (unsigned long)(&ftrace_graph_caller)) #define FTRACE_GRAPH_CALL_IP ((unsigned long)(&ftrace_graph_call)) int ftrace_enable_ftrace_graph_caller(void) { return ftrace_modify_code(FTRACE_GRAPH_CALL_IP, - CALL_FTRACE_GRAPH_CALLER); + insn_j_ftrace_graph_caller); } int ftrace_disable_ftrace_graph_caller(void) { - return ftrace_modify_code(FTRACE_GRAPH_CALL_IP, ftrace_nop); + return ftrace_modify_code(FTRACE_GRAPH_CALL_IP, INSN_NOP); } - -#endif /* !CONFIG_DYNAMIC_FTRACE */ +#endif /* !CONFIG_DYNAMIC_FTRACE */ #ifndef KBUILD_MCOUNT_RA_ADDRESS -#define S_RA_SP (0xafbf << 16) /* s{d,w} ra, offset(sp) */ -#define S_R_SP (0xafb0 << 16) /* s{d,w} R, offset(sp) */ -#define OFFSET_MASK 0xffff /* stack offset range: 0 ~ PT_SIZE */ +#define INSN_S_RA_SP 0xafbf0000 /* s{d,w} ra, offset(sp) */ +#define STACK_OFFSET_MASK 0xffff /* stack offset range: 0 ~ PT_SIZE */ unsigned long ftrace_get_parent_addr(unsigned long self_addr, unsigned long parent, @@ -162,35 +210,35 @@ unsigned long ftrace_get_parent_addr(unsigned long self_addr, unsigned int code; int faulted; - /* in module or kernel? */ - if (self_addr & 0x40000000) { - /* module: move to the instruction "lui v1, HI_16BIT_OF_MCOUNT" */ - ip = self_addr - 20; - } else { - /* kernel: move to the instruction "move ra, at" */ - ip = self_addr - 12; - } + /* + * For module, move the ip from calling site of mcount to the + * instruction "lui v1, hi_16bit_of_mcount"(offset is 20), but for + * kernel, move to the instruction "move ra, at"(offset is 12) + */ + ip = self_addr - (in_module(self_addr) ? 20 : 12); - /* search the text until finding the non-store instruction or "s{d,w} - * ra, offset(sp)" instruction */ + /* + * search the text until finding the non-store instruction or "s{d,w} + * ra, offset(sp)" instruction + */ do { ip -= 4; /* get the code at "ip": code = *(unsigned int *)ip; */ safe_load_code(code, ip, faulted); - if (unlikely(faulted)) return 0; - /* If we hit the non-store instruction before finding where the + /* + * If we hit the non-store instruction before finding where the * ra is stored, then this is a leaf function and it does not - * store the ra on the stack. */ - if ((code & S_R_SP) != S_R_SP) + * store the ra on the stack. + */ + if ((code & INSN_S_R_SP) != INSN_S_R_SP) return parent_addr; + } while (((code & INSN_S_RA_SP) != INSN_S_RA_SP)); - } while (((code & S_RA_SP) != S_RA_SP)); - - sp = fp + (code & OFFSET_MASK); + sp = fp + (code & STACK_OFFSET_MASK); /* ra = *(unsigned long *)sp; */ safe_load_stack(ra, sp, faulted); @@ -201,8 +249,7 @@ unsigned long ftrace_get_parent_addr(unsigned long self_addr, return sp; return 0; } - -#endif +#endif /* !KBUILD_MCOUNT_RA_ADDRESS */ /* * Hook the return address and push it in the stack of return addrs @@ -211,16 +258,17 @@ unsigned long ftrace_get_parent_addr(unsigned long self_addr, void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr, unsigned long fp) { + int faulted; unsigned long old; struct ftrace_graph_ent trace; unsigned long return_hooker = (unsigned long) &return_to_handler; - int faulted; if (unlikely(atomic_read(¤t->tracing_graph_pause))) return; - /* "parent" is the stack address saved the return address of the caller + /* + * "parent" is the stack address saved the return address of the caller * of _mcount. * * if the gcc < 4.5, a leaf function does not save the return address @@ -241,12 +289,14 @@ void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr, if (unlikely(faulted)) goto out; #ifndef KBUILD_MCOUNT_RA_ADDRESS - parent = (unsigned long *)ftrace_get_parent_addr(self_addr, old, - (unsigned long)parent, - fp); - /* If fails when getting the stack address of the non-leaf function's - * ra, stop function graph tracer and return */ - if (parent == 0) + parent = (unsigned long *)ftrace_get_parent_addr( + self_addr, old, (unsigned long)parent, fp); + + /* + * If fails on getting the stack address of the non-leaf function's ra, + * stop function graph tracer and return + */ + if (unlikely(parent == 0)) goto out; #endif /* *parent = return_hooker; */ @@ -272,4 +322,4 @@ out: ftrace_graph_stop(); WARN_ON(1); } -#endif /* CONFIG_FUNCTION_GRAPH_TRACER */ +#endif /* CONFIG_FUNCTION_GRAPH_TRACER */ -- 1.7.0.1 From ralf@linux-mips.org Tue Mar 16 15:03:31 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Tue, 16 Mar 2010 15:03:35 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:52998 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492230Ab0CPODb (ORCPT ); Tue, 16 Mar 2010 15:03:31 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o2GE3UxG031216; Tue, 16 Mar 2010 15:03:30 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o2GE3Thv031211; Tue, 16 Mar 2010 15:03:29 +0100 Date: Tue, 16 Mar 2010 15:03:28 +0100 From: Ralf Baechle To: Florian Fainelli Cc: linux-mips@linux-mips.org, peter fuerst Subject: Re: [PATCH v2] MIPS: make CAC_ADDR and UNCAC_ADDR account for PHYS_OFFSET Message-ID: <20100316140328.GA20160@linux-mips.org> References: <201003100951.10028.florian@openwrt.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <201003100951.10028.florian@openwrt.org> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26240 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Wed, Mar 10, 2010 at 09:51:09AM +0100, Florian Fainelli wrote: > On AR7, we already redefine PHYS_OFFSET to match the system specifities, it is > however not sufficient when unsing dma_{map,unmap}_single, specifically in the > Ethernet driver, we must also adjust CAC_ADDR and UNCAC_ADDR for DMA to work > correctly. This patch fixes the following issue, seen in cpmac_open: Thanks, applied. Ralf From ralf@linux-mips.org Tue Mar 16 18:34:06 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Tue, 16 Mar 2010 18:34:10 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:55168 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492423Ab0CPReG (ORCPT ); Tue, 16 Mar 2010 18:34:06 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o2GHY2re004509; Tue, 16 Mar 2010 18:34:03 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o2GHY03S004501; Tue, 16 Mar 2010 18:34:00 +0100 Date: Tue, 16 Mar 2010 18:33:59 +0100 From: Ralf Baechle To: Andreas Ferber Cc: Markus Wigge , Michael Buesch , linux-mips@linux-mips.org Subject: Re: [PATCH] mips: Fix SSB PCIcore IO resource management Message-ID: <20100316173359.GB20160@linux-mips.org> References: <20100316113551.GB19241@marcant.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20100316113551.GB19241@marcant.net> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26241 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Tue, Mar 16, 2010 at 12:35:51PM +0100, Andreas Ferber wrote: > The SSB PCIcore code reused the IO resource fixup code from the original > 2.4.x Broadcom patch for BCM47xx based devices, which was a quick hack > for doing PCI IO resource configuration back then (the boot loader > doesn't configure PCI devices on this platform). > > However, this code is no longer necessary since the kernel now can do > PCI resource management fine all by itself, so remove the old code. > > When removing the code, it becomes obvious that the mem_offset setting > in the PCIcore driver was wrong, however this was masked by the fixup > code before, except in a few cases involving yenta_socket. For > BCM47xx, the correct offset is 0, and since this is the only device > using PCIcore in host mode, the offset can simply be removed > unconditionally. Thanks, applied. Ralf From ralf@linux-mips.org Tue Mar 16 19:09:55 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Tue, 16 Mar 2010 19:09:58 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:44124 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492429Ab0CPSJy (ORCPT ); Tue, 16 Mar 2010 19:09:54 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o2GI9oJT005527; Tue, 16 Mar 2010 19:09:50 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o2GI9kX9005524; Tue, 16 Mar 2010 19:09:46 +0100 Date: Tue, 16 Mar 2010 19:09:46 +0100 From: Ralf Baechle To: Yang Shi Cc: ddaney@caviumnetworks.com, ben-linux@fluff.org, khali@linux-fr.org, linux-mips@linux-mips.org, linux-i2c@vger.kernel.org Subject: Re: [PATCH V2] MIPS: Octeon: Register EEPROM device on the I2C bus Message-ID: <20100316180946.GC20160@linux-mips.org> References: <1268026190-18300-1-git-send-email-yang.shi@windriver.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1268026190-18300-1-git-send-email-yang.shi@windriver.com> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26242 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Mon, Mar 08, 2010 at 01:29:50PM +0800, Yang Shi wrote: > An SPD resides on 0x50 of the I2C bus on CN56xx/57xx board, > register this device. I wonder what the use case for this patch is? Normally Linux doesn't care about SPD. I also wonder how this will work for configurations with multiple memory modules thus multiple SPD EEPROMS. Ralf From khali@linux-fr.org Tue Mar 16 20:06:49 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Tue, 16 Mar 2010 20:06:54 +0100 (CET) Received: from poutre.nerim.net ([62.4.16.124]:60343 "EHLO poutre.nerim.net" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492315Ab0CPTGt (ORCPT ); Tue, 16 Mar 2010 20:06:49 +0100 Received: from localhost (localhost [127.0.0.1]) by poutre.nerim.net (Postfix) with ESMTP id BFF8439DFE0; Tue, 16 Mar 2010 20:06:45 +0100 (CET) X-Virus-Scanned: amavisd-new at nerim.net Received: from poutre.nerim.net ([127.0.0.1]) by localhost (poutre.nerim.net [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 7KvgxTmKmJqe; Tue, 16 Mar 2010 20:06:44 +0100 (CET) Received: from hyperion.delvare (jdelvare.pck.nerim.net [62.212.121.182]) by poutre.nerim.net (Postfix) with ESMTP id 860B939DFE6; Tue, 16 Mar 2010 20:06:44 +0100 (CET) Date: Tue, 16 Mar 2010 20:06:47 +0100 From: Jean Delvare To: Ralf Baechle Cc: Yang Shi , ddaney@caviumnetworks.com, ben-linux@fluff.org, linux-mips@linux-mips.org, linux-i2c@vger.kernel.org Subject: Re: [PATCH V2] MIPS: Octeon: Register EEPROM device on the I2C bus Message-ID: <20100316200647.3803edf1@hyperion.delvare> In-Reply-To: <20100316180946.GC20160@linux-mips.org> References: <1268026190-18300-1-git-send-email-yang.shi@windriver.com> <20100316180946.GC20160@linux-mips.org> X-Mailer: Claws Mail 3.5.0 (GTK+ 2.14.4; i586-suse-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26243 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: khali@linux-fr.org Precedence: bulk X-list: linux-mips Hi Ralf, On Tue, 16 Mar 2010 19:09:46 +0100, Ralf Baechle wrote: > On Mon, Mar 08, 2010 at 01:29:50PM +0800, Yang Shi wrote: > > > An SPD resides on 0x50 of the I2C bus on CN56xx/57xx board, > > register this device. > > I wonder what the use case for this patch is? Normally Linux doesn't care > about SPD. The Linux kernel doesn't care, but user-space may. As a matter of fact, there is a script out there (decode-dimms, in the i2c-tools package) decoding the SPD data and presenting it to the user. Some people want to know the details about their memory modules. > I also wonder how this will work for configurations with multiple memory > modules thus multiple SPD EEPROMS. The kernel code should instantiate one spd device per memory module (assuming they are all reachable.) Obviously this can't be done in a static way. -- Jean Delvare From ralf@linux-mips.org Tue Mar 16 20:55:32 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Tue, 16 Mar 2010 20:55:36 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:50123 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492434Ab0CPTzc (ORCPT ); Tue, 16 Mar 2010 20:55:32 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o2GJtUP4008332; Tue, 16 Mar 2010 20:55:30 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o2GJtTgZ008325; Tue, 16 Mar 2010 20:55:29 +0100 Date: Tue, 16 Mar 2010 20:55:29 +0100 From: Ralf Baechle To: David Daney Cc: linux-mips@linux-mips.org Subject: Re: [PATCH 1/3] MIPS: Add SYSCALL to uasm. Message-ID: <20100316195529.GD20160@linux-mips.org> References: <1266538385-29088-1-git-send-email-ddaney@caviumnetworks.com> <1266538385-29088-2-git-send-email-ddaney@caviumnetworks.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1266538385-29088-2-git-send-email-ddaney@caviumnetworks.com> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26244 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Thu, Feb 18, 2010 at 04:13:03PM -0800, David Daney wrote: Thanks, applied. Ralf From ralf@linux-mips.org Tue Mar 16 20:55:54 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Tue, 16 Mar 2010 20:56:01 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:50131 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492504Ab0CPTzx (ORCPT ); Tue, 16 Mar 2010 20:55:53 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o2GJtpAM008347; Tue, 16 Mar 2010 20:55:51 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o2GJtpP1008346; Tue, 16 Mar 2010 20:55:51 +0100 Date: Tue, 16 Mar 2010 20:55:51 +0100 From: Ralf Baechle To: David Daney Cc: linux-mips@linux-mips.org Subject: Re: [PATCH 2/3] MIPS: Preliminary vdso. Message-ID: <20100316195551.GE20160@linux-mips.org> References: <1266538385-29088-1-git-send-email-ddaney@caviumnetworks.com> <1266538385-29088-3-git-send-email-ddaney@caviumnetworks.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1266538385-29088-3-git-send-email-ddaney@caviumnetworks.com> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26245 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Thu, Feb 18, 2010 at 04:13:04PM -0800, David Daney wrote: > This is a preliminary patch to add a vdso to all user processes. > Still missing are ELF headers and .eh_frame information. But it is > enough to allow us to move signal trampolines off of the stack. Note > that emulation of branch delay slots in the FPU emulator still > requires the stack. > > We allocate a single page (the vdso) and write all possible signal > trampolines into it. The stack is moved down by one page and the vdso > is mapped into this space. Thanks, applied. Ralf From ralf@linux-mips.org Tue Mar 16 20:56:22 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Tue, 16 Mar 2010 20:56:26 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:50338 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492490Ab0CPT4W (ORCPT ); Tue, 16 Mar 2010 20:56:22 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o2GJuKgw008382; Tue, 16 Mar 2010 20:56:20 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o2GJuKcS008381; Tue, 16 Mar 2010 20:56:20 +0100 Date: Tue, 16 Mar 2010 20:56:20 +0100 From: Ralf Baechle To: David Daney Cc: linux-mips@linux-mips.org Subject: Re: [PATCH 3/3] MIPS: Move signal trampolines off of the stack. Message-ID: <20100316195619.GF20160@linux-mips.org> References: <1266538385-29088-1-git-send-email-ddaney@caviumnetworks.com> <1266538385-29088-4-git-send-email-ddaney@caviumnetworks.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1266538385-29088-4-git-send-email-ddaney@caviumnetworks.com> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26246 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Thu, Feb 18, 2010 at 04:13:05PM -0800, David Daney wrote: > This is a follow on to the vdso patch. > > Since all processes now have signal trampolines permanently mapped, we > can use those instead of putting the trampoline on the stack and > invalidating the corresponding icache across all CPUs. We also get > rid of a bunch of ICACHE_REFILLS_WORKAROUND_WAR code. Thanks, applied. Ralf From rkota@broadcom.com Wed Mar 17 06:27:46 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 17 Mar 2010 06:27:50 +0100 (CET) Received: from mms3.broadcom.com ([216.31.210.19]:2870 "EHLO MMS3.broadcom.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491033Ab0CQF1q convert rfc822-to-8bit (ORCPT ); Wed, 17 Mar 2010 06:27:46 +0100 Received: from [10.16.192.224] by MMS3.broadcom.com with ESMTP (Broadcom SMTP Relay (Email Firewall v6.3.2)); Tue, 16 Mar 2010 22:27:26 -0700 X-Server-Uuid: B55A25B1-5D7D-41F8-BC53-C57E7AD3C201 Received: from SJEXCHCCR01.corp.ad.broadcom.com ([10.252.49.130]) by SJEXCHHUB01.corp.ad.broadcom.com ([10.16.192.224]) with mapi; Tue, 16 Mar 2010 22:27:26 -0700 From: "Ramgopal Kota" To: "linux-mips@linux-mips.org" Date: Tue, 16 Mar 2010 22:27:24 -0700 Subject: RE: MIPS raw_local_irq_restore flags Thread-Topic: MIPS raw_local_irq_restore flags Thread-Index: AcrBK3wkk2FBNl+9TaOxNHDqJxz80AEZusdw Message-ID: <6C370B347C3FE8438C9692873287D2E1109DDF0057@SJEXCHCCR01.corp.ad.broadcom.com> References: <4BEA3FF3CAA35E408EA55C7BE2E61D0546AC862322@xmail3.se.axis.com> In-Reply-To: <4BEA3FF3CAA35E408EA55C7BE2E61D0546AC862322@xmail3.se.axis.com> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: acceptlanguage: en-US MIME-Version: 1.0 X-WSS-ID: 67BEB7B431G62138311-01-01 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 8BIT Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26247 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: rkota@broadcom.com Precedence: bulk X-list: linux-mips Hi , Is there any moderation done to the messages posted to this mailing list. I posted couple of messages but none has come , just wondering what is wrong ? Thanks & Regards, Ramgopal Kota -----Original Message----- From: linux-mips-bounce@linux-mips.org [mailto:linux-mips-bounce@linux-mips.org] On Behalf Of Mikael Starvik Sent: Thursday, March 11, 2010 8:30 PM To: linux-mips@linux-mips.org Subject: MIPS raw_local_irq_restore flags For the common case CONFIG_CPU_MIPSR2 && CONFIG_IRQ_CPU raw_local_irq_restore_flags is defined as: " beqz \\flags, 1f \n" " di \n" " ei \n" "1: \n" Doesn't this imply that you can't do recursive local_irq_save() (with different locks ofcourse)? Best Regards /Mikael From rkota@broadcom.com Wed Mar 17 07:05:03 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 17 Mar 2010 07:05:07 +0100 (CET) Received: from mms3.broadcom.com ([216.31.210.19]:3756 "EHLO MMS3.broadcom.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491080Ab0CQGFD convert rfc822-to-8bit (ORCPT ); Wed, 17 Mar 2010 07:05:03 +0100 Received: from [10.16.192.232] by MMS3.broadcom.com with ESMTP (Broadcom SMTP Relay (Email Firewall v6.3.2)); Tue, 16 Mar 2010 23:04:51 -0700 X-Server-Uuid: B55A25B1-5D7D-41F8-BC53-C57E7AD3C201 Received: from SJEXCHCCR01.corp.ad.broadcom.com ([10.252.49.130]) by SJEXCHHUB02.corp.ad.broadcom.com ([10.16.192.232]) with mapi; Tue, 16 Mar 2010 23:04:51 -0700 From: "Ramgopal Kota" To: "linux-mips@linux-mips.org" Date: Tue, 16 Mar 2010 23:04:49 -0700 Subject: HIGHMEM not working. Thread-Topic: HIGHMEM not working. Thread-Index: AcrBK3wkk2FBNl+9TaOxNHDqJxz80AEZusdwAAEpc2A= Message-ID: <6C370B347C3FE8438C9692873287D2E1109DDF0073@SJEXCHCCR01.corp.ad.broadcom.com> References: <4BEA3FF3CAA35E408EA55C7BE2E61D0546AC862322@xmail3.se.axis.com> <6C370B347C3FE8438C9692873287D2E1109DDF0057@SJEXCHCCR01.corp.ad.broadcom.com> In-Reply-To: <6C370B347C3FE8438C9692873287D2E1109DDF0057@SJEXCHCCR01.corp.ad.broadcom.com> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: acceptlanguage: en-US MIME-Version: 1.0 X-WSS-ID: 67BEAE8931G62159373-01-01 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 8BIT Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26248 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: rkota@broadcom.com Precedence: bulk X-list: linux-mips Hi, I need help in running CPU with HIGH_MEM. Below is the processor memory map .. It is base on MIPS64K Core. 0x0000_0000 -- 0x07FF_FFFF -> RAM (Region 1) 0x0800_0000 -- 0x0FFF_FFFF -> PCI BUS 1 0x2000_0000 -- 0x2FFF_FFFF -> Flash Region 0x4000_0000 -- 0x47FF_FFFF -> PCI BUS 2 0x8000_0000 -- 0xCFFF_FFFF -> RAM (Region 2) the first 128MB is remapped into 0x8000_0000 .. The system has 256MB of PHYSICAL RAM .. I configured TLB's and added memory regions. The system boots up correctly .. If I show cat /proc/meminfo it shows 256MB , it could allocate memory ( I verified with a simple program allocating the memory and writing some data) There is a PCI MAC device on PCI BUS 2. The CPU is not able to perform DMA to/from device. The MAC driver is allocating DMA buffers using the following code .. { phy = virt_to_bus(high_memory); << Resulting in 0xa000_0000; virt = ioremap(pbase); } The Linux configuration has no ZONE_DMA configuration .. Few questions .. A) HIGH_MEMORY START Address is configured to 0x2000_0000 , is it correct value. B) Are there any data which can be dumped to know more information ? C) Do I need to enable MEMORY Discontinuous Config also ? ( As I did not see a reason to enable this as memory in Region 2 is contiguous) D) I read in some articles on internet that if HIGH_MEM is used,PCI bounce buffers needs to be used ? Can some body point me how to set this up ? Thanks & Regards, Ramgopal Kota From jiang.adam@gmail.com Wed Mar 17 12:30:03 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 17 Mar 2010 12:30:07 +0100 (CET) Received: from mail-qy0-f180.google.com ([209.85.221.180]:62829 "EHLO mail-qy0-f180.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491972Ab0CQLaD convert rfc822-to-8bit (ORCPT ); Wed, 17 Mar 2010 12:30:03 +0100 Received: by qyk10 with SMTP id 10so540188qyk.6 for ; Wed, 17 Mar 2010 04:29:55 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:in-reply-to:references :date:message-id:subject:from:to:cc:content-type :content-transfer-encoding; bh=nda2sQZEqHrRes3iHgkimqtRd9vN6MwYSZ91qktZ0HE=; b=jJZaBlYwZWZ5FBh8rDkwxnqxiGgyp5puNBVVsLND5w/Iyrs6X1oZPeMefTT/VJaFF1 MIM1Zkze7nY/ZkL0BqI/gKaf1aF1rzONBOSddqLtmMJCjGzyycOaUB8cnwg2hEb0ZMNq v7m5huyjsF/dfIH5LneVkx1Pkko0wQIZNZYdg= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type:content-transfer-encoding; b=LgBMiPVQGFYt5ZlTI3MIGpW2LuK2Zn6L/JHT3YGfAdmDTPZ6bzAV1QoVPpCu4yLEdM B99bdUedYi2jClFrt3s31+UQTrVD9nPgjK+zBHlaffIKlAMTHJ2tu0A9Ee0ycHP4/I9y Rs6ur/q38jWxnA8p83KIW2UAjD6yepGX6jY6U= MIME-Version: 1.0 Received: by 10.224.87.159 with SMTP id w31mr210887qal.50.1268825393655; Wed, 17 Mar 2010 04:29:53 -0700 (PDT) In-Reply-To: <6C370B347C3FE8438C9692873287D2E1109DDF015B@SJEXCHCCR01.corp.ad.broadcom.com> References: <4BEA3FF3CAA35E408EA55C7BE2E61D0546AC862322@xmail3.se.axis.com> <6C370B347C3FE8438C9692873287D2E1109DDF0057@SJEXCHCCR01.corp.ad.broadcom.com> <6C370B347C3FE8438C9692873287D2E1109DDF0073@SJEXCHCCR01.corp.ad.broadcom.com> <1864acd31003170006q602602a9j5675964b886b0a55@mail.gmail.com> <6C370B347C3FE8438C9692873287D2E1109DDF015B@SJEXCHCCR01.corp.ad.broadcom.com> Date: Wed, 17 Mar 2010 20:29:53 +0900 Message-ID: <1864acd31003170429l7b0972fcn46a6404674ad62d9@mail.gmail.com> Subject: Re: HIGHMEM not working. From: Adam Jiang To: "linux-mips@linux-mips.org" Cc: Ramgopal Kota Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8BIT Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26249 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: jiang.adam@gmail.com Precedence: bulk X-list: linux-mips 2010/3/17 Ramgopal Kota > > Adam, > > Thanks for the response ... > > We enabled CONFIG_SPARSEMEM in place of CONFIG_DISCONTIGMEM .. It is specified that SPARSEMEM will eventually replaces DISCONTIGMEM > (http://marc.info/?l=lhms-devel&m=111451976115580&w=2) Yes. It seems a better way to do this. > We have configured zones as below in paging_init function .. > max_zone_pfns[ZONE_DMA] = MAX_DMA_PFN; > max_zone_pfns[ZONE_NORMAL] = max_low_pfn;   <<  0x20000 > max_zone_pfns[ZONE_HIGHMEM] = highend_pfn;  <<  0x8FFFF > We added 2 memory regions as below .. > add_memory_region(0x00000000, 128MB, BOOT_MEM_RAM); > add_memory_region(0x88000000  ,128MB, BOOT_MEM_RAM); > Is this what you are refering as 2 memory nodes ?? > OK. Physical memory is managed in nodes and zones. Because you have mentioned that there would be 2 ranges of memory on your architecture like Node 0   [0x0000 0000, 0x0800 0000] Node 1   [0x8000 0000, 0xcfff ffff] I guess you can define these two nodes by add_memory_region function but I am not sure... However, the start_addr of the second node would be 0x8000 0000 rather than 0x8800 0000. On the other hand, I think the following code is not correct > max_zone_pfns[ZONE_DMA] = MAX_DMA_PFN; > max_zone_pfns[ZONE_NORMAL] = max_low_pfn;   <<  0x20000 > max_zone_pfns[ZONE_HIGHMEM] = highend_pfn;  <<  0x8FFFF if max_low_pfn is 0x20000, then your ZONE_NORMAL will end up at 0x2000 0000(PFN x PAGE_SIZE); but according to the information about physical memory address you provided. 0x0800 0000 ~ 0x0fff ffff is used as PCI device addresses. Em~, the highend_pfn is not correct either. Please double check these address. Bounce buffers are required for devices that cannnot access the full range of memory available to the CPU, such as old ISA cards with only 24bit address. Is this the situation you have to face? If it is needed you may look into function create_bounce_buffer() in the other drivers for details, however, I still think you would not need it. By the way, high memory just mapped to low memory by kmap() function. You don't need to define a ZONE_HIGH in your configuration. Because 0x8000_0000 ~ 0x8fff ffff is kseg0 on MIPS. You can handle them as ZONE_NORMAL in much easier way. Best regards, /Adam > > What is the significance of MAX_PHYSMEM_BITS  & SECTION_SIZE_BITS .. Sorry for troubling with novoice questions .. > > Thanks & Regards, > Ramgopal Kota > ________________________________ > From: Adam Jiang [mailto:jiang.adam@gmail.com] > Sent: Wednesday, March 17, 2010 12:36 PM > To: Ramgopal Kota > Subject: Re: HIGHMEM not working. > > 2010/3/17 Ramgopal Kota >> >>  Hi, >> >> I need help in running CPU with HIGH_MEM. >> Below is the processor memory map .. It is base on MIPS64K Core. >> >> 0x0000_0000  -- 0x07FF_FFFF   -> RAM (Region 1) >> 0x0800_0000  -- 0x0FFF_FFFF   -> PCI BUS 1 >> 0x2000_0000  -- 0x2FFF_FFFF   -> Flash Region >> 0x4000_0000  -- 0x47FF_FFFF   -> PCI BUS 2 >> 0x8000_0000  -- 0xCFFF_FFFF   -> RAM (Region 2) the first 128MB is remapped into 0x8000_0000 .. >> >> The system has 256MB of PHYSICAL RAM .. >> >> I configured TLB's and added memory regions. The system boots up correctly .. >> If I show cat /proc/meminfo it shows 256MB , it could allocate memory ( I verified with a simple program allocating the memory and writing some data) >> >> There is a PCI MAC device on PCI BUS 2. The CPU is not able to perform DMA to/from device. >> The MAC driver is allocating DMA buffers using the following code .. >> { >>    phy = virt_to_bus(high_memory);  << Resulting in 0xa000_0000; >>    virt = ioremap(pbase); >> } >> The Linux configuration has no ZONE_DMA configuration .. >> >> >> Few questions .. >>  A) HIGH_MEMORY START Address is configured to 0x2000_0000 , is it correct value. > > I don't think 0x2000_0000 is a correct value for HIGH_MEM in general meaning. The value depends on which architecture you're working on. And its start address could be initialed in > LNX/arch/(arch)/kernel/setup.c > LNX/arch/(arch)/mm/init.c > in your case, I thing you can use all your memory as ZONE_NORMAL or ZONE_DMA but not ZONE_HIGH. > >> >>  B) Are there any data which can be dumped to know more information ? > > Yes, of course. You could try to add some printk in the above two files in bootmem_init() function or function free_area_init_core() to see what's the memory layout exactly. > >> >>  C) Do I need to enable MEMORY Discontinuous Config also ?  ( As I did not see a reason to enable this as memory in Region 2 is contiguous) > > Since you have two separated range fsyc memory on you board, I think you do really need CONFIG_DISCONTIGMEM_*; You may also need to define two memory nodes in bootmem. > >> >>  D) I read in some articles on internet that if HIGH_MEM is used,PCI bounce buffers needs to be used ? Can some body point me how to set this up ? > > Usually, DMA data transferring could perform directly on fysc memory address. You may not need bounce buffer or scatter/gather operations if you can config all your memory to ZONE_NORMAL. ZONE_DMA is not necessary because this area is only required by kmalloc call with GFP_DMA flags. > Best regards, > /Adam > >> >> Thanks & Regards, >> Ramgopal Kota >> >> > > -- > Adam Jiang > ----------------------------------- > e-mail:jiang.adam@gmail.com > http://www.adamjiang.com > ----------------------------------- -- Adam Jiang ----------------------------------- e-mail:jiang.adam@gmail.com http://www.adamjiang.com ----------------------------------- From rkota@broadcom.com Wed Mar 17 13:07:13 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 17 Mar 2010 13:07:16 +0100 (CET) Received: from mms3.broadcom.com ([216.31.210.19]:3622 "EHLO MMS3.broadcom.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491768Ab0CQMHN convert rfc822-to-8bit (ORCPT ); Wed, 17 Mar 2010 13:07:13 +0100 Received: from [10.16.192.224] by MMS3.broadcom.com with ESMTP (Broadcom SMTP Relay (Email Firewall v6.3.2)); Wed, 17 Mar 2010 05:06:50 -0700 X-Server-Uuid: B55A25B1-5D7D-41F8-BC53-C57E7AD3C201 Received: from SJEXCHCCR01.corp.ad.broadcom.com ([10.252.49.130]) by SJEXCHHUB01.corp.ad.broadcom.com ([10.16.192.224]) with mapi; Wed, 17 Mar 2010 05:06:50 -0700 From: "Ramgopal Kota" To: "Adam Jiang" , "linux-mips@linux-mips.org" Date: Wed, 17 Mar 2010 05:06:47 -0700 Subject: RE: HIGHMEM not working. Thread-Topic: HIGHMEM not working. Thread-Index: AcrFxTYLt1ryOz20RK+pVqNgRQr2PwAA8FPw Message-ID: <6C370B347C3FE8438C9692873287D2E1109DDF01AD@SJEXCHCCR01.corp.ad.broadcom.com> References: <4BEA3FF3CAA35E408EA55C7BE2E61D0546AC862322@xmail3.se.axis.com> <6C370B347C3FE8438C9692873287D2E1109DDF0057@SJEXCHCCR01.corp.ad.broadcom.com> <6C370B347C3FE8438C9692873287D2E1109DDF0073@SJEXCHCCR01.corp.ad.broadcom.com> <1864acd31003170006q602602a9j5675964b886b0a55@mail.gmail.com> <6C370B347C3FE8438C9692873287D2E1109DDF015B@SJEXCHCCR01.corp.ad.broadcom.com> <1864acd31003170429l7b0972fcn46a6404674ad62d9@mail.gmail.com> In-Reply-To: <1864acd31003170429l7b0972fcn46a6404674ad62d9@mail.gmail.com> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: acceptlanguage: en-US MIME-Version: 1.0 X-WSS-ID: 67BE1A5031G62437712-01-01 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8BIT Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26250 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: rkota@broadcom.com Precedence: bulk X-list: linux-mips Adam, The first 128MB of 0x8000_0000 is mapped to 0x0000_0000 address, hence I added them as 2 nodes with 0 & 0x8800_0000 .. max_zone_pfns[ZONE_NORMAL] should be assigned with maximum physical address of the memory i.e. address >> page_size ... As the physical address is 0x07FF_FFFF .. The ZONE_NORMAL PFN should be 0x07FFF .. As ZONE_HIGH max address is 0x8FFF_FFFF .. The ZONE_HIGH PFN should be 0x8FFFF .. Why do you think this is wrong ? I did not get how I can handle with out ZONE_HIGH ?? Can you explain me in more detail please .. Thanks & Regards, Ramgopal Kota -----Original Message----- From: Adam Jiang [mailto:jiang.adam@gmail.com] Sent: Wednesday, March 17, 2010 5:00 PM To: linux-mips@linux-mips.org Cc: Ramgopal Kota Subject: Re: HIGHMEM not working. 2010/3/17 Ramgopal Kota > > Adam, > > Thanks for the response ... > > We enabled CONFIG_SPARSEMEM in place of CONFIG_DISCONTIGMEM .. It is specified that SPARSEMEM will eventually replaces DISCONTIGMEM > (http://marc.info/?l=lhms-devel&m=111451976115580&w=2) Yes. It seems a better way to do this. > We have configured zones as below in paging_init function .. > max_zone_pfns[ZONE_DMA] = MAX_DMA_PFN; > max_zone_pfns[ZONE_NORMAL] = max_low_pfn;   <<  0x20000 > max_zone_pfns[ZONE_HIGHMEM] = highend_pfn;  <<  0x8FFFF > We added 2 memory regions as below .. > add_memory_region(0x00000000, 128MB, BOOT_MEM_RAM); > add_memory_region(0x88000000  ,128MB, BOOT_MEM_RAM); > Is this what you are refering as 2 memory nodes ?? > OK. Physical memory is managed in nodes and zones. Because you have mentioned that there would be 2 ranges of memory on your architecture like Node 0   [0x0000 0000, 0x0800 0000] Node 1   [0x8000 0000, 0xcfff ffff] I guess you can define these two nodes by add_memory_region function but I am not sure... However, the start_addr of the second node would be 0x8000 0000 rather than 0x8800 0000. On the other hand, I think the following code is not correct > max_zone_pfns[ZONE_DMA] = MAX_DMA_PFN; > max_zone_pfns[ZONE_NORMAL] = max_low_pfn;   <<  0x20000 > max_zone_pfns[ZONE_HIGHMEM] = highend_pfn;  <<  0x8FFFF if max_low_pfn is 0x20000, then your ZONE_NORMAL will end up at 0x2000 0000(PFN x PAGE_SIZE); but according to the information about physical memory address you provided. 0x0800 0000 ~ 0x0fff ffff is used as PCI device addresses. Em~, the highend_pfn is not correct either. Please double check these address. Bounce buffers are required for devices that cannnot access the full range of memory available to the CPU, such as old ISA cards with only 24bit address. Is this the situation you have to face? If it is needed you may look into function create_bounce_buffer() in the other drivers for details, however, I still think you would not need it. By the way, high memory just mapped to low memory by kmap() function. You don't need to define a ZONE_HIGH in your configuration. Because 0x8000_0000 ~ 0x8fff ffff is kseg0 on MIPS. You can handle them as ZONE_NORMAL in much easier way. Best regards, /Adam > > What is the significance of MAX_PHYSMEM_BITS  & SECTION_SIZE_BITS .. Sorry for troubling with novoice questions .. > > Thanks & Regards, > Ramgopal Kota > ________________________________ > From: Adam Jiang [mailto:jiang.adam@gmail.com] > Sent: Wednesday, March 17, 2010 12:36 PM > To: Ramgopal Kota > Subject: Re: HIGHMEM not working. > > 2010/3/17 Ramgopal Kota >> >>  Hi, >> >> I need help in running CPU with HIGH_MEM. >> Below is the processor memory map .. It is base on MIPS64K Core. >> >> 0x0000_0000  -- 0x07FF_FFFF   -> RAM (Region 1) >> 0x0800_0000  -- 0x0FFF_FFFF   -> PCI BUS 1 >> 0x2000_0000  -- 0x2FFF_FFFF   -> Flash Region >> 0x4000_0000  -- 0x47FF_FFFF   -> PCI BUS 2 >> 0x8000_0000  -- 0xCFFF_FFFF   -> RAM (Region 2) the first 128MB is remapped into 0x8000_0000 .. >> >> The system has 256MB of PHYSICAL RAM .. >> >> I configured TLB's and added memory regions. The system boots up correctly .. >> If I show cat /proc/meminfo it shows 256MB , it could allocate memory ( I verified with a simple program allocating the memory and writing some data) >> >> There is a PCI MAC device on PCI BUS 2. The CPU is not able to perform DMA to/from device. >> The MAC driver is allocating DMA buffers using the following code .. >> { >>    phy = virt_to_bus(high_memory);  << Resulting in 0xa000_0000; >>    virt = ioremap(pbase); >> } >> The Linux configuration has no ZONE_DMA configuration .. >> >> >> Few questions .. >>  A) HIGH_MEMORY START Address is configured to 0x2000_0000 , is it correct value. > > I don't think 0x2000_0000 is a correct value for HIGH_MEM in general meaning. The value depends on which architecture you're working on. And its start address could be initialed in > LNX/arch/(arch)/kernel/setup.c > LNX/arch/(arch)/mm/init.c > in your case, I thing you can use all your memory as ZONE_NORMAL or ZONE_DMA but not ZONE_HIGH. > >> >>  B) Are there any data which can be dumped to know more information ? > > Yes, of course. You could try to add some printk in the above two files in bootmem_init() function or function free_area_init_core() to see what's the memory layout exactly. > >> >>  C) Do I need to enable MEMORY Discontinuous Config also ?  ( As I did not see a reason to enable this as memory in Region 2 is contiguous) > > Since you have two separated range fsyc memory on you board, I think you do really need CONFIG_DISCONTIGMEM_*; You may also need to define two memory nodes in bootmem. > >> >>  D) I read in some articles on internet that if HIGH_MEM is used,PCI bounce buffers needs to be used ? Can some body point me how to set this up ? > > Usually, DMA data transferring could perform directly on fysc memory address. You may not need bounce buffer or scatter/gather operations if you can config all your memory to ZONE_NORMAL. ZONE_DMA is not necessary because this area is only required by kmalloc call with GFP_DMA flags. > Best regards, > /Adam > >> >> Thanks & Regards, >> Ramgopal Kota >> >> > > -- > Adam Jiang > ----------------------------------- > e-mail:jiang.adam@gmail.com > http://www.adamjiang.com > ----------------------------------- -- Adam Jiang ----------------------------------- e-mail:jiang.adam@gmail.com http://www.adamjiang.com ----------------------------------- From ralf@linux-mips.org Wed Mar 17 14:03:26 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 17 Mar 2010 14:03:29 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:56401 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1491753Ab0CQND0 (ORCPT ); Wed, 17 Mar 2010 14:03:26 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o2HD3OKA003694; Wed, 17 Mar 2010 14:03:24 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o2HD3NCA003692; Wed, 17 Mar 2010 14:03:23 +0100 Date: Wed, 17 Mar 2010 14:03:23 +0100 From: Ralf Baechle To: Ramgopal Kota Cc: "linux-mips@linux-mips.org" Subject: Re: MIPS raw_local_irq_restore flags Message-ID: <20100317130323.GG20160@linux-mips.org> References: <4BEA3FF3CAA35E408EA55C7BE2E61D0546AC862322@xmail3.se.axis.com> <6C370B347C3FE8438C9692873287D2E1109DDF0057@SJEXCHCCR01.corp.ad.broadcom.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <6C370B347C3FE8438C9692873287D2E1109DDF0057@SJEXCHCCR01.corp.ad.broadcom.com> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26251 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Tue, Mar 16, 2010 at 10:27:24PM -0700, Ramgopal Kota wrote: > Is there any moderation done to the messages posted to this mailing list. I posted couple of messages but none has come , just wondering what is wrong ? There is no general moderation for any of the lists on linux-mips but your postings were false positives of the spamfilter and as such ended up in the moderator queue (which I religiously ignore due to volume and it being beyond 99% spam) anyway. This was not the first time this happened so I've disabled the tests. And as a general rant, if you want to post some new topic to the list then don't reply to another posting. That totally screws mail threading so start a new mail thread please. Ralf From ralf@linux-mips.org Wed Mar 17 14:52:31 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 17 Mar 2010 14:52:36 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:34974 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492194Ab0CQNwb (ORCPT ); Wed, 17 Mar 2010 14:52:31 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o2HDqP30004844; Wed, 17 Mar 2010 14:52:26 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o2HDqNEb004841; Wed, 17 Mar 2010 14:52:23 +0100 Date: Wed, 17 Mar 2010 14:52:23 +0100 From: Ralf Baechle To: Wu Zhangjin Cc: linux-mips@linux-mips.org, Shinya Kuribayashi , Zhang Le , Wu Zhangjin Subject: Re: [PATCH v3 2/3] Loongson-2F: Enable fixups of binutils 2.20.1 Message-ID: <20100317135223.GA4554@linux-mips.org> References: MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26252 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Sat, Mar 13, 2010 at 12:34:16PM +0800, Wu Zhangjin wrote: > diff --git a/arch/mips/Makefile b/arch/mips/Makefile > index 2f2eac2..5ae342e 100644 > --- a/arch/mips/Makefile > +++ b/arch/mips/Makefile > @@ -135,7 +135,9 @@ cflags-$(CONFIG_CPU_LOONGSON2) += -Wa,--trap > cflags-$(CONFIG_CPU_LOONGSON2E) += \ > $(call cc-option,-march=loongson2e,-march=r4600) > cflags-$(CONFIG_CPU_LOONGSON2F) += \ > - $(call cc-option,-march=loongson2f,-march=r4600) > + $(call cc-option,-march=loongson2f,-march=r4600) \ > + $(call as-option,-Wa$(comma)-mfix-loongson2f-nop,) \ > + $(call as-option,-Wa$(comma)-mfix-loongson2f-jump,) Shouldn't these options be used unconditionally? It seems a kernel build should rather fail than a possibly unreliable kernel be built - possibly even without the user noticing the problem. Ralf From ralf@linux-mips.org Wed Mar 17 16:02:11 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 17 Mar 2010 16:02:14 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:34724 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492386Ab0CQPCL (ORCPT ); Wed, 17 Mar 2010 16:02:11 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o2HF28H5006552; Wed, 17 Mar 2010 16:02:09 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o2HF28ef006550; Wed, 17 Mar 2010 16:02:08 +0100 Date: Wed, 17 Mar 2010 16:02:07 +0100 From: Ralf Baechle To: Wu Zhangjin Cc: linux-mips@linux-mips.org, Shinya Kuribayashi Subject: Re: [PATCH v3 3/3] Loongson-2F: Fixup of problems introduced by -mfix-loongson2f-jump of binutils 2.20.1 Message-ID: <20100317150207.GB4554@linux-mips.org> References: <169f2daa3d623fe56c5b0be30feeda10bc79a478.1268453720.git.wuzhangjin@gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <169f2daa3d623fe56c5b0be30feeda10bc79a478.1268453720.git.wuzhangjin@gmail.com> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26253 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Sat, Mar 13, 2010 at 12:34:17PM +0800, Wu Zhangjin wrote: > - /* reboot via jumping to boot base address */ > + /* reboot via jumping to boot base address > + * > + * ".set noat" and ".set at" are used to ensure the address not break > + * by the -mfix-loongson2f-jump option provided by binutils 2.20.1 (or > + * higher version) which try to change the jumping address to "addr & > + * 0xcfffffff" via the at($1) register, this is totally wrong for > + * 0xbfc00000 (LOONGSON_BOOT_BASE). > + */ > + __asm__ __volatile__(".set noat\n"); > ((void (*)(void))ioremap_nocache(LOONGSON_BOOT_BASE, 4)) (); > + __asm__ __volatile__(".set at\n"); Ouch. This is fragile and totally puts the kernels fate at the mercy of gcc and the ioremap_nocache() implementation. GCC might emit a .set noat at any time. Something like void loongson_restart(char *command) { void (*func)(void); /* do preparation for reboot */ mach_prepare_reboot(); /* reboot via jumping to boot base address */ func = (void *) ioremap_nocache(LOONGSON_BOOT_BASE, 4); __asm__ __volatile__( " .set noat \n" " jr %[func] \n" " .set at \n" : /* No outputs */ : [func] "r" (func)); } should be safe against -mfix-loongson2f-jump I think. The workaround in http://sourceware.org/ml/binutils/2009-11/msg00387.html will also cause problems calling functions with bits 28..29 set, that is in the ranges 0x81000000..0xbffffff and 0xd0000000..0xffffffff. The first range is not much of a problem as only the kernel proper resides there and the kernel load address is manually selected in the Makefile. The 2nd range might be used for under certain circumstances. Ralf From ralf@linux-mips.org Wed Mar 17 16:11:26 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 17 Mar 2010 16:11:42 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:43811 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492386Ab0CQPLZ (ORCPT ); Wed, 17 Mar 2010 16:11:25 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o2HFBOYY006921; Wed, 17 Mar 2010 16:11:24 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o2HFBO8q006919; Wed, 17 Mar 2010 16:11:24 +0100 Date: Wed, 17 Mar 2010 16:11:24 +0100 From: Ralf Baechle To: Wu Zhangjin Cc: linux-mips@linux-mips.org, Shinya Kuribayashi Subject: Re: [PATCH v3 1/3] Loongson-2F: Flush the branch target history such as BTB and RAS Message-ID: <20100317151124.GC4554@linux-mips.org> References: <05e2ba2596f23fa4dda64d63ce2480504b1be4ac.1268453720.git.wuzhangjin@gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: <05e2ba2596f23fa4dda64d63ce2480504b1be4ac.1268453720.git.wuzhangjin@gmail.com> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26254 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Sat, Mar 13, 2010 at 12:34:15PM +0800, Wu Zhangjin wrote: > From: Wu Zhangjin > > As the Chapter 15: "Errata: Issue of Out-of-order in loongson"[1] shows, to > workaround the Issue of Loongson-2F，We need to do: > > "When switching from user mode to kernel mode, you should flush the > branch target history such as BTB and RAS." > > This patch did clear BTB(branch target buffer), forbid RAS(return > address stack) via Loongson-2F's 64bit diagnostic register. > > [1] Chinese Version: http://www.loongson.cn/uploadfile/file/200808211 > [2] English Version of Chapter 15: > http://groups.google.com.hk/group/loongson-dev/msg/e0d2e220958f10a6?dmode=source Thanks, applied. Ralf From ralf@linux-mips.org Wed Mar 17 16:39:24 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 17 Mar 2010 16:39:28 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:58802 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492467Ab0CQPjY (ORCPT ); Wed, 17 Mar 2010 16:39:24 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o2HFdMSJ007814; Wed, 17 Mar 2010 16:39:22 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o2HFdJK4007807; Wed, 17 Mar 2010 16:39:20 +0100 Date: Wed, 17 Mar 2010 16:39:18 +0100 From: Ralf Baechle To: Andrea Gelmini Cc: linux-kernel@vger.kernel.org, Paul Mundt , linux-mips@linux-mips.org, linux-sh@vger.kernel.org Subject: Re: [PATCH 42/66] arch/mips/lib/libgcc.h: Checkpatch cleanup Message-ID: <20100317153918.GD4554@linux-mips.org> References: <1267289508-31031-1-git-send-email-andrea.gelmini@gelma.net> <1267289508-31031-43-git-send-email-andrea.gelmini@gelma.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1267289508-31031-43-git-send-email-andrea.gelmini@gelma.net> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26255 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Sat, Feb 27, 2010 at 05:51:23PM +0100, Andrea Gelmini wrote: > From: Andrea Gelmini > Date: Sat, 27 Feb 2010 17:51:23 +0100 > To: linux-kernel@vger.kernel.org > Cc: Andrea Gelmini , > Ralf Baechle , > Paul Mundt , linux-mips@linux-mips.org, > linux-sh@vger.kernel.org > Subject: [PATCH 42/66] arch/mips/lib/libgcc.h: Checkpatch cleanup > > arch/mips/lib/libgcc.h:21: ERROR: open brace '{' following union go on the same line > > Signed-off-by: Andrea Gelmini > --- > arch/mips/lib/libgcc.h | 3 +-- > arch/sh/lib/libgcc.h | 3 +-- > 2 files changed, 2 insertions(+), 4 deletions(-) I've applied the MIPS portion of the patch, thanks. Ralf From ralf@linux-mips.org Wed Mar 17 16:51:46 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 17 Mar 2010 16:51:50 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:48630 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492478Ab0CQPvq (ORCPT ); Wed, 17 Mar 2010 16:51:46 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o2HFpa1d008172; Wed, 17 Mar 2010 16:51:37 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o2HFpVXZ008159; Wed, 17 Mar 2010 16:51:31 +0100 Date: Wed, 17 Mar 2010 16:51:28 +0100 From: Ralf Baechle To: Andrea Gelmini Cc: Paul Mundt , linux-kernel@vger.kernel.org, linux-mips@linux-mips.org, linux-sh@vger.kernel.org Subject: Re: [PATCH 42/66] arch/mips/lib/libgcc.h: Checkpatch cleanup Message-ID: <20100317155126.GE4554@linux-mips.org> References: <1267289508-31031-1-git-send-email-andrea.gelmini@gelma.net> <1267289508-31031-43-git-send-email-andrea.gelmini@gelma.net> <20100301023316.GB23086@linux-sh.org> <9cdbb57f1003020713x4e897af9ka87e348bf782f380@mail.gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <9cdbb57f1003020713x4e897af9ka87e348bf782f380@mail.gmail.com> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26256 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Tue, Mar 02, 2010 at 04:13:11PM +0100, Andrea Gelmini wrote: > 2010/3/1 Paul Mundt : > Hi Paul, > thanks for your reply. > > > I'll apply the sh part by itself, and I suppose Ralf will do the same for > > MIPS. I'm a bit confused as to why these were lumped together when 65 out > > of 66 oneliner patches made no use of ad-hoc grouping. > > After the experience with GregKH, I was looking for others maintainer > interested in code cleanup. So I sent some patches over random files. > If you care, I can checkpatch.pl all mips subtree and send you patches (if any). Feel free to - but remember that the coding style document is a guideline not something to be followed unconditionally. Ralf From ralf@linux-mips.org Wed Mar 17 17:17:48 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 17 Mar 2010 17:17:51 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:59471 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492660Ab0CQQRs (ORCPT ); Wed, 17 Mar 2010 17:17:48 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o2HGHkGk009086; Wed, 17 Mar 2010 17:17:47 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o2HGHjWM009084; Wed, 17 Mar 2010 17:17:45 +0100 Date: Wed, 17 Mar 2010 17:17:44 +0100 From: Ralf Baechle To: Florian Fainelli Cc: linux-mips@linux-mips.org Subject: Re: [PATCH 4/4] bcm63xx: fix BCM6338 and BCM6345 gpio count Message-ID: <20100317161744.GF4554@linux-mips.org> References: <201003012336.32430.florian@openwrt.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <201003012336.32430.florian@openwrt.org> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26257 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Mon, Mar 01, 2010 at 11:36:32PM +0100, Florian Fainelli wrote: > The number of GPIOs on BCM6338 is 8, while BCM6345 has only 16 GPIOs > available. Thanks, applied. Ralf From jiang.adam@gmail.com Wed Mar 17 17:58:00 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 17 Mar 2010 17:58:05 +0100 (CET) Received: from mail-qy0-f180.google.com ([209.85.221.180]:50585 "EHLO mail-qy0-f180.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492617Ab0CQQ6A convert rfc822-to-8bit (ORCPT ); Wed, 17 Mar 2010 17:58:00 +0100 Received: by qyk10 with SMTP id 10so659129qyk.6 for ; Wed, 17 Mar 2010 09:57:53 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:in-reply-to:references :date:message-id:subject:from:to:cc:content-type :content-transfer-encoding; bh=a0bZUCZT3rUQgCv51z10F3TusFFIhTMqdmb2NJNU5JM=; b=VVOedspMuAcUQJlbfeYdOc8oMB1eqJgJMt9gUxsLKBmC2PT8YlPRJXYUIHoUfQml6V kKrlxPk6DPplo2Q7zaBXt4vOo7JssbT5Tl8TTbnTpE/CqkFyDjIZHY8+Kq0Otf2YzZHW F+uHeIvRPd/bl8lcAqBWQcxc4psSZVizF+1io= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type:content-transfer-encoding; b=QQeKbn1JuZcrIY63OKSuHnsfnT6BqoTPe9IEp0AO/pHT4cZ+egsGdJeMwoVQxP2d6Y lUarduEatpzLBBr0EBmH5EbIMw8OwRZGA1LfnvCUBWwNV0z3ULSR9KUt1BE8cyBqTpjV nHyRyv0GzptvR2tX+3IEznP5CQBAG7ZpNLoqQ= MIME-Version: 1.0 Received: by 10.224.87.66 with SMTP id v2mr351822qal.343.1268845071409; Wed, 17 Mar 2010 09:57:51 -0700 (PDT) In-Reply-To: <6C370B347C3FE8438C9692873287D2E1109DDF01AD@SJEXCHCCR01.corp.ad.broadcom.com> References: <4BEA3FF3CAA35E408EA55C7BE2E61D0546AC862322@xmail3.se.axis.com> <6C370B347C3FE8438C9692873287D2E1109DDF0057@SJEXCHCCR01.corp.ad.broadcom.com> <6C370B347C3FE8438C9692873287D2E1109DDF0073@SJEXCHCCR01.corp.ad.broadcom.com> <1864acd31003170006q602602a9j5675964b886b0a55@mail.gmail.com> <6C370B347C3FE8438C9692873287D2E1109DDF015B@SJEXCHCCR01.corp.ad.broadcom.com> <1864acd31003170429l7b0972fcn46a6404674ad62d9@mail.gmail.com> <6C370B347C3FE8438C9692873287D2E1109DDF01AD@SJEXCHCCR01.corp.ad.broadcom.com> Date: Thu, 18 Mar 2010 01:57:51 +0900 Message-ID: <1864acd31003170957hc3abf4dx1fbcb243049a51ca@mail.gmail.com> Subject: Re: HIGHMEM not working. From: Adam Jiang To: "linux-mips@linux-mips.org" Cc: Ramgopal Kota Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8BIT Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26258 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: jiang.adam@gmail.com Precedence: bulk X-list: linux-mips Hi, Ramgopal 2010/3/17 Ramgopal Kota : > The first 128MB of 0x8000_0000 is mapped to 0x0000_0000 address, hence I added them as 2 nodes with 0 & 0x8800_0000 .. > max_zone_pfns[ZONE_NORMAL] should be assigned with maximum physical address of the memory i.e. address >> page_size ... > As the physical address is 0x07FF_FFFF .. The ZONE_NORMAL PFN should be 0x07FFF .. > As ZONE_HIGH max address is 0x8FFF_FFFF .. The ZONE_HIGH PFN should be 0x8FFFF .. Why do you think this is wrong ? I get confusing with your memory mapping... What do you mean ZONE_HIGH max address is 0x8fff ffff? Could you please clarify your memory layout within physical address instead of mixed mapped virtual address? And, what's the virtual addresses you planed to map them? > I did not get how I can handle with out ZONE_HIGH ?? Can you explain me in more detail please .. You are using a MIPS64 core, aren't you? If Linux kernel could only access the virtual memory address lower than PAGE_OFFSET, you don't need to define a ZONE_HIGH because you have only 128M + 128M memory < (PAGE_OFFSET) on your target. You don't need map those memory because you can access all your memory directly, I think. If you have a 32-bit PCI controller on you board, you may need bounce buffers because you CPU just has 64-bit address. I missed the most important information you have provided. You can find out some sample code on ARM at the following URL: http://lxr.linux.no/linux+v2.6.33/arch/arm/common/dmabounce.c#L5 there're common routines you can try to port it to MIPS if you need them. Regards, /Adam > > Thanks & Regards, > Ramgopal Kota > > -----Original Message----- > From: Adam Jiang [mailto:jiang.adam@gmail.com] > Sent: Wednesday, March 17, 2010 5:00 PM > To: linux-mips@linux-mips.org > Cc: Ramgopal Kota > Subject: Re: HIGHMEM not working. > > 2010/3/17 Ramgopal Kota >> >> Adam, >> >> Thanks for the response ... >> >> We enabled CONFIG_SPARSEMEM in place of CONFIG_DISCONTIGMEM .. It is specified that SPARSEMEM will eventually replaces DISCONTIGMEM >> (http://marc.info/?l=lhms-devel&m=111451976115580&w=2) > > Yes. It seems a better way to do this. > >> We have configured zones as below in paging_init function .. >> max_zone_pfns[ZONE_DMA] = MAX_DMA_PFN; >> max_zone_pfns[ZONE_NORMAL] = max_low_pfn;   <<  0x20000 >> max_zone_pfns[ZONE_HIGHMEM] = highend_pfn;  <<  0x8FFFF >> We added 2 memory regions as below .. >> add_memory_region(0x00000000, 128MB, BOOT_MEM_RAM); >> add_memory_region(0x88000000  ,128MB, BOOT_MEM_RAM); >> Is this what you are refering as 2 memory nodes ?? >> > > OK. Physical memory is managed in nodes and zones. Because you have > mentioned that there would be 2 ranges of memory on your architecture > like > > Node 0   [0x0000 0000, 0x0800 0000] > Node 1   [0x8000 0000, 0xcfff ffff] > > I guess you can define these two nodes by add_memory_region function > but I am not sure... However, the start_addr of the second node would > be 0x8000 0000 rather than 0x8800 0000. On the other hand, I think the > following code is not correct > >> max_zone_pfns[ZONE_DMA] = MAX_DMA_PFN; >> max_zone_pfns[ZONE_NORMAL] = max_low_pfn;   <<  0x20000 >> max_zone_pfns[ZONE_HIGHMEM] = highend_pfn;  <<  0x8FFFF > > if max_low_pfn is 0x20000, then your ZONE_NORMAL will end up at 0x2000 > 0000(PFN x PAGE_SIZE); but according to the information about physical > memory address you provided. 0x0800 0000 ~ 0x0fff ffff is used as PCI > device addresses. Em~, the highend_pfn is not correct either. Please > double check these address. > > Bounce buffers are required for devices that cannnot access the full > range of memory available to the CPU, such as old ISA cards with only > 24bit address. Is this the situation you have to face? If it is needed > you may look into function create_bounce_buffer() in the other drivers > for details, however, I still think you would not need it. By the way, > high memory just mapped to low memory by kmap() function. You don't > need to define a ZONE_HIGH in your configuration. Because 0x8000_0000 > ~ 0x8fff ffff is kseg0 on MIPS. You can handle them as ZONE_NORMAL in > much easier way. > > Best regards, > /Adam > >> >> What is the significance of MAX_PHYSMEM_BITS  & SECTION_SIZE_BITS .. Sorry for troubling with novoice questions .. >> >> Thanks & Regards, >> Ramgopal Kota >> ________________________________ >> From: Adam Jiang [mailto:jiang.adam@gmail.com] >> Sent: Wednesday, March 17, 2010 12:36 PM >> To: Ramgopal Kota >> Subject: Re: HIGHMEM not working. >> >> 2010/3/17 Ramgopal Kota >>> >>>  Hi, >>> >>> I need help in running CPU with HIGH_MEM. >>> Below is the processor memory map .. It is base on MIPS64K Core. >>> >>> 0x0000_0000  -- 0x07FF_FFFF   -> RAM (Region 1) >>> 0x0800_0000  -- 0x0FFF_FFFF   -> PCI BUS 1 >>> 0x2000_0000  -- 0x2FFF_FFFF   -> Flash Region >>> 0x4000_0000  -- 0x47FF_FFFF   -> PCI BUS 2 >>> 0x8000_0000  -- 0xCFFF_FFFF   -> RAM (Region 2) the first 128MB is remapped into 0x8000_0000 .. >>> >>> The system has 256MB of PHYSICAL RAM .. >>> >>> I configured TLB's and added memory regions. The system boots up correctly .. >>> If I show cat /proc/meminfo it shows 256MB , it could allocate memory ( I verified with a simple program allocating the memory and writing some data) >>> >>> There is a PCI MAC device on PCI BUS 2. The CPU is not able to perform DMA to/from device. >>> The MAC driver is allocating DMA buffers using the following code .. >>> { >>>    phy = virt_to_bus(high_memory);  << Resulting in 0xa000_0000; >>>    virt = ioremap(pbase); >>> } >>> The Linux configuration has no ZONE_DMA configuration .. >>> >>> >>> Few questions .. >>>  A) HIGH_MEMORY START Address is configured to 0x2000_0000 , is it correct value. >> >> I don't think 0x2000_0000 is a correct value for HIGH_MEM in general meaning. The value depends on which architecture you're working on. And its start address could be initialed in >> LNX/arch/(arch)/kernel/setup.c >> LNX/arch/(arch)/mm/init.c >> in your case, I thing you can use all your memory as ZONE_NORMAL or ZONE_DMA but not ZONE_HIGH. >> >>> >>>  B) Are there any data which can be dumped to know more information ? >> >> Yes, of course. You could try to add some printk in the above two files in bootmem_init() function or function free_area_init_core() to see what's the memory layout exactly. >> >>> >>>  C) Do I need to enable MEMORY Discontinuous Config also ?  ( As I did not see a reason to enable this as memory in Region 2 is contiguous) >> >> Since you have two separated range fsyc memory on you board, I think you do really need CONFIG_DISCONTIGMEM_*; You may also need to define two memory nodes in bootmem. >> >>> >>>  D) I read in some articles on internet that if HIGH_MEM is used,PCI bounce buffers needs to be used ? Can some body point me how to set this up ? >> >> Usually, DMA data transferring could perform directly on fysc memory address. You may not need bounce buffer or scatter/gather operations if you can config all your memory to ZONE_NORMAL. ZONE_DMA is not necessary because this area is only required by kmalloc call with GFP_DMA flags. >> Best regards, >> /Adam >> >>> >>> Thanks & Regards, >>> Ramgopal Kota >>> >>> >> >> -- >> Adam Jiang >> ----------------------------------- >> e-mail:jiang.adam@gmail.com >> http://www.adamjiang.com >> ----------------------------------- > > > > -- > Adam Jiang > ----------------------------------- > e-mail:jiang.adam@gmail.com > http://www.adamjiang.com > ----------------------------------- > > > -- Adam Jiang ----------------------------------- e-mail:jiang.adam@gmail.com http://www.adamjiang.com ----------------------------------- From ralf@linux-mips.org Wed Mar 17 21:33:50 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Wed, 17 Mar 2010 21:33:54 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:59923 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492617Ab0CQUdu (ORCPT ); Wed, 17 Mar 2010 21:33:50 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o2HKXmDV015454; Wed, 17 Mar 2010 21:33:49 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o2HKXlMr015437; Wed, 17 Mar 2010 21:33:47 +0100 Date: Wed, 17 Mar 2010 21:33:46 +0100 From: Ralf Baechle To: Florian Fainelli Cc: linux-mips@linux-mips.org Subject: Re: [PATCH 1/4] bcm63xx: fix tabs vs spaces in board_bcm963xx.c Message-ID: <20100317203346.GG4554@linux-mips.org> References: <201003012336.15400.florian@openwrt.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <201003012336.15400.florian@openwrt.org> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26259 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Mon, Mar 01, 2010 at 11:36:15PM +0100, Florian Fainelli wrote: Patch rejects. From ralf@linux-mips.org Thu Mar 18 16:01:35 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Thu, 18 Mar 2010 16:01:38 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:45540 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492015Ab0CRPBf (ORCPT ); Thu, 18 Mar 2010 16:01:35 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o2IF1XAY010543; Thu, 18 Mar 2010 16:01:34 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o2IF1Xv5010542; Thu, 18 Mar 2010 16:01:33 +0100 Date: Thu, 18 Mar 2010 16:01:32 +0100 From: Ralf Baechle To: Florian Fainelli Cc: linux-mips@linux-mips.org Subject: Re: [PATCH 2/4] bcm63xx: add rta1025w_16 board Message-ID: <20100318150132.GH4554@linux-mips.org> References: <201003012336.22578.florian@openwrt.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <201003012336.22578.florian@openwrt.org> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26260 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Mon, Mar 01, 2010 at 11:36:22PM +0100, Florian Fainelli wrote: Applied, thanks. Ralf From ralf@linux-mips.org Thu Mar 18 16:02:10 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Thu, 18 Mar 2010 16:02:17 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:45547 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492015Ab0CRPB6 (ORCPT ); Thu, 18 Mar 2010 16:01:58 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o2IF1wnK010571; Thu, 18 Mar 2010 16:01:58 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o2IF1vJ2010570; Thu, 18 Mar 2010 16:01:57 +0100 Date: Thu, 18 Mar 2010 16:01:57 +0100 From: Ralf Baechle To: Florian Fainelli Cc: linux-mips@linux-mips.org Subject: Re: [PATCH 3/4] bcm63xx: add DWVS0 board Message-ID: <20100318150157.GI4554@linux-mips.org> References: <201003012336.28071.florian@openwrt.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <201003012336.28071.florian@openwrt.org> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26261 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips Applied, thanks. Ralf From ralf@linux-mips.org Thu Mar 18 18:00:38 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Thu, 18 Mar 2010 18:00:42 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:57017 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492193Ab0CRRAi (ORCPT ); Thu, 18 Mar 2010 18:00:38 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o2IH0YKx013522; Thu, 18 Mar 2010 18:00:34 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o2IH0VcS013520; Thu, 18 Mar 2010 18:00:31 +0100 Date: Thu, 18 Mar 2010 18:00:30 +0100 From: Ralf Baechle To: Jean Delvare Cc: Yang Shi , ddaney@caviumnetworks.com, ben-linux@fluff.org, linux-mips@linux-mips.org, linux-i2c@vger.kernel.org Subject: Re: [PATCH V2] MIPS: Octeon: Register EEPROM device on the I2C bus Message-ID: <20100318170030.GJ4554@linux-mips.org> References: <1268026190-18300-1-git-send-email-yang.shi@windriver.com> <20100316180946.GC20160@linux-mips.org> <20100316200647.3803edf1@hyperion.delvare> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20100316200647.3803edf1@hyperion.delvare> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26262 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Tue, Mar 16, 2010 at 08:06:47PM +0100, Jean Delvare wrote: > The Linux kernel doesn't care, but user-space may. As a matter of fact, > there is a script out there (decode-dimms, in the i2c-tools package) > decoding the SPD data and presenting it to the user. Some people want > to know the details about their memory modules. > > > I also wonder how this will work for configurations with multiple memory > > modules thus multiple SPD EEPROMS. > > The kernel code should instantiate one spd device per memory module > (assuming they are all reachable.) Obviously this can't be done in a > static way. SPD is virtually omnipresent these days so I wonder if maybe there already is some probing functionality already available? Ralf From khali@linux-fr.org Thu Mar 18 18:48:51 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Thu, 18 Mar 2010 18:48:54 +0100 (CET) Received: from poutre.nerim.net ([62.4.16.124]:55922 "EHLO poutre.nerim.net" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492182Ab0CRRsv (ORCPT ); Thu, 18 Mar 2010 18:48:51 +0100 Received: from localhost (localhost [127.0.0.1]) by poutre.nerim.net (Postfix) with ESMTP id 7E67339DE93; Thu, 18 Mar 2010 18:48:47 +0100 (CET) X-Virus-Scanned: amavisd-new at nerim.net Received: from poutre.nerim.net ([127.0.0.1]) by localhost (poutre.nerim.net [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 1UPixwFqrkFu; Thu, 18 Mar 2010 18:48:46 +0100 (CET) Received: from hyperion.delvare (jdelvare.pck.nerim.net [62.212.121.182]) by poutre.nerim.net (Postfix) with ESMTP id 572A939DE7C; Thu, 18 Mar 2010 18:48:46 +0100 (CET) Date: Thu, 18 Mar 2010 18:48:48 +0100 From: Jean Delvare To: Ralf Baechle Cc: Yang Shi , ddaney@caviumnetworks.com, ben-linux@fluff.org, linux-mips@linux-mips.org, linux-i2c@vger.kernel.org Subject: Re: [PATCH V2] MIPS: Octeon: Register EEPROM device on the I2C bus Message-ID: <20100318184848.4ec62327@hyperion.delvare> In-Reply-To: <20100318170030.GJ4554@linux-mips.org> References: <1268026190-18300-1-git-send-email-yang.shi@windriver.com> <20100316180946.GC20160@linux-mips.org> <20100316200647.3803edf1@hyperion.delvare> <20100318170030.GJ4554@linux-mips.org> X-Mailer: Claws Mail 3.5.0 (GTK+ 2.14.4; i586-suse-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26263 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: khali@linux-fr.org Precedence: bulk X-list: linux-mips Hi Ralf, On Thu, 18 Mar 2010 18:00:30 +0100, Ralf Baechle wrote: > On Tue, Mar 16, 2010 at 08:06:47PM +0100, Jean Delvare wrote: > > > The Linux kernel doesn't care, but user-space may. As a matter of fact, > > there is a script out there (decode-dimms, in the i2c-tools package) > > decoding the SPD data and presenting it to the user. Some people want > > to know the details about their memory modules. > > > > > I also wonder how this will work for configurations with multiple memory > > > modules thus multiple SPD EEPROMS. > > > > The kernel code should instantiate one spd device per memory module > > (assuming they are all reachable.) Obviously this can't be done in a > > static way. > > SPD is virtually omnipresent these days so I wonder if maybe there already > is some probing functionality already available? Not that I aware of, except for the fact that the "eeprom" driver will probe all i2c adapters with class bit I2C_CLASS_SPD set for chips at addresses 0x50-0x57, and it will instantiate a device for each of them. This will instantiate devices for non-SPD EEPROM, and even non-EEPROMs living at these addresses, so this can't really be considered a generic solution. A generic solution would let the platform (not the eeprom driver) probe for devices at selected I2C addresses (the ones where SPD EEPROMs can live of the specific platform) and instantiate spd devices (possibly after integrity checking). This would avoid false positives. -- Jean Delvare From p2@psychaos.be Thu Mar 18 19:17:35 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Thu, 18 Mar 2010 19:17:38 +0100 (CET) Received: from apfelkorn.psychaos.be ([195.144.77.38]:57108 "EHLO apfelkorn.psychaos.be" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491065Ab0CRSRf (ORCPT ); Thu, 18 Mar 2010 19:17:35 +0100 Received: from p2 by apfelkorn.psychaos.be with local (Exim 4.69) (envelope-from ) id 1NsKHm-0005xm-6R; Thu, 18 Mar 2010 20:17:34 +0200 Date: Thu, 18 Mar 2010 20:17:34 +0200 From: Peter 'p2' De Schrijver To: Jan Rovins Cc: linux-mips@linux-mips.org Subject: Re: linux 2.6.33 on movidis x16 Message-ID: <20100318181734.GG2437@apfelkorn> References: <20100305141113.GD2437@apfelkorn> <4B9EB45D.8050106@adax.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <4B9EB45D.8050106@adax.com> X-Unexpected-Header: The spanish inquisition ! X-mate: Mate, mann gewohnt sich an alles X-Paddo: Munch, Munch User-Agent: Mutt/1.5.18 (2008-05-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26264 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: p2@debian.org Precedence: bulk X-list: linux-mips On 2010-03-15 18:27:41 (-0400), Jan Rovins wrote: > Peter 'p2' De Schrijver wrote: >> Hi, >> >> We are trying to get linux 2.6.33 to work on the movidis x16 board. Main thing >> missing is the addresses of the PHYs. Does anyone know those ? >> Unfortunately I don't have physical access to the board so I can't just look >> for the PHY components being used. >> >> Thanks, >> >> Peter. >> >> > Do you mean the Movidis x19 ? Maybe. AFAIK ours is called x16, and uses a cavium octeon CN3860 chip. > We have 2 of these in our lab. They are running the old 2.6.21.7 from > the CnUsers 1.8.1 tool chain. > They are currently being used by other developers for some application > testing, but I can grab the serial console boot-up messages and send > them to you, if the PHY info is in there. That would be helpful. Otherwise you could also try running ethtool on all the interfaces. That should give you the PHY address as well. Thanks ! Peter. From David.Daney@caviumnetworks.com Thu Mar 18 19:26:42 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Thu, 18 Mar 2010 19:26:45 +0100 (CET) Received: from mail3.caviumnetworks.com ([12.108.191.235]:9101 "EHLO mail3.caviumnetworks.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492193Ab0CRS0m (ORCPT ); Thu, 18 Mar 2010 19:26:42 +0100 Received: from caexch01.caveonetworks.com (Not Verified[192.168.16.9]) by mail3.caviumnetworks.com with MailMarshal (v6,7,2,8378) id ; Thu, 18 Mar 2010 11:26:48 -0700 Received: from caexch01.caveonetworks.com ([192.168.16.9]) by caexch01.caveonetworks.com with Microsoft SMTPSVC(6.0.3790.3959); Thu, 18 Mar 2010 11:26:17 -0700 Received: from dd1.caveonetworks.com ([12.108.191.236]) by caexch01.caveonetworks.com over TLS secured channel with Microsoft SMTPSVC(6.0.3790.3959); Thu, 18 Mar 2010 11:26:17 -0700 Message-ID: <4BA27048.2010707@caviumnetworks.com> Date: Thu, 18 Mar 2010 11:26:16 -0700 From: David Daney User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.8) Gecko/20100301 Fedora/3.0.3-1.fc12 Thunderbird/3.0.3 MIME-Version: 1.0 To: Peter 'p2' De Schrijver CC: Jan Rovins , linux-mips@linux-mips.org Subject: Re: linux 2.6.33 on movidis x16 References: <20100305141113.GD2437@apfelkorn> <4B9EB45D.8050106@adax.com> <20100318181734.GG2437@apfelkorn> In-Reply-To: <20100318181734.GG2437@apfelkorn> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-OriginalArrivalTime: 18 Mar 2010 18:26:17.0338 (UTC) FILETIME=[7FCCFDA0:01CAC6C8] Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26265 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ddaney@caviumnetworks.com Precedence: bulk X-list: linux-mips On 03/18/2010 11:17 AM, Peter 'p2' De Schrijver wrote: > On 2010-03-15 18:27:41 (-0400), Jan Rovins wrote: >> Peter 'p2' De Schrijver wrote: >>> Hi, >>> >>> We are trying to get linux 2.6.33 to work on the movidis x16 board. Main thing >>> missing is the addresses of the PHYs. Does anyone know those ? >>> Unfortunately I don't have physical access to the board so I can't just look >>> for the PHY components being used. >>> >>> Thanks, >>> >>> Peter. >>> >>> >> Do you mean the Movidis x19 ? > > Maybe. AFAIK ours is called x16, and uses a cavium octeon CN3860 chip. > >> We have 2 of these in our lab. They are running the old 2.6.21.7 from >> the CnUsers 1.8.1 tool chain. >> They are currently being used by other developers for some application >> testing, but I can grab the serial console boot-up messages and send >> them to you, if the PHY info is in there. > > That would be helpful. Otherwise you could also try running ethtool on all the > interfaces. That should give you the PHY address as well. > I just found one of these. Current theory is that the PHY addresses are 0-7 I don't think the PHY numbers are printed anywhere. David Daney From janr@adax.com Thu Mar 18 19:59:23 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Thu, 18 Mar 2010 19:59:27 +0100 (CET) Received: from mail1.adax.com ([208.201.231.104]:22267 "EHLO mail1.adax.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492435Ab0CRS7X (ORCPT ); Thu, 18 Mar 2010 19:59:23 +0100 Received: from static-151-204-189-187.pskn.east.verizon.net (static-151-204-189-187.pskn.east.verizon.net [151.204.189.187]) by mail1.adax.com (Postfix) with ESMTP id 93CE5120AB9; Thu, 18 Mar 2010 11:59:14 -0700 (PDT) Received: from localhost (localhost [127.0.0.1]) by static-151-204-189-187.pskn.east.verizon.net (Postfix) with ESMTP id 9E825400589; Thu, 18 Mar 2010 14:59:13 -0400 (EDT) X-Virus-Scanned: amavisd-new at pskn.east.verizon.net Received: from static-151-204-189-187.pskn.east.verizon.net ([127.0.0.1]) by localhost (static-151-204-189-187.pskn.east.verizon.net [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id SNDwZvZ66lil; Thu, 18 Mar 2010 14:58:52 -0400 (EDT) Received: from [192.168.1.76] (jr001327.mtl-nj.adax [192.168.1.76]) by static-151-204-189-187.pskn.east.verizon.net (Postfix) with ESMTP id D95F1400588; Thu, 18 Mar 2010 14:58:51 -0400 (EDT) Message-ID: <4BA277FB.6080808@adax.com> Date: Thu, 18 Mar 2010 14:59:07 -0400 From: Jan Rovins User-Agent: Thunderbird 2.0.0.24 (Windows/20100228) MIME-Version: 1.0 To: David Daney CC: Peter 'p2' De Schrijver , linux-mips@linux-mips.org Subject: Re: linux 2.6.33 on movidis x16 References: <20100305141113.GD2437@apfelkorn> <4B9EB45D.8050106@adax.com> <20100318181734.GG2437@apfelkorn> <4BA27048.2010707@caviumnetworks.com> In-Reply-To: <4BA27048.2010707@caviumnetworks.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26266 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: janr@adax.com Precedence: bulk X-list: linux-mips David Daney wrote: > On 03/18/2010 11:17 AM, Peter 'p2' De Schrijver wrote: >> On 2010-03-15 18:27:41 (-0400), Jan Rovins wrote: >>> Peter 'p2' De Schrijver wrote: >>>> Hi, >>>> >>>> We are trying to get linux 2.6.33 to work on the movidis x16 board. >>>> Main thing >>>> missing is the addresses of the PHYs. Does anyone know those ? >>>> Unfortunately I don't have physical access to the board so I can't >>>> just look >>>> for the PHY components being used. >>>> >>>> Thanks, >>>> >>>> Peter. >>>> >>>> >>> Do you mean the Movidis x19 ? >> >> Maybe. AFAIK ours is called x16, and uses a cavium octeon CN3860 chip. >> >>> We have 2 of these in our lab. They are running the old 2.6.21.7 from >>> the CnUsers 1.8.1 tool chain. >>> They are currently being used by other developers for some application >>> testing, but I can grab the serial console boot-up messages and send >>> them to you, if the PHY info is in there. >> >> That would be helpful. Otherwise you could also try running ethtool >> on all the >> interfaces. That should give you the PHY address as well. >> > > I just found one of these. Current theory is that the PHY addresses > are 0-7 > > I don't think the PHY numbers are printed anywhere. > > David Daney You are right, it is called an x16. My memory is failing me, I should have looked at the machine first. from dmesg: cavium-ethernet: Cavium Networks Octeon SDK version 1.8.1, build 294 Interface 0 has 4 ports (RGMII) Interface 1 has 4 ports (RGMII) Interface 2 has 4 ports (NPI) Uniform Multi-Platform E-IDE driver Revision: 7.00alpha2 looks like the PHYAD field has what you are loohing for, ethtool output: --------------------------- for i in 0 1 2 3 4 5 6 7; do ethtool eth$i; done Settings for eth0: Supported ports: [ TP MII ] Supported link modes: 10baseT/Half 10baseT/Full 100baseT/Half 100baseT/Full 1000baseT/Half 1000baseT/Full Supports auto-negotiation: Yes Advertised link modes: 10baseT/Half 10baseT/Full 100baseT/Half 100baseT/Full 1000baseT/Half 1000baseT/Full Advertised auto-negotiation: Yes Speed: 10Mb/s Duplex: Half Port: MII PHYAD: 1 Transceiver: internal Auto-negotiation: on Link detected: no Settings for eth1: Supported ports: [ TP MII ] Supported link modes: 10baseT/Half 10baseT/Full 100baseT/Half 100baseT/Full 1000baseT/Half 1000baseT/Full Supports auto-negotiation: Yes Advertised link modes: 10baseT/Half 10baseT/Full 100baseT/Half 100baseT/Full 1000baseT/Half 1000baseT/Full Advertised auto-negotiation: Yes Speed: 10Mb/s Duplex: Half Port: MII PHYAD: 2 Transceiver: internal Auto-negotiation: on Link detected: no Settings for eth2: Supported ports: [ TP MII ] Supported link modes: 10baseT/Half 10baseT/Full 100baseT/Half 100baseT/Full 1000baseT/Half 1000baseT/Full Supports auto-negotiation: Yes Advertised link modes: 10baseT/Half 10baseT/Full 100baseT/Half 100baseT/Full 1000baseT/Half 1000baseT/Full Advertised auto-negotiation: Yes Speed: 10Mb/s Duplex: Half Port: MII PHYAD: 3 Transceiver: internal Auto-negotiation: on Link detected: no Settings for eth3: Supported ports: [ TP MII ] Supported link modes: 10baseT/Half 10baseT/Full 100baseT/Half 100baseT/Full 1000baseT/Half 1000baseT/Full Supports auto-negotiation: Yes Advertised link modes: 10baseT/Half 10baseT/Full 100baseT/Half 100baseT/Full 1000baseT/Half 1000baseT/Full Advertised auto-negotiation: Yes Speed: 10Mb/s Duplex: Half Port: MII PHYAD: 4 Transceiver: internal Auto-negotiation: on Link detected: no Settings for eth4: Supported ports: [ TP MII ] Supported link modes: 10baseT/Half 10baseT/Full 100baseT/Half 100baseT/Full Supports auto-negotiation: Yes Advertised link modes: 10baseT/Half 10baseT/Full 100baseT/Half 100baseT/Full Advertised auto-negotiation: Yes Speed: 100Mb/s Duplex: Full Port: MII PHYAD: 0 Transceiver: internal Auto-negotiation: on Link detected: yes Settings for eth5: Supported ports: [ TP MII ] Supported link modes: 10baseT/Half 10baseT/Full 100baseT/Half 100baseT/Full Supports auto-negotiation: Yes Advertised link modes: 10baseT/Half 10baseT/Full 100baseT/Half 100baseT/Full Advertised auto-negotiation: Yes Speed: 100Mb/s Duplex: Full Port: MII PHYAD: 0 Transceiver: internal Auto-negotiation: on Link detected: yes Settings for eth6: Supported ports: [ TP MII ] Supported link modes: 10baseT/Half 10baseT/Full 100baseT/Half 100baseT/Full Supports auto-negotiation: Yes Advertised link modes: 10baseT/Half 10baseT/Full 100baseT/Half 100baseT/Full Advertised auto-negotiation: Yes Speed: 100Mb/s Duplex: Full Port: MII PHYAD: 0 Transceiver: internal Auto-negotiation: on Link detected: yes Settings for eth7: Supported ports: [ TP MII ] Supported link modes: 10baseT/Half 10baseT/Full 100baseT/Half 100baseT/Full Supports auto-negotiation: Yes Advertised link modes: 10baseT/Half 10baseT/Full 100baseT/Half 100baseT/Full Advertised auto-negotiation: Yes Speed: 100Mb/s Duplex: Full Port: MII PHYAD: 0 Transceiver: internal Auto-negotiation: on Link detected: yes ---------------------------------------------------------- You may want to re-map the interfaces so that they show up in a sane order, there is a cluster of 8 ethernet ports, 4 per row. Right now on our boxes , eth0 shows up on the bottom row, second from the right. Jan From ralf@linux-mips.org Fri Mar 19 13:00:56 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Fri, 19 Mar 2010 13:00:59 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:37030 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492058Ab0CSMAz (ORCPT ); Fri, 19 Mar 2010 13:00:55 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o2JC0idh015774; Fri, 19 Mar 2010 13:00:45 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o2JC0dM6015772; Fri, 19 Mar 2010 13:00:40 +0100 Date: Fri, 19 Mar 2010 13:00:38 +0100 From: Ralf Baechle To: Maxime Bizon Cc: linux-mips@linux-mips.org Subject: Re: [PATCH 3/7] MIPS: bcm63xx: fix double gpio registration. Message-ID: <20100319120037.GK4554@linux-mips.org> References: <1264872898-28149-1-git-send-email-mbizon@freebox.fr> <1264872898-28149-4-git-send-email-mbizon@freebox.fr> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1264872898-28149-4-git-send-email-mbizon@freebox.fr> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26267 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Sat, Jan 30, 2010 at 06:34:54PM +0100, Maxime Bizon wrote: > bcm63xx_gpio_init is already called from prom_init to allow board to > use them early, so we can remove the unneeded arch_initcall. Applied, thanks. Ralf From ralf@linux-mips.org Fri Mar 19 13:19:27 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Fri, 19 Mar 2010 13:19:31 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:57055 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492066Ab0CSMT1 (ORCPT ); Fri, 19 Mar 2010 13:19:27 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o2JCJPbC016396; Fri, 19 Mar 2010 13:19:25 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o2JCJOfR016393; Fri, 19 Mar 2010 13:19:24 +0100 Date: Fri, 19 Mar 2010 13:19:23 +0100 From: Ralf Baechle To: Maxime Bizon Cc: linux-mips@linux-mips.org Subject: Re: [PATCH 4/7] MIPS: bcm63xx: add support for second uart. Message-ID: <20100319121923.GL4554@linux-mips.org> References: <1264872898-28149-1-git-send-email-mbizon@freebox.fr> <1264872898-28149-5-git-send-email-mbizon@freebox.fr> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1264872898-28149-5-git-send-email-mbizon@freebox.fr> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26268 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Sat, Jan 30, 2010 at 06:34:55PM +0100, Maxime Bizon wrote: > The bcm63xx SOC has two uarts, some boards use the second one for > bluetooth. This patch changes platform device registration code to > handle this. Changes to the uart driver are in a separate patch. Applied. Thanks, Ralf From ralf@linux-mips.org Fri Mar 19 13:23:41 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Fri, 19 Mar 2010 13:23:46 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:37153 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492066Ab0CSMXl (ORCPT ); Fri, 19 Mar 2010 13:23:41 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o2JCNdKM016579; Fri, 19 Mar 2010 13:23:39 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o2JCNdnP016577; Fri, 19 Mar 2010 13:23:39 +0100 Date: Fri, 19 Mar 2010 13:23:39 +0100 From: Ralf Baechle To: Maxime Bizon Cc: linux-mips@linux-mips.org Subject: Re: [PATCH 5/7] MIPS: bcm63xx: fix typo in cpu-feature-overrides file. Message-ID: <20100319122339.GM4554@linux-mips.org> References: <1264872898-28149-1-git-send-email-mbizon@freebox.fr> <1264872898-28149-6-git-send-email-mbizon@freebox.fr> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1264872898-28149-6-git-send-email-mbizon@freebox.fr> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26269 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Sat, Jan 30, 2010 at 06:34:56PM +0100, Maxime Bizon wrote: > Fix typo: CONFIG_BCMCPU_IS_63xx does not exists, > CONFIG_BCM63XX_CPU_63xx is the valid config option. Thanks, applied. Ralf From ralf@linux-mips.org Fri Mar 19 13:34:43 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Fri, 19 Mar 2010 13:34:46 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:55425 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492058Ab0CSMen (ORCPT ); Fri, 19 Mar 2010 13:34:43 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o2JCYZuS016883; Fri, 19 Mar 2010 13:34:36 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o2JCYXUI016881; Fri, 19 Mar 2010 13:34:33 +0100 Date: Fri, 19 Mar 2010 13:34:31 +0100 From: Ralf Baechle To: Maxime Bizon Cc: linux-mips@linux-mips.org Subject: Re: [PATCH 6/7] MIPS: bcm63xx: call board_register_device from device_initcall() Message-ID: <20100319123430.GN4554@linux-mips.org> References: <1264872898-28149-1-git-send-email-mbizon@freebox.fr> <1264872898-28149-7-git-send-email-mbizon@freebox.fr> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1264872898-28149-7-git-send-email-mbizon@freebox.fr> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26270 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Sat, Jan 30, 2010 at 06:34:57PM +0100, Maxime Bizon wrote: > Some device registration (eg leds), expect subsystem initcall to be > run first, so move board device registration to device_initcall(). Applied. Thanks, Ralf From ralf@linux-mips.org Fri Mar 19 13:44:56 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Fri, 19 Mar 2010 13:44:59 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:56318 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492292Ab0CSMo4 (ORCPT ); Fri, 19 Mar 2010 13:44:56 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o2JCiscv017101; Fri, 19 Mar 2010 13:44:54 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o2JCirPa017099; Fri, 19 Mar 2010 13:44:53 +0100 Date: Fri, 19 Mar 2010 13:44:53 +0100 From: Ralf Baechle To: Maxime Bizon Cc: linux-mips@linux-mips.org Subject: Re: [PATCH 6/7] MIPS: bcm63xx: call board_register_device from device_initcall() Message-ID: <20100319124453.GO4554@linux-mips.org> References: <1264872898-28149-1-git-send-email-mbizon@freebox.fr> <1264872898-28149-7-git-send-email-mbizon@freebox.fr> <20100319123430.GN4554@linux-mips.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20100319123430.GN4554@linux-mips.org> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26271 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Fri, Mar 19, 2010 at 01:34:31PM +0100, Ralf Baechle wrote: > > Some device registration (eg leds), expect subsystem initcall to be > > run first, so move board device registration to device_initcall(). > > Applied. Thanks, And pulled again. Florian posted his NACK together with a new patch which get sorted separately by patchworks so I didn't see the NACK right away. To avoid that sort of things do me a favor, post comments to an old patch and submissions of new patches in separate emails. Thanks. Ralf From ralf@linux-mips.org Fri Mar 19 13:47:47 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Fri, 19 Mar 2010 13:47:52 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:37322 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492292Ab0CSMrr (ORCPT ); Fri, 19 Mar 2010 13:47:47 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o2JCljNO017239; Fri, 19 Mar 2010 13:47:45 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o2JCliVY017237; Fri, 19 Mar 2010 13:47:44 +0100 Date: Fri, 19 Mar 2010 13:47:44 +0100 From: Ralf Baechle To: Florian Fainelli Cc: Maxime Bizon , linux-mips@linux-mips.org Subject: Re: [PATCH 6/7] MIPS: bcm63xx: call board_register_device from device_initcall() Message-ID: <20100319124744.GP4554@linux-mips.org> References: <1264872898-28149-1-git-send-email-mbizon@freebox.fr> <1264872898-28149-7-git-send-email-mbizon@freebox.fr> <201003021438.47105.ffainelli@freebox.fr> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <201003021438.47105.ffainelli@freebox.fr> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26272 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Tue, Mar 02, 2010 at 02:38:47PM +0100, Florian Fainelli wrote: And applied this one. Thanks, Ralf From ralf@linux-mips.org Fri Mar 19 14:22:45 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Fri, 19 Mar 2010 14:22:49 +0100 (CET) Received: from localhost.localdomain ([127.0.0.1]:38809 "EHLO h5.dl5rb.org.uk" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S1492155Ab0CSNWm (ORCPT ); Fri, 19 Mar 2010 14:22:42 +0100 Received: from h5.dl5rb.org.uk (localhost.localdomain [127.0.0.1]) by h5.dl5rb.org.uk (8.14.3/8.14.3) with ESMTP id o2JDMWBo018169; Fri, 19 Mar 2010 14:22:35 +0100 Received: (from ralf@localhost) by h5.dl5rb.org.uk (8.14.3/8.14.3/Submit) id o2JDMSuF018166; Fri, 19 Mar 2010 14:22:28 +0100 Date: Fri, 19 Mar 2010 14:22:25 +0100 From: Ralf Baechle To: Maxime Bizon Cc: linux-mips@linux-mips.org Subject: Re: [PATCH 7/7] MIPS: bcm63xx: initialize gpio_out_low & out_high to current value at boot. Message-ID: <20100319132223.GQ4554@linux-mips.org> References: <1264872898-28149-1-git-send-email-mbizon@freebox.fr> <1264872898-28149-8-git-send-email-mbizon@freebox.fr> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1264872898-28149-8-git-send-email-mbizon@freebox.fr> User-Agent: Mutt/1.5.20 (2009-08-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26273 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ralf@linux-mips.org Precedence: bulk X-list: linux-mips On Sat, Jan 30, 2010 at 06:34:58PM +0100, Maxime Bizon wrote: > To avoid glitch during gpio initialisation, fetch gpio output > registers values left by bootloader. Applied as well. Thanks, Ralf From p2@psychaos.be Fri Mar 19 18:52:49 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Fri, 19 Mar 2010 18:52:53 +0100 (CET) Received: from apfelkorn.psychaos.be ([195.144.77.38]:55974 "EHLO apfelkorn.psychaos.be" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492061Ab0CSRwt (ORCPT ); Fri, 19 Mar 2010 18:52:49 +0100 Received: from p2 by apfelkorn.psychaos.be with local (Exim 4.69) (envelope-from ) id 1NsgNJ-0007Af-52; Fri, 19 Mar 2010 19:52:45 +0200 Date: Fri, 19 Mar 2010 19:52:45 +0200 From: Peter 'p2' De Schrijver To: Jan Rovins Cc: David Daney , linux-mips@linux-mips.org Subject: Re: linux 2.6.33 on movidis x16 Message-ID: <20100319175244.GH2437@apfelkorn> References: <20100305141113.GD2437@apfelkorn> <4B9EB45D.8050106@adax.com> <20100318181734.GG2437@apfelkorn> <4BA27048.2010707@caviumnetworks.com> <4BA277FB.6080808@adax.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <4BA277FB.6080808@adax.com> X-Unexpected-Header: The spanish inquisition ! X-mate: Mate, mann gewohnt sich an alles X-Paddo: Munch, Munch User-Agent: Mutt/1.5.18 (2008-05-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26274 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: p2@debian.org Precedence: bulk X-list: linux-mips Hi, > You are right, it is called an x16. My memory is failing me, I should > have looked at the machine first. > Ok :) > from dmesg: > > cavium-ethernet: Cavium Networks Octeon SDK version 1.8.1, build 294 > Interface 0 has 4 ports (RGMII) > Interface 1 has 4 ports (RGMII) > Interface 2 has 4 ports (NPI) > Uniform Multi-Platform E-IDE driver Revision: 7.00alpha2 > > > looks like the PHYAD field has what you are loohing for, > ethtool output: --------------------------- > > for i in 0 1 2 3 4 5 6 7; do ethtool eth$i; done > Settings for eth0: > Supported ports: [ TP MII ] > Supported link modes: 10baseT/Half 10baseT/Full > 100baseT/Half 100baseT/Full > 1000baseT/Half 1000baseT/Full > Supports auto-negotiation: Yes > Advertised link modes: 10baseT/Half 10baseT/Full > 100baseT/Half 100baseT/Full > 1000baseT/Half 1000baseT/Full > Advertised auto-negotiation: Yes > Speed: 10Mb/s > Duplex: Half > Port: MII > PHYAD: 1 > Transceiver: internal > Auto-negotiation: on > Link detected: no > Settings for eth1: > Supported ports: [ TP MII ] > Supported link modes: 10baseT/Half 10baseT/Full > 100baseT/Half 100baseT/Full > 1000baseT/Half 1000baseT/Full > Supports auto-negotiation: Yes > Advertised link modes: 10baseT/Half 10baseT/Full > 100baseT/Half 100baseT/Full > 1000baseT/Half 1000baseT/Full > Advertised auto-negotiation: Yes > Speed: 10Mb/s > Duplex: Half > Port: MII > PHYAD: 2 > Transceiver: internal > Auto-negotiation: on > Link detected: no > Settings for eth2: > Supported ports: [ TP MII ] > Supported link modes: 10baseT/Half 10baseT/Full > 100baseT/Half 100baseT/Full > 1000baseT/Half 1000baseT/Full > Supports auto-negotiation: Yes > Advertised link modes: 10baseT/Half 10baseT/Full > 100baseT/Half 100baseT/Full > 1000baseT/Half 1000baseT/Full > Advertised auto-negotiation: Yes > Speed: 10Mb/s > Duplex: Half > Port: MII > PHYAD: 3 > Transceiver: internal > Auto-negotiation: on > Link detected: no > Settings for eth3: > Supported ports: [ TP MII ] > Supported link modes: 10baseT/Half 10baseT/Full > 100baseT/Half 100baseT/Full > 1000baseT/Half 1000baseT/Full > Supports auto-negotiation: Yes > Advertised link modes: 10baseT/Half 10baseT/Full > 100baseT/Half 100baseT/Full > 1000baseT/Half 1000baseT/Full > Advertised auto-negotiation: Yes > Speed: 10Mb/s > Duplex: Half > Port: MII > PHYAD: 4 > Transceiver: internal > Auto-negotiation: on > Link detected: no > Settings for eth4: > Supported ports: [ TP MII ] > Supported link modes: 10baseT/Half 10baseT/Full > 100baseT/Half 100baseT/Full > Supports auto-negotiation: Yes > Advertised link modes: 10baseT/Half 10baseT/Full > 100baseT/Half 100baseT/Full > Advertised auto-negotiation: Yes > Speed: 100Mb/s > Duplex: Full > Port: MII > PHYAD: 0 > Transceiver: internal > Auto-negotiation: on > Link detected: yes > Settings for eth5: > Supported ports: [ TP MII ] > Supported link modes: 10baseT/Half 10baseT/Full > 100baseT/Half 100baseT/Full > Supports auto-negotiation: Yes > Advertised link modes: 10baseT/Half 10baseT/Full > 100baseT/Half 100baseT/Full > Advertised auto-negotiation: Yes > Speed: 100Mb/s > Duplex: Full > Port: MII > PHYAD: 0 > Transceiver: internal > Auto-negotiation: on > Link detected: yes > Settings for eth6: > Supported ports: [ TP MII ] > Supported link modes: 10baseT/Half 10baseT/Full > 100baseT/Half 100baseT/Full > Supports auto-negotiation: Yes > Advertised link modes: 10baseT/Half 10baseT/Full > 100baseT/Half 100baseT/Full > Advertised auto-negotiation: Yes > Speed: 100Mb/s > Duplex: Full > Port: MII > PHYAD: 0 > Transceiver: internal > Auto-negotiation: on > Link detected: yes > Settings for eth7: > Supported ports: [ TP MII ] > Supported link modes: 10baseT/Half 10baseT/Full > 100baseT/Half 100baseT/Full > Supports auto-negotiation: Yes > Advertised link modes: 10baseT/Half 10baseT/Full > 100baseT/Half 100baseT/Full > Advertised auto-negotiation: Yes > Speed: 100Mb/s > Duplex: Full > Port: MII > PHYAD: 0 > Transceiver: internal > Auto-negotiation: on > Link detected: yes > PHYAD 0 for ports 4 - 7 seems a bit strange to me. Do you have the sources of the kernel you're running on those machines ? If so, could you look at int cvmx_helper_board_get_mii_address(int ipd_port) in cvmx-helper-board.c ? It contains a large switch statement. case CVMX_BOARD_TYPE_CUST_WSX16: is what I'm interested in. > ---------------------------------------------------------- > You may want to re-map the interfaces so that they show up in a sane > order, there is a cluster of 8 ethernet ports, 4 per row. Right now on > our boxes , eth0 shows up on the bottom row, second from the right. Thanks, Peter. From justinmattock@gmail.com Fri Mar 19 20:52:00 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Fri, 19 Mar 2010 20:52:06 +0100 (CET) Received: from mail-bw0-f215.google.com ([209.85.218.215]:61491 "EHLO mail-bw0-f215.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492061Ab0CSTwA (ORCPT ); Fri, 19 Mar 2010 20:52:00 +0100 Received: by bwz7 with SMTP id 7so2714922bwz.24 for ; Fri, 19 Mar 2010 12:51:53 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:from:to:cc:subject:date :message-id:x-mailer; bh=MqF5NRIF6EC8JcoeAYqFPct26gL8hfcEZdVjIrv+QnE=; b=C0BFcctQSFqfWjgLXqSdasNx2vPSpVd6KMRn1JbMazHXK3kUm22w8ROEhu7YPgnDyW /f7RZFZzo585jQoS6fRvFZjCistAaNdql/CYBuv4Wf+lWTa5jUBwmvIAH2ugssu8ZrsM 43+ba6+OGlEecHj/GX176Qb5D3TeZroxOoh48= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=from:to:cc:subject:date:message-id:x-mailer; b=VldEDSE3vEK6ybdwxWMbnmCEpbiV1FDawd5KeFG7Hig1DuG5y2OxkjQQoQTnXbgEr8 uR9/XoZ6xyLryr27A9vf/NFIrJbdblrvmPSeobH2TYdrSmpdklfaqNFIbNkD0/8/a8aU qwIvsbpwA8B37EageFEMztL0lvmmIplY1MQ8Q= Received: by 10.204.83.132 with SMTP id f4mr1232665bkl.73.1269028312977; Fri, 19 Mar 2010 12:51:52 -0700 (PDT) Received: from localhost.localdomain (cpe-76-173-26-187.socal.res.rr.com [76.173.26.187]) by mx.google.com with ESMTPS id 16sm910351bwz.1.2010.03.19.12.51.47 (version=TLSv1/SSLv3 cipher=RC4-MD5); Fri, 19 Mar 2010 12:51:51 -0700 (PDT) From: "Justin P. Mattock" To: trivial@kernel.org Cc: linux-arm-kernel@lists.infradead.org, linux-audit@redhat.com, uclinux-dist-devel@blackfin.uclinux.org, linux-cris-kernel@axis.com, linux-ia64@vger.kernel.org, linux-mips@linux-mips.org, linux-parisc@vger.kernel.org, linux-s390@vger.kernel.org, selinux@tycho.nsa.gov, sparclinux@vger.kernel.org, x86@kernel.org, netdev@vger.kernel.org, linux-kernel@vger.kernel.org, "Justin P. Mattock" Subject: [PATCH] Cosmetic:Partially remove deprecated __initcall() and change to Date: Fri, 19 Mar 2010 12:51:31 -0700 Message-Id: <1269028291-9103-1-git-send-email-justinmattock@gmail.com> X-Mailer: git-send-email 1.6.5.GIT Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26275 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: justinmattock@gmail.com Precedence: bulk X-list: linux-mips After doing some things on a small issue, I noticed through web surfing, that there were patches submitted pertaining that __initcall is deprecated, and device_initcall should be used. So as a change of subject(since what I was looking at was frustrating me),I decided to grep the whole tree and make the change(partially). Currently I'm running this patch on my system, kernel compiles without any errors or warnings.(thought there would be a speed increase but didn't see much(if any)). Biggest problem I have though is testing this on other hardware types (I only have a macbook,and an iMac). So please if you have the access to other arch/hardware types please test. Now what I mean by partially is the __initcall function is still there, so(if any) userspace apps/libs depend on this it's there so they dont break and/or any other subsystem, that needs time to make the changes. Note: the remaining files that still have __initcall in them are: (according to grep) arch/um/include/shared/init.h include/linux/init.h scripts/checkpatch.pl either I or somebody else, can change this(although a bit concerned about breaking things). Signed-off-by: Justin P. Mattock --- Documentation/DocBook/kernel-hacking.tmpl | 4 ++-- Documentation/cpu-freq/cpu-drivers.txt | 2 +- Documentation/kbuild/makefiles.txt | 2 +- arch/arm/mach-at91/leds.c | 2 +- arch/arm/mach-clps711x/p720t.c | 2 +- arch/arm/mach-ebsa110/leds.c | 2 +- arch/arm/mach-footbridge/cats-hw.c | 2 +- arch/arm/mach-footbridge/ebsa285-leds.c | 2 +- arch/arm/mach-footbridge/netwinder-hw.c | 2 +- arch/arm/mach-footbridge/netwinder-leds.c | 2 +- arch/arm/mach-ks8695/leds.c | 2 +- arch/arm/mach-omap1/leds.c | 2 +- arch/arm/mach-omap1/pm.c | 2 +- arch/arm/mach-orion5x/db88f5281-setup.c | 2 +- arch/arm/mach-orion5x/rd88f5182-setup.c | 2 +- arch/arm/mach-pxa/generic.c | 2 +- arch/arm/mach-pxa/pxa25x.c | 2 +- arch/arm/mach-shark/leds.c | 2 +- arch/blackfin/kernel/bfin_gpio.c | 2 +- arch/blackfin/mach-common/pm.c | 2 +- arch/cris/arch-v10/kernel/debugport.c | 2 +- arch/cris/arch-v10/kernel/fasttimer.c | 2 +- arch/cris/arch-v10/mm/init.c | 2 +- arch/cris/arch-v32/kernel/fasttimer.c | 2 +- arch/cris/arch-v32/kernel/pinmux.c | 2 +- arch/cris/arch-v32/kernel/signal.c | 2 +- arch/cris/arch-v32/mach-a3/io.c | 2 +- arch/cris/arch-v32/mach-a3/pinmux.c | 2 +- arch/cris/arch-v32/mach-fs/io.c | 2 +- arch/cris/arch-v32/mach-fs/pinmux.c | 2 +- arch/cris/kernel/profile.c | 2 +- arch/cris/kernel/time.c | 2 +- arch/cris/kernel/traps.c | 2 +- arch/frv/kernel/gdb-stub.c | 2 +- arch/frv/kernel/pm-mb93093.c | 2 +- arch/frv/kernel/pm.c | 2 +- arch/frv/kernel/sysctl.c | 2 +- arch/h8300/kernel/gpio.c | 2 +- arch/ia64/hp/sim/simeth.c | 2 +- arch/ia64/hp/sim/simserial.c | 2 +- arch/ia64/kernel/audit.c | 2 +- arch/ia64/kernel/crash.c | 2 +- arch/ia64/kernel/cyclone.c | 2 +- arch/ia64/kernel/perfmon.c | 2 +- arch/ia64/kernel/setup.c | 2 +- arch/ia64/kernel/uncached.c | 2 +- arch/ia64/kernel/unwind.c | 2 +- arch/ia64/mm/init.c | 2 +- arch/mips/Makefile | 2 +- arch/mips/kernel/unaligned.c | 2 +- arch/mips/lasat/sysctl.c | 2 +- arch/mips/math-emu/cp1emu.c | 2 +- arch/mips/nxp/pnx8550/common/proc.c | 2 +- arch/mips/sibyte/sb1250/bus_watcher.c | 2 +- arch/mn10300/kernel/gdb-stub.c | 2 +- arch/mn10300/kernel/mn10300-serial.c | 2 +- arch/mn10300/kernel/profile.c | 2 +- arch/parisc/kernel/pci-dma.c | 2 +- arch/parisc/kernel/pdc_chassis.c | 2 +- arch/powerpc/kernel/audit.c | 2 +- arch/powerpc/kernel/idle.c | 2 +- arch/powerpc/kernel/irq.c | 2 +- arch/powerpc/kernel/proc_powerpc.c | 2 +- arch/powerpc/kernel/prom.c | 4 ++-- arch/powerpc/kernel/rtas-proc.c | 2 +- arch/powerpc/kernel/rtasd.c | 2 +- arch/powerpc/kernel/sysfs.c | 2 +- arch/powerpc/kernel/tau_6xx.c | 2 +- arch/powerpc/kernel/vio.c | 2 +- arch/powerpc/platforms/iseries/lpevents.c | 2 +- arch/powerpc/platforms/iseries/mf.c | 2 +- arch/powerpc/platforms/iseries/proc.c | 2 +- arch/powerpc/platforms/iseries/viopath.c | 2 +- arch/powerpc/platforms/pseries/eeh.c | 2 +- arch/powerpc/platforms/pseries/hvCall_inst.c | 2 +- arch/powerpc/platforms/pseries/power.c | 2 +- arch/powerpc/platforms/pseries/ras.c | 2 +- arch/powerpc/platforms/pseries/reconfig.c | 2 +- arch/powerpc/xmon/xmon.c | 2 +- arch/s390/appldata/appldata_base.c | 2 +- arch/s390/kernel/audit.c | 2 +- arch/s390/kernel/compat_exec_domain.c | 2 +- arch/s390/kernel/ipl.c | 2 +- arch/s390/kernel/topology.c | 2 +- arch/sh/boards/board-edosk7760.c | 2 +- arch/sh/boards/board-sh7785lcr.c | 2 +- arch/sh/boards/mach-cayman/setup.c | 2 +- arch/sh/boards/mach-landisk/setup.c | 2 +- arch/sh/boards/mach-r2d/setup.c | 2 +- arch/sh/boards/mach-sdk7786/setup.c | 2 +- arch/sh/boards/mach-se/7206/setup.c | 2 +- arch/sh/boards/mach-se/7751/setup.c | 2 +- arch/sh/boards/mach-sh03/setup.c | 2 +- arch/sh/kernel/traps_64.c | 2 +- arch/sparc/kernel/apc.c | 2 +- arch/sparc/kernel/audit.c | 2 +- arch/sparc/kernel/mdesc.c | 2 +- arch/sparc/kernel/pmc.c | 2 +- arch/um/drivers/mconsole_kern.c | 8 ++++---- arch/um/drivers/net_kern.c | 2 +- arch/um/drivers/stderr_console.c | 2 +- arch/um/drivers/ubd_kern.c | 4 ++-- arch/um/kernel/exitcode.c | 2 +- arch/um/kernel/physmem.c | 2 +- arch/um/os-Linux/aio.c | 4 ++-- arch/um/os-Linux/skas/mem.c | 2 +- arch/um/os-Linux/skas/process.c | 2 +- arch/um/os-Linux/umid.c | 2 +- arch/um/sys-i386/tls.c | 2 +- arch/x86/kernel/audit_64.c | 2 +- arch/x86/kernel/tlb_uv.c | 4 ++-- arch/x86/kernel/vsyscall_64.c | 4 ++-- arch/x86/mm/dump_pagetables.c | 2 +- arch/x86/vdso/vdso32-setup.c | 4 ++-- arch/x86/vdso/vma.c | 2 +- arch/xtensa/platforms/iss/console.c | 2 +- drivers/net/arm/am79c961a.c | 2 +- drivers/net/hamradio/baycom_epp.c | 1 + drivers/net/hamradio/baycom_par.c | 1 + drivers/net/hamradio/baycom_ser_fdx.c | 1 + drivers/net/hamradio/baycom_ser_hdx.c | 1 + drivers/s390/char/sclp_cmd.c | 2 +- drivers/s390/char/sclp_config.c | 2 +- drivers/s390/char/sclp_cpi_sys.c | 2 +- drivers/s390/char/sclp_vt220.c | 2 +- drivers/s390/cio/blacklist.c | 2 +- drivers/staging/rtl8192u/ieee80211/api.c | 2 +- fs/aio.c | 2 +- fs/compat_ioctl.c | 2 +- ipc/ipc_sysctl.c | 2 +- ipc/mqueue.c | 2 +- ipc/util.c | 2 +- kernel/audit.c | 2 +- kernel/audit_tree.c | 2 +- kernel/dma.c | 2 +- kernel/futex.c | 2 +- kernel/lockdep_proc.c | 2 +- kernel/pid_namespace.c | 2 +- kernel/posix-cpu-timers.c | 2 +- kernel/posix-timers.c | 2 +- kernel/resource.c | 2 +- kernel/sched_debug.c | 2 +- kernel/time/timer_list.c | 2 +- kernel/time/timer_stats.c | 2 +- kernel/tracepoint.c | 2 +- kernel/utsname_sysctl.c | 2 +- lib/audit.c | 2 +- lib/debugobjects.c | 2 +- mm/bounce.c | 2 +- mm/memory.c | 2 +- mm/mm_init.c | 2 +- mm/slab.c | 2 +- mm/slub.c | 2 +- mm/swapfile.c | 2 +- net/ipv4/syncookies.c | 2 +- net/ipv4/sysctl_net_ipv4.c | 2 +- security/keys/proc.c | 2 +- security/selinux/hooks.c | 2 +- security/selinux/netif.c | 2 +- security/selinux/netlink.c | 2 +- security/selinux/netnode.c | 2 +- security/selinux/netport.c | 2 +- security/selinux/selinuxfs.c | 2 +- security/selinux/ss/services.c | 2 +- security/smack/smackfs.c | 2 +- sound/last.c | 2 +- 166 files changed, 176 insertions(+), 172 deletions(-) diff --git a/Documentation/DocBook/kernel-hacking.tmpl b/Documentation/DocBook/kernel-hacking.tmpl index 7b3f493..a682709 100644 --- a/Documentation/DocBook/kernel-hacking.tmpl +++ b/Documentation/DocBook/kernel-hacking.tmpl @@ -734,7 +734,7 @@ printk(KERN_INFO "my ip: %pI4\n", &ipaddress); - <function>__initcall()</function>/<function>module_init()</function> + <title><function>device_initcall()</function>/<function>module_init()</function> <filename class="headerfile">include/linux/init.h</filename> Many parts of the kernel are well served as a module @@ -750,7 +750,7 @@ printk(KERN_INFO "my ip: %pI4\n", &ipaddress); function is to be called at module insertion time (if the file is compiled as a module), or at boot time: if the file is not compiled as a module the module_init() macro - becomes equivalent to __initcall(), which + becomes equivalent to device_initcall(), which through linker magic ensures that the function is called on boot. diff --git a/Documentation/cpu-freq/cpu-drivers.txt b/Documentation/cpu-freq/cpu-drivers.txt index 6c30e93..12bb6e4 100644 --- a/Documentation/cpu-freq/cpu-drivers.txt +++ b/Documentation/cpu-freq/cpu-drivers.txt @@ -41,7 +41,7 @@ on what is necessary: 1.1 Initialization ------------------ -First of all, in an __initcall level 7 (module_init()) or later +First of all, in an device_initcall level 7 (module_init()) or later function check whether this kernel runs on the right CPU and the right chipset. If so, register a struct cpufreq_driver with the CPUfreq core using cpufreq_register_driver() diff --git a/Documentation/kbuild/makefiles.txt b/Documentation/kbuild/makefiles.txt index 71c602d..645ea71 100644 --- a/Documentation/kbuild/makefiles.txt +++ b/Documentation/kbuild/makefiles.txt @@ -159,7 +159,7 @@ more details, with real examples. built-in.o and succeeding instances will be ignored. Link order is significant, because certain functions - (module_init() / __initcall) will be called during boot in the + (module_init() / device_initcall) will be called during boot in the order they appear. So keep in mind that changing the link order may e.g. change the order in which your SCSI controllers are detected, and thus your disks are renumbered. diff --git a/arch/arm/mach-at91/leds.c b/arch/arm/mach-at91/leds.c index 0415a83..08e342d 100644 --- a/arch/arm/mach-at91/leds.c +++ b/arch/arm/mach-at91/leds.c @@ -179,7 +179,7 @@ static int __init leds_init(void) return 0; } -__initcall(leds_init); +device_initcall(leds_init); void __init at91_init_leds(u8 cpu_led, u8 timer_led) diff --git a/arch/arm/mach-clps711x/p720t.c b/arch/arm/mach-clps711x/p720t.c index 0d94a30..1047b23 100644 --- a/arch/arm/mach-clps711x/p720t.c +++ b/arch/arm/mach-clps711x/p720t.c @@ -121,5 +121,5 @@ static int p720t_hw_init(void) return 0; } -__initcall(p720t_hw_init); +device_initcall(p720t_hw_init); diff --git a/arch/arm/mach-ebsa110/leds.c b/arch/arm/mach-ebsa110/leds.c index 6a6ea57..8b57907 100644 --- a/arch/arm/mach-ebsa110/leds.c +++ b/arch/arm/mach-ebsa110/leds.c @@ -48,4 +48,4 @@ static int __init leds_init(void) return 0; } -__initcall(leds_init); +device_initcall(leds_init); diff --git a/arch/arm/mach-footbridge/cats-hw.c b/arch/arm/mach-footbridge/cats-hw.c index 1b996b2..1b616e5 100644 --- a/arch/arm/mach-footbridge/cats-hw.c +++ b/arch/arm/mach-footbridge/cats-hw.c @@ -69,7 +69,7 @@ static int __init cats_hw_init(void) return 0; } -__initcall(cats_hw_init); +device_initcall(cats_hw_init); /* * CATS uses soft-reboot by default, since diff --git a/arch/arm/mach-footbridge/ebsa285-leds.c b/arch/arm/mach-footbridge/ebsa285-leds.c index 4e10090..f162fb2 100644 --- a/arch/arm/mach-footbridge/ebsa285-leds.c +++ b/arch/arm/mach-footbridge/ebsa285-leds.c @@ -136,4 +136,4 @@ static int __init leds_init(void) return 0; } -__initcall(leds_init); +device_initcall(leds_init); diff --git a/arch/arm/mach-footbridge/netwinder-hw.c b/arch/arm/mach-footbridge/netwinder-hw.c index ac7ffa6..090b4cf 100644 --- a/arch/arm/mach-footbridge/netwinder-hw.c +++ b/arch/arm/mach-footbridge/netwinder-hw.c @@ -623,7 +623,7 @@ static int __init nw_hw_init(void) return 0; } -__initcall(nw_hw_init); +device_initcall(nw_hw_init); /* * Older NeTTroms either do not provide a parameters diff --git a/arch/arm/mach-footbridge/netwinder-leds.c b/arch/arm/mach-footbridge/netwinder-leds.c index 00269fe..f504682 100644 --- a/arch/arm/mach-footbridge/netwinder-leds.c +++ b/arch/arm/mach-footbridge/netwinder-leds.c @@ -136,4 +136,4 @@ static int __init leds_init(void) return 0; } -__initcall(leds_init); +device_initcall(leds_init); diff --git a/arch/arm/mach-ks8695/leds.c b/arch/arm/mach-ks8695/leds.c index 184ef74..97c3828 100644 --- a/arch/arm/mach-ks8695/leds.c +++ b/arch/arm/mach-ks8695/leds.c @@ -90,4 +90,4 @@ static int __init leds_init(void) return 0; } -__initcall(leds_init); +device_initcall(leds_init); diff --git a/arch/arm/mach-omap1/leds.c b/arch/arm/mach-omap1/leds.c index 277f356..bd447fa 100644 --- a/arch/arm/mach-omap1/leds.c +++ b/arch/arm/mach-omap1/leds.c @@ -63,4 +63,4 @@ omap_leds_init(void) return 0; } -__initcall(omap_leds_init); +device_initcall(omap_leds_init); diff --git a/arch/arm/mach-omap1/pm.c b/arch/arm/mach-omap1/pm.c index b1d3f9f..fda5f60 100644 --- a/arch/arm/mach-omap1/pm.c +++ b/arch/arm/mach-omap1/pm.c @@ -724,4 +724,4 @@ static int __init omap_pm_init(void) return 0; } -__initcall(omap_pm_init); +device_initcall(omap_pm_init); diff --git a/arch/arm/mach-orion5x/db88f5281-setup.c b/arch/arm/mach-orion5x/db88f5281-setup.c index d318bea..0c76893 100644 --- a/arch/arm/mach-orion5x/db88f5281-setup.c +++ b/arch/arm/mach-orion5x/db88f5281-setup.c @@ -197,7 +197,7 @@ static int __init db88f5281_7seg_init(void) return 0; } -__initcall(db88f5281_7seg_init); +device_initcall(db88f5281_7seg_init); /***************************************************************************** * PCI diff --git a/arch/arm/mach-orion5x/rd88f5182-setup.c b/arch/arm/mach-orion5x/rd88f5182-setup.c index a04f9e4..43e9a49 100644 --- a/arch/arm/mach-orion5x/rd88f5182-setup.c +++ b/arch/arm/mach-orion5x/rd88f5182-setup.c @@ -130,7 +130,7 @@ static int __init rd88f5182_dbgled_init(void) return 0; } -__initcall(rd88f5182_dbgled_init); +device_initcall(rd88f5182_dbgled_init); #endif diff --git a/arch/arm/mach-pxa/generic.c b/arch/arm/mach-pxa/generic.c index 3126a35..1d44a9a 100644 --- a/arch/arm/mach-pxa/generic.c +++ b/arch/arm/mach-pxa/generic.c @@ -12,7 +12,7 @@ * published by the Free Software Foundation. * * Since this file should be linked before any other machine specific file, - * the __initcall() here will be executed first. This serves as default + * the device_initcall() here will be executed first. This serves as default * initialization stuff for PXA machines which can be overridden later if * need be. */ diff --git a/arch/arm/mach-pxa/pxa25x.c b/arch/arm/mach-pxa/pxa25x.c index 0b9ad30..b0df545 100644 --- a/arch/arm/mach-pxa/pxa25x.c +++ b/arch/arm/mach-pxa/pxa25x.c @@ -12,7 +12,7 @@ * published by the Free Software Foundation. * * Since this file should be linked before any other machine specific file, - * the __initcall() here will be executed first. This serves as default + * the device_initcall() here will be executed first. This serves as default * initialization stuff for PXA machines which can be overridden later if * need be. */ diff --git a/arch/arm/mach-shark/leds.c b/arch/arm/mach-shark/leds.c index c9e32de..511693a 100644 --- a/arch/arm/mach-shark/leds.c +++ b/arch/arm/mach-shark/leds.c @@ -163,4 +163,4 @@ static int __init leds_init(void) return 0; } -__initcall(leds_init); +device_initcall(leds_init); diff --git a/arch/blackfin/kernel/bfin_gpio.c b/arch/blackfin/kernel/bfin_gpio.c index a174596..bc5ba54 100644 --- a/arch/blackfin/kernel/bfin_gpio.c +++ b/arch/blackfin/kernel/bfin_gpio.c @@ -1285,7 +1285,7 @@ static __init int gpio_register_proc(void) proc_gpio->read_proc = gpio_proc_read; return proc_gpio != NULL; } -__initcall(gpio_register_proc); +device_initcall(gpio_register_proc); #endif #ifdef CONFIG_GPIOLIB diff --git a/arch/blackfin/mach-common/pm.c b/arch/blackfin/mach-common/pm.c index 8837be4..dc36d3a 100644 --- a/arch/blackfin/mach-common/pm.c +++ b/arch/blackfin/mach-common/pm.c @@ -266,4 +266,4 @@ static int __init bfin_pm_init(void) return 0; } -__initcall(bfin_pm_init); +device_initcall(bfin_pm_init); diff --git a/arch/cris/arch-v10/kernel/debugport.c b/arch/cris/arch-v10/kernel/debugport.c index 99851ba..0348977 100644 --- a/arch/cris/arch-v10/kernel/debugport.c +++ b/arch/cris/arch-v10/kernel/debugport.c @@ -564,4 +564,4 @@ init_etrax_debug(void) #endif return 0; } -__initcall(init_etrax_debug); +device_initcall(init_etrax_debug); diff --git a/arch/cris/arch-v10/kernel/fasttimer.c b/arch/cris/arch-v10/kernel/fasttimer.c index 5ff08a8..b991462 100644 --- a/arch/cris/arch-v10/kernel/fasttimer.c +++ b/arch/cris/arch-v10/kernel/fasttimer.c @@ -879,4 +879,4 @@ int fast_timer_init(void) } return 0; } -__initcall(fast_timer_init); +device_initcall(fast_timer_init); diff --git a/arch/cris/arch-v10/mm/init.c b/arch/cris/arch-v10/mm/init.c index baa746c..908d829 100644 --- a/arch/cris/arch-v10/mm/init.c +++ b/arch/cris/arch-v10/mm/init.c @@ -217,7 +217,7 @@ __init init_ioremap(void) return 0; } -__initcall(init_ioremap); +device_initcall(init_ioremap); /* Helper function for the two below */ diff --git a/arch/cris/arch-v32/kernel/fasttimer.c b/arch/cris/arch-v32/kernel/fasttimer.c index 111caa1..221bcea 100644 --- a/arch/cris/arch-v32/kernel/fasttimer.c +++ b/arch/cris/arch-v32/kernel/fasttimer.c @@ -833,4 +833,4 @@ int fast_timer_init(void) } return 0; } -__initcall(fast_timer_init); +device_initcall(fast_timer_init); diff --git a/arch/cris/arch-v32/kernel/pinmux.c b/arch/cris/arch-v32/kernel/pinmux.c index f6f3637..c2dfcd3 100644 --- a/arch/cris/arch-v32/kernel/pinmux.c +++ b/arch/cris/arch-v32/kernel/pinmux.c @@ -226,4 +226,4 @@ crisv32_pinmux_dump(void) } } -__initcall(crisv32_pinmux_init); +device_initcall(crisv32_pinmux_init); diff --git a/arch/cris/arch-v32/kernel/signal.c b/arch/cris/arch-v32/kernel/signal.c index 372d0ca..d1e1b4e 100644 --- a/arch/cris/arch-v32/kernel/signal.c +++ b/arch/cris/arch-v32/kernel/signal.c @@ -672,4 +672,4 @@ cris_init_signal(void) return 0; } -__initcall(cris_init_signal); +device_initcall(cris_init_signal); diff --git a/arch/cris/arch-v32/mach-a3/io.c b/arch/cris/arch-v32/mach-a3/io.c index 090ceb9..ea40a9b 100644 --- a/arch/cris/arch-v32/mach-a3/io.c +++ b/arch/cris/arch-v32/mach-a3/io.c @@ -95,7 +95,7 @@ static int __init crisv32_io_init(void) return ret; } -__initcall(crisv32_io_init); +device_initcall(crisv32_io_init); int crisv32_io_get(struct crisv32_iopin *iopin, unsigned int port, unsigned int pin) diff --git a/arch/cris/arch-v32/mach-a3/pinmux.c b/arch/cris/arch-v32/mach-a3/pinmux.c index 18648ef..14bbc78 100644 --- a/arch/cris/arch-v32/mach-a3/pinmux.c +++ b/arch/cris/arch-v32/mach-a3/pinmux.c @@ -383,4 +383,4 @@ crisv32_pinmux_dump(void) } } -__initcall(crisv32_pinmux_init); +device_initcall(crisv32_pinmux_init); diff --git a/arch/cris/arch-v32/mach-fs/io.c b/arch/cris/arch-v32/mach-fs/io.c index a695866..fce8438 100644 --- a/arch/cris/arch-v32/mach-fs/io.c +++ b/arch/cris/arch-v32/mach-fs/io.c @@ -128,7 +128,7 @@ static int __init crisv32_io_init(void) return ret; } -__initcall(crisv32_io_init); +device_initcall(crisv32_io_init); int crisv32_io_get(struct crisv32_iopin *iopin, unsigned int port, unsigned int pin) diff --git a/arch/cris/arch-v32/mach-fs/pinmux.c b/arch/cris/arch-v32/mach-fs/pinmux.c index 38f29ee..751f991 100644 --- a/arch/cris/arch-v32/mach-fs/pinmux.c +++ b/arch/cris/arch-v32/mach-fs/pinmux.c @@ -306,4 +306,4 @@ void crisv32_pinmux_dump(void) } } -__initcall(crisv32_pinmux_init); +device_initcall(crisv32_pinmux_init); diff --git a/arch/cris/kernel/profile.c b/arch/cris/kernel/profile.c index 9aa5711..2472f3f 100644 --- a/arch/cris/kernel/profile.c +++ b/arch/cris/kernel/profile.c @@ -82,4 +82,4 @@ __init init_cris_profile(void) return 0; } -__initcall(init_cris_profile); +device_initcall(init_cris_profile); diff --git a/arch/cris/kernel/time.c b/arch/cris/kernel/time.c index a05dd31..7554e96 100644 --- a/arch/cris/kernel/time.c +++ b/arch/cris/kernel/time.c @@ -160,4 +160,4 @@ __init init_udelay(void) return 0; } -__initcall(init_udelay); +device_initcall(init_udelay); diff --git a/arch/cris/kernel/traps.c b/arch/cris/kernel/traps.c index 541efbf..8f77b68 100644 --- a/arch/cris/kernel/traps.c +++ b/arch/cris/kernel/traps.c @@ -177,7 +177,7 @@ oops_nmi_register(void) return 0; } -__initcall(oops_nmi_register); +device_initcall(oops_nmi_register); #endif diff --git a/arch/frv/kernel/gdb-stub.c b/arch/frv/kernel/gdb-stub.c index 7ca8a6b..71d564c 100644 --- a/arch/frv/kernel/gdb-stub.c +++ b/arch/frv/kernel/gdb-stub.c @@ -2017,7 +2017,7 @@ static int __init gdbstub_postinit(void) return 0; } /* end gdbstub_postinit() */ -__initcall(gdbstub_postinit); +device_initcall(gdbstub_postinit); #endif /*****************************************************************************/ diff --git a/arch/frv/kernel/pm-mb93093.c b/arch/frv/kernel/pm-mb93093.c index eaa7b58..3caa348 100644 --- a/arch/frv/kernel/pm-mb93093.c +++ b/arch/frv/kernel/pm-mb93093.c @@ -61,5 +61,5 @@ static int __init mb93093_pm_init(void) return 0; } -__initcall(mb93093_pm_init); +device_initcall(mb93093_pm_init); diff --git a/arch/frv/kernel/pm.c b/arch/frv/kernel/pm.c index 5fa3889..b1aad48 100644 --- a/arch/frv/kernel/pm.c +++ b/arch/frv/kernel/pm.c @@ -348,6 +348,6 @@ static int __init pm_init(void) return 0; } -__initcall(pm_init); +device_initcall(pm_init); #endif diff --git a/arch/frv/kernel/sysctl.c b/arch/frv/kernel/sysctl.c index 035516c..3280f8b 100644 --- a/arch/frv/kernel/sysctl.c +++ b/arch/frv/kernel/sysctl.c @@ -217,4 +217,4 @@ static int __init frv_sysctl_init(void) return 0; } -__initcall(frv_sysctl_init); +device_initcall(frv_sysctl_init); diff --git a/arch/h8300/kernel/gpio.c b/arch/h8300/kernel/gpio.c index 6a25dd5..6ea7dd9 100644 --- a/arch/h8300/kernel/gpio.c +++ b/arch/h8300/kernel/gpio.c @@ -164,7 +164,7 @@ static __init int register_proc(void) return proc_gpio != NULL; } -__initcall(register_proc); +device_initcall(register_proc); #endif void __init h8300_gpio_init(void) diff --git a/arch/ia64/hp/sim/simeth.c b/arch/ia64/hp/sim/simeth.c index 7e81966..e669908 100644 --- a/arch/ia64/hp/sim/simeth.c +++ b/arch/ia64/hp/sim/simeth.c @@ -524,4 +524,4 @@ set_multicast_list(struct net_device *dev) printk(KERN_WARNING "%s: set_multicast_list called\n", dev->name); } -__initcall(simeth_probe); +device_initcall(simeth_probe); diff --git a/arch/ia64/hp/sim/simserial.c b/arch/ia64/hp/sim/simserial.c index 2bef526..470c0a1 100644 --- a/arch/ia64/hp/sim/simserial.c +++ b/arch/ia64/hp/sim/simserial.c @@ -980,5 +980,5 @@ simrs_init (void) } #ifndef MODULE -__initcall(simrs_init); +device_initcall(simrs_init); #endif diff --git a/arch/ia64/kernel/audit.c b/arch/ia64/kernel/audit.c index 96a9d18..c7d79ba 100644 --- a/arch/ia64/kernel/audit.c +++ b/arch/ia64/kernel/audit.c @@ -57,4 +57,4 @@ static int __init audit_classes_init(void) return 0; } -__initcall(audit_classes_init); +device_initcall(audit_classes_init); diff --git a/arch/ia64/kernel/crash.c b/arch/ia64/kernel/crash.c index b942f40..07218af 100644 --- a/arch/ia64/kernel/crash.c +++ b/arch/ia64/kernel/crash.c @@ -282,5 +282,5 @@ machine_crash_setup(void) return 0; } -__initcall(machine_crash_setup); +device_initcall(machine_crash_setup); diff --git a/arch/ia64/kernel/cyclone.c b/arch/ia64/kernel/cyclone.c index 71e3586..66fb97e 100644 --- a/arch/ia64/kernel/cyclone.c +++ b/arch/ia64/kernel/cyclone.c @@ -125,4 +125,4 @@ int __init init_cyclone_clock(void) return 0; } -__initcall(init_cyclone_clock); +device_initcall(init_cyclone_clock); diff --git a/arch/ia64/kernel/perfmon.c b/arch/ia64/kernel/perfmon.c index 703062c..cb7199a 100644 --- a/arch/ia64/kernel/perfmon.c +++ b/arch/ia64/kernel/perfmon.c @@ -6700,7 +6700,7 @@ pfm_init(void) return 0; } -__initcall(pfm_init); +device_initcall(pfm_init); /* * this function is called before pfm_init() diff --git a/arch/ia64/kernel/setup.c b/arch/ia64/kernel/setup.c index 41ae6a5..8322a01 100644 --- a/arch/ia64/kernel/setup.c +++ b/arch/ia64/kernel/setup.c @@ -243,7 +243,7 @@ static int __init register_memory(void) return 0; } -__initcall(register_memory); +device_initcall(register_memory); #ifdef CONFIG_KEXEC diff --git a/arch/ia64/kernel/uncached.c b/arch/ia64/kernel/uncached.c index a595823..4a9812f 100644 --- a/arch/ia64/kernel/uncached.c +++ b/arch/ia64/kernel/uncached.c @@ -279,4 +279,4 @@ static int __init uncached_init(void) return 0; } -__initcall(uncached_init); +device_initcall(uncached_init); diff --git a/arch/ia64/kernel/unwind.c b/arch/ia64/kernel/unwind.c index b6c0e63..62de896 100644 --- a/arch/ia64/kernel/unwind.c +++ b/arch/ia64/kernel/unwind.c @@ -2223,7 +2223,7 @@ create_gate_table (void) return 0; } -__initcall(create_gate_table); +device_initcall(create_gate_table); void __init unw_init (void) diff --git a/arch/ia64/mm/init.c b/arch/ia64/mm/init.c index ed41759..da31b31 100644 --- a/arch/ia64/mm/init.c +++ b/arch/ia64/mm/init.c @@ -716,4 +716,4 @@ per_linux32_init(void) return 0; } -__initcall(per_linux32_init); +device_initcall(per_linux32_init); diff --git a/arch/mips/Makefile b/arch/mips/Makefile index 2f2eac2..0341a9e 100644 --- a/arch/mips/Makefile +++ b/arch/mips/Makefile @@ -534,7 +534,7 @@ load-$(CONFIG_SGI_IP32) += 0xffffffff80004000 # # This is a LIB so that it links at the end, and initcalls are later # the sequence; but it is built as an object so that modules don't get -# removed (as happens, even if they have __initcall/module_init) +# removed (as happens, even if they have device_initcall/module_init) # core-$(CONFIG_SIBYTE_BCM112X) += arch/mips/sibyte/sb1250/ core-$(CONFIG_SIBYTE_BCM112X) += arch/mips/sibyte/common/ diff --git a/arch/mips/kernel/unaligned.c b/arch/mips/kernel/unaligned.c index 69b039c..7233ac8 100644 --- a/arch/mips/kernel/unaligned.c +++ b/arch/mips/kernel/unaligned.c @@ -567,5 +567,5 @@ static int __init debugfs_unaligned(void) return -ENOMEM; return 0; } -__initcall(debugfs_unaligned); +device_initcall(debugfs_unaligned); #endif diff --git a/arch/mips/lasat/sysctl.c b/arch/mips/lasat/sysctl.c index d87ffd0..6cc8d2f 100644 --- a/arch/mips/lasat/sysctl.c +++ b/arch/mips/lasat/sysctl.c @@ -285,4 +285,4 @@ static int __init lasat_register_sysctl(void) return 0; } -__initcall(lasat_register_sysctl); +device_initcall(lasat_register_sysctl); diff --git a/arch/mips/math-emu/cp1emu.c b/arch/mips/math-emu/cp1emu.c index 8f2f8e9..3936be4 100644 --- a/arch/mips/math-emu/cp1emu.c +++ b/arch/mips/math-emu/cp1emu.c @@ -1325,5 +1325,5 @@ static int __init debugfs_fpuemu(void) return 0; } -__initcall(debugfs_fpuemu); +device_initcall(debugfs_fpuemu); #endif diff --git a/arch/mips/nxp/pnx8550/common/proc.c b/arch/mips/nxp/pnx8550/common/proc.c index af094cd..9a10d8c 100644 --- a/arch/mips/nxp/pnx8550/common/proc.c +++ b/arch/mips/nxp/pnx8550/common/proc.c @@ -108,4 +108,4 @@ static int pnx8550_proc_init( void ) return 0; } -__initcall(pnx8550_proc_init); +device_initcall(pnx8550_proc_init); diff --git a/arch/mips/sibyte/sb1250/bus_watcher.c b/arch/mips/sibyte/sb1250/bus_watcher.c index 45274bd..e8e7f23 100644 --- a/arch/mips/sibyte/sb1250/bus_watcher.c +++ b/arch/mips/sibyte/sb1250/bus_watcher.c @@ -255,4 +255,4 @@ int __init sibyte_bus_watcher(void) return 0; } -__initcall(sibyte_bus_watcher); +device_initcall(sibyte_bus_watcher); diff --git a/arch/mn10300/kernel/gdb-stub.c b/arch/mn10300/kernel/gdb-stub.c index 41b1170..22e0613 100644 --- a/arch/mn10300/kernel/gdb-stub.c +++ b/arch/mn10300/kernel/gdb-stub.c @@ -1884,7 +1884,7 @@ static int __init gdbstub_postinit(void) return 0; } -__initcall(gdbstub_postinit); +device_initcall(gdbstub_postinit); #endif /* diff --git a/arch/mn10300/kernel/mn10300-serial.c b/arch/mn10300/kernel/mn10300-serial.c index ef34d5a..9cada2f 100644 --- a/arch/mn10300/kernel/mn10300-serial.c +++ b/arch/mn10300/kernel/mn10300-serial.c @@ -1350,7 +1350,7 @@ static int __init mn10300_serial_init(void) return ret; } -__initcall(mn10300_serial_init); +device_initcall(mn10300_serial_init); #ifdef CONFIG_MN10300_TTYSM_CONSOLE diff --git a/arch/mn10300/kernel/profile.c b/arch/mn10300/kernel/profile.c index 20d7d03..e799032 100644 --- a/arch/mn10300/kernel/profile.c +++ b/arch/mn10300/kernel/profile.c @@ -48,4 +48,4 @@ static __init int profile_init(void) return 0; } -__initcall(profile_init); +device_initcall(profile_init); diff --git a/arch/parisc/kernel/pci-dma.c b/arch/parisc/kernel/pci-dma.c index c07f618..e7c12c0 100644 --- a/arch/parisc/kernel/pci-dma.c +++ b/arch/parisc/kernel/pci-dma.c @@ -410,7 +410,7 @@ pcxl_dma_init(void) return 0; } -__initcall(pcxl_dma_init); +device_initcall(pcxl_dma_init); static void * pa11_dma_alloc_consistent (struct device *dev, size_t size, dma_addr_t *dma_handle, gfp_t flag) { diff --git a/arch/parisc/kernel/pdc_chassis.c b/arch/parisc/kernel/pdc_chassis.c index d47ba1a..35ad4cb 100644 --- a/arch/parisc/kernel/pdc_chassis.c +++ b/arch/parisc/kernel/pdc_chassis.c @@ -295,7 +295,7 @@ static int __init pdc_chassis_create_procfs(void) return 0; } -__initcall(pdc_chassis_create_procfs); +device_initcall(pdc_chassis_create_procfs); #endif /* CONFIG_PROC_FS */ #endif /* CONFIG_PDC_CHASSIS_WARN */ diff --git a/arch/powerpc/kernel/audit.c b/arch/powerpc/kernel/audit.c index a4dab7c..e721e2e 100644 --- a/arch/powerpc/kernel/audit.c +++ b/arch/powerpc/kernel/audit.c @@ -80,4 +80,4 @@ static int __init audit_classes_init(void) return 0; } -__initcall(audit_classes_init); +device_initcall(audit_classes_init); diff --git a/arch/powerpc/kernel/idle.c b/arch/powerpc/kernel/idle.c index 049dda6..694861f 100644 --- a/arch/powerpc/kernel/idle.c +++ b/arch/powerpc/kernel/idle.c @@ -134,5 +134,5 @@ register_powersave_nap_sysctl(void) return 0; } -__initcall(register_powersave_nap_sysctl); +device_initcall(register_powersave_nap_sysctl); #endif diff --git a/arch/powerpc/kernel/irq.c b/arch/powerpc/kernel/irq.c index 64f6f20..2e97ae9 100644 --- a/arch/powerpc/kernel/irq.c +++ b/arch/powerpc/kernel/irq.c @@ -1203,7 +1203,7 @@ static int __init irq_debugfs_init(void) return 0; } -__initcall(irq_debugfs_init); +device_initcall(irq_debugfs_init); #endif /* CONFIG_VIRQ_DEBUG */ #ifdef CONFIG_PPC64 diff --git a/arch/powerpc/kernel/proc_powerpc.c b/arch/powerpc/kernel/proc_powerpc.c index 1ed3b8d..f53bdca 100644 --- a/arch/powerpc/kernel/proc_powerpc.c +++ b/arch/powerpc/kernel/proc_powerpc.c @@ -91,7 +91,7 @@ static int __init proc_ppc64_init(void) return 0; } -__initcall(proc_ppc64_init); +device_initcall(proc_ppc64_init); #endif /* CONFIG_PPC64 */ diff --git a/arch/powerpc/kernel/prom.c b/arch/powerpc/kernel/prom.c index 05131d6..628d6df 100644 --- a/arch/powerpc/kernel/prom.c +++ b/arch/powerpc/kernel/prom.c @@ -845,7 +845,7 @@ static int __init prom_reconfig_setup(void) { return pSeries_reconfig_notifier_register(&prom_reconfig_nb); } -__initcall(prom_reconfig_setup); +device_initcall(prom_reconfig_setup); #endif /* Find the device node for a given logical cpu number, also returns the cpu @@ -909,5 +909,5 @@ static int __init export_flat_device_tree(void) return 0; } -__initcall(export_flat_device_tree); +device_initcall(export_flat_device_tree); #endif diff --git a/arch/powerpc/kernel/rtas-proc.c b/arch/powerpc/kernel/rtas-proc.c index 8777fb0..7578df5 100644 --- a/arch/powerpc/kernel/rtas-proc.c +++ b/arch/powerpc/kernel/rtas-proc.c @@ -279,7 +279,7 @@ static int __init proc_rtas_init(void) return 0; } -__initcall(proc_rtas_init); +device_initcall(proc_rtas_init); static int parse_number(const char __user *p, size_t count, unsigned long *val) { diff --git a/arch/powerpc/kernel/rtasd.c b/arch/powerpc/kernel/rtasd.c index 2e4832a..1bfd919 100644 --- a/arch/powerpc/kernel/rtasd.c +++ b/arch/powerpc/kernel/rtasd.c @@ -508,7 +508,7 @@ static int __init rtas_init(void) return 0; } -__initcall(rtas_init); +device_initcall(rtas_init); static int __init surveillance_setup(char *str) { diff --git a/arch/powerpc/kernel/sysfs.c b/arch/powerpc/kernel/sysfs.c index e235e52..6ddb889 100644 --- a/arch/powerpc/kernel/sysfs.c +++ b/arch/powerpc/kernel/sysfs.c @@ -92,7 +92,7 @@ static int __init smt_setup(void) of_node_put(options); return 0; } -__initcall(smt_setup); +device_initcall(smt_setup); static int __init setup_smt_snooze_delay(char *str) { diff --git a/arch/powerpc/kernel/tau_6xx.c b/arch/powerpc/kernel/tau_6xx.c index a753b72..a38d207 100644 --- a/arch/powerpc/kernel/tau_6xx.c +++ b/arch/powerpc/kernel/tau_6xx.c @@ -248,7 +248,7 @@ int __init TAU_init(void) return 0; } -__initcall(TAU_init); +device_initcall(TAU_init); /* * return current temp diff --git a/arch/powerpc/kernel/vio.c b/arch/powerpc/kernel/vio.c index 77f6421..c74aae9 100644 --- a/arch/powerpc/kernel/vio.c +++ b/arch/powerpc/kernel/vio.c @@ -1303,7 +1303,7 @@ static int __init vio_bus_init(void) return 0; } -__initcall(vio_bus_init); +device_initcall(vio_bus_init); static ssize_t name_show(struct device *dev, struct device_attribute *attr, char *buf) diff --git a/arch/powerpc/platforms/iseries/lpevents.c b/arch/powerpc/platforms/iseries/lpevents.c index b0f8a85..66bc6ff 100644 --- a/arch/powerpc/platforms/iseries/lpevents.c +++ b/arch/powerpc/platforms/iseries/lpevents.c @@ -337,5 +337,5 @@ static int __init proc_lpevents_init(void) &proc_lpevents_operations); return 0; } -__initcall(proc_lpevents_init); +device_initcall(proc_lpevents_init); diff --git a/arch/powerpc/platforms/iseries/mf.c b/arch/powerpc/platforms/iseries/mf.c index 6617915..d140c09 100644 --- a/arch/powerpc/platforms/iseries/mf.c +++ b/arch/powerpc/platforms/iseries/mf.c @@ -1299,7 +1299,7 @@ static int __init mf_proc_init(void) return 0; } -__initcall(mf_proc_init); +device_initcall(mf_proc_init); #endif /* CONFIG_PROC_FS */ diff --git a/arch/powerpc/platforms/iseries/proc.c b/arch/powerpc/platforms/iseries/proc.c index 0676368..1e9ac90 100644 --- a/arch/powerpc/platforms/iseries/proc.c +++ b/arch/powerpc/platforms/iseries/proc.c @@ -117,4 +117,4 @@ static int __init iseries_proc_init(void) &proc_titantod_operations); return 0; } -__initcall(iseries_proc_init); +device_initcall(iseries_proc_init); diff --git a/arch/powerpc/platforms/iseries/viopath.c b/arch/powerpc/platforms/iseries/viopath.c index 5aea94f..daac760 100644 --- a/arch/powerpc/platforms/iseries/viopath.c +++ b/arch/powerpc/platforms/iseries/viopath.c @@ -186,7 +186,7 @@ static int __init vio_proc_init(void) proc_create("iSeries/config", 0, NULL, &proc_viopath_operations); return 0; } -__initcall(vio_proc_init); +device_initcall(vio_proc_init); /* See if a given LP is active. Allow for invalid lps to be passed in * and just return invalid diff --git a/arch/powerpc/platforms/pseries/eeh.c b/arch/powerpc/platforms/pseries/eeh.c index 7df7fbb..cb72726 100644 --- a/arch/powerpc/platforms/pseries/eeh.c +++ b/arch/powerpc/platforms/pseries/eeh.c @@ -1287,4 +1287,4 @@ static int __init eeh_init_proc(void) proc_create("ppc64/eeh", 0, NULL, &proc_eeh_operations); return 0; } -__initcall(eeh_init_proc); +device_initcall(eeh_init_proc); diff --git a/arch/powerpc/platforms/pseries/hvCall_inst.c b/arch/powerpc/platforms/pseries/hvCall_inst.c index 1fefae7..ca1b4e2 100644 --- a/arch/powerpc/platforms/pseries/hvCall_inst.c +++ b/arch/powerpc/platforms/pseries/hvCall_inst.c @@ -164,4 +164,4 @@ static int __init hcall_inst_init(void) return 0; } -__initcall(hcall_inst_init); +device_initcall(hcall_inst_init); diff --git a/arch/powerpc/platforms/pseries/power.c b/arch/powerpc/platforms/pseries/power.c index 6d62662..0eafaf8 100644 --- a/arch/powerpc/platforms/pseries/power.c +++ b/arch/powerpc/platforms/pseries/power.c @@ -77,5 +77,5 @@ static int __init apo_pm_init(void) { return (sysfs_create_file(power_kobj, &auto_poweron_attr.attr)); } -__initcall(apo_pm_init); +device_initcall(apo_pm_init); #endif diff --git a/arch/powerpc/platforms/pseries/ras.c b/arch/powerpc/platforms/pseries/ras.c index d20b96e..247bad9 100644 --- a/arch/powerpc/platforms/pseries/ras.c +++ b/arch/powerpc/platforms/pseries/ras.c @@ -152,7 +152,7 @@ static int __init init_ras_IRQ(void) return 0; } -__initcall(init_ras_IRQ); +device_initcall(init_ras_IRQ); /* * Handle power subsystem events (EPOW). diff --git a/arch/powerpc/platforms/pseries/reconfig.c b/arch/powerpc/platforms/pseries/reconfig.c index a2305d2..d82ca6c 100644 --- a/arch/powerpc/platforms/pseries/reconfig.c +++ b/arch/powerpc/platforms/pseries/reconfig.c @@ -557,4 +557,4 @@ static int proc_ppc64_create_ofdt(void) return 0; } -__initcall(proc_ppc64_create_ofdt); +device_initcall(proc_ppc64_create_ofdt); diff --git a/arch/powerpc/xmon/xmon.c b/arch/powerpc/xmon/xmon.c index 8bad7d5..cd1ce39 100644 --- a/arch/powerpc/xmon/xmon.c +++ b/arch/powerpc/xmon/xmon.c @@ -2751,7 +2751,7 @@ static int __init setup_xmon_sysrq(void) register_sysrq_key('x', &sysrq_xmon_op); return 0; } -__initcall(setup_xmon_sysrq); +device_initcall(setup_xmon_sysrq); #endif /* CONFIG_MAGIC_SYSRQ */ static int __initdata xmon_early, xmon_off; diff --git a/arch/s390/appldata/appldata_base.c b/arch/s390/appldata/appldata_base.c index 5c91995..af7ccb3 100644 --- a/arch/s390/appldata/appldata_base.c +++ b/arch/s390/appldata/appldata_base.c @@ -659,7 +659,7 @@ out_driver: return rc; } -__initcall(appldata_init); +device_initcall(appldata_init); /**************************** init / exit ******************************/ diff --git a/arch/s390/kernel/audit.c b/arch/s390/kernel/audit.c index f4932c2..8a9023a 100644 --- a/arch/s390/kernel/audit.c +++ b/arch/s390/kernel/audit.c @@ -75,4 +75,4 @@ static int __init audit_classes_init(void) return 0; } -__initcall(audit_classes_init); +device_initcall(audit_classes_init); diff --git a/arch/s390/kernel/compat_exec_domain.c b/arch/s390/kernel/compat_exec_domain.c index 914d494..e0715f6 100644 --- a/arch/s390/kernel/compat_exec_domain.c +++ b/arch/s390/kernel/compat_exec_domain.c @@ -26,4 +26,4 @@ static int __init s390_init (void) return 0; } -__initcall(s390_init); +device_initcall(s390_init); diff --git a/arch/s390/kernel/ipl.c b/arch/s390/kernel/ipl.c index 7eedbbc..19906a7 100644 --- a/arch/s390/kernel/ipl.c +++ b/arch/s390/kernel/ipl.c @@ -1806,7 +1806,7 @@ static int __init s390_ipl_init(void) return 0; } -__initcall(s390_ipl_init); +device_initcall(s390_ipl_init); static void __init strncpy_skip_quote(char *dst, char *src, int n) { diff --git a/arch/s390/kernel/topology.c b/arch/s390/kernel/topology.c index 14ef6f0..1a1d7ee 100644 --- a/arch/s390/kernel/topology.c +++ b/arch/s390/kernel/topology.c @@ -291,7 +291,7 @@ out: update_cpu_core_map(); return rc; } -__initcall(init_topology_update); +device_initcall(init_topology_update); void __init s390_init_cpu_topology(void) { diff --git a/arch/sh/boards/board-edosk7760.c b/arch/sh/boards/board-edosk7760.c index 35dc099..f47ac82 100644 --- a/arch/sh/boards/board-edosk7760.c +++ b/arch/sh/boards/board-edosk7760.c @@ -182,7 +182,7 @@ static int __init init_edosk7760_devices(void) return platform_add_devices(edosk7760_devices, ARRAY_SIZE(edosk7760_devices)); } -__initcall(init_edosk7760_devices); +device_initcall(init_edosk7760_devices); /* * The Machine Vector diff --git a/arch/sh/boards/board-sh7785lcr.c b/arch/sh/boards/board-sh7785lcr.c index fe7e686..ee65ff0 100644 --- a/arch/sh/boards/board-sh7785lcr.c +++ b/arch/sh/boards/board-sh7785lcr.c @@ -284,7 +284,7 @@ static int __init sh7785lcr_devices_setup(void) return platform_add_devices(sh7785lcr_devices, ARRAY_SIZE(sh7785lcr_devices)); } -__initcall(sh7785lcr_devices_setup); +device_initcall(sh7785lcr_devices_setup); /* Initialize IRQ setting */ void __init init_sh7785lcr_IRQ(void) diff --git a/arch/sh/boards/mach-cayman/setup.c b/arch/sh/boards/mach-cayman/setup.c index 7e8216a..e89e8e1 100644 --- a/arch/sh/boards/mach-cayman/setup.c +++ b/arch/sh/boards/mach-cayman/setup.c @@ -165,7 +165,7 @@ static int __init smsc_superio_setup(void) return 0; } -__initcall(smsc_superio_setup); +device_initcall(smsc_superio_setup); static void __iomem *cayman_ioport_map(unsigned long port, unsigned int len) { diff --git a/arch/sh/boards/mach-landisk/setup.c b/arch/sh/boards/mach-landisk/setup.c index 50337ac..62f26f4 100644 --- a/arch/sh/boards/mach-landisk/setup.c +++ b/arch/sh/boards/mach-landisk/setup.c @@ -83,7 +83,7 @@ static int __init landisk_devices_setup(void) ARRAY_SIZE(landisk_devices)); } -__initcall(landisk_devices_setup); +device_initcall(landisk_devices_setup); static void __init landisk_setup(char **cmdline_p) { diff --git a/arch/sh/boards/mach-r2d/setup.c b/arch/sh/boards/mach-r2d/setup.c index b84df6a..4b98a52 100644 --- a/arch/sh/boards/mach-r2d/setup.c +++ b/arch/sh/boards/mach-r2d/setup.c @@ -258,7 +258,7 @@ static int __init rts7751r2d_devices_setup(void) return platform_add_devices(rts7751r2d_devices, ARRAY_SIZE(rts7751r2d_devices)); } -__initcall(rts7751r2d_devices_setup); +device_initcall(rts7751r2d_devices_setup); static void rts7751r2d_power_off(void) { diff --git a/arch/sh/boards/mach-sdk7786/setup.c b/arch/sh/boards/mach-sdk7786/setup.c index f094ea2..ddcc73d 100644 --- a/arch/sh/boards/mach-sdk7786/setup.c +++ b/arch/sh/boards/mach-sdk7786/setup.c @@ -132,7 +132,7 @@ static int __init sdk7786_devices_setup(void) return sdk7786_i2c_setup(); } -__initcall(sdk7786_devices_setup); +device_initcall(sdk7786_devices_setup); static int sdk7786_mode_pins(void) { diff --git a/arch/sh/boards/mach-se/7206/setup.c b/arch/sh/boards/mach-se/7206/setup.c index 8f5c65d..91d0823 100644 --- a/arch/sh/boards/mach-se/7206/setup.c +++ b/arch/sh/boards/mach-se/7206/setup.c @@ -77,7 +77,7 @@ static int __init se7206_devices_setup(void) { return platform_add_devices(se7206_devices, ARRAY_SIZE(se7206_devices)); } -__initcall(se7206_devices_setup); +device_initcall(se7206_devices_setup); /* * The Machine Vector diff --git a/arch/sh/boards/mach-se/7751/setup.c b/arch/sh/boards/mach-se/7751/setup.c index 5057251..69e1479 100644 --- a/arch/sh/boards/mach-se/7751/setup.c +++ b/arch/sh/boards/mach-se/7751/setup.c @@ -48,7 +48,7 @@ static int __init se7751_devices_setup(void) { return platform_add_devices(se7751_devices, ARRAY_SIZE(se7751_devices)); } -__initcall(se7751_devices_setup); +device_initcall(se7751_devices_setup); /* * The Machine Vector diff --git a/arch/sh/boards/mach-sh03/setup.c b/arch/sh/boards/mach-sh03/setup.c index af4a0c0..d4f79b2 100644 --- a/arch/sh/boards/mach-sh03/setup.c +++ b/arch/sh/boards/mach-sh03/setup.c @@ -96,7 +96,7 @@ static int __init sh03_devices_setup(void) return platform_add_devices(sh03_devices, ARRAY_SIZE(sh03_devices)); } -__initcall(sh03_devices_setup); +device_initcall(sh03_devices_setup); static struct sh_machine_vector mv_sh03 __initmv = { .mv_name = "Interface (CTP/PCI-SH03)", diff --git a/arch/sh/kernel/traps_64.c b/arch/sh/kernel/traps_64.c index e3f92eb..0868d8d 100644 --- a/arch/sh/kernel/traps_64.c +++ b/arch/sh/kernel/traps_64.c @@ -923,7 +923,7 @@ static int __init init_sysctl(void) return 0; } -__initcall(init_sysctl); +device_initcall(init_sysctl); asmlinkage void do_debug_interrupt(unsigned long code, struct pt_regs *regs) diff --git a/arch/sparc/kernel/apc.c b/arch/sparc/kernel/apc.c index 71ec90b..768b51b 100644 --- a/arch/sparc/kernel/apc.c +++ b/arch/sparc/kernel/apc.c @@ -188,4 +188,4 @@ static int __init apc_init(void) * and is easiest to ioremap when SBus is already * initialized, so we install ourselves thusly: */ -__initcall(apc_init); +device_initcall(apc_init); diff --git a/arch/sparc/kernel/audit.c b/arch/sparc/kernel/audit.c index 8fff0ac..4be9a78 100644 --- a/arch/sparc/kernel/audit.c +++ b/arch/sparc/kernel/audit.c @@ -80,4 +80,4 @@ static int __init audit_classes_init(void) return 0; } -__initcall(audit_classes_init); +device_initcall(audit_classes_init); diff --git a/arch/sparc/kernel/mdesc.c b/arch/sparc/kernel/mdesc.c index cdc91d9..0fb5a9c 100644 --- a/arch/sparc/kernel/mdesc.c +++ b/arch/sparc/kernel/mdesc.c @@ -903,7 +903,7 @@ static int __init mdesc_misc_init(void) return misc_register(&mdesc_misc); } -__initcall(mdesc_misc_init); +device_initcall(mdesc_misc_init); void __init sun4v_mdesc_init(void) { diff --git a/arch/sparc/kernel/pmc.c b/arch/sparc/kernel/pmc.c index 5e4563d..74a99b5 100644 --- a/arch/sparc/kernel/pmc.c +++ b/arch/sparc/kernel/pmc.c @@ -93,4 +93,4 @@ static int __init pmc_init(void) * and is easiest to ioremap when SBus is already * initialized, so we install ourselves thusly: */ -__initcall(pmc_init); +device_initcall(pmc_init); diff --git a/arch/um/drivers/mconsole_kern.c b/arch/um/drivers/mconsole_kern.c index de317d0..c55fa3a 100644 --- a/arch/um/drivers/mconsole_kern.c +++ b/arch/um/drivers/mconsole_kern.c @@ -493,7 +493,7 @@ static int __init mem_mc_init(void) return 0; } -__initcall(mem_mc_init); +device_initcall(mem_mc_init); #define CONFIG_BUF_SIZE 64 @@ -816,7 +816,7 @@ static int __init mconsole_init(void) return 1; } -__initcall(mconsole_init); +device_initcall(mconsole_init); static ssize_t mconsole_proc_write(struct file *file, const char __user *buffer, size_t count, loff_t *pos) @@ -873,7 +873,7 @@ void unlock_notify(void) spin_unlock(¬ify_spinlock); } -__initcall(create_proc_mconsole); +device_initcall(create_proc_mconsole); #define NOTIFY "notify:" @@ -923,7 +923,7 @@ static int add_notifier(void) return 0; } -__initcall(add_notifier); +device_initcall(add_notifier); char *mconsole_notify_socket(void) { diff --git a/arch/um/drivers/net_kern.c b/arch/um/drivers/net_kern.c index a74245a..3864903 100644 --- a/arch/um/drivers/net_kern.c +++ b/arch/um/drivers/net_kern.c @@ -824,7 +824,7 @@ static int uml_net_init(void) return 0; } -__initcall(uml_net_init); +device_initcall(uml_net_init); static void close_devices(void) { diff --git a/arch/um/drivers/stderr_console.c b/arch/um/drivers/stderr_console.c index d07a97f..29eb7ce 100644 --- a/arch/um/drivers/stderr_console.c +++ b/arch/um/drivers/stderr_console.c @@ -59,4 +59,4 @@ static int __init unregister_stderr(void) return 0; } -__initcall(unregister_stderr); +device_initcall(unregister_stderr); diff --git a/arch/um/drivers/ubd_kern.c b/arch/um/drivers/ubd_kern.c index c1ff690..d5de090 100644 --- a/arch/um/drivers/ubd_kern.c +++ b/arch/um/drivers/ubd_kern.c @@ -1010,7 +1010,7 @@ static int __init ubd_mc_init(void) return 0; } -__initcall(ubd_mc_init); +device_initcall(ubd_mc_init); static int __init ubd0_init(void) { @@ -1024,7 +1024,7 @@ static int __init ubd0_init(void) return 0; } -__initcall(ubd0_init); +device_initcall(ubd0_init); /* Used in ubd_init, which is an initcall */ static struct platform_driver ubd_driver = { diff --git a/arch/um/kernel/exitcode.c b/arch/um/kernel/exitcode.c index 829df49..cb05ff8 100644 --- a/arch/um/kernel/exitcode.c +++ b/arch/um/kernel/exitcode.c @@ -75,4 +75,4 @@ static int make_proc_exitcode(void) return 0; } -__initcall(make_proc_exitcode); +device_initcall(make_proc_exitcode); diff --git a/arch/um/kernel/physmem.c b/arch/um/kernel/physmem.c index a1a9090..87537f0 100644 --- a/arch/um/kernel/physmem.c +++ b/arch/um/kernel/physmem.c @@ -209,4 +209,4 @@ static int setup_iomem(void) return 0; } -__initcall(setup_iomem); +device_initcall(setup_iomem); diff --git a/arch/um/os-Linux/aio.c b/arch/um/os-Linux/aio.c index 57e3d46..337437e 100644 --- a/arch/um/os-Linux/aio.c +++ b/arch/um/os-Linux/aio.c @@ -343,13 +343,13 @@ static int init_aio(void) } /* - * The reason for the __initcall/__uml_exitcall asymmetry is that init_aio + * The reason for the device_initcall/__uml_exitcall asymmetry is that init_aio * needs to be called when the kernel is running because it calls run_helper, * which needs get_free_page. exit_aio is a __uml_exitcall because the generic * kernel does not run __exitcalls on shutdown, and can't because many of them * break when called outside of module unloading. */ -__initcall(init_aio); +device_initcall(init_aio); static void exit_aio(void) { diff --git a/arch/um/os-Linux/skas/mem.c b/arch/um/os-Linux/skas/mem.c index d261f17..a0a816a 100644 --- a/arch/um/os-Linux/skas/mem.c +++ b/arch/um/os-Linux/skas/mem.c @@ -46,7 +46,7 @@ static int __init init_syscall_regs(void) return 0; } -__initcall(init_syscall_regs); +device_initcall(init_syscall_regs); extern int proc_mm; diff --git a/arch/um/os-Linux/skas/process.c b/arch/um/os-Linux/skas/process.c index d6e0a22..25ef142 100644 --- a/arch/um/os-Linux/skas/process.c +++ b/arch/um/os-Linux/skas/process.c @@ -473,7 +473,7 @@ static int __init init_thread_regs(void) return 0; } -__initcall(init_thread_regs); +device_initcall(init_thread_regs); int copy_context_skas0(unsigned long new_stack, int pid) { diff --git a/arch/um/os-Linux/umid.c b/arch/um/os-Linux/umid.c index a27defb..23f7b37 100644 --- a/arch/um/os-Linux/umid.c +++ b/arch/um/os-Linux/umid.c @@ -326,7 +326,7 @@ static int __init make_umid_init(void) return 0; } -__initcall(make_umid_init); +device_initcall(make_umid_init); int __init umid_file_name(char *name, char *buf, int len) { diff --git a/arch/um/sys-i386/tls.c b/arch/um/sys-i386/tls.c index c6c7131..b285fd7 100644 --- a/arch/um/sys-i386/tls.c +++ b/arch/um/sys-i386/tls.c @@ -393,4 +393,4 @@ static int __init __setup_host_supports_tls(void) return 0; } -__initcall(__setup_host_supports_tls); +device_initcall(__setup_host_supports_tls); diff --git a/arch/x86/kernel/audit_64.c b/arch/x86/kernel/audit_64.c index 06d3e5a..bab6ddc 100644 --- a/arch/x86/kernel/audit_64.c +++ b/arch/x86/kernel/audit_64.c @@ -78,4 +78,4 @@ static int __init audit_classes_init(void) return 0; } -__initcall(audit_classes_init); +device_initcall(audit_classes_init); diff --git a/arch/x86/kernel/tlb_uv.c b/arch/x86/kernel/tlb_uv.c index 364d015..eb71685 100644 --- a/arch/x86/kernel/tlb_uv.c +++ b/arch/x86/kernel/tlb_uv.c @@ -860,5 +860,5 @@ static int __init uv_bau_init(void) return 0; } -__initcall(uv_bau_init); -__initcall(uv_ptc_init); +device_initcall(uv_bau_init); +device_initcall(uv_ptc_init); diff --git a/arch/x86/kernel/vsyscall_64.c b/arch/x86/kernel/vsyscall_64.c index 1c0c6ab..df45aaf 100644 --- a/arch/x86/kernel/vsyscall_64.c +++ b/arch/x86/kernel/vsyscall_64.c @@ -244,7 +244,7 @@ static ctl_table kernel_root_table2[] = { }; #endif -/* Assume __initcall executes before all user space. Hopefully kmod +/* Assume device_initcall executes before all user space. Hopefully kmod doesn't violate that. We'll find out if it does. */ static void __cpuinit vsyscall_set_cpu(int cpu) { @@ -306,4 +306,4 @@ static int __init vsyscall_init(void) return 0; } -__initcall(vsyscall_init); +device_initcall(vsyscall_init); diff --git a/arch/x86/mm/dump_pagetables.c b/arch/x86/mm/dump_pagetables.c index a725b7f..a84535d 100644 --- a/arch/x86/mm/dump_pagetables.c +++ b/arch/x86/mm/dump_pagetables.c @@ -349,7 +349,7 @@ static int pt_dump_init(void) return 0; } -__initcall(pt_dump_init); +device_initcall(pt_dump_init); MODULE_LICENSE("GPL"); MODULE_AUTHOR("Arjan van de Ven "); MODULE_DESCRIPTION("Kernel debugging helper that dumps pagetables"); diff --git a/arch/x86/vdso/vdso32-setup.c b/arch/x86/vdso/vdso32-setup.c index 02b442e..6c73b85 100644 --- a/arch/x86/vdso/vdso32-setup.c +++ b/arch/x86/vdso/vdso32-setup.c @@ -374,7 +374,7 @@ int arch_setup_additional_pages(struct linux_binprm *bprm, int uses_interp) #ifdef CONFIG_X86_64 -__initcall(sysenter_setup); +device_initcall(sysenter_setup); #ifdef CONFIG_SYSCTL /* Register vsyscall32 into the ABI table */ @@ -405,7 +405,7 @@ static __init int ia32_binfmt_init(void) register_sysctl_table(abi_root_table2); return 0; } -__initcall(ia32_binfmt_init); +device_initcall(ia32_binfmt_init); #endif #else /* CONFIG_X86_32 */ diff --git a/arch/x86/vdso/vma.c b/arch/x86/vdso/vma.c index 21e1aeb..860addc 100644 --- a/arch/x86/vdso/vma.c +++ b/arch/x86/vdso/vma.c @@ -73,7 +73,7 @@ static int __init init_vdso_vars(void) vdso_enabled = 0; return -ENOMEM; } -__initcall(init_vdso_vars); +device_initcall(init_vdso_vars); struct linux_binprm; diff --git a/arch/xtensa/platforms/iss/console.c b/arch/xtensa/platforms/iss/console.c index e60a1f5..cdf1715 100644 --- a/arch/xtensa/platforms/iss/console.c +++ b/arch/xtensa/platforms/iss/console.c @@ -248,7 +248,7 @@ static __exit void rs_exit(void) } -/* We use `late_initcall' instead of just `__initcall' as a workaround for +/* We use `late_initcall' instead of just `device_initcall' as a workaround for * the fact that (1) simcons_tty_init can't be called before tty_init, * (2) tty_init is called via `module_init', (3) if statically linked, * module_init == device_init, and (4) there's no ordering of init lists. diff --git a/drivers/net/arm/am79c961a.c b/drivers/net/arm/am79c961a.c index f1f58c5..cc2c734 100644 --- a/drivers/net/arm/am79c961a.c +++ b/drivers/net/arm/am79c961a.c @@ -775,4 +775,4 @@ static int __init am79c961_init(void) return platform_driver_register(&am79c961_driver); } -__initcall(am79c961_init); +device_initcall(am79c961_init); diff --git a/drivers/net/hamradio/baycom_epp.c b/drivers/net/hamradio/baycom_epp.c index a3c0dc9..fc6f80e 100644 --- a/drivers/net/hamradio/baycom_epp.c +++ b/drivers/net/hamradio/baycom_epp.c @@ -35,6 +35,7 @@ * removed some pre-2.2 kernel compatibility cruft * 0.6 10.08.1999 Check if parport can do SPP and is safe to access during interrupt contexts * 0.7 12.02.2000 adapted to softnet driver interface + * 0.8 03.18.2010 update deprecated __initcall to device_initcall * */ diff --git a/drivers/net/hamradio/baycom_par.c b/drivers/net/hamradio/baycom_par.c index 5f5af9a..f2ddffa 100644 --- a/drivers/net/hamradio/baycom_par.c +++ b/drivers/net/hamradio/baycom_par.c @@ -64,6 +64,7 @@ * 0.8 12.02.2000 adapted to softnet driver interface * removed direct parport access, uses parport driver methods * 0.9 03.07.2000 fix interface name handling + * 0.10 03.18.2010 update deprecated __initcall to device_initcall */ /*****************************************************************************/ diff --git a/drivers/net/hamradio/baycom_ser_fdx.c b/drivers/net/hamradio/baycom_ser_fdx.c index 0cab992..d90ad93 100644 --- a/drivers/net/hamradio/baycom_ser_fdx.c +++ b/drivers/net/hamradio/baycom_ser_fdx.c @@ -67,6 +67,7 @@ * 0.8 10.08.1999 use module_init/module_exit * 0.9 12.02.2000 adapted to softnet driver interface * 0.10 03.07.2000 fix interface name handling + * 0.11 03.18.2010 update deprecated __initcall to device_initcall */ /*****************************************************************************/ diff --git a/drivers/net/hamradio/baycom_ser_hdx.c b/drivers/net/hamradio/baycom_ser_hdx.c index 1686f6d..3033cae 100644 --- a/drivers/net/hamradio/baycom_ser_hdx.c +++ b/drivers/net/hamradio/baycom_ser_hdx.c @@ -57,6 +57,7 @@ * 0.8 10.08.1999 use module_init/module_exit * 0.9 12.02.2000 adapted to softnet driver interface * 0.10 03.07.2000 fix interface name handling + * 0.11 03.18.2010 update deprecated __initcall to device_initcall */ /*****************************************************************************/ diff --git a/drivers/s390/char/sclp_cmd.c b/drivers/s390/char/sclp_cmd.c index b3beab6..642b3bf 100644 --- a/drivers/s390/char/sclp_cmd.c +++ b/drivers/s390/char/sclp_cmd.c @@ -628,7 +628,7 @@ out: free_page((unsigned long) sccb); return rc; } -__initcall(sclp_detect_standby_memory); +device_initcall(sclp_detect_standby_memory); #endif /* CONFIG_MEMORY_HOTPLUG */ diff --git a/drivers/s390/char/sclp_config.c b/drivers/s390/char/sclp_config.c index b497afe..c257eb3 100644 --- a/drivers/s390/char/sclp_config.c +++ b/drivers/s390/char/sclp_config.c @@ -87,4 +87,4 @@ static int __init sclp_conf_init(void) return rc; } -__initcall(sclp_conf_init); +device_initcall(sclp_conf_init); diff --git a/drivers/s390/char/sclp_cpi_sys.c b/drivers/s390/char/sclp_cpi_sys.c index 62c2647..1ebb83c 100644 --- a/drivers/s390/char/sclp_cpi_sys.c +++ b/drivers/s390/char/sclp_cpi_sys.c @@ -426,4 +426,4 @@ static int __init cpi_init(void) return rc; } -__initcall(cpi_init); +device_initcall(cpi_init); diff --git a/drivers/s390/char/sclp_vt220.c b/drivers/s390/char/sclp_vt220.c index 3796ffd..deff553 100644 --- a/drivers/s390/char/sclp_vt220.c +++ b/drivers/s390/char/sclp_vt220.c @@ -707,7 +707,7 @@ out_driver: put_tty_driver(driver); return rc; } -__initcall(sclp_vt220_tty_init); +device_initcall(sclp_vt220_tty_init); static void __sclp_vt220_flush_buffer(void) { diff --git a/drivers/s390/cio/blacklist.c b/drivers/s390/cio/blacklist.c index 7eab9ab..363fd02 100644 --- a/drivers/s390/cio/blacklist.c +++ b/drivers/s390/cio/blacklist.c @@ -398,6 +398,6 @@ cio_ignore_proc_init (void) return 0; } -__initcall (cio_ignore_proc_init); +device_initcall(cio_ignore_proc_init); #endif /* CONFIG_PROC_FS */ diff --git a/drivers/staging/rtl8192u/ieee80211/api.c b/drivers/staging/rtl8192u/ieee80211/api.c index c627d02..156d867 100644 --- a/drivers/staging/rtl8192u/ieee80211/api.c +++ b/drivers/staging/rtl8192u/ieee80211/api.c @@ -229,7 +229,7 @@ static int __init init_crypto(void) return 0; } -__initcall(init_crypto); +device_initcall(init_crypto); /* EXPORT_SYMBOL_GPL(crypto_register_alg); diff --git a/fs/aio.c b/fs/aio.c index 1cf12b3..80ddf44 100644 --- a/fs/aio.c +++ b/fs/aio.c @@ -92,7 +92,7 @@ static int __init aio_setup(void) return 0; } -__initcall(aio_setup); +device_initcall(aio_setup); static void aio_free_ring(struct kioctx *ctx) { diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c index 6d55b61..b579a82 100644 --- a/fs/compat_ioctl.c +++ b/fs/compat_ioctl.c @@ -1780,4 +1780,4 @@ static int __init init_sys32_ioctl(void) init_sys32_ioctl_cmp, NULL); return 0; } -__initcall(init_sys32_ioctl); +device_initcall(init_sys32_ioctl); diff --git a/ipc/ipc_sysctl.c b/ipc/ipc_sysctl.c index 56410fa..3d59d44 100644 --- a/ipc/ipc_sysctl.c +++ b/ipc/ipc_sysctl.c @@ -209,4 +209,4 @@ static int __init ipc_sysctl_init(void) return 0; } -__initcall(ipc_sysctl_init); +device_initcall(ipc_sysctl_init); diff --git a/ipc/mqueue.c b/ipc/mqueue.c index e4e3f04..c58ca57 100644 --- a/ipc/mqueue.c +++ b/ipc/mqueue.c @@ -1322,4 +1322,4 @@ out_sysctl: return error; } -__initcall(init_mqueue_fs); +device_initcall(init_mqueue_fs); diff --git a/ipc/util.c b/ipc/util.c index 79ce84e..d82ec6c 100644 --- a/ipc/util.c +++ b/ipc/util.c @@ -106,7 +106,7 @@ static int __init ipc_init(void) register_ipcns_notifier(&init_ipc_ns); return 0; } -__initcall(ipc_init); +device_initcall(ipc_init); /** * ipc_init_ids - initialise IPC identifiers diff --git a/kernel/audit.c b/kernel/audit.c index 78f7f86..d489f47 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -989,7 +989,7 @@ static int __init audit_init(void) return 0; } -__initcall(audit_init); +device_initcall(audit_init); /* Process kernel command-line parameter at boot time. audit=0 or audit=1. */ static int __init audit_enable(char *str) diff --git a/kernel/audit_tree.c b/kernel/audit_tree.c index 028e856..e54d2c6 100644 --- a/kernel/audit_tree.c +++ b/kernel/audit_tree.c @@ -919,4 +919,4 @@ static int __init audit_tree_init(void) return 0; } -__initcall(audit_tree_init); +device_initcall(audit_tree_init); diff --git a/kernel/dma.c b/kernel/dma.c index f903189..b54e161 100644 --- a/kernel/dma.c +++ b/kernel/dma.c @@ -153,7 +153,7 @@ static int __init proc_dma_init(void) return 0; } -__initcall(proc_dma_init); +device_initcall(proc_dma_init); #endif EXPORT_SYMBOL(request_dma); diff --git a/kernel/futex.c b/kernel/futex.c index e7a35f1..94efb8c 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -2670,4 +2670,4 @@ static int __init futex_init(void) return 0; } -__initcall(futex_init); +device_initcall(futex_init); diff --git a/kernel/lockdep_proc.c b/kernel/lockdep_proc.c index d4aba4f..a4088c9 100644 --- a/kernel/lockdep_proc.c +++ b/kernel/lockdep_proc.c @@ -687,5 +687,5 @@ static int __init lockdep_proc_init(void) return 0; } -__initcall(lockdep_proc_init); +device_initcall(lockdep_proc_init); diff --git a/kernel/pid_namespace.c b/kernel/pid_namespace.c index 79aac93..2c7b5e7 100644 --- a/kernel/pid_namespace.c +++ b/kernel/pid_namespace.c @@ -189,4 +189,4 @@ static __init int pid_namespaces_init(void) return 0; } -__initcall(pid_namespaces_init); +device_initcall(pid_namespaces_init); diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c index 1a22dfd..036d64e 100644 --- a/kernel/posix-cpu-timers.c +++ b/kernel/posix-cpu-timers.c @@ -1727,4 +1727,4 @@ static __init int init_posix_cpu_timers(void) return 0; } -__initcall(init_posix_cpu_timers); +device_initcall(init_posix_cpu_timers); diff --git a/kernel/posix-timers.c b/kernel/posix-timers.c index 00d1fda..e8c2d04 100644 --- a/kernel/posix-timers.c +++ b/kernel/posix-timers.c @@ -309,7 +309,7 @@ static __init int init_posix_timers(void) return 0; } -__initcall(init_posix_timers); +device_initcall(init_posix_timers); static void schedule_next_timer(struct k_itimer *timr) { diff --git a/kernel/resource.c b/kernel/resource.c index 2d5be5d..e678177 100644 --- a/kernel/resource.c +++ b/kernel/resource.c @@ -136,7 +136,7 @@ static int __init ioresources_init(void) proc_create("iomem", 0, NULL, &proc_iomem_operations); return 0; } -__initcall(ioresources_init); +device_initcall(ioresources_init); #endif /* CONFIG_PROC_FS */ diff --git a/kernel/sched_debug.c b/kernel/sched_debug.c index 67f95aa..c17988b 100644 --- a/kernel/sched_debug.c +++ b/kernel/sched_debug.c @@ -379,7 +379,7 @@ static int __init init_sched_debug_procfs(void) return 0; } -__initcall(init_sched_debug_procfs); +device_initcall(init_sched_debug_procfs); void proc_sched_show_task(struct task_struct *p, struct seq_file *m) { diff --git a/kernel/time/timer_list.c b/kernel/time/timer_list.c index bdfb8dd..962b3b2 100644 --- a/kernel/time/timer_list.c +++ b/kernel/time/timer_list.c @@ -296,4 +296,4 @@ static int __init init_timer_list_procfs(void) return -ENOMEM; return 0; } -__initcall(init_timer_list_procfs); +device_initcall(init_timer_list_procfs); diff --git a/kernel/time/timer_stats.c b/kernel/time/timer_stats.c index 2f3b585..517552b 100644 --- a/kernel/time/timer_stats.c +++ b/kernel/time/timer_stats.c @@ -422,4 +422,4 @@ static int __init init_tstats_procfs(void) return -ENOMEM; return 0; } -__initcall(init_tstats_procfs); +device_initcall(init_tstats_procfs); diff --git a/kernel/tracepoint.c b/kernel/tracepoint.c index cc89be5..c4d12e8 100644 --- a/kernel/tracepoint.c +++ b/kernel/tracepoint.c @@ -580,7 +580,7 @@ static int init_tracepoints(void) { return register_module_notifier(&tracepoint_module_nb); } -__initcall(init_tracepoints); +device_initcall(init_tracepoints); #endif /* CONFIG_MODULES */ diff --git a/kernel/utsname_sysctl.c b/kernel/utsname_sysctl.c index a2cd77e..7cbb001 100644 --- a/kernel/utsname_sysctl.c +++ b/kernel/utsname_sysctl.c @@ -111,4 +111,4 @@ static int __init utsname_sysctl_init(void) return 0; } -__initcall(utsname_sysctl_init); +device_initcall(utsname_sysctl_init); diff --git a/lib/audit.c b/lib/audit.c index 8e7dc1c..26f41a9 100644 --- a/lib/audit.c +++ b/lib/audit.c @@ -63,4 +63,4 @@ static int __init audit_classes_init(void) return 0; } -__initcall(audit_classes_init); +device_initcall(audit_classes_init); diff --git a/lib/debugobjects.c b/lib/debugobjects.c index a9a8996..ee41495 100644 --- a/lib/debugobjects.c +++ b/lib/debugobjects.c @@ -666,7 +666,7 @@ err: return -ENOMEM; } -__initcall(debug_objects_init_debugfs); +device_initcall(debug_objects_init_debugfs); #else static inline void debug_objects_init_debugfs(void) { } diff --git a/mm/bounce.c b/mm/bounce.c index a2b76a5..b6df140 100644 --- a/mm/bounce.c +++ b/mm/bounce.c @@ -39,7 +39,7 @@ static __init int init_emergency_pool(void) return 0; } -__initcall(init_emergency_pool); +device_initcall(init_emergency_pool); /* * highmem version, map in to vec diff --git a/mm/memory.c b/mm/memory.c index 5b7f200..b39f34e 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -3231,7 +3231,7 @@ static int __init gate_vma_init(void) gate_vma.vm_flags |= VM_ALWAYSDUMP; return 0; } -__initcall(gate_vma_init); +device_initcall(gate_vma_init); #endif struct vm_area_struct *get_gate_vma(struct task_struct *tsk) diff --git a/mm/mm_init.c b/mm/mm_init.c index 4e0e265..6586ce5 100644 --- a/mm/mm_init.c +++ b/mm/mm_init.c @@ -149,4 +149,4 @@ static int __init mm_sysfs_init(void) return 0; } -__initcall(mm_sysfs_init); +device_initcall(mm_sysfs_init); diff --git a/mm/slab.c b/mm/slab.c index a9f325b..891218c 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -1597,7 +1597,7 @@ static int __init cpucache_init(void) start_cpu_timer(cpu); return 0; } -__initcall(cpucache_init); +device_initcall(cpucache_init); /* * Interface to system's page allocator. No need to hold the cache-lock. diff --git a/mm/slub.c b/mm/slub.c index b364844..a048aa5 100644 --- a/mm/slub.c +++ b/mm/slub.c @@ -4577,7 +4577,7 @@ static int __init slab_sysfs_init(void) return 0; } -__initcall(slab_sysfs_init); +device_initcall(slab_sysfs_init); #endif /* diff --git a/mm/swapfile.c b/mm/swapfile.c index 6cd0a8f..c1e2b6d 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -1764,7 +1764,7 @@ static int __init procswaps_init(void) proc_create("swaps", 0, NULL, &proc_swaps_operations); return 0; } -__initcall(procswaps_init); +device_initcall(procswaps_init); #endif /* CONFIG_PROC_FS */ #ifdef MAX_SWAPFILES_CHECK diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c index 5c24db4..5bdd7d2 100644 --- a/net/ipv4/syncookies.c +++ b/net/ipv4/syncookies.c @@ -32,7 +32,7 @@ static __init int init_syncookies(void) get_random_bytes(syncookie_secret, sizeof(syncookie_secret)); return 0; } -__initcall(init_syncookies); +device_initcall(init_syncookies); #define COOKIEBITS 24 /* Upper bits store count */ #define COOKIEMASK (((__u32)1 << COOKIEBITS) - 1) diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index c1bc074..6e61075 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -748,4 +748,4 @@ static __init int sysctl_ipv4_init(void) return 0; } -__initcall(sysctl_ipv4_init); +device_initcall(sysctl_ipv4_init); diff --git a/security/keys/proc.c b/security/keys/proc.c index 9d01021..9412316 100644 --- a/security/keys/proc.c +++ b/security/keys/proc.c @@ -83,7 +83,7 @@ static int __init key_proc_init(void) } /* end key_proc_init() */ -__initcall(key_proc_init); +device_initcall(key_proc_init); /*****************************************************************************/ /* diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c index 5feecb4..db02af5 100644 --- a/security/selinux/hooks.c +++ b/security/selinux/hooks.c @@ -5783,7 +5783,7 @@ out: return err; } -__initcall(selinux_nf_ip_init); +device_initcall(selinux_nf_ip_init); #ifdef CONFIG_SECURITY_SELINUX_DISABLE static void selinux_nf_ip_exit(void) diff --git a/security/selinux/netif.c b/security/selinux/netif.c index b4e14bc..9f89a94 100644 --- a/security/selinux/netif.c +++ b/security/selinux/netif.c @@ -315,5 +315,5 @@ static __init int sel_netif_init(void) return err; } -__initcall(sel_netif_init); +device_initcall(sel_netif_init); diff --git a/security/selinux/netlink.c b/security/selinux/netlink.c index 1ae5564..403b707 100644 --- a/security/selinux/netlink.c +++ b/security/selinux/netlink.c @@ -113,4 +113,4 @@ static int __init selnl_init(void) return 0; } -__initcall(selnl_init); +device_initcall(selnl_init); diff --git a/security/selinux/netnode.c b/security/selinux/netnode.c index 7100072..e7ae484 100644 --- a/security/selinux/netnode.c +++ b/security/selinux/netnode.c @@ -344,4 +344,4 @@ static __init int sel_netnode_init(void) return ret; } -__initcall(sel_netnode_init); +device_initcall(sel_netnode_init); diff --git a/security/selinux/netport.c b/security/selinux/netport.c index fe7fba6..94c29f4 100644 --- a/security/selinux/netport.c +++ b/security/selinux/netport.c @@ -278,4 +278,4 @@ static __init int sel_netport_init(void) return ret; } -__initcall(sel_netport_init); +device_initcall(sel_netport_init); diff --git a/security/selinux/selinuxfs.c b/security/selinux/selinuxfs.c index cd191bb..03eb7d8 100644 --- a/security/selinux/selinuxfs.c +++ b/security/selinux/selinuxfs.c @@ -1730,7 +1730,7 @@ static int __init init_sel_fs(void) return err; } -__initcall(init_sel_fs); +device_initcall(init_sel_fs); #ifdef CONFIG_SECURITY_SELINUX_DISABLE void exit_sel_fs(void) diff --git a/security/selinux/ss/services.c b/security/selinux/ss/services.c index cf27b3e..e7eb598 100644 --- a/security/selinux/ss/services.c +++ b/security/selinux/ss/services.c @@ -2964,7 +2964,7 @@ static int __init aurule_init(void) return err; } -__initcall(aurule_init); +device_initcall(aurule_init); #ifdef CONFIG_NETLABEL /** diff --git a/security/smack/smackfs.c b/security/smack/smackfs.c index aeead75..8faf43e 100644 --- a/security/smack/smackfs.c +++ b/security/smack/smackfs.c @@ -1366,4 +1366,4 @@ static int __init init_smk_fs(void) return err; } -__initcall(init_smk_fs); +device_initcall(init_smk_fs); diff --git a/sound/last.c b/sound/last.c index bdd0857..5efc26a 100644 --- a/sound/last.c +++ b/sound/last.c @@ -38,4 +38,4 @@ static int __init alsa_sound_last_init(void) return 0; } -__initcall(alsa_sound_last_init); +device_initcall(alsa_sound_last_init); -- 1.6.5.GIT From janr@adax.com Fri Mar 19 21:27:46 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Fri, 19 Mar 2010 21:27:49 +0100 (CET) Received: from mail1.adax.com ([208.201.231.104]:8954 "EHLO mail1.adax.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492562Ab0CSU1q (ORCPT ); Fri, 19 Mar 2010 21:27:46 +0100 Received: from static-151-204-189-187.pskn.east.verizon.net (static-151-204-189-187.pskn.east.verizon.net [151.204.189.187]) by mail1.adax.com (Postfix) with ESMTP id 1C97F1209A9; Fri, 19 Mar 2010 13:27:42 -0700 (PDT) Received: from localhost (localhost [127.0.0.1]) by static-151-204-189-187.pskn.east.verizon.net (Postfix) with ESMTP id 35FA3400571; Fri, 19 Mar 2010 16:27:41 -0400 (EDT) X-Virus-Scanned: amavisd-new at pskn.east.verizon.net Received: from static-151-204-189-187.pskn.east.verizon.net ([127.0.0.1]) by localhost (static-151-204-189-187.pskn.east.verizon.net [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id oV-iuyrlJkkD; Fri, 19 Mar 2010 16:27:09 -0400 (EDT) Received: from [192.168.1.76] (jr001327.mtl-nj.adax [192.168.1.76]) by static-151-204-189-187.pskn.east.verizon.net (Postfix) with ESMTP id 35DD4400589; Fri, 19 Mar 2010 16:27:09 -0400 (EDT) Message-ID: <4BA3DE2D.1070801@adax.com> Date: Fri, 19 Mar 2010 16:27:25 -0400 From: Jan Rovins User-Agent: Thunderbird 2.0.0.24 (Windows/20100228) MIME-Version: 1.0 To: Peter 'p2' De Schrijver CC: David Daney , linux-mips@linux-mips.org Subject: Re: linux 2.6.33 on movidis x16 References: <20100305141113.GD2437@apfelkorn> <4B9EB45D.8050106@adax.com> <20100318181734.GG2437@apfelkorn> <4BA27048.2010707@caviumnetworks.com> <4BA277FB.6080808@adax.com> <20100319175244.GH2437@apfelkorn> In-Reply-To: <20100319175244.GH2437@apfelkorn> Content-Type: multipart/mixed; boundary="------------010407080309010906040908" Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26276 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: janr@adax.com Precedence: bulk X-list: linux-mips This is a multi-part message in MIME format. --------------010407080309010906040908 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Peter 'p2' De Schrijver wrote: > Hi, > > >> You are right, it is called an x16. My memory is failing me, I should >> have looked at the machine first. >> >> > > Ok :) > > >> from dmesg: >> >> cavium-ethernet: Cavium Networks Octeon SDK version 1.8.1, build 294 >> Interface 0 has 4 ports (RGMII) >> Interface 1 has 4 ports (RGMII) >> Interface 2 has 4 ports (NPI) >> Uniform Multi-Platform E-IDE driver Revision: 7.00alpha2 >> >> >> looks like the PHYAD field has what you are loohing for, >> ethtool output: --------------------------- >> >> for i in 0 1 2 3 4 5 6 7; do ethtool eth$i; done >> Settings for eth0: >> Supported ports: [ TP MII ] >> Supported link modes: 10baseT/Half 10baseT/Full >> 100baseT/Half 100baseT/Full >> 1000baseT/Half 1000baseT/Full >> Supports auto-negotiation: Yes >> Advertised link modes: 10baseT/Half 10baseT/Full >> 100baseT/Half 100baseT/Full >> 1000baseT/Half 1000baseT/Full >> Advertised auto-negotiation: Yes >> Speed: 10Mb/s >> Duplex: Half >> Port: MII >> PHYAD: 1 >> Transceiver: internal >> Auto-negotiation: on >> Link detected: no >> Settings for eth1: >> Supported ports: [ TP MII ] >> Supported link modes: 10baseT/Half 10baseT/Full >> 100baseT/Half 100baseT/Full >> 1000baseT/Half 1000baseT/Full >> Supports auto-negotiation: Yes >> Advertised link modes: 10baseT/Half 10baseT/Full >> 100baseT/Half 100baseT/Full >> 1000baseT/Half 1000baseT/Full >> Advertised auto-negotiation: Yes >> Speed: 10Mb/s >> Duplex: Half >> Port: MII >> PHYAD: 2 >> Transceiver: internal >> Auto-negotiation: on >> Link detected: no >> Settings for eth2: >> Supported ports: [ TP MII ] >> Supported link modes: 10baseT/Half 10baseT/Full >> 100baseT/Half 100baseT/Full >> 1000baseT/Half 1000baseT/Full >> Supports auto-negotiation: Yes >> Advertised link modes: 10baseT/Half 10baseT/Full >> 100baseT/Half 100baseT/Full >> 1000baseT/Half 1000baseT/Full >> Advertised auto-negotiation: Yes >> Speed: 10Mb/s >> Duplex: Half >> Port: MII >> PHYAD: 3 >> Transceiver: internal >> Auto-negotiation: on >> Link detected: no >> Settings for eth3: >> Supported ports: [ TP MII ] >> Supported link modes: 10baseT/Half 10baseT/Full >> 100baseT/Half 100baseT/Full >> 1000baseT/Half 1000baseT/Full >> Supports auto-negotiation: Yes >> Advertised link modes: 10baseT/Half 10baseT/Full >> 100baseT/Half 100baseT/Full >> 1000baseT/Half 1000baseT/Full >> Advertised auto-negotiation: Yes >> Speed: 10Mb/s >> Duplex: Half >> Port: MII >> PHYAD: 4 >> Transceiver: internal >> Auto-negotiation: on >> Link detected: no >> Settings for eth4: >> Supported ports: [ TP MII ] >> Supported link modes: 10baseT/Half 10baseT/Full >> 100baseT/Half 100baseT/Full >> Supports auto-negotiation: Yes >> Advertised link modes: 10baseT/Half 10baseT/Full >> 100baseT/Half 100baseT/Full >> Advertised auto-negotiation: Yes >> Speed: 100Mb/s >> Duplex: Full >> Port: MII >> PHYAD: 0 >> Transceiver: internal >> Auto-negotiation: on >> Link detected: yes >> Settings for eth5: >> Supported ports: [ TP MII ] >> Supported link modes: 10baseT/Half 10baseT/Full >> 100baseT/Half 100baseT/Full >> Supports auto-negotiation: Yes >> Advertised link modes: 10baseT/Half 10baseT/Full >> 100baseT/Half 100baseT/Full >> Advertised auto-negotiation: Yes >> Speed: 100Mb/s >> Duplex: Full >> Port: MII >> PHYAD: 0 >> Transceiver: internal >> Auto-negotiation: on >> Link detected: yes >> Settings for eth6: >> Supported ports: [ TP MII ] >> Supported link modes: 10baseT/Half 10baseT/Full >> 100baseT/Half 100baseT/Full >> Supports auto-negotiation: Yes >> Advertised link modes: 10baseT/Half 10baseT/Full >> 100baseT/Half 100baseT/Full >> Advertised auto-negotiation: Yes >> Speed: 100Mb/s >> Duplex: Full >> Port: MII >> PHYAD: 0 >> Transceiver: internal >> Auto-negotiation: on >> Link detected: yes >> Settings for eth7: >> Supported ports: [ TP MII ] >> Supported link modes: 10baseT/Half 10baseT/Full >> 100baseT/Half 100baseT/Full >> Supports auto-negotiation: Yes >> Advertised link modes: 10baseT/Half 10baseT/Full >> 100baseT/Half 100baseT/Full >> Advertised auto-negotiation: Yes >> Speed: 100Mb/s >> Duplex: Full >> Port: MII >> PHYAD: 0 >> Transceiver: internal >> Auto-negotiation: on >> Link detected: yes >> >> > > PHYAD 0 for ports 4 - 7 seems a bit strange to me. Do you have the sources of the > kernel you're running on those machines ? If so, could you look at > int cvmx_helper_board_get_mii_address(int ipd_port) in cvmx-helper-board.c ? > It contains a large switch statement. case CVMX_BOARD_TYPE_CUST_WSX16: is what I'm > interested in. > > I have attached the source of cvmx_helper_board_get_mii_address(). In looking at it, it is all coming back to me now. I had to modify that function to get the Ethernet to work. The Movidis folk had a working kernel based on the CaviumNetworks 1.6.X or 1.7.X toolchain. I ported it up to the 1.8.1 toolchain, and had to add the CVMX_BOARD_TYPE_CUST_WSX16 entry to get the ethernet working, so the strangeness is my fault, but it got the first 4 Ehernets working. I thought I patterned it after what they did in their earlier kernels, but perhaps I never saw the Movidis mods, and just did it by trial & error. (it was a few years ago). If I have a chance I will dig around to see if I can find the source to the origonal 1.6.1 Movidis kernel that they provided. > >> ---------------------------------------------------------- >> You may want to re-map the interfaces so that they show up in a sane >> order, there is a cluster of 8 ethernet ports, 4 per row. Right now on >> our boxes , eth0 shows up on the bottom row, second from the right. >> > > Thanks, > > Peter. > --------------010407080309010906040908 Content-Type: text/plain; name="cvmx_helper_board_get_mii_address.txt" Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="cvmx_helper_board_get_mii_address.txt" /** * Return the MII PHY address associated with the given IPD * port. A result of -1 means there isn't a MII capable PHY * connected to this port. On chips supporting multiple MII * busses the bus number is encoded in bits <15:8>. * * This function must be modified for every new Octeon board. * Internally it uses switch statements based on the cvmx_sysinfo * data to determine board types and revisions. It replies on the * fact that every Octeon board receives a unique board type * enumeration from the bootloader. * * @param ipd_port Octeon IPD port to get the MII address for. * * @return MII PHY address and bus number or -1. */ int cvmx_helper_board_get_mii_address(int ipd_port) { switch (cvmx_sysinfo_get()->board_type) { case CVMX_BOARD_TYPE_SIM: /* Simulator doesn't have MII */ return -1; case CVMX_BOARD_TYPE_EBT3000: case CVMX_BOARD_TYPE_EBT5800: case CVMX_BOARD_TYPE_THUNDER: case CVMX_BOARD_TYPE_NICPRO2: /* Interface 0 is SPI4, interface 1 is RGMII */ if ((ipd_port >= 16) && (ipd_port < 20)) return ipd_port - 16; else return -1; case CVMX_BOARD_TYPE_KODAMA: case CVMX_BOARD_TYPE_EBH3100: case CVMX_BOARD_TYPE_HIKARI: case CVMX_BOARD_TYPE_CN3010_EVB_HS5: case CVMX_BOARD_TYPE_CN3005_EVB_HS5: case CVMX_BOARD_TYPE_CN3020_EVB_HS5: /* Port 0 is WAN connected to a PHY, Port 1 is GMII connected to a switch */ if (ipd_port == 0) return 4; else if (ipd_port == 1) return 9; else return -1; case CVMX_BOARD_TYPE_NAC38: /* Board has 8 RGMII ports PHYs are 0-7 */ if ((ipd_port >= 0) && (ipd_port < 4)) return ipd_port; else if ((ipd_port >= 16) && (ipd_port < 20)) return ipd_port - 16 + 4; else return -1; case CVMX_BOARD_TYPE_EBH3000: /* Board has dual SPI4 and no PHYs */ return -1; case CVMX_BOARD_TYPE_EBH5200: case CVMX_BOARD_TYPE_EBH5201: case CVMX_BOARD_TYPE_EBT5200: /* Board has 4 SGMII ports. The PHYs start right after the MII ports MII0 = 0, MII1 = 1, SGMII = 2-5 */ if ((ipd_port >= 0) && (ipd_port < 4)) return ipd_port+2; else return -1; case CVMX_BOARD_TYPE_EBH5600: case CVMX_BOARD_TYPE_EBH5601: case CVMX_BOARD_TYPE_KBP: /* JLR first Movidis X16 */ case CVMX_BOARD_TYPE_CUST_WSX16: /* JLR second Movidis X16 */ /* Board has 8 SGMII ports. 4 connect out, two connect to a switch, and 2 loop to each other */ if ((ipd_port >= 0) && (ipd_port < 4)) return ipd_port+1; else return -1; case CVMX_BOARD_TYPE_CUST_NB5: if (ipd_port == 2) return 4; else return -1; case CVMX_BOARD_TYPE_NIC_XLE_4G: /* Board has 4 SGMII ports. connected QLM3(interface 1) */ if ((ipd_port >= 16) && (ipd_port < 20)) return ipd_port - 16 + 1; else return -1; case CVMX_BOARD_TYPE_BBGW_REF: return -1; /* No PHYs are connected to Octeon, everything is through switch */ } /* Some unknown board. Somebody forgot to update this function... */ cvmx_dprintf("cvmx_helper_board_get_mii_address: Unknown board type %d\n", cvmx_sysinfo_get()->board_type); return -1; } --------------010407080309010906040908-- From David.Daney@caviumnetworks.com Fri Mar 19 23:04:15 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Fri, 19 Mar 2010 23:04:18 +0100 (CET) Received: from mail3.caviumnetworks.com ([12.108.191.235]:14479 "EHLO mail3.caviumnetworks.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491070Ab0CSWEP (ORCPT ); Fri, 19 Mar 2010 23:04:15 +0100 Received: from caexch01.caveonetworks.com (Not Verified[192.168.16.9]) by mail3.caviumnetworks.com with MailMarshal (v6,7,2,8378) id ; Fri, 19 Mar 2010 15:04:22 -0700 Received: from caexch01.caveonetworks.com ([192.168.16.9]) by caexch01.caveonetworks.com with Microsoft SMTPSVC(6.0.3790.3959); Fri, 19 Mar 2010 15:03:29 -0700 Received: from dd1.caveonetworks.com ([12.108.191.236]) by caexch01.caveonetworks.com over TLS secured channel with Microsoft SMTPSVC(6.0.3790.3959); Fri, 19 Mar 2010 15:03:29 -0700 Message-ID: <4BA3F4AC.5000307@caviumnetworks.com> Date: Fri, 19 Mar 2010 15:03:24 -0700 From: David Daney User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.8) Gecko/20100301 Fedora/3.0.3-1.fc12 Thunderbird/3.0.3 MIME-Version: 1.0 To: Jan Rovins CC: Peter 'p2' De Schrijver , linux-mips@linux-mips.org Subject: Re: linux 2.6.33 on movidis x16 References: <20100305141113.GD2437@apfelkorn> <4B9EB45D.8050106@adax.com> <20100318181734.GG2437@apfelkorn> <4BA27048.2010707@caviumnetworks.com> <4BA277FB.6080808@adax.com> <20100319175244.GH2437@apfelkorn> <4BA3DE2D.1070801@adax.com> In-Reply-To: <4BA3DE2D.1070801@adax.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-OriginalArrivalTime: 19 Mar 2010 22:03:29.0626 (UTC) FILETIME=[021013A0:01CAC7B0] Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26277 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: ddaney@caviumnetworks.com Precedence: bulk X-list: linux-mips On 03/19/2010 01:27 PM, Jan Rovins wrote: > case CVMX_BOARD_TYPE_KBP: /* JLR first Movidis X16 */ > case CVMX_BOARD_TYPE_CUST_WSX16: /* JLR second Movidis X16 */ > /* Board has 8 SGMII ports. 4 connect out, two connect to a switch, > and 2 loop to each other */ > if ((ipd_port >= 0) && (ipd_port < 4)) > return ipd_port+1; > else > return -1; I don't think that is correct. I just verified that that PHYs are on addresses 0-7. That code makes them start at 1. The Octeon mdio bus driver also needs a small change. I am working on a patch. David Daney From janr@adax.com Fri Mar 19 23:38:27 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Fri, 19 Mar 2010 23:38:30 +0100 (CET) Received: from mail1.adax.com ([208.201.231.104]:22426 "EHLO mail1.adax.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492508Ab0CSWi1 (ORCPT ); Fri, 19 Mar 2010 23:38:27 +0100 Received: from static-151-204-189-187.pskn.east.verizon.net (static-151-204-189-187.pskn.east.verizon.net [151.204.189.187]) by mail1.adax.com (Postfix) with ESMTP id 260EB120A4C; Fri, 19 Mar 2010 15:38:24 -0700 (PDT) Received: from localhost (localhost [127.0.0.1]) by static-151-204-189-187.pskn.east.verizon.net (Postfix) with ESMTP id 0580F400589; Fri, 19 Mar 2010 18:38:23 -0400 (EDT) X-Virus-Scanned: amavisd-new at pskn.east.verizon.net Received: from static-151-204-189-187.pskn.east.verizon.net ([127.0.0.1]) by localhost (static-151-204-189-187.pskn.east.verizon.net [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id o+aA-rMOoT3d; Fri, 19 Mar 2010 18:38:02 -0400 (EDT) Received: from [192.168.1.76] (jr001327.mtl-nj.adax [192.168.1.76]) by static-151-204-189-187.pskn.east.verizon.net (Postfix) with ESMTP id EBF29400571; Fri, 19 Mar 2010 18:38:01 -0400 (EDT) Message-ID: <4BA3FCD9.2020205@adax.com> Date: Fri, 19 Mar 2010 18:38:17 -0400 From: Jan Rovins User-Agent: Thunderbird 2.0.0.24 (Windows/20100228) MIME-Version: 1.0 To: David Daney CC: Peter 'p2' De Schrijver , linux-mips@linux-mips.org Subject: Re: linux 2.6.33 on movidis x16 References: <20100305141113.GD2437@apfelkorn> <4B9EB45D.8050106@adax.com> <20100318181734.GG2437@apfelkorn> <4BA27048.2010707@caviumnetworks.com> <4BA277FB.6080808@adax.com> <20100319175244.GH2437@apfelkorn> <4BA3DE2D.1070801@adax.com> <4BA3F4AC.5000307@caviumnetworks.com> In-Reply-To: <4BA3F4AC.5000307@caviumnetworks.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26278 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: janr@adax.com Precedence: bulk X-list: linux-mips David Daney wrote: > On 03/19/2010 01:27 PM, Jan Rovins wrote: > > >> case CVMX_BOARD_TYPE_KBP: /* JLR first Movidis X16 */ >> case CVMX_BOARD_TYPE_CUST_WSX16: /* JLR second Movidis X16 */ >> /* Board has 8 SGMII ports. 4 connect out, two connect to >> a switch, >> and 2 loop to each other */ >> if ((ipd_port >= 0) && (ipd_port < 4)) >> return ipd_port+1; >> else >> return -1; > > I don't think that is correct. I just verified that that PHYs are on > addresses 0-7. That code makes them start at 1. > > The Octeon mdio bus driver also needs a small change. > > I am working on a patch. You are most likely right. using ipd_port+1 is probably what made my eth0 show up on the 2nd RJ45 port. We were just using those boxes for early development, to verify basic Octeon functionality of some of our hardware drivers, so once an Ethernet port started working, I did not look into it any further, since it was a temporary setup just to get things going. I looked for the original source from Movidis, and all I found in their kernel source were broken symlinks to a tool chain they that they did not give me :-( Jan > > David Daney > From rdunlap@xenotime.net Fri Mar 19 23:56:31 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Fri, 19 Mar 2010 23:56:35 +0100 (CET) Received: from xenotime.net ([72.52.64.118]:35494 "HELO xenotime.net" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with SMTP id S1492282Ab0CSW4b (ORCPT ); Fri, 19 Mar 2010 23:56:31 +0100 Received: from chimera.site ([71.245.98.113]) by xenotime.net for ; Fri, 19 Mar 2010 15:56:22 -0700 Message-ID: <4BA40115.2000509@xenotime.net> Date: Fri, 19 Mar 2010 15:56:21 -0700 From: Randy Dunlap Organization: YPO4 User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.5) Gecko/20091209 Fedora/3.0-3.fc11 Thunderbird/3.0 MIME-Version: 1.0 To: "Justin P. Mattock" CC: trivial@kernel.org, linux-arm-kernel@lists.infradead.org, linux-audit@redhat.com, uclinux-dist-devel@blackfin.uclinux.org, linux-cris-kernel@axis.com, linux-ia64@vger.kernel.org, linux-mips@linux-mips.org, linux-parisc@vger.kernel.org, linux-s390@vger.kernel.org, selinux@tycho.nsa.gov, sparclinux@vger.kernel.org, x86@kernel.org, netdev@vger.kernel.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH] Cosmetic:Partially remove deprecated __initcall() and change to References: <1269028291-9103-1-git-send-email-justinmattock@gmail.com> In-Reply-To: <1269028291-9103-1-git-send-email-justinmattock@gmail.com> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26279 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: rdunlap@xenotime.net Precedence: bulk X-list: linux-mips On 03/19/10 12:51, Justin P. Mattock wrote: > After doing some things on a small issue, > I noticed through web surfing, that there were patches > submitted pertaining that __initcall is deprecated, > and device_initcall should be used. Where was this discussion? Do you have any pointers to it? I don't see any mention of __initcall being deprecated in Linus' mainline git tree, or in linux-next, or in mmotm. Where are those patches? > So as a change of subject(since what I was looking at > was frustrating me),I decided to grep the whole tree > and make the change(partially). > > Currently I'm running this patch on my system, kernel compiles > without any errors or warnings.(thought there would be a speed increase > but didn't see much(if any)). No, __initcall(x) is just a shorthand version of typing device_initcall(x). They do the same thing. > Biggest problem I have though is testing this on other hardware types > (I only have a macbook,and an iMac). > So please if you have the access to other arch/hardware types please > test. > > Now what I mean by partially is the __initcall function is still > there, so(if any) userspace apps/libs depend on this it's there > so they dont break and/or any other subsystem, that needs time > to make the changes. The only thing that might be affected is building out-of-tree drivers, but those are easy to fix. > Note: > the remaining files that still have __initcall in them are: > (according to grep) > > arch/um/include/shared/init.h > include/linux/init.h > scripts/checkpatch.pl > > either I or somebody else, can change this(although a bit > concerned about breaking things). > > Signed-off-by: Justin P. Mattock > --- > Documentation/DocBook/kernel-hacking.tmpl | 4 ++-- > Documentation/cpu-freq/cpu-drivers.txt | 2 +- > Documentation/kbuild/makefiles.txt | 2 +- > arch/arm/mach-at91/leds.c | 2 +- > arch/arm/mach-clps711x/p720t.c | 2 +- > arch/arm/mach-ebsa110/leds.c | 2 +- > arch/arm/mach-footbridge/cats-hw.c | 2 +- > arch/arm/mach-footbridge/ebsa285-leds.c | 2 +- > arch/arm/mach-footbridge/netwinder-hw.c | 2 +- > arch/arm/mach-footbridge/netwinder-leds.c | 2 +- > arch/arm/mach-ks8695/leds.c | 2 +- > arch/arm/mach-omap1/leds.c | 2 +- > arch/arm/mach-omap1/pm.c | 2 +- > arch/arm/mach-orion5x/db88f5281-setup.c | 2 +- > arch/arm/mach-orion5x/rd88f5182-setup.c | 2 +- > arch/arm/mach-pxa/generic.c | 2 +- > arch/arm/mach-pxa/pxa25x.c | 2 +- > arch/arm/mach-shark/leds.c | 2 +- > arch/blackfin/kernel/bfin_gpio.c | 2 +- > arch/blackfin/mach-common/pm.c | 2 +- > arch/cris/arch-v10/kernel/debugport.c | 2 +- > arch/cris/arch-v10/kernel/fasttimer.c | 2 +- > arch/cris/arch-v10/mm/init.c | 2 +- > arch/cris/arch-v32/kernel/fasttimer.c | 2 +- > arch/cris/arch-v32/kernel/pinmux.c | 2 +- > arch/cris/arch-v32/kernel/signal.c | 2 +- > arch/cris/arch-v32/mach-a3/io.c | 2 +- > arch/cris/arch-v32/mach-a3/pinmux.c | 2 +- > arch/cris/arch-v32/mach-fs/io.c | 2 +- > arch/cris/arch-v32/mach-fs/pinmux.c | 2 +- > arch/cris/kernel/profile.c | 2 +- > arch/cris/kernel/time.c | 2 +- > arch/cris/kernel/traps.c | 2 +- > arch/frv/kernel/gdb-stub.c | 2 +- > arch/frv/kernel/pm-mb93093.c | 2 +- > arch/frv/kernel/pm.c | 2 +- > arch/frv/kernel/sysctl.c | 2 +- > arch/h8300/kernel/gpio.c | 2 +- > arch/ia64/hp/sim/simeth.c | 2 +- > arch/ia64/hp/sim/simserial.c | 2 +- > arch/ia64/kernel/audit.c | 2 +- > arch/ia64/kernel/crash.c | 2 +- > arch/ia64/kernel/cyclone.c | 2 +- > arch/ia64/kernel/perfmon.c | 2 +- > arch/ia64/kernel/setup.c | 2 +- > arch/ia64/kernel/uncached.c | 2 +- > arch/ia64/kernel/unwind.c | 2 +- > arch/ia64/mm/init.c | 2 +- > arch/mips/Makefile | 2 +- > arch/mips/kernel/unaligned.c | 2 +- > arch/mips/lasat/sysctl.c | 2 +- > arch/mips/math-emu/cp1emu.c | 2 +- > arch/mips/nxp/pnx8550/common/proc.c | 2 +- > arch/mips/sibyte/sb1250/bus_watcher.c | 2 +- > arch/mn10300/kernel/gdb-stub.c | 2 +- > arch/mn10300/kernel/mn10300-serial.c | 2 +- > arch/mn10300/kernel/profile.c | 2 +- > arch/parisc/kernel/pci-dma.c | 2 +- > arch/parisc/kernel/pdc_chassis.c | 2 +- > arch/powerpc/kernel/audit.c | 2 +- > arch/powerpc/kernel/idle.c | 2 +- > arch/powerpc/kernel/irq.c | 2 +- > arch/powerpc/kernel/proc_powerpc.c | 2 +- > arch/powerpc/kernel/prom.c | 4 ++-- > arch/powerpc/kernel/rtas-proc.c | 2 +- > arch/powerpc/kernel/rtasd.c | 2 +- > arch/powerpc/kernel/sysfs.c | 2 +- > arch/powerpc/kernel/tau_6xx.c | 2 +- > arch/powerpc/kernel/vio.c | 2 +- > arch/powerpc/platforms/iseries/lpevents.c | 2 +- > arch/powerpc/platforms/iseries/mf.c | 2 +- > arch/powerpc/platforms/iseries/proc.c | 2 +- > arch/powerpc/platforms/iseries/viopath.c | 2 +- > arch/powerpc/platforms/pseries/eeh.c | 2 +- > arch/powerpc/platforms/pseries/hvCall_inst.c | 2 +- > arch/powerpc/platforms/pseries/power.c | 2 +- > arch/powerpc/platforms/pseries/ras.c | 2 +- > arch/powerpc/platforms/pseries/reconfig.c | 2 +- > arch/powerpc/xmon/xmon.c | 2 +- > arch/s390/appldata/appldata_base.c | 2 +- > arch/s390/kernel/audit.c | 2 +- > arch/s390/kernel/compat_exec_domain.c | 2 +- > arch/s390/kernel/ipl.c | 2 +- > arch/s390/kernel/topology.c | 2 +- > arch/sh/boards/board-edosk7760.c | 2 +- > arch/sh/boards/board-sh7785lcr.c | 2 +- > arch/sh/boards/mach-cayman/setup.c | 2 +- > arch/sh/boards/mach-landisk/setup.c | 2 +- > arch/sh/boards/mach-r2d/setup.c | 2 +- > arch/sh/boards/mach-sdk7786/setup.c | 2 +- > arch/sh/boards/mach-se/7206/setup.c | 2 +- > arch/sh/boards/mach-se/7751/setup.c | 2 +- > arch/sh/boards/mach-sh03/setup.c | 2 +- > arch/sh/kernel/traps_64.c | 2 +- > arch/sparc/kernel/apc.c | 2 +- > arch/sparc/kernel/audit.c | 2 +- > arch/sparc/kernel/mdesc.c | 2 +- > arch/sparc/kernel/pmc.c | 2 +- > arch/um/drivers/mconsole_kern.c | 8 ++++---- > arch/um/drivers/net_kern.c | 2 +- > arch/um/drivers/stderr_console.c | 2 +- > arch/um/drivers/ubd_kern.c | 4 ++-- > arch/um/kernel/exitcode.c | 2 +- > arch/um/kernel/physmem.c | 2 +- > arch/um/os-Linux/aio.c | 4 ++-- > arch/um/os-Linux/skas/mem.c | 2 +- > arch/um/os-Linux/skas/process.c | 2 +- > arch/um/os-Linux/umid.c | 2 +- > arch/um/sys-i386/tls.c | 2 +- > arch/x86/kernel/audit_64.c | 2 +- > arch/x86/kernel/tlb_uv.c | 4 ++-- > arch/x86/kernel/vsyscall_64.c | 4 ++-- > arch/x86/mm/dump_pagetables.c | 2 +- > arch/x86/vdso/vdso32-setup.c | 4 ++-- > arch/x86/vdso/vma.c | 2 +- > arch/xtensa/platforms/iss/console.c | 2 +- > drivers/net/arm/am79c961a.c | 2 +- > drivers/net/hamradio/baycom_epp.c | 1 + > drivers/net/hamradio/baycom_par.c | 1 + > drivers/net/hamradio/baycom_ser_fdx.c | 1 + > drivers/net/hamradio/baycom_ser_hdx.c | 1 + > drivers/s390/char/sclp_cmd.c | 2 +- > drivers/s390/char/sclp_config.c | 2 +- > drivers/s390/char/sclp_cpi_sys.c | 2 +- > drivers/s390/char/sclp_vt220.c | 2 +- > drivers/s390/cio/blacklist.c | 2 +- > drivers/staging/rtl8192u/ieee80211/api.c | 2 +- > fs/aio.c | 2 +- > fs/compat_ioctl.c | 2 +- > ipc/ipc_sysctl.c | 2 +- > ipc/mqueue.c | 2 +- > ipc/util.c | 2 +- > kernel/audit.c | 2 +- > kernel/audit_tree.c | 2 +- > kernel/dma.c | 2 +- > kernel/futex.c | 2 +- > kernel/lockdep_proc.c | 2 +- > kernel/pid_namespace.c | 2 +- > kernel/posix-cpu-timers.c | 2 +- > kernel/posix-timers.c | 2 +- > kernel/resource.c | 2 +- > kernel/sched_debug.c | 2 +- > kernel/time/timer_list.c | 2 +- > kernel/time/timer_stats.c | 2 +- > kernel/tracepoint.c | 2 +- > kernel/utsname_sysctl.c | 2 +- > lib/audit.c | 2 +- > lib/debugobjects.c | 2 +- > mm/bounce.c | 2 +- > mm/memory.c | 2 +- > mm/mm_init.c | 2 +- > mm/slab.c | 2 +- > mm/slub.c | 2 +- > mm/swapfile.c | 2 +- > net/ipv4/syncookies.c | 2 +- > net/ipv4/sysctl_net_ipv4.c | 2 +- > security/keys/proc.c | 2 +- > security/selinux/hooks.c | 2 +- > security/selinux/netif.c | 2 +- > security/selinux/netlink.c | 2 +- > security/selinux/netnode.c | 2 +- > security/selinux/netport.c | 2 +- > security/selinux/selinuxfs.c | 2 +- > security/selinux/ss/services.c | 2 +- > security/smack/smackfs.c | 2 +- > sound/last.c | 2 +- > 166 files changed, 176 insertions(+), 172 deletions(-) -- ~Randy From david.s.daney@gmail.com Sat Mar 20 00:00:55 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Sat, 20 Mar 2010 00:00:59 +0100 (CET) Received: from mail-pw0-f49.google.com ([209.85.160.49]:48702 "EHLO mail-pw0-f49.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492635Ab0CSXAz (ORCPT ); Sat, 20 Mar 2010 00:00:55 +0100 Received: by pwj2 with SMTP id 2so2619500pwj.36 for ; Fri, 19 Mar 2010 16:00:45 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:message-id:date:from :user-agent:mime-version:to:cc:subject:references:in-reply-to :content-type; bh=1tzDxYAfqAtLoMHwYFgcWq9dbfsZKUf2DWMvSgF5PgA=; b=hQwPJ/Qxt5gDFY37hkHo1YlBHvU0VkJGAPfXnk4B7Qi0/LnzZes2OKllJ+W1L865ia g+mLrGyAoCIA3Z1OHEJgDPDtEZtJaMBRtR5AxDGcVmsMAdgt12Comrz3czwqD2prdzaZ AWHu0DgW3VMa3Zz2yBihITh6ToDjThoY12hH0= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=message-id:date:from:user-agent:mime-version:to:cc:subject :references:in-reply-to:content-type; b=pDm70kyKr35R/G320D0rV1YlfB4M/1InCqfM2Iy3MR8gePY4mYeLVlXDuQ7InQ2x1i pYs7FO/pbi9Mxa0IBJeEdP+tu+jEIOPObsWOPcty2C+tUfVcVZQgLIyo1u+VJfEfC5Ig eE3wal9xCIvNG91jxQhvZJ31EbztwUkgdjsME= Received: by 10.141.53.15 with SMTP id f15mr4747847rvk.119.1269039645697; Fri, 19 Mar 2010 16:00:45 -0700 (PDT) Received: from dd_xps.caveonetworks.com ([12.108.191.226]) by mx.google.com with ESMTPS id 20sm1260736pzk.3.2010.03.19.16.00.44 (version=TLSv1/SSLv3 cipher=RC4-MD5); Fri, 19 Mar 2010 16:00:44 -0700 (PDT) Message-ID: <4BA4021B.4010905@gmail.com> Date: Fri, 19 Mar 2010 16:00:43 -0700 From: David Daney User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.8) Gecko/20100301 Fedora/3.0.3-1.fc11 Thunderbird/3.0.3 MIME-Version: 1.0 To: David Daney CC: Peter 'p2' De Schrijver , Jan Rovins , linux-mips@linux-mips.org Subject: Re: linux 2.6.33 on movidis x16 References: <20100305141113.GD2437@apfelkorn> <4B9EB45D.8050106@adax.com> <20100318181734.GG2437@apfelkorn> <4BA27048.2010707@caviumnetworks.com> In-Reply-To: <4BA27048.2010707@caviumnetworks.com> Content-Type: multipart/mixed; boundary="------------040609090508020401050503" Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26280 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: david.s.daney@gmail.com Precedence: bulk X-list: linux-mips This is a multi-part message in MIME format. --------------040609090508020401050503 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit On 03/18/2010 11:26 AM, David Daney wrote: > On 03/18/2010 11:17 AM, Peter 'p2' De Schrijver wrote: >> On 2010-03-15 18:27:41 (-0400), Jan Rovins wrote: >>> Peter 'p2' De Schrijver wrote: >>>> Hi, >>>> >>>> We are trying to get linux 2.6.33 to work on the movidis x16 board. >>>> Main thing >>>> missing is the addresses of the PHYs. Does anyone know those ? >>>> Unfortunately I don't have physical access to the board so I can't >>>> just look >>>> for the PHY components being used. >>>> >>>> Thanks, >>>> >>>> Peter. >>>> >>>> >>> Do you mean the Movidis x19 ? >> >> Maybe. AFAIK ours is called x16, and uses a cavium octeon CN3860 chip. >> >>> We have 2 of these in our lab. They are running the old 2.6.21.7 from >>> the CnUsers 1.8.1 tool chain. >>> They are currently being used by other developers for some application >>> testing, but I can grab the serial console boot-up messages and send >>> them to you, if the PHY info is in there. >> >> That would be helpful. Otherwise you could also try running ethtool >> on all the >> interfaces. That should give you the PHY address as well. >> > > I just found one of these. Current theory is that the PHY addresses > are 0-7 > > I don't think the PHY numbers are printed anywhere. Can you try the attached patch? from the back of the box we have: --------------------------------------- | eth5 | eth7 | | eth1 | eth3 | --------------------------------------- | eth4 | eth6 | | eth0 | eth2 | --------------------------------------- I was able to boot 2.6.34-rc1 to a Debian root nfs mounted via eth1. I didn't verify that eth4 - eth7 worked. David Daney --------------040609090508020401050503 Content-Type: text/plain; name="movidis.diff" Content-Transfer-Encoding: 7bit Content-Disposition: attachment; filename="movidis.diff" diff --git a/drivers/net/phy/mdio-octeon.c b/drivers/net/phy/mdio-octeon.c index 61a4461..dfaaf30 100644 --- a/drivers/net/phy/mdio-octeon.c +++ b/drivers/net/phy/mdio-octeon.c @@ -87,6 +87,7 @@ static int octeon_mdiobus_write(struct mii_bus *bus, int phy_id, static int __init octeon_mdiobus_probe(struct platform_device *pdev) { struct octeon_mdiobus *bus; + union cvmx_smix_en smi_en; int i; int err = -ENOENT; @@ -102,6 +103,10 @@ static int __init octeon_mdiobus_probe(struct platform_device *pdev) if (!bus->mii_bus) goto err; + smi_en.u64 = 0; + smi_en.s.en = 1; + cvmx_write_csr(CVMX_SMIX_EN(bus->unit), smi_en.u64); + /* * Standard Octeon evaluation boards don't support phy * interrupts, we need to poll. @@ -132,17 +137,22 @@ err_register: err: devm_kfree(&pdev->dev, bus); + smi_en.u64 = 0; + cvmx_write_csr(CVMX_SMIX_EN(bus->unit), smi_en.u64); return err; } static int __exit octeon_mdiobus_remove(struct platform_device *pdev) { struct octeon_mdiobus *bus; + union cvmx_smix_en smi_en; bus = dev_get_drvdata(&pdev->dev); mdiobus_unregister(bus->mii_bus); mdiobus_free(bus->mii_bus); + smi_en.u64 = 0; + cvmx_write_csr(CVMX_SMIX_EN(bus->unit), smi_en.u64); return 0; } diff --git a/drivers/staging/octeon/cvmx-helper-board.c b/drivers/staging/octeon/cvmx-helper-board.c index 3085e38..fb8d3ee 100644 --- a/drivers/staging/octeon/cvmx-helper-board.c +++ b/drivers/staging/octeon/cvmx-helper-board.c @@ -153,6 +153,13 @@ int cvmx_helper_board_get_mii_address(int ipd_port) * through switch. */ return -1; + case CVMX_BOARD_TYPE_CUST_WSX16: + if (ipd_port >= 0 && ipd_port <= 3) + return ipd_port; + else if (ipd_port >= 16 && ipd_port <= 19) + return ipd_port - 16 + 4; + else + return -1; } /* Some unknown board. Somebody forgot to update this function... */ --------------040609090508020401050503-- From justinmattock@gmail.com Sat Mar 20 00:24:33 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Sat, 20 Mar 2010 00:24:37 +0100 (CET) Received: from mail-gy0-f177.google.com ([209.85.160.177]:58465 "EHLO mail-gy0-f177.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492517Ab0CSXYd (ORCPT ); Sat, 20 Mar 2010 00:24:33 +0100 Received: by gyg10 with SMTP id 10so1822975gyg.36 for ; Fri, 19 Mar 2010 16:24:25 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:message-id:date:from :user-agent:mime-version:to:cc:subject:references:in-reply-to :content-type:content-transfer-encoding; bh=n9HgQ/Q1wl9/cIMVFDMNNyP7wJhGBkdkjmlagsmOJ1o=; b=vVQ5ZYZVXBwZNqF9nbHBheAHIdO3U29W+QZZKabtGa6gm5taeNMYwYjeq5ZpQ7rzm4 kDOv6sUgq+9SYro0WDLIhVh03HtBd/fCmttiYhUWZNpVEDDF6ydbVAG/kXzaJXjWUarZ xjIXwnmcmhDqZKhX2f040+aqoWvtJYCe8NeoI= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=message-id:date:from:user-agent:mime-version:to:cc:subject :references:in-reply-to:content-type:content-transfer-encoding; b=wlqq23bzBYV7qtOMFDzentuJYxNQJo1qTOU40tC2fjvxhK22V+7nBiiLCbbpGv0S3g KHwpM/4x1YGlU8XHxso0/4Q2EL4nAZXAD/Z6Hf3/3NwA9+c9e/QW24cmvOrTrhPryBYn 5i10pM3aWRlkoQ+G3jnRIHXo3LAUjYEOOMtrY= Received: by 10.151.25.21 with SMTP id c21mr66845ybj.169.1269041065750; Fri, 19 Mar 2010 16:24:25 -0700 (PDT) Received: from [10.0.0.92] (cpe-76-173-26-187.socal.res.rr.com [76.173.26.187]) by mx.google.com with ESMTPS id 8sm500153yxg.67.2010.03.19.16.24.23 (version=SSLv3 cipher=RC4-MD5); Fri, 19 Mar 2010 16:24:25 -0700 (PDT) Message-ID: <4BA407AC.5030506@gmail.com> Date: Fri, 19 Mar 2010 16:24:28 -0700 From: "Justin P. Mattock" User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.4pre) Gecko/20091114 Lightning/1.0pre Thunderbird/3.0b4 MIME-Version: 1.0 To: Randy Dunlap CC: trivial@kernel.org, linux-arm-kernel@lists.infradead.org, linux-audit@redhat.com, uclinux-dist-devel@blackfin.uclinux.org, linux-cris-kernel@axis.com, linux-ia64@vger.kernel.org, linux-mips@linux-mips.org, linux-parisc@vger.kernel.org, linux-s390@vger.kernel.org, selinux@tycho.nsa.gov, sparclinux@vger.kernel.org, x86@kernel.org, netdev@vger.kernel.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH] Cosmetic:Partially remove deprecated __initcall() and change to References: <1269028291-9103-1-git-send-email-justinmattock@gmail.com> <4BA40115.2000509@xenotime.net> In-Reply-To: <4BA40115.2000509@xenotime.net> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26281 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: justinmattock@gmail.com Precedence: bulk X-list: linux-mips On 03/19/2010 03:56 PM, Randy Dunlap wrote: > On 03/19/10 12:51, Justin P. Mattock wrote: >> After doing some things on a small issue, >> I noticed through web surfing, that there were patches >> submitted pertaining that __initcall is deprecated, >> and device_initcall should be used. > > Where was this discussion? Do you have any pointers to it? > The best info on this is scripts/checkpatch.pl line #2664 when I found this I just did a quick search of __initcall (which gives hits here and there) https://patchwork.kernel.org/patch/23344/ (also found others at around 2008 or so) > I don't see any mention of __initcall being deprecated in > Linus' mainline git tree, or in linux-next, or in mmotm. > Where are those patches? > I don't know(I'm out of the social pipeline when it comes to Linux news, and updates).. like in the explanation part of the patch I was looking into something else, then ran into this, so as a break(from what I was originally doing) decided to do this and submit. > >> So as a change of subject(since what I was looking at >> was frustrating me),I decided to grep the whole tree >> and make the change(partially). >> >> Currently I'm running this patch on my system, kernel compiles >> without any errors or warnings.(thought there would be a speed increase >> but didn't see much(if any)). > > No, __initcall(x) is just a shorthand version of typing > device_initcall(x). They do the same thing. > yep, that's what I found out as well(reason for the cosmetic in the subject line). >> Biggest problem I have though is testing this on other hardware types >> (I only have a macbook,and an iMac). >> So please if you have the access to other arch/hardware types please >> test. >> >> Now what I mean by partially is the __initcall function is still >> there, so(if any) userspace apps/libs depend on this it's there >> so they dont break and/or any other subsystem, that needs time >> to make the changes. > > The only thing that might be affected is building out-of-tree drivers, > but those are easy to fix. > alright..main concern is making sure things don't break in the kernel(even though things do). I can have a go at the header files, submit then if it's something people agree they want to do, then they can go from there(if not it's fine as well). >> Note: >> the remaining files that still have __initcall in them are: >> (according to grep) >> >> arch/um/include/shared/init.h >> include/linux/init.h >> scripts/checkpatch.pl >> >> either I or somebody else, can change this(although a bit >> concerned about breaking things). >> >> Signed-off-by: Justin P. Mattock >> --- >> Documentation/DocBook/kernel-hacking.tmpl | 4 ++-- >> Documentation/cpu-freq/cpu-drivers.txt | 2 +- >> Documentation/kbuild/makefiles.txt | 2 +- >> arch/arm/mach-at91/leds.c | 2 +- >> arch/arm/mach-clps711x/p720t.c | 2 +- >> arch/arm/mach-ebsa110/leds.c | 2 +- >> arch/arm/mach-footbridge/cats-hw.c | 2 +- >> arch/arm/mach-footbridge/ebsa285-leds.c | 2 +- >> arch/arm/mach-footbridge/netwinder-hw.c | 2 +- >> arch/arm/mach-footbridge/netwinder-leds.c | 2 +- >> arch/arm/mach-ks8695/leds.c | 2 +- >> arch/arm/mach-omap1/leds.c | 2 +- >> arch/arm/mach-omap1/pm.c | 2 +- >> arch/arm/mach-orion5x/db88f5281-setup.c | 2 +- >> arch/arm/mach-orion5x/rd88f5182-setup.c | 2 +- >> arch/arm/mach-pxa/generic.c | 2 +- >> arch/arm/mach-pxa/pxa25x.c | 2 +- >> arch/arm/mach-shark/leds.c | 2 +- >> arch/blackfin/kernel/bfin_gpio.c | 2 +- >> arch/blackfin/mach-common/pm.c | 2 +- >> arch/cris/arch-v10/kernel/debugport.c | 2 +- >> arch/cris/arch-v10/kernel/fasttimer.c | 2 +- >> arch/cris/arch-v10/mm/init.c | 2 +- >> arch/cris/arch-v32/kernel/fasttimer.c | 2 +- >> arch/cris/arch-v32/kernel/pinmux.c | 2 +- >> arch/cris/arch-v32/kernel/signal.c | 2 +- >> arch/cris/arch-v32/mach-a3/io.c | 2 +- >> arch/cris/arch-v32/mach-a3/pinmux.c | 2 +- >> arch/cris/arch-v32/mach-fs/io.c | 2 +- >> arch/cris/arch-v32/mach-fs/pinmux.c | 2 +- >> arch/cris/kernel/profile.c | 2 +- >> arch/cris/kernel/time.c | 2 +- >> arch/cris/kernel/traps.c | 2 +- >> arch/frv/kernel/gdb-stub.c | 2 +- >> arch/frv/kernel/pm-mb93093.c | 2 +- >> arch/frv/kernel/pm.c | 2 +- >> arch/frv/kernel/sysctl.c | 2 +- >> arch/h8300/kernel/gpio.c | 2 +- >> arch/ia64/hp/sim/simeth.c | 2 +- >> arch/ia64/hp/sim/simserial.c | 2 +- >> arch/ia64/kernel/audit.c | 2 +- >> arch/ia64/kernel/crash.c | 2 +- >> arch/ia64/kernel/cyclone.c | 2 +- >> arch/ia64/kernel/perfmon.c | 2 +- >> arch/ia64/kernel/setup.c | 2 +- >> arch/ia64/kernel/uncached.c | 2 +- >> arch/ia64/kernel/unwind.c | 2 +- >> arch/ia64/mm/init.c | 2 +- >> arch/mips/Makefile | 2 +- >> arch/mips/kernel/unaligned.c | 2 +- >> arch/mips/lasat/sysctl.c | 2 +- >> arch/mips/math-emu/cp1emu.c | 2 +- >> arch/mips/nxp/pnx8550/common/proc.c | 2 +- >> arch/mips/sibyte/sb1250/bus_watcher.c | 2 +- >> arch/mn10300/kernel/gdb-stub.c | 2 +- >> arch/mn10300/kernel/mn10300-serial.c | 2 +- >> arch/mn10300/kernel/profile.c | 2 +- >> arch/parisc/kernel/pci-dma.c | 2 +- >> arch/parisc/kernel/pdc_chassis.c | 2 +- >> arch/powerpc/kernel/audit.c | 2 +- >> arch/powerpc/kernel/idle.c | 2 +- >> arch/powerpc/kernel/irq.c | 2 +- >> arch/powerpc/kernel/proc_powerpc.c | 2 +- >> arch/powerpc/kernel/prom.c | 4 ++-- >> arch/powerpc/kernel/rtas-proc.c | 2 +- >> arch/powerpc/kernel/rtasd.c | 2 +- >> arch/powerpc/kernel/sysfs.c | 2 +- >> arch/powerpc/kernel/tau_6xx.c | 2 +- >> arch/powerpc/kernel/vio.c | 2 +- >> arch/powerpc/platforms/iseries/lpevents.c | 2 +- >> arch/powerpc/platforms/iseries/mf.c | 2 +- >> arch/powerpc/platforms/iseries/proc.c | 2 +- >> arch/powerpc/platforms/iseries/viopath.c | 2 +- >> arch/powerpc/platforms/pseries/eeh.c | 2 +- >> arch/powerpc/platforms/pseries/hvCall_inst.c | 2 +- >> arch/powerpc/platforms/pseries/power.c | 2 +- >> arch/powerpc/platforms/pseries/ras.c | 2 +- >> arch/powerpc/platforms/pseries/reconfig.c | 2 +- >> arch/powerpc/xmon/xmon.c | 2 +- >> arch/s390/appldata/appldata_base.c | 2 +- >> arch/s390/kernel/audit.c | 2 +- >> arch/s390/kernel/compat_exec_domain.c | 2 +- >> arch/s390/kernel/ipl.c | 2 +- >> arch/s390/kernel/topology.c | 2 +- >> arch/sh/boards/board-edosk7760.c | 2 +- >> arch/sh/boards/board-sh7785lcr.c | 2 +- >> arch/sh/boards/mach-cayman/setup.c | 2 +- >> arch/sh/boards/mach-landisk/setup.c | 2 +- >> arch/sh/boards/mach-r2d/setup.c | 2 +- >> arch/sh/boards/mach-sdk7786/setup.c | 2 +- >> arch/sh/boards/mach-se/7206/setup.c | 2 +- >> arch/sh/boards/mach-se/7751/setup.c | 2 +- >> arch/sh/boards/mach-sh03/setup.c | 2 +- >> arch/sh/kernel/traps_64.c | 2 +- >> arch/sparc/kernel/apc.c | 2 +- >> arch/sparc/kernel/audit.c | 2 +- >> arch/sparc/kernel/mdesc.c | 2 +- >> arch/sparc/kernel/pmc.c | 2 +- >> arch/um/drivers/mconsole_kern.c | 8 ++++---- >> arch/um/drivers/net_kern.c | 2 +- >> arch/um/drivers/stderr_console.c | 2 +- >> arch/um/drivers/ubd_kern.c | 4 ++-- >> arch/um/kernel/exitcode.c | 2 +- >> arch/um/kernel/physmem.c | 2 +- >> arch/um/os-Linux/aio.c | 4 ++-- >> arch/um/os-Linux/skas/mem.c | 2 +- >> arch/um/os-Linux/skas/process.c | 2 +- >> arch/um/os-Linux/umid.c | 2 +- >> arch/um/sys-i386/tls.c | 2 +- >> arch/x86/kernel/audit_64.c | 2 +- >> arch/x86/kernel/tlb_uv.c | 4 ++-- >> arch/x86/kernel/vsyscall_64.c | 4 ++-- >> arch/x86/mm/dump_pagetables.c | 2 +- >> arch/x86/vdso/vdso32-setup.c | 4 ++-- >> arch/x86/vdso/vma.c | 2 +- >> arch/xtensa/platforms/iss/console.c | 2 +- >> drivers/net/arm/am79c961a.c | 2 +- >> drivers/net/hamradio/baycom_epp.c | 1 + >> drivers/net/hamradio/baycom_par.c | 1 + >> drivers/net/hamradio/baycom_ser_fdx.c | 1 + >> drivers/net/hamradio/baycom_ser_hdx.c | 1 + >> drivers/s390/char/sclp_cmd.c | 2 +- >> drivers/s390/char/sclp_config.c | 2 +- >> drivers/s390/char/sclp_cpi_sys.c | 2 +- >> drivers/s390/char/sclp_vt220.c | 2 +- >> drivers/s390/cio/blacklist.c | 2 +- >> drivers/staging/rtl8192u/ieee80211/api.c | 2 +- >> fs/aio.c | 2 +- >> fs/compat_ioctl.c | 2 +- >> ipc/ipc_sysctl.c | 2 +- >> ipc/mqueue.c | 2 +- >> ipc/util.c | 2 +- >> kernel/audit.c | 2 +- >> kernel/audit_tree.c | 2 +- >> kernel/dma.c | 2 +- >> kernel/futex.c | 2 +- >> kernel/lockdep_proc.c | 2 +- >> kernel/pid_namespace.c | 2 +- >> kernel/posix-cpu-timers.c | 2 +- >> kernel/posix-timers.c | 2 +- >> kernel/resource.c | 2 +- >> kernel/sched_debug.c | 2 +- >> kernel/time/timer_list.c | 2 +- >> kernel/time/timer_stats.c | 2 +- >> kernel/tracepoint.c | 2 +- >> kernel/utsname_sysctl.c | 2 +- >> lib/audit.c | 2 +- >> lib/debugobjects.c | 2 +- >> mm/bounce.c | 2 +- >> mm/memory.c | 2 +- >> mm/mm_init.c | 2 +- >> mm/slab.c | 2 +- >> mm/slub.c | 2 +- >> mm/swapfile.c | 2 +- >> net/ipv4/syncookies.c | 2 +- >> net/ipv4/sysctl_net_ipv4.c | 2 +- >> security/keys/proc.c | 2 +- >> security/selinux/hooks.c | 2 +- >> security/selinux/netif.c | 2 +- >> security/selinux/netlink.c | 2 +- >> security/selinux/netnode.c | 2 +- >> security/selinux/netport.c | 2 +- >> security/selinux/selinuxfs.c | 2 +- >> security/selinux/ss/services.c | 2 +- >> security/smack/smackfs.c | 2 +- >> sound/last.c | 2 +- >> 166 files changed, 176 insertions(+), 172 deletions(-) > > From sebastian@breakpoint.cc Sun Mar 21 12:02:43 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Sun, 21 Mar 2010 12:02:46 +0100 (CET) Received: from Chamillionaire.breakpoint.cc ([85.10.199.196]:56209 "EHLO Chamillionaire.breakpoint.cc" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491204Ab0CULCn (ORCPT ); Sun, 21 Mar 2010 12:02:43 +0100 Received: id: bigeasy by Chamillionaire.breakpoint.cc with local (easymta 1.00 BETA 1) id 1NtIvZ-0006es-GP; Sun, 21 Mar 2010 12:02:41 +0100 Date: Sun, 21 Mar 2010 12:02:41 +0100 From: Sebastian Andrzej Siewior To: Ralf Baechle Cc: linux-mips@linux-mips.org Subject: [PATCH] mips/mm: fix module support on SiByte Message-ID: <20100321110241.GA25569@Chamillionaire.breakpoint.cc> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-15 Content-Disposition: inline X-Key-Id: FE3F4706 X-Key-Fingerprint: FFDA BBBB 3563 1B27 75C9 925B 98D5 5C1C FE3F 4706 User-Agent: Mutt/1.5.20 (2009-06-14) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26282 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: sebastian@breakpoint.cc Precedence: bulk X-list: linux-mips Since commit 656be92f aka "Load modules to CKSEG0 if CONFIG_BUILD_ELF64=n" module support is broken on 64bit. Since then modules arr loaded into 32bit compat adresses which are sign extended 64bit addresses. The SiByte war handler was not updated and those addresses were not recognized by the TLB hadling. This patch fixes this by shifting away the upper bits including the R and Fill bits. Now we compare VPN2 of C0_ENTRYHI against the matching bits at C0_BADVADDR. Cc: Signed-off-by: Sebastian Andrzej Siewior --- arch/mips/mm/tlbex.c | 8 ++++++++ 1 files changed, 8 insertions(+), 0 deletions(-) diff --git a/arch/mips/mm/tlbex.c b/arch/mips/mm/tlbex.c index badcf5e..47faeb4 100644 --- a/arch/mips/mm/tlbex.c +++ b/arch/mips/mm/tlbex.c @@ -745,6 +745,10 @@ static void __cpuinit build_r4000_tlb_refill_handler(void) UASM_i_MFC0(&p, K1, C0_ENTRYHI); uasm_i_xor(&p, K0, K0, K1); UASM_i_SRL(&p, K0, K0, PAGE_SHIFT + 1); +#ifdef CONFIG_64BIT + /* Make sure we have here just VPN2 */ + uasm_i_dsll32(&p, K0, K0, PAGE_SHIFT + 1 + 24 - 32); +#endif uasm_il_bnez(&p, &r, K0, label_leave); /* No need for uasm_i_nop */ } @@ -1264,6 +1268,10 @@ static void __cpuinit build_r4000_tlb_load_handler(void) UASM_i_MFC0(&p, K1, C0_ENTRYHI); uasm_i_xor(&p, K0, K0, K1); UASM_i_SRL(&p, K0, K0, PAGE_SHIFT + 1); +#ifdef CONFIG_64BIT + /* Make sure we have here just VPN2 */ + uasm_i_dsll32(&p, K0, K0, PAGE_SHIFT + 1 + 24 - 32); +#endif uasm_il_bnez(&p, &r, K0, label_leave); /* No need for uasm_i_nop */ } -- 1.6.6 From p2@psychaos.be Sun Mar 21 17:00:32 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Sun, 21 Mar 2010 17:00:35 +0100 (CET) Received: from apfelkorn.psychaos.be ([195.144.77.38]:39721 "EHLO apfelkorn.psychaos.be" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492705Ab0CUQAc (ORCPT ); Sun, 21 Mar 2010 17:00:32 +0100 Received: from p2 by apfelkorn.psychaos.be with local (Exim 4.69) (envelope-from ) id 1NtNZk-0000Ab-0J; Sun, 21 Mar 2010 18:00:28 +0200 Date: Sun, 21 Mar 2010 18:00:27 +0200 From: Peter 'p2' De Schrijver To: David Daney Cc: David Daney , Jan Rovins , linux-mips@linux-mips.org Subject: Re: linux 2.6.33 on movidis x16 Message-ID: <20100321160027.GJ2437@apfelkorn> References: <20100305141113.GD2437@apfelkorn> <4B9EB45D.8050106@adax.com> <20100318181734.GG2437@apfelkorn> <4BA27048.2010707@caviumnetworks.com> <4BA4021B.4010905@gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <4BA4021B.4010905@gmail.com> X-Unexpected-Header: The spanish inquisition ! X-mate: Mate, mann gewohnt sich an alles X-Paddo: Munch, Munch User-Agent: Mutt/1.5.18 (2008-05-17) Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26283 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: p2@debian.org Precedence: bulk X-list: linux-mips > > Can you try the attached patch? > > > from the back of the box we have: > > --------------------------------------- > | eth5 | eth7 | | eth1 | eth3 | > --------------------------------------- > | eth4 | eth6 | | eth0 | eth2 | > --------------------------------------- > > I was able to boot 2.6.34-rc1 to a Debian root nfs mounted via eth1. > > I didn't verify that eth4 - eth7 worked. > Yes ! That seems to work. We intend to use the machine as buildd, so 1 working ethernet is fine :) Thanks, Peter. From adebex@gmail.com Sun Mar 21 18:08:08 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Sun, 21 Mar 2010 18:08:15 +0100 (CET) Received: from mail-fx0-f216.google.com ([209.85.220.216]:43334 "EHLO mail-fx0-f216.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492428Ab0CURIH (ORCPT ); Sun, 21 Mar 2010 18:08:07 +0100 Received: by fxm8 with SMTP id 8so4000945fxm.25 for ; Sun, 21 Mar 2010 10:08:01 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:from:to:subject:date :user-agent:mime-version:content-type:content-transfer-encoding :message-id; bh=ABnITTmOU2e6D+SffjQ5RmkzU9xomC82KJLJXaD+gw8=; b=cPJ7R6FkTFb25EfOQN6txJ9GAsAmnv8xM3mUfaQp8zhOO2weoixgThPzrzDYNn4vet 58yDzSMRq6+RcXucxlGMekzpfMAV5W+D2gPgZAePlFcvF2QTCZ91ZLcqh8F25kN2kU0r ZLkMcQHRjpUZEzPRdiAHDT+QoHk3oGMVUo6xA= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=from:to:subject:date:user-agent:mime-version:content-type :content-transfer-encoding:message-id; b=JfVYWmMqWrt+8byAgScaO/McwFGCiR1dc2HN1+AC/4xGjUvtqqc6hDaN39etd1euWg l8FYVyr/O3Ol4pEANA1WHUtZKvFJKl2M2F2vHw94VRIc/TYFy+i+GCVwQbfe8t/gszrT NA1Kz3rNYavUIZDtasvbH+S9/A65CClX6+QIc= Received: by 10.103.7.31 with SMTP id k31mr9414507mui.132.1269191280861; Sun, 21 Mar 2010 10:08:00 -0700 (PDT) Received: from debian.localnet (chello087206211117.chello.pl [87.206.211.117]) by mx.google.com with ESMTPS id n7sm17969484mue.15.2010.03.21.10.07.59 (version=TLSv1/SSLv3 cipher=RC4-MD5); Sun, 21 Mar 2010 10:08:00 -0700 (PDT) From: Adrian Byszuk To: linux-mips@linux-mips.org Subject: [PATCH] Fix kexec call on MIPS platform Date: Sun, 21 Mar 2010 17:06:47 +0000 User-Agent: KMail/1.12.4 (Linux/2.6.33.1-custom; KDE/4.3.4; i686; ; ) MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit Message-Id: <201003211706.47373.adebex@gmail.com> Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26284 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: adebex@gmail.com Precedence: bulk X-list: linux-mips Dear developers, This kernel patch fixes problems with kexec call on some devices. I tested it on Asus WL-500gP v2. I suppose it would behave well on all MIPS machines. Applicable to 2.6.32 and 2.6.33 Signed-off-by: Adrian Byszuk --- --- a/arch/mips/kernel/machine_kexec.c 2010-03-15 15:52:04.000000000 +0000 +++ b/arch/mips/kernel/machine_kexec.c 2010-03-21 15:25:13.953615489 +0000 @@ -52,7 +52,8 @@ reboot_code_buffer = (unsigned long)page_address(image->control_code_page); - kexec_start_address = image->start; + kexec_start_address = + (unsigned long) phys_to_virt(image->start); kexec_indirection_page = (unsigned long) phys_to_virt(image->head & PAGE_MASK); From rdunlap@xenotime.net Mon Mar 22 01:34:06 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Mon, 22 Mar 2010 01:34:13 +0100 (CET) Received: from xenotime.net ([72.52.64.118]:58625 "HELO xenotime.net" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with SMTP id S1492498Ab0CVAeG (ORCPT ); Mon, 22 Mar 2010 01:34:06 +0100 Received: from chimera.site ([71.245.98.113]) by xenotime.net for ; Sun, 21 Mar 2010 17:33:56 -0700 Message-ID: <4BA6BAF3.10603@xenotime.net> Date: Sun, 21 Mar 2010 17:33:55 -0700 From: Randy Dunlap Organization: YPO4 User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.5) Gecko/20091209 Fedora/3.0-3.fc11 Thunderbird/3.0 MIME-Version: 1.0 To: "Justin P. Mattock" CC: trivial@kernel.org, linux-arm-kernel@lists.infradead.org, linux-audit@redhat.com, uclinux-dist-devel@blackfin.uclinux.org, linux-cris-kernel@axis.com, linux-ia64@vger.kernel.org, linux-mips@linux-mips.org, linux-parisc@vger.kernel.org, linux-s390@vger.kernel.org, selinux@tycho.nsa.gov, sparclinux@vger.kernel.org, x86@kernel.org, netdev@vger.kernel.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH] Cosmetic:Partially remove deprecated __initcall() and change to References: <1269028291-9103-1-git-send-email-justinmattock@gmail.com> <4BA40115.2000509@xenotime.net> <4BA407AC.5030506@gmail.com> In-Reply-To: <4BA407AC.5030506@gmail.com> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26285 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: rdunlap@xenotime.net Precedence: bulk X-list: linux-mips On 03/19/10 16:24, Justin P. Mattock wrote: > On 03/19/2010 03:56 PM, Randy Dunlap wrote: >> On 03/19/10 12:51, Justin P. Mattock wrote: >>> After doing some things on a small issue, >>> I noticed through web surfing, that there were patches >>> submitted pertaining that __initcall is deprecated, >>> and device_initcall should be used. >> >> Where was this discussion? Do you have any pointers to it? >> > > The best info on this is scripts/checkpatch.pl > line #2664 > > when I found this I just did a quick search of __initcall > (which gives hits here and there) > https://patchwork.kernel.org/patch/23344/ > (also found others at around 2008 or so) Thanks. IMO there should be something in the kernel source tree that says explicitly that __initcall is deprecated and should be replaced by using . That's missing. >> I don't see any mention of __initcall being deprecated in >> Linus' mainline git tree, or in linux-next, or in mmotm. >> Where are those patches? >> > > I don't know(I'm out of the social pipeline when it comes to Linux news, > and updates).. > like in the explanation part of the patch > I was looking into something else, then ran into this, > so as a break(from what I was originally doing) > decided to do this and submit. > >> >>> So as a change of subject(since what I was looking at >>> was frustrating me),I decided to grep the whole tree >>> and make the change(partially). >>> >>> Currently I'm running this patch on my system, kernel compiles >>> without any errors or warnings.(thought there would be a speed increase >>> but didn't see much(if any)). >> >> No, __initcall(x) is just a shorthand version of typing >> device_initcall(x). They do the same thing. >> > > yep, that's what I found out as well(reason for the cosmetic > in the subject line). > >>> Biggest problem I have though is testing this on other hardware types >>> (I only have a macbook,and an iMac). >>> So please if you have the access to other arch/hardware types please >>> test. >>> >>> Now what I mean by partially is the __initcall function is still >>> there, so(if any) userspace apps/libs depend on this it's there >>> so they dont break and/or any other subsystem, that needs time >>> to make the changes. >> >> The only thing that might be affected is building out-of-tree drivers, >> but those are easy to fix. >> > > alright..main concern is making sure things don't break in the > kernel(even though things do). > > I can have a go at the header files, submit > then if it's something people agree they want to do, then they > can go from there(if not it's fine as well). > >>> Note: >>> the remaining files that still have __initcall in them are: >>> (according to grep) >>> >>> arch/um/include/shared/init.h >>> include/linux/init.h >>> scripts/checkpatch.pl >>> >>> either I or somebody else, can change this(although a bit >>> concerned about breaking things). >>> >>> Signed-off-by: Justin P. Mattock >>> --- -- ~Randy From justinmattock@gmail.com Mon Mar 22 02:04:05 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Mon, 22 Mar 2010 02:04:08 +0100 (CET) Received: from mail-gy0-f177.google.com ([209.85.160.177]:43461 "EHLO mail-gy0-f177.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1492711Ab0CVBEF (ORCPT ); Mon, 22 Mar 2010 02:04:05 +0100 Received: by gyg10 with SMTP id 10so2548256gyg.36 for ; Sun, 21 Mar 2010 18:03:57 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:message-id:date:from :user-agent:mime-version:to:cc:subject:references:in-reply-to :content-type:content-transfer-encoding; bh=TCSZf5E5P183cdHnzF8ZAIoZzey6b/GHx5IIVby3KO0=; b=qZvN6bDuxyMTMCDzPuEWWMrs7I6AxCbUrquAR4HdOh1QI1eBj7IBLpw99bZ5g3G3LC KZnmIkrz4S5dwgN7L/6xchl7SIUxIjpmgRNlhei63VYHtWxyYOtvaDF3TqFIBqeZe1cd +F8pyYV40zZFwE3hV0/dexd91FJ3+UCr8JOWg= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=message-id:date:from:user-agent:mime-version:to:cc:subject :references:in-reply-to:content-type:content-transfer-encoding; b=lHmvoGfPDpjG7udp6eriKaMhsIOikHVoS5ryHr7O+4tZr+Zqu/noX7gT+n43Pdhdvh CcQgXWVsF85Oh2gvZsMtm2f8f9K92hQQYHr64VdcgxnZyWUj0ToPT9jAfxxUGq05cgtj Y+Z5i/Vs43b/+U8u+ssUji/bg81DvxMws7y20= Received: by 10.90.22.18 with SMTP id 18mr220143agv.77.1269219837553; Sun, 21 Mar 2010 18:03:57 -0700 (PDT) Received: from [10.0.0.96] (cpe-76-173-26-187.socal.res.rr.com [76.173.26.187]) by mx.google.com with ESMTPS id 34sm1506366yxf.54.2010.03.21.18.03.55 (version=SSLv3 cipher=RC4-MD5); Sun, 21 Mar 2010 18:03:56 -0700 (PDT) Message-ID: <4BA6C208.6010000@gmail.com> Date: Sun, 21 Mar 2010 18:04:08 -0700 From: "Justin P. mattock" User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.4pre) Gecko/20091114 Lightning/1.0pre Thunderbird/3.0b4 MIME-Version: 1.0 To: Randy Dunlap CC: trivial@kernel.org, linux-arm-kernel@lists.infradead.org, linux-audit@redhat.com, uclinux-dist-devel@blackfin.uclinux.org, linux-cris-kernel@axis.com, linux-ia64@vger.kernel.org, linux-mips@linux-mips.org, linux-parisc@vger.kernel.org, linux-s390@vger.kernel.org, selinux@tycho.nsa.gov, sparclinux@vger.kernel.org, x86@kernel.org, netdev@vger.kernel.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH] Cosmetic:Partially remove deprecated __initcall() and change to References: <1269028291-9103-1-git-send-email-justinmattock@gmail.com> <4BA40115.2000509@xenotime.net> <4BA407AC.5030506@gmail.com> <4BA6BAF3.10603@xenotime.net> In-Reply-To: <4BA6BAF3.10603@xenotime.net> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26286 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: justinmattock@gmail.com Precedence: bulk X-list: linux-mips On 03/21/2010 05:33 PM, Randy Dunlap wrote: > On 03/19/10 16:24, Justin P. Mattock wrote: >> On 03/19/2010 03:56 PM, Randy Dunlap wrote: >>> On 03/19/10 12:51, Justin P. Mattock wrote: >>>> After doing some things on a small issue, >>>> I noticed through web surfing, that there were patches >>>> submitted pertaining that __initcall is deprecated, >>>> and device_initcall should be used. >>> >>> Where was this discussion? Do you have any pointers to it? >>> >> >> The best info on this is scripts/checkpatch.pl >> line #2664 >> >> when I found this I just did a quick search of __initcall >> (which gives hits here and there) >> https://patchwork.kernel.org/patch/23344/ >> (also found others at around 2008 or so) > > Thanks. IMO there should be something in the kernel source tree > that says explicitly that __initcall is deprecated and should be > replaced by using. That's missing. > agree.. I should of searched the source tree for something that says deprecated and so forth before doing anything(the checkpatch.pl was something I noticed down the line but doesn't say deprecated(say's explicitly). > >>> I don't see any mention of __initcall being deprecated in >>> Linus' mainline git tree, or in linux-next, or in mmotm. >>> Where are those patches? >>> >> >> I don't know(I'm out of the social pipeline when it comes to Linux news, >> and updates).. >> like in the explanation part of the patch >> I was looking into something else, then ran into this, >> so as a break(from what I was originally doing) >> decided to do this and submit. >> >>> >>>> So as a change of subject(since what I was looking at >>>> was frustrating me),I decided to grep the whole tree >>>> and make the change(partially). >>>> >>>> Currently I'm running this patch on my system, kernel compiles >>>> without any errors or warnings.(thought there would be a speed increase >>>> but didn't see much(if any)). >>> >>> No, __initcall(x) is just a shorthand version of typing >>> device_initcall(x). They do the same thing. >>> >> >> yep, that's what I found out as well(reason for the cosmetic >> in the subject line). >> >>>> Biggest problem I have though is testing this on other hardware types >>>> (I only have a macbook,and an iMac). >>>> So please if you have the access to other arch/hardware types please >>>> test. >>>> >>>> Now what I mean by partially is the __initcall function is still >>>> there, so(if any) userspace apps/libs depend on this it's there >>>> so they dont break and/or any other subsystem, that needs time >>>> to make the changes. >>> >>> The only thing that might be affected is building out-of-tree drivers, >>> but those are easy to fix. >>> >> >> alright..main concern is making sure things don't break in the >> kernel(even though things do). >> >> I can have a go at the header files, submit >> then if it's something people agree they want to do, then they >> can go from there(if not it's fine as well). >> >>>> Note: >>>> the remaining files that still have __initcall in them are: >>>> (according to grep) >>>> >>>> arch/um/include/shared/init.h >>>> include/linux/init.h >>>> scripts/checkpatch.pl >>>> >>>> either I or somebody else, can change this(although a bit >>>> concerned about breaking things). >>>> >>>> Signed-off-by: Justin P. Mattock >>>> --- > > In any case I'll leave this to you guys to decide. The patch is in cyberspace now, so if/when this ever is decided it's there(patch), then can be used then. Justin P. Mattock From mlistz@gmail.com Mon Mar 22 08:01:30 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Mon, 22 Mar 2010 08:01:33 +0100 (CET) Received: from mail-qy0-f173.google.com ([209.85.221.173]:51040 "EHLO mail-qy0-f173.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491927Ab0CVHB3 (ORCPT ); Mon, 22 Mar 2010 08:01:29 +0100 Received: by qyk4 with SMTP id 4so1273219qyk.24 for ; Mon, 22 Mar 2010 00:01:22 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:date:message-id:subject :from:to:cc:content-type; bh=RoRqYXQCGk7KMCCAS7LTw5Uy3n9/MtZN3KIvWiMnEdA=; b=l5gzE00y6XJVfi+/sbJzqd0AN3A+dMZZ0NLbIjw6Z6NXvdHS4+VSfYdIW2jarRiBZM MRVEVo0AklSyBpdYAi7ykJfrowbH5FuaLRaNK2HnzgOz7t5nlSGqRBLQflawRZFJxpYL 407QMHgZleaM7yppiqHH6QV9c6pjYgki9jFgY= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:date:message-id:subject:from:to:cc:content-type; b=YKLwRsDmBmR8e3onpev3aKruRFYgacoADJDFLm+M3S1VmFXkC/Ck7moFj3JB08d9bJ IuCFpQ+4B80pomLnFhz2Et9nUcXt1kozz4vRRbVMNv7FGtkTidVNWBqQUlvL9DQwIuAj OX9NrNPJIbP406kEqPyarNwmznsnsdfFDbnHs= MIME-Version: 1.0 Received: by 10.229.190.133 with SMTP id di5mr211493qcb.23.1269241281957; Mon, 22 Mar 2010 00:01:21 -0700 (PDT) Date: Mon, 22 Mar 2010 15:01:21 +0800 Message-ID: Subject: [BUG?] cavium cn56xx and dma_map_single warning From: Zhuang Yuyao To: David Daney Cc: linux-mips@linux-mips.org Content-Type: text/plain; charset=ISO-8859-1 Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26287 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: mlistz@gmail.com Precedence: bulk X-list: linux-mips Hi, I've got a strange dma error on my cavium cn56xx board. ...... dma_map_single: Warning: Mapping memory address that might conflict with devices 0xfbb8f000-0xfbb8ffff dma_map_single: Warning: Mapping memory address that might conflict with devices 0xfb2ef000-0xfb2f0fff dma_map_single: Warning: Mapping memory address that might conflict with devices 0xfb3de000-0xfb3defff dma_map_single: Warning: Mapping memory address that might conflict with devices 0xfb3dd000-0xfb3ddfff dma_map_single: Warning: Mapping memory address that might conflict with devices 0xfac28000-0xfac29fff dma_map_single: Warning: Mapping memory address that might conflict with devices 0xfad1a000-0xfad1bfff ..... this error appeared when I plugged 4G or 8G ram on the board, if there is only 2G ram on the board, the error never happened. an adaptec 3405 pci-e raid card is plugged on the board, and a harddisk attached. the program i am running is a postgresql server with more than 20 million record, so do a count(*) will let postgresql consume almost all of the available memory. while there is 4G ram, the 'free' command reported that the free memory is below 18M. then the error kept printing on the console for a while, then kernel panic. the warning message is come from arch/mips/cavium-octeon/dma-octeon.c http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=blob;f=arch/mips/cavium-octeon/dma-octeon.c;h=be531ec1f2064b590b58dfe8b4db4f5534999bab;hb=HEAD i tried several raid card from different producer, including LSI and Adaptec. the same error always happened while there are more than or equal to 4G memory on the board. Would you please give some advice on this issue? any help will be appreciated. Thanks. Zhuang Yuyao From mlistz@gmail.com Mon Mar 22 10:31:50 2010 Received: with ECARTIS (v1.0.0; list linux-mips); Mon, 22 Mar 2010 10:31:53 +0100 (CET) Received: from qw-out-1920.google.com ([74.125.92.145]:10376 "EHLO qw-out-1920.google.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1491852Ab0CVJbt (ORCPT ); Mon, 22 Mar 2010 10:31:49 +0100 Received: by qw-out-1920.google.com with SMTP id 4so751135qwk.54 for ; Mon, 22 Mar 2010 02:31:48 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:in-reply-to:references :date:message-id:subject:from:to:cc:content-type; bh=spf8Mntd4dSkC+mNF4NOvaZfr9VQwe9kIiCm2+2I+Sg=; b=Y19N5f0XOdXe/hQxxLjadB84v1l1Kbb5NPm+t8juTU0IckQLJV+HmYFlboTtlR1718 o5t9rNKT/pjyHLnpL/C8AcST0pEY6jSTcOd5q/LkyXCEr1HEz3ePDGgZiX3a/n1sdh7c us+QszZ1ecOs98cDodeHhTtHN9MYIbalmRiOA= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; b=jMqh9CT95RjMzegNK1vcEXd15N8EMdF4w42RReQ/1lIXoienFIVuPW4pbCF4tq+CDM HcSvb1js6yhUBUE2Tg1OoSxew7R4iE5nNcob8Ag9nw1bsdKf6Q2uif/85XmdkzZg722w HSbvAfzQzSc1xZf5uS3SN7jXX9pIXjHUHDots= MIME-Version: 1.0 Received: by 10.229.112.2 with SMTP id u2mr2260202qcp.0.1269250308589; Mon, 22 Mar 2010 02:31:48 -0700 (PDT) In-Reply-To: References: Date: Mon, 22 Mar 2010 17:31:48 +0800 Message-ID: Subject: Re: [BUG?] cavium cn56xx and dma_map_single warning From: Zhuang Yuyao To: David Daney Cc: linux-mips@linux-mips.org Content-Type: text/plain; charset=ISO-8859-1 Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 26288 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: mlistz@gmail.com Precedence: bulk X-list: linux-mips Here are some additional information.I added a dump_stack() where the warning happened and changed pr_warning() to panic(). here is the output. I am using linux 2.6.32.9. Is it scsi drivers where i should look into, to make the scsi do not request for memory reserved by octeon? [] dump_stack+0x8/0x34 [] octeon_map_dma_mem+0x4dc/0x538 [] dma_map_sg+0xa0/0xe0 [] scsi_dma_map+0x40/0x50 [] aac_build_sgraw+0x44/0x1a0 [] aac_read_raw_io+0x98/0x120 [] aac_scsi_cmd+0xb48/0x14b8 [] aac_queuecommand+0x94/0xa8 [] scsi_dispatch_cmd+0x104/0x260 [] scsi_request_fn+0x374/0x430 [] generic_unplug_device+0x38/0x50 [] sync_page+0x4c/0x70 [] sync_page_killable+0x10/0x48 [] __wait_on_bit_lock+0xcc/0x158 [] __lock_page_killable+0x50/0x60 [] generic_file_aio_read+0x444/0x6c0 [] do_sync_read+0xbc/0x120 [] vfs_read+0xb4/0x178 [] SyS_read+0x48/0xa0 [] handle_sys64+0x44/0x60 Kernel panic - not syncing: dma_map_single: conflict 0xffc00000 On Mon, Mar 22, 2010 at 3:01 PM, Zhuang Yuyao wrote: > Hi, > > I've got a strange dma error on my cavium cn56xx board. > > ...... >