From pefo@enea.se Fri Sep 1 13:51:37 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id NAA02820 for ; Fri, 1 Sep 1995 13:51:36 -0700 Received: from outside.enea.se by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA10329; Fri, 1 Sep 1995 13:57:11 +0200 (MET) Received: by outside.enea.se (8.6.9/8.6.9) id NAA09821; Fri, 1 Sep 1995 13:58:50 +0200 Message-Id: <199509011158.NAA09821@outside.enea.se> Received: by outside via smap (V1.3) id sma009817; Fri Sep 1 13:58:49 1995 Received: by enea.se (4.1/SMI-4.0) id AA14909; Fri, 1 Sep 95 13:56:38 +0200 From: pefo@enea.se (Per Fogelstrom) Subject: Re: SIGEMT To: linux-mips@fnet.fr Date: Fri, 1 Sep 1995 13:56:37 +0200 (MET DST) Cc: andy@waldorf-gmbh.de (Andreas Busse) In-Reply-To: <199508311342.PAA14417@newton.waldorf-gmbh.de> from "Andreas Busse" at Aug 31, 95 03:42:07 pm X-Mailer: ELM [version 2.4 PL23] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 232 Lines: 9 > > > Hi all, > > I wonder why gdb thinks that signal 7 is SIGEMT. In Linux > this is SIGBUS. Any ideas, anyone ? > Because 7 is SIGEMT and have been for eons. (Read PDP11 times) SIGBUS is 10. At least on SYSV. And on BSD also. From ralf@waldorf-gmbh.de Fri Sep 1 20:26:19 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id UAA06960 for ; Fri, 1 Sep 1995 20:26:18 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA17556; Fri, 1 Sep 1995 20:34:48 +0200 (MET) From: Ralf Baechle Message-Id: <199509011834.UAA07682@scotty.waldorf-gmbh.de> Received: from localhost by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id UAA07682; Fri, 1 Sep 1995 20:34:36 +0200 Subject: Re: SIGEMT To: linux-mips@fnet.fr Date: Fri, 1 Sep 1995 20:34:36 +0200 (MET DST) In-Reply-To: <199509011158.NAA09821@outside.enea.se> from "Per Fogelstrom" at Sep 1, 95 01:56:37 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 559 Lines: 15 Hi all, > > I wonder why gdb thinks that signal 7 is SIGEMT. In Linux > > this is SIGBUS. Any ideas, anyone ? > > > Because 7 is SIGEMT and have been for eons. (Read PDP11 times) > SIGBUS is 10. At least on SYSV. And on BSD also. Relying on a fixed relation number <-> SYMBOLIC named isn't very portable. For Linux and SIGBUS I have to note that SIGBUS didn't exist for long times under Linux. Since the i386 doesn't generate a bus error exception you should expect the unexpected :-( See also signal(7) which only documents the i386 behaviour. Ralf From ralf@waldorf-gmbh.de Fri Sep 1 20:54:07 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id UAA07254 for ; Fri, 1 Sep 1995 20:54:07 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA17967; Fri, 1 Sep 1995 21:02:37 +0200 (MET) From: Ralf Baechle Message-Id: <199509011902.VAA08131@scotty.waldorf-gmbh.de> Received: from localhost by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id VAA08131; Fri, 1 Sep 1995 21:02:26 +0200 Subject: Re: kgdb and Linux/MIPS To: linux-mips@fnet.fr Date: Fri, 1 Sep 1995 21:02:25 +0200 (MET DST) In-Reply-To: <199508301836.UAA10759@newton.waldorf-gmbh.de> from "Andreas Busse" at Aug 30, 95 08:36:53 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1463 Lines: 41 Hi all, > I have some (limited) success with kgdb. At least, both machines > connect, the GDB on my Intel box happily talks to the Mips and > gets response. > So far, so good. The bad thing is that the usual exception handlers > of Linux/MIPS do not save all registers, nor in the order GDB > expects them. That leads of course to some mysterious behaviour :-) > > Question is: Would it hurt to change the struct pt_regs plus the > the macros SAVE_ALL, RESTORE_ALL and everything related so that > all required registers are saved, or shall I write my own > SAVE_ALL_GDB and RESTORE_ALL_GDB macros and a new struct containing > the regs ? > > GDB expects following regs in this order: > > $0 ... $31 ($0 isn't worth to store, I know :-)) > CP0_STATUS, > CP0_EPC, > LO, HI, > BAD_VADDR > > Most of them are availabe thru pt_regs. Only zero ($0), > k0 ($26) and k1 ($27) are missing. > > So we have a couple of choices: > a) change struct pt_regs and all associated macros and functions > b) fake the values of zero (always 0 anyway) and k0,k1 > c) change GDB > > I would favourize a), but b) would also work. Changing GDB > isn't a good idea, I believe. I intend to change struct pt_regs & SAVE_ALL/RESTORE_ALL anyway. They really save to much on the stack. It's probably the best when the remote debugging support in the kernel that you have to hack anyway fakes the stack order for GDB or so. Will have to dive into the internal of GDB ... Ralf From ralf@waldorf-gmbh.de Fri Sep 1 20:57:51 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id UAA07294 for ; Fri, 1 Sep 1995 20:57:50 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA18007; Fri, 1 Sep 1995 21:06:21 +0200 (MET) From: Ralf Baechle Message-Id: <199509011906.VAA08201@scotty.waldorf-gmbh.de> Received: from localhost by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id VAA08201; Fri, 1 Sep 1995 21:06:08 +0200 Subject: Re: kgdb and Linux/MIPS To: linux-mips@fnet.fr Date: Fri, 1 Sep 1995 21:06:07 +0200 (MET DST) In-Reply-To: <199508302016.OAA04854@rover.village.org> from "Warner Losh" at Aug 30, 95 02:16:36 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 551 Lines: 17 Hi, > : a) change struct pt_regs and all associated macros and functions > > I like this option the best. I like > > : b) fake the values of zero (always 0 anyway) and k0,k1 > > less, but if $0 and $27 and $26 never change, then it would be > acceptible. You need quite fancy routines if you want to save ALL general purpose registers. That works only if you use an additional register, eg. the "linked address register". Which again destroys that register. GDB wants to have all registers in a stackframe - but that's impossible ... Ralf From davem@caip.rutgers.edu Fri Sep 1 22:08:09 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id WAA08017 for ; Fri, 1 Sep 1995 22:08:08 -0700 Received: from caipfs.rutgers.edu by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA19309; Fri, 1 Sep 1995 22:16:38 +0200 (MET) Received: from huahaga.rutgers.edu (huahaga.rutgers.edu [128.6.155.53]) by caipfs.rutgers.edu (8.6.9+bestmx+oldruq+newsunq+grosshack/8.6.9) with ESMTP id QAA17752 for ; Fri, 1 Sep 1995 16:16:55 -0400 Received: (davem@localhost) by huahaga.rutgers.edu (8.6.9+bestmx+oldruq+newsunq+grosshack/8.6.9) id QAA27683; Fri, 1 Sep 1995 16:16:54 -0400 Date: Fri, 1 Sep 1995 16:16:54 -0400 Message-Id: <199509012016.QAA27683@huahaga.rutgers.edu> From: "David S. Miller" To: linux-mips@fnet.fr In-Reply-To: <199508311342.PAA14417@newton.waldorf-gmbh.de> (message from Andreas Busse on Thu, 31 Aug 1995 15:42:07 +0200) Subject: Grrr: SIGEMT Content-Length: 707 Lines: 17 Date: Thu, 31 Aug 1995 15:42:07 +0200 From: Andreas Busse I wonder why gdb thinks that signal 7 is SIGEMT. In Linux this is SIGBUS. Any ideas, anyone ? SIGEMT me thinks is a BSD thing, I have to provide it for SunOS compatibility if that means anything... And whats even more strange is that the NetBSD Sparc code sends this signal to a process if it gets a tagged addition overflow trap ??!?!?! Possibly Ultrix or IRIX or some other MIPS based Unix has this? As another point of reference my Sparc gdb stub happens to use SIGEMT for fp disabled traps so it seems as if SIGEMT is another one of the MAGIC Unix things ;) Later, David S. Miller davem@caip.rutgers.edu From imp@rover.village.org Sat Sep 2 01:33:07 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id BAA10088 for ; Sat, 2 Sep 1995 01:33:01 -0700 Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA23843; Sat, 2 Sep 1995 01:41:31 +0200 (MET) Received: from LOCALHOST (LOCALHOST [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id RAA22064 for ; Fri, 1 Sep 1995 17:41:41 -0600 Message-Id: <199509012341.RAA22064@rover.village.org> To: linux-mips@fnet.fr Subject: Hrumph! Date: Fri, 01 Sep 1995 17:41:41 -0600 From: Warner Losh Content-Length: 631 Lines: 14 As you may have guess from my subject, I've found why my machine doesn't print anything when linux boots. At least I think I have. Looking at drivers/char/console.c:con_init() tells me why. It will only use the boot_info record when ACER_PICA_61 or MIPS_MAGNUM_4000 are enabled and when booted on those machines. Otherwise it will do the default PC things. However, my video card isn't at SLOTSPACE + 0xb8000; like the kernel just blindly assumes :-(. It doesn't use the address in the boot_info variable, like maybe it should. So I thought I'd bounce it off the list, while I go off and try to fix this. Comments? Warner From ralf@waldorf-gmbh.de Sat Sep 2 02:20:40 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id CAA11896 for ; Sat, 2 Sep 1995 02:20:39 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA24350; Sat, 2 Sep 1995 02:29:10 +0200 (MET) From: Ralf Baechle Message-Id: <199509020029.CAA13843@scotty.waldorf-gmbh.de> Received: from localhost by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id CAA13843; Sat, 2 Sep 1995 02:29:07 +0200 Subject: Re: Hrumph! To: linux-mips@fnet.fr Date: Sat, 2 Sep 1995 02:29:06 +0200 (MET DST) In-Reply-To: <199509012341.RAA22064@rover.village.org> from "Warner Losh" at Sep 1, 95 05:41:41 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 910 Lines: 23 Hi, > As you may have guess from my subject, I've found why my machine > doesn't print anything when linux boots. At least I think I have. > > Looking at drivers/char/console.c:con_init() tells me why. It will > only use the boot_info record when ACER_PICA_61 or MIPS_MAGNUM_4000 > are enabled and when booted on those machines. Otherwise it will do > the default PC things. However, my video card isn't at SLOTSPACE + > 0xb8000; like the kernel just blindly assumes :-(. It doesn't use the > address in the boot_info variable, like maybe it should. > > So I thought I'd bounce it off the list, while I go off and try to fix > this. Comments? Sorry Warner, the default assumption with SLOTSPACE + 0xb8000 is just what the Wreckstation Tyne (Oh, sorry ... a freudian typo ... the faulty board I had just killed my nerves for too long) needs. Fixing it for your machine should be trivial. Ralf From ralf@waldorf-gmbh.de Sat Sep 2 02:23:07 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id CAA11936 for ; Sat, 2 Sep 1995 02:23:06 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA24393; Sat, 2 Sep 1995 02:31:38 +0200 (MET) From: Ralf Baechle Message-Id: <199509020031.CAA13904@scotty.waldorf-gmbh.de> Received: from localhost by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id CAA13904; Sat, 2 Sep 1995 02:31:36 +0200 Subject: Re: Grrr: SIGEMT To: linux-mips@fnet.fr Date: Sat, 2 Sep 1995 02:31:35 +0200 (MET DST) In-Reply-To: <199509012016.QAA27683@huahaga.rutgers.edu> from "David S. Miller" at Sep 1, 95 04:16:54 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 619 Lines: 14 Hi, > SIGEMT me thinks is a BSD thing, I have to provide it for SunOS > compatibility if that means anything... And whats even more strange > is that the NetBSD Sparc code sends this signal to a process if it > gets a tagged addition overflow trap ??!?!?! Possibly Ultrix or IRIX > or some other MIPS based Unix has this? As another point of reference > my Sparc gdb stub happens to use SIGEMT for fp disabled traps so it > seems as if SIGEMT is another one of the MAGIC Unix things ;) Hmpf... I think I'm gonna do some tests with Waldorf's RISC/os machines and take a look at the BSD 4.4 Lite sources. Ralf From imp@rover.village.org Sat Sep 2 02:27:22 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id CAA11996 for ; Sat, 2 Sep 1995 02:27:22 -0700 Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA24434; Sat, 2 Sep 1995 02:35:50 +0200 (MET) Received: from LOCALHOST (LOCALHOST [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id SAA25174 for ; Fri, 1 Sep 1995 18:36:06 -0600 Message-Id: <199509020036.SAA25174@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Hrumph! In-Reply-To: Your message of Sat, 02 Sep 1995 02:29:06 +0200 Date: Fri, 01 Sep 1995 18:36:05 -0600 From: Warner Losh Content-Length: 441 Lines: 10 : the default assumption with SLOTSPACE + 0xb8000 is just what the : Wreckstation Tyne (Oh, sorry ... a freudian typo ... the faulty board I : had just killed my nerves for too long) needs. Fixing it for your machine : should be trivial. It should be, but it is proving to be frustrating. I guess I've had a long week and am a little crispy and I should just go off and do something else. Crispy in general, not at Linux/MIPS. Warner From andy@waldorf-gmbh.de Sat Sep 2 10:32:02 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id KAA16895 for ; Sat, 2 Sep 1995 10:32:01 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA00345; Sat, 2 Sep 1995 10:40:32 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id KAA20787; Sat, 2 Sep 1995 10:41:35 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id KAA19137; Sat, 2 Sep 1995 10:40:51 +0200 Date: Sat, 2 Sep 1995 10:40:51 +0200 From: Andreas Busse Message-Id: <199509020840.KAA19137@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Re: It linked! In-Reply-To: <199508311919.NAA12985@rover.village.org> X-Mailer: [XMailTool v3.1.2b] Content-Length: 617 Lines: 21 > > OK. It linked! I just copied the tyne.S to rpc44.S, hacked the > Makefile and it built! Now, it is time to go off and boot it and see > if things work well or not. I don't think it will, but I can start > trying now :-) Congrats!!! > > After disabling a few things, I was able to get the kernel size below > 640K, so I'll be able to boot it w/o problems. Well, I'll be able to > load it into memory and it will not have truncation problems :-) > Interesting. No matter what I do, the Magnum code is never smaller than 700k. Certainly not when compiled with debugging info :-) Cheers, Andy From andy@waldorf-gmbh.de Sat Sep 2 10:42:31 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id KAA17012 for ; Sat, 2 Sep 1995 10:42:30 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA00468; Sat, 2 Sep 1995 10:51:02 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id KAA20976; Sat, 2 Sep 1995 10:52:05 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id KAA19170; Sat, 2 Sep 1995 10:51:21 +0200 Date: Sat, 2 Sep 1995 10:51:21 +0200 From: Andreas Busse Message-Id: <199509020851.KAA19170@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Re: Hrumph! In-Reply-To: <199509012341.RAA22064@rover.village.org> X-Mailer: [XMailTool v3.1.2b] Content-Length: 1127 Lines: 36 > > As you may have guess from my subject, I've found why my machine > doesn't print anything when linux boots. At least I think I have. > > Looking at drivers/char/console.c:con_init() tells me why. It will > only use the boot_info record when ACER_PICA_61 or MIPS_MAGNUM_4000 > are enabled and when booted on those machines. Otherwise it will do > the default PC things. However, my video card isn't at SLOTSPACE + > 0xb8000; like the kernel just blindly assumes :-(. It doesn't use the > address in the boot_info variable, like maybe it should. > > So I thought I'd bounce it off the list, while I go off and try to fix > this. Comments? > Sure. Do it the Magnum/PICA way: #ifdef CONFIG_DESKSTATION_RPC44 if (boot_info.machtype == MACH_DESKSTATION_RPC44) { ... set rpc44 video information here ... } else #endif You probably don't need more than this to make the console driver working. At least, nothing more was needed for the PICA. Also, make sure that you have a correct map0_table in kernel/head.S for the RPC44. Without a correct initial mapping nothing will work! Cheers, Andy From andy@waldorf-gmbh.de Sat Sep 2 10:46:25 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id KAA17061 for ; Sat, 2 Sep 1995 10:46:25 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA00506; Sat, 2 Sep 1995 10:54:56 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id KAA21060; Sat, 2 Sep 1995 10:55:57 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id KAA19185; Sat, 2 Sep 1995 10:55:13 +0200 Date: Sat, 2 Sep 1995 10:55:13 +0200 From: Andreas Busse Message-Id: <199509020855.KAA19185@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Re: Grrr: SIGEMT In-Reply-To: <199509020031.CAA13904@scotty.waldorf-gmbh.de> X-Mailer: [XMailTool v3.1.2b] Content-Length: 947 Lines: 25 > > Hi, > > > SIGEMT me thinks is a BSD thing, I have to provide it for SunOS > > compatibility if that means anything... And whats even more strange > > is that the NetBSD Sparc code sends this signal to a process if it > > gets a tagged addition overflow trap ??!?!?! Possibly Ultrix or IRIX > > or some other MIPS based Unix has this? As another point of reference > > my Sparc gdb stub happens to use SIGEMT for fp disabled traps so it > > seems as if SIGEMT is another one of the MAGIC Unix things ;) > > Hmpf... I think I'm gonna do some tests with Waldorf's RISC/os machines > and take a look at the BSD 4.4 Lite sources. > No matter why, I believe that this is a GDB bug. It should take care of include/signal.h, or needs a way to configure the signal meanings. But it's also the question why Linux assigns bus error to signal 7 while all others assign this to signal 10. Should be corrected, IMHO. Cheers, Andy From andy@waldorf-gmbh.de Sat Sep 2 10:54:58 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id KAA17150 for ; Sat, 2 Sep 1995 10:54:57 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA00641; Sat, 2 Sep 1995 11:03:28 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id LAA21210; Sat, 2 Sep 1995 11:04:30 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id LAA19218; Sat, 2 Sep 1995 11:03:46 +0200 Date: Sat, 2 Sep 1995 11:03:46 +0200 From: Andreas Busse Message-Id: <199509020903.LAA19218@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Re: Grrr: SIGEMT In-Reply-To: <199509020031.CAA13904@scotty.waldorf-gmbh.de> X-Mailer: [XMailTool v3.1.2b] Content-Length: 323 Lines: 11 > > Hmpf... I think I'm gonna do some tests with Waldorf's RISC/os machines > and take a look at the BSD 4.4 Lite sources. > Just done that. In all flavours of RiscOS (BSD43, SVR3 and SVR4) SIGEMT is 7 and SIGBUS is 10. So, Linux's signal mapping is *wrong*, or at least non-standard. We should change that. Andy From andy@waldorf-gmbh.de Sat Sep 2 11:03:34 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id LAA17259 for ; Sat, 2 Sep 1995 11:03:33 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA00766; Sat, 2 Sep 1995 11:12:05 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id LAA21349; Sat, 2 Sep 1995 11:13:08 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id LAA19237; Sat, 2 Sep 1995 11:12:23 +0200 Date: Sat, 2 Sep 1995 11:12:23 +0200 From: Andreas Busse Message-Id: <199509020912.LAA19237@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Re: Grrr: SIGEMT In-Reply-To: <199509020855.KAA19185@newton.waldorf-gmbh.de> X-Mailer: [XMailTool v3.1.2b] Content-Length: 839 Lines: 18 > > > > > SIGEMT me thinks is a BSD thing, I have to provide it for SunOS > > > compatibility if that means anything... And whats even more strange > > > is that the NetBSD Sparc code sends this signal to a process if it > > > gets a tagged addition overflow trap ??!?!?! Possibly Ultrix or IRIX > > > or some other MIPS based Unix has this? As another point of reference > > > my Sparc gdb stub happens to use SIGEMT for fp disabled traps so it > > > seems as if SIGEMT is another one of the MAGIC Unix things ;) > > Actually, that seems perfectly OK. SIGEMT is "emulation trap", that is, something needs to be emulated. In this case, the FPU :-) Interestingly Linux doesn't have SIGEMT, although there is a FPU emulator in the 386 kernel. Never took a look at it, so I don't know how that works. Cheers, Andy From andy@waldorf-gmbh.de Sat Sep 2 12:55:46 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id MAA18399 for ; Sat, 2 Sep 1995 12:55:45 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA01989; Sat, 2 Sep 1995 13:04:16 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id NAA22890; Sat, 2 Sep 1995 13:05:20 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id NAA19454; Sat, 2 Sep 1995 13:04:32 +0200 Date: Sat, 2 Sep 1995 13:04:32 +0200 From: Andreas Busse Message-Id: <199509021104.NAA19454@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Re: kgdb and Linux/MIPS In-Reply-To: <199509011902.VAA08131@scotty.waldorf-gmbh.de> X-Mailer: [XMailTool v3.1.2b] Content-Length: 545 Lines: 17 > > I intend to change struct pt_regs & SAVE_ALL/RESTORE_ALL anyway. > They really save to much on the stack. It's probably the best > when the remote debugging support in the kernel that you have to > hack anyway fakes the stack order for GDB or so. Will have to dive > into the internal of GDB ... > What exactly do you want to leave away? I cannot imagine a single register that could be left away... Anyway, if you plan to change pt_regs I really better write my own code to save the regs. Thanks for saving my time :-( Andy From tiv@vma.bme.hu Sat Sep 2 13:14:12 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id NAA18608 for ; Sat, 2 Sep 1995 13:14:11 -0700 Received: from nic.bme.hu by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA02259; Sat, 2 Sep 1995 13:22:41 +0200 (MET) Received: from euromath.vma.bme.hu (euromath.vma.bme.hu [152.66.83.1]) by nic.bme.hu (8.6.10/8.6.10) with ESMTP id NAA30604 for ; Sat, 2 Sep 1995 13:22:56 +0200 Received: (from tiv@localhost) by euromath.vma.bme.hu (8.6.10/8.6.9) id NAA19751 for linux-mips@fnet.fr; Sat, 2 Sep 1995 13:22:55 +0200 From: Szemethy Tivadar Message-Id: <199509021122.NAA19751@euromath.vma.bme.hu> Subject: binutils question To: linux-mips@fnet.fr Date: Sat, 2 Sep 1995 13:22:55 +0200 (MET DST) X-Mailer: ELM [version 2.4 PL22] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 677 Lines: 16 Hi all, Last night I tried to install binutils (with the patches)on a linux box (as cross-tools), and realized that the configuration doesn't support anything like mipsel-linux or similar (mipsllinux), only mips-elf. Is that true ? Is it entirely ELF now ? Anyway, I'm trying with mips-ecoff now. I just need COFF files to play with the pmax bootloader. I thought that mips-ecoff and mips-linuxaout would be the same, or at least very similar, but it looks like they aren't. Or am I completely wrong ? If I am, I speak executable files of type '407' where the .data and .text can be loaded at once (according to my netbsd sources) Someone plase enlighten me... thanks, tiv From andy@waldorf-gmbh.de Sat Sep 2 17:26:47 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id RAA21170 for ; Sat, 2 Sep 1995 17:26:45 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA04879; Sat, 2 Sep 1995 17:35:17 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id RAA26631; Sat, 2 Sep 1995 17:36:17 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id RAA19782; Sat, 2 Sep 1995 17:35:14 +0200 Date: Sat, 2 Sep 1995 17:35:14 +0200 From: Andreas Busse Message-Id: <199509021535.RAA19782@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Re: binutils question In-Reply-To: <199509021122.NAA19751@euromath.vma.bme.hu> X-Mailer: [XMailTool v3.1.2b] Content-Length: 1335 Lines: 32 > > Last night I tried to install binutils (with the patches)on a linux box > (as cross-tools), and realized that the configuration doesn't support anything > like mipsel-linux or similar (mipsllinux), only mips-elf. > Is that true ? Is it entirely ELF now ? Actually not. Last time when I compiled the crossdev stuff I had no problems with the configuration. What you need to patch binutils is binutils-2.5.2-2.diffs.gz, when I remember right. Then configure binutils with --host=i486-linux and --target=mipsel-linux. That should work right out of the box. > Anyway, I'm trying with mips-ecoff now. I just need COFF files to play > with the pmax bootloader. > I thought that mips-ecoff and mips-linuxaout would be the same, or at least > very similar, but it looks like they aren't. > Or am I completely wrong ? If I am, I speak executable files of type '407' > where the .data and .text can be loaded at once (according to my netbsd sources) > Someone plase enlighten me... > The binutils-2.5.2 with patch set -2 should support both a.out and ecoff, but a.out is the default. If you want ecoff output, you need to tell the linker about. Take a look at Milo, which is linked into an ecoff binary. You can also download the gcc and binutils binaries from ftp.waldorf-gmbh.de, if you like. Hope this helps, Andy From imp@rover.village.org Sat Sep 2 17:42:18 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id RAA21353 for ; Sat, 2 Sep 1995 17:42:17 -0700 Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA05058; Sat, 2 Sep 1995 17:50:42 +0200 (MET) Received: from LOCALHOST (LOCALHOST [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id JAA15944 for ; Sat, 2 Sep 1995 09:50:54 -0600 Message-Id: <199509021550.JAA15944@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Grrr: SIGEMT In-Reply-To: Your message of Sat, 02 Sep 1995 11:03:46 +0200 Date: Sat, 02 Sep 1995 09:50:53 -0600 From: Warner Losh Content-Length: 372 Lines: 12 : Just done that. In all flavours of RiscOS (BSD43, SVR3 and SVR4) : SIGEMT is 7 and SIGBUS is 10. So, Linux's signal mapping is *wrong*, : or at least non-standard. We should change that. Don't even get me started on this.... It is acceptible under POSIX. Posix just defines the names, but not the symbolic mappings. Not that I think it is right, mind you... Warner From imp@rover.village.org Sat Sep 2 17:43:22 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id RAA21374 for ; Sat, 2 Sep 1995 17:43:21 -0700 Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA05077; Sat, 2 Sep 1995 17:51:44 +0200 (MET) Received: from LOCALHOST (LOCALHOST [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id JAA15961 for ; Sat, 2 Sep 1995 09:51:55 -0600 Message-Id: <199509021551.JAA15961@rover.village.org> To: linux-mips@fnet.fr Subject: Re: binutils question In-Reply-To: Your message of Sat, 02 Sep 1995 13:22:55 +0200 Date: Sat, 02 Sep 1995 09:51:55 -0600 From: Warner Losh Content-Length: 142 Lines: 6 : like mipsel-linux or similar (mipsllinux), only mips-elf. You need the patches to binutils. They make mipsel-*linux a known type. Warner From imp@rover.village.org Sat Sep 2 17:47:36 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id RAA21423 for ; Sat, 2 Sep 1995 17:47:35 -0700 Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA05164; Sat, 2 Sep 1995 17:56:05 +0200 (MET) Received: from LOCALHOST (LOCALHOST [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id JAA15992 for ; Sat, 2 Sep 1995 09:56:20 -0600 Message-Id: <199509021556.JAA15992@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Hrumph! In-Reply-To: Your message of Sat, 02 Sep 1995 10:51:21 +0200 Date: Sat, 02 Sep 1995 09:56:19 -0600 From: Warner Losh Content-Length: 1134 Lines: 34 : Sure. Do it the Magnum/PICA way: : : #ifdef CONFIG_DESKSTATION_RPC44 : if (boot_info.machtype == MACH_DESKSTATION_RPC44) { : ... : set rpc44 video information here : ... : } else : #endif Yes. I did that, and the driver still isn't working, so I must have done something wrong. : Also, make sure that you have a correct map0_table in kernel/head.S : for the RPC44. Without a correct initial mapping nothing will work! That's the problem, I think. What is supposed to be mapped where? I've stumbled around a bit trying to find it all, but more keeps popping up. Also, when MILO boots, is it in translated (using virutal addresses) or untranslated mode? Does the MIPS even have an untranslated mode? I know that in MILO the addresses 0x8000000 and 0xa0000000 start the range of memory and the ISA bus is "overmapped" from 0xn00a0000 to 0xn00fffff. And that I/O ports start at 0xB0000000. Hmmm, just got mail from Ralf suggesting I use tlbdump to see where these guys are mapped and then put that into the map statement. Hmmm, I'll see if that works, but I thought I had this down :-(. More hacking to go. Warner From ralf@waldorf-gmbh.de Sun Sep 3 01:35:35 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id BAA26115 for ; Sun, 3 Sep 1995 01:35:34 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA11779; Sun, 3 Sep 1995 01:44:05 +0200 (MET) From: Ralf Baechle Message-Id: <199509022345.BAA01790@scotty.waldorf-gmbh.de> Received: from localhost by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id BAA01790; Sun, 3 Sep 1995 01:45:16 +0200 Subject: Re: binutils question To: linux-mips@fnet.fr Date: Sun, 3 Sep 1995 01:45:14 +0200 (MET DST) In-Reply-To: <199509021122.NAA19751@euromath.vma.bme.hu> from "Szemethy Tivadar" at Sep 2, 95 01:22:55 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 2327 Lines: 53 Hi all, > Last night I tried to install binutils (with the patches)on a linux box > (as cross-tools), and realized that the configuration doesn't support > anything like mipsel-linux or similar (mipsllinux), only mips-elf. You must have made something wrong. Something like the following should work for you, too: tar zxf /pub/gnu/binutils-2.5.2.tar.gz cd binutils-2.5.2 gzip -cd /pub/linux/mips/crossdev/src/binutils-2.5.2-6.diffs.gz | patch -p1 ./configure --prefix=/usr/local --host=sparcling-table-water \ -target=mipsel-linux make CC=gcc CFLAGS="-O2 -vomit-frame-pointer -keep-programmer-inline" \ LDFLAGS=-s make install Maybe you just omited the -p option from patch? You must ALWAYS give a -p option to patch or new files will be created in the current directory! > Is that true ? Is it entirely ELF now ? ELF is only a option that you get when configuring for mips{el}-linuxelf. mips{el}-linux is just the same, but with a.out format. Please note that all Linux ports have a special a.out flavour. The .text section starts not 4096 bytes (as usual) but only 1024 bytes from the begining of the file. The generic a.out configuration for mips will therefore NOT work. For the generation of ELF binaries the mips{el}-elf configuration is sufficient. The linux configurations have additional support for a.out and COFF (required for Milo, eg.). And - back to your question - the big ELF day is still waiting to come ... > Anyway, I'm trying with mips-ecoff now. I just need COFF files to play > with the pmax bootloader. > I thought that mips-ecoff and mips-linuxaout would be the same, or at least > very similar, but it looks like they aren't. > Or am I completely wrong ? If I am, I speak executable files of type '407' > where the .data and .text can be loaded at once (according to my netbsd sources) > Someone plase enlighten me... Both executable types support the 0407 (octal!) type. It mean impure executable and for a.out also possibly object file. Though a.out and (E)COFF are completly different formats. Just as a warning: when you use the crossbinutils' strip command to strip a impure (0407) COFF executable, a bug in strip will convert it into a demand paged (0413) COFF executable. This bug was quite nervragging when I still used the Tyne board. Ralf From ralf@waldorf-gmbh.de Sun Sep 3 01:37:14 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id BAA26147 for ; Sun, 3 Sep 1995 01:37:14 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA11807; Sun, 3 Sep 1995 01:45:44 +0200 (MET) From: Ralf Baechle Message-Id: <199509022346.BAA01810@scotty.waldorf-gmbh.de> Received: from localhost by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id BAA01810; Sun, 3 Sep 1995 01:46:54 +0200 Subject: Re: It linked! To: linux-mips@fnet.fr Date: Sun, 3 Sep 1995 01:46:53 +0200 (MET DST) In-Reply-To: <199509020840.KAA19137@newton.waldorf-gmbh.de> from "Andreas Busse" at Sep 2, 95 10:40:51 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 388 Lines: 12 Hi, > > After disabling a few things, I was able to get the kernel size below > > 640K, so I'll be able to boot it w/o problems. Well, I'll be able to > > load it into memory and it will not have truncation problems :-) > > Interesting. No matter what I do, the Magnum code is never smaller > than 700k. Certainly not when compiled with debugging info :-) Blame the font. Ralf From ralf@waldorf-gmbh.de Sun Sep 3 02:06:31 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id CAA27768 for ; Sun, 3 Sep 1995 02:06:30 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA12107; Sun, 3 Sep 1995 02:15:02 +0200 (MET) From: Ralf Baechle Message-Id: <199509030016.CAA02280@scotty.waldorf-gmbh.de> Received: from localhost by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id CAA02280; Sun, 3 Sep 1995 02:16:14 +0200 Subject: Re: Hrumph! To: linux-mips@fnet.fr Date: Sun, 3 Sep 1995 02:16:12 +0200 (MET DST) In-Reply-To: <199509021556.JAA15992@rover.village.org> from "Warner Losh" at Sep 2, 95 09:56:19 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 2608 Lines: 67 Hi, > Yes. I did that, and the driver still isn't working, so I must have > done something wrong. > > : Also, make sure that you have a correct map0_table in kernel/head.S > : for the RPC44. Without a correct initial mapping nothing will work! > > That's the problem, I think. What is supposed to be mapped where? > I've stumbled around a bit trying to find it all, but more keeps > popping up. The mappings for the Tyne should be almost right for your purposes. Just adjust the physical addresses for the rPC44. The mappings should be as follows: table entry virtual address -> physical address #1 0xe0000000 0x0, uncached #2 0xe2000000 0x10000000, uncached The ENTRYLO1 field of both mappings is unused. > Also, when MILO boots, is it in translated (using virtual addresses) > or untranslated mode? Does the MIPS even have an untranslated mode? Depending of the system Milo either uses KSEG0 or KSEG1. Both are mapped. Both KSEG0 and KSEG1 map a 512mb segment from 0x80000000/ 0xa0000000 to physical address 0x0. There is no real untranslated mode as on 68k when you switch off the address translation. > I know that in MILO the addresses 0x8000000 and 0xa0000000 start the > range of memory and the ISA bus is "overmapped" from 0xn00a0000 to > 0xn00fffff. And that I/O ports start at 0xB0000000. With a bit of luck this might be all you have to know about your machine's hardware! If your I/O ports start at 0xb0000000, which is physical address 0x10000000, you just need to map them to virtual address 0xe2000000 and port accesses using in/out macros should work. If you didn't do this right, the kernel probably crashed with an invalid (tlbl) exception when it first tried to use in/out. You should probably try to do something like lw t0,0xe2000061 xori t0,3 sw t0,0xe2000061 after setting up the wired mappings - just to verify that your mappings are right. This code tries to access port 0x61 which controls a certain buzzing sound device on normal PCs. Or even easier for you - post your map0 mapping table. > Hmmm, just got mail from Ralf suggesting I use tlbdump to see where > these guys are mapped and then put that into the map statement. Hmmm, > I'll see if that works, but I thought I had this down :-(. Well, I thought this more as a verification that your initial mappings have been installed right. We also used an additional primitive console driver for debugging output. It is being used in Milo's dummy vmlinux if I remember right and was quite usefull for initial debugging. Just in case ... Ralf From ralf@waldorf-gmbh.de Sun Sep 3 02:15:33 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id CAA27865 for ; Sun, 3 Sep 1995 02:15:32 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA12225; Sun, 3 Sep 1995 02:24:04 +0200 (MET) From: Ralf Baechle Message-Id: <199509030025.CAA02417@scotty.waldorf-gmbh.de> Received: from localhost by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id CAA02417; Sun, 3 Sep 1995 02:25:17 +0200 Subject: Re: Grrr: SIGEMT To: linux-mips@fnet.fr Date: Sun, 3 Sep 1995 02:25:16 +0200 (MET DST) In-Reply-To: <199509020903.LAA19218@newton.waldorf-gmbh.de> from "Andreas Busse" at Sep 2, 95 11:03:46 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 499 Lines: 13 Hi, > Just done that. In all flavours of RiscOS (BSD43, SVR3 and SVR4) > SIGEMT is 7 and SIGBUS is 10. So, Linux's signal mapping is *wrong*, > or at least non-standard. We should change that. Have fun with all the other packages that then will have to be changed, too. In my opinion it's GDB's fault anyway. If everything fails you might write a program that #includes and evaluates it's contents. Just like dozens of packages do. Yes, it's ugly but the right thing (TM). Ralf From ralf@waldorf-gmbh.de Sun Sep 3 02:20:46 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id CAA27925 for ; Sun, 3 Sep 1995 02:20:42 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA12263; Sun, 3 Sep 1995 02:29:13 +0200 (MET) From: Ralf Baechle Message-Id: <199509030030.CAA02493@scotty.waldorf-gmbh.de> Received: from localhost by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id CAA02493; Sun, 3 Sep 1995 02:30:24 +0200 Subject: Re: Grrr: SIGEMT To: linux-mips@fnet.fr Date: Sun, 3 Sep 1995 02:30:24 +0200 (MET DST) In-Reply-To: <199509021550.JAA15944@rover.village.org> from "Warner Losh" at Sep 2, 95 09:50:53 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 757 Lines: 19 > > : Just done that. In all flavours of RiscOS (BSD43, SVR3 and SVR4) > : SIGEMT is 7 and SIGBUS is 10. So, Linux's signal mapping is *wrong*, > : or at least non-standard. We should change that. > > Don't even get me started on this.... > > It is acceptible under POSIX. Posix just defines the names, but not > the symbolic mappings. > > Not that I think it is right, mind you... Oh well, the POSIX stuff is perfect as long as you only need to map signal names to signal numbers. Unfortunately they forgot the reverse relation. As I wrote before one has to write a program that examines how the system maps signal names to signal numbers. Not nice but portable. But UNIX has never been a nice OS - it has been reliable, user friedly ... Ralf From ralf@waldorf-gmbh.de Sun Sep 3 02:25:33 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id CAA27984 for ; Sun, 3 Sep 1995 02:25:32 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA12320; Sun, 3 Sep 1995 02:34:03 +0200 (MET) From: Ralf Baechle Message-Id: <199509030035.CAA02582@scotty.waldorf-gmbh.de> Received: from localhost by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id CAA02582; Sun, 3 Sep 1995 02:35:15 +0200 Subject: Re: kgdb and Linux/MIPS To: linux-mips@fnet.fr Date: Sun, 3 Sep 1995 02:35:15 +0200 (MET DST) In-Reply-To: <199509021104.NAA19454@newton.waldorf-gmbh.de> from "Andreas Busse" at Sep 2, 95 01:04:32 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 683 Lines: 17 Hi, > > I intend to change struct pt_regs & SAVE_ALL/RESTORE_ALL anyway. > > They really save to much on the stack. It's probably the best > > when the remote debugging support in the kernel that you have to > > hack anyway fakes the stack order for GDB or so. Will have to dive > > into the internal of GDB ... > > What exactly do you want to leave away? I cannot imagine a single > register that could be left away... All the s-registers are being saved by the C compiler. It's therefore useless work to save them on the stack. Only the switch_to() stuff really needs to save them. And switch_to() is MUCH less often being executed than SAVE_ALL/RESTORE_ALL. Ralf From andy@waldorf-gmbh.de Sun Sep 3 09:13:14 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id JAA02227 for ; Sun, 3 Sep 1995 09:13:14 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA17287; Sun, 3 Sep 1995 09:21:46 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id JAA08236; Sun, 3 Sep 1995 09:23:09 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id JAA21632; Sun, 3 Sep 1995 09:22:04 +0200 Date: Sun, 3 Sep 1995 09:22:04 +0200 From: Andreas Busse Message-Id: <199509030722.JAA21632@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Re: Hrumph! In-Reply-To: <199509021556.JAA15992@rover.village.org> X-Mailer: [XMailTool v3.1.2b] Content-Length: 1979 Lines: 57 > > : Sure. Do it the Magnum/PICA way: > : > : #ifdef CONFIG_DESKSTATION_RPC44 > : if (boot_info.machtype == MACH_DESKSTATION_RPC44) { > : ... > : set rpc44 video information here > : ... > : } else > : #endif > > Yes. I did that, and the driver still isn't working, so I must have > done something wrong. > But you already had the console driver working for the dummy vmlinux, right ? So you know where the video adapter lives, and that should actually be sufficient for know. > : Also, make sure that you have a correct map0_table in kernel/head.S > : for the RPC44. Without a correct initial mapping nothing will work! > > That's the problem, I think. What is supposed to be mapped where? > I've stumbled around a bit trying to find it all, but more keeps > popping up. > For now, it should be enough to provide an empty table. However, a mapping table must be there since the initialisation probably fails. > Also, when MILO boots, is it in translated (using virutal addresses) > or untranslated mode? Does the MIPS even have an untranslated mode? Milo runs totally untranslated in KSEG1. The BIOS itself may translate hardware addresses, though. > > I know that in MILO the addresses 0x8000000 and 0xa0000000 start the > range of memory and the ISA bus is "overmapped" from 0xn00a0000 to > 0xn00fffff. And that I/O ports start at 0xB0000000. > > Hmmm, just got mail from Ralf suggesting I use tlbdump to see where > these guys are mapped and then put that into the map statement. Hmmm, > I'll see if that works, but I thought I had this down :-(. > Well, if you still don't succeed, hardcode some video output in head.S. This is what I've done for the Magnum and the Acer to make sure that it gets thru the TLB initialization. Output numbers or letter to the video ram (using hardcoded addresses) after each code block in head.S. This way you can figure out where the problem is. Ugly, but works. Cheers, Andy From andy@waldorf-gmbh.de Sun Sep 3 09:20:53 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id JAA02306 for ; Sun, 3 Sep 1995 09:20:52 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA17354; Sun, 3 Sep 1995 09:29:25 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id JAA08328; Sun, 3 Sep 1995 09:30:48 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id JAA21651; Sun, 3 Sep 1995 09:29:44 +0200 Date: Sun, 3 Sep 1995 09:29:44 +0200 From: Andreas Busse Message-Id: <199509030729.JAA21651@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Re: It linked! In-Reply-To: <199509022346.BAA01810@scotty.waldorf-gmbh.de> X-Mailer: [XMailTool v3.1.2b] Content-Length: 222 Lines: 11 > > Interesting. No matter what I do, the Magnum code is never smaller > > than 700k. Certainly not when compiled with debugging info :-) > > Blame the font. > Oh, you're right... I've forgotten :-) Cheers, Andy From andy@waldorf-gmbh.de Sun Sep 3 09:27:07 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id JAA02384 for ; Sun, 3 Sep 1995 09:27:06 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA17458; Sun, 3 Sep 1995 09:35:38 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id JAA08421; Sun, 3 Sep 1995 09:37:01 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id JAA21664; Sun, 3 Sep 1995 09:35:56 +0200 Date: Sun, 3 Sep 1995 09:35:56 +0200 From: Andreas Busse Message-Id: <199509030735.JAA21664@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Re: Grrr: SIGEMT In-Reply-To: <199509030025.CAA02417@scotty.waldorf-gmbh.de> X-Mailer: [XMailTool v3.1.2b] Content-Length: 690 Lines: 21 > > Have fun with all the other packages that then will have to be changed, > too. In my opinion it's GDB's fault anyway. If everything fails you > might write a program that #includes and evaluates it's > contents. Just like dozens of packages do. Yes, it's ugly but the > right thing (TM). > It's a) GDB's fault, which I believe too *and* b) Linux's fault not to have SIGEMT and to reverse the locations of SIGBUS and SIGUSR1 (which should be SIGEMT). I don't mind to hack a workaround, but shouldn't we contact Linus regarding that? I'm interested to hear why Linux has mapped the signals exactly this way! Or do you know any possible reason ? Cheers, Andy From davem@caip.rutgers.edu Sun Sep 3 09:56:06 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id JAA02700 for ; Sun, 3 Sep 1995 09:56:05 -0700 Received: from caipfs.rutgers.edu by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA17772; Sun, 3 Sep 1995 10:04:37 +0200 (MET) Received: from huahaga.rutgers.edu (huahaga.rutgers.edu [128.6.155.53]) by caipfs.rutgers.edu (8.6.9+bestmx+oldruq+newsunq+grosshack/8.6.9) with ESMTP id EAA28442 for ; Sun, 3 Sep 1995 04:04:53 -0400 Received: (davem@localhost) by huahaga.rutgers.edu (8.6.9+bestmx+oldruq+newsunq+grosshack/8.6.9) id EAA04866; Sun, 3 Sep 1995 04:04:52 -0400 Date: Sun, 3 Sep 1995 04:04:52 -0400 Message-Id: <199509030804.EAA04866@huahaga.rutgers.edu> From: "David S. Miller" To: linux-mips@fnet.fr Cc: linux-mips@fnet.fr In-Reply-To: <199509030735.JAA21664@newton.waldorf-gmbh.de> (message from Andreas Busse on Sun, 3 Sep 1995 09:35:56 +0200) Subject: Re: SIGEMT Content-Length: 1015 Lines: 22 Date: Sun, 3 Sep 1995 09:35:56 +0200 From: Andreas Busse It's a) GDB's fault, which I believe too *and* b) Linux's fault not to have SIGEMT and to reverse the locations of SIGBUS and SIGUSR1 (which should be SIGEMT). I don't mind to hack a workaround, but shouldn't we contact Linus regarding that? I'm interested to hear why Linux has mapped the signals exactly this way! Or do you know any possible reason ? Ummm... I don't know if you guys have noticed, but in the 1.3.x series there is an linux/include/asm-arch/signal.h file now... also for ioctl.h errno.h sockopt.h shmparam.h fcntl.h... you guys are missing out on a lot of very clean solutions to the 'compatability thing' by not syncing up with the 1.3.x trees, at least recently. SunOS binary compatability is a snap with the way things are set up now. So the answer to your question is that Linux-Mips can map the signals any damn way it feels like it ;-) Later, David S. Miller davem@caip.rutgers.edu From andy@waldorf-gmbh.de Sun Sep 3 11:09:30 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id LAA03438 for ; Sun, 3 Sep 1995 11:09:29 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA18469; Sun, 3 Sep 1995 11:18:02 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id LAA09777; Sun, 3 Sep 1995 11:19:28 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id LAA21841; Sun, 3 Sep 1995 11:18:20 +0200 Date: Sun, 3 Sep 1995 11:18:20 +0200 From: Andreas Busse Message-Id: <199509030918.LAA21841@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Re: SIGEMT In-Reply-To: <199509030804.EAA04866@huahaga.rutgers.edu> X-Mailer: [XMailTool v3.1.2b] Content-Length: 1149 Lines: 27 > > Ummm... I don't know if you guys have noticed, but in the 1.3.x series > there is an linux/include/asm-arch/signal.h file now... also for > ioctl.h errno.h sockopt.h shmparam.h fcntl.h... you guys are missing > out on a lot of very clean solutions to the 'compatability thing' by > not syncing up with the 1.3.x trees, at least recently. SunOS binary > compatability is a snap with the way things are set up now. So the > answer to your question is that Linux-Mips can map the signals any > damn way it feels like it ;-) > Sounds good! But the reason to stay with 1.2.x was that we wanted to have a reliable source base until we've finished things like memory management, scheduling etc. I believe the time to switch to 1.3.x comes soon, but no yet. I don't know how others think about, but I'd like to see user code running on all current targets before we switch over and don't know if bugs are our own or if they have been introduced from the 1.3.x source tree in general. So back to signal.h: I'll change SIGBUS to 10 and declare SIGEMT as signal 7. Agreed, or is there anything else I should take care about? Cheers, Andy From imp@rover.village.org Sun Sep 3 18:43:55 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id SAA08037 for ; Sun, 3 Sep 1995 18:43:50 -0700 Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA23586; Sun, 3 Sep 1995 18:52:20 +0200 (MET) Received: from LOCALHOST (LOCALHOST [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id KAA18392 for ; Sun, 3 Sep 1995 10:52:36 -0600 Message-Id: <199509031652.KAA18392@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Hrumph! In-Reply-To: Your message of Sun, 03 Sep 1995 09:22:04 +0200 Date: Sun, 03 Sep 1995 10:52:36 -0600 From: Warner Losh Content-Length: 742 Lines: 18 : But you already had the console driver working for the dummy vmlinux, : right ? So you know where the video adapter lives, and that should : actually be sufficient for know. Yes. I thought so too, but I've make a slipup somewhere. I'm looking into it right now.... : Well, if you still don't succeed, hardcode some video output : in head.S. This is what I've done for the Magnum and the Acer : to make sure that it gets thru the TLB initialization. Output : numbers or letter to the video ram (using hardcoded addresses) : after each code block in head.S. This way you can figure out : where the problem is. Ugly, but works. I had thought of doing either that, or making the LEDs on the keyboard do a Sun-3/50-like cylon too. Warner From andy@waldorf-gmbh.de Sun Sep 3 19:03:43 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id TAA08250 for ; Sun, 3 Sep 1995 19:03:42 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA23833; Sun, 3 Sep 1995 19:12:02 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id TAA03104; Sun, 3 Sep 1995 19:13:19 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id TAA00450; Sun, 3 Sep 1995 19:12:13 +0200 Date: Sun, 3 Sep 1995 19:12:13 +0200 From: Andreas Busse Message-Id: <199509031712.TAA00450@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Re: Hrumph! In-Reply-To: <199509031652.KAA18392@rover.village.org> X-Mailer: [XMailTool v3.1.2b] Content-Length: 464 Lines: 17 > > I had thought of doing either that, or making the LEDs on the keyboard > do a Sun-3/50-like cylon too. > Also nice :-) Anyway, if you aren't sure what works and what not, comment out everything in init/main.c that isn't essential. No drivers except console, no IRQs, no nothing. Perhaps even paging_init() etc. Well, guess you got it. Strip the kernel down to nearly nothing, and then try again to enable features again, step by step. Cheers, Andy From imp@rover.village.org Sun Sep 3 22:39:39 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id WAA10417 for ; Sun, 3 Sep 1995 22:39:38 -0700 Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA25972; Sun, 3 Sep 1995 22:48:05 +0200 (MET) Received: from LOCALHOST (LOCALHOST [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id OAA20268 for ; Sun, 3 Sep 1995 14:48:19 -0600 Message-Id: <199509032048.OAA20268@rover.village.org> To: linux-mips@fnet.fr Subject: Non-verbose boot? Date: Sun, 03 Sep 1995 14:48:18 -0600 From: Warner Losh Content-Length: 341 Lines: 7 Is anybody interested in some patches to milo that do a non-verbose boot? That is, it looks more like the NT boot loader. Prints a . for each sector loaded, etc? I've been hacking on MILO to streamline the process of booting (since I figured out how to make LINUX the default boot), and thought I'd ask if people were interested. Warner From imp@rover.village.org Sun Sep 3 22:43:45 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id WAA10468 for ; Sun, 3 Sep 1995 22:43:45 -0700 Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA26044; Sun, 3 Sep 1995 22:52:16 +0200 (MET) Received: from LOCALHOST (LOCALHOST [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id OAA20554 for ; Sun, 3 Sep 1995 14:52:34 -0600 Message-Id: <199509032052.OAA20554@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Hrumph! In-Reply-To: Your message of Sun, 03 Sep 1995 19:12:13 +0200 Date: Sun, 03 Sep 1995 14:52:32 -0600 From: Warner Losh Content-Length: 1426 Lines: 38 : Anyway, if you aren't sure what works and what not, comment out : everything in init/main.c that isn't essential. No drivers except : console, no IRQs, no nothing. Perhaps even paging_init() etc. : Well, guess you got it. Strip the kernel down to nearly nothing, : and then try again to enable features again, step by step. Well, I have the main code that looks like: *(char *) 0xA00A0000 = '1'; setup_arch(&command_line, &memory_start, &memory_end); *(char *) 0xA00A0000 = '2'; memory_start = paging_init(memory_start,memory_end); *(char *) 0xA00A0000 = '3'; trap_init(); *(char *) 0xA00A0000 = '4'; init_IRQ(); *(char *) 0xA00A0000 = '5'; sched_init(); *(char *) 0xA00A0000 = '6'; parse_options(command_line); *(char *) 0xA00A0000 = '7'; init_modules(); *(char *) 0xA00A0000 = '8'; Which is a *BIG* hammer (but much easier than the cylon[*] light). I've found that setup_arch never returns.... Just thought I'd share this technique with the list. I'd bet most people know about it that are actively working on Linux, but maybe some of the lurkers might not. I get to '1', but I never see any other numbers. Either setup_arch is unmapping the ununmappable, or I'm on to something :-). Anyway, this gives me a big clue what is going on, so I'll keep working on it. Just wanted to let people BTW, I also wanted to thank everyone that has been sending me tips. They are most helpful and useful. Warner From imp@rover.village.org Sun Sep 3 23:35:13 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id XAA11015 for ; Sun, 3 Sep 1995 23:35:12 -0700 Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA26592; Sun, 3 Sep 1995 23:43:42 +0200 (MET) Received: from LOCALHOST (LOCALHOST [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id PAA21104 for ; Sun, 3 Sep 1995 15:44:00 -0600 Message-Id: <199509032144.PAA21104@rover.village.org> To: linux-mips@fnet.fr Subject: More progress Date: Sun, 03 Sep 1995 15:43:59 -0600 From: Warner Losh Content-Length: 688 Lines: 20 OK. Problem #1: panic() doesn't work until things are farther along in the boot process :-(. Linux has been panicing all over the place, but I can't see the panic message :-(. Problem #2: The config name is DESKSTATION_RPC44, but in some places it was _RPC (note, no 44). Fixed these. Problem #3: The IRQ stuff is panicing because there is no case for RPC44. I'm fixing it as well. etc. Anyway, I've hacked PANIC to print stuff when things aren't too smart yet. Again, another tip: If you don't get any output, then maybe Linux is panicing before it has enough brains to print error messages. Maybe we should use the BIOS routines if no video address is yet set???? Warner From ralf@waldorf-gmbh.de Mon Sep 4 01:10:23 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id BAA11968 for ; Mon, 4 Sep 1995 01:10:22 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA29933; Mon, 4 Sep 1995 01:18:54 +0200 (MET) From: Ralf Baechle Message-Id: <199509032320.BAA09226@scotty.waldorf-gmbh.de> Received: from localhost by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id BAA09226; Mon, 4 Sep 1995 01:20:41 +0200 Subject: Re: More progress To: linux-mips@fnet.fr Date: Mon, 4 Sep 1995 01:20:40 +0200 (MET DST) In-Reply-To: <199509032144.PAA21104@rover.village.org> from "Warner Losh" at Sep 3, 95 03:43:59 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1528 Lines: 42 Hi, > OK. Problem #1: > panic() doesn't work until things are farther along in the > boot process :-(. Linux has been panicing all over the place, > but I can't see the panic message :-(. Known problem; therefore I suggested to use the tiny console code from Milo. Luckily I didn't have to worry about that kind of problems for looong times. You'll soon reach that point, too - won't you? > Problem #2: > The config name is DESKSTATION_RPC44, but in some places it > was _RPC (note, no 44). Fixed these. Sorry, I just tried to provide some hooks for the rPC44 - and failed :-( > Problem #3: > The IRQ stuff is panicing because there is no case for RPC44. > I'm fixing it as well. > etc. Probably a copy of the Tyne stuff should work for you. > Anyway, I've hacked PANIC to print stuff when things aren't too smart > yet. > > Again, another tip: If you don't get any output, then maybe Linux is > panicing before it has enough brains to print error messages. Maybe > we should use the BIOS routines if no video address is yet set???? The problem is that the ARC BIOS enters the "program state" when Milo calls the launch() subroutine. This also means that calling the BIOS has no defined result anymore. In fact at that time the kernel is where important BIOS variables are supposed to be, eg. You'd have to hack your own hardware level print routines. The Linux kernel once had these but I removed them ... I'll look through my collection of old source code snipets and send you what I find. Ralf From ralf@waldorf-gmbh.de Mon Sep 4 01:19:59 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id BAA12081 for ; Mon, 4 Sep 1995 01:19:58 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA00160; Mon, 4 Sep 1995 01:28:31 +0200 (MET) From: Ralf Baechle Message-Id: <199509032330.BAA09379@scotty.waldorf-gmbh.de> Received: from localhost by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id BAA09379; Mon, 4 Sep 1995 01:30:18 +0200 Subject: Re: Non-verbose boot? To: linux-mips@fnet.fr Date: Mon, 4 Sep 1995 01:30:18 +0200 (MET DST) In-Reply-To: <199509032048.OAA20268@rover.village.org> from "Warner Losh" at Sep 3, 95 02:48:18 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 597 Lines: 14 Hi, > Is anybody interested in some patches to milo that do a non-verbose > boot? That is, it looks more like the NT boot loader. Prints a . for > each sector loaded, etc? I've been hacking on MILO to streamline the > process of booting (since I figured out how to make LINUX the default > boot), and thought I'd ask if people were interested. Well, I don't believe that Linux should necessarily try to look like NT. The current Milo is hardwired to verbose mode as debugging aid. Since it didn't make trouble for quite some time we should switch it to non-verbose mode as you say. Ralf From imp@rover.village.org Mon Sep 4 06:25:49 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id GAA16475 for ; Mon, 4 Sep 1995 06:25:47 -0700 Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA04011; Mon, 4 Sep 1995 06:34:10 +0200 (MET) Received: from LOCALHOST (LOCALHOST [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id WAA22270 for ; Sun, 3 Sep 1995 22:34:28 -0600 Message-Id: <199509040434.WAA22270@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Non-verbose boot? In-Reply-To: Your message of Mon, 04 Sep 1995 01:30:18 +0200 Date: Sun, 03 Sep 1995 22:34:28 -0600 From: Warner Losh Content-Length: 964 Lines: 20 : Well, I don't believe that Linux should necessarily try to look like NT. Nor I. However, Milo is currently a bit chatty for its function. As more and more of the kernel is working, the relative importance of MILO is diminishing, so its output should also. Don't get me wrong, MILO is a fine work of software, but I'm not convinced it needs to blow its horn quite so loudly :-). Most other boot loaders are very quiet most of the time (lilo, FreeBSD's, Boot Easy, etc). : The current Milo is hardwired to verbose mode as debugging aid. : Since it didn't make trouble for quite some time we should switch it : to non-verbose mode as you say. OK. I'll put the patches together and send them off when I get a bootable Deskstation rPC 44 (at least one that prints the first kernel greeting, and can position the cursor :-). I'm to kernel point '4' now, and hope to be at '5' within the hour and to '8' by the end of this long (for us Yanks) weekend. Warner From imp@rover.village.org Mon Sep 4 06:48:07 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id GAA16711 for ; Mon, 4 Sep 1995 06:48:06 -0700 Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA04213; Mon, 4 Sep 1995 06:56:36 +0200 (MET) Received: from LOCALHOST (LOCALHOST [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id WAA22474 for ; Sun, 3 Sep 1995 22:56:54 -0600 Message-Id: <199509040456.WAA22474@rover.village.org> To: linux-mips@fnet.fr Subject: Life is good :-) Date: Sun, 03 Sep 1995 22:56:52 -0600 From: Warner Losh Content-Length: 1662 Lines: 51 OK. I rebuilt the kernel -DPORT_BASE=0xB0000000 (since the memory map isn't completely working yet), hacked it here and there, and have some good news to report: I have "dim" output! As you likely know, this means that the kernel is printing things, and MILO isn't :-). For those of you that care, the output says: Console: colour CGA 80x50, 1 virtual console (max 63) Life is good [*] bios32 is done [*] Calibrarting delay loop.. ok 50.08 bogo mips keyboard unrecognized scan code (70) ignored tty00 at 0x3f8 (irq = 4) is a 16450 tty00 at 0x2f8 (irq = 3) is a 16450 Floppy drive(s): fd0 is 1.44M Unable to handle kernel paging request at virtual address 600003f2 Opps: 00000001 Status: 00011003 Cause: 0000830c Int: 0 Process swapper (pid: 0 stackpage=80070dbb) kfree of non-kmalloced memory: 80072d80, next = 00000000, order = 0 task[0] (swapper) killed: unable to recover Kernel Panic: trying to free up swapper memory space in swapper task - not syncing. Also, there is something bogus with they keyboard driver, since all the keys yield bogus keyscans. Looks like progress to me :-). Time to checkpoint my changes, and mail them to ralf, along with the milo changes. Some of these changes are ugly, but they have gotten me this far. Tasks: 0) Make patches for Ralf 1) Work on the memory map 2) Work on the "hole" in memory 3) Work on the "copy down the exception hanlders" code and load the code at 0x80100000 4) Play it by ear from here :-) Thanks for everybody's help to get me here. Looks like I won't need the minimal console support after all. A few, well placed debugs saved the day. Warner From andy@waldorf-gmbh.de Mon Sep 4 13:13:25 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id NAA20651 for ; Mon, 4 Sep 1995 13:13:23 -0700 Received: from chenas.inria.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA10704; Mon, 4 Sep 1995 13:21:57 +0200 (MET) Received: from scotty.waldorf-gmbh.de by chenas.inria.fr (5.65c8d/92.02.29) via EUnet-France id AA09032; Mon, 4 Sep 1995 13:22:11 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id HAA14601; Mon, 4 Sep 1995 07:57:50 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id HAA02096; Mon, 4 Sep 1995 07:56:26 +0200 Date: Mon, 4 Sep 1995 07:56:26 +0200 From: Andreas Busse Message-Id: <199509040556.HAA02096@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Re: Non-verbose boot? In-Reply-To: <199509032048.OAA20268@rover.village.org> X-Mailer: [XMailTool v3.1.2b] Content-Length: 595 Lines: 16 > > Is anybody interested in some patches to milo that do a non-verbose > boot? That is, it looks more like the NT boot loader. Prints a . for > each sector loaded, etc? I've been hacking on MILO to streamline the > process of booting (since I figured out how to make LINUX the default > boot), and thought I'd ask if people were interested. > Yes, I am! I was also thinking about printing a dot for each sector, but now that you've done that already... :-) Send me the patches, and I'll integrate them into the current version. Stoned, any other patches from you ? Cheers, Andy From andy@waldorf-gmbh.de Mon Sep 4 13:13:34 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id NAA20678 for ; Mon, 4 Sep 1995 13:13:33 -0700 Received: from chenas.inria.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA10716; Mon, 4 Sep 1995 13:22:06 +0200 (MET) Received: from scotty.waldorf-gmbh.de by chenas.inria.fr (5.65c8d/92.02.29) via EUnet-France id AA09042; Mon, 4 Sep 1995 13:22:23 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id IAA14747; Mon, 4 Sep 1995 08:05:19 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id IAA02107; Mon, 4 Sep 1995 08:03:55 +0200 Date: Mon, 4 Sep 1995 08:03:55 +0200 From: Andreas Busse Message-Id: <199509040603.IAA02107@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Re: Hrumph! In-Reply-To: <199509032052.OAA20554@rover.village.org> X-Mailer: [XMailTool v3.1.2b] Content-Length: 1331 Lines: 39 > > Well, I have the main code that looks like: > *(char *) 0xA00A0000 = '1'; > setup_arch(&command_line, &memory_start, &memory_end); > [...] > That's very similar to the debug stuff I used to make the Magnum and Pica initialization working. Although I even needed some assembler stuff in head.S. > Which is a *BIG* hammer (but much easier than the cylon[*] light). > I've found that setup_arch never returns.... Just thought I'd share > this technique with the list. I'd bet most people know about it that > are actively working on Linux, but maybe some of the lurkers might > not. > But at least you know that you're coming thru the low-level init stuff. That's way better than nothing! > I get to '1', but I never see any other numbers. Either setup_arch is > unmapping the ununmappable, or I'm on to something :-). Anyway, this > gives me a big clue what is going on, so I'll keep working on it. > Just wanted to let people No, I don't think that setup_arch() is doing some magic. Take a look at it and you'll know that is actually nearly empty. Perhaps you need some further debug output there. > > BTW, I also wanted to thank everyone that has been sending me tips. > They are most helpful and useful. > No problem at all! It's great to see some success on the RPC44 ! Cheers, Andy From andy@waldorf-gmbh.de Mon Sep 4 13:13:29 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id NAA20663 for ; Mon, 4 Sep 1995 13:13:29 -0700 Received: from chenas.inria.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA10711; Mon, 4 Sep 1995 13:22:01 +0200 (MET) Received: from scotty.waldorf-gmbh.de by chenas.inria.fr (5.65c8d/92.02.29) via EUnet-France id AA09037; Mon, 4 Sep 1995 13:22:17 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id IAA14769; Mon, 4 Sep 1995 08:07:27 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id IAA02113; Mon, 4 Sep 1995 08:06:04 +0200 Date: Mon, 4 Sep 1995 08:06:04 +0200 From: Andreas Busse Message-Id: <199509040606.IAA02113@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Re: More progress In-Reply-To: <199509032144.PAA21104@rover.village.org> X-Mailer: [XMailTool v3.1.2b] Content-Length: 892 Lines: 30 > > OK. Problem #1: > panic() doesn't work until things are farther along in the > boot process :-(. Linux has been panicing all over the place, > but I can't see the panic message :-(. > Problem #2: > The config name is DESKSTATION_RPC44, but in some places it > was _RPC (note, no 44). Fixed these. > Problem #3: > The IRQ stuff is panicing because there is no case for RPC44. > I'm fixing it as well. > etc. > > Anyway, I've hacked PANIC to print stuff when things aren't too smart > yet. > That explains a lot. > Again, another tip: If you don't get any output, then maybe Linux is > panicing before it has enough brains to print error messages. Maybe > we should use the BIOS routines if no video address is yet set???? > That won't work because the kernel overwrites the BIOS. I guess the fix for panic() you've done is sufficient? Cheers, Andy From Gilles.Gravier@France.Sun.COM Mon Sep 4 12:19:23 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id MAA20001 for ; Mon, 4 Sep 1995 12:19:22 -0700 Received: from mercury.Sun.COM by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA09639; Mon, 4 Sep 1995 12:27:53 +0200 (MET) Received: from snail.Sun.COM by mercury.Sun.COM (Sun.COM) id DAA23299; Mon, 4 Sep 1995 03:28:10 -0700 Received: from France (isunfra) by snail.Sun.COM (4.1/SMI-4.1) id AA04264; Mon, 4 Sep 95 03:28:01 PDT Received: from elektra.France.Sun.COM by France (4.1/SMI-4.1) id AA16301; Mon, 4 Sep 95 12:27:56 +0200 Received: from tatooine.France.Sun.COM by elektra.France.Sun.COM (5.0/SMI-SVR4 (1/24/94)) id AA02377; Mon, 4 Sep 1995 12:27:56 --100 Received: by tatooine.France.Sun.COM (5.x/SMI-SVR4) id AA02198; Mon, 4 Sep 1995 12:27:52 +0200 Date: Mon, 4 Sep 1995 12:27:52 +0200 From: Gilles.Gravier@France.Sun.COM (Gilles Gravier - ITC) Message-Id: <9509041027.AA02198@tatooine.France.Sun.COM> To: linux-mips@fnet.fr Subject: Re: Hello X-Sun-Charset: US-ASCII Content-Length: 1990 Lines: 49 Hi! Just a little question... How far is the Linux/Sun port? (not that I want to replace Solaris on my office workstation, but, well, you know, at home...). Hopefully, someday, the PC, the MIPS, and the Sun will all run Linux at home! Gilles G. /----------------+-------------------------------------------\ | /\ | Gilles Gravier | | \\ \ | Pre-Sales Technical Support | | \ \\ / | Sun Microsystems France | | / \/ / / | 13 avenue Maurane Saulnier | | / / \//\ | B.P. 53 | | \//\ / / | 78142 Velizy CEDEX | | / / /\ / | FRANCE | | / \\ \ | Email: Gilles.Gravier@France.Sun.COM | | \ \\ | Phone: +33-1-30-67-52-81 | | \/ | Fax: +33-1-30-67-53-06 | \----------------+-------------------------------------------/ - From linux-mips-request@guadalquivir.fnet.fr Mon Aug 28 20:43 MET 1995 - Resent-Date: Mon, 28 Aug 1995 20:33:23 -0700 - Date: Mon, 28 Aug 1995 14:38:53 -0400 - From: "David S. Miller" - To: linux-mips@fnet.fr - Subject: Hello - Resent-Message-Id: <"HwEeh.0.QA.3geGm"@guadalquivir.fnet.fr> - Resent-From: linux-mips@guadalquivir.fnet.fr - X-Mailing-List: archive/latest/22 - X-Loop: linux-mips@guadalquivir.fnet.fr - Resent-Sender: linux-mips-request@guadalquivir.fnet.fr - - - Hi, maybe some of you know me and that I am the chief coordinator of - the Sparc port to Linux at this time. I am wondering where the main - workers of the Mips port communicate and if I can get in on some of - the conversations somehow as I imaging many of the problems/issues - facing your team are similar to what I have/will be encountering. - - Thanks, and happy hacking ;) - - Later, - David S. Miller - davem@caip.rutgers.edu - - From andy@waldorf-gmbh.de Mon Sep 4 13:37:35 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id NAA21139 for ; Mon, 4 Sep 1995 13:37:34 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA11110; Mon, 4 Sep 1995 13:45:58 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id NAA19690; Mon, 4 Sep 1995 13:47:53 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id NAA02589; Mon, 4 Sep 1995 13:46:22 +0200 Date: Mon, 4 Sep 1995 13:46:22 +0200 From: Andreas Busse Message-Id: <199509041146.NAA02589@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Re: Life is good :-) In-Reply-To: <199509040456.WAA22474@rover.village.org> X-Mailer: [XMailTool v3.1.2b] Content-Length: 2088 Lines: 68 > > OK. I rebuilt the kernel -DPORT_BASE=0xB0000000 (since the memory map > isn't completely working yet), hacked it here and there, and have some > good news to report: > > I have "dim" output! As you likely know, this means that the kernel > is printing things, and MILO isn't :-). > > For those of you that care, the output says: > > Console: colour CGA 80x50, 1 virtual console (max 63) > Life is good [*] > bios32 is done [*] > Calibrarting delay loop.. ok 50.08 bogo mips Aha. 50 Mhz system clock :-) > keyboard unrecognized scan code (70) ignored > tty00 at 0x3f8 (irq = 4) is a 16450 > tty00 at 0x2f8 (irq = 3) is a 16450 With my 1.2.10.4 patches ? I'm excited !? > Floppy drive(s): fd0 is 1.44M And again! Thanks for checking it on a non-Magnum :-) > Unable to handle kernel paging request at virtual address 600003f2 Opps: 00000001 > > [...] > > Also, there is something bogus with they keyboard driver, since all > the keys yield bogus keyscans. > > Looks like progress to me :-). > In fact! And you were fast too! Congrats! > Time to checkpoint my changes, and mail them to ralf, along with the > milo changes. Some of these changes are ugly, but they have gotten me > this far. > > Tasks: > 0) Make patches for Ralf I'm not sure about how we handle that. I'm working on patch set #5 (kgdb support and other "minor" changes) so it might also be ok if you send them to me. Don't know on what Ralf is working right now. > 1) Work on the memory map > 2) Work on the "hole" in memory > 3) Work on the "copy down the exception hanlders" code and > load the code at 0x80100000 > 4) Play it by ear from here :-) > > Thanks for everybody's help to get me here. Looks like I won't need > the minimal console support after all. A few, well placed debugs > saved the day. > printk() is our best friend. If it's not available, *((char *) 0x...) = 'x' is also quite nice. But wait for the GDB stub! This works long before the kernel even thinks about consoles and such :-) Cheers, Andy From Stoned.Elipot@univ-evry.fr Mon Sep 4 13:42:35 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id NAA21201 for ; Mon, 4 Sep 1995 13:42:34 -0700 Received: from amiante.univ-evry.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA11196; Mon, 4 Sep 1995 13:51:07 +0200 (MET) Received: from bandsept.univ-evry.fr (bandsept.univ-evry.fr [192.93.214.19]) by amiante.univ-evry.fr with ESMTP (8.6.10/94.01.10); id NAA29522; Mon, 4 Sep 1995 13:51:27 +0200 Received: (seb@localhost) by bandsept.univ-evry.fr (8.6.12/94.01.26); id NAA00756; Mon, 4 Sep 1995 13:49:53 +0200 Message-Id: <199509041149.NAA00756@bandsept.univ-evry.fr> Date: Mon, 4 Sep 1995 13:49:53 +0200 From: Stoned Elipot To: linux-mips@fnet.fr In-Reply-To: <199509040556.HAA02096@newton.waldorf-gmbh.de> (message from Andreas Busse on Mon, 4 Sep 1995 07:56:26 +0200) Subject: Re: Non-verbose boot? X-Mailer: GNU Emacs [19.29] Mail Mode X-Attribution: seb X-Pgp-Key: ask a good public key server Reply-To: Stoned.Elipot@univ-evry.fr Content-Length: 1850 Lines: 40 >>>>> "Andreas" == Andreas Busse writes: [SNIP] Andreas> Yes, I am! I was also thinking about printing a dot for each Andreas> sector, but now that you've done that already... :-) Send me Andreas> the patches, and I'll integrate them into the current Andreas> version. Stoned, any other patches from you ? Not yet, go ahead if you wnt to release a new milo, I'm working on Oily RAM detection, and thinking hard about the NAPS (new argument passing scheme, I like this expression :) between Milo/Kernel. BTW Ralf, we should read the tags list in setup_arch() just as bootinfo struct's infos are used in currents kernel version, right ? So What to do if we have larger stuff to extract from the tags list (aka G364 font table :)? My problem is that I can't see where to put this table: I don't want to waste the memory used by the tag list, so I should made this piece of RAM availble for the _init(start_mem, end_mem) routines calls in main.c. But the font table this quite "huge", so what if we tune the creation of tags list by Milo, in order to have such big item, one after another, at the very beginning of the list. Something like: --------+---------+-----+ ... --+----------- kernel | ramdisk | font| | available mem | | | | | | | | --------+---------+-----+- ... -+----------- | | | |<- whole tag list ->| |<- ->| This remains in place |<- This is ->| available as free mem after setup_arch() This way, the font - or other stuff we don't yet know about :) - it at a well defined place, and only few octets are wasted (the tag header of the font). Comments ? Cheers, Stoned. From kah@terma.dk Mon Sep 4 15:39:30 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id PAA24441 for ; Mon, 4 Sep 1995 15:39:30 -0700 Received: from ns.dknet.dk by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA13256; Mon, 4 Sep 1995 15:47:52 +0200 (MET) Received: from terma by ns.dknet.dk with UUCP id AA23196 (5.65c8/IDA-1.4.4j for linux-mips@fnet.fr); Mon, 4 Sep 1995 15:47:48 +0200 Message-Id: <199509041347.AA23196@ns.dknet.dk> Received: by lulu.terma.dk (1.38.193.4/16.2) id AA13256; Mon, 4 Sep 1995 15:23:55 +0200 From: Kai Harrekilde-Petersen Subject: disasm typo correction To: linux-mips@fnet.fr (Linux/MIPS mailing list) Date: Mon, 4 Sep 95 15:23:55 +0200 Mailer: Elm [revision: 70.85] X-Charset: ASCII X-Char-Esc: 29 Content-Length: 1700 Lines: 49 Hi guys, Just a typo correction (yes, the difference between sub and subu is big enough to cause concern) in disasm. Otherwise, we're happily testing away the Sonic and P9100 at the moment. Chris (a colleague of mine) is currently taking the Sonic up to "test level 1" (ie internal loopback), to test that the Sonic can both read and write from/to memory. The P9100 is flying (we've had the first bw GIF on the screen :-) The disasm diff follows after the signature. Kai -- Kai Harrekilde-Petersen Linux: The choice of a GNU generation Terma Elektronik, Surveillance Systems Division #include PGP Key fingerprint = 9A 70 8A FF A3 31 3E DC FB 9C A2 50 D7 FC 4F C8 --- disasm-0.34.c Fri Aug 18 11:33:11 1995 +++ disasm.c Mon Sep 4 16:27:29 1995 @@ -2,8 +2,8 @@ * Copyright (C) 1995 TERMA Elektronik A/S, Lystrup, Denmark * Written by Kai Harrekilde-Petersen * - * Version: 0.33 - * Date: 15. August 1995 + * Version: 0.34a + * Date: 4. September 1995 * * 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 @@ -27,7 +27,7 @@ #include #include -#define VERSION "0.33" +#define VERSION "0.34a" /* instruction classes */ #define TYPE_NONE -1 /* opcode is invalid */ @@ -132,7 +132,7 @@ {"dmult",TYPE_MOVE}, {"dmultu",TYPE_MOVE}, {"ddiv",TYPE_MOVE}, {"ddivu",TYPE_MOVE}, {"add",TYPE_REG}, {"addu",TYPE_REG}, - {"sub",TYPE_REG}, {"sub",TYPE_REG}, + {"sub",TYPE_REG}, {"subu",TYPE_REG}, {"and",TYPE_REG}, {"or",TYPE_REG}, {"xor",TYPE_REG}, {"nor",TYPE_REG}, {"",TYPE_NONE}, {"",TYPE_NONE}, From imp@rover.village.org Mon Sep 4 18:11:52 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id SAA27435 for ; Mon, 4 Sep 1995 18:11:51 -0700 Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA16885; Mon, 4 Sep 1995 18:20:13 +0200 (MET) Received: from LOCALHOST (LOCALHOST [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id KAA23678 for ; Mon, 4 Sep 1995 10:20:10 -0600 Message-Id: <199509041620.KAA23678@rover.village.org> To: linux-mips@fnet.fr Subject: Re: More progress In-Reply-To: Your message of Mon, 04 Sep 1995 08:06:04 +0200 Date: Mon, 04 Sep 1995 10:20:09 -0600 From: Warner Losh Content-Length: 145 Lines: 6 : That won't work because the kernel overwrites the BIOS. I guess : the fix for panic() you've done is sufficient? Yes. It was enough! Warner From ralf@waldorf-gmbh.de Mon Sep 4 18:39:12 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id SAA27718 for ; Mon, 4 Sep 1995 18:39:12 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA17473; Mon, 4 Sep 1995 18:47:41 +0200 (MET) From: Ralf Baechle Message-Id: <199509041648.SAA24784@scotty.waldorf-gmbh.de> Received: from localhost by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id SAA24784; Mon, 4 Sep 1995 18:48:41 +0200 Subject: Re: Non-verbose boot? To: linux-mips@fnet.fr Date: Mon, 4 Sep 1995 18:48:41 +0200 (MET DST) In-Reply-To: <199509040556.HAA02096@newton.waldorf-gmbh.de> from "Andreas Busse" at Sep 4, 95 07:56:26 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 677 Lines: 16 Hi, > > Is anybody interested in some patches to milo that do a non-verbose > > boot? That is, it looks more like the NT boot loader. Prints a . for > > each sector loaded, etc? I've been hacking on MILO to streamline the > > process of booting (since I figured out how to make LINUX the default > > boot), and thought I'd ask if people were interested. > > Yes, I am! I was also thinking about printing a dot for each > sector, but now that you've done that already... :-) > Send me the patches, and I'll integrate them into the current > version. Stoned, any other patches from you ? I also made some changes mostly reflecting changes in other packages. Ralf From andy@waldorf-gmbh.de Mon Sep 4 19:40:13 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id TAA28339 for ; Mon, 4 Sep 1995 19:40:12 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA18430; Mon, 4 Sep 1995 19:48:41 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id TAA25729; Mon, 4 Sep 1995 19:49:44 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id TAA05415; Mon, 4 Sep 1995 19:49:12 +0200 Date: Mon, 4 Sep 1995 19:49:12 +0200 From: Andreas Busse Message-Id: <199509041749.TAA05415@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Re: Non-verbose boot? In-Reply-To: <199509041648.SAA24784@scotty.waldorf-gmbh.de> X-Mailer: [XMailTool v3.1.2b] Content-Length: 127 Lines: 9 > > I also made some changes mostly reflecting changes in other packages. > Ok, then send me patches too. Cheers, Andy From imp@rover.village.org Tue Sep 5 00:26:21 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id AAA01320 for ; Tue, 5 Sep 1995 00:26:20 -0700 Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA23274; Tue, 5 Sep 1995 00:34:50 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id QAA00187 for ; Mon, 4 Sep 1995 16:34:52 -0600 Message-Id: <199509042234.QAA00187@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Non-verbose boot? In-Reply-To: Your message of Mon, 04 Sep 1995 13:49:53 +0200 Date: Mon, 04 Sep 1995 16:34:51 -0600 From: Warner Losh Content-Length: 463 Lines: 11 : This way, the font - or other stuff we don't yet know about :) - it at : a well defined place, and only few octets are wasted (the tag header : of the font). Comments ? My only concern with this would be that the Deskstation rPC44 has a hole in memory between 0xa0000 and 0xfffff. Other than that, it looks OK. Oh, you might run into a problem if you just kfree it, since the kernel will complain that you are freeing memory not returned by kmalloc. Warner From ralf@waldorf-gmbh.de Tue Sep 5 00:36:00 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id AAA01417 for ; Tue, 5 Sep 1995 00:35:59 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA23538; Tue, 5 Sep 1995 00:44:30 +0200 (MET) From: Ralf Baechle Message-Id: <199509042245.AAA30959@scotty.waldorf-gmbh.de> Received: from localhost by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id AAA30959; Tue, 5 Sep 1995 00:45:41 +0200 Subject: Re: Non-verbose boot? To: linux-mips@fnet.fr Date: Tue, 5 Sep 1995 00:45:40 +0200 (MET DST) In-Reply-To: <199509042234.QAA00187@rover.village.org> from "Warner Losh" at Sep 4, 95 04:34:51 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 903 Lines: 22 Hi, > : This way, the font - or other stuff we don't yet know about :) - it at > : a well defined place, and only few octets are wasted (the tag header > : of the font). Comments ? > > My only concern with this would be that the Deskstation rPC44 has a > hole in memory between 0xa0000 and 0xfffff. Other than that, it looks > OK. Oh, you might run into a problem if you just kfree it, since the > kernel will complain that you are freeing memory not returned by > kmalloc. How about copying the stuff to the end of available memory. During startup in setup_arch() the kernel decides what it needs and what not. All the required stuff will be copied down just after _end. The memory range from behind the last junk to the end of memory is then returned and initialized as normal. This makes it necessary that the kernel makes it's decissions very early. I think we can live with that. Ralf From andy@waldorf-gmbh.de Tue Sep 5 18:53:00 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id SAA13987 for ; Tue, 5 Sep 1995 18:52:59 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA12257; Tue, 5 Sep 1995 19:01:20 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id TAA13445; Tue, 5 Sep 1995 19:02:12 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id TAA10602; Tue, 5 Sep 1995 19:01:39 +0200 Date: Tue, 5 Sep 1995 19:01:39 +0200 From: Andreas Busse Message-Id: <199509051701.TAA10602@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Bugs... X-Mailer: [XMailTool v3.1.2b] Content-Length: 1481 Lines: 47 Hi all, I'm fighting with the GDB stub and (who wonders) think mysterious things happening... The problem I have that the gdb stub works, but not for a BREAK instruction. In exactly this case, the CP0_EPC and CP0_CAUSE are exchanged, and the CP0_STATUS register on the stackframe contains nonsense. Before you say: check your handler I say: It's the same for all exceptions, *including* BREAK. So I took a look into the R4000 V2.2 and 3.0 errata. Not that I found something regarding this problem, but something else instead: [quoted] The following conditions cause the R4000 to operate incorrectly: 1) An exception is taken from user code 2) On the ERET instruction of the exception handler, a CacheErr is taken The R4000 takes the CacheErr exception correctly but returns to user code instead of the ERET of the exception handler. This will then cause an Address Error exception. WORKAROUND: Use the following code sequence as the last three instructions of the exception handler: eret nop eret The CacheErr handler must add 4 to the ErrorEPC to return to the nop. [end of quoted material] I guess we should take this into account. Any comments? Cheers, Andy ----------------------------------------------------------- Andreas Busse | andy@waldorf-gmbh.de Waldorf Electronics GmbH, R&D Dep. | Phone: +49 2636-970105 Neustrasse 9-12, D-53498 Waldorf | Fax: +49 2636-970106 ----------------------------------------------------------- From andy@waldorf-gmbh.de Tue Sep 5 19:24:03 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id TAA14337 for ; Tue, 5 Sep 1995 19:24:02 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA12747; Tue, 5 Sep 1995 19:32:32 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id TAA13909; Tue, 5 Sep 1995 19:33:36 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id TAA10955; Tue, 5 Sep 1995 19:33:03 +0200 Date: Tue, 5 Sep 1995 19:33:03 +0200 From: Andreas Busse Message-Id: <199509051733.TAA10955@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: S3 email ? X-Mailer: [XMailTool v3.1.2b] Content-Length: 79 Lines: 7 Hi all, does someone has an email address of S3, by accident ? Thanks, Andy From andy@waldorf-gmbh.de Tue Sep 5 20:53:46 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id UAA15383 for ; Tue, 5 Sep 1995 20:53:45 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA13991; Tue, 5 Sep 1995 21:02:16 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id VAA15239; Tue, 5 Sep 1995 21:03:24 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id VAA11867; Tue, 5 Sep 1995 21:02:49 +0200 Date: Tue, 5 Sep 1995 21:02:49 +0200 From: Andreas Busse Message-Id: <199509051902.VAA11867@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Remote-GDB for Linux/MIPS X-Mailer: [XMailTool v3.1.2b] Content-Length: 1202 Lines: 38 Hi all! So. The GDB stub works so far. There are some minor problems during startup of the host gdb, but as soon vmlinux executes a break instruction and enters the gdb exception handler, the communication syncs. Things supported: - full source code debugging (yes, I compiled the kernel with -g) - breakpoint setting - register dumps (incl. FPU, if enabled) - disassembly by reading memory on the target - memory dumps - etc. Things not yet supported: - Modifying registers, except for breakpoint handling - other baudrates and devices than 9600 on ttyS0 - kill, reset - Single Step - and many other things being possible at least in theory However, if demand is high I will concentrate on adding missing features. Question is: Someone interested in this current version? If this is the case, I'll upload kernel patches, gdb patches as well as gdb binaries to ftp.waldorf-gmbh.de. Cheers, Andy ----------------------------------------------------------- Andreas Busse | andy@waldorf-gmbh.de Waldorf Electronics GmbH, R&D Dep. | Phone: +49 2636-970105 Neustrasse 9-12, D-53498 Waldorf | Fax: +49 2636-970106 ----------------------------------------------------------- From Stoned.Elipot@univ-evry.fr Tue Sep 5 21:18:20 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id VAA15651 for ; Tue, 5 Sep 1995 21:18:19 -0700 Received: from amiante.univ-evry.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA14421; Tue, 5 Sep 1995 21:26:46 +0200 (MET) Received: from bandsept.univ-evry.fr (bandsept.univ-evry.fr [192.93.214.19]) by amiante.univ-evry.fr with ESMTP (8.6.10/94.01.10); id VAA01633; Tue, 5 Sep 1995 21:27:04 +0200 Received: (seb@localhost) by bandsept.univ-evry.fr (8.6.12/94.01.26); id VAA02466; Tue, 5 Sep 1995 21:25:30 +0200 Message-Id: <199509051925.VAA02466@bandsept.univ-evry.fr> Date: Tue, 5 Sep 1995 21:25:30 +0200 From: Stoned Elipot To: linux-mips@fnet.fr In-Reply-To: <199509051902.VAA11867@newton.waldorf-gmbh.de> (message from Andreas Busse on Tue, 5 Sep 1995 21:02:49 +0200) Subject: Re: Remote-GDB for Linux/MIPS X-Mailer: GNU Emacs [19.29] Mail Mode X-Attribution: seb X-Pgp-Key: ask a good public key server Reply-To: Stoned.Elipot@univ-evry.fr Content-Length: 338 Lines: 12 Hi, >>>>> "Andreas" == Andreas Busse writes: [SNIP] Andreas> Question is: Someone interested in this current version? If Andreas> this is the case, I'll upload kernel patches, gdb patches as Andreas> well as gdb binaries to ftp.waldorf-gmbh.de. Great work Andy. Yes indeed, I'm interested ! :) Cheers, Stoned. From davem@caip.rutgers.edu Tue Sep 5 23:05:49 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id XAA16764 for ; Tue, 5 Sep 1995 23:05:48 -0700 Received: from caipfs.rutgers.edu by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA16034; Tue, 5 Sep 1995 23:14:19 +0200 (MET) Received: from huahaga.rutgers.edu (huahaga.rutgers.edu [128.6.155.53]) by caipfs.rutgers.edu (8.6.9+bestmx+oldruq+newsunq+grosshack/8.6.9) with ESMTP id RAA14511 for ; Tue, 5 Sep 1995 17:14:37 -0400 Received: (davem@localhost) by huahaga.rutgers.edu (8.6.9+bestmx+oldruq+newsunq+grosshack/8.6.9) id RAA26265; Tue, 5 Sep 1995 17:14:37 -0400 Date: Tue, 5 Sep 1995 17:14:37 -0400 Message-Id: <199509052114.RAA26265@huahaga.rutgers.edu> From: "David S. Miller" To: linux-mips@fnet.fr In-Reply-To: <199509051902.VAA11867@newton.waldorf-gmbh.de> (message from Andreas Busse on Tue, 5 Sep 1995 21:02:49 +0200) Subject: Re: Remote-GDB for Linux/MIPS Content-Length: 1055 Lines: 28 Date: Tue, 5 Sep 1995 21:02:49 +0200 From: Andreas Busse So. The GDB stub works so far. There are some minor problems during startup of the host gdb, but as soon vmlinux executes a break instruction and enters the gdb exception handler, the communication syncs. Andy, I guess since direct modification of registers is not working you cannot call arbitrary functions in the kernel as of yet? I find it a life saver at times to be able to do (gdb) print show_free_areas() and similar ;-) Also, you can do things like: (gdb) print panic("gdb panics") but only if you provide a 'malloc()' function in the kernel for the stub to call to get a memory pool to store the string into. I basically on the sparc allocate a page for gdb to alloc from and wrote simple memory pool routines and a malloc() which is static in the stub file itself, once the page is used up malloc() returns null. I find this very useful personally (especially for debugging the mm code). Later, David S. Miller davem@caip.rutgers.edu From andy@waldorf-gmbh.de Wed Sep 6 07:27:30 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id HAA23237 for ; Wed, 6 Sep 1995 07:27:29 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA24186; Wed, 6 Sep 1995 07:36:01 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id HAA25399; Wed, 6 Sep 1995 07:37:02 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id HAA13340; Wed, 6 Sep 1995 07:36:33 +0200 Date: Wed, 6 Sep 1995 07:36:33 +0200 From: Andreas Busse Message-Id: <199509060536.HAA13340@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Re: Remote-GDB for Linux/MIPS In-Reply-To: <199509051925.VAA02466@bandsept.univ-evry.fr> X-Mailer: [XMailTool v3.1.2b] Content-Length: 360 Lines: 13 > Andreas> Question is: Someone interested in this current version? If > Andreas> this is the case, I'll upload kernel patches, gdb patches as > Andreas> well as gdb binaries to ftp.waldorf-gmbh.de. > > Great work Andy. Yes indeed, I'm interested ! :) > Ok. I'll do some fine tuning and then upload it. Expect everything for tomorrow. Cheers, Andy From andy@waldorf-gmbh.de Wed Sep 6 07:29:43 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id HAA23266 for ; Wed, 6 Sep 1995 07:29:41 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA24206; Wed, 6 Sep 1995 07:38:12 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id HAA25425; Wed, 6 Sep 1995 07:39:13 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id HAA13349; Wed, 6 Sep 1995 07:38:45 +0200 Date: Wed, 6 Sep 1995 07:38:45 +0200 From: Andreas Busse Message-Id: <199509060538.HAA13349@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Re: Remote-GDB for Linux/MIPS In-Reply-To: <199509052114.RAA26265@huahaga.rutgers.edu> X-Mailer: [XMailTool v3.1.2b] Content-Length: 985 Lines: 26 > > Andy, I guess since direct modification of registers is not working > you cannot call arbitrary functions in the kernel as of yet? I find > it a life saver at times to be able to do > > (gdb) print show_free_areas() > > and similar ;-) Also, you can do things like: > > (gdb) print panic("gdb panics") > > but only if you provide a 'malloc()' function in the kernel for the > stub to call to get a memory pool to store the string into. I > basically on the sparc allocate a page for gdb to alloc from and wrote > simple memory pool routines and a malloc() which is static in the stub > file itself, once the page is used up malloc() returns null. I find > this very useful personally (especially for debugging the mm code). > Well, it's probably not that much effort to make this working too. However, I'm quite overloaded these days and probably don't find the time. As I said, if demand is high I'll add it sooner. Otherwise later :-) Cheers, Andy From andy@waldorf-gmbh.de Wed Sep 6 14:06:55 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id OAA27240 for ; Wed, 6 Sep 1995 14:01:14 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA01141; Wed, 6 Sep 1995 14:03:14 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id OAA31870; Wed, 6 Sep 1995 14:04:22 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id OAA17748; Wed, 6 Sep 1995 14:03:44 +0200 Date: Wed, 6 Sep 1995 14:03:44 +0200 From: Andreas Busse Message-Id: <199509061203.OAA17748@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Linux-1.2.10.5 patches X-Mailer: [XMailTool v3.1.2b] Content-Length: 1711 Lines: 44 Hi all, I've uploaded patch set #5 to ftp.waldorf-gmbh.de, in /pub/linux/mips/private/Incoming/linux-1.2.10.4-1.2.10.5.diffs.gz These patches contain: - the remote GDB stub, in 4 parts: arch/mips/kernel/gdb-low.S the low-level exception handler arch/mips/kernel/gdb-stub.c the "high-level" part of the stub incluse/asm-mips/gdb-stub.h include file for the gdb stub drivers/char/serial.c primitive serial driver for ttyS0 - some minor changes to include files - new configuration option CONFIG_REMOTE_DEBUG Some DONT's: - You can put the gdb initialization (a call to set_debug_traps()) whereever you want, but never before init_traps() since init_traps() would overwrite the exception handler pointers set_debug_traps() has set. - You cannot use the GDB stub together with serial line support. The GDB stub has its own very primitive serial driver which works without interrupts, queues, timers etc. Enabling the normal serial drivers will confuse the communication between host and target. I will change this later, but for now you must choose between debugging support or serial lines. - There's no way to disable remote debugging while it's compiled into the kernel. Due to the lack of a command line interface the stub is enabled as soon you turn on CONFIG_REMOTE_DEBUG. I'll change that as soon we have a working command line interface. GDB patches and binaries follow soon. Cheers, Andy ----------------------------------------------------------- Andreas Busse | andy@waldorf-gmbh.de Waldorf Electronics GmbH, R&D Dep. | Phone: +49 2636-970105 Neustrasse 9-12, D-53498 Waldorf | Fax: +49 2636-970106 ----------------------------------------------------------- From andy@waldorf-gmbh.de Wed Sep 6 14:40:09 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id OAA27627 for ; Wed, 6 Sep 1995 14:40:08 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA01937; Wed, 6 Sep 1995 14:48:37 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id OAA32531; Wed, 6 Sep 1995 14:49:48 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id OAA21256; Wed, 6 Sep 1995 14:49:10 +0200 Date: Wed, 6 Sep 1995 14:49:10 +0200 From: Andreas Busse Message-Id: <199509061249.OAA21256@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: gdb-4.14.1 patches X-Mailer: [XMailTool v3.1.2b] Content-Length: 1016 Lines: 28 Hi all, I've uploaded patches to GDB to ftp.waldorf-gmbh.de in /pub/linux/mips/crossdev/src/gdb-4.14-4.14.1.diffs.gz These patches contain the necessary changes to run GDB as a remote debugger for the Linux/MIPS kernel. Note that this is tested for a.out kernels only. ELF support is not built in, but should be possible by applying the binutils/bfd patches for ELF to the BFD stuff of GDB too. Something I forgot to say to the kernel patches: They contain a .gdbinit file and a short script named "mkdisk". If you want to use gdb as a remote debugger, copy both files from arch/mips to the kernel top directory. "mkdisk" copies the kernel to a temp file, strips it and then writes it to floppy. Cheers, Andy ----------------------------------------------------------- Andreas Busse | andy@waldorf-gmbh.de Waldorf Electronics GmbH, R&D Dep. | Phone: +49 2636-970105 Neustrasse 9-12, D-53498 Waldorf | Fax: +49 2636-970106 ----------------------------------------------------------- From andy@waldorf-gmbh.de Wed Sep 6 15:42:57 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id PAA28329 for ; Wed, 6 Sep 1995 15:42:56 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA03081; Wed, 6 Sep 1995 15:51:22 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id PAA01020; Wed, 6 Sep 1995 15:52:30 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id PAA24172; Wed, 6 Sep 1995 15:51:50 +0200 Date: Wed, 6 Sep 1995 15:51:50 +0200 From: Andreas Busse Message-Id: <199509061351.PAA24172@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: gdb-4.14.1 binary X-Mailer: [XMailTool v3.1.2b] Content-Length: 245 Lines: 15 Finally... I've uploaded a binary of gdb-4.14.1 to ftp.waldorf-gmbh.de in /pub/linux/mips/crossdev/i486-linux/gdb-4.14.1.tar.gz. As you might expect, this is an a.out binary package. If you want ELF, compile it yourself! :-) Cheers, Andy From andy@waldorf-gmbh.de Wed Sep 6 16:07:32 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id QAA28592 for ; Wed, 6 Sep 1995 16:07:30 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA03691; Wed, 6 Sep 1995 16:15:55 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id QAA01393; Wed, 6 Sep 1995 16:16:32 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id QAA24220; Wed, 6 Sep 1995 16:15:53 +0200 Date: Wed, 6 Sep 1995 16:15:53 +0200 From: Andreas Busse Message-Id: <199509061415.QAA24220@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: TODO list (update) X-Mailer: [XMailTool v3.1.2b] Content-Length: 2619 Lines: 124 Hi again! Here it is again, the lovely TODO list. Please let me know if I can remove something :-) Cheers, Andy ------------------------snip-------------------------- The Linux/MIPS TODO list 95/09/06 This list is split into several parts: Loader: Bootstrap loader related things. Kernel: Kernel-related topics, such as drivers needed, bugs to fix etc. Crossdev: Cross-compiler and linker related things to do. Library: Linux/MIPS C Library related topics. Applications: All user code related things. Tools: Needed tools, for porting, debugging and so on. If you think you can do something on Linux/MIPS, first check what you would most prefer or what is most suited to you. Then see below if there is something needed. If you find something you like to work on, please let us know so that work won't be done twice. Thanks! Please send submissions, suggestions etc. directly to me. Andy NOTE: New Items are marked with an asterix (*) LOADER ====== 1. DECstation 3100/5000 loaders, both for harddisk and network boot. 2. Mips Magnum 3000 and RC3xxx boot loaders 3. Milo: Magnum 4000 and Olivetti M700 video mode detection needed. 4. Milo: Support for new bootinfo item `MACHGROUP' needed. 5: Milo: Add support for Deskstation rPC44. KERNEL ====== 1. Integrate support for Olivetti M700. Magnum 4000 code can be used. 2. Fix JAZZ SONIC driver 3. Fix TCP/IP and UDP protocol stacks 4. SCSI driver for Magnum 4000, Olivetti M700 and Acer PICA-61 needed. 5. Test parallel driver for Magnum 4000, Olivetti M700 and Acer PICA-61 6. Combine interrupt handlers of Acer PICA, Mips Magnum 4000 and Olivetti M700 into a single source file, if possible. 7. Add support for the Deskstation rPC44 interrupt handler. 8. Add support for i386 ISA drivers to the kernel, mostly for Deskstation rPC44 9. Add support for EISA bus on the rPC44. 10. Complete GDB stub for remote debugging (Mostly done now) * 11. Add command line parser to arch/mips/kernel/setup_arch() CROSSDEV ======== 1. Fix ELF support for target. 2. Provide binary crossdev distribution for ELF- and a.out-based Linux/i386 3. Create instructions for building cross compilers for your platform. LIBRARY ======= 1. Check and debug C library 2. Create shared library TOOLS ===== 1. Fix ELF support in GDB, if necessary. APPLICATIONS (if one can say so at this time :-)) ============ 1. Release source and binary of basic apps, such as a shell, fs tools and other essential programs. 2. Create rootdisk/bootdisk sets for all target machines From imp@rover.village.org Thu Sep 7 08:09:21 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id IAA09792 for ; Thu, 7 Sep 1995 08:09:18 -0700 Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA19160; Thu, 7 Sep 1995 08:17:42 +0200 (MET) Received: from LOCALHOST (LOCALHOST [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id AAA04994 for ; Thu, 7 Sep 1995 00:17:58 -0600 Message-Id: <199509070617.AAA04994@rover.village.org> To: linux-mips@fnet.fr Subject: MIPS Assembler help Date: Thu, 07 Sep 1995 00:17:58 -0600 From: Warner Losh Content-Length: 3605 Lines: 102 What follows is very detailed, but I hope that some of the "old-timers" can take a look at it and let me know where I'm going astray. I'm having a crash in the floppy init code. I'm starting to think that gcc just might be generating bogus code, but I can't be sure since I don't swim in MIPS R4x00 assembler just yet. Before I start, my kernel is compiled -DPORT_BASE=0xB0000000 rather than using the default address for that, since I don't have my maps working quite right yet. I guess I'm having trouble with delay slots. R -> R operations take place right away, right? The 'C' code: for(i=0; i< N_FDC; i++){ if(FDCS->address != -1){ fdc = i; reset_fdc_info(1); fd_out(FDCS->dor, FD_DOR); /* outb_p(fdc_state[fdc].dor, FD_DOR) */ } } The fd_out line expands to outb_p(fdc_state[fdc].dor, 2 + fdc_state[fdc].address); and then to ((__builtin_constant_p((2 + fdc_state[fdc].address)) && (2 + fdc_state[fdc].address) < 32768) ? __outbc_p((fdc_state[fdc].dor),(2 + fdc_state[fdc].address)) : __outb_p((fdc_state[fdc].dor),(2 + fdc_state[fdc].address))); which should optimize to __outb_p((fdc_state[fdc].dor),(2 + fdc_state[fdc].address)); since __buildin_constant_p is going to be false. [[ Hmmm, I've just hacked this part of the code, so maybe that's the probelm :-( ]] which translates to: extern inline void __outb_p(unsigned int value, unsigned int port) { __asm__ __volatile__ ("s" #b "\t%0,%1(%2)" : : "r" (value), "i" (0), "r" (PORT_BASE+port)); __asm__ __volatile__( "sb\t$0,0x80(%0)" : : "r" (PORT_BASE)); __asm__ __volatile__( "sb\t$0,0x80(%0)" : : "r" (PORT_BASE)); __asm__ __volatile__( "sb\t$0,0x80(%0)" : : "r" (PORT_BASE)); __asm__ __volatile__( "sb\t$0,0x80(%0)" : : "r" (PORT_BASE)); } Which then gets expanded out to the following by gcc: ... li $19,-1342177280 # 0xb0000000 # $18 seems to be 0x80030b8 which is in the bss section at fdc_state # fdc is at 0x800830d0, but that is likely not relevant ... lw $3,fdc #nop # $3 = 0 sll $2,$3,1 # $2 = 0 addu $2,$2,$3 # $2 = 0 sll $2,$2,3 # $2 = 0 (which is fdc * 24, btw) lw $3,fdc_state+16($2) # Loads the contents of 0x800830e0 into $3. # Why n+16???? addu $2,$2,$18 # $2 = 80030b8 lbu $2,13($2) # Load the DOR byte into $2 addu $3,$3,2 # $3 should have the address in it now addu $3,$3,$19 # Add 2 and also add the BASE_PORT to $3 #APP sb $2,0($3) # *** $3 = 0x600003f2 #NO_APP li $9,-1342177280 # 0xb0000000 #APP sb $0,0x80($9) sb $0,0x80($9) sb $0,0x80($9) sb $0,0x80($9) #NO_APP ... At the three stared point in the code, $3 is 0x600003f2. It really should be 0xB00003f2. I'm just not seeing where things are going amiss. address = 0xb000003f0 dor = 0x4. FD_DOR = 0xb000003f2 FD_DOR is interesting, since it is BASE_PORT + FD_IOPORT + 2, but I think it should really only be FD_IOPORT + 2 = 0x3f0. That would make the macro expand correctly, and 0xb0000000 + 0xb0000000 = 0x160000000 or 0x60000000, which is the virtual address in my crash. So I'm confused. Why is it defined like that? I would guess that is because that's the way it is on the ACER, MAGNUM and OLIVETTY boards (fixed at location 0xe00030000), but on all others it is defined to be PORT_BASE + 0x3f0. Why not just 0x3f0? On the Deskstation, that seems to be the right thing to define it as, but I'm not keen on breaking anybody else's ports. Wo whats the right thing to do here? When I change it to be just 0xf30 I get farther in my boot (it hangs afer printing FDC 0 is a 8272A, which is more progress). Is this a valid change, or should I be doing something different? Warner From andy@waldorf-gmbh.de Thu Sep 7 08:42:43 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id IAA10132 for ; Thu, 7 Sep 1995 08:42:42 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA19661; Thu, 7 Sep 1995 08:51:14 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id IAA16646; Thu, 7 Sep 1995 08:52:20 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id IAA28048; Thu, 7 Sep 1995 08:51:48 +0200 Date: Thu, 7 Sep 1995 08:51:48 +0200 From: Andreas Busse Message-Id: <199509070651.IAA28048@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Re: MIPS Assembler help In-Reply-To: <199509070617.AAA04994@rover.village.org> X-Mailer: [XMailTool v3.1.2b] Content-Length: 3080 Lines: 87 > > What follows is very detailed, but I hope that some of the > "old-timers" can take a look at it and let me know where I'm going > astray. > > I'm having a crash in the floppy init code. I'm starting to think > that gcc just might be generating bogus code, but I can't be sure > since I don't swim in MIPS R4x00 assembler just yet. > > Before I start, my kernel is compiled -DPORT_BASE=0xB0000000 rather > than using the default address for that, since I don't have my maps > working quite right yet. > > I guess I'm having trouble with delay slots. R -> R operations take > place right away, right? Yes, as long you don't use CP0 registers. But that's another case... > > The 'C' code: > for(i=0; i< N_FDC; i++){ > if(FDCS->address != -1){ > fdc = i; > reset_fdc_info(1); > fd_out(FDCS->dor, FD_DOR); /* outb_p(fdc_state[fdc].dor, FD_DOR) */ > } > } > > [ Code deleted ] > > > At the three stared point in the code, $3 is 0x600003f2. It really > should be 0xB00003f2. I'm just not seeing where things are going > amiss. > > address = 0xb000003f0 > dor = 0x4. > > FD_DOR = 0xb000003f2 > > FD_DOR is interesting, since it is BASE_PORT + FD_IOPORT + 2, but I > think it should really only be FD_IOPORT + 2 = 0x3f0. That would make > the macro expand correctly, and 0xb0000000 + 0xb0000000 = 0x160000000 > or 0x60000000, which is the virtual address in my crash. > That sounds familiar. Actually, it's the same problem I had with the Magnum. The out## and in## macros add PORT_BASE by default. So if you say FD_IOPORT is 0xb00003f0 and PORT_BASE is 0xb0000000 you finally get 0x600003f0 as address. It might be my fault. Take a look at floppy.c, somewhere at the top: #define FDC1 ((boot_info.machtype == MACH_ACER_PICA_61 || \ boot_info.machtype == MACH_MIPS_MAGNUM_4000 || \ boot_info.machtype == MACH_OLIVETTI_M700) ? \ 0xe0003000 : PORT_BASE + 0x3f0) Change that into ... boot_info.machtype == MACH_OLIVETTI_M700) ? \ 0xe0003000 : 0x3f0) and everything should be fine. I was thinking about changing the in## and out## macros at all and might have left that in by accident or lazyness. Sorry. > So I'm confused. Why is it defined like that? I would guess that is > because that's the way it is on the ACER, MAGNUM and OLIVETTY boards > (fixed at location 0xe00030000), but on all others it is defined to be > PORT_BASE + 0x3f0. Why not just 0x3f0? On the Deskstation, that > seems to be the right thing to define it as, but I'm not keen on > breaking anybody else's ports. > > Wo whats the right thing to do here? When I change it to be just > 0xf30 I get farther in my boot (it hangs afer printing FDC 0 is a > 8272A, which is more progress). Is this a valid change, or should I > be doing something different? > It's a valid change. See above. I'll correct that in my source tree too. I must admit that my changes to the floppy driver are totally untested for any other box than Jazz's. It might even break on Intel boxes... Sorry again, Andy From andy@waldorf-gmbh.de Thu Sep 7 20:00:28 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id UAA17029 for ; Thu, 7 Sep 1995 20:00:27 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA01749; Thu, 7 Sep 1995 20:08:57 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id MAA20666; Thu, 7 Sep 1995 12:31:48 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id MAA28866; Thu, 7 Sep 1995 12:31:11 +0200 Date: Thu, 7 Sep 1995 12:31:11 +0200 From: Andreas Busse Message-Id: <199509071031.MAA28866@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Speedup Milo X-Mailer: [XMailTool v3.1.2b] Content-Length: 630 Lines: 19 Hi all, I'm tired of always waiting 5 minutes to boot a kernel on my Magnum, really. So I've played around with buffered file I/O, but even this didn't help. I believe the problem is that as soon I do a seek() floppy thruput decreases to nearly zero. Reading a file just from the beginning to the end is way faster. So I'd like to re-organize the way Milo loads the kernel. I'm thinking of loading the whole thing at once into a large buffer, and then setting appropriate pointers to the symbols, text and data spaces. Before I start writing that I'd like to hear comments about the other's plans. Are there any? Cheers, Andy From andy@waldorf-gmbh.de Thu Sep 7 19:57:16 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id TAA16983 for ; Thu, 7 Sep 1995 19:57:15 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA01666; Thu, 7 Sep 1995 20:05:45 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id OAA22224; Thu, 7 Sep 1995 14:12:35 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id OAA30294; Thu, 7 Sep 1995 14:11:54 +0200 Date: Thu, 7 Sep 1995 14:11:54 +0200 From: Andreas Busse Message-Id: <199509071211.OAA30294@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Milo Speedup X-Mailer: [XMailTool v3.1.2b] Content-Length: 922 Lines: 35 Hi again, I got no comments -- doesn't matter :-) Anyway, I hacked Milo to load the kernel file with a single read() call, and what shall I say... Kernel load time for 728019 bytes on my Magnum 4000: before: 300 seconds after: 26 seconds and Milo is some kbytes smaller now, too. To the ELF hackers: elf_open_kernel() isn't necessary anymore. You just need to write elf_load_kernel the same way as aout_load_kernel() now works: - get kernel filesize - malloc filesize bytes - read complete file - copy exec header into struct kexec - set pointer to symbols - return pointer to kernel text That's it. Cheers, Andy ----------------------------------------------------------- Andreas Busse | andy@waldorf-gmbh.de Waldorf Electronics GmbH, R&D Dep. | Phone: +49 2636-970105 Neustrasse 9-12, D-53498 Waldorf | Fax: +49 2636-970106 ----------------------------------------------------------- From Stoned.Elipot@univ-evry.fr Thu Sep 7 20:14:20 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id UAA17169 for ; Thu, 7 Sep 1995 20:14:20 -0700 Received: from amiante.univ-evry.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA02039; Thu, 7 Sep 1995 20:22:52 +0200 (MET) Received: from bandsept.univ-evry.fr (bandsept.univ-evry.fr [192.93.214.19]) by amiante.univ-evry.fr with ESMTP (8.6.10/94.01.10); id UAA05134; Thu, 7 Sep 1995 20:23:12 +0200 Received: (seb@localhost) by bandsept.univ-evry.fr (8.6.12/94.01.26); id UAA06502; Thu, 7 Sep 1995 20:21:37 +0200 Message-Id: <199509071821.UAA06502@bandsept.univ-evry.fr> Date: Thu, 7 Sep 1995 20:21:37 +0200 From: Stoned Elipot To: linux-mips@fnet.fr In-Reply-To: <199509071211.OAA30294@newton.waldorf-gmbh.de> (message from Andreas Busse on Thu, 7 Sep 1995 14:11:54 +0200) Subject: Re: Milo Speedup X-Mailer: GNU Emacs [19.29] Mail Mode X-Attribution: seb X-Pgp-Key: ask a good public key server Reply-To: Stoned.Elipot@univ-evry.fr Content-Length: 469 Lines: 16 Hi Andy, Hi you all, >>>>> "Andreas" == Andreas Busse writes: Andreas> Hi again, Andreas> I got no comments -- doesn't matter :-) Yes indeed, guadalquivir receive your first mail 3 minutes *after* your second one !! And that was about 5 hours after you sent the first one !! Ah, information superhighways ! :) [SNIP] Andreas> before: 300 seconds after: 26 seconds Wouah,... So is the Magnum BIOS's read() broken or not ??? Cheers, stoned. From andy@waldorf-gmbh.de Thu Sep 7 21:38:44 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id VAA18030 for ; Thu, 7 Sep 1995 21:38:43 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA03199; Thu, 7 Sep 1995 21:47:15 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id VAA31455; Thu, 7 Sep 1995 21:48:27 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id VAA32386; Thu, 7 Sep 1995 21:47:48 +0200 Date: Thu, 7 Sep 1995 21:47:48 +0200 From: Andreas Busse Message-Id: <199509071947.VAA32386@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Re: Milo Speedup In-Reply-To: <199509071821.UAA06502@bandsept.univ-evry.fr> X-Mailer: [XMailTool v3.1.2b] Content-Length: 1161 Lines: 32 Hi all! > > Yes indeed, guadalquivir receive your first mail 3 minutes *after* your > second one !! And that was about 5 hours after you sent the first one > !! Ah, information superhighways ! :) No, *this* time it wasn't a problem of the so-called "data highways". Some crazu people placed a bomb (NO JOKE!!!) at the main Telekom office in Koblenz, where we connect to the Internet, resulting in a total failure of all ISDN, cable TV, BTX and wide-area network services around Koblenz. I must say, this time the German Telecoma worked really fast. It took only half a day to setup sufficient backup lines and switching systems! > [SNIP] > Andreas> before: 300 seconds after: 26 seconds > Wouah,... So is the Magnum BIOS's read() broken or not ??? > It is. Definitely. Do a single fseek() and performance goes down the tubes :-) Cheers, Andy ----------------------------------------------------------- Andreas Busse | andy@waldorf-gmbh.de Waldorf Electronics GmbH, R&D Dep. | Phone: +49 2636-970105 Neustrasse 9-12, D-53498 Waldorf | Fax: +49 2636-970106 ----------------------------------------------------------- From imp@rover.village.org Fri Sep 8 04:51:25 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id EAA23762 for ; Fri, 8 Sep 1995 04:51:23 -0700 Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA10843; Fri, 8 Sep 1995 04:59:48 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id VAA06967 for ; Thu, 7 Sep 1995 21:00:09 -0600 Message-Id: <199509080300.VAA06967@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Milo Speedup In-Reply-To: Your message of Thu, 07 Sep 1995 14:11:54 +0200 Date: Thu, 07 Sep 1995 21:00:08 -0600 From: Warner Losh Content-Length: 1462 Lines: 30 : I got no comments -- doesn't matter :-) Hmmm, guess this might make some of my patches to milo (but not all) obsolete. :-(. Also, this assumes that you can load the kernel at the right place in physical memory. The location for the Deskstation rpc44 is different than other machines, I fear, due to the 640K low memory limit. I also think that it may be wasteful of space since it will keep the kernel symbols in memory if you aren't careful. How does the kernel get relocated to its right place? Are you copying it down from high memory where the malloc returns memory? Anyway, despite these misgivings (which are likely just misunderstandings on my part), I think this is a great idea! This would mean that we could put the data decompression into MILO itself with very little effort if we load the kernel high enough. The whole thing could be read in off disk (which should be even faster, I would imagine) and then decompressed into the right location. I'm not sure how that would impact MILO's load time, since it would have to be bigger by a small amount. I'd love to be able to have the same load address for all machines, but fear that we might not be able to do that easily. Well, we could make it 0x80010000 on all machines, and have Linux be smart enough to put either the low 640K or the low 1M of memory depending on the machine into its free pool of memory. This sounds like the most portible way to cope with the problem... Warner From andy@waldorf-gmbh.de Fri Sep 8 07:19:40 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id HAA25259 for ; Fri, 8 Sep 1995 07:19:39 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA13080; Fri, 8 Sep 1995 07:28:10 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id HAA07971; Fri, 8 Sep 1995 07:29:16 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id HAA01216; Fri, 8 Sep 1995 07:28:45 +0200 Date: Fri, 8 Sep 1995 07:28:45 +0200 From: Andreas Busse Message-Id: <199509080528.HAA01216@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Re: Milo Speedup In-Reply-To: <199509080300.VAA06967@rover.village.org> X-Mailer: [XMailTool v3.1.2b] Content-Length: 2354 Lines: 54 Hi Warner, > > Hmmm, guess this might make some of my patches to milo (but not all) > obsolete. :-(. > No, don't worry. My changes don't affect the basic way Milo works, it just avoids tons of seeks during kernel load. > Also, this assumes that you can load the kernel at the right place in > physical memory. The location for the Deskstation rpc44 is different > than other machines, I fear, due to the 640K low memory limit. I also > think that it may be wasteful of space since it will keep the kernel > symbols in memory if you aren't careful. How does the kernel get > relocated to its right place? Are you copying it down from high > memory where the malloc returns memory? Yes, the kernel including all symbols etc. is loaded to high memory first and then copied down to it's final location by using the same and unchanged functions as before. Useless kernel symbols are free'd as soon the kernel starts. They occupy memory only while Milo is running. Actually, as it was before. > > Anyway, despite these misgivings (which are likely just > misunderstandings on my part), I think this is a great idea! This > would mean that we could put the data decompression into MILO itself > with very little effort if we load the kernel high enough. The whole > thing could be read in off disk (which should be even faster, I would > imagine) and then decompressed into the right location. I'm not sure > how that would impact MILO's load time, since it would have to be > bigger by a small amount. You're right. The kernel could be compressed completely which certainly makes a difference when you have lots of symbols. BTW, for source level debugging I boot a mostly stripped kernel and use the unstripped (3 Meg) kernel only for GDB. I couldn't write it on a floppy anyway :-) > > I'd love to be able to have the same load address for all machines, > but fear that we might not be able to do that easily. Well, we could > make it 0x80010000 on all machines, and have Linux be smart enough to > put either the low 640K or the low 1M of memory depending on the > machine into its free pool of memory. This sounds like the most > portible way to cope with the problem... > Hmmm. That would work, but actually I don't like this unnecessary memory fragmentation. But if the other's agree... Cheers, Andy From imp@rover.village.org Fri Sep 8 07:29:26 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id HAA25364 for ; Fri, 8 Sep 1995 07:29:26 -0700 Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA13187; Fri, 8 Sep 1995 07:37:57 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id XAA09358 for ; Thu, 7 Sep 1995 23:38:19 -0600 Message-Id: <199509080538.XAA09358@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Milo Speedup In-Reply-To: Your message of Fri, 08 Sep 1995 07:28:45 +0200 Date: Thu, 07 Sep 1995 23:38:18 -0600 From: Warner Losh Content-Length: 1293 Lines: 25 : Yes, the kernel including all symbols etc. is loaded to high memory : first and then copied down to it's final location by using the : same and unchanged functions as before. Useless kernel symbols : are free'd as soon the kernel starts. They occupy memory only while : Milo is running. Actually, as it was before. Hmmm. So I did slightly misunderstand how things were being done. : Hmmm. That would work, but actually I don't like this unnecessary : memory fragmentation. But if the other's agree... I think that it wouldn't matter much at all. I believe that we could reduce the amount of fragmentation if we loaded the data and, if possible, the bss below the 640K mark. Also, as things start up, I would imagine that they would use up at least 1M of memory (and certainly 640K) so that wouldn't be a huge deal, I don't think. We have plenty of time to get to know this problem. There are a few places in the kernel that are #ifdef'd based on the machine type that is configured, which means we can't have a single, unified kernel for ARCBIOS machines at this stage anyway. For the moment, I'll just hack MILO to move the kernel to different places based on the machine type and to perform a basic sanity check to make sure that the kernel is LD'd to that address, etc. Warner From andy@waldorf-gmbh.de Fri Sep 8 08:25:04 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id IAA25954 for ; Fri, 8 Sep 1995 08:25:03 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA13885; Fri, 8 Sep 1995 08:33:35 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id IAA08875; Fri, 8 Sep 1995 08:34:41 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id IAA01313; Fri, 8 Sep 1995 08:34:09 +0200 Date: Fri, 8 Sep 1995 08:34:09 +0200 From: Andreas Busse Message-Id: <199509080634.IAA01313@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Re: Milo Speedup In-Reply-To: <199509080538.XAA09358@rover.village.org> X-Mailer: [XMailTool v3.1.2b] Content-Length: 1615 Lines: 38 Hi again! > > Hmmm. So I did slightly misunderstand how things were being done. > Doesn't matter :-) > : Hmmm. That would work, but actually I don't like this unnecessary > : memory fragmentation. But if the other's agree... > > I think that it wouldn't matter much at all. I believe that we could > reduce the amount of fragmentation if we loaded the data and, if > possible, the bss below the 640K mark. Also, as things start up, I > would imagine that they would use up at least 1M of memory (and > certainly 640K) so that wouldn't be a huge deal, I don't think. We > have plenty of time to get to know this problem. There are a few > places in the kernel that are #ifdef'd based on the machine type that > is configured, which means we can't have a single, unified kernel for > ARCBIOS machines at this stage anyway. For the moment, I'll just hack > MILO to move the kernel to different places based on the machine type > and to perform a basic sanity check to make sure that the kernel is > LD'd to that address, etc. > Ok. If you're ready with that, can you send me patches again? Meanwhile I'll try to build in your latest changes, and together with the above changes we could release Milo 0.26. Agreed ? Regarding kernel load address: The best way would of course be a PIC kernel, but I don't know how reliable PIC currently is. I would say, leave the "standard" kernel 0x80000000 for know, just load the RPC44 kernel to a different place. It doesn't hurt that much not to have a unified kernel for the moment. It's more important to have a unified loader :-) Cheers, Andy From imp@rover.village.org Fri Sep 8 08:35:30 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id IAA26059 for ; Fri, 8 Sep 1995 08:35:28 -0700 Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA14003; Fri, 8 Sep 1995 08:43:58 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id AAA09820 for ; Fri, 8 Sep 1995 00:44:13 -0600 Message-Id: <199509080644.AAA09820@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Milo Speedup In-Reply-To: Your message of Fri, 08 Sep 1995 08:34:09 +0200 Date: Fri, 08 Sep 1995 00:44:12 -0600 From: Warner Losh Content-Length: 1206 Lines: 29 : Ok. If you're ready with that, can you send me patches again? I've not had MILO do these things yet, so I can't send you the patches just yet :-). : Meanwhile I'll try to build in your latest changes, and together : with the above changes we could release Milo 0.26. Agreed ? Sounds good to me. I'll send you the patches as soon as I've compeleted them. I'll also try to send patches to the kernel because I think I'm to the load the RAM disk stage in my kernel booting, which sounds promising. It just hangs there for reasons I'm not sure of (likely looking for my non-existant hard disk :-). : Regarding kernel load address: The best way would of course : be a PIC kernel, but I don't know how reliable PIC currently is. I'd rather not do PIC just yet. I had problems with PIC and ELF when I tried to do that, so I'm a little leery of a PIC kernel. : I would say, leave the "standard" kernel 0x80000000 for know, : just load the RPC44 kernel to a different place. It doesn't hurt : that much not to have a unified kernel for the moment. It's more : important to have a unified loader :-) Sounds cool to me. Is there some way we can mark these so we can go back later and find them? Warner From andy@waldorf-gmbh.de Fri Sep 8 08:42:44 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id IAA26149 for ; Fri, 8 Sep 1995 08:42:43 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA14104; Fri, 8 Sep 1995 08:51:15 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id IAA09167; Fri, 8 Sep 1995 08:52:23 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id IAA01359; Fri, 8 Sep 1995 08:51:51 +0200 Date: Fri, 8 Sep 1995 08:51:51 +0200 From: Andreas Busse Message-Id: <199509080651.IAA01359@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Re: Milo Speedup In-Reply-To: <199509080644.AAA09820@rover.village.org> X-Mailer: [XMailTool v3.1.2b] Content-Length: 1190 Lines: 36 Hi again, > > I've not had MILO do these things yet, so I can't send you the patches > just yet :-). > I know :-) > : Meanwhile I'll try to build in your latest changes, and together > : with the above changes we could release Milo 0.26. Agreed ? > > Sounds good to me. I'll send you the patches as soon as I've > compeleted them. I'll also try to send patches to the kernel because > I think I'm to the load the RAM disk stage in my kernel booting, which > sounds promising. It just hangs there for reasons I'm not sure of > (likely looking for my non-existant hard disk :-). Hmm. IDE or SCSI enabled? Better turn it off. I'm sure it produces only trouble and headache :-) > : I would say, leave the "standard" kernel 0x80000000 for know, > : just load the RPC44 kernel to a different place. It doesn't hurt > : that much not to have a unified kernel for the moment. It's more > : important to have a unified loader :-) > > Sounds cool to me. Is there some way we can mark these so we can go > back later and find them? > Well, I can put it on the TODO list. As long nobody removes this item, we will be remembered at least once a week. Cheers, Andy From andy@waldorf-gmbh.de Fri Sep 8 09:03:19 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id JAA26396 for ; Fri, 8 Sep 1995 09:03:18 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA14442; Fri, 8 Sep 1995 09:11:49 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id JAA09438; Fri, 8 Sep 1995 09:12:57 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id JAA01393; Fri, 8 Sep 1995 09:12:04 +0200 Date: Fri, 8 Sep 1995 09:12:04 +0200 From: Andreas Busse Message-Id: <199509080712.JAA01393@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: move to 1.3 X-Mailer: [XMailTool v3.1.2b] Content-Length: 1343 Lines: 36 Hi all, the demand to move to the 1.3 kernel seems to increase. There are quite a few people on the list asking for this, and I must admit that it would have several advantages. However, I'd like to see user code running on all current targets, that is the Pica, Magnum, Olivetti and RPC44 boxes before moving over to 1.3. I believe that if we change before having all targets more or less at the same state we would produce more problems than needed. It might even happen that people are working on different kernel releases, which is certainly no good idea. So what needs to be done (I'll move that to the TODO list as soon we agreed over that) ? - add my patches #1...#5 to the current kernel - add Olivetti support to the kernel and Milo, where it's still missing. I don't know what is missing, but I'm convinced that there are many locations missing some "if ( ... || MACH_OLIVETTI_M700)" or similar lines... - add floppy driver support to the PICA kernel. - add RPC44 support to Milo and kernel - add Linux/i386 1.2.11 kernel patches - check user code on all systems (at least /bin/bash should work) - release Linux/MIPS kernel 1.2.11 I know, that's a bit oversimplified. However, I believe that we need to have a common kernel functionality before we can change to 1.3. Ok, tell me what you think and what I've forgotten. Andy From Stoned.Elipot@univ-evry.fr Fri Sep 8 11:29:15 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id LAA27862 for ; Fri, 8 Sep 1995 11:29:14 -0700 Received: from amiante.univ-evry.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA16992; Fri, 8 Sep 1995 11:37:24 +0200 (MET) Received: from bandsept.univ-evry.fr (bandsept.univ-evry.fr [192.93.214.19]) by amiante.univ-evry.fr with ESMTP (8.6.10/94.01.10); id LAA06123; Fri, 8 Sep 1995 11:37:47 +0200 Received: (seb@localhost) by bandsept.univ-evry.fr (8.6.12/94.01.26); id LAA07203; Fri, 8 Sep 1995 11:36:13 +0200 Message-Id: <199509080936.LAA07203@bandsept.univ-evry.fr> Date: Fri, 8 Sep 1995 11:36:13 +0200 From: Stoned Elipot To: linux-mips@fnet.fr In-Reply-To: <199509080712.JAA01393@newton.waldorf-gmbh.de> (message from Andreas Busse on Fri, 8 Sep 1995 09:12:04 +0200) Subject: Re: move to 1.3 X-Mailer: GNU Emacs [19.29] Mail Mode X-Attribution: seb X-Pgp-Key: ask a good public key server Reply-To: Stoned.Elipot@univ-evry.fr Content-Length: 658 Lines: 17 Hi all, >>>>> "Andreas" == Andreas Busse writes: [SNIP] Andreas> - add Olivetti support to the kernel and Milo, where it's Andreas> still missing. I don't know what is missing, but I'm Andreas> convinced that there are many locations missing some "if ( Andreas> ... || MACH_OLIVETTI_M700)" or similar lines... Hey, you're right, I'm sorry but I didn't find 1 hour in the last 3 weeks :( But what about those 2 German Oily boxes ?? :) Anyway, I'm pretty sure that with all the stuff you and Ralf had done we could have ash running: definitevely I'll take a whole day this week-end, and I will died or succed !! :) Cheers, stoned. From imp@rover.village.org Fri Sep 8 16:51:01 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id QAA01726 for ; Fri, 8 Sep 1995 16:51:00 -0700 Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA22742; Fri, 8 Sep 1995 16:59:29 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id IAA10533 for ; Fri, 8 Sep 1995 08:59:50 -0600 Message-Id: <199509081459.IAA10533@rover.village.org> To: linux-mips@fnet.fr Subject: Re: move to 1.3 In-Reply-To: Your message of Fri, 08 Sep 1995 09:12:04 +0200 Date: Fri, 08 Sep 1995 08:59:49 -0600 From: Warner Losh Content-Length: 1231 Lines: 33 : the demand to move to the 1.3 kernel seems to increase. : There are quite a few people on the list asking for this, : and I must admit that it would have several advantages. I'd love to be able to just run 1.3.x out of the box :-). : However, I'd like to see user code running on all current : targets, that is the Pica, Magnum, Olivetti and RPC44 boxes : before moving over to 1.3. I'm happy with where I'm at with the RPC44 box that if it didn't run user code, I wouldn't complain much. : I believe that if we change before having all targets : more or less at the same state we would produce more : problems than needed. It might even happen that people are : working on different kernel releases, which is certainly : no good idea. This is a valid concern. I'm not sure why I can't get past the kernel messages phase of the boot, but I'm not overly worried about it. : - add RPC44 support to Milo and kernel OK. I'll try to get a set of diffs against the 1.2.10.4 kernel that I've been using as a base, and we can iterate on them. : - check user code on all systems (at least /bin/bash should work) As an aside: What's the easiest way to check this? Can I just put /bin/bash on a msdos formatted floppy? Warner From imp@rover.village.org Fri Sep 8 16:54:14 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id QAA01776 for ; Fri, 8 Sep 1995 16:54:13 -0700 Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA22818; Fri, 8 Sep 1995 17:02:37 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id JAA10559 for ; Fri, 8 Sep 1995 09:02:56 -0600 Message-Id: <199509081502.JAA10559@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Milo Speedup In-Reply-To: Your message of Fri, 08 Sep 1995 08:51:51 +0200 Date: Fri, 08 Sep 1995 09:02:56 -0600 From: Warner Losh Content-Length: 777 Lines: 19 : Hmm. IDE or SCSI enabled? Better turn it off. I'm sure it produces : only trouble and headache :-) I think I may have left IDE enabled. However, I didn't think it would be a problem since I don't have a IDE controller in my machine. It should just detect that it isn't there and not do anything. : Well, I can put it on the TODO list. As long nobody removes this : item, we will be remembered at least once a week. It isn't the fact that we need a unified kernel, it is marking the places in the source as people add/notice them that I'm talking about here. Something like: /* XXX Fix for unified kernel */ or some stock phrase we can grep for. That way, when we do go to fix it, we have a higher confidence level that we've caught all the interesting parts. Warner From andy@waldorf-gmbh.de Sat Sep 9 18:35:13 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id SAA05514 for ; Sat, 9 Sep 1995 18:35:11 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA15448; Sat, 9 Sep 1995 18:43:40 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id SAA06783; Sat, 9 Sep 1995 18:44:47 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id SAA05063; Sat, 9 Sep 1995 18:44:12 +0200 Date: Sat, 9 Sep 1995 18:44:12 +0200 From: Andreas Busse Message-Id: <199509091644.SAA05063@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Waldorf changes connection X-Mailer: [XMailTool v3.1.2b] Content-Length: 701 Lines: 20 Hi all, on Monday, Sep. 11 Waldorf Electronics changes its Internet connection to a 14.4kBaud analogue leased line. I'm currently testing this link and expect a bit trouble with thruput. If you wanna download something, better do it now. Cheers, Andy P.S.: Sorry to those whose mail to me isn't answered yet. I was fighting with this f*cking line, modems and packet drivers since Thursday evening. ----------------------------------------------------------- Andreas Busse | andy@waldorf-gmbh.de Waldorf Electronics GmbH, R&D Dep. | Phone: +49 2636-970105 Neustrasse 9-12, D-53498 Waldorf | Fax: +49 2636-970106 ----------------------------------------------------------- From andy@waldorf-gmbh.de Sat Sep 9 18:40:15 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id SAA05576 for ; Sat, 9 Sep 1995 18:40:14 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA15502; Sat, 9 Sep 1995 18:48:46 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id SAA06839; Sat, 9 Sep 1995 18:49:55 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id SAA05075; Sat, 9 Sep 1995 18:49:17 +0200 Date: Sat, 9 Sep 1995 18:49:17 +0200 From: Andreas Busse Message-Id: <199509091649.SAA05075@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Re: move to 1.3 In-Reply-To: <199509080936.LAA07203@bandsept.univ-evry.fr> X-Mailer: [XMailTool v3.1.2b] Content-Length: 661 Lines: 19 Hi Stoned and all, > > Hey, you're right, I'm sorry but I didn't find 1 hour in the last 3 > weeks :( But what about those 2 German Oily boxes ?? :) Anyway, I'm > pretty sure that with all the stuff you and Ralf had done we could > have ash running: definitevely I'll take a whole day this week-end, > and I will died or succed !! :) > That was no offence, really! However, the both german Oily boxes seem to get dusty... Stefan (stenz@waldorf-gmbh.de) isn't a kernel hacker at all, and I guess Wayne (hodgen@uni-koblenz.de) still has no RAM for hix box. So don't expect too much help from there. Or am I wrong, Wayne and Stefan ? :-) Cheers, Andy From andy@waldorf-gmbh.de Sat Sep 9 18:45:23 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id SAA05644 for ; Sat, 9 Sep 1995 18:45:22 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA15582; Sat, 9 Sep 1995 18:53:55 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id SAA06939; Sat, 9 Sep 1995 18:55:03 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id SAA05088; Sat, 9 Sep 1995 18:54:27 +0200 Date: Sat, 9 Sep 1995 18:54:27 +0200 From: Andreas Busse Message-Id: <199509091654.SAA05088@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Re: move to 1.3 In-Reply-To: <199509081459.IAA10533@rover.village.org> X-Mailer: [XMailTool v3.1.2b] Content-Length: 955 Lines: 33 Hi again, > > I'd love to be able to just run 1.3.x out of the box :-). > I must admit that I'm still running 1.2.11 on my Intel boxes and never tried 1.3.x. > > This is a valid concern. I'm not sure why I can't get past the kernel > messages phase of the boot, but I'm not overly worried about it. > Where is it stopping ? Perhaps you have a problem with page_color() ? This macro needs to be changed depending on your Icache linesize, when I remember right. > > OK. I'll try to get a set of diffs against the 1.2.10.4 kernel that > I've been using as a base, and we can iterate on them. > > As an aside: What's the easiest way to check this? Can I just put > /bin/bash on a msdos formatted floppy? No, it should be on a ramdisk image, which itself must be on the boot floppy as long you don't have the floppy driver working. I believe there's such an image on ftp.waldorf-gmbh.de. Shall I check it for you ? Cheers, Andy From andy@waldorf-gmbh.de Sat Sep 9 19:15:27 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id TAA05958 for ; Sat, 9 Sep 1995 19:15:27 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA15916; Sat, 9 Sep 1995 19:23:53 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id TAA07425; Sat, 9 Sep 1995 19:24:56 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id TAA05127; Sat, 9 Sep 1995 19:24:06 +0200 Date: Sat, 9 Sep 1995 19:24:06 +0200 From: Andreas Busse Message-Id: <199509091724.TAA05127@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Re: Milo Speedup In-Reply-To: <199509081502.JAA10559@rover.village.org> X-Mailer: [XMailTool v3.1.2b] Content-Length: 874 Lines: 27 Hi again, > > I think I may have left IDE enabled. However, I didn't think it would > be a problem since I don't have a IDE controller in my machine. It > should just detect that it isn't there and not do anything. Actually yes. But who knows... > > : Well, I can put it on the TODO list. As long nobody removes this > : item, we will be remembered at least once a week. > > It isn't the fact that we need a unified kernel, it is marking the > places in the source as people add/notice them that I'm talking about > here. Something like: > /* XXX Fix for unified kernel */ > or some stock phrase we can grep for. That way, when we do go to fix > it, we have a higher confidence level that we've caught all the > interesting parts. > No problem. Mark the locations where you think changes will be necessary and we'll let the marks in. OK? Andy From ralf@waldorf-gmbh.de Sat Sep 9 19:44:21 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id TAA06267 for ; Sat, 9 Sep 1995 19:44:20 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA16194; Sat, 9 Sep 1995 19:52:48 +0200 (MET) From: Ralf Baechle Message-Id: <199509091753.TAA07904@scotty.waldorf-gmbh.de> Received: from localhost by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id TAA07904; Sat, 9 Sep 1995 19:53:59 +0200 Subject: Re: move to 1.3 To: linux-mips@fnet.fr Date: Sat, 9 Sep 1995 19:53:59 +0200 (MET DST) In-Reply-To: <199509091654.SAA05088@newton.waldorf-gmbh.de> from "Andreas Busse" at Sep 9, 95 06:54:27 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 844 Lines: 23 Hi again, > I must admit that I'm still running 1.2.11 on my Intel boxes > and never tried 1.3.x. I tried it yesterday on a company machine and have to admit that it *blows* 1.2.13 away. Well ignoring the reliability ... > Where is it stopping ? Perhaps you have a problem with page_color() ? > This macro needs to be changed depending on your Icache linesize, > when I remember right. Page_colour() needs to be changed according to the L1 cache size (and for R8000 / R10000 according to the changed caching policies). I've fixed this finally in my private source tree for the R4xx0 CPUs. This has nothing to do with the cache line sizes. > > /bin/bash on a msdos formatted floppy? Bash is a huge beast ... Consider that even the statically linked ash is about 200k. I suggest ash for now when your trying to use disks. Ralf From imp@rover.village.org Sat Sep 9 21:54:55 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id VAA07592 for ; Sat, 9 Sep 1995 21:54:54 -0700 Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA17444; Sat, 9 Sep 1995 22:03:18 +0200 (MET) Received: from LOCALHOST (LOCALHOST [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id OAA14688 for ; Sat, 9 Sep 1995 14:03:34 -0600 Message-Id: <199509092003.OAA14688@rover.village.org> To: linux-mips@fnet.fr Subject: Re: move to 1.3 In-Reply-To: Your message of Sat, 09 Sep 1995 18:54:27 +0200 Date: Sat, 09 Sep 1995 14:03:34 -0600 From: Warner Losh Content-Length: 2062 Lines: 46 : I must admit that I'm still running 1.2.11 on my Intel boxes : and never tried 1.3.x. Local Linux experts suggest that 1.3 is much faster than 1.2, at least that's what I recall from the rapid fire discussion that took place at pub-night a while ago. Not a big concern just yet. : > This is a valid concern. I'm not sure why I can't get past the kernel : > messages phase of the boot, but I'm not overly worried about it. : > : : Where is it stopping ? Perhaps you have a problem with page_color() ? : This macro needs to be changed depending on your Icache linesize, : when I remember right. I don't know. Haven't taken the time to track down this problem. I'm having difficulties generating a 1.2.10.4 tree to generate a patch from, mostly disk space problems. I do know that it gets to my last printk in floppy_init() and seems to return from it. I've not quantified it further at this time. : No, it should be on a ramdisk image, which itself must be on the : boot floppy as long you don't have the floppy driver working. I : believe there's such an image on ftp.waldorf-gmbh.de. Shall I : check it for you ? I do have a copy of that disk image, so you needn't do anything. I think I need to do the hole in memory and the load at a high address stuff first, since I'm right on the edge as it is with memory, and something like a ramdisk image will push me over the edge into video rom/ram :-(. Also, I'm starting to batch up patches, which I'm not sure that I like doing, so I'd like to clear my queue, lest something gets lost in the shuffle. My game plan: 1) Iterate on patches to MILO and send them off to Andy as part of 0.26. 2) Get the proper mappings working so I don't have to compile with as many non-standard hacks (this is TLB stuff) 3) Generate Deskstation patches to date for the Linux kernel 4) Make Linux and MILO grok loading LINUX at 0x80010000 and have Linux copy down the interrupt handlers, etc. 5) Make MILO grok compressed kernels 6) Start worrying about RAMDISKS and/or SCSI disks, etc. Warner From phil%zipmail.co.uk%kiss.demon.co.uk%wbsmail.zipmail.co.uk@post.demon.co.uk Sun Sep 10 01:31:25 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id BAA09804 for ; Sun, 10 Sep 1995 01:31:24 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA22111; Sun, 10 Sep 1995 01:39:57 +0200 (MET) Date: Sun, 10 Sep 1995 01:41:07 +0200 Received: from disperse.demon.co.uk by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id BAA14087; Sun, 10 Sep 1995 01:41:07 +0200 Received: from post.demon.co.uk by disperse.demon.co.uk id aa20976; 9 Sep 95 23:13 +0100 Received: from wbsmail.zipmail.co.uk by post.demon.co.uk id aa25399; 9 Sep 95 23:11 +0100 Received: from kiss.demon.co.uk by wbsmail.zipmail.co.uk id aa06369; 9 Sep 95 23:13 BST X-Sender: phil@wbsmail.zipmail.co.uk (Unverified) X-Mailer: Windows Eudora Version 1.4.4 Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" To: linux@waldorf-gmbh.de From: Phil Taylor Subject: Linux MIPS Message-Id: <9509092313.aa06369@wbsmail.zipmail.co.uk> Content-Length: 701 Lines: 15 Just a quick note to see how the MIPS ports are going, I am using an Olivetti M700-10, if i can be of any help with testing etc please let me know.... Thanks, ------------------------------------- Tear Here ------------------------------------- Phil Taylor Lan Systems phil@zipmail.co.uk 150 London Road, Leicester ENGLAND, LE2 1AN Network & Internet Consultancy Tel : +44 (0)116 255 9961 Software & Hardware Fax : +44 (0)116 255 8861 " Just because I'm Paranoid, Doesn't mean the world isn't out to get me " From andy@waldorf-gmbh.de Sun Sep 10 08:59:36 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id IAA15695 for ; Sun, 10 Sep 1995 08:59:35 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA27182; Sun, 10 Sep 1995 09:08:06 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id JAA20368; Sun, 10 Sep 1995 09:09:04 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id JAA07123; Sun, 10 Sep 1995 09:08:14 +0200 Date: Sun, 10 Sep 1995 09:08:14 +0200 From: Andreas Busse Message-Id: <199509100708.JAA07123@newton.waldorf-gmbh.de> To: Phil Taylor Subject: Re: Linux MIPS In-Reply-To: <9509092313.aa06369@wbsmail.zipmail.co.uk> Cc: linux-mips@fnet.fr X-Mailer: [XMailTool v3.1.2b] Content-Length: 1024 Lines: 31 Hi Phil! > > Just a quick note to see how the MIPS ports are going, > > I am using an Olivetti M700-10, if i can be of any help with testing etc > please let me know.... > Sure! Do you think of testing only, or would you also be able to spend some time into kernel hacking? The problem is that although the Olivetti box belongs to our "best supported" architecture (if one can say so at this moment) its kernel lacks of several small things that could be easily done and tested. Just let me know! I also guess that you want to subscribe to our mailing list. Please let me know if I shall put you on. BTW, What kind of OS are you running on that box? WinNT or perhaps Mips RiscOS ? Thanks for your offer, Andy ----------------------------------------------------------- Andreas Busse | andy@waldorf-gmbh.de Waldorf Electronics GmbH, R&D Dep. | Phone: +49 2636-970105 Neustrasse 9-12, D-53498 Waldorf | Fax: +49 2636-970106 ----------------------------------------------------------- From ralf@waldorf-gmbh.de Sun Sep 10 13:45:49 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id NAA18614 for ; Sun, 10 Sep 1995 13:45:47 -0700 Received: from rio.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA00245; Sun, 10 Sep 1995 13:54:18 +0200 (MET) From: ralf@waldorf-gmbh.de (Ralf Baechle) Message-Id: <199509101134.NAA00153@rio.waldorf-gmbh.de> Received: by rio.waldorf-gmbh.de (8.6.11/GEN-1.0.8) id NAA00153; Sun, 10 Sep 1995 13:34:27 +0200 Subject: GUUG exhibition To: linux-mips@fnet.fr Date: Sun, 10 Sep 1995 13:34:24 +0200 (MET DST) Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 242 Lines: 8 Hi all, I and maybe Andy too, will be on the GUUG exhibition in Wiesbaden, Germany on the next Wednesday, September 13. Would be great if we could meet there. So if someone is there just drop an email so that we can make a date. Ralf From ralf@waldorf-gmbh.de Sun Sep 10 20:58:31 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id UAA23054 for ; Sun, 10 Sep 1995 20:58:30 -0700 Received: from rio.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA05297; Sun, 10 Sep 1995 21:07:02 +0200 (MET) From: ralf@waldorf-gmbh.de (Ralf Baechle) Message-Id: <199509101832.UAA03584@rio.waldorf-gmbh.de> Received: by rio.waldorf-gmbh.de (8.6.11/GEN-1.0.8) id UAA03584; Sun, 10 Sep 1995 20:32:36 +0200 Subject: Remote debugger To: linux-mips@fnet.fr Date: Sun, 10 Sep 1995 20:32:34 +0200 (MET DST) Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 399 Lines: 11 Hi, I just tried to debug a kernel using Andy's remote debugging support. The bad new is that debugging info needs that much memory that it's definately unusable for small machines. When linking on my 486DX/33 still wasn't finished after over 10 minutes I gave up. Ps showed for that kernel with lots of options turned on about 12MB memory in use. Printk() will stay my best friend ... Ralf From pefo@enea.se Sun Sep 10 22:43:59 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id WAA24171 for ; Sun, 10 Sep 1995 22:43:59 -0700 Received: from outside.enea.se by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA06429; Sun, 10 Sep 1995 22:52:33 +0200 (MET) Received: by outside.enea.se (8.6.9/8.6.9) id WAA01249 for ; Sun, 10 Sep 1995 22:54:10 +0200 Message-Id: <199509102054.WAA01249@outside.enea.se> Received: by outside via smap (V1.3) id sma001247; Sun Sep 10 22:53:54 1995 Received: by enea.se (4.1/SMI-4.0) id AA12692; Sun, 10 Sep 95 22:51:54 +0200 From: pefo@enea.se (Per Fogelstrom) Subject: Re: Remote debugger To: linux-mips@fnet.fr Date: Sun, 10 Sep 1995 22:51:54 +0200 (MET DST) In-Reply-To: <199509101832.UAA03584@rio.waldorf-gmbh.de> from "Ralf Baechle" at Sep 10, 95 08:32:34 pm X-Mailer: ELM [version 2.4 PL23] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 732 Lines: 20 > > Hi, > > I just tried to debug a kernel using Andy's remote debugging support. > The bad new is that debugging info needs that much memory that it's > definately unusable for small machines. When linking on my 486DX/33 > still wasn't finished after over 10 minutes I gave up. Ps showed for > that kernel with lots of options turned on about 12MB memory in use. > > Printk() will stay my best friend ... > > You should really consider getting my debugstuff to work..... :-) I've been hacking kernels for more than 15 years now (maby more) and i have never missed the features in gdb. Kernel bugs, often impossible to find just looking in the source, but doing a few single steps in the machine code..... Just a thought.. From imp@rover.village.org Mon Sep 11 00:31:58 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id AAA25269 for ; Mon, 11 Sep 1995 00:31:56 -0700 Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA09426; Mon, 11 Sep 1995 00:40:28 +0200 (MET) Received: from LOCALHOST (LOCALHOST [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id QAA16434 for ; Sun, 10 Sep 1995 16:40:49 -0600 Message-Id: <199509102240.QAA16434@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Remote debugger In-Reply-To: Your message of Sun, 10 Sep 1995 20:32:34 +0200 Date: Sun, 10 Sep 1995 16:40:48 -0600 From: Warner Losh Content-Length: 950 Lines: 20 : I just tried to debug a kernel using Andy's remote debugging support. : The bad new is that debugging info needs that much memory that it's : definately unusable for small machines. When linking on my 486DX/33 : still wasn't finished after over 10 minutes I gave up. Ps showed for : that kernel with lots of options turned on about 12MB memory in use. How much memory are we talking about here? I have 32M on my machine and was wondering if that was enough... Also, if kgdb is like gdb, you needn't compile the entire kernel -g. We had a similar problem with OI (-g == 30-80M depending on the compiler). When we needed to debug an entry field, we'd just compile entfld.C -g, and the rest of the library w/o -g. This seemed to work well in practice. Sure, you'd get hip deep into the debugging session and realize you need dtech1.C also, but usually with a little planning these sorts of things didn't matter much. Just a though... Warner From andy@waldorf-gmbh.de Mon Sep 11 07:24:20 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id HAA00917 for ; Mon, 11 Sep 1995 07:24:19 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA14761; Mon, 11 Sep 1995 07:32:51 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id HAA08288; Mon, 11 Sep 1995 07:33:59 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id HAA10048; Mon, 11 Sep 1995 07:33:22 +0200 Date: Mon, 11 Sep 1995 07:33:22 +0200 From: Andreas Busse Message-Id: <199509110533.HAA10048@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Re: Remote debugger In-Reply-To: <199509102240.QAA16434@rover.village.org> X-Mailer: [XMailTool v3.1.2b] Content-Length: 919 Lines: 22 > > How much memory are we talking about here? I have 32M on my machine > and was wondering if that was enough... > > Also, if kgdb is like gdb, you needn't compile the entire kernel -g. > We had a similar problem with OI (-g == 30-80M depending on the > compiler). When we needed to debug an entry field, we'd just compile > entfld.C -g, and the rest of the library w/o -g. This seemed to work > well in practice. Sure, you'd get hip deep into the debugging session > and realize you need dtech1.C also, but usually with a little planning > these sorts of things didn't matter much. > I have a 32Meg machine too and that's perfect. However, Warner is right in what he's saying. You don't need to turn on -g for the complete kernel tree, if you don't like that. For example, if you expect bugs in the drivers, add -g to the driver's CFLAGS, and nowhere else. I've tried that, and it works. Andy From andy@waldorf-gmbh.de Mon Sep 11 07:28:10 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id HAA00966 for ; Mon, 11 Sep 1995 07:28:09 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA14791; Mon, 11 Sep 1995 07:36:40 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id HAA08340; Mon, 11 Sep 1995 07:37:41 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id HAA10059; Mon, 11 Sep 1995 07:37:07 +0200 Date: Mon, 11 Sep 1995 07:37:07 +0200 From: Andreas Busse Message-Id: <199509110537.HAA10059@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Re: Remote debugger In-Reply-To: <199509101832.UAA03584@rio.waldorf-gmbh.de> X-Mailer: [XMailTool v3.1.2b] Content-Length: 750 Lines: 19 > > I just tried to debug a kernel using Andy's remote debugging support. > The bad new is that debugging info needs that much memory that it's > definately unusable for small machines. When linking on my 486DX/33 > still wasn't finished after over 10 minutes I gave up. Ps showed for > that kernel with lots of options turned on about 12MB memory in use. > > Printk() will stay my best friend ... > Well, as suggested, try to add -g to the parts where you expect the bug and turn it off anywhere else. Even if you still prefer printk() then I'd like to know if it works. Have you checked the serial lines before? They need to work, although they can't be used for anything else but debugging as soon CONFIG_REMOTE_DEBUG is on. Andy From andy@waldorf-gmbh.de Mon Sep 11 07:35:28 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id HAA01044 for ; Mon, 11 Sep 1995 07:35:27 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA14870; Mon, 11 Sep 1995 07:43:55 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id HAA08453; Mon, 11 Sep 1995 07:44:57 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id HAA10074; Mon, 11 Sep 1995 07:44:16 +0200 Date: Mon, 11 Sep 1995 07:44:16 +0200 From: Andreas Busse Message-Id: <199509110544.HAA10074@newton.waldorf-gmbh.de> To: phil@zipmail.co.uk, Andreas Busse Subject: Re: Linux MIPS Cc: linux-mips@fnet.fr X-Mailer: [XMailTool v3.1.2b] Content-Length: 1881 Lines: 58 Hi Phil! > > If I can be of any help kernel hacking let me know, although I have done > quite a bit of windoze/dos C programming, some unix admin stuff and a few > bug fixes in freebsd I have never done any kernel hacking (good time to > start I suppose !!!) > Sure :-) I wasn't a kernel hacker too before I simply couldn't resist to work on it... > Please let me know what I can do !!!!! > > > >I also guess that you want to subscribe to our mailing list. > >Please let me know if I shall put you on. > > yes, please do put me on the mailing list. > > OK. I'll subscribe you. The list's address is . What you can and should to then is to ask Stoned Elipot if and how you can help him with the Olivetti support. Also, if there are any questions, please feel free to post on the list. > >BTW, What kind of OS are you running on that box? WinNT or > >perhaps Mips RiscOS ? > > At the moment, nothing !! It is a bare machine that I picked up quite cheaply. > Ah, guess you got it from Applied Personal Computing ? > I was going to install WinNT but I would rather put a unix variant on it, I > thought about port PMAX for NETBSD but as It does not yet have native > binaries and does not directly support the Olivetti I thought I would see > how the Linux Port is doing > > I know that this may not be very popular but my main unix machine is > currently running FreeBSD (sorry) any chance of a cross compiler ???? > Well, we have crosscompiler patches and binaries available. It shouldn't be a problem to make mipsel-linux-gcc running on a NetBSD box. > I realise I ought to install i386/Linux really though (g) > Well, it would be useful, of course. But I don't mind if you continue to run NetBSD. No religions on this list :-) Well... Welcome to Linux/MIPS! Have fun! Cheers, Andy From marius@prilnari.com Mon Sep 11 07:52:09 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id HAA01293 for ; Mon, 11 Sep 1995 07:52:09 -0700 Received: from yakko.prilnari.com by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA15175; Mon, 11 Sep 1995 08:00:39 +0200 (MET) Received: (from marius@localhost) by yakko.prilnari.com (8.6.12/8.6.12) id CAA00161; Mon, 11 Sep 1995 02:05:32 -0400 Date: Mon, 11 Sep 1995 02:05:31 -0400 From: Matt Messier Subject: DECstation news To: linux-mips@fnet.fr Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 680 Lines: 15 I'm all excited right now. Its not really a major step, but I'm damned proud of myself :-) For those of you interested in the DECstation (pmax) port, I've apparently successfully figured out the structure of boot information. I've hacked in some code for the i386 kernel to read the DECstation partition tables and all of that fun stuff. So now I have the ability to mount the DECstation drive. There is a ufs package available somewhere that supports ufs (ro). I've mounted and can look around. By the way, has anyone else been working on the DS stuff at all? If so, I'd like to find out how far along you are. Matt Messier The Prilnari Project Lincoln, RI, USA From andy@waldorf-gmbh.de Mon Sep 11 08:21:54 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id IAA01607 for ; Mon, 11 Sep 1995 08:21:54 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA15565; Mon, 11 Sep 1995 08:30:27 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id IAA09117; Mon, 11 Sep 1995 08:31:32 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id IAA10146; Mon, 11 Sep 1995 08:30:55 +0200 Date: Mon, 11 Sep 1995 08:30:55 +0200 From: Andreas Busse Message-Id: <199509110630.IAA10146@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Re: DECstation news In-Reply-To: X-Mailer: [XMailTool v3.1.2b] Content-Length: 983 Lines: 26 Hi! > > I'm all excited right now. Its not really a major step, but I'm damned > proud of myself :-) For those of you interested in the DECstation (pmax) > port, I've apparently successfully figured out the structure of boot > information. I've hacked in some code for the i386 kernel to read the > DECstation partition tables and all of that fun stuff. So now I have the > ability to mount the DECstation drive. There is a ufs package available > somewhere that supports ufs (ro). I've mounted and can look around. > Congrats! Good to hear that there's some movement in the DEC department of Linux/MIPS :-) > By the way, has anyone else been working on the DS stuff at all? If so, > I'd like to find out how far along you are. So far I know there's a halfway working network loader, and when I remember right, there was also some progress with disk booting. Anyway, I would also be interested in the current state of this part of the project. Andy From marius@prilnari.com Mon Sep 11 08:35:00 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id IAA01740 for ; Mon, 11 Sep 1995 08:34:59 -0700 Received: from yakko.prilnari.com by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA15762; Mon, 11 Sep 1995 08:43:28 +0200 (MET) Received: (from marius@localhost) by yakko.prilnari.com (8.6.12/8.6.12) id CAA00425; Mon, 11 Sep 1995 02:43:52 -0400 Date: Mon, 11 Sep 1995 02:43:52 -0400 From: Matt Messier Subject: Re: DECstation news To: linux-mips@fnet.fr In-Reply-To: <199509110630.IAA10146@newton.waldorf-gmbh.de> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 1411 Lines: 32 On Mon, 11 Sep 1995, Andreas Busse wrote: > Congrats! Good to hear that there's some movement in the DEC department > of Linux/MIPS :-) You know, I often wonder if I'm supposed to be working on this project. If it isn't one thing, it is another. I never seem to be able to work on it at all. Right now I'm short a keyboard. I will hopefully be able to get down to the store tommorrow :-) I've been putting it off for over a month now, but but I have newly found inspiration. > > By the way, has anyone else been working on the DS stuff at all? If so, > > I'd like to find out how far along you are. > > So far I know there's a halfway working network loader, and when > I remember right, there was also some progress with disk booting. > Anyway, I would also be interested in the current state of this > part of the project. Great. For the time being I will be using Ultrix' boot loader to play around with kernels simply because the only means I have of getting anything onto the drive really is via ethernet. If I lose the Ultrix boot loader, I'm all done. The more I'm finding out, the less difficult it looks like the boot loader is going to be to write. By the way, last I knew, there wasn't actually anything for the boot loader written yet, but it had been talked about, much the talk being mine. With any luck, this has changed :-) Matt Messier The Prilnari Project Lincoln, RI, USA From andy@waldorf-gmbh.de Mon Sep 11 08:46:04 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id IAA01872 for ; Mon, 11 Sep 1995 08:46:04 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA15975; Mon, 11 Sep 1995 08:54:34 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id IAA09628; Mon, 11 Sep 1995 08:55:40 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id IAA10216; Mon, 11 Sep 1995 08:55:05 +0200 Date: Mon, 11 Sep 1995 08:55:05 +0200 From: Andreas Busse Message-Id: <199509110655.IAA10216@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Re: DECstation news In-Reply-To: X-Mailer: [XMailTool v3.1.2b] Content-Length: 1343 Lines: 33 > > You know, I often wonder if I'm supposed to be working on this project. If > it isn't one thing, it is another. I never seem to be able to work on it at > all. Right now I'm short a keyboard. I will hopefully be able to get down > to the store tommorrow :-) I've been putting it off for over a month now, but > but I have newly found inspiration. > Actually, it's the same for me. But I feel guilty if I stop working on Linux/MIPS for more than a week, so I give me a kick and feel good when I've done something :-) > > Great. For the time being I will be using Ultrix' boot loader to play around > with kernels simply because the only means I have of getting anything onto > the drive really is via ethernet. If I lose the Ultrix boot loader, I'm all > done. The more I'm finding out, the less difficult it looks like the boot > loader is going to be to write. > > By the way, last I knew, there wasn't actually anything for the boot loader > written yet, but it had been talked about, much the talk being mine. With > any luck, this has changed :-) > Yes, you're probably right. There was so much talk that I might have mixed up something. A summarize of the state of the DEC port would be really useful :-) BTW, would it be hard to add write support to the UFS filesystem ? Can anyone help? Andy From hodgen@mailhost.uni-koblenz.de Mon Sep 11 09:53:57 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id JAA02567 for ; Mon, 11 Sep 1995 09:53:56 -0700 Received: from infko.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA17194; Mon, 11 Sep 1995 10:02:16 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by infko.uni-koblenz.de (8.6.9/8.6.9) with SMTP id KAA06074 for ; Mon, 11 Sep 1995 10:03:45 +0200 From: Wayne Hodgen Message-Id: <199509110803.KAA06074@infko.uni-koblenz.de> Received: by ozzy (5.x/KO-2.0) id AA14176; Mon, 11 Sep 1995 10:01:47 +0200 Subject: German Ollis To: linux-mips@guadalquivir.fnet.fr Date: Mon, 11 Sep 1995 10:01:46 +0200 (MET DST) Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1249 Lines: 27 Hi, > > weeks :( But what about those 2 German Oily boxes ?? :) Anyway, I'm > > pretty sure that with all the stuff you and Ralf had done we could > > have ash running: definitevely I'll take a whole day this week-end, > > and I will died or succed !! :) > > > > That was no offence, really! However, the both german Oily boxes > seem to get dusty... Stefan (stenz@waldorf-gmbh.de) isn't a kernel > hacker at all, and I guess Wayne (hodgen@uni-koblenz.de) still has > no RAM for hix box. So don't expect too much help from there. > Or am I wrong, Wayne and Stefan ? :-) Unfortunately not wrong, the story of my RAM has become a farce of biblical proportions! 2 weeks ago we sent them an ultimatum, correct RAM or money our money back. They were so immpressed, they went on holiday :( So, unless I get my stuff this week, does anyone know a reliable source for baseball bats or other heavy club-like weapons? ;) Wayne. -- Wayne Hodgen | hodgen@informatik.uni-koblenz.de | #include From marius@prilnari.com Mon Sep 11 13:45:59 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id NAA06198 for ; Mon, 11 Sep 1995 13:45:58 -0700 Received: from yakko.prilnari.com by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA21346; Mon, 11 Sep 1995 13:54:28 +0200 (MET) Received: (from marius@localhost) by yakko.prilnari.com (8.6.12/8.6.12) id HAA00692; Mon, 11 Sep 1995 07:54:52 -0400 Date: Mon, 11 Sep 1995 07:54:52 -0400 From: Matt Messier Subject: Re: DECstation news To: linux-mips@fnet.fr In-Reply-To: <199509110655.IAA10216@newton.waldorf-gmbh.de> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 629 Lines: 15 On Mon, 11 Sep 1995, Andreas Busse wrote: > BTW, would it be hard to add write support to the UFS filesystem ? > Can anyone help? There is code that exists now to read a UFS file system. More specifically it works with Berkeley's Fast File System I believe it is. Both are horrible filesystems so I rarely bother to differentiate :-) In any case, Yoshi Gottleib(sp?) wrote the code for reading. I sent him an email last night to see if he's done anything with writing at all. If not, I'll probably end up doing it, if for no other reason than to appease myself. *smile* Matt Messier The Prilnari Project Lincoln, RI, USA From davem@caip.rutgers.edu Mon Sep 11 15:25:31 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id PAA07228 for ; Mon, 11 Sep 1995 15:25:30 -0700 Received: from caipfs.rutgers.edu by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA23123; Mon, 11 Sep 1995 15:33:57 +0200 (MET) Received: from huahaga.rutgers.edu (huahaga.rutgers.edu [128.6.155.53]) by caipfs.rutgers.edu (8.6.9+bestmx+oldruq+newsunq+grosshack/8.6.9) with ESMTP id JAA13727 for ; Mon, 11 Sep 1995 09:34:07 -0400 Received: (davem@localhost) by huahaga.rutgers.edu (8.6.9+bestmx+oldruq+newsunq+grosshack/8.6.9) id JAA12094; Mon, 11 Sep 1995 09:34:06 -0400 Date: Mon, 11 Sep 1995 09:34:06 -0400 Message-Id: <199509111334.JAA12094@huahaga.rutgers.edu> From: "David S. Miller" To: linux-mips@fnet.fr In-Reply-To: <199509110655.IAA10216@newton.waldorf-gmbh.de> (message from Andreas Busse on Mon, 11 Sep 1995 08:55:05 +0200) Subject: Re: DECstation news Content-Length: 655 Lines: 17 Date: Mon, 11 Sep 1995 08:55:05 +0200 From: Andreas Busse BTW, would it be hard to add write support to the UFS filesystem ? Can anyone help? Note that I have somebody working on ufs code for my Sparc port. Linus needs it to mount osf native partitions also. Write support is the hardest part, however my developer has started from scratch. He is writing the code the right way, he placed a Solaris UFS partition on a floppy, copies the filesystem into a file and emacs's it to learn how it "really works". I can hook somebody up with him if collaboration is desired. Later, David S. Miller davem@caip.rutgers.edu From imp@rover.village.org Mon Sep 11 16:39:52 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id QAA07992 for ; Mon, 11 Sep 1995 16:39:51 -0700 Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA24755; Mon, 11 Sep 1995 16:48:19 +0200 (MET) Received: from LOCALHOST (LOCALHOST [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id IAA17128; Mon, 11 Sep 1995 08:48:33 -0600 Message-Id: <199509111448.IAA17128@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Linux MIPS Cc: phil@zipmail.co.uk, Andreas Busse In-Reply-To: Your message of Mon, 11 Sep 1995 07:44:16 +0200 Date: Mon, 11 Sep 1995 08:48:32 -0600 From: Warner Losh Content-Length: 247 Lines: 7 > I know that this may not be very popular but my main unix machine is > currently running FreeBSD (sorry) any chance of a cross compiler ???? That's what I run. Cross compiler isn't hard to build, or I could get tar balls out the door. Warner From tiv@vma.bme.hu Mon Sep 11 18:16:02 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id SAA08986 for ; Mon, 11 Sep 1995 18:16:01 -0700 Received: from nic.bme.hu by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA26773; Mon, 11 Sep 1995 18:24:31 +0200 (MET) Received: from euromath.vma.bme.hu (euromath.vma.bme.hu [152.66.83.1]) by nic.bme.hu (8.6.10/8.6.10) with ESMTP id SAA12561 for ; Mon, 11 Sep 1995 18:24:53 +0200 Received: (from tiv@localhost) by euromath.vma.bme.hu (8.6.10/8.6.9) id SAA22314 for linux-mips@fnet.fr; Mon, 11 Sep 1995 18:24:51 +0200 From: Szemethy Tivadar Message-Id: <199509111624.SAA22314@euromath.vma.bme.hu> Subject: Re: DECstation news To: linux-mips@fnet.fr Date: Mon, 11 Sep 1995 18:24:50 +0200 (MET DST) In-Reply-To: from "Matt Messier" at Sep 11, 95 02:05:31 am X-Mailer: ELM [version 2.4 PL22] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 438 Lines: 15 hello, Sorry, I'm awfully busy these days. What I'm going to do in the DS section: - improve the netboot loader (almost ready) - write the SCSI disk (aka 'rz') bootblocks - we'll see what comes next. For the first time, the boot loaders will be able to load mipsel 407 executables (from ufs in the 'rz' case), but they should be aware of the ELF/ext2fs combination too. I'll show up when I'll have something to show... tivadar From marius@prilnari.com Mon Sep 11 18:29:49 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id SAA09119 for ; Mon, 11 Sep 1995 18:29:49 -0700 Received: from yakko.prilnari.com by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA27010; Mon, 11 Sep 1995 18:38:14 +0200 (MET) Received: (from marius@localhost) by yakko.prilnari.com (8.6.12/8.6.12) id MAA01024; Mon, 11 Sep 1995 12:38:36 -0400 Date: Mon, 11 Sep 1995 12:38:36 -0400 From: Matt Messier Subject: Re: DECstation news To: linux-mips@fnet.fr In-Reply-To: <199509111624.SAA22314@euromath.vma.bme.hu> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 1248 Lines: 28 On Mon, 11 Sep 1995, Szemethy Tivadar wrote: > What I'm going to do in the DS section: > - improve the netboot loader (almost ready) > - write the SCSI disk (aka 'rz') bootblocks > - we'll see what comes next. > > For the first time, the boot loaders will be able to load mipsel 407 executables > (from ufs in the 'rz' case), but they should be aware of the ELF/ext2fs > combination too. The boot blocks and disk label is trivial. What is really going to be a pain is writing the boot loader module that gets stuffed into the boot blocks. I've more or less got the code to write these pieces already. It might also be a better idea to use xiafs or ext2fs as opposed to ufs since these filesystems already exist for Linux, whereas ufs does not. It is the boot loader that really needs to be able to recognize what type of file system it is. If we use the Ultrix bootloader (as I was planning to do for my work), we more or less have to use ufs. As it turns out, I found a good price on a 240mb drive so I'll probably pick that up and work on the bootloader. In all honesty, I don't think that it will be too terribly long before we can actually start working on the kernel itself. Matt Messier The Prilnari Project Lincoln, RI, USA From ralf@waldorf-gmbh.de Thu Sep 14 23:17:49 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id XAA04629 for ; Thu, 14 Sep 1995 23:17:49 -0700 Received: from rio.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA18498; Thu, 14 Sep 1995 23:26:20 +0200 (MET) From: ralf@waldorf-gmbh.de (Ralf Baechle) Message-Id: <199509120041.CAA00858@rio.waldorf-gmbh.de> Received: by rio.waldorf-gmbh.de (8.6.11/GEN-1.0.8) id CAA00858; Tue, 12 Sep 1995 02:41:02 +0200 Subject: Re: DECstation news To: linux-mips@fnet.fr Date: Tue, 12 Sep 1995 02:41:01 +0200 (MET DST) In-Reply-To: <199509111624.SAA22314@euromath.vma.bme.hu> from "Szemethy Tivadar" at Sep 11, 95 06:24:50 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 449 Lines: 12 hello, > For the first time, the boot loaders will be able to load mipsel 407 executables > (from ufs in the 'rz' case), but they should be aware of the ELF/ext2fs > combination too. If that makes live easier for you - the ELF linker has no problem in linking ELF files together and then producing an a.out file. This is ugly and has several disadvantages but could make live easier as long as we don't have real ELF loader code ready. Ralf From ralf@waldorf-gmbh.de Thu Sep 14 23:17:31 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id XAA04603 for ; Thu, 14 Sep 1995 23:17:31 -0700 Received: from rio.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA18486; Thu, 14 Sep 1995 23:25:59 +0200 (MET) From: ralf@waldorf-gmbh.de (Ralf Baechle) Message-Id: <199509120048.CAA00873@rio.waldorf-gmbh.de> Received: by rio.waldorf-gmbh.de (8.6.11/GEN-1.0.8) id CAA00873; Tue, 12 Sep 1995 02:48:35 +0200 Subject: Re: German Ollis To: linux-mips@fnet.fr Date: Tue, 12 Sep 1995 02:48:33 +0200 (MET DST) In-Reply-To: <199509110803.KAA06074@infko.uni-koblenz.de> from "Wayne Hodgen" at Sep 11, 95 10:01:46 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 209 Lines: 8 Hi, > So, unless I get my stuff this week, does anyone know a reliable source > for baseball bats or other heavy club-like weapons? ;) Remember that you're the BOFH of the university in Koblenz ;-) Ralf From ralf@waldorf-gmbh.de Thu Sep 14 23:17:44 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id XAA04616 for ; Thu, 14 Sep 1995 23:17:44 -0700 Received: from rio.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA18495; Thu, 14 Sep 1995 23:26:10 +0200 (MET) From: ralf@waldorf-gmbh.de (Ralf Baechle) Message-Id: <199509120054.CAA00882@rio.waldorf-gmbh.de> Received: by rio.waldorf-gmbh.de (8.6.11/GEN-1.0.8) id CAA00882; Tue, 12 Sep 1995 02:54:13 +0200 Subject: Re: DECstation news To: linux-mips@fnet.fr Date: Tue, 12 Sep 1995 02:54:12 +0200 (MET DST) In-Reply-To: <199509110655.IAA10216@newton.waldorf-gmbh.de> from "Andreas Busse" at Sep 11, 95 08:55:05 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 409 Lines: 12 Hi, > BTW, would it be hard to add write support to the UFS filesystem ? > Can anyone help? What exist is read only, not absoloutly reliable and not tuned for performance. If I recall right UFS is more complex than Ext2fs and I therefore think doing the thing *really*'d be a big challenge. Nevertheless worth to be done as lots of people are waiting for religion or compatibility reasons for it. Ralf From andy@waldorf-gmbh.de Tue Sep 12 12:38:33 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id MAA00551 for ; Tue, 12 Sep 1995 12:38:18 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA16618; Tue, 12 Sep 1995 12:46:50 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id MAA04126; Tue, 12 Sep 1995 12:48:05 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id MAA13921; Tue, 12 Sep 1995 12:47:26 +0200 Date: Tue, 12 Sep 1995 12:47:26 +0200 From: Andreas Busse Message-Id: <199509121047.MAA13921@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Milo patches ? X-Mailer: [XMailTool v3.1.2b] Content-Length: 111 Lines: 8 Hi all, I'm working on a 0.26 release of Milo. If someone has patches, please send them to me. Thanks, Andy From imp@rover.village.org Wed Sep 13 06:08:31 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id GAA10739 for ; Wed, 13 Sep 1995 06:08:29 -0700 Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA05759; Wed, 13 Sep 1995 06:17:00 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id WAA01488 for ; Tue, 12 Sep 1995 22:17:20 -0600 Message-Id: <199509130417.WAA01488@rover.village.org> To: linux-mips@fnet.fr Subject: Deskstation Date: Tue, 12 Sep 1995 22:17:20 -0600 From: Warner Losh Content-Length: 847 Lines: 20 I just called the deskstation people. Basically, they sell 98% Alpha boxes, and 2% MIPS boxes, so they aren't real interested in working with me, except on the side like I've been able to do here and there to date. This likely explains a lot of the level of support that we've seen for their machines in this group. They did invite me to work with them, but set my expectations very low. I'll let you know if meaningful discussions do happen with them. They were rather impressed that I had gotten as far as I had with as little help from them as I had needed, so there may be some hope in this relationship working out. I think they want to work with me as their schedules allow, but I also think that I shouldn't expect much from them in the way of support. Maybe if I get a MIPS box running, they'll send me an alpha to try :-). Warner From andy@waldorf-gmbh.de Wed Sep 13 07:07:03 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id HAA11248 for ; Wed, 13 Sep 1995 07:07:03 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA06451; Wed, 13 Sep 1995 07:15:34 +0200 (MET) Received: from newton.waldorf-gmbh.de by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id HAA21939; Wed, 13 Sep 1995 07:17:00 +0200 Received: (from andy@localhost) by newton.waldorf-gmbh.de (8.6.12/8.6.9) id HAA16591; Wed, 13 Sep 1995 07:16:08 +0200 Date: Wed, 13 Sep 1995 07:16:08 +0200 From: Andreas Busse Message-Id: <199509130516.HAA16591@newton.waldorf-gmbh.de> To: linux-mips@fnet.fr Subject: Re: Deskstation In-Reply-To: <199509130417.WAA01488@rover.village.org> X-Mailer: [XMailTool v3.1.2b] Content-Length: 1294 Lines: 36 Hi, > > I just called the deskstation people. Basically, they sell 98% Alpha > boxes, and 2% MIPS boxes, so they aren't real interested in working > with me, except on the side like I've been able to do here and there > to date. Interesting... Looks like MIPS should do some advertising, perhaps even offer a low-cost board as DEC does. Not our fault that Alpha is the winner, but our disadvantage... > > This likely explains a lot of the level of support that we've seen for > their machines in this group. They did invite me to work with them, > but set my expectations very low. I'll let you know if meaningful > discussions do happen with them. > > They were rather impressed that I had gotten as far as I had with as > little help from them as I had needed, so there may be some hope in > this relationship working out. I think they want to work with me as > their schedules allow, but I also think that I shouldn't expect much > from them in the way of support. > I can only confirm that. The support I got for the Tyne board was as the same level. Just enough to make the basic things running. > Maybe if I get a MIPS box running, they'll send me an alpha to try > :-). Just a question out of curiousity: Did you mention your relationship to me? Andy From ralf@waldorf-gmbh.de Thu Sep 14 16:33:55 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id QAA00795 for ; Thu, 14 Sep 1995 16:33:53 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA11394; Thu, 14 Sep 1995 16:42:23 +0200 (MET) From: Ralf Baechle Message-Id: <199509141443.QAA20071@scotty.waldorf-gmbh.de> Received: from localhost by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id QAA20071; Thu, 14 Sep 1995 16:43:27 +0200 Subject: Re: Deskstation To: linux-mips@fnet.fr Date: Thu, 14 Sep 1995 16:43:27 +0200 (MET DST) In-Reply-To: <199509130516.HAA16591@newton.waldorf-gmbh.de> from "Andreas Busse" at Sep 13, 95 07:16:08 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 841 Lines: 17 Hi, > Interesting... Looks like MIPS should do some advertising, perhaps > even offer a low-cost board as DEC does. Not our fault that Alpha > is the winner, but our disadvantage... Yesterday I've met quite some people from the Linux congress in Berlin. We found that some companies sales (namely Wordperfect) for Linux already exceed the sales for SCO. As someone found out about every third customer that visited Star Division asked for a (nonexisting) Linux version of their StarWriter. Based on the number of sold CDROMs Linux and how often each one has probably been installed Linux is third most installed operating system in the US. This and the fact that the vice president of Caldera showed strong interest in Linux/MIPS should probably more than enough justification to for Mips and SGI to be interested in Linux/MIPS. Ralf From zaitcev@lab.ipmce.su Fri Sep 15 07:56:38 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id HAA10821 for ; Fri, 15 Sep 1995 07:56:37 -0700 Received: from eunet.EU.net by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA27520; Fri, 15 Sep 1995 08:05:12 +0200 (MET) Received: from kremvax.demos.su (kremvax.demos.su [192.91.186.200]) by eunet.EU.net (8.6.10/8.6.10) with ESMTP id IAA06335 for ; Fri, 15 Sep 1995 08:05:37 +0200 Received: by kremvax.demos.su (8.6.12/D) from root@localhost for linux-mips@fnet.fr id KAA14252; Fri, 15 Sep 1995 10:04:58 +0400 Received: from bone.UUCP by ipmce.su (8.6.9/SMI-RJMc) for linux-mips@fnet.fr id JAA01842; Fri, 15 Sep 1995 09:55:11 +0400 Received: from lab.ipmce.su (terra.lab.ipmce.su) by tretyak.ipmce.su (5.x/SMI-SVR4) id AA03093; Fri, 15 Sep 1995 09:55:23 +0400 Received: from jamica.lab.ipmce.su by lab.ipmce.su (4.1/SMI-4.1) id AA09347; Fri, 15 Sep 95 09:53:46 MSD Received: from mellorn.lab.ipmce.su by jamica.lab.ipmce.su (5.x/SMI-SVR4) id AA18228; Fri, 15 Sep 1995 09:53:44 +0400 Received: by mellorn.lab.ipmce.su (5.x/SMI-SVR4) id AA00242; Fri, 15 Sep 1995 09:53:42 +0400 Date: Fri, 15 Sep 1995 09:53:42 +0400 From: zaitcev@lab.ipmce.su (Pete A. Zaitcev) Message-Id: <9509150553.AA00242@mellorn.lab.ipmce.su> To: linux-mips@fnet.fr Subject: Re: DECstation news X-Sun-Charset: US-ASCII Content-Length: 1085 Lines: 26 Dear Collegues, SPARC PROM requires a.out images to boot and the rest of Linux/SPARC was planned to be ELF-only. But I had some troubles with ELF-->a.out linkage with binutils-2.5.8 for SPARC. I wrote a simple filter which converts a bootable image of a kernel form ELF to a.out. It is somewhere at vger.rutgers.edu under /pub/linux/Sparc/... Sorry for disturbtion if this note is not interesting. I wonder which version of gld/BFD can link ELF object code with a.out result without a coredumps (at MIPS). IMHO, this solution is better than ad hoc tools. I want to adopt it for SPARC. Why Ralf called it 'ugly'? Best wishes, Pete >> For the first time, the boot loaders will be able to load mipsel 407 executables >> (from ufs in the 'rz' case), but they should be aware of the ELF/ext2fs >> combination too. > >If that makes live easier for you - the ELF linker has no problem in linking >ELF files together and then producing an a.out file. This is ugly and >has several disadvantages but could make live easier as long as we don't >have real ELF loader code ready. > > Ralf From ralf@waldorf-gmbh.de Fri Sep 15 09:40:31 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id JAA11850 for ; Fri, 15 Sep 1995 09:40:30 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA29219; Fri, 15 Sep 1995 09:49:01 +0200 (MET) From: Ralf Baechle Message-Id: <199509150750.JAA04063@scotty.waldorf-gmbh.de> Received: from localhost by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id JAA04063; Fri, 15 Sep 1995 09:50:17 +0200 Subject: Re: DECstation news To: linux-mips@fnet.fr Date: Fri, 15 Sep 1995 09:50:17 +0200 (MET DST) Cc: zaitcev@lab.ipmce.su In-Reply-To: <9509150553.AA00242@mellorn.lab.ipmce.su> from "Pete A. Zaitcev" at Sep 15, 95 09:53:42 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1105 Lines: 25 Hi all, > Dear Collegues, > > SPARC PROM requires a.out images to boot and the rest of Linux/SPARC > was planned to be ELF-only. But I had some troubles with ELF-->a.out linkage > with binutils-2.5.8 for SPARC. I wrote a simple filter which converts a > bootable image of a kernel form ELF to a.out. It is somewhere at > vger.rutgers.edu under /pub/linux/Sparc/... Sorry for disturbtion if > this note is not interesting. Thanks, I'll give it a try! > I wonder which version of gld/BFD can link ELF object code with a.out > result without a coredumps (at MIPS). IMHO, this solution is better than > ad hoc tools. I want to adopt it for SPARC. Why Ralf called it 'ugly'? Sorry, I missexpressed myself. What may cause is linking code from different object file formats. The target object file format of the linker is less a problem. There are some problems with mixing multiple format left in the current release of the binutils. This is not of big interest as they result from a extensions in the a.out assembler (handling %HI, %hi and %lo) that are not yet present in the ELF assembler. Ralf From imp@rover.village.org Fri Sep 15 18:37:46 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id SAA17082 for ; Fri, 15 Sep 1995 18:37:45 -0700 Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA08961; Fri, 15 Sep 1995 18:46:17 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id KAA02564 for ; Fri, 15 Sep 1995 10:08:55 -0600 Message-Id: <199509151608.KAA02564@rover.village.org> To: linux-mips@fnet.fr Subject: Cross compilers for MIPS NT? Date: Fri, 15 Sep 1995 10:08:55 -0600 From: Warner Losh Content-Length: 107 Lines: 5 I didn't see one last time I looked, but... Has any body brewed a cross compiler for MIPS NT 3.5? Warner From Stoned.Elipot@univ-evry.fr Fri Sep 15 18:17:55 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id SAA16842 for ; Fri, 15 Sep 1995 18:17:55 -0700 Received: from amiante.univ-evry.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA08546; Fri, 15 Sep 1995 18:26:30 +0200 (MET) Received: from bandsept.univ-evry.fr (bandsept.univ-evry.fr [192.93.214.19]) by amiante.univ-evry.fr with ESMTP (8.6.10/94.01.10); id SAA16621; Fri, 15 Sep 1995 18:27:01 +0200 Received: (seb@localhost) by bandsept.univ-evry.fr (8.6.12/94.01.26); id SAA15266; Fri, 15 Sep 1995 18:25:15 +0200 Message-Id: <199509151625.SAA15266@bandsept.univ-evry.fr> Date: Fri, 15 Sep 1995 18:25:15 +0200 From: Stoned Elipot To: linux-mips@fnet.fr Subject: Andy, Ralf ?? X-Mailer: GNU Emacs [19.29] Mail Mode X-Attribution: seb X-Pgp-Key: ask a good public key server Reply-To: Stoned.Elipot@univ-evry.fr Content-Length: 140 Lines: 4 Hi all, I didn't succed in getting in touch via email with Andy and Ralf today (MAIlER DAEMON unhappy). Does someone did ? Thanks, Stoned. From jens@jens1.whu-koblenz.de Fri Sep 15 18:24:30 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id SAA16936 for ; Fri, 15 Sep 1995 18:24:28 -0700 Received: from obelix.WHU-Koblenz.DE by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA08655; Fri, 15 Sep 1995 18:33:01 +0200 (MET) Received: from jens1.whu-koblenz.de (jens@jens1.WHU-Koblenz.DE [193.141.48.51]) by obelix.WHU-Koblenz.de (8.6.8.1/8.6.4) with ESMTP id SAA06817 for ; Fri, 15 Sep 1995 18:33:30 +0200 Received: (from jens@localhost) by jens1.whu-koblenz.de (8.6.12/8.6.9) id SAA03092 for linux-mips@fnet.fr; Fri, 15 Sep 1995 18:31:21 +0200 From: Jens Hoffmann Message-Id: <199509151631.SAA03092@jens1.whu-koblenz.de> Subject: Re: Andy, Ralf ?? To: linux-mips@fnet.fr Date: Fri, 15 Sep 1995 18:31:17 +0200 (MET DST) In-Reply-To: <199509151625.SAA15266@bandsept.univ-evry.fr> from "Stoned Elipot" at Sep 15, 95 06:25:15 pm Reply-To: jens@waldorf-gmbh.de X-Mailer: ELM [version 2.4 PL24] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 340 Lines: 14 Hi Stoned, Stoned Elipot wrote: > Hi all, I didn't succed in getting in touch via email with Andy and > Ralf today (MAIlER DAEMON unhappy). Does someone did ? Andy is configuring his site today. hard-n-soft.de is on it's way. Try again, sit should work soon. (His new domain not, cause the DNS needs some more time) Greetings, Jens From andy@newton.soft-n-hard.de Fri Sep 15 18:22:20 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id SAA16904 for ; Fri, 15 Sep 1995 18:22:20 -0700 Received: from bnc.net by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA08620; Fri, 15 Sep 1995 18:30:40 +0200 (MET) Received: from newton.soft-n-hard.de ([194.163.66.33]) by bnc.net (8.6.11/8.6.9) with ESMTP id SAA14741 for ; Fri, 15 Sep 1995 18:32:12 +0200 Received: (from andy@localhost) by newton.soft-n-hard.de (8.6.12/8.6.9) id SAA00743; Fri, 15 Sep 1995 18:31:27 +0200 Date: Fri, 15 Sep 1995 18:31:27 +0200 From: Andreas Busse Message-Id: <199509151631.SAA00743@newton.soft-n-hard.de> To: linux-mips@fnet.fr Subject: Re: Andy, Ralf ?? In-Reply-To: <199509151625.SAA15266@bandsept.univ-evry.fr> Cc: andy@newton.soft-n-hard.de, ralf@waldorf-gmbh.de X-Mailer: [XMailTool v3.1.2b] Content-Length: 692 Lines: 24 Hi, > > Hi all, I didn't succed in getting in touch via email with Andy and > Ralf today (MAIlER DAEMON unhappy). Does someone did ? > Perhaps because I changed my domain :-) You can either send mail to or (hopefully :-)) Better use the old address. The new domain and route might not be internationally available yet... Cheers, Andy ----------------------------------------------------------- Andreas Busse | andy@soft-n-hard.de Soft N Hard GbR | Phone: +49 2636-970105 Im Hufen Boden 16, D-53498 Waldorf | Fax: +49 2636-970106 ----------------------------------------------------------- From andy@newton.soft-n-hard.de Fri Sep 15 18:33:50 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id SAA17034 for ; Fri, 15 Sep 1995 18:33:49 -0700 Received: from bnc.net by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA08859; Fri, 15 Sep 1995 18:42:09 +0200 (MET) Received: from newton.soft-n-hard.de ([194.163.66.33]) by bnc.net (8.6.11/8.6.9) with ESMTP id SAA14868 for ; Fri, 15 Sep 1995 18:43:45 +0200 Received: (from andy@localhost) by newton.soft-n-hard.de (8.6.12/8.6.9) id SAA00771; Fri, 15 Sep 1995 18:43:00 +0200 Date: Fri, 15 Sep 1995 18:43:00 +0200 From: Andreas Busse Message-Id: <199509151643.SAA00771@newton.soft-n-hard.de> To: linux-mips@fnet.fr Subject: Re: Andy, Ralf ?? In-Reply-To: <199509151631.SAA03092@jens1.whu-koblenz.de> X-Mailer: [XMailTool v3.1.2b] Content-Length: 620 Lines: 21 Hi all, > > Andy is configuring his site today. > hard-n-soft.de is on it's way. > Try again, sit should work soon. > (His new domain not, cause the DNS needs some more time) > Jens is absolutely right. And it's real fun to do the routing for Waldorf and my own net to two different providers :-) Cheers, Andy ----------------------------------------------------------- Andreas Busse | andy@soft-n-hard.de Soft N Hard GbR | Phone: +49 2636-970105 Im Hufen Boden 16, D-53498 Waldorf | Fax: +49 2636-970106 ----------------------------------------------------------- From ralf@waldorf-gmbh.de Fri Sep 15 22:49:47 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id WAA19492 for ; Fri, 15 Sep 1995 22:49:46 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA13447; Fri, 15 Sep 1995 22:58:17 +0200 (MET) From: Ralf Baechle Message-Id: <199509152059.WAA17034@scotty.waldorf-gmbh.de> Received: from localhost by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id WAA17034; Fri, 15 Sep 1995 22:59:40 +0200 Subject: Re: Cross compilers for MIPS NT? To: linux-mips@fnet.fr Date: Fri, 15 Sep 1995 22:59:40 +0200 (MET DST) In-Reply-To: <199509151608.KAA02564@rover.village.org> from "Warner Losh" at Sep 15, 95 10:08:55 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 379 Lines: 13 Hi, > I didn't see one last time I looked, but... > > Has any body brewed a cross compiler for MIPS NT 3.5? Sorry, nothing yet heared about such a beast. There is however some work in progress in brewing WIN32 support. This and the POSIX features of NT should make it fairly easy to brew such a thing. Don't know how far this applies to the configuration scripts. Ralf From ralf@waldorf-gmbh.de Fri Sep 15 23:22:24 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id XAA19818 for ; Fri, 15 Sep 1995 23:22:23 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA14122; Fri, 15 Sep 1995 23:30:54 +0200 (MET) From: Ralf Baechle Message-Id: <199509152132.XAA17602@scotty.waldorf-gmbh.de> Received: from localhost by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id XAA17602; Fri, 15 Sep 1995 23:32:16 +0200 Subject: Re: Andy, Ralf ?? To: linux-mips@fnet.fr Date: Fri, 15 Sep 1995 23:32:15 +0200 (MET DST) In-Reply-To: <199509151625.SAA15266@bandsept.univ-evry.fr> from "Stoned Elipot" at Sep 15, 95 06:25:15 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 545 Lines: 18 Hi, > Hi all, I didn't succed in getting in touch via email with Andy and > Ralf today (MAIlER DAEMON unhappy). Does someone did ? > > Thanks, Stoned. Well, mailer trouble may occour sometimes. If someone tries to reach me and fail he might try one of the following accounts: ralf@informatik.uni-koblenz.de (Only reachable from .de!) linux@informatik.uni-koblenz.de (This is a group account!) baechle@oldtimer.cs.uni-sb.de (64k limit for mailsize) rbaechle@ernie.mi.uni-koeln.de So just in case everything breaks ... Ralf From pab@rmnet.it Fri Sep 15 23:41:22 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.6.12/8.6.12) with SMTP id XAA20008 for ; Fri, 15 Sep 1995 23:41:18 -0700 Received: from snoop.uni.net by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA14568; Fri, 15 Sep 1995 23:49:46 +0200 (MET) Received: from pag (pag.RMnet.it [194.20.98.41]) by snoop.uni.net (8.6.11/8.6.9) with SMTP id XAA08356 for ; Fri, 15 Sep 1995 23:37:08 +0100 Date: Fri, 15 Sep 1995 23:37:08 +0100 Message-Id: <199509152237.XAA08356@snoop.uni.net> X-Sender: pab@relay.uni.net X-Mailer: Windows Eudora Light Version 1.5.2b1 Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" To: linux-mips@fnet.fr From: Paolo Bevilacqua Subject: Re: Andy, Ralf ?? Content-Length: 393 Lines: 14 Ralf wrote: > ralf@informatik.uni-koblenz.de (Only reachable from .de!) > linux@informatik.uni-koblenz.de (This is a group account!) > baechle@oldtimer.cs.uni-sb.de (64k limit for mailsize) > rbaechle@ernie.mi.uni-koeln.de > >So just in case everything breaks ... > My only fear is that when the Internet will fall down, i can be under a stack of heavvy routers .... /pab From imp@rover.village.org Mon Sep 18 20:39:42 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id UAA00943 for ; Mon, 18 Sep 1995 20:39:41 -0700 Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA16017; Mon, 18 Sep 1995 20:47:42 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id MAA09416 for ; Mon, 18 Sep 1995 12:48:13 -0600 Message-Id: <199509181848.MAA09416@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Cross compilers for MIPS NT? Date: Mon, 18 Sep 1995 12:48:13 -0600 From: Warner Losh Content-Length: 776 Lines: 18 : Sorry, nothing yet heared about such a beast. There is however : some work in progress in brewing WIN32 support. This and the : POSIX features of NT should make it fairly easy to brew such : a thing. Don't know how far this applies to the configuration : scripts. Just need to be able to build console mode apps, which I know that the NT Intel port of GCC does right now. I'd *NEVER* use the POSIX side of NT. It is a toy that is there as a checklist item for the government contracts, and will never be functional, imho. I'll have to blaze this trail myself then.... Looks like the configure script for the latest gcc snapshot groks only alpha--winnt and ix86--winnt. I'll see how hard it is to add mipsel--winnt :-). Would anybody else find this useful? Warner From censjg@caledonia.hw.ac.uk Wed Sep 20 12:39:37 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id MAA07591 for ; Wed, 20 Sep 1995 12:39:36 -0700 From: censjg@caledonia.hw.ac.uk Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA29798; Wed, 20 Sep 1995 12:47:29 +0200 (MET) Received: from caledonias1.caledonia.hw.ac.uk by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id MAA06840; Wed, 20 Sep 1995 12:48:11 +0200 Received: from cend1c4.caledonia.hw.ac.uk by caledonias1.caledonia.hw.ac.uk with SMTP id AA11042; Wed, 20 Sep 1995 11:47:10 +0100 Received: by cend1c4.caledonia.hw.ac.uk (5.65/4.7) id AA10718; Wed, 20 Sep 1995 11:47:09 +0100 Date: Wed, 20 Sep 1995 11:47:09 +0100 Message-Id: <9509201047.AA10718@cend1c4.caledonia.hw.ac.uk> To: linux@waldorf-gmbh.de Subject: DECStation R3000 Linux Content-Length: 824 Lines: 25 Hi, The Linux-MIPS howto says you want to hear from people running DECstations, I run a network of around 80, currently running Ultrix, here at Heriot-Watt University in Edinburgh, UK. I'm running Linux on my PC, and on ~12 PC's at an Internet Cafe where I also work. I'm very interested in the idea of running Linux on some of the mips machines, (all 5000 series, models 240, 133, 125 and 25), especially since the demise of Ultrix. I don't know if I will be able to offer much time for porting etc, though I'd like to. Please let me know the state of support for R3000, and suggestions for how I might be able to help the development process. I have access to hp9000, SunOS 4.1.n, Solaris 2.4, DEC OSF/1 V3.2, Linux etc. -- Simon Greaves censjg@caledonia.hw.ac.uk DDI: +44 (0)131 451 3265 Fax: +44 (0)131 451 3261 From ralf@waldorf-gmbh.de Wed Sep 20 17:25:38 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id RAA08010 for ; Wed, 20 Sep 1995 17:25:37 -0700 Received: from scotty.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA05524; Wed, 20 Sep 1995 17:33:33 +0200 (MET) From: Ralf Baechle Message-Id: <199509201534.RAA13165@scotty.waldorf-gmbh.de> Received: from localhost by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id RAA13165; Wed, 20 Sep 1995 17:34:34 +0200 Subject: Re: DECStation R3000 Linux To: censjg@caledonia.hw.ac.uk Date: Wed, 20 Sep 1995 17:34:33 +0200 (MET DST) Cc: linux@waldorf-gmbh.de In-Reply-To: <9509201047.AA10718@cend1c4.caledonia.hw.ac.uk> from "censjg@caledonia.hw.ac.uk" at Sep 20, 95 11:47:09 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 2596 Lines: 57 Hi, > The Linux-MIPS howto says you want to hear from people running > DECstations, I run a network of around 80, currently running Ultrix, > here at Heriot-Watt University in Edinburgh, UK. Hey, that's nice; never heared before of someone with that many DECs. I guess you really want to swap Ultrix for Linux on your machines, don't you? > I'm running Linux on my PC, and on ~12 PC's at an Internet Cafe where > I also work. > > I'm very interested in the idea of running Linux on some of the mips > machines, (all 5000 series, models 240, 133, 125 and 25), especially > since the demise of Ultrix. > > I don't know if I will be able to offer much time for porting etc, > though I'd like to. Please let me know the state of support for R3000, > and suggestions for how I might be able to help the development > process. The most work till now (and that isn't very much ...) has been invested into support for the 3000 series. Jim Paradis of DEC has offered to give us the required information for an port. A bootserver for booting a DECstation 3000 (dunno about 5000) from a Linux host has been written. I personally have a R4400 workstation. Having worked only for the R4xx0 series in the beginning I've changed most of the stuff so that pluging in the R3000 code at the right place should suffice. A short answer for your questions but I'm in hurry ... I don't know what development works you did yet; if you did ever some assembler hacking you're probably the right man for the R3000 port. If not don't - no problem. Linux/MIPS is my first MIPS program, too. > I have access to hp9000, SunOS 4.1.n, Solaris 2.4, DEC OSF/1 V3.2, > Linux etc. Linux is the preferable operatiog system for cross development as it can write the required ext2fs root filesystems etc. and all the cross development software compiles easily. Nevertheless all the other UNIX style OS should be fine. An NT crossdevelopment kit it currently in the works. I'd be more than happy to see some real move on the DECstation side; the demand for Linux as replacement OS for Ultrix (aka Buglix on this mailing list ...) is big if not even HUGE. If I understand you right you're from an academic institution. Maybe some students are happy to work with you, too? Please feel free to ask me any questions about the MIPS port; as the main kernel hacker I'm the best person to answer your questions. On the mailing list you can reach the other DECstation hackers. For subscription to the mailing list please mail to Stoned.Elipot@univ-evry.fr; he is maintaining the subscription list. Ralf (in hurry ...) From imp@rover.village.org Wed Sep 20 18:28:15 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id SAA08113 for ; Wed, 20 Sep 1995 18:28:15 -0700 Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA06866; Wed, 20 Sep 1995 18:36:16 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id KAA21023; Wed, 20 Sep 1995 10:36:46 -0600 Message-Id: <199509201636.KAA21023@rover.village.org> To: linux-mips@fnet.fr Subject: Re: DECStation R3000 Linux Cc: censjg@caledonia.hw.ac.uk, linux@waldorf-gmbh.de In-Reply-To: Your message of Wed, 20 Sep 1995 17:34:33 +0200 Date: Wed, 20 Sep 1995 10:36:45 -0600 From: Warner Losh Content-Length: 114 Lines: 5 : An NT crossdevelopment kit it currently in the works. Slowly, and haltingly, but it is in the works... Warner From andy@soft-n-hard.de Wed Sep 20 18:29:16 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id SAA08126 for ; Wed, 20 Sep 1995 18:29:13 -0700 Received: from bnc.net by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA06889; Wed, 20 Sep 1995 18:37:10 +0200 (MET) Received: from newton.soft-n-hard.de ([194.163.66.33]) by bnc.net (8.6.11/8.6.9) with ESMTP id SAA23304 for ; Wed, 20 Sep 1995 18:38:53 +0200 Received: (from andy@localhost) by newton.soft-n-hard.de (8.6.12/8.6.9) id SAA05490; Wed, 20 Sep 1995 18:38:12 +0200 Date: Wed, 20 Sep 1995 18:38:12 +0200 From: Andreas Busse Message-Id: <199509201638.SAA05490@newton.soft-n-hard.de> To: linux-mips@fnet.fr Subject: Back on earth again X-Mailer: [XMailTool v3.1.2b] Content-Length: 1312 Lines: 37 Hi all, I'm back on earth again, after some fights with my new provider, the DE-NIC, german Telecom and so on. That was real fun, including flame wars :-) Stoned, could you please change my subscription on the linux-mips list to ? Thanks! Next: The Linux/MIPS archive has been nearly completely moved to ftp://ftp.fnet.fr/linux-mips. I'll notify all mirror site administrators that ftp.waldorf-gmbh.de isn't active anymore. Please also do not up/download files to and from ftp.waldorf-gmbh.de anymore since this archive will be closed soon, as well as many other archives on this host. Active mirror sites are ftp.uni-mainz.de:/pub/Linux/arch/mips ftp.mcc.ac.uk:/pub/linux/MIPS I'm not sure about what happened to the mirror on sunsite.unc.edu but I'll check that soon. The Web pages will also be moved soon, and as a temporary solution I'll setup links from www.waldorf-gmbh.de to ftp.fnet.fr. Have I missed something important in the last few days ? Cheers, Andy ----------------------------------------------------------- Andreas Busse | andy@soft-n-hard.de Soft N Hard GbR | Phone: +49 2636-970105 Im Hufen Boden 16, D-53498 Waldorf | Fax: +49 2636-970106 ----------------------------------------------------------- From andy@soft-n-hard.de Fri Sep 22 17:40:28 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id RAA27653 for ; Fri, 22 Sep 1995 17:40:27 -0700 Received: from bnc.net by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA29409; Fri, 22 Sep 1995 17:48:26 +0200 (MET) Received: from newton.soft-n-hard.de ([194.163.66.33]) by bnc.net (8.6.11/8.6.9) with ESMTP id RAA13080 for ; Fri, 22 Sep 1995 17:49:58 +0200 Received: (from andy@localhost) by newton.soft-n-hard.de (8.6.12/8.6.9) id RAA11103; Fri, 22 Sep 1995 17:49:10 +0200 Date: Fri, 22 Sep 1995 17:49:10 +0200 From: Andreas Busse Message-Id: <199509221549.RAA11103@newton.soft-n-hard.de> To: linux-mips@fnet.fr Subject: Re: Andy, Ralf ?? In-Reply-To: <199509152237.XAA08356@snoop.uni.net> X-Mailer: [XMailTool v3.1.2b] Content-Length: 602 Lines: 18 > > My only fear is that when the Internet will fall down, i can be under a > stack of heavvy routers .... > Sorry for the late answer, Paolo, but I guess you're wrong. You won't be hit by heavy routers and cables but by fat, lazy, dumb and brain damaged system adminstrators :->>> Cheers, Andy ----------------------------------------------------------- Andreas Busse | andy@soft-n-hard.de Soft N Hard GbR | Phone: +49 2636-970105 Im Hufen Boden 16, D-53498 Waldorf | Fax: +49 2636-970106 ----------------------------------------------------------- From andy@soft-n-hard.de Fri Sep 22 17:48:08 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id RAA27739 for ; Fri, 22 Sep 1995 17:48:07 -0700 Received: from bnc.net by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA29575; Fri, 22 Sep 1995 17:56:05 +0200 (MET) Received: from newton.soft-n-hard.de ([194.163.66.33]) by bnc.net (8.6.11/8.6.9) with ESMTP id RAA13223; Fri, 22 Sep 1995 17:57:46 +0200 Received: (from andy@localhost) by newton.soft-n-hard.de (8.6.12/8.6.9) id RAA11119; Fri, 22 Sep 1995 17:56:59 +0200 Date: Fri, 22 Sep 1995 17:56:59 +0200 From: Andreas Busse Message-Id: <199509221556.RAA11119@newton.soft-n-hard.de> To: linux-mips@fnet.fr, censjg@caledonia.hw.ac.uk Subject: Re: DECStation R3000 Linux In-Reply-To: <199509201534.RAA13165@scotty.waldorf-gmbh.de> X-Mailer: [XMailTool v3.1.2b] Content-Length: 934 Lines: 30 > > Hi, > > > The Linux-MIPS howto says you want to hear from people running > > DECstations, I run a network of around 80, currently running Ultrix, > > here at Heriot-Watt University in Edinburgh, UK. > > Hey, that's nice; never heared before of someone with that many DECs. > I guess you really want to swap Ultrix for Linux on your machines, don't > you? > > [...] Hi, sorry for my late response -- usually I'm the one who answers all these queries, but I had some trouble with changing my Internet provider and had no time to answer all my mail yet. So... Is there anything I can do ? Please let me know, Andy ----------------------------------------------------------- Andreas Busse | andy@soft-n-hard.de Soft N Hard GbR | Phone: +49 2636-970105 Im Hufen Boden 16, D-53498 Waldorf | Fax: +49 2636-970106 ----------------------------------------------------------- From andy@soft-n-hard.de Fri Sep 22 19:35:37 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id TAA28792 for ; Fri, 22 Sep 1995 19:35:37 -0700 Received: from bnc.net by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA02259; Fri, 22 Sep 1995 19:43:28 +0200 (MET) Received: from newton.soft-n-hard.de ([194.163.66.33]) by bnc.net (8.6.11/8.6.9) with ESMTP id TAA16534 for ; Fri, 22 Sep 1995 19:45:08 +0200 Received: (from andy@localhost) by newton.soft-n-hard.de (8.6.12/8.6.9) id TAA11308; Fri, 22 Sep 1995 19:44:27 +0200 Date: Fri, 22 Sep 1995 19:44:27 +0200 From: Andreas Busse Message-Id: <199509221744.TAA11308@newton.soft-n-hard.de> To: linux-mips@fnet.fr Subject: Sunsite Mirror X-Mailer: [XMailTool v3.1.2b] Content-Length: 518 Lines: 17 Hi all, I just checked some of the Sunsite Mirrors regarding the Linux/MIPS archive. It seems uptodate in comparison to the archive on ftp.waldorf-gmbh.de. Hopefully they will soon swap over to FNET.FR. Cheers, Andy ----------------------------------------------------------- Andreas Busse | andy@soft-n-hard.de Soft N Hard GbR | Phone: +49 2636-970105 Im Hufen Boden 16, D-53498 Waldorf | Fax: +49 2636-970106 ----------------------------------------------------------- From andy@soft-n-hard.de Fri Sep 22 19:46:24 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id TAA28917 for ; Fri, 22 Sep 1995 19:46:23 -0700 Received: from bnc.net by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA02398; Fri, 22 Sep 1995 19:54:15 +0200 (MET) Received: from newton.soft-n-hard.de ([194.163.66.33]) by bnc.net (8.6.11/8.6.9) with ESMTP id TAA16808 for ; Fri, 22 Sep 1995 19:55:57 +0200 Received: (from andy@localhost) by newton.soft-n-hard.de (8.6.12/8.6.9) id TAA11331; Fri, 22 Sep 1995 19:55:15 +0200 Date: Fri, 22 Sep 1995 19:55:15 +0200 From: Andreas Busse Message-Id: <199509221755.TAA11331@newton.soft-n-hard.de> To: linux-mips@fnet.fr Subject: mcc.ac.uk mirror X-Mailer: [XMailTool v3.1.2b] Content-Length: 572 Lines: 18 Hi again, good news: the Linux/MIPS mirror on ftp.mcc.ac.uk is uptodate, including the now complete doc/ tree with all MIPS documents from sgigate.sgi.com. Sunsite is mirroring ftp.mcc.ac.uk, so we can expect a perfect distribution within a couple of days. Cheers, Andy ----------------------------------------------------------- Andreas Busse | andy@soft-n-hard.de Soft N Hard GbR | Phone: +49 2636-970105 Im Hufen Boden 16, D-53498 Waldorf | Fax: +49 2636-970106 ----------------------------------------------------------- From andy@soft-n-hard.de Mon Sep 25 10:12:02 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id KAA09083 for ; Mon, 25 Sep 1995 10:12:01 -0700 Received: from bnc.net by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA25699; Mon, 25 Sep 1995 09:20:05 +0100 (MET) Received: from newton.soft-n-hard.de ([194.163.66.33]) by bnc.net (8.6.11/8.6.9) with ESMTP id JAA03478 for ; Mon, 25 Sep 1995 09:21:54 +0100 Received: (from andy@localhost) by newton.soft-n-hard.de (8.6.12/8.6.9) id JAA20330; Mon, 25 Sep 1995 09:21:12 +0100 Date: Mon, 25 Sep 1995 09:21:12 +0100 From: Andreas Busse Message-Id: <199509250821.JAA20330@newton.soft-n-hard.de> To: linux-mips@fnet.fr Subject: Linux-1.2.11 patches X-Mailer: [XMailTool v3.1.2b] Content-Length: 895 Lines: 26 Hi all, today patches from Linux-1.2.10 to Linux-1.2.11 will appear on ftp.fnet.fr in /linux-mips/kernel/src/v1.2/linux-1.2.10-1.2.11.diffs.gz I will also integrate my last changes to the 1.2.10 source tree and upload a 1.2.11-1.2.11.1 patch. Also, to make the update to 1.3.x more easy, I'll remove the NFS boot patches again and upload a 1.2.11.1-1.2.11.2 patch. If someone has made any changes to the kernel, please send me patches now. The move to 1.3.x will come soon and I really would like to include support for the Olivetti and Deskstation boxes before this change. Thanks, Andy ----------------------------------------------------------- Andreas Busse | andy@soft-n-hard.de Soft N Hard GbR | Phone: +49 2636-970105 Im Hufen Boden 16, D-53498 Waldorf | Fax: +49 2636-970106 ----------------------------------------------------------- From andy@soft-n-hard.de Mon Sep 25 11:15:54 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id LAA09693 for ; Mon, 25 Sep 1995 11:15:53 -0700 Received: from bnc.net by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA27281; Mon, 25 Sep 1995 10:23:51 +0100 (MET) Received: from newton.soft-n-hard.de ([194.163.66.33]) by bnc.net (8.6.11/8.6.9) with ESMTP id KAA05134 for ; Mon, 25 Sep 1995 10:25:29 +0100 Received: (from andy@localhost) by newton.soft-n-hard.de (8.6.12/8.6.9) id KAA21522; Mon, 25 Sep 1995 10:24:48 +0100 Date: Mon, 25 Sep 1995 10:24:48 +0100 From: Andreas Busse Message-Id: <199509250924.KAA21522@newton.soft-n-hard.de> To: linux-mips@fnet.fr Subject: Again: Kernel patches X-Mailer: [XMailTool v3.1.2b] Content-Length: 646 Lines: 19 Hi again, I've sent the 1.2.11.1 and 1.2.11.2 patches to Luc. Patch set #1 contains some minor changes only, whereas patch set #2 removes the NFS boot/root option from the kernel. It was necessary to remove this since the 1.3.x patches otherwise break. Both patches should appear on ftp.fnet.fr in /linux-mips/private/Incoming. Cheers, Andy ----------------------------------------------------------- Andreas Busse | andy@soft-n-hard.de Soft N Hard GbR | Phone: +49 2636-970105 Im Hufen Boden 16, D-53498 Waldorf | Fax: +49 2636-970106 ----------------------------------------------------------- From Stoned.Elipot@univ-evry.fr Mon Sep 25 12:23:48 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id MAA10519 for ; Mon, 25 Sep 1995 12:23:48 -0700 Received: from amiante.univ-evry.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA28592; Mon, 25 Sep 1995 11:31:54 +0100 (MET) Received: from bandsept.univ-evry.fr (bandsept.univ-evry.fr [192.93.214.19]) by amiante.univ-evry.fr with ESMTP (8.6.10/94.01.10); id LAA00603; Mon, 25 Sep 1995 11:32:32 +0100 Received: (seb@localhost) by bandsept.univ-evry.fr (8.6.12/94.01.26); id MAA05078; Mon, 25 Sep 1995 12:30:24 +0200 Message-Id: <199509251030.MAA05078@bandsept.univ-evry.fr> Date: Mon, 25 Sep 1995 12:30:24 +0200 From: Stoned Elipot To: linux-mips@fnet.fr In-Reply-To: <199509250821.JAA20330@newton.soft-n-hard.de> (message from Andreas Busse on Mon, 25 Sep 1995 09:21:12 +0100) Subject: Re: Linux-1.2.11 patches X-Mailer: GNU Emacs [19.29] Mail Mode X-Attribution: seb X-Pgp-Key: ask a good public key server Reply-To: Stoned.Elipot@univ-evry.fr Content-Length: 530 Lines: 14 Hi All, >>>>> "Andreas" == Andreas Busse writes: [SNIP] Andreas> If someone has made any changes to the kernel, please send me Andreas> patches now. The move to 1.3.x will come soon and I really Andreas> would like to include support for the Olivetti and Andreas> Deskstation boxes before this change. Hum,... all right, I've got to something then. I'll try to do my best. Cheers, Stoned - afraid to make changes in kernel and tearing on the time hacking milo was enought to make the Oily stuff go on :) From andy@soft-n-hard.de Mon Sep 25 14:21:47 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id OAA01406 for ; Mon, 25 Sep 1995 14:21:47 -0700 Received: from bnc.net by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA00902; Mon, 25 Sep 1995 13:29:44 +0100 (MET) Received: from newton.soft-n-hard.de ([194.163.66.33]) by bnc.net (8.6.11/8.6.9) with ESMTP id LAA07398 for ; Mon, 25 Sep 1995 11:50:33 +0100 Received: (from andy@localhost) by newton.soft-n-hard.de (8.6.12/8.6.9) id LAA21668; Mon, 25 Sep 1995 11:49:50 +0100 Date: Mon, 25 Sep 1995 11:49:50 +0100 From: Andreas Busse Message-Id: <199509251049.LAA21668@newton.soft-n-hard.de> To: linux-mips@fnet.fr Subject: Re: Linux-1.2.11 patches In-Reply-To: <199509251030.MAA05078@bandsept.univ-evry.fr> X-Mailer: [XMailTool v3.1.2b] Content-Length: 1848 Lines: 57 Hi again, > > Hum,... all right, I've got to something then. I'll try to do my best. > Wonderful ! :-) > Cheers, Stoned - afraid to make changes in kernel and tearing on the > time hacking milo was enought to make the Oily stuff go on :) > Well, it's probably a bit more than just hacking Milo, but it's not that hard as you might think now. Step 1: Edit arch/mips/kernel/head.S - Copy the map0_magnum4000 to map0_olivetti_m700 - Add map0_olivetti_m700 to the map0table (Note that the position in map0table MUST correspond to the MACHTYPE ID of the Oily box) Step 2: Create arch/mips/kernel/olivetti.S - Copy magnum4000.S into this file - Change all names accordingly - Edit Makefile Step 3: Edit arch/mips/kernel/irq.c - In init_IRQ, add a case statement for the Olivetti. Copy either the code of the PICA or the Magnum. Better comment out any unnecessary interrupts. Step 4: Edit arch/mips/kernel/setup.c - Add CONFIG stuff for the Olivetti. Step 5: Edit drivers/char/console.c - Whereever you find CONFIG_MAGNUM_4000, add the Olivetti too. Since both uses the G364 graphics controller, this should work right out of the box. Sorry, we never got patches for the console driver, so the Olivetti support has been removed again silently. As a rule of thumb: Whenever you made something working, release patches! Otherwise your work goes down the tubes :-) That, besides some other minor changes that might be necessary, should be sufficient at least to boot the kernel and to see some console output. Cheers, Andy ----------------------------------------------------------- Andreas Busse | andy@soft-n-hard.de Soft N Hard GbR | Phone: +49 2636-970105 Im Hufen Boden 16, D-53498 Waldorf | Fax: +49 2636-970106 ----------------------------------------------------------- From Stoned.Elipot@univ-evry.fr Mon Sep 25 15:05:00 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id PAA01836 for ; Mon, 25 Sep 1995 15:05:00 -0700 Received: from amiante.univ-evry.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA01815; Mon, 25 Sep 1995 14:13:06 +0100 (MET) Received: from bandsept.univ-evry.fr (bandsept.univ-evry.fr [192.93.214.19]) by amiante.univ-evry.fr with ESMTP (8.6.10/94.01.10); id OAA01142; Mon, 25 Sep 1995 14:13:43 +0100 Received: (seb@localhost) by bandsept.univ-evry.fr (8.6.12/94.01.26); id PAA07989; Mon, 25 Sep 1995 15:11:33 +0200 Message-Id: <199509251311.PAA07989@bandsept.univ-evry.fr> Date: Mon, 25 Sep 1995 15:11:33 +0200 From: Stoned Elipot To: linux-mips@fnet.fr In-Reply-To: <199509251049.LAA21668@newton.soft-n-hard.de> (message from Andreas Busse on Mon, 25 Sep 1995 11:49:50 +0100) Subject: Re: Linux-1.2.11 patches X-Mailer: GNU Emacs [19.29] Mail Mode X-Attribution: seb X-Pgp-Key: ask a good public key server Reply-To: Stoned.Elipot@univ-evry.fr Content-Length: 722 Lines: 21 Hi again, (very boring introduction, isn't it ? :) >>>>> "Andreas" == Andreas Busse writes: [SNIP] I got the checklist, I already done half the work, but something is wrong between 1.2.10.2 and 1.2.10.5, as you had suggested centuries ago, I go slowly towards 1.2.10.5... Andreas> Sorry, we never got patches for the console driver, so the Andreas> Olivetti support has been removed again silently. As a rule Andreas> of thumb: Whenever you made something working, release Andreas> patches! Otherwise your work goes down the tubes :-) Hey, that's Luc's job ! Well, he promised to work on it in the next weeks, it was nearly impossible for various reasons the last 3 month :( Cheers, Stoned. From Stoned.Elipot@univ-evry.fr Mon Sep 25 15:32:58 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id PAA02117 for ; Mon, 25 Sep 1995 15:32:57 -0700 Received: from amiante.univ-evry.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA02408; Mon, 25 Sep 1995 14:41:02 +0100 (MET) Received: from bandsept.univ-evry.fr (bandsept.univ-evry.fr [192.93.214.19]) by amiante.univ-evry.fr with ESMTP (8.6.10/94.01.10); id OAA01172; Mon, 25 Sep 1995 14:41:38 +0100 Received: (seb@localhost) by bandsept.univ-evry.fr (8.6.12/94.01.26); id PAA08033; Mon, 25 Sep 1995 15:39:29 +0200 Message-Id: <199509251339.PAA08033@bandsept.univ-evry.fr> Date: Mon, 25 Sep 1995 15:39:29 +0200 From: Stoned Elipot To: linux-mips@fnet.fr In-Reply-To: (message from Matt Messier on Thu, 1 Jan 1970 08:30:55 -0500) Subject: Re: DECstation status X-Mailer: GNU Emacs [19.29] Mail Mode X-Attribution: seb X-Pgp-Key: ask a good public key server Reply-To: Stoned.Elipot@univ-evry.fr Content-Length: 733 Lines: 18 Hi all, >>>>> "Matt" == Matt Messier writes: Matt> Does anybody out there with a DEC 3100 or 5000 have the ability Matt> to test this stuff? It is a little bit difficult for me at the Matt> moment because I need to swap drives around all of the time to Matt> test things. I'm quite willing to work on writing the code if Matt> someone can test it easily for me. Once things get a bit Matt> further along and we know that it is working somewhat reasonably Matt> well, I'll be a bit more willing to spend half an hour at a shot Matt> to test a change :-) I got an unplugged personnal DEC 5000 station with Ultrix 4.4 on its disks at about 50cm on my right hand, can it be of any help ?? :) Cheers, Stoned. From andy@soft-n-hard.de Mon Sep 25 15:36:57 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id PAA02169 for ; Mon, 25 Sep 1995 15:36:56 -0700 Received: from bnc.net by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA02491; Mon, 25 Sep 1995 14:44:51 +0100 (MET) Received: from newton.soft-n-hard.de ([194.163.66.33]) by bnc.net (8.6.11/8.6.9) with ESMTP id OAA12243 for ; Mon, 25 Sep 1995 14:46:40 +0100 Received: (from andy@localhost) by newton.soft-n-hard.de (8.6.12/8.6.9) id OAA22055; Mon, 25 Sep 1995 14:45:56 +0100 Date: Mon, 25 Sep 1995 14:45:56 +0100 From: Andreas Busse Message-Id: <199509251345.OAA22055@newton.soft-n-hard.de> To: linux-mips@fnet.fr Subject: Re: DECstation status In-Reply-To: X-Mailer: [XMailTool v3.1.2b] Content-Length: 1809 Lines: 47 > > This weekend I had an opportunity to get some work done on the DECstation > port. Admittedly I haven't gotten too terribly far, but I've got basic > boot code written. It is as yet untested, but I have written code to > install the bootloader onto the scsi device, and I have gotten a start on > the bootloader itself. > Good work! > Does anybody out there with a DEC 3100 or 5000 have the ability to test > this stuff? It is a little bit difficult for me at the moment because I > need to swap drives around all of the time to test things. I'm quite > willing to work on writing the code if someone can test it easily for > me. Once things get a bit further along and we know that it is working > somewhat reasonably well, I'll be a bit more willing to spend half an > hour at a shot to test a change :-) > Yes, please contact Simon Greaves . Today he wrote: > >So... Is there anything I can do ? > > Just keep me informed of the status of Linux for MIPS R3000 (ie DECstation > 5000 series). > > Might be able to offer some help with the port, depending on what needs done > etc. > Also, what I would like to do is to have the bootloader recognize an ext2 > partition as opposed to a ufs partition (or possibly xiafs). There are a > number of reasons for this. Is anyone opposed? > No, I would prefer to use the ext2fs on a DEC too if I had such a box. It's the standard Linux fs, and actually quite necessary! Cheers, Andy ----------------------------------------------------------- Andreas Busse | andy@soft-n-hard.de Soft N Hard GbR | Phone: +49 2636-970105 Im Hufen Boden 16, D-53498 Waldorf | Fax: +49 2636-970106 ----------------------------------------------------------- From andy@soft-n-hard.de Mon Sep 25 15:38:54 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id PAA02198 for ; Mon, 25 Sep 1995 15:38:53 -0700 Received: from bnc.net by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA02541; Mon, 25 Sep 1995 14:46:53 +0100 (MET) Received: from newton.soft-n-hard.de ([194.163.66.33]) by bnc.net (8.6.11/8.6.9) with ESMTP id OAA12298 for ; Mon, 25 Sep 1995 14:48:22 +0100 Received: (from andy@localhost) by newton.soft-n-hard.de (8.6.12/8.6.9) id OAA22068; Mon, 25 Sep 1995 14:47:39 +0100 Date: Mon, 25 Sep 1995 14:47:39 +0100 From: Andreas Busse Message-Id: <199509251347.OAA22068@newton.soft-n-hard.de> To: linux-mips@fnet.fr Subject: Re: DECstation status In-Reply-To: <199509251339.PAA08033@bandsept.univ-evry.fr> X-Mailer: [XMailTool v3.1.2b] Content-Length: 610 Lines: 20 > > I got an unplugged personnal DEC 5000 station with Ultrix 4.4 on its > disks at about 50cm on my right hand, can it be of any help ?? :) > > Cheers, Stoned. > Not that I think you shouldn't do anything regarding the DEC port, but didn't you say you already have enough load ? :->>> Cheers, Andy ----------------------------------------------------------- Andreas Busse | andy@soft-n-hard.de Soft N Hard GbR | Phone: +49 2636-970105 Im Hufen Boden 16, D-53498 Waldorf | Fax: +49 2636-970106 ----------------------------------------------------------- From Stoned.Elipot@univ-evry.fr Mon Sep 25 15:48:30 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id PAA02315 for ; Mon, 25 Sep 1995 15:48:29 -0700 Received: from amiante.univ-evry.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA02762; Mon, 25 Sep 1995 14:56:34 +0100 (MET) Received: from bandsept.univ-evry.fr (bandsept.univ-evry.fr [192.93.214.19]) by amiante.univ-evry.fr with ESMTP (8.6.10/94.01.10); id OAA01207; Mon, 25 Sep 1995 14:57:10 +0100 Received: (seb@localhost) by bandsept.univ-evry.fr (8.6.12/94.01.26); id PAA08194; Mon, 25 Sep 1995 15:55:01 +0200 Message-Id: <199509251355.PAA08194@bandsept.univ-evry.fr> Date: Mon, 25 Sep 1995 15:55:01 +0200 From: Stoned Elipot To: linux-mips@fnet.fr In-Reply-To: <199509251347.OAA22068@newton.soft-n-hard.de> (message from Andreas Busse on Mon, 25 Sep 1995 14:47:39 +0100) Subject: Re: DECstation status X-Mailer: GNU Emacs [19.29] Mail Mode X-Attribution: seb X-Pgp-Key: ask a good public key server Reply-To: Stoned.Elipot@univ-evry.fr Content-Length: 600 Lines: 18 hi, >>>>> "Andreas" == Andreas Busse writes: [SNIP] Andreas> Not that I think you shouldn't do anything regarding the DEC Andreas> port, but didn't you say you already have enough load ? :->>> Oh yes, but not because of Linux/MIPS :( Ouin,... And besides this I know a little about disk swapping - it's boring: that was before I got a floppy drive and an transceiver for the Oily,.... yes I must admit it, I used vindoze NT to do the downloading of milo,kernel,... on this box, shame on me :) Well, that's why I think I can put on this little DEC... :) :) Cheers, Stoned. From marius@prilnari.com Mon Sep 25 16:17:00 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id QAA02592 for ; Mon, 25 Sep 1995 16:16:59 -0700 Received: from yakko.prilnari.com by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA03484; Mon, 25 Sep 1995 15:24:59 +0100 (MET) Received: (from marius@localhost) by yakko.prilnari.com (8.6.12/8.6.12) id KAA06896; Mon, 25 Sep 1995 10:25:27 -0400 Date: Mon, 25 Sep 1995 10:25:27 -0400 From: Matt Messier Subject: Re: DECstation status To: linux-mips@fnet.fr In-Reply-To: <199509251339.PAA08033@bandsept.univ-evry.fr> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 2348 Lines: 44 On Mon, 25 Sep 1995, Stoned Elipot wrote: > >>>>> "Matt" == Matt Messier writes: > > Matt> Does anybody out there with a DEC 3100 or 5000 have the ability > Matt> to test this stuff? It is a little bit difficult for me at the > Matt> moment because I need to swap drives around all of the time to > Matt> test things. I'm quite willing to work on writing the code if > Matt> someone can test it easily for me. Once things get a bit > Matt> further along and we know that it is working somewhat reasonably > Matt> well, I'll be a bit more willing to spend half an hour at a shot > Matt> to test a change :-) > > I got an unplugged personnal DEC 5000 station with Ultrix 4.4 on its > disks at about 50cm on my right hand, can it be of any help ?? :) Quite possibly :-) The only problem is that once you run this, you'll lose your Ultrix bootloader. Also, my program to install the bootloader is written for Linux/i386, but should work without much difficulty under Ultrix. The machine that I am working with has a barebones install of Ultrix 4.2 on it. I don't have a compiler, or much of anything for it for that matter. My primary means of getting things onto that machine is to nfs mount from my 486 Linux box. Once I start testing, I lose that ability and have to start swapping the drive back and forth. At any rate, I have determined that our cross development tools are producing binaries with valid headers. I haven't been able to get anything to execute in application mode, but I haven't tried real hard either. I am at work right now so I don't have time to put the stuff together for testing elsewhere, but I will do so tonight. By the way, where should I ftp this stuff to? What shall it be called, and should it be separated from Milo? Personally I think the answer to the Milo question is yes. The text and data segments need to fit within 15360 bytes. All that needs to happen in the loader (at least for now) is to load the kernel and jump to it's entry point. It also needs to properly handle prom restarts, etc. Fairly straight forward, but it also needs to have knowledge of the filesystem so that it can find the kernel. What I have now does little more than print some messages out to the console via the firmware callback vector. Matt Messier The Prilnari Project Lincoln, RI, USA From andy@soft-n-hard.de Mon Sep 25 16:25:44 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id QAA02689 for ; Mon, 25 Sep 1995 16:25:44 -0700 Received: from bnc.net by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA03626; Mon, 25 Sep 1995 15:33:42 +0100 (MET) Received: from newton.soft-n-hard.de ([194.163.66.33]) by bnc.net (8.6.11/8.6.9) with ESMTP id PAA13682 for ; Mon, 25 Sep 1995 15:35:30 +0100 Received: (from andy@localhost) by newton.soft-n-hard.de (8.6.12/8.6.9) id PAA22141; Mon, 25 Sep 1995 15:34:47 +0100 Date: Mon, 25 Sep 1995 15:34:47 +0100 From: Andreas Busse Message-Id: <199509251434.PAA22141@newton.soft-n-hard.de> To: linux-mips@fnet.fr Subject: Re: DECstation status In-Reply-To: X-Mailer: [XMailTool v3.1.2b] Content-Length: 1480 Lines: 30 > > I am at work right now so I don't have time to put the stuff together for > testing elsewhere, but I will do so tonight. By the way, where should I > ftp this stuff to? What shall it be called, and should it be separated from > Milo? Personally I think the answer to the Milo question is yes. The text > and data segments need to fit within 15360 bytes. All that needs to happen in > the loader (at least for now) is to load the kernel and jump to it's entry > point. It also needs to properly handle prom restarts, etc. Fairly straight > forward, but it also needs to have knowledge of the filesystem so that it can > find the kernel. What I have now does little more than print some messages > out to the console via the firmware callback vector. > As a temporary solution, please mail tar'ed gzip'ed and uuencoded files to Luc Beurton and tell him where to put the files. Regarding Milo or not: Name it DECLD or whatever, but at a certain point it would be a good idea to integrate your sources into the milo source tree, but not into milo itself. This way a build of milo can produce all necessary loaders, you know ? Cheers, Andy ----------------------------------------------------------- Andreas Busse | andy@soft-n-hard.de Soft N Hard GbR | Phone: +49 2636-970105 Im Hufen Boden 16, D-53498 Waldorf | Fax: +49 2636-970106 ----------------------------------------------------------- From marius@prilnari.com Mon Sep 25 16:37:31 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id QAA02814 for ; Mon, 25 Sep 1995 16:37:31 -0700 Received: from yakko.prilnari.com by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA03878; Mon, 25 Sep 1995 15:45:27 +0100 (MET) Received: (from marius@localhost) by yakko.prilnari.com (8.6.12/8.6.12) id KAA06958; Mon, 25 Sep 1995 10:46:03 -0400 Date: Mon, 25 Sep 1995 10:46:03 -0400 From: Matt Messier Subject: Re: DECstation status To: linux-mips@fnet.fr In-Reply-To: <199509251434.PAA22141@newton.soft-n-hard.de> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 1798 Lines: 36 On Mon, 25 Sep 1995, Andreas Busse wrote: > > I am at work right now so I don't have time to put the stuff together for > > testing elsewhere, but I will do so tonight. By the way, where should I > > ftp this stuff to? What shall it be called, and should it be separated from > > Milo? Personally I think the answer to the Milo question is yes. The text > > and data segments need to fit within 15360 bytes. All that needs to happen in > > the loader (at least for now) is to load the kernel and jump to it's entry > > point. It also needs to properly handle prom restarts, etc. Fairly straight > > forward, but it also needs to have knowledge of the filesystem so that it can > > find the kernel. What I have now does little more than print some messages > > out to the console via the firmware callback vector. > > > > As a temporary solution, please mail tar'ed gzip'ed and uuencoded > files to Luc Beurton and tell him where to put > the files. Ok. That works for me. I'll send some stuff his way tonight. Where should it be put (i.e., where should I tell him to put it)? > Regarding Milo or not: Name it DECLD or whatever, but at a certain > point it would be a good idea to integrate your sources into the > milo source tree, but not into milo itself. This way a build of > milo can produce all necessary loaders, you know ? Ok. For the time being I'm going to call it LinuxBoot following the UltrixBoot naming convention. This can change at any time. I'm not attached to that name in any way :-) Also makes sense to migrate it into the milo source tree eventually (once it is deemed to be working). For now I would prefer to keep it separate for ease of development and testing purposes. Matt Messier The Prilnari Project Lincoln, RI, USA From andy@soft-n-hard.de Mon Sep 25 16:45:49 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id QAA02903 for ; Mon, 25 Sep 1995 16:45:48 -0700 Received: from bnc.net by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA04051; Mon, 25 Sep 1995 15:53:45 +0100 (MET) Received: from newton.soft-n-hard.de ([194.163.66.33]) by bnc.net (8.6.11/8.6.9) with ESMTP id PAA14225 for ; Mon, 25 Sep 1995 15:55:24 +0100 Received: (from andy@localhost) by newton.soft-n-hard.de (8.6.12/8.6.9) id PAA22183; Mon, 25 Sep 1995 15:54:41 +0100 Date: Mon, 25 Sep 1995 15:54:41 +0100 From: Andreas Busse Message-Id: <199509251454.PAA22183@newton.soft-n-hard.de> To: linux-mips@fnet.fr Subject: Re: DECstation status In-Reply-To: X-Mailer: [XMailTool v3.1.2b] Content-Length: 1092 Lines: 29 Hi again! > > Ok. That works for me. I'll send some stuff his way tonight. Where should > it be put (i.e., where should I tell him to put it)? > As long it's in the beta (or alpha) phase, please put it in the private/Incoming directory, except you want it make public thru all mirror sites. In this case, somewhere in the src/boot tree. > Ok. For the time being I'm going to call it LinuxBoot following the > UltrixBoot naming convention. This can change at any time. I'm not > attached to that name in any way :-) Also makes sense to migrate it into the > milo source tree eventually (once it is deemed to be working). For now I > would prefer to keep it separate for ease of development and testing > purposes. Sure, no problem! Thanks for your effort! Cheers, Andy ----------------------------------------------------------- Andreas Busse | andy@soft-n-hard.de Soft N Hard GbR | Phone: +49 2636-970105 Im Hufen Boden 16, D-53498 Waldorf | Fax: +49 2636-970106 ----------------------------------------------------------- From marius@prilnari.com Mon Sep 25 15:23:20 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id PAA02020 for ; Mon, 25 Sep 1995 15:23:19 -0700 Received: from scotty.waldorf-gmbh.de ([193.98.169.1]) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA02193; Mon, 25 Sep 1995 14:31:18 +0100 (MET) Received: from yakko.prilnari.com by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id PAA04286; Mon, 25 Sep 1995 15:32:09 +0100 Received: (from marius@localhost) by yakko.prilnari.com (8.6.12/8.6.12) id IAA06768; Mon, 25 Sep 1995 15:23:20 -0500 Date: Mon, 25 Sep 1995 15:23:20 -0500 From: Matt Messier Subject: DECstation status To: Linux/MIPS Developers Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 1038 Lines: 22 This weekend I had an opportunity to get some work done on the DECstation port. Admittedly I haven't gotten too terribly far, but I've got basic boot code written. It is as yet untested, but I have written code to install the bootloader onto the scsi device, and I have gotten a start on the bootloader itself. Does anybody out there with a DEC 3100 or 5000 have the ability to test this stuff? It is a little bit difficult for me at the moment because I need to swap drives around all of the time to test things. I'm quite willing to work on writing the code if someone can test it easily for me. Once things get a bit further along and we know that it is working somewhat reasonably well, I'll be a bit more willing to spend half an hour at a shot to test a change :-) Also, what I would like to do is to have the bootloader recognize an ext2 partition as opposed to a ufs partition (or possibly xiafs). There are a number of reasons for this. Is anyone opposed? Matt Messier The Prilnari Project Lincoln, RI, USA From linux@mailhost.uni-koblenz.de Tue Sep 26 03:07:29 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id DAA10269 for ; Tue, 26 Sep 1995 03:07:29 -0700 Received: from infko.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA11708; Tue, 26 Sep 1995 02:16:14 +0100 (MET) Received: from goethe (linux@goethe.uni-koblenz.de [141.26.4.30]) by infko.uni-koblenz.de (8.6.9/8.6.9) with SMTP id CAA29429 for ; Tue, 26 Sep 1995 02:19:15 +0100 From: Systemkennung Linux Message-Id: <199509260119.CAA29429@infko.uni-koblenz.de> Received: by goethe (4.1/KO-2.0) id AA16425; Tue, 26 Sep 95 02:15:34 +0100 Subject: Re: DECstation status To: linux-mips@fnet.fr Date: Tue, 26 Sep 1995 02:15:34 +0100 (MET) In-Reply-To: from "Matt Messier" at Jan 1, 70 08:30:55 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 379 Lines: 11 Hi, > Also, what I would like to do is to have the bootloader recognize an ext2 > partition as opposed to a ufs partition (or possibly xiafs). There are a > number of reasons for this. Is anyone opposed? I suggest to use the libraries included with the e2fsutils for linux. The should provide enough and well tested code to manipulate ext2 filesystems for Linux. Ralf From andy@soft-n-hard.de Thu Sep 28 08:56:32 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id IAA06490 for ; Thu, 28 Sep 1995 08:56:32 +0100 Received: from bnc.net by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA13104; Thu, 28 Sep 1995 08:56:41 +0100 (MET) Received: from newton.soft-n-hard.de (soft-n-hard.de [194.163.66.33]) by bnc.net (8.6.11/8.6.9) with ESMTP id IAA11756 for ; Thu, 28 Sep 1995 08:56:37 +0100 Received: (from andy@localhost) by newton.soft-n-hard.de (8.6.12/8.6.9) id IAA25492; Thu, 28 Sep 1995 08:57:09 +0100 Date: Thu, 28 Sep 1995 08:57:09 +0100 From: Andreas Busse Message-Id: <199509280757.IAA25492@newton.soft-n-hard.de> To: linux-mips@fnet.fr Subject: Re: Cross compilers for MIPS NT? In-Reply-To: <199509151608.KAA02564@rover.village.org> X-Mailer: [XMailTool v3.1.2b] Content-Length: 560 Lines: 20 > > I didn't see one last time I looked, but... > > Has any body brewed a cross compiler for MIPS NT 3.5? > > Warner > Did you notice that there's some limited support for Win-NT in gcc-2.7.0? Only as native compiler for Alpha and x86, but... Andy ----------------------------------------------------------- Andreas Busse | andy@soft-n-hard.de Soft N Hard GbR | Phone: +49 2636-970105 Im Hufen Boden 16, D-53498 Waldorf | Fax: +49 2636-970106 ----------------------------------------------------------- From andy@soft-n-hard.de Thu Sep 28 10:57:06 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id KAA11197 for ; Thu, 28 Sep 1995 10:57:04 +0100 Received: from bnc.net by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA15272; Thu, 28 Sep 1995 10:57:09 +0100 (MET) Received: from newton.soft-n-hard.de (soft-n-hard.de [194.163.66.33]) by bnc.net (8.6.11/8.6.9) with ESMTP id KAA14873 for ; Thu, 28 Sep 1995 10:56:24 +0100 Received: (from andy@localhost) by newton.soft-n-hard.de (8.6.12/8.6.9) id KAA25817; Thu, 28 Sep 1995 10:56:54 +0100 Date: Thu, 28 Sep 1995 10:56:54 +0100 From: Andreas Busse Message-Id: <199509280956.KAA25817@newton.soft-n-hard.de> To: linux-mips@fnet.fr Subject: /bin/sh for DOS ? X-Mailer: [XMailTool v3.1.2b] Content-Length: 431 Lines: 16 Hi all! Yes, I really ask for a sh-alike for DOS, preferably free or GPLed. If someone knows about, please tell me! Thanks, Andy ----------------------------------------------------------- Andreas Busse | andy@soft-n-hard.de Soft N Hard GbR | Phone: +49 2636-970105 Im Hufen Boden 16, D-53498 Waldorf | Fax: +49 2636-970106 ----------------------------------------------------------- From jens@jens1.whu-koblenz.de Thu Sep 28 11:22:10 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id LAA12963 for ; Thu, 28 Sep 1995 11:22:08 +0100 Received: from obelix.WHU-Koblenz.DE by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA15769; Thu, 28 Sep 1995 11:22:20 +0100 (MET) Received: from jens1.whu-koblenz.de (jens@jens1.WHU-Koblenz.DE [193.141.48.51]) by obelix.WHU-Koblenz.de (8.6.8.1/8.6.4) with ESMTP id LAA21077 for ; Thu, 28 Sep 1995 11:22:30 +0100 Received: (from jens@localhost) by jens1.whu-koblenz.de (8.6.12/8.6.9) id LAA03043 for linux-mips@fnet.fr; Thu, 28 Sep 1995 11:19:13 +0100 From: Jens Hoffmann Message-Id: <199509281019.LAA03043@jens1.whu-koblenz.de> Subject: Re: /bin/sh for DOS ? To: linux-mips@fnet.fr Date: Thu, 28 Sep 1995 11:19:09 +0100 (MET) In-Reply-To: <199509280956.KAA25817@newton.soft-n-hard.de> from "Andreas Busse" at Sep 28, 95 10:56:54 am X-Mailer: ELM [version 2.4 PL24] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 371 Lines: 17 Hi Andy, Andreas Busse wrote: > Hi all! > Yes, I really ask for a sh-alike for DOS, preferably > free or GPLed. If someone knows about, please tell > me! There is a package from PVS (sp?). It has a complete (around 300 utilities) set of Unix-style commands. I think it is primarily marketed as a revision control system :-)) There is a ksh in it! Greetings, Jens From andy@soft-n-hard.de Thu Sep 28 11:30:43 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id LAA13060 for ; Thu, 28 Sep 1995 11:30:41 +0100 Received: from bnc.net by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA15968; Thu, 28 Sep 1995 11:30:45 +0100 (MET) Received: from newton.soft-n-hard.de (soft-n-hard.de [194.163.66.33]) by bnc.net (8.6.11/8.6.9) with ESMTP id LAA15911 for ; Thu, 28 Sep 1995 11:30:36 +0100 Received: (from andy@localhost) by newton.soft-n-hard.de (8.6.12/8.6.9) id LAA25979; Thu, 28 Sep 1995 11:31:05 +0100 Date: Thu, 28 Sep 1995 11:31:05 +0100 From: Andreas Busse Message-Id: <199509281031.LAA25979@newton.soft-n-hard.de> To: linux-mips@fnet.fr Subject: Re: /bin/sh for DOS ? In-Reply-To: <199509281019.LAA03043@jens1.whu-koblenz.de> X-Mailer: [XMailTool v3.1.2b] Content-Length: 592 Lines: 21 Hi, > There is a package from PVS (sp?). > It has a complete (around 300 utilities) set of Unix-style commands. > I think it is primarily marketed as a revision control system :-)) > > There is a ksh in it! > What is PVS ? Where can I get it ? What does it cost ? :-) Cheers, Andy ----------------------------------------------------------- Andreas Busse | andy@soft-n-hard.de Soft N Hard GbR | Phone: +49 2636-970105 Im Hufen Boden 16, D-53498 Waldorf | Fax: +49 2636-970106 ----------------------------------------------------------- From kah@terma.dk Thu Sep 28 14:32:23 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id OAA14963 for ; Thu, 28 Sep 1995 14:32:18 +0100 Received: from ns.dknet.dk by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA19629; Thu, 28 Sep 1995 14:32:24 +0100 (MET) Received: from terma by ns.dknet.dk with UUCP id AA14012 (5.65c8/IDA-1.4.4j for linux-mips@fnet.fr); Thu, 28 Sep 1995 14:32:16 +0100 Message-Id: <199509281332.AA14012@ns.dknet.dk> Received: from lulu.terma.dk (lulu) by vera.terma.dk; Thu, 28 Sep 1995 14:02:40 +0200 Received: by lulu.terma.dk (1.38.193.4/16.2) id AA00470; Thu, 28 Sep 1995 12:49:32 +0100 From: Kai Harrekilde-Petersen Subject: Re: /bin/sh for DOS ? To: andy@soft-n-hard.de Date: Thu, 28 Sep 95 12:49:31 MET Cc: linux-mips@fnet.fr (Linux/MIPS mailing list) In-Reply-To: from "uucp" at Sep 28, 95 1:30 pm Mailer: Elm [revision: 70.85] X-Charset: ASCII X-Char-Esc: 29 Content-Length: 733 Lines: 18 > > There is a package from PVS (sp?). > > It has a complete (around 300 utilities) set of Unix-style commands. > > I think it is primarily marketed as a revision control system :-)) > > > > There is a ksh in it! > > > What is PVS ? Where can I get it ? What does it cost ? :-) I think you're talking of PCVS (or is it called PCMS? can't remember). I have a "backup copy" somewhere in my ancient stack of 5 1/4" 360Kb disks at home, that I got from a company that I worked for in '89. Kai -- Kai Harrekilde-Petersen Linux: The choice of a GNU generation Terma Elektronik, Surveillance Systems Division #include PGP Key fingerprint = 9A 70 8A FF A3 31 3E DC FB 9C A2 50 D7 FC 4F C8 From pab@uni.net Thu Sep 28 14:51:59 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id OAA15184 for ; Thu, 28 Sep 1995 14:51:59 +0100 Received: from snoop.uni.net by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA20059; Thu, 28 Sep 1995 14:52:12 +0100 (MET) Received: from pag (pag.RMnet.it [194.20.98.41]) by snoop.uni.net (8.6.11/8.6.9) with SMTP id OAA06117; Thu, 28 Sep 1995 14:48:00 +0100 Date: Thu, 28 Sep 1995 14:48:00 +0100 Message-Id: <199509281348.OAA06117@snoop.uni.net> X-Sender: pab@relay.uni.net X-Mailer: Windows Eudora Light Version 1.5.2b1 Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" To: linux-mips@fnet.fr, linux-mips@fnet.fr From: Paolo Bevilacqua Subject: GCC 2.7.0 (Was: Cross compilers for MIPS NT?) Content-Length: 1477 Lines: 43 At 08:57 AM 9/28/95 +0100, Andreas Busse wrote: > > > > > I didn't see one last time I looked, but... > > > > Has any body brewed a cross compiler for MIPS NT 3.5? > > > > Warner > > > >Did you notice that there's some limited support for >Win-NT in gcc-2.7.0? Only as native compiler for Alpha >and x86, but... > > I've installed GCC 2.7.0 on my 486 box and had a lot of trouble compiling a 1.2.13 kernel. After spending a lot of time removing leading underscores in assembler symbols to match his idea 'ld' outputted and ELF image that xtract refuse to deal with. I've the added -oformat a.out-i386-link to the linker options, but now it says that this would output an unrepresentable section, and refuses to dump a such (oscene in my toughts) image. There is to say that i'm totally dummie for GNUism, may be i was evertytime very lucky in past ? And Unidata hardware trashing departement striked again. Now there is an Alpha (Jensen, not -that- Jensen :) board floating around. Gee, i thin i will go soon (6 months) to crash some kernel on it. Final subject: I will be in Amsterdam 11,12,13 october for a RIPE meeting. The meeting itself is usually boring (except that i will introduce the new NAP in Rome :), but the city isn't. If you are in floating beer mood, whistle. Would be nice to meet all you fine hacker in a recreactional context. Cheers all! /pab -- Paolo Bevilacqua RMnet Communications ph/fax: +39 6 85302774 http://www.rmnet.it From Stoned.Elipot@univ-evry.fr Thu Sep 28 16:14:09 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id QAA16226 for ; Thu, 28 Sep 1995 16:14:09 +0100 Received: from amiante.univ-evry.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA21661; Thu, 28 Sep 1995 16:14:23 +0100 (MET) Received: from bandsept.univ-evry.fr (bandsept.univ-evry.fr [192.93.214.19]) by amiante.univ-evry.fr with ESMTP (8.6.10/94.01.10); id QAA08349; Thu, 28 Sep 1995 16:14:25 +0100 Received: (seb@localhost) by bandsept.univ-evry.fr (8.6.12/94.01.26); id QAA17226; Thu, 28 Sep 1995 16:14:24 +0200 Message-Id: <199509281414.QAA17226@bandsept.univ-evry.fr> Date: Thu, 28 Sep 1995 16:14:24 +0200 From: Stoned Elipot To: linux-mips@fnet.fr Subject: test X-Mailer: GNU Emacs [19.29] Mail Mode X-Attribution: seb X-Pgp-Key: ask a good public key server Reply-To: Stoned.Elipot@univ-evry.fr Content-Length: 52 Lines: 3 Just testing brand new archiving scheme :) Stined. From andy@soft-n-hard.de Thu Sep 28 15:23:20 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id PAA15574 for ; Thu, 28 Sep 1995 15:23:19 +0100 Received: from bnc.net by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA20581; Thu, 28 Sep 1995 15:23:25 +0100 (MET) Received: from newton.soft-n-hard.de (soft-n-hard.de [194.163.66.33]) by bnc.net (8.6.11/8.6.9) with ESMTP id PAA21610 for ; Thu, 28 Sep 1995 15:23:23 +0100 Received: (from andy@localhost) by newton.soft-n-hard.de (8.6.12/8.6.9) id PAA26582; Thu, 28 Sep 1995 15:23:53 +0100 Date: Thu, 28 Sep 1995 15:23:53 +0100 From: Andreas Busse Message-Id: <199509281423.PAA26582@newton.soft-n-hard.de> To: linux-mips@fnet.fr Subject: Re: GCC 2.7.0 (Was: Cross compilers for MIPS NT?) In-Reply-To: <199509281348.OAA06117@snoop.uni.net> X-Mailer: [XMailTool v3.1.2b] Content-Length: 1709 Lines: 36 > > I've installed GCC 2.7.0 on my 486 box and had a lot of trouble compiling a > 1.2.13 kernel. After spending a lot of time removing leading underscores in > assembler symbols to match his idea 'ld' outputted and ELF image that xtract > refuse to deal with. I've the added -oformat a.out-i386-link to the linker > options, but now it says that this would output an unrepresentable section, > and refuses to dump a such (oscene in my toughts) image. There is to say > that i'm totally dummie for GNUism, may be i was evertytime very lucky in past ? > Hmmm. I must say that gcc-2.7.0-3 and binutils 2.5.2-6, ie. the FSF releases plus the Linux/MIPS patches from Ralf work really fine as cross compilers from Linux to DOS (yeah!) and Linux/MIPS. I had lots more trouble with any other release. > And Unidata hardware trashing departement striked again. Now there is an > Alpha (Jensen, not -that- Jensen :) board floating around. Gee, i thin i > will go soon (6 months) to crash some kernel on it. > > Final subject: > I will be in Amsterdam 11,12,13 october for a RIPE meeting. The meeting > itself is usually boring (except that i will introduce the new NAP in Rome > :), but the city isn't. If you are in floating beer mood, whistle. Would be > nice to meet all you fine hacker in a recreactional context. Amsterdam. Hmm, that's not that far away from here. So why not! Cheers, Andy ----------------------------------------------------------- Andreas Busse | andy@soft-n-hard.de Soft N Hard GbR | Phone: +49 2636-970105 Im Hufen Boden 16, D-53498 Waldorf | Fax: +49 2636-970106 ----------------------------------------------------------- From Stoned.Elipot@univ-evry.fr Thu Sep 28 18:24:30 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id SAA17793 for ; Thu, 28 Sep 1995 18:24:29 +0100 Received: from amiante.univ-evry.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA24325; Thu, 28 Sep 1995 18:24:42 +0100 (MET) Received: from bandsept.univ-evry.fr (bandsept.univ-evry.fr [192.93.214.19]) by amiante.univ-evry.fr with ESMTP (8.6.10/94.01.10); id SAA08616; Thu, 28 Sep 1995 18:24:43 +0100 Received: (seb@localhost) by bandsept.univ-evry.fr (8.6.12/94.01.26); id SAA17837; Thu, 28 Sep 1995 18:24:42 +0200 Message-Id: <199509281624.SAA17837@bandsept.univ-evry.fr> Date: Thu, 28 Sep 1995 18:24:42 +0200 From: Stoned Elipot To: linux-mips@fnet.fr Subject: test2 X-Mailer: GNU Emacs [19.29] Mail Mode X-Attribution: seb X-Pgp-Key: ask a good public key server Reply-To: Stoned.Elipot@univ-evry.fr Content-Length: 31 Lines: 2 Just another, sorry :) Stoned. From andy@soft-n-hard.de Thu Sep 28 18:32:34 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id SAA17938 for ; Thu, 28 Sep 1995 18:32:34 +0100 Received: from bnc.net by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA24462; Thu, 28 Sep 1995 18:32:33 +0100 (MET) Received: from newton.soft-n-hard.de (soft-n-hard.de [194.163.66.33]) by bnc.net (8.6.11/8.6.9) with ESMTP id SAA26187 for ; Thu, 28 Sep 1995 18:32:31 +0100 Received: (from andy@localhost) by newton.soft-n-hard.de (8.6.12/8.6.9) id SAA27705; Thu, 28 Sep 1995 18:33:01 +0100 Date: Thu, 28 Sep 1995 18:33:01 +0100 From: Andreas Busse Message-Id: <199509281733.SAA27705@newton.soft-n-hard.de> To: linux-mips@fnet.fr Subject: Re: /bin/sh for DOS ? In-Reply-To: <199509281332.AA14012@ns.dknet.dk> X-Mailer: [XMailTool v3.1.2b] Content-Length: 613 Lines: 16 > I have a "backup copy" somewhere in my ancient stack of 5 1/4" 360Kb > disks at home, that I got from a company that I worked for in '89. Well, I hacked a /bin/ash for MSDOS. Doesn't do much yet, but at least poor man's ls (echo *) works already :-) And it prints the DOS environment variables! Cheers, Andy ----------------------------------------------------------- Andreas Busse | andy@soft-n-hard.de Soft N Hard GbR | Phone: +49 2636-970105 Im Hufen Boden 16, D-53498 Waldorf | Fax: +49 2636-970106 ----------------------------------------------------------- From ralf@waldorf-gmbh.de Thu Sep 28 22:03:06 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id WAA20210 for ; Thu, 28 Sep 1995 22:03:05 +0100 Received: from scotty.waldorf-gmbh.de ([193.98.169.1]) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA29042; Thu, 28 Sep 1995 22:03:15 +0100 (MET) From: Ralf Baechle Message-Id: <199509282102.WAA12156@scotty.waldorf-gmbh.de> Received: from localhost by scotty.waldorf-gmbh.de (8.6.4/WE-1.0.1) id WAA12156; Thu, 28 Sep 1995 22:02:39 +0100 Subject: Re: GCC 2.7.0 (Was: Cross compilers for MIPS NT?) To: linux-mips@fnet.fr Date: Thu, 28 Sep 1995 22:02:39 +0100 (MET) In-Reply-To: <199509281348.OAA06117@snoop.uni.net> from "Paolo Bevilacqua" at Sep 28, 95 02:48:00 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1561 Lines: 30 Hi, > I've installed GCC 2.7.0 on my 486 box and had a lot of trouble compiling a > 1.2.13 kernel. After spending a lot of time removing leading underscores in > assembler symbols to match his idea 'ld' outputted and ELF image that xtract > refuse to deal with. I've the added -oformat a.out-i386-link to the linker > options, but now it says that this would output an unrepresentable section, > and refuses to dump a such (oscene in my toughts) image. There is to say > that i'm totally dummie for GNUism, may be i was evertytime very lucky in > past ? Sounds like you were using a Linux/i386 box. From release 2.7.0 on GCC configures to use ELF output for i[345]86-*-linux* and to use a.out only for i[345]86-*-linuxaout* targets. This is due to the change to ELF for all Linux implementations. If you want to switch to ELF you have to install Libc binaries, eventually ld.so and binutils. There is a good chance to nuke the systems for people that aren't used tweaking that part of the system. Currently an fullautomatic kit for conversion to ELF is being developed. For more information see the GCC mailing list. > Final subject: > I will be in Amsterdam 11,12,13 october for a RIPE meeting. The meeting > itself is usually boring (except that i will introduce the new NAP in Rome > :), but the city isn't. If you are in floating beer mood, whistle. Would be > nice to meet all you fine hacker in a recreactional context. Sounds like I'll have to travel to Amsterdam in a recreactional context ... Who else ist intrested to come with me? Ralf From pab@RMnet.IT Thu Sep 28 23:03:55 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id XAA20853 for ; Thu, 28 Sep 1995 23:03:52 +0100 Received: from malcolm.RMnet.it by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA00248; Thu, 28 Sep 1995 23:03:58 +0100 (MET) Received: from pag (pag.RMnet.it [194.20.98.41]) by malcolm.rmnet.it (8.6.11/8.6.9) with SMTP id UAA21465 for ; Thu, 28 Sep 1995 20:03:59 -0400 From: "Paolo Bevilacqua" Message-Id: <950928230125.ZM4799@pag> Date: Thu, 28 Sep 1995 23:01:24 +0100 In-Reply-To: Ralf Baechle "Re: GCC 2.7.0 (Was: Cross compilers for MIPS NT?)" (Sep 28, 10:02pm) References: <199509282102.WAA12156@scotty.waldorf-gmbh.de> X-Mailer: Z-Mail 4.0 (4.0.0 Aug 21 1995) To: linux-mips@fnet.fr Subject: Re: GCC 2.7.0 (Was: Cross compilers for MIPS NT?) Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Length: 984 Lines: 27 On Sep 28, 10:02pm, Ralf Baechle wrote: > > Sounds like you were using a Linux/i386 box. From release 2.7.0 on GCC > configures to use ELF output for i[345]86-*-linux* and to use a.out only > for i[345]86-*-linuxaout* targets. This is due to the change to ELF for > all Linux implementations. If you want to switch to ELF you have to > install Libc binaries, eventually ld.so and binutils. There is a good > chance to nuke the systems for people that aren't used tweaking that > part of the system. Currently an fullautomatic kit for conversion to ELF > is being developed. For more information see the GCC mailing list. > Yes, it is an Intel box. I've installed all the suite you recommended and user stuff compiles and runs Ok. Only the kernel gave me this problem. But doesn't matter, i'll stick with the current one :) > Sounds like I'll have to travel to Amsterdam in a recreactional context ... > Who else ist intrested to come with me? > yeah, let's amsty! From ralf@waldorf-gmbh.de Sat Sep 30 23:36:39 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id XAA24861 for ; Sat, 30 Sep 1995 23:36:39 +0100 Received: from rio.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA16460; Sat, 30 Sep 1995 23:36:29 +0100 (MET) From: ralf@waldorf-gmbh.de (Ralf Baechle) Message-Id: <199509290248.DAA00822@rio.waldorf-gmbh.de> Received: by rio.waldorf-gmbh.de (8.6.11/GEN-1.0.8) id DAA00822; Fri, 29 Sep 1995 03:48:21 +0100 Subject: Linux 1.3.0 To: linux-mips@fnet.fr Date: Fri, 29 Sep 1995 03:48:19 +0100 (MET) Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 805 Lines: 18 Hi, I'm about to do the switch to the 1.3 kernel series. 1.3 introduced per some new architecure specific file that open an elegant way to integrate compatibility with other operating system. I've choosen to make Linux/MIPS as compatible reasonable and possible with MIPS ABI compliant systems and IRIX. The idea is to get binary compatible with these operating systems. The required changes of structures will break the compatibility with the old binaries but I think it will be worth it. Due to the switch to ELF this shouldn't hurt that much anyway ... I'll try to release the 1.3 series kernels in a series as fast as possible, so don't really try to run one of these kernels; I won't introduce new features or do real debugging before I catched up the leap to Linus' kernel releases. Ralf From jens@jens1.whu-koblenz.de Fri Sep 29 10:52:50 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id KAA29614 for ; Fri, 29 Sep 1995 10:52:49 +0100 Received: from obelix.WHU-Koblenz.DE by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA12569; Fri, 29 Sep 1995 10:53:01 +0100 (MET) Received: from jens1.whu-koblenz.de (jens@jens1.WHU-Koblenz.DE [193.141.48.51]) by obelix.WHU-Koblenz.de (8.6.8.1/8.6.4) with ESMTP id KAA23884 for ; Fri, 29 Sep 1995 10:53:00 +0100 Received: (from jens@localhost) by jens1.whu-koblenz.de (8.6.12/8.6.9) id KAA03465 for linux-mips@fnet.fr; Fri, 29 Sep 1995 10:49:52 +0100 From: Jens Hoffmann Message-Id: <199509290949.KAA03465@jens1.whu-koblenz.de> Subject: Re: /bin/sh for DOS ? To: linux-mips@fnet.fr Date: Fri, 29 Sep 1995 10:49:47 +0100 (MET) In-Reply-To: <199509281332.AA14012@ns.dknet.dk> from "Kai Harrekilde-Petersen" at Sep 28, 95 12:49:31 pm X-Mailer: ELM [version 2.4 PL24] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 432 Lines: 16 Hi Kai, Kai Harrekilde-Petersen wrote: > > What is PVS ? Where can I get it ? What does it cost ? :-) > I think you're talking of PCVS (or is it called PCMS? can't remember). > I have a "backup copy" somewhere in my ancient stack of 5 1/4" 360Kb > disks at home, that I got from a company that I worked for in '89. We do mean the same package! (But there are newer versions, I am sure). Hmm. Wasn't it from MKS? Gruss, Jens From andy@soft-n-hard.de Fri Sep 29 10:58:31 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id KAA29709 for ; Fri, 29 Sep 1995 10:58:30 +0100 Received: from bnc.net by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA12662; Fri, 29 Sep 1995 10:58:39 +0100 (MET) Received: from newton.soft-n-hard.de (soft-n-hard.de [194.163.66.33]) by bnc.net (8.6.11/8.6.9) with ESMTP id KAA20002 for ; Fri, 29 Sep 1995 10:58:35 +0100 Received: (from andy@localhost) by newton.soft-n-hard.de (8.6.12/8.6.9) id KAA31948; Fri, 29 Sep 1995 10:59:04 +0100 Date: Fri, 29 Sep 1995 10:59:04 +0100 From: Andreas Busse Message-Id: <199509290959.KAA31948@newton.soft-n-hard.de> To: linux-mips@fnet.fr Subject: Re: /bin/sh for DOS ? In-Reply-To: <199509290949.KAA03465@jens1.whu-koblenz.de> X-Mailer: [XMailTool v3.1.2b] Content-Length: 480 Lines: 18 > > We do mean the same package! (But there are newer versions, I am sure). > Hmm. > > Wasn't it from MKS? > Possibly yes. However, I really prefer sources. Cheers, Andy ----------------------------------------------------------- Andreas Busse | andy@soft-n-hard.de Soft N Hard GbR | Phone: +49 2636-970105 Im Hufen Boden 16, D-53498 Waldorf | Fax: +49 2636-970106 ----------------------------------------------------------- From kah@terma.dk Fri Sep 29 15:33:17 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id PAA01377 for ; Fri, 29 Sep 1995 15:33:16 +0100 Received: from ns.dknet.dk by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA18039; Fri, 29 Sep 1995 15:33:20 +0100 (MET) Received: from terma by ns.dknet.dk with UUCP id AA03361 (5.65c8/IDA-1.4.4j for linux-mips@fnet.fr); Fri, 29 Sep 1995 15:33:00 +0100 Message-Id: <199509291433.AA03361@ns.dknet.dk> Received: from lulu.terma.dk (lulu) by vera.terma.dk; Fri, 29 Sep 1995 15:32:17 +0200 Received: by lulu.terma.dk (1.38.193.4/16.2) id AA04502; Fri, 29 Sep 1995 14:19:00 +0100 From: Kai Harrekilde-Petersen Subject: Re: /bin/sh for DOS? To: andy@soft-n-hard.de (Andreas Busse) Date: Fri, 29 Sep 95 14:18:59 MET Cc: linux-mips@fnet.fr (Linux/MIPS mailing list) Mailer: Elm [revision: 70.85] X-Charset: ASCII X-Char-Esc: 29 Content-Length: 1039 Lines: 33 -----BEGIN PGP SIGNED MESSAGE----- Jens wrote: > > We do mean the same package! (But there are newer versions, I am sure). > > Hmm. > > > > Wasn't it from MKS? > > I have no recollection of which SW company it was from. > Possibly yes. However, I really prefer sources. As promised in private email, I looked through my archive of GNUish DOS, and the Bash that I have there seems to be a vanilla bash v1.12 source dist (~860Kb tar.gz). I'll look at the shell/file utils package during the weekend. Kai - -- Kai Harrekilde-Petersen Linux: The choice of a GNU generation Terma Elektronik, Surveillance Systems Division #include PGP Key fingerprint = 9A 70 8A FF A3 31 3E DC FB 9C A2 50 D7 FC 4F C8 -----BEGIN PGP SIGNATURE----- Version: 2.6.2 iQCVAgUBMGvyN/fpFDG15Qu5AQHFzQP+KNp7XONuiL/bxDaNXBlmVhmm9qRQh7gH oKPrlU0gFHLrqVPG5V2BI75QpU0k4qutzIX2hJLugJNw+DDs9w1Y03gMsDPrpyzU l5EtBCdoZcYcSlFtFZMoovbXT98KfGTxl58HkIK627zlMXSIqi4OLq3oq30UXGtc JcjhQTz9LU8= =p8C8 -----END PGP SIGNATURE----- From ralf@waldorf-gmbh.de Sat Sep 30 23:36:00 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id XAA24810 for ; Sat, 30 Sep 1995 23:35:59 +0100 Received: from rio.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA16359; Sat, 30 Sep 1995 23:25:24 +0100 (MET) From: ralf@waldorf-gmbh.de (Ralf Baechle) Message-Id: <199509302107.WAA05469@rio.waldorf-gmbh.de> Received: by rio.waldorf-gmbh.de (8.6.11/GEN-1.0.8) id WAA05469; Sat, 30 Sep 1995 22:07:35 +0100 Subject: Linux 1.3.0 To: linux-mips@fnet.fr Date: Sat, 30 Sep 1995 22:07:35 +0100 (MET) Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 777 Lines: 25 -----BEGIN PGP SIGNED MESSAGE----- Hi all, I've uploaded Linux 1.3.0 to ftp.fnet.fr. As I'm currently trying to catch up with Linus' kernel releases I didn't really test this one out; all I guarantee for is that it compiles on my machine. Consider 1.3.0 a snapshot of what I'm working on. I won't upload a diff file since this is almost senseless due to the large size (2.5mb unpacked/664kb packed). Happy hacking, Ralf d917caaee5a93bf9f4d48842f6528ddc linux-1.3.0.tar.gz -----BEGIN PGP SIGNATURE----- Version: 2.6.2i iQCVAwUBMG2xlEckbl6vezDBAQHsKwP/bjaLE6cN2MDnrG7Mw/p8TNAgg8zxJO3T MBnSvXm915/Cfa5b17iXaV5rdzOlsdrj9Rqtor+Z5FVLK0T7ezybrsv3StznNcOO PHLZVTyLZy63Bk6JbUC2IiirTK3k3JY5Z9iineYmgAz4oa9PVCeDk4Tp70ixujeJ dsyl+NfRA8U= =FjEq -----END PGP SIGNATURE----- From ralf@waldorf-gmbh.de Sat Sep 30 23:36:31 1995 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by pactole.fnet.fr (8.6.12/8.6.12) with SMTP id XAA24842 for ; Sat, 30 Sep 1995 23:36:31 +0100 Received: from chenas.inria.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA16464; Sat, 30 Sep 1995 23:36:46 +0100 (MET) Received: from rio.waldorf-gmbh.de by chenas.inria.fr (5.65c8d/92.02.29) via EUnet-France id AA18959; Sat, 30 Sep 1995 23:29:27 +0100 (MET) From: ralf@waldorf-gmbh.de (Ralf Baechle) Message-Id: <199509302216.XAA06763@rio.waldorf-gmbh.de> Received: by rio.waldorf-gmbh.de (8.6.11/GEN-1.0.8) id XAA06763; Sat, 30 Sep 1995 23:16:47 +0100 Subject: Linux 1.3.1 To: linux-mips@fnet.fr Date: Sat, 30 Sep 1995 23:16:45 +0100 (MET) Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1048 Lines: 36 -----BEGIN PGP SIGNED MESSAGE----- Hi all, to miscite someone - Is it a bird ... Is it a plane ... No it's KernelMan, faster than a speeding bullet, to your rescue. [actually it's more than a hour so I'm allowed to ...] I'm uploading the kernel patch for Linux 1.3.1 to ftp.fnet.fr. As for 1.3.0 the same non-guarantee applies. Stoned, Luc: I didn't regularily upload a full source tree to ftp.waldorf-gmbh.de to keep both diskspace and bandwith consumption by mirroring low but I think most people will prefer not having to apply more than a dozen patches in a row to build an uptodate kernel source. How much is diskspace/bandwith consumption a consideration on ftp.fnet.fr? Ralf 038e8d74cea12a8302f9d4ce2e3c273b linux-1.3.0-1.3.1.diffs.gz -----BEGIN PGP SIGNATURE----- Version: 2.6.2i iQCVAwUBMG3Bykckbl6vezDBAQGwdQP/dVhTFrgbuDPATCQSV94UTVzkJsU7VmK2 KXJeLBvck+w4dOAGpXHouDJ1y+5PBvIJ2ZyLA9yBEprWeocnq6nrDwHPnjqtxQVN L8jjMePmf2m8H6JFVhj9axlbLPQuEPsclX86VFQlX0LXPN94G7vuOabMNYdRfR31 KYbr2JJtZM4= =lSJ9 -----END PGP SIGNATURE-----