From linux@informatik.uni-koblenz.de Thu May 1 02:00:44 1997 Received-Date: Thu, 1 May 1997 02:00:42 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA09301; Thu, 1 May 1997 01:57:22 +0200 (MET) Received: from thoma (linux@thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.8.5/8.6.9) with SMTP id CAA17267 for ; Thu, 1 May 1997 02:00:30 +0200 (MEST) From: Systemkennung Linux Message-Id: <199705010000.CAA17267@informatik.uni-koblenz.de> Received: by thoma (SMI-8.6/KO-2.0) id CAA07117; Thu, 1 May 1997 02:00:26 +0200 Subject: Re: DECStation update... To: linux-mips@guadalquivir.fnet.fr Date: Thu, 1 May 1997 02:00:25 +0200 (MET DST) In-Reply-To: <199704302222.WAA08701@suede.sw.oz.au> from "Paul Antoine" at May 1, 97 08:22:40 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 389 Lines: 11 Hi, > Would you please check that it boots at least to the 'Added tags' point in > the kernel? It's not getting past the code at kernel_entry in head.S due to > R4K'isms in that portion of the code (due to R4K coders setting the assembler > to 'noreorder' then assuming all MIPS CPU's have 'some' interlocks as does > the R4K... :-) Could you tell me where these R4k'isms are? Ralf From linux@informatik.uni-koblenz.de Thu May 1 02:38:59 1997 Received-Date: Thu, 1 May 1997 02:38:52 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA10325; Thu, 1 May 1997 02:35:25 +0200 (MET) Received: from thoma (linux@thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.8.5/8.6.9) with SMTP id CAA24300 for ; Thu, 1 May 1997 02:38:34 +0200 (MEST) From: Systemkennung Linux Message-Id: <199705010038.CAA24300@informatik.uni-koblenz.de> Received: by thoma (SMI-8.6/KO-2.0) id CAA07130; Thu, 1 May 1997 02:38:31 +0200 Subject: Re: Linux/MIPS 2.1.36 uploaded To: linux-mips@guadalquivir.fnet.fr Date: Thu, 1 May 1997 02:38:29 +0200 (MET DST) In-Reply-To: <199704302246.WAA08836@suede.sw.oz.au> from "Paul Antoine" at May 1, 97 08:46:07 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 2180 Lines: 50 Hi, > > The number of changes compared to my last real release 2.1.14 is too > > large to be enumerated and people still having patches against this > > release will probably have their fun with them ... > > Arrrrrgggggghhhhhhh! Can you highlight the big changes from 2.1.14?? I > had foolishly assumed that your next release would be closer to 2.1.14 than > the previous 'official' release... Not at all. David did invest quite a lot of work into the low level mm stuff and did a nice job in un-disgustifing (head.S ...) the code. Some of the changes: - partional Indy support is in. - IRIX binary compatility code is in. - massive cleanup; lots of code has been moved around. - Support for R4000/R4400 SC/MC style second level caches. This means Linux should now be working (modulo a stupid bug that kills all machines that are wireing TLB entries on startup.) for machines like the Mips Magnum 4000SC and the Millenium. We should get rid of these mappings anyway; TLB entries are a very scarce resource and we should not waste them for hardware we access just now and then. Things are different on the R3000 where we always have 8 wired entries but on the R4k where we can select the number ourselves we want as few of them as possible. - All the 64bit-isms for R4k kernels are out. I'll add them back somewhen later. - PCI support. - ptrace(2) support is now in. GDB to come rsn. - Linus' patches upto 2.1.36 applied. Kernel NFS server is my favourite inovation :-) - Lots of changes to the code, especially the IRIX compat code code to adapt to the new mechanisms of Linux 2.1.x for accessing userspace from the kernel space. - Introduced kernel locking in the MIPS code as required for SMP. > Please tell me you won't do another release for a little while... so I can > get things stable on the DS's again and send you the code... :-) Oh, I'm just waiting for Linus to release 2.1.37. Even more more fun :-) The good news for you is that I'll be in the Silicon Valley next week, so use this week ;-) Ralf (If that shock was to hard for you I owe you a beer. En Bruxelles avec notres amis Luc et Stoned ;-) From paul@suede.sw.oz.au Thu May 1 04:28:11 1997 Received-Date: Thu, 1 May 1997 04:28:10 +0200 (MET DST) Received: from staff.cs.su.OZ.AU by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA13146; Thu, 1 May 1997 04:24:53 +0200 (MET) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id CAA23855; Thu, 1 May 1997 02:27:34 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id CAA10074; Thu, 1 May 1997 02:27:33 GMT (SMI-8.6/1.34) (from paul for linux-mips@fnet.fr) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705010227.CAA10074@suede.sw.oz.au> Subject: Re: Linux/MIPS 2.1.36 uploaded To: linux-mips@guadalquivir.fnet.fr Date: Thu, 1 May 1997 12:27:32 +1000 (EST) In-Reply-To: <199705010038.CAA24300@informatik.uni-koblenz.de> from "Systemkennung Linux" at May 1, 97 02:38:29 am Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 2312 Lines: 61 Ralf wrote: > Not at all. David did invest quite a lot of work into the low level mm > stuff and did a nice job in un-disgustifing (head.S ...) the code. Yes, I noticed some of this in 2.1.14... well done David! > Some of the changes: > > - IRIX binary compatility code is in. Run IRIX binaries on the DECStation... that'd be cute! :-) > - massive cleanup; lots of code has been moved around. More moves than were in 2.1.14? > - Support for R4000/R4400 SC/MC style second level caches. This means > Linux should now be working (modulo a stupid bug that kills all machines > that are wireing TLB entries on startup.) for machines like the > Mips Magnum 4000SC and the Millenium. Cool! Now all I need to do is be able to boot my Magnum 4000SC via the MIPS proprietary boot protocol in BIG ENDIAN mode so I don't need an HD or a video card... > - Linus' patches upto 2.1.36 applied. Kernel NFS server is my favourite > inovation :-) Sounds good. > > Please tell me you won't do another release for a little while... so I can > > get things stable on the DS's again and send you the code... :-) > > Oh, I'm just waiting for Linus to release 2.1.37. Even more more fun :-) > > The good news for you is that I'll be in the Silicon Valley next week, > so use this week ;-) Can you reveal to the anxious audience what you'll be doing in that fun part of the world?? :-) > (If that shock was to hard for you I owe you a beer. En Bruxelles avec > notres amis Luc et Stoned ;-) Sounds good to me! I should find out when/where the next European High Performance Computing conference is so I can get back to Europe. It may even be in France or Germany this time so you don't have to drive for 8 hours to get to Bruxelles! It would also be nice to see notres amis Luc et Stoned, and find out what they've been up to... Regards, Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From paul@suede.sw.oz.au Thu May 1 04:36:24 1997 Received-Date: Thu, 1 May 1997 04:36:24 +0200 (MET DST) Received: from staff.cs.su.OZ.AU by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA13380; Thu, 1 May 1997 04:33:07 +0200 (MET) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id CAA24026; Thu, 1 May 1997 02:35:51 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id CAA10123; Thu, 1 May 1997 02:35:50 GMT (SMI-8.6/1.34) (from paul for linux-mips@fnet.fr) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705010235.CAA10123@suede.sw.oz.au> Subject: Re: DECStation update... To: linux-mips@guadalquivir.fnet.fr Date: Thu, 1 May 1997 12:35:49 +1000 (EST) In-Reply-To: <199705010000.CAA17267@informatik.uni-koblenz.de> from "Systemkennung Linux" at May 1, 97 02:00:25 am Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 1194 Lines: 26 Ralf wrote: > > the kernel? It's not getting past the code at kernel_entry in head.S due to > > R4K'isms in that portion of the code (due to R4K coders setting the > > assembler to 'noreorder' then assuming all MIPS CPU's have 'some' > > interlocks as does the R4K... :-) > > Could you tell me where these R4k'isms are? I don't know yet where the problem may lie in head.S, as I haven't got beyond deciding that this area is the right place to look. I am now very familiar with the R3000 CPU's tendency to crash in weird ways when encountering such situations, and it seems to me that this is what's happening at the moment. I have already had to fix R4K stuff in header files such as system.h for the cli/sti macros... Regards, Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From tschroed@cheetah.wsc.edu Thu May 1 08:02:01 1997 Received-Date: Thu, 1 May 1997 08:02:00 +0200 (MET DST) Received: from cheetah.wsc.edu by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA23129; Thu, 1 May 1997 07:58:44 +0200 (MET) Received: by cheetah.wsc.edu (5.65/DEC-Ultrix/4.4) id AA18463; Thu, 1 May 1997 01:01:28 -0500 Other-Stuff: localhost tschroed@cheetah.wsc.edu Trevor Schroeder tschroed Date: Thu, 1 May 1997 01:01:28 -0500 (CDT) From: Trevor Schroeder To: linux-mips@guadalquivir.fnet.fr Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 247 Lines: 7 ____________________________________________________________ The NOOP command always succeeds. It does nothing. Trevor Schroeder tschroed@cheetah.wsc.edu ------------------------------------------------------------ From frist@lehr.chem.tu-berlin.de Thu May 1 12:16:25 1997 Received-Date: Thu, 1 May 1997 12:16:24 +0200 (MET DST) Received: from lehr.chem.TU-Berlin.DE by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA29469; Thu, 1 May 1997 12:13:09 +0200 (MET) Received: (from frist@localhost) by lehr.chem.tu-berlin.de (8.7.4/8.7.3) id MAA14946; Thu, 1 May 1997 12:16:20 +0200 Date: Thu, 1 May 1997 12:16:20 +0200 (MET DST) From: Frieder Streffer To: linux-mips@guadalquivir.fnet.fr Subject: Re: DECStation update... In-Reply-To: <199704302222.WAA08701@suede.sw.oz.au> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 2019 Lines: 52 On Thu, 1 May 1997, Paul Antoine wrote > > I've just loaded the kernel binary onto Softway's FTP, and Luc's wonder-script > will pull that back to the main ftp site sometime today (I think this happens > hourly but can't remember). I've created a link to it in the web page, > and have updated the Port Status and the Development Environment Notes to > reflect this. > > The test kernel binary should be available at: > > ftp://ftp.fnet.fr/linux-mips/ftp.softway.com.au/dec_vmlinux.2.1.14.test.gz > > Would you please check that it boots at least to the 'Added tags' point in > the kernel? It's not getting past the code at kernel_entry in head.S due to > R4K'isms in that portion of the code (due to R4K coders setting the assembler > to 'noreorder' then assuming all MIPS CPU's have 'some' interlocks as does > the R4K... :-) > > The new DECStation code uses the 2100/3100 PROM routines prior to > initialising the driver code you both wrote. It also should correctly > determine the amount of memory present... tell me if it doesn't, as I've > had to call either your code or the original DS5000 code that queries the PROM > depending upon which machine it detects... hopefully your code still works!! > > Regards, > Paul > _______________________________________________________________________________ > Paul M. Antoine, Net: paul@sw.oz.au > Softway Pty Ltd WWW: www.softway.com.au > PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 > Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 > > "It is the lack of acceptance of diversity which threatens to > destroy society, NOT the free expression of it." - Me. > Hi Paul, no, sorry the kernel craches immedeatly after the boot with Excptn : Excptn pc : 0x300800d CReg : 0x10 SReg : 0x30080000 Vaddr : 0x304800d Sp : 0xa17fc0000 I don't get any message of the kernel, so something with the PROM call might be wrong. regards, Frieder From amitm@amitpc.iso.dec.com Thu May 1 14:18:33 1997 Received-Date: Thu, 1 May 1997 14:18:32 +0200 (MET DST) Received: from server21.digital.fr by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA02611; Thu, 1 May 1997 14:15:17 +0200 (MET) Received: from tavosf.iso.dec.com (tavosf.iso.dec.com [16.190.64.193]) by server21.digital.fr (8.7.5/8.7) with SMTP id OAA15908; Thu, 1 May 1997 14:24:12 +0200 (MET DST) Received: from amitpc by tavosf.iso.dec.com (5.65v3.2/1.1.10.5/24Mar97-0101PM) id AA03088; Thu, 1 May 1997 15:12:39 +0300 Received: by amitpc.iso.dec.com id (Debian Smail-3.2 1996-Jul-4 #2); Thu, 1 May 1997 15:11:20 +0300 (IDT) Sender: amitm@amitpc.iso.dec.com (Amit-Margalit) Message-Id: X-Mailer: XFMail 1.1 [p0] on Linux Sender: amitm@amitpc.iso.dec.com Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 8bit Mime-Version: 1.0 Date: Thu, 01 May 1997 15:09:52 +0300 (IDT) Organization: Digital Equipment Corp. From: Amit Margalit To: Linux-MIPS Mailing List , Paul Antoine Subject: Booted your latest kernel... Content-Length: 1166 Lines: 40 Hi Paul! Here are the results of booting that last kernel you released on my DS5000/150... Amit -------------------------------------------------------------------- Copyright (C) Paul... Linux version 2.1.14 (paul@slither)... #161 Sun Apr 27 20:21:21 EST 1997 Found memory size: 98304MB - shouldn't this say KB? I-Cache size: OK D-Cache size: OK System has firmware type: TCF0 This DECStation is a: DS5000/1x0 Registering console! Registered console! Registered console! Got the following for the osconsole env. variable: 1,3 No idea what console to use! Added tags Setup complete, launching kernel... Kernel panic: Yeee, unsupported mmu/cache architecture. And there it hung. *** This time it rebooted Ultrix fine on first attempt, without init'ing at the >> prompt. Amit ------------------------------------------------------------------------ Amit Margalit - Author of XBanner: http://physics.fullerton.edu/XBanner/ ------------------------------------------------------------------------ All information in the above mail message is personal, and does not imply on my employer. None of it should be attributed to my employer. From ericz@fikus.com Thu May 1 16:13:37 1997 Received-Date: Thu, 1 May 1997 16:13:36 +0200 (MET DST) Received: from bw.webex.net by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA05623; Thu, 1 May 1997 16:10:21 +0200 (MET) Received: from avalon (user163.webex.net [204.249.170.163]) by bw.webex.net (8.8.5/8.6.12) with ESMTP id KAA24359 for ; Thu, 1 May 1997 10:11:42 -0400 (EDT) Message-Id: <199705011411.KAA24359@bw.webex.net> Reply-To: From: "Eric Zidovec" To: Subject: Linux for Olivetti M700-10 Date: Thu, 1 May 1997 10:13:32 -0400 X-Msmail-Priority: Normal X-Priority: 3 X-Mailer: Microsoft Internet Mail 4.70.1160 Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 416 Lines: 13 Hey there folks. Haven't checked in for a while but does anyone here know about the current status of the Olivetti port that was going on about 6 months ago. I know there's been a big flurry of activity in the Decstation port (congrats!) and some towards the SGI. The last time I mail-tagged with Ralf, he mentioned something about the on-board devices were still not working. Has that changed? Regards, Eric From linux@informatik.uni-koblenz.de Thu May 1 16:30:20 1997 Received-Date: Thu, 1 May 1997 16:30:19 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA06016; Thu, 1 May 1997 16:27:03 +0200 (MET) Received: from thoma (linux@thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.8.5/8.6.9) with SMTP id QAA28378 for ; Thu, 1 May 1997 16:30:10 +0200 (MEST) From: Systemkennung Linux Message-Id: <199705011430.QAA28378@informatik.uni-koblenz.de> Received: by thoma (SMI-8.6/KO-2.0) id QAA07786; Thu, 1 May 1997 16:30:08 +0200 Subject: Re: Linux/MIPS 2.1.36 uploaded To: linux-mips@guadalquivir.fnet.fr Date: Thu, 1 May 1997 16:30:08 +0200 (MET DST) In-Reply-To: <199705010227.CAA10074@suede.sw.oz.au> from "Paul Antoine" at May 1, 97 12:27:32 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1078 Lines: 34 Hi, > Run IRIX binaries on the DECStation... that'd be cute! :-) Will work only on big endian machines. > > - massive cleanup; lots of code has been moved around. > > More moves than were in 2.1.14? > > - Linus' patches upto 2.1.36 applied. Kernel NFS server is my favourite > > inovation :-) > > Sounds good. Yep. On the Linux Expo they demonstrated 5 or 6 mb/s NFS server throughput over the wire :-) > Can you reveal to the anxious audience what you'll be doing in that fun > part of the world?? :-) As you say, having fun. And working on Linus "world domination, fast" project :-) > > (If that shock was to hard for you I owe you a beer. En Bruxelles avec > > notres amis Luc et Stoned ;-) > > Sounds good to me! I should find out when/where the next European High > Performance Computing conference is so I can get back to Europe. It may > even be in France or Germany this time so you don't have to drive for 8 hours > to get to Bruxelles! It would also be nice to see notres amis Luc et Stoned, > and find out what they've been up to... Ralf From tschroed@cheetah.wsc.edu Thu May 1 19:27:25 1997 Received-Date: Thu, 1 May 1997 19:27:24 +0200 (MET DST) Received: from cheetah.wsc.edu by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA08906; Thu, 1 May 1997 19:24:08 +0200 (MET) Received: by cheetah.wsc.edu (5.65/DEC-Ultrix/4.4) id AA19284; Thu, 1 May 1997 12:26:57 -0500 Other-Stuff: localhost tschroed@cheetah.wsc.edu Trevor Schroeder tschroed Date: Thu, 1 May 1997 12:26:57 -0500 (CDT) From: Trevor Schroeder To: linux-mips@guadalquivir.fnet.fr Subject: Ooops... Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 451 Lines: 11 I just realized that I sent a blank message the last time (it was late, what can I say ;). Anyhow, I would like to be added to the Linux/MIPS mailing list, I wanna have Linux on my DECstation. :) Thanks ____________________________________________________________ The NOOP command always succeeds. It does nothing. Trevor Schroeder tschroed@cheetah.wsc.edu ------------------------------------------------------------ From ralf@informatik.uni-koblenz.de Thu May 1 23:03:51 1997 Received-Date: Thu, 1 May 1997 23:03:50 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA11446; Thu, 1 May 1997 23:00:33 +0200 (MET) Received: from thoma (ralf@thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.8.5/8.6.9) with SMTP id XAA12766 for ; Thu, 1 May 1997 23:03:45 +0200 (MEST) From: Ralf Baechle Message-Id: <199705012103.XAA12766@informatik.uni-koblenz.de> Received: by thoma (SMI-8.6/KO-2.0) id XAA08177; Thu, 1 May 1997 23:03:43 +0200 Subject: DEC Documentation References (fwd) To: linux-mips@guadalquivir.fnet.fr Date: Thu, 1 May 1997 23:03:42 +0200 (MET DST) Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 577 Lines: 19 For what it's worth; I just received the message below on another list. Ralf ----------------------------------------------------------------- There is a collection of DIGITAL documentation references available at NETBSD.ORG. Some of it is apparently online, in PostScript format, including references on DECstation 3000 and DECstation 5000, which are MIPS-based. In addition there are technical references of the ALPHA systems, chips and their support circuitry. ftp://ftp.netbsd.org/pub/NetBSD/misc/dec-docs/index.html Thought that might be interesting... Dominik From tschroed@cheetah.wsc.edu Thu May 1 23:15:04 1997 Received-Date: Thu, 1 May 1997 23:15:04 +0200 (MET DST) Received: from cheetah.wsc.edu by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA11712; Thu, 1 May 1997 23:11:41 +0200 (MET) Received: by cheetah.wsc.edu (5.65/DEC-Ultrix/4.4) id AA19604; Thu, 1 May 1997 16:14:23 -0500 Other-Stuff: localhost tschroed@cheetah.wsc.edu Trevor Schroeder tschroed Date: Thu, 1 May 1997 16:14:21 -0500 (CDT) From: Trevor Schroeder To: linux-mips@guadalquivir.fnet.fr Subject: Re: DEC Documentation References (fwd) In-Reply-To: <199705012103.XAA12766@informatik.uni-koblenz.de> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 827 Lines: 22 On Thu, 1 May 1997, Ralf Baechle wrote: > There is a collection of DIGITAL documentation references available > at NETBSD.ORG. > > Some of it is apparently online, in PostScript format, including > references on DECstation 3000 and DECstation 5000, which are > MIPS-based. In addition there are technical references of the ALPHA > systems, chips and their support circuitry. > > ftp://ftp.netbsd.org/pub/NetBSD/misc/dec-docs/index.html additionally, the DECstation 5000 5000 Hardware Maintenance Guide is at ftp://ftp.wsc.edu/pub/maxine/maxine.Z Also compressed postscript. ____________________________________________________________ The NOOP command always succeeds. It does nothing. Trevor Schroeder tschroed@cheetah.wsc.edu ------------------------------------------------------------ From linux@informatik.uni-koblenz.de Fri May 2 02:59:17 1997 Received-Date: Fri, 2 May 1997 02:59:15 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA17285; Fri, 2 May 1997 02:55:58 +0200 (MET) Received: from thoma (linux@thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.8.5/8.6.9) with SMTP id CAA01262; Fri, 2 May 1997 02:59:02 +0200 (MEST) From: Systemkennung Linux Message-Id: <199705020059.CAA01262@informatik.uni-koblenz.de> Received: by thoma (SMI-8.6/KO-2.0) id CAA08495; Fri, 2 May 1997 02:58:59 +0200 Subject: Re: DECStation update... To: stuarta@foxln.com.au Date: Fri, 2 May 1997 02:58:58 +0200 (MET DST) Cc: linux-mips@guadalquivir.fnet.fr In-Reply-To: <9705012241.AA01824@gormenghast.foxln.com.au> from "Stuart Auchterlonie" at May 2, 97 00:42:38 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 2362 Lines: 47 Hi, > On another note, I have been trying to setup my build environment > on a Sparc 5 running Solaris 2.4. I built my own binaries after applying > the patches off ftp.fnet.fr (gcc 2.7.2.2 & binutil 2.7.0.9). The patches are intended to be applied to the vanilla FSF versions of GCC 2.7.2 (2.7.2.2 should work except harmless rejects, haven't checked). But in no case use HJ Lu's binutils 2.7.0.9. It will generate unuseable kernels with high probability. > When I try and build a mips kernel (ralfs 2.1.14 version) the first > problem I have is trying to build the new mkdep version. For some reason > this uses a shared memory segment and is not well liked by solaris. > If you do manage to get it to compile, (if I remember correctly, with a > bit of help from the bsd compatibility libs) then it crashes the second > time it is run :( Yes, unfortunately you're right. > Now if we force the issue a bit and touch .depend in the root of the source > tree we can actually start compiling the kernel. Most of this works out > okay except when it gets to the low level stuff and it gives me the > following rude message: > > make[1]: Entering directory `/opt/home/stuarta/mips/linux-2.1.14/arch/mips/mips1' > mipsel-linux-gcc -D__KERNEL__ -I/opt/home/stuarta/mips/linux-2.1.14/include -I/opt/home/stuarta/mips/include -Wall -Wstrict-prototypes -O2 -fomit-frame-pointer -G 0 -mno-abicalls -fno-pic -mcpu=r3000 -mips1 -c r3000.S -o r3000.o > r3000.S: Assembler messages: > r3000.S:141: Warning: Macro used $at after ".set noat" > r3000.S:230: Warning: Macro used $at after ".set noat" > r3000.S:277: Error: illegal operands `sw' > r3000.S:337: Warning: Macro used $at after ".set noat" > r3000.S:365: Warning: Macro used $at after ".set noat" > r3000.S:526: Warning: Macro used $at after ".set noat" > make[1]: *** [r3000.o] Error 1 > make[1]: Leaving directory `/opt/home/stuarta/mips/linux-2.1.14/arch/mips/mips1' > make: *** [linuxsubdirs] Error 2 > > Suspecting this was something to do with the way I installed the binutils, > I grab the binutil binaries for solaris 2.5 and gave them a go but this > didn't improve matters.... Well, the problem is the source. Paul's DECstation source tree is probably completly out of sync with mine. We'll fix that rsn. Paul, maybe you have a tree more suitable for DECstations for ftp at Softway? Ralf From linux@informatik.uni-koblenz.de Fri May 2 03:12:18 1997 Received-Date: Fri, 2 May 1997 03:12:17 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA17634; Fri, 2 May 1997 03:08:57 +0200 (MET) Received: from thoma (linux@thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.8.5/8.6.9) with SMTP id DAA01730 for ; Fri, 2 May 1997 03:12:07 +0200 (MEST) From: Systemkennung Linux Message-Id: <199705020112.DAA01730@informatik.uni-koblenz.de> Received: by thoma (SMI-8.6/KO-2.0) id DAA08539; Fri, 2 May 1997 03:12:04 +0200 Subject: Re: DEC Documentation References (fwd) To: linux-mips@guadalquivir.fnet.fr Date: Fri, 2 May 1997 03:12:04 +0200 (MET DST) In-Reply-To: from "Trevor Schroeder" at May 1, 97 04:14:21 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 810 Lines: 23 > > There is a collection of DIGITAL documentation references available > > at NETBSD.ORG. > > > > Some of it is apparently online, in PostScript format, including > > references on DECstation 3000 and DECstation 5000, which are > > MIPS-based. In addition there are technical references of the ALPHA > > systems, chips and their support circuitry. > > > > ftp://ftp.netbsd.org/pub/NetBSD/misc/dec-docs/index.html > > additionally, the DECstation 5000 5000 Hardware Maintenance Guide is > at ftp://ftp.wsc.edu/pub/maxine/maxine.Z > > Also compressed postscript. Paul, you check these files out? Maybe they should go into our documentation collection at FNet. > The NOOP command always succeeds. It does nothing. It wastes time, bandwidth and bloats. A dead NOOP is a good NOOP ;-) Ralf From linux@informatik.uni-koblenz.de Fri May 2 09:48:54 1997 Received-Date: Fri, 2 May 1997 09:48:53 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA29838; Fri, 2 May 1997 09:45:23 +0200 (MET) Received: from thoma (linux@thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.8.5/8.6.9) with SMTP id JAA15679; Fri, 2 May 1997 09:48:29 +0200 (MEST) From: Systemkennung Linux Message-Id: <199705020748.JAA15679@informatik.uni-koblenz.de> Received: by thoma (SMI-8.6/KO-2.0) id JAA09532; Fri, 2 May 1997 09:48:27 +0200 Subject: Re: DECStation update... To: stuarta@foxln.com.au Date: Fri, 2 May 1997 09:48:26 +0200 (MET DST) Cc: linux-mips@guadalquivir.fnet.fr In-Reply-To: <9705020737.AA12054@gormenghast.foxln.com.au> from "Stuart Auchterlonie" at May 2, 97 09:39:10 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 670 Lines: 19 Hi, > The patches for gcc went I cleanly onto standard 2.7.2.2 as far as I > can remember... and it built okay. > I suspect I used the standard binutils, but it was too long ago :( You can find out the exact version with 'ld --version'. > > Well, the problem is the source. Paul's DECstation source tree is probably > > completly out of sync with mine. We'll fix that rsn. Paul, maybe you > > have a tree more suitable for DECstations for ftp at Softway? > > I am still not totally convinced it is the source. > I have just tried playing around with paul's 2.1.1 dec source, and > I still managed to get errors from the assembler...... It definately is. Ralf From ryan@ravage.labs.gmu.edu Fri May 2 22:06:20 1997 Received-Date: Fri, 2 May 1997 22:06:19 +0200 (MET DST) Received: from ravage.labs.gmu.edu ([129.174.166.55]) by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA16568; Fri, 2 May 1997 22:03:02 +0200 (MET) Received: from localhost (ryan@localhost) by ravage.labs.gmu.edu (8.8.3/8.8.3) with SMTP id EAA03649 for ; Fri, 2 May 1997 04:06:33 -0400 Date: Fri, 2 May 1997 04:06:33 -0400 (EDT) From: Ryan Rafferty To: linux-mips@guadalquivir.fnet.fr Subject: Re: Linux/MIPS 2.1.36 uploaded In-Reply-To: <199705010038.CAA24300@informatik.uni-koblenz.de> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 1064 Lines: 32 On Thu, 1 May 1997, Systemkennung Linux wrote: > - Support for R4000/R4400 SC/MC style second level caches. This means > Linux should now be working (modulo a stupid bug that kills all machines > that are wireing TLB entries on startup.) for machines like the > Mips Magnum 4000SC and the Millenium. If the bug is simple, how hard would it be for me to fix it? Is it trivial? I think the gentleman who was inquiring about the NEC Mips server he had would be interested in the 4400 SC support, since apparently his machine is very similar to the Magnum 4000SC. Oh yeah, does anyone on this list have a Magnum 4000SC? If so, do you know how it compares in performance to the 4000PC? > The good news for you is that I'll be in the Silicon Valley next week, > so use this week ;-) I hope you enjoy your trip! Now's definitely the time to travel there, though--any later and it would be too hot, IMHO. Ciao, Ryan Rafferty > > Ralf > > (If that shock was to hard for you I owe you a beer. En Bruxelles avec > notres amis Luc et Stoned ;-) > From paul@suede.sw.oz.au Fri May 2 14:49:32 1997 Received-Date: Fri, 2 May 1997 14:49:31 +0200 (MET DST) Received: from staff.cs.su.OZ.AU by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA08150; Fri, 2 May 1997 14:46:05 +0200 (MET) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id MAA16388; Fri, 2 May 1997 12:49:01 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id MAA26892; Fri, 2 May 1997 12:49:00 GMT (SMI-8.6/1.34) (from paul for linux-mips@fnet.fr) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705021249.MAA26892@suede.sw.oz.au> Subject: Re: DEC Documentation References (fwd) To: linux-mips@guadalquivir.fnet.fr Date: Fri, 2 May 1997 22:48:59 +1000 (EST) In-Reply-To: <199705012103.XAA12766@informatik.uni-koblenz.de> from "Ralf Baechle" at May 1, 97 11:03:42 pm Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 1355 Lines: 34 > For what it's worth; I just received the message below on another list. Thanks! > ----------------------------------------------------------------- > > There is a collection of DIGITAL documentation references available > at NETBSD.ORG. > > Some of it is apparently online, in PostScript format, including > references on DECstation 3000 and DECstation 5000, which are > MIPS-based. In addition there are technical references of the ALPHA > systems, chips and their support circuitry. > > ftp://ftp.netbsd.org/pub/NetBSD/misc/dec-docs/index.html > > Thought that might be interesting... Unfortunately, it's all the same doco we've seen before (though it has been and remains good information!)... if only we could find the equivalent to the KN02 Functional Spec. for the other KN02-XX models... writing a keyboard driver for the maxine may prove difficult without it, for instance. Regards, Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From paul@suede.sw.oz.au Fri May 2 14:57:34 1997 Received-Date: Fri, 2 May 1997 14:57:33 +0200 (MET DST) Received: from staff.cs.su.OZ.AU by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA08372; Fri, 2 May 1997 14:54:15 +0200 (MET) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id MAA16426; Fri, 2 May 1997 12:57:01 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id MAA26910; Fri, 2 May 1997 12:57:00 GMT (SMI-8.6/1.34) (from paul for linux-mips@fnet.fr) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705021257.MAA26910@suede.sw.oz.au> Subject: Re: DECStation update... To: linux-mips@guadalquivir.fnet.fr Date: Fri, 2 May 1997 22:56:59 +1000 (EST) In-Reply-To: <199705020748.JAA15679@informatik.uni-koblenz.de> from "Systemkennung Linux" at May 2, 97 09:48:26 am Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 1379 Lines: 36 Hi folks, Ralf and Trevor have been saying: > > > Well, the problem is the source. Paul's DECstation source tree is > > > probably completly out of sync with mine. We'll fix that rsn. Paul, > > > maybe you have a tree more suitable for DECstations for ftp at Softway? All the Softway source is mirrored up at fnet at the address: /linux-mips/ftp.softway.com.au/ There are source trees and sample binaries in there, including one called linux.2.1.1.dec.tar.gz which is a version of Ralf's 2.1.1 that actually compiles on an R3000... though I have a fuzzy memory that the problem you're seeing was solved some other way, and can't remember it just yet. > > I have just tried playing around with paul's 2.1.1 dec source, and > > I still managed to get errors from the assembler...... So you got 2.1.1.dec.tar.gz?? > It definately is. I'm fairly certain it hasn't been in the past... just can't remember! Regards, Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From tschroed@cheetah.wsc.edu Fri May 2 16:34:23 1997 Received-Date: Fri, 2 May 1997 16:34:22 +0200 (MET DST) Received: from cheetah.wsc.edu by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA11158; Fri, 2 May 1997 16:30:55 +0200 (MET) Received: by cheetah.wsc.edu (5.65/DEC-Ultrix/4.4) id AA20700; Fri, 2 May 1997 09:33:45 -0500 Other-Stuff: localhost tschroed@cheetah.wsc.edu Trevor Schroeder tschroed Date: Fri, 2 May 1997 09:33:44 -0500 (CDT) From: Trevor Schroeder To: linux-mips@guadalquivir.fnet.fr Subject: Re: DECStation update... In-Reply-To: <199705021257.MAA26910@suede.sw.oz.au> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 831 Lines: 18 On Fri, 2 May 1997, Paul Antoine wrote: > There are source trees and sample binaries in there, including one called > linux.2.1.1.dec.tar.gz which is a version of Ralf's 2.1.1 that actually > compiles on an R3000... though I have a fuzzy memory that the problem you're > seeing was solved some other way, and can't remember it just yet. > There is a minor problem with the 2.1.1.dec tree: arch/mips/kernel/bitags.c (If I recall correctly) doesn't #include , so memcpy isn't defined and it will puke if you try to compile it out of the box. Otherwise it works fine. ____________________________________________________________ The NOOP command always succeeds. It does nothing. Trevor Schroeder tschroed@cheetah.wsc.edu ------------------------------------------------------------ From marks@sun470.rd.qms.com Fri May 2 18:20:43 1997 Received-Date: Fri, 2 May 1997 18:20:42 +0200 (MET DST) Received: from gatekeeper.qms.com by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA13726; Fri, 2 May 1997 18:17:24 +0200 (MET) Received: from rd.qms.com (sun470.qms.com) by gatekeeper.qms.com (4.1/SMI-4.1) id AA24284; Fri, 2 May 97 11:20:36 CDT Received: from speedy.qms.com.rd.qms.com by rd.qms.com (4.1/SMI-4.1) id AA16641; Fri, 2 May 97 11:20:35 CDT Received: by speedy.qms.com.rd.qms.com (8.8.2) id LAA05571; Fri, 2 May 1997 11:20:35 -0500 Date: Fri, 2 May 1997 11:20:35 -0500 Message-Id: <199705021620.LAA05571@speedy.qms.com.rd.qms.com> From: Mark Salter To: linux-mips@guadalquivir.fnet.fr Subject: mailing list enquiry Content-Length: 791 Lines: 24 Hi, I work for a company which manufactures embedded systems based on MIPS cpus. Primarily, the 4300 but we have an R5000 controller in prototype and we are soon to have an R5400 controller. "Embedded" is a bit misleading in that typical controllers for us have an IDE or SCSI hard drive and 32 - 128 MB of RAM and TCP/IP hardware. We are currently exploring a linux port to our controllers which is the reason I am sending this note. I have a firm committment from management to allow me a great deal of time to attempt to bring up linux on our controller. I would like to be added to the linux-mips developers mailing list in order to establish contact with the people involved with the ongoing mips port. Thank You, Mark Salter Member of Technical Staff QMS, Inc. Mobile, AL 36695 From extractor@mail-response.com Sat May 3 02:19:38 1997 Received-Date: Sat, 3 May 1997 02:19:37 +0200 (MET DST) From: extractor@mail-response.com Received: from gatekeeper.waldorf-gmbh.de by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA20662; Sat, 3 May 1997 02:16:19 +0200 (MET) Received: from scotty (root@scotty.waldorf-gmbh.de [193.98.169.1]) by gatekeeper.waldorf-gmbh.de (8.6.12/8.6.9) with ESMTP id DAA18210 for ; Sat, 3 May 1997 03:20:55 +0200 Received: from mail-response.com ([205.254.167.57]) by scotty (8.6.12/8.6.12) with ESMTP id BAA30808 for ; Sat, 3 May 1997 01:18:37 +0200 Received: by mail-response.com (8.8.5/8.8.5.69) with SMTP id QAA20017; Fri, 2 May 1997 16:14:26 -0400 Date: Fri, 2 May 1997 16:14:26 -0400 Message-Id: <199705022014.QAA20017@mail-response.com> To: extractor@mail-response.com Subject: re: your website hits Content-Length: 510 Lines: 15 I would like to help you get more people to your website. There are tens of thousands of internet users who would love to visit your website but don't even know it exists. I would like to tell you how simple it is for you to let the world know about your site and what you have to offer. This can all be done for less than what most people spend on internet access in two months. Please reply with the words, "Website Promotion" and I'll get that information out to you right away. Thank you, Simone From linux@informatik.uni-koblenz.de Sat May 3 01:47:23 1997 Received-Date: Sat, 3 May 1997 01:47:22 +0200 (MET DST) Received: from informatik.uni-koblenz.de ([141.26.4.1]) by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA19999; Sat, 3 May 1997 01:44:01 +0200 (MET) Received: from thoma (linux@thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.8.5/8.6.9) with SMTP id BAA24974 for ; Sat, 3 May 1997 01:47:11 +0200 (MEST) From: Systemkennung Linux Message-Id: <199705022347.BAA24974@informatik.uni-koblenz.de> Received: by thoma (SMI-8.6/KO-2.0) id BAA11961; Sat, 3 May 1997 01:47:05 +0200 Subject: Re: Linux/MIPS 2.1.36 uploaded To: linux-mips@guadalquivir.fnet.fr Date: Sat, 3 May 1997 01:47:04 +0200 (MET DST) In-Reply-To: from "Ryan Rafferty" at May 2, 97 04:06:33 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 646 Lines: 18 Hi, > > - Support for R4000/R4400 SC/MC style second level caches. This means > > Linux should now be working (modulo a stupid bug that kills all machines > > that are wireing TLB entries on startup.) for machines like the > > Mips Magnum 4000SC and the Millenium. > > If the bug is simple, how hard would it be for me to fix it? Is it > trivial? The "trivial" fix: take the code that loads the wired entries into the TLB in version 2.1.14 and put it into 2.1.36. The right thing is to get rid of the wired entries and map the stuff using normal page tables. Not trivial; these are some special issues when fixing this. Ralf From paul@suede.sw.oz.au Sat May 3 02:49:52 1997 Received-Date: Sat, 3 May 1997 02:49:51 +0200 (MET DST) Received: from staff.cs.su.OZ.AU by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA21467; Sat, 3 May 1997 02:46:33 +0200 (MET) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id AAA27428; Sat, 3 May 1997 00:49:15 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id AAA00211; Sat, 3 May 1997 00:49:13 GMT (SMI-8.6/1.34) (from paul for linux-mips@fnet.fr) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705030049.AAA00211@suede.sw.oz.au> Subject: Re: Linux/MIPS 2.1.36 uploaded To: linux-mips@guadalquivir.fnet.fr Date: Sat, 3 May 1997 10:49:13 +1000 (EST) In-Reply-To: from "Ryan Rafferty" at May 2, 97 04:06:33 am Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 1062 Lines: 22 Ryan wrote: > I think the gentleman who was inquiring about the NEC Mips server he had > would be interested in the 4400 SC support, since apparently his machine > is very similar to the Magnum 4000SC. Oh yeah, does anyone on this list > have a Magnum 4000SC? If so, do you know how it compares in performance > to the 4000PC? I am told that it is *much* faster than the PC version (without the R4000-controlled second-level cache)... though without a program to respond to the MIPS boot protocol, I can't test it as I haven't got a video card that would let me use the NT little-endian boot PROM. Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From paul@suede.sw.oz.au Sat May 3 02:53:01 1997 Received-Date: Sat, 3 May 1997 02:53:00 +0200 (MET DST) Received: from staff.cs.su.OZ.AU by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA21638; Sat, 3 May 1997 02:49:42 +0200 (MET) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id AAA27476; Sat, 3 May 1997 00:52:34 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id AAA00229; Sat, 3 May 1997 00:52:33 GMT (SMI-8.6/1.34) (from paul for linux-mips@fnet.fr) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705030052.AAA00229@suede.sw.oz.au> Subject: Re: DECStation update... To: linux-mips@guadalquivir.fnet.fr Date: Sat, 3 May 1997 10:52:32 +1000 (EST) In-Reply-To: from "Trevor Schroeder" at May 2, 97 09:33:44 am Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 725 Lines: 18 Trevor wrote: > There is a minor problem with the 2.1.1.dec tree: arch/mips/kernel/bitags.c > (If I recall correctly) doesn't #include , so memcpy isn't > defined and it will puke if you try to compile it out of the box. Otherwise > it works fine. Cool. Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From davidm@fuzzbox.psrg.cs.usyd.edu.au Sat May 3 04:45:44 1997 Received-Date: Sat, 3 May 1997 04:45:43 +0200 (MET DST) Received: from fuzzbox.psrg.cs.usyd.edu.AU by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA24526; Sat, 3 May 1997 04:42:24 +0200 (MET) Received: from fuzzbox.psrg.cs.usyd.edu.au (davidm@fuzzbox.psrg.cs.usyd.edu.au [129.78.97.193]) by fuzzbox.psrg.cs.usyd.edu.au (8.8.4/8.8.4) with SMTP id MAA02192 for ; Sat, 3 May 1997 12:45:07 +1000 Date: Sat, 3 May 1997 12:45:07 +1000 (EST) From: David Monro Sender: davidm@fuzzbox.psrg.cs.usyd.edu.au Reply-To: David Monro Subject: Magnum 4000SC To: linux-mips@guadalquivir.fnet.fr Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-Length: 727 Lines: 13 People were enquiring about the 4000-SC - I have a 4000SC-50 which I am about to start experimenting with. I also have a protocol description for the bfs boot code stuff so I'll be writing a basic bfs server rsn. I have no graphics card for this thing so I'll be using a serial console at least for the forseeable future. I've also still got one running RISC/OS which helps (that for example is where I got the bfs protocol description) btw, how is the port going for the magnum 3000 series? I've also got one of them (a 3020 I think). I'm hoping that sooner or later I can get both it and the 4000 running the same binaries - is this likely to happen? (ie I really would prefer to run the 4000SC in big endian mode). David From paul@suede.sw.oz.au Sat May 3 07:11:27 1997 Received-Date: Sat, 3 May 1997 07:11:26 +0200 (MET DST) Received: from staff.cs.su.OZ.AU by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA01818; Sat, 3 May 1997 07:08:08 +0200 (MET) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id FAA00680; Sat, 3 May 1997 05:10:51 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id FAA00722; Sat, 3 May 1997 05:10:49 GMT (SMI-8.6/1.34) (from paul for linux-mips@fnet.fr) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705030510.FAA00722@suede.sw.oz.au> Subject: Re: Magnum 4000SC To: linux-mips@guadalquivir.fnet.fr Date: Sat, 3 May 1997 15:10:49 +1000 (EST) In-Reply-To: from "David Monro" at May 3, 97 12:45:07 pm Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 1757 Lines: 34 David wrote: > People were enquiring about the 4000-SC - I have a 4000SC-50 which I am about > to start experimenting with. I also have a protocol description for the bfs > boot code stuff so I'll be writing a basic bfs server rsn. I have no graphics > card for this thing so I'll be using a serial console at least for the > forseeable future. I've also still got one running RISC/OS which helps (that > for example is where I got the bfs protocol description) Is the protocol description adequate? Is it from the man pages for RISC/OS ver 5.01?? I will certainly help you test the bfsd you write (see below)... > btw, how is the port going for the magnum 3000 series? I've also got one of > them (a 3020 I think). I'm hoping that sooner or later I can get both it and > the 4000 running the same binaries - is this likely to happen? (ie I really > would prefer to run the 4000SC in big endian mode). I have a 4000SC and a 3230, and had intended to run them both big endian as I too am lacking a video card for the 4000SC. Big endian shouldn't be too much trouble, as much of the little->big work was done for the SPARC, and used again on the SGI Indy port. With any luck there won't be too much to do... We should even be able to run SGI binaries! Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From davidm@fuzzbox.psrg.cs.usyd.edu.au Sun May 4 19:15:39 1997 Received-Date: Sun, 4 May 1997 19:15:38 +0200 (MET DST) Received: from fuzzbox.psrg.cs.usyd.edu.AU by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA17007; Sun, 4 May 1997 19:12:17 +0200 (MET) Received: from fuzzbox.psrg.cs.usyd.edu.au (davidm@fuzzbox.psrg.cs.usyd.edu.au [129.78.97.193]) by fuzzbox.psrg.cs.usyd.edu.au (8.8.4/8.8.4) with SMTP id DAA17775; Mon, 5 May 1997 03:15:18 +1000 Date: Mon, 5 May 1997 03:15:18 +1000 (EST) From: David Monro Sender: davidm@fuzzbox.psrg.cs.usyd.edu.au Reply-To: David Monro Subject: bfsd-1.0 released To: linux-mips@guadalquivir.fnet.fr Cc: paul@suede.sw.oz.au Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-Length: 856 Lines: 19 For those of us whe are messing with the Magnum 4000SC, I have implemented and tested a bfs server for it. It seems capable of getting the machine to run sash and to boot the RISCos kernel (which promptly falls over because I don't have any disks on the machine :-)) The place to get it is: ftp://ftp.cs.usyd.edu.au/davidm/bfsd-1.0.tar.gz Should compile on most posix systems I think, but developed under linux/x86. Bugs etc please let me know... (btw I now have enough gear set up to develop for the 4000 from home - the serial console on it is plugged into the serial port on another machine and I can powercycle it (if I get desperate) using an x10 controller from another linux box. I compile on a third linux box and tcpdump on a 4th which happens to be on the same piece of thinnet. I then talk to it all via my ppp link - and it works!). David From allens@vivanet.com Mon May 5 05:08:04 1997 Received-Date: Mon, 5 May 1997 05:08:03 +0200 (MET DST) Received: from vivanet.com ([192.108.245.18]) by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA28070; Mon, 5 May 1997 05:04:44 +0200 (MET) Received: from shell10.vivanet.com (allens@shell10.vivanet.com [192.108.245.8]) by vivanet.com (8.7.5/8.7.3) with ESMTP id XAA26888; Sun, 4 May 1997 23:11:12 -0400 (EDT) From: Stu Allen Received: (from allens@localhost) by shell10.vivanet.com (8.7.5/8.7.3) id XAA08654; Sun, 4 May 1997 23:10:20 -0400 (EDT) Message-Id: <199705050310.XAA08654@shell10.vivanet.com> Subject: Re: DECStation update... To: linux-mips@guadalquivir.fnet.fr Date: Sun, 4 May 1997 23:10:20 -0400 (EDT) Cc: allens@vivanet.com (Stu Allen) In-Reply-To: <199704302222.WAA08701@suede.sw.oz.au> from "Paul Antoine" at May 1, 97 08:22:40 am X-Mailer: ELM [version 2.4 PL25] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 1262 Lines: 45 Paul (and others!), Well, I grabbed the 2.1.14 test kernel and tried it on my lonely little 5100. (Which I accidently booted into Ultrix and discovered hadn't been up for 153 days!) I was disappointed, but not suprised, to get an immediate register dump, just as with the previous kernels. I'm going to type it all in, with hopes that you (or someone else on the list) might be able to make some sense of it: Here's what I get: >> boot -f tftp(0,6)/tftpdir/dec_vmlinux 2..1.. 615360+82736+108976 entry 0x80086e60 KN230 Exception v0 = 00000001 v1 = 00000000 a0 = a000eeb0 a1 = 00000001 a2 = a0003e80 a3 = 80086e60 s0 = a07ff800 s1 = a0800000 s2 = bc000000 s3 = 00000000 s4 = bfc39454 s5 = bfc37f10 s6 = 00000001 s7 = 10003328 k0 = a0010000 k1 = 0000fc04 sp = a000ee80 gp = 80183900 fp = 00000000 ra = bfc00a40 badvaddr = bfc0099f epc = bc000000 sr = b0400000 cause = 30000028 processor revision: 2.32 KN230 V1.4 08-00-2b-1d-3a-20 0x800000 >> Well, I _do_ have a copy of the MIPS architecture book, so maybe I'll take a look and see if I can make any sense at all of this. Of course, I may not be able to get far without a map of the kernel. Thanks!! Stu Allen Linux kernel-hacker-wannabee! From paul@suede.sw.oz.au Mon May 5 14:44:23 1997 Received-Date: Mon, 5 May 1997 14:44:22 +0200 (MET DST) Received: from staff.cs.su.OZ.AU by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA18912; Mon, 5 May 1997 14:40:52 +0200 (MET) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id MAA08636; Mon, 5 May 1997 12:43:51 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id MAA22177; Mon, 5 May 1997 12:43:50 GMT (SMI-8.6/1.34) (from paul for linux-mips@fnet.fr) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705051243.MAA22177@suede.sw.oz.au> Subject: Frieder's 3100 errors To: linux-mips@guadalquivir.fnet.fr Date: Mon, 5 May 1997 22:43:50 +1000 (EST) In-Reply-To: from "Frieder Streffer" at May 1, 97 12:16:20 pm Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 906 Lines: 27 Frieder wrote: > no, sorry the kernel craches immedeatly after the boot with > > Excptn : > Excptn pc : 0x300800d > CReg : 0x10 > SReg : 0x30080000 > Vaddr : 0x304800d > Sp : 0xa17fc0000 > > I don't get any message of the kernel, so something with the PROM call > might be wrong. Indeed it probably is. I'll send some information to let you do some tests of the PROM call info. I extracted out of Mach. Regards, Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From frist@lehr.chem.tu-berlin.de Mon May 5 15:40:44 1997 Received-Date: Mon, 5 May 1997 15:40:42 +0200 (MET DST) Received: from lehr.chem.TU-Berlin.DE by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA20918; Mon, 5 May 1997 15:37:21 +0200 (MET) Received: (from frist@localhost) by lehr.chem.tu-berlin.de (8.7.4/8.7.3) id PAA17850; Mon, 5 May 1997 15:40:35 +0200 Date: Mon, 5 May 1997 15:40:34 +0200 (MET DST) From: Frieder Streffer To: linux-mips@guadalquivir.fnet.fr Subject: Re: Frieder's 3100 errors In-Reply-To: <199705051243.MAA22177@suede.sw.oz.au> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 1393 Lines: 42 On Mon, 5 May 1997, Paul Antoine wrote: > Frieder wrote: > > > no, sorry the kernel craches immedeatly after the boot with > > > > Excptn : > > Excptn pc : 0x300800d > > CReg : 0x10 > > SReg : 0x30080000 > > Vaddr : 0x304800d > > Sp : 0xa17fc0000 > > > > I don't get any message of the kernel, so something with the PROM call > > might be wrong. > > Indeed it probably is. I'll send some information to let you do some > tests of the PROM call info. I extracted out of Mach. > > Regards, > Paul > _______________________________________________________________________________ > Paul M. Antoine, Net: paul@sw.oz.au > Softway Pty Ltd WWW: www.softway.com.au > PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 > Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 > > "It is the lack of acceptance of diversity which threatens to > destroy society, NOT the free expression of it." - Me. > Hi, yep, I will have a look to it. I recently searched the netbsd sources for their PROM calling code, which is quite easy. Hopefully, I will be able to cook something out of it. I don't have any info/doc about the DS3100 PROM except for the entry of its jumptable! Has someone information aboute the routines further than the comments in dec_prom.h from netbsd?? regards, Frieder From neuffer@goofy.zdv.Uni-Mainz.de Mon May 5 16:22:39 1997 Received-Date: Mon, 5 May 1997 16:22:38 +0200 (MET DST) Received: from kralle.zdv.Uni-Mainz.DE by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA22307; Mon, 5 May 1997 16:19:18 +0200 (MET) Received: from localhost (neuffer@localhost) by kralle.zdv.Uni-Mainz.DE (8.8.5/8.8.5) with ESMTP id QAA22507 for ; Mon, 5 May 1997 16:22:27 +0200 (MET DST) X-Authentication-Warning: kralle.zdv.Uni-Mainz.DE: neuffer owned process doing -bs X-Received: from nomis.i-connect.net (qmailr@nomis.i-Connect.Net [206.190.143.10]) by mail.Uni-Mainz.DE (8.8.5/8.8.4) with SMTP id QAA17717 for ; Mon, 5 May 1997 16:05:09 +0200 (MET DST) X-Received: (qmail 24057 invoked by uid 106); 5 May 1997 14:05:56 -0000 Delivered-To: neuffer@nomis.i-Connect.Net X-Received: (qmail 24055 invoked from network); 5 May 1997 14:05:55 -0000 X-Received: from wolverine.hq.cic.net (198.87.19.70) by nomis.i-connect.net with SMTP; 5 May 1997 14:05:55 -0000 X-Received: from vger.rutgers.edu (root@vger.rutgers.edu [128.6.190.2]) by wolverine.hq.cic.net (8.8.5/8.6.12) with ESMTP id KAA17009; Mon, 5 May 1997 10:04:47 -0400 X-Received: by vger.rutgers.edu id <971349-248>; Mon, 5 May 1997 09:54:02 -0400 X-Received: from vuser.vu.union.edu ([149.106.37.197]) by vger.rutgers.edu with ESMTP id <971355-248>; Mon, 5 May 1997 09:53:45 -0400 X-Received: from [149.106.149.82] ([149.106.149.82]) by vuser.vu.union.edu (8.8.5/8.8.2) with ESMTP id JAA00919 for ; Mon, 5 May 1997 09:55:04 -0400 X-Sender: hagopiar@mail.vu.union.edu Message-Id: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Mon, 5 May 1997 09:54:30 -0400 To: linux-kernel@vger.rutgers.edu From: Rob Hagopian Subject: Pentium II Math Bug Sender: owner-linux-kernel@vger.rutgers.edu Precedence: bulk Resent-Date: Mon, 5 May 1997 16:22:23 +0200 (MET DST) Resent-From: Michael Neuffer Resent-To: linux-mips@guadalquivir.fnet.fr Resent-Message-Id: Content-Length: 119 Lines: 6 Check out http://www.x86.org/ after 9:00A PST for details. Glad I'm PPC person personally ;-) -Rob Hagopian From davidm@fuzzbox.psrg.cs.usyd.edu.au Mon May 5 16:22:07 1997 Received-Date: Mon, 5 May 1997 16:22:07 +0200 (MET DST) Received: from fuzzbox.psrg.cs.usyd.edu.AU by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA22301; Mon, 5 May 1997 16:18:44 +0200 (MET) Received: from fuzzbox.psrg.cs.usyd.edu.au (davidm@fuzzbox.psrg.cs.usyd.edu.au [129.78.97.193]) by fuzzbox.psrg.cs.usyd.edu.au (8.8.4/8.8.4) with SMTP id AAA29867; Tue, 6 May 1997 00:21:33 +1000 Date: Tue, 6 May 1997 00:21:33 +1000 (EST) From: David Monro Sender: davidm@fuzzbox.psrg.cs.usyd.edu.au Reply-To: David Monro Subject: Oops, bfsd-1.01 released To: linux-mips@guadalquivir.fnet.fr Cc: Paul Antoine Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-Length: 495 Lines: 11 NEW! This time with documentation! Thanks to Paul Antoine, who pointed out that I hadn't supplied a man page (somehow it got left out of the original tarball). The code hasn't changed. You can find it at ftp://ftp.cs.usyd.edu.au/davidm/bfsd-1.01.tar.gz (If anybody missed the original announcement, bfsd implements a server for the MIPS boot file system protocol which is the only protocol supported by a magnum 4000SC for netbooting - don't ask me why bootp/tftp wasn't good enough). David From paul@suede.sw.oz.au Mon May 5 16:26:40 1997 Received-Date: Mon, 5 May 1997 16:26:37 +0200 (MET DST) Received: from staff.cs.su.OZ.AU by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA22391; Mon, 5 May 1997 16:22:58 +0200 (MET) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id OAA10352; Mon, 5 May 1997 14:25:50 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id OAA22992; Mon, 5 May 1997 14:25:48 GMT (SMI-8.6/1.34) (from paul for linux-mips@fnet.fr) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705051425.OAA22992@suede.sw.oz.au> Subject: Another DECStations test kernel To: linux-mips@guadalquivir.fnet.fr (Linux MIPS mailing list) Date: Tue, 6 May 1997 00:25:48 +1000 (EST) Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 1036 Lines: 26 Hi folks, For those of you with DS5000/xx[x]'s you may like to try the new kernel I've uploaded at: ftp://ftp.fnet.fr/linux-mips/softway.com.au/dec_vmlinux.2.1.14.test2.gz It now shows BogoMIPS... but is not yet using real console code (tm). I'm sorry about the large number of debugging printk's, but hey, it's a test kernel, right?? Those with DS3100's will have to be a little patient as Frieder and I figure out how to use the boot PROM's for console (I'm afraid the real console code will take some debugging now I've munged... err... moved it around a bit! :-) Regards, Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From paul@suede.sw.oz.au Mon May 5 23:39:38 1997 Received-Date: Mon, 5 May 1997 23:39:37 +0200 (MET DST) Received: from staff.cs.su.OZ.AU by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA00518; Mon, 5 May 1997 23:36:17 +0200 (MET) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id VAA15817; Mon, 5 May 1997 21:39:18 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id VAA25462; Mon, 5 May 1997 21:39:17 GMT (SMI-8.6/1.34) (from paul for ) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705052139.VAA25462@suede.sw.oz.au> Subject: Re: New DECStation kernel To: Luc.Beurton@guadalquivir.fnet.fr (Luc Beurton) Date: Tue, 6 May 1997 07:39:17 +1000 (EST) Cc: linux-mips@guadalquivir.fnet.fr (Linux MIPS mailing list) In-Reply-To: <199705051753.RAA20779@orenoque.fnet.fr> from "Luc Beurton" at May 5, 97 07:53:22 pm Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 622 Lines: 16 Luc wrote: > I can't get the file dec_vmlinux.2.1.14.test2.gz (0660). > I'm impatient to test it :-)) Oopps! It was late when I uploaded it... it should be fine now. Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From Stoned.Elipot@univ-evry.fr Tue May 6 18:39:14 1997 Received-Date: Tue, 6 May 1997 18:39:13 +0200 (MET DST) Received: from bandsept.univ-evry.fr (localhost.univ-evry.fr [127.0.0.1]) by bandsept.univ-evry.fr with ESMTP (8.8.5/97.01.21/bandsept); id SAA04653; Tue, 6 May 1997 18:39:06 +0200 (MET DST) Message-Id: <199705061639.SAA04653@bandsept.univ-evry.fr> To: linux-mips@fnet.fr Subject: Linux-mips net services down for few hours, Fnet is moving... X-It: bandsept (better OS but not yet a 4.4, and better CPU) X-Mailer: MH 6.8.4 Date: Tue, 06 May 1997 18:39:05 +0200 From: Stoned Elipot Content-Length: 463 Lines: 11 Hi all, Luc is in the process of moving all the Fnet machines, it should start in about an hour and hopefully this night (MET TZ) Fnet services will be back, hosted on the same machines but 10km away from the actual site. During this time, this mailing-list will be down as well as linux-mips FTP and WWW servers. Regarding FTP, you can find at ftp://ftp.univ-evry.fr/pub/system/linux/linux-mips an uptodate mirror of Fnet's linux-mips area. Cheers, Stoned. From paul@suede.sw.oz.au Thu May 8 16:58:33 1997 Received-Date: Thu, 8 May 1997 16:58:27 +0200 (MET DST) Received: from staff.cs.su.OZ.AU by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA18708; Wed, 7 May 1997 04:45:22 +0200 (MET) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id VAA04167; Tue, 6 May 1997 21:52:11 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id VAA07332; Tue, 6 May 1997 21:52:10 GMT (SMI-8.6/1.34) (from paul for ) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705062152.VAA07332@suede.sw.oz.au> Subject: Re: Linux/MIPS -- DECstation To: jeffm@eznet.net Date: Wed, 7 May 1997 07:52:10 +1000 (EST) Cc: linux-mips@guadalquivir.fnet.fr (Linux MIPS mailing list) In-Reply-To: <199705061439.KAA02474@wks.eznet.net> from "Jeff Mahoney" at May 6, 97 10:39:56 am Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 3591 Lines: 90 Hi Jeff, > Hi. I was wondering if you might be able to answer a few questions for me, > regarding Linux/DECstation? Sure. > 1) What sort of binary compatibility will be supported? Whatever we can manage - clearly we will work towards Ultrix binary compat. however one of the reasons for moving away from Ultrix is to have dynamic libraries as they're so much smaller... so I imagine you'd only want Ultrix compat. for programs you can't recompile such as commercial binaries. > 2) What drivers are currently in progress? The basic ones for the moment, as we've not even got a complete boot of Linux yet! Though I think we're close... and we do appear to be making much better progress now that we have more kernel hackers on board. The drivers we ultimately support are determined by the availability of appropriate documentation as much as what hardware people have on which to test. We've been pestering DEC, and working on other avenues of documentation such as source code for machines for which we have nothing. > I ask because our student group as many DECstations in use, using ULTRIX. > Being that we're a student group, we're looking to find a free OS to run > on these machines. You'd be surprised how many commercial organisations want to run a free OS with dynamic libraries... :-) > Currently, I've been working on getting NetBSD/pmax up on them, but > there is a problem with our using it: It doesn't support the PMAG-[DE] > PXG 3-D dual-turbochannel card. We use these cards in almost all our > machines (that aren't servers). Ahhh... this could be a problem. As far as I know, I have *no* documentation whatsoever on them! I'll have a poke around... > If you need a testing opportunity, we have the following equipment: > > DECstation 5000/{120,125,133,200,240} > PMAZB-AA Dual SCSI bus controller > DEFZA FDDI Interface (CDDI) > DEC LANCE Ethernet Interface (Expansion module, in addition to internal) > PrestoServe NVRAM > > Framebuffers: > PMAG-B > PMAGB-B (67 MHz) (1024x864) > PMAGB-B (199 MHz) (1280x1024) > PMAG-C (1280x1024) > PMAG-[DE] (1280x1024) Dual TurboChannel, Gobs of memory This is a fairly impressive range of hardware. It'd be good to be able to support it all. Does the University have a relationship with DEC that you could use to get some documentation? The kind of thing we need is a hardware functional specification for each device. We have them for: DS2100/3100 DS5000/200 PMAG-B PMAZ...? Basic SCSI controller Basic expansion board ethernet controller Similarities between these and other boards, and the fact that the TurboChannel helps to identify and describe their basic traits will help us support more than these. > If there's anything I can do to help your efforts, I'll see what I can do. Well, as you're in the US, and close to DEC's HQ in Nashua you in a better position to get documentation out of the folks at DEC. Both the NetBSD and Mach operating system source code mentions documents we've never seen... they're the ones we need, in addition to those for your FDDI, NVRAM, Dual SCSI, 3D Graphics etc. cards. Does this help? Regards, Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From frist@lehr.chem.tu-berlin.de Thu May 8 16:57:09 1997 Received-Date: Thu, 8 May 1997 16:57:07 +0200 (MET DST) Received: from lehr.chem.TU-Berlin.DE by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA00105; Wed, 7 May 1997 10:17:38 +0200 (MET) Received: (from frist@localhost) by lehr.chem.tu-berlin.de (8.7.4/8.7.3) id KAA00926; Wed, 7 May 1997 10:20:53 +0200 Date: Wed, 7 May 1997 10:20:53 +0200 (MET DST) From: Frieder Streffer To: linux-mips@guadalquivir.fnet.fr Subject: Re: Another DECStations test kernel In-Reply-To: <199705051425.OAA22992@suede.sw.oz.au> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 1058 Lines: 33 On Tue, 6 May 1997, Paul Antoine wrote: > Hi folks, > > For those of you with DS5000/xx[x]'s you may like to try the new kernel I've > uploaded at: > > ftp://ftp.fnet.fr/linux-mips/softway.com.au/dec_vmlinux.2.1.14.test2.gz > > It now shows BogoMIPS... but is not yet using real console code (tm). I'm > sorry about the large number of debugging printk's, but hey, it's a test > kernel, right?? > > Those with DS3100's will have to be a little patient as Frieder and I > figure out how to use the boot PROM's for console (I'm afraid the real > console code will take some debugging now I've munged... err... moved it > around a bit! :-) > > Regards, > Paul > Hi Paul, I managed to figure out a PROM calling scheme based on netbsd files. The idea behind is just having a interface to the rex_prom. If the rex_prom is not availible (i.e. the PROM_MAGIC is not correct) one tries to rebuild the essential functions with DS3100 ROM calls. I will be off from tomorrow until monday, so I will send you the files in the evening. regards, Frieder From paul@suede.sw.oz.au Thu May 8 16:33:24 1997 Received-Date: Thu, 8 May 1997 16:33:17 +0200 (MET DST) Received: from staff.cs.su.OZ.AU by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA21342; Thu, 8 May 1997 01:12:09 +0200 (MET) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id XAA15852; Wed, 7 May 1997 23:15:18 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id XAA19808; Wed, 7 May 1997 23:15:15 GMT (SMI-8.6/1.34) (from paul for linux-mips@fnet.fr) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705072315.XAA19808@suede.sw.oz.au> Subject: Re: DECStation update... To: linux-mips@guadalquivir.fnet.fr Date: Thu, 8 May 1997 09:15:14 +1000 (EST) In-Reply-To: <199705050310.XAA08654@shell10.vivanet.com> from "Stu Allen" at May 4, 97 11:10:20 pm Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 2518 Lines: 67 Stu Allen wrote: > Well, I grabbed the 2.1.14 test kernel and tried it on my lonely > little 5100. > I was disappointed, but not suprised, to get an immediate register > dump, just as with the previous kernels. Hmmm... I'm feeling better about the possibility of getting your machine going, at least to the point of making something print on the console! In peering at the Mach 3.0 code, and the header files in the standard RISC/OS kernel binary release, it became apparent that early DEC MIPS designs were based *very* heavily on the workstations from MIPS themselves. The DS3100's prom calls are near identical to those of the MIPS 3230's... So it would strike me as entirely possible that the 5100 is very similar to a 3100. > I'm going to type it all in, with hopes that you (or someone else > on the list) might be able to make some sense of it: I'll peer at this more closely when I'm in front of my workstation at home, but one of the first things the kernel does is attempt to use a PROM function to print a string. That function presently bombs on a 3100 as well, and Frieder and I are fixing it. Thinking about your problem some more: it's entirely possible that in the boot PROM of your machine is a jump table of useful functions, but that jump table may be a little different from the 3100's (albeit modelled on the MIPS machines). We should probably write a small program to disassemble the PROM... Another thing you could do is describe the hardware of your machine a bit. Have you taken the lid off your machines? Can you describe how the boards are arranged? Does it have a motherboard with the CPU, ether etc., and optional QBUS slots, or are devices like SCSI and ether on QBUS cards?? > KN230 V1.4 > 08-00-2b-1d-3a-20 > 0x800000 > >> Do you have a list of the commands that the boot prom accepts? What does it print when you type '?' or 'help'? > Stu Allen > Linux kernel-hacker-wannabee! Resistance is useless... :-) Regards, Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From harald.koerfgen@netcologne.de Thu May 8 16:30:47 1997 Received-Date: Thu, 8 May 1997 16:30:46 +0200 (MET DST) From: harald.koerfgen@netcologne.de Received: from ns1.netcologne.de by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA15213; Thu, 8 May 1997 15:41:12 +0200 (MET) Received: from franz.netcologne.de by ns1.netcologne.de (8.6.12/NetCologne/marvin/netsafe-a0020) id ; Thu, 8 May 1997 15:34:08 +0200 with SMTP X-Ncc-Regid: de.netcologne Message-Id: X-Mailer: XFMail 1.0 [p0] on Linux Sender: harry@franz.netcologne.de Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 8bit Mime-Version: 1.0 In-Reply-To: <199705051425.OAA22992@suede.sw.oz.au> Date: Thu, 08 May 1997 15:32:00 +0200 (MEST) To: (Paul Antoine) Subject: RE: Another DECStations test kernel Cc: linux-mips@guadalquivir.fnet.fr Content-Length: 616 Lines: 23 On 05-May-97 Paul Antoine wrote: >Hi folks, > >For those of you with DS5000/xx[x]'s you may like to try the new kernel I've >uploaded at: > > ftp://ftp.fnet.fr/linux-mips/softway.com.au/dec_vmlinux.2.1.14.test2.gz > >It now shows BogoMIPS... but is not yet using real console code (tm). I'm >sorry about the large number of debugging printk's, but hey, it's a test >kernel, right?? Hi, I've tested dec_vmlinux.2.1.14.test2 yesterday on my 5000/133 and it does not get very far. Immediately after displaying the I-Cache and D-Cache sizes it jumps back to the PROM prompt :-(. -- Regards, Harald Koerfgen From allens@vivanet.com Thu May 8 19:23:00 1997 Received-Date: Thu, 8 May 1997 19:22:53 +0200 (MET DST) Received: from vivanet.com (mail.vivanet.com) by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA18887; Thu, 8 May 1997 19:19:24 +0200 (MET) Received: from shell10.vivanet.com (allens@shell10.vivanet.com [192.108.245.8]) by vivanet.com (8.7.5/8.7.3) with ESMTP id NAA07684; Thu, 8 May 1997 13:29:39 -0400 (EDT) From: Stu Allen Received: (from allens@localhost) by shell10.vivanet.com (8.7.5/8.7.3) id NAA21888; Thu, 8 May 1997 13:28:51 -0400 (EDT) Message-Id: <199705081728.NAA21888@shell10.vivanet.com> Subject: Re: DECStation update... To: linux-mips@guadalquivir.fnet.fr Date: Thu, 8 May 1997 13:28:50 -0400 (EDT) In-Reply-To: <199705072315.XAA19808@suede.sw.oz.au> from "Paul Antoine" at May 8, 97 09:15:14 am X-Mailer: ELM [version 2.4 PL25] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 3457 Lines: 90 > > Paul Antoine wrote: > > > not turned on!> > That is _amazling_ accurate ... it's sitting under the table in the corner of my "computer lab" at home. It's probably for the best though - if it were up on top it could see all the other machines that get to actually do something! :-) > Hmmm... I'm feeling better about the possibility of getting your > machine going, at least to the point of making something print on the Cool! > console! In peering at the Mach 3.0 code, and the header files in the > standard RISC/OS kernel binary release, it became apparent that early > DEC MIPS designs were based *very* heavily on the workstations from > MIPS themselves. The DS3100's prom calls are near identical to those > of the MIPS 3230's... > > So it would strike me as entirely possible that the 5100 is very > similar to a 3100. > That would make a lot of sense, since as I remember DEC positioned this beast as the replacement for the 3100, or at least to _us_ they did! I'm actually the one who bought the thing in the first place, and had originally gotten quotes on a 3100 and was all set to order one when the 5100 came out, so I ended up with this instead. > (snip) > Thinking about your problem some more: it's entirely possible that in > the boot PROM of your machine is a jump table of useful functions, but > that jump table may be a little different from the 3100's (albeit > modelled on the MIPS machines). We should probably write a small > program to disassemble the PROM... Based on what's been on the list, and my experiece with the previous kernels, I've suspected all along that a PROM call was probably the culprit. Of course, the lack of any output is also a clue! I _do_ have a working Ultrix 4.5 system on this box, complete with GCC and friends, so I would be more than happy to try compiling and running a "PROM finder" program. > > Another thing you could do is describe the hardware of your machine a > bit. Have you taken the lid off your machines? Can you describe how > the boards are arranged? Does it have a motherboard with the CPU, > ether etc., and optional QBUS slots, or are devices like SCSI and > ether on QBUS cards?? OK this is from my memory of when I last opened the beast, and that was over a year ago. One thing for certain - there is no QBUS (or any other bus) in the box. (I think the web page is wrong on this point BTW.) The only slot it has is for a serial interface card, which I believe was specific to this one system. It added like 4 async and 1 sync interface I believe (it already has 4 async ports). Everything else is on the motherboard - Ethernet (10Base2 & AUI), SCSI, etc. I'll try and open it up again tonight if I get a chance. > > > KN230 V1.4 > > 08-00-2b-1d-3a-20 > > 0x800000 > > >> > > Do you have a list of the commands that the boot prom accepts? What > does it print when you type '?' or 'help'? > I'll have to get you this .... Oh - one other clue - the command to boot using TFTP follows the 2100/3100 convention, which perhaps suggests that the PROM is similar to the 3100's. > > Linux kernel-hacker-wannabee! > > Resistance is useless... :-) > It's funny really. This box only has 8 (!) meg of RAM (I think - can't remeber!) and 2 207 meg hard disks, so chances are it will never really do anything "useful". But it sure will be cool to see it running Linux! From davidm@fuzzbox.psrg.cs.usyd.edu.au Thu May 8 20:07:26 1997 Received-Date: Thu, 8 May 1997 20:07:06 +0200 (MET DST) Received: from fuzzbox.psrg.cs.usyd.edu.au ([129.78.97.193]) by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA19479; Thu, 8 May 1997 20:03:20 +0200 (MET) Received: from fuzzbox.psrg.cs.usyd.edu.au (davidm@fuzzbox.psrg.cs.usyd.edu.au [129.78.97.193]) by fuzzbox.psrg.cs.usyd.edu.au (8.8.4/8.8.4) with SMTP id EAA18673 for ; Fri, 9 May 1997 04:04:59 +1000 Date: Fri, 9 May 1997 04:04:59 +1000 (EST) From: David Monro Sender: davidm@fuzzbox.psrg.cs.usyd.edu.au Reply-To: David Monro Subject: fading support for mips1 cpus :-( To: linux-mips@guadalquivir.fnet.fr Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-Length: 632 Lines: 22 Has the code for mips1 (ie R3000) cpus been neglected recently? I'm trying to compile a kernel for a Magnum 3000 (yes I don't expect it to work yet), and I'm running into trouble. First, asm/atomic.h, we have typedef struct { int counter; } atomic_t and then we do thinks like - atomic_t *v; *v += i; which doesn't compile. Changing all the *v things to v->counter fixes this (this is in the _MIPS_ISA_MIPS1 section). The next problem is that someone has used ll and sc instructions (mips2+) in asm/softirq.c: clear_active_bhs() I have no idea what to do with this one - someone like to write a mips1 version for me? David From paul@suede.sw.oz.au Thu May 8 23:39:31 1997 Received-Date: Thu, 8 May 1997 23:39:29 +0200 (MET DST) Received: from staff.cs.su.OZ.AU by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA22081; Thu, 8 May 1997 23:36:03 +0200 (MET) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id VAA03609; Thu, 8 May 1997 21:39:12 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id VAA05158; Thu, 8 May 1997 21:39:10 GMT (SMI-8.6/1.34) (from paul for linux-mips@fnet.fr) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705082139.VAA05158@suede.sw.oz.au> Subject: Re: fading support for mips1 cpus :-( To: linux-mips@guadalquivir.fnet.fr Date: Fri, 9 May 1997 07:39:10 +1000 (EST) In-Reply-To: from "David Monro" at May 9, 97 04:04:59 am Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 1661 Lines: 34 David Monro wrote: > Has the code for mips1 (ie R3000) cpus been neglected recently? I'm trying to > compile a kernel for a Magnum 3000 (yes I don't expect it to work yet), and > I'm running into trouble. Which source tree are you working from? If it's anything other than the 2.1.1.dec tree, then the 3000 code will almost certainly be very flakey. Most development on the kernel has been done for R4K... with the DECStation port being the only one presently focussing on the R3K (although several people, including myself, wish to get going on the MIPSen now you've written a bfsd boot daemon!) Consequently, it's very easy for the R4k coders to "forget" the limitations of their poorer cousins when writing new code and integrating the DECStation code into the kernel. I.E. The code you have has probably never been compiled to an R3K target... I have fixed a *lot* of R4K-isms in the tree I'm working on, and once that's patched into 2.1.36, I'll release it to Ralf for integrating into the main tree. In the meantime, you may wish to play with the aforementioned 2.1.1.dec source tree... which should be at: ftp://ftp.fnet.fr/linux-mips/ftp.softway.com.au/linux-2.1.1.dec.tar.gz Regards, Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From paul@suede.sw.oz.au Fri May 9 00:40:03 1997 Received-Date: Fri, 9 May 1997 00:40:00 +0200 (MET DST) Received: from staff.cs.su.OZ.AU by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA23790; Fri, 9 May 1997 00:36:36 +0200 (MET) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id WAA05100; Thu, 8 May 1997 22:39:46 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id WAA05492; Thu, 8 May 1997 22:39:46 GMT (SMI-8.6/1.34) (from paul for linux-mips@fnet.fr) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705082239.WAA05492@suede.sw.oz.au> Subject: Re: DECStation update... To: linux-mips@guadalquivir.fnet.fr Date: Fri, 9 May 1997 08:39:45 +1000 (EST) In-Reply-To: <199705081728.NAA21888@shell10.vivanet.com> from "Stu Allen" at May 8, 97 01:28:50 pm Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 3522 Lines: 83 Stu Allen wrote: > > > not turned on!> > > > > That is _amazling_ accurate ... it's sitting under the table in the corner > of my "computer lab" at home. It's probably for the best though - if it > were up on top it could see all the other machines that get to actually do > something! :-) That's very wise; you don't want a jealous computer on your hands... :-) > > So it would strike me as entirely possible that the 5100 is very > > similar to a 3100. > > > > That would make a lot of sense, since as I remember DEC positioned this beast > as the replacement for the 3100, or at least to _us_ they did! I'm actually > the one who bought the thing in the first place, and had originally gotten > quotes on a 3100 and was all set to order one when the 5100 came out, so I > ended up with this instead. Ahhh... this is even better. Can you remember why it was that they thought the 5100 was better? Was it just that it has an R3K CPU?? I suspect it may just be an R3K 3100... > Based on what's been on the list, and my experiece with the previous > kernels, I've suspected all along that a PROM call was probably the > culprit. Of course, the lack of any output is also a clue! Well, once we get the 3100 PROM calls working, you may be pleasantly surprised! > I _do_ have a working Ultrix 4.5 system on this box, complete with GCC and > friends, so I would be more than happy to try compiling and running a > "PROM finder" program. Cool - we'll do this as the next step once the 3100 code is known not to work. > OK this is from my memory of when I last opened the beast, and that was > over a year ago. One thing for certain - there is no QBUS (or any other > bus) in the box. (I think the web page is wrong on this point BTW.) Ohhh... I think I've confused the 5100 with the 5400... > The only slot it has is for a serial interface card, which I believe was > specific to this one system. It added like 4 async and 1 sync interface > I believe (it already has 4 async ports). Everything else is on the > motherboard - Ethernet (10Base2 & AUI), SCSI, etc. That would be consistant too... if it's a server version of the 3100, then one obvious criticism of the 3100 was the lack of serial ports for 'real' server uses. Remember when machines had *lots* of serial ports if they were servers to drive all those dumb terminals??? :-) Our Pyramid at University had 64!! > I'll try and open it up again tonight if I get a chance. Cool... look at the major chips, and note the numbers etc. > Oh - one other clue - the command to boot using TFTP follows the 2100/3100 > convention, which perhaps suggests that the PROM is similar to the 3100's. Yet more evidence... > It's funny really. This box only has 8 (!) meg of RAM (I think - can't > remeber!) and 2 207 meg hard disks, so chances are it will never really do > anything "useful". But it sure will be cool to see it running Linux! You may find that it takes 3100 memory... and as the disks are scsi... Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From allens@vivanet.com Fri May 9 13:59:14 1997 Received-Date: Fri, 9 May 1997 13:59:12 +0200 (MET DST) Received: from vivanet.com (mail.vivanet.com) by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA17416; Fri, 9 May 1997 13:55:48 +0200 (MET) Received: from shell10.vivanet.com (allens@shell10.vivanet.com [192.108.245.8]) by vivanet.com (8.7.5/8.7.3) with ESMTP id IAA17403 for ; Fri, 9 May 1997 08:06:04 -0400 (EDT) From: Stu Allen Received: (from allens@localhost) by shell10.vivanet.com (8.7.5/8.7.3) id IAA03152 for linux-mips@fnet.fr; Fri, 9 May 1997 08:05:15 -0400 (EDT) Message-Id: <199705091205.IAA03152@shell10.vivanet.com> Subject: Re: DECStation update... To: linux-mips@guadalquivir.fnet.fr Date: Fri, 9 May 1997 08:05:14 -0400 (EDT) In-Reply-To: <199705082239.WAA05492@suede.sw.oz.au> from "Paul Antoine" at May 9, 97 08:39:45 am X-Mailer: ELM [version 2.4 PL25] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 747 Lines: 30 OK, here's the output of the "?" command at the >> PROM prompt: (I _really_ need to hook up something in place of my VT320 for a console!) >> ? CMD: boot [-f FILE] [-(s|m)] [-n] [ARG...] conf [-(b|f|m|s|g)] continue d [[[-(b|h|w)] [ADDR]] | [-H REG}} VAL disable DEV dump [-H] | {[[[-(b|h|w)] [-(o|d|u|x|c|B)]]|[-I]] RNG] e [-(b|h|w)] ADDR enable DEV fill [-(b|h|w)] [-v VAL] RNG go [PC] help [CMD] init passwd -(c|s|u) prcache -(u|f|z) printenv [EVAR...] scsi [?] | [cmd [unit] [parm]] setenv EVAR STR test [?] | [[-v] [-(f|h)] [-r[REPS]] [-s] TNUM [parm1 parm2 parm3 ....]] unsetenv EVAR ? [CMD] RNG: [ADDR#CNT] | [ADDR:ADDR] >> Hopefully I'll get a chance to open the box later today or tonight .... From Luc.Beurton@fnet.fr Fri May 9 19:19:41 1997 Received-Date: Fri, 9 May 1997 19:19:41 +0200 (MET DST) Received: from orenoque.fnet.fr (localhost.fnet.fr [127.0.0.1]) by orenoque.fnet.fr with ESMTP (8.8.5/97.02.12/Orenoque); id RAA08050; Fri, 9 May 1997 17:19:40 GMT Message-Id: <199705091719.RAA08050@orenoque.fnet.fr> To: linux-mips@fnet.fr Subject: Fnet is back. X-Mailer: MH 6.8.4 Date: Fri, 09 May 1997 19:19:40 +0200 From: Luc Beurton Content-Length: 118 Lines: 6 Hi all, The mailling-list, ftp and wouaib is up. France Telecomm pass this tree days to connect us !!!! Sorry. Luc. From gb141@columbia.edu Fri May 9 22:11:14 1997 Received-Date: Fri, 9 May 1997 22:11:13 +0200 (MET DST) Received: from mailrelay1.cc.columbia.edu by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA27430; Fri, 9 May 1997 22:07:50 +0200 (MET) Received: from dogma.columbia.edu (dialup-22-9.cc.columbia.edu [128.59.47.93]) by mailrelay1.cc.columbia.edu (8.8.5/8.8.5) with SMTP id QAA19484 for ; Fri, 9 May 1997 16:11:08 -0400 (EDT) Message-Id: <33738444.1FB7@columbia.edu> Date: Fri, 09 May 1997 16:08:37 -0400 From: Gregory Boduch Reply-To: gb141@columbia.edu Organization: http://www.juilliard.edu http://www.columbia.edu X-Mailer: Mozilla 3.01Gold (Win95; I) Mime-Version: 1.0 To: linux-mips@guadalquivir.fnet.fr Subject: Linux-MIPS on SGI O2 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Length: 155 Lines: 9 Hi, I did not find any references to running Linux-MIPS on any of the SGI machines, especially on R5000 O2s? Could you explain. Thanks in advance, GB From harald.koerfgen@netcologne.de Fri May 9 23:15:38 1997 Received-Date: Fri, 9 May 1997 23:15:37 +0200 (MET DST) From: harald.koerfgen@netcologne.de Received: from ns1.netcologne.de by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA28707; Fri, 9 May 1997 23:12:13 +0200 (MET) Received: from franz.netcologne.de by ns1.netcologne.de (8.6.12/NetCologne/marvin/netsafe-a0020) id ; Fri, 9 May 1997 23:04:55 +0200 with SMTP X-Ncc-Regid: de.netcologne Message-Id: X-Mailer: XFMail 1.0 [p0] on Linux Sender: harry@franz.netcologne.de Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 8bit Mime-Version: 1.0 In-Reply-To: <199705082229.WAA05440@suede.sw.oz.au> Date: Fri, 09 May 1997 22:35:17 +0200 (MEST) To: (Paul Antoine) , linux-mips@guadalquivir.fnet.fr Subject: Re: Another DECStations test kernel Content-Length: 1349 Lines: 40 Hello all, Paul wrote: >Harald wrote: > >> I've tested dec_vmlinux.2.1.14.test2 yesterday on my 5000/133 and it does not >> get very far. Immediately after displaying the I-Cache and D-Cache sizes it >> jumps back to the PROM prompt :-(. > >Does it do this every time? I'll have a look at what it's doing, but it is >strange that it dies there, as it really shouldn't... there are lots more >exciting places for it to die :-) > >Can anyone else with a 5000/1xx confirm this?? > >Paul Paul, I have removed my PMAG-E Video Card (one of those fancy 3D-Cards with an i860, lots of VRAM and everything...) from my DECstation and the test2 kernel is running until showing the "calibration delay loop ..." message, but no BogoMIPS ;-). I remember having read of similar problems in the NetBSD Mailing List Archive. This lead me to the assumption, that the Vertical Retrace Interrupts from the PMAG-E are coming through. On your maxine the TC Slot Interrupts are handled by the IOAsic, on my kmin the TC Interrupts are directly connected to the CPU. Does the test2 kernel unmask the IRQ0 and IRQ1 bits in the CP0 Status? If it does, that could explain this unexpected behaviour. -- Regards, Harald Koerfgen The first time, it's a KLUDGE! The second, a trick. Later, it's a well-established technique! -- Mike Broido, Intermetrics From paul@suede.sw.oz.au Sat May 10 13:08:39 1997 Received-Date: Sat, 10 May 1997 13:08:37 +0200 (MET DST) Received: from staff.cs.su.OZ.AU by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA21190; Sat, 10 May 1997 13:05:11 +0200 (MET) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id HAA29371; Sat, 10 May 1997 07:42:23 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id HAA18655; Sat, 10 May 1997 07:42:22 GMT (SMI-8.6/1.34) (from paul for linux-mips@fnet.fr) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705100742.HAA18655@suede.sw.oz.au> Subject: Re: DECStation update... To: linux-mips@guadalquivir.fnet.fr Date: Sat, 10 May 1997 17:42:21 +1000 (EST) In-Reply-To: <199705091205.IAA03152@shell10.vivanet.com> from "Stu Allen" at May 9, 97 08:05:14 am Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 1306 Lines: 46 Stu Allen wrote: > OK, here's the output of the "?" command at the >> PROM prompt: > > (I _really_ need to hook up something in place of my VT320 for a console!) Yes! > >> ? > CMD: > boot [-f FILE] [-(s|m)] [-n] [ARG...] > conf [-(b|f|m|s|g)] > continue > d [[[-(b|h|w)] [ADDR]] | [-H REG}} VAL > disable DEV > dump [-H] | {[[[-(b|h|w)] [-(o|d|u|x|c|B)]]|[-I]] RNG] > e [-(b|h|w)] ADDR > enable DEV > fill [-(b|h|w)] [-v VAL] RNG > go [PC] > help [CMD] > init > passwd -(c|s|u) > prcache -(u|f|z) > printenv [EVAR...] > scsi [?] | [cmd [unit] [parm]] > setenv EVAR STR > test [?] | [[-v] [-(f|h)] [-r[REPS]] [-s] TNUM [parm1 parm2 parm3 ....]] > unsetenv EVAR > ? [CMD] > RNG: [ADDR#CNT] | [ADDR:ADDR] > >> Can anyone with a 3100 confirm that this is approximately what is available on their boot PROM's?? Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From paul@suede.sw.oz.au Sat May 10 15:59:19 1997 Received-Date: Sat, 10 May 1997 15:59:17 +0200 (MET DST) Received: from staff.cs.su.OZ.AU by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA25109; Sat, 10 May 1997 15:55:51 +0200 (MET) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id NAA03809; Sat, 10 May 1997 13:58:47 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id NAA19648; Sat, 10 May 1997 13:58:46 GMT (SMI-8.6/1.34) (from paul for linux-mips@fnet.fr) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705101358.NAA19648@suede.sw.oz.au> Subject: vmalloc help?? To: linux-mips@guadalquivir.fnet.fr (Linux MIPS mailing list) Date: Sat, 10 May 1997 23:58:45 +1000 (EST) Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 789 Lines: 19 Hi folks, I've discovered the latest point at which the DECStation kernel is failing: namely in accessing KSEG2 after having vmalloc'd a hash table for the buffers. Can anyone with good knowledge of this help? It seems that vmalloc is allocating an address of 0xc0000000 for the table, and that any access of this hangs... Regards, Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From davidm@fuzzbox.psrg.cs.usyd.edu.au Sat May 10 16:29:41 1997 Received-Date: Sat, 10 May 1997 16:29:40 +0200 (MET DST) Received: from fuzzbox.psrg.cs.usyd.edu.AU by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA25770; Sat, 10 May 1997 16:26:15 +0200 (MET) Received: from fuzzbox.psrg.cs.usyd.edu.au (davidm@fuzzbox.psrg.cs.usyd.edu.au [129.78.97.193]) by fuzzbox.psrg.cs.usyd.edu.au (8.8.4/8.8.4) with SMTP id AAA05013 for ; Sun, 11 May 1997 00:24:52 +1000 Date: Sun, 11 May 1997 00:24:52 +1000 (EST) From: David Monro Sender: davidm@fuzzbox.psrg.cs.usyd.edu.au Reply-To: David Monro Subject: help with asm errors To: linux-mips@guadalquivir.fnet.fr Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-Length: 919 Lines: 19 Background - I'm trying to compile Paul's decstation kernel for a Magnum 3000. I'm using Paul's kernel because the 2.1.36 has too many r40000isms. I'm using the include/asm/bitops.h from 2.1.36 (since the 2.1.1 bitops.h fails for a big endian compile). I get most of the way there and then it dies with the following error: mips-linux-gcc -D__KERNEL__ -I/amd/milawa/root/u2/staff/davidm/mips/li nux-2.1.1.dec/include -Wall -Wstrict-prototypes -O2 -fomit-frame-point er -G 0 -mno-abicalls -fno-pic -mcpu=r3000 -mips1 -c head.S -o head.o head.S: Assembler messages: head.S:665: Warning: Used $at without ".set noat" head.S:705: Error: attempt to .org/.space backwards? (-53796) head.S:705: Fatal error: Cannot write to output file. make[1]: *** [head.o] Error 1 Line 665 is SAVE_ALL, and line 705 is the end of file. Not being intimately familiar with mips asm, I currently don't have much clue here... help? David From tschroed@cheetah.wsc.edu Sat May 10 22:48:17 1997 Received-Date: Sat, 10 May 1997 22:48:16 +0200 (MET DST) Received: from cheetah.wsc.edu by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA29945; Sat, 10 May 1997 22:44:52 +0200 (MET) Received: by cheetah.wsc.edu (5.65/DEC-Ultrix/4.4) id AA02190; Sat, 10 May 1997 15:47:57 -0500 Other-Stuff: localhost tschroed@cheetah.wsc.edu Trevor Schroeder tschroed Date: Sat, 10 May 1997 15:47:57 -0500 (CDT) From: Trevor Schroeder To: Linux/MIPS list Subject: DECstation stuff Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 561 Lines: 11 This may seem kind of wacky, but lacking any nice docs like the one describing KN02, why not poke around /usr/sys/machine/mips/kn*h on ULTRIX to get a feel for the other firmwares? Or am I missing something? ____________________________________________________________ "One unerring mark of the love of truth is not entertaining any propositions with greater assurance than the proofs it is built upon will warrant" -- John Locke, 1690 Trevor Schroeder tschroed@cheetah.wsc.edu ------------------------------------------------------------ From paul@suede.sw.oz.au Sun May 11 00:18:23 1997 Received-Date: Sun, 11 May 1997 00:18:21 +0200 (MET DST) Received: from staff.cs.su.OZ.AU by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA02216; Sun, 11 May 1997 00:14:56 +0200 (MET) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id WAA08711; Sat, 10 May 1997 22:17:53 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id WAA20759; Sat, 10 May 1997 22:17:52 GMT (SMI-8.6/1.34) (from paul for linux-mips@fnet.fr) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705102217.WAA20759@suede.sw.oz.au> Subject: Re: DECstation stuff To: linux-mips@guadalquivir.fnet.fr Date: Sun, 11 May 1997 08:17:51 +1000 (EST) In-Reply-To: from "Trevor Schroeder" at May 10, 97 03:47:57 pm Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 1277 Lines: 25 Trevor Schroeder wrote: > This may seem kind of wacky, but lacking any nice docs like the one describing > KN02, why not poke around /usr/sys/machine/mips/kn*h on ULTRIX to get a feel > for the other firmwares? Or am I missing something? This kind of information is fine, but doesn't tell you how to talk to the chips, for instance, just where they are. Common chips like the LANCE ethernet chip are fine anyway - we do just need to know where it is in the address map - however talking to the Personal DECStation's Access.Bus keyboard and mouse interface is another thing entirely... let alone the special ASIC's used in some models (many functions of which are obvious, fortunately!). If someone would tar up the Ultrix header files, I'd appreciate it... I forgot to do this prior to erasing my Ultrix partition. :-) Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From tschroed@cheetah.wsc.edu Sun May 11 00:30:03 1997 Received-Date: Sun, 11 May 1997 00:30:02 +0200 (MET DST) Received: from cheetah.wsc.edu by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA02579; Sun, 11 May 1997 00:26:38 +0200 (MET) Received: by cheetah.wsc.edu (5.65/DEC-Ultrix/4.4) id AA02298; Sat, 10 May 1997 17:29:44 -0500 Other-Stuff: localhost tschroed@cheetah.wsc.edu Trevor Schroeder tschroed Date: Sat, 10 May 1997 17:29:43 -0500 (CDT) From: Trevor Schroeder To: linux-mips@guadalquivir.fnet.fr Subject: Re: DECstation stuff In-Reply-To: <199705102217.WAA20759@suede.sw.oz.au> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 1394 Lines: 30 On Sun, 11 May 1997, Paul Antoine wrote: > This kind of information is fine, but doesn't tell you how to talk to the > chips, for instance, just where they are. Common chips like the LANCE > ethernet chip are fine anyway - we do just need to know where it is in the > address map - however talking to the Personal DECStation's Access.Bus keyboard *nods* Right, that was kind of what I meant. But a lot of these guys have the same(ish) hardware, just different offsets for functions. It's better than nothing. And besides, using the same trick, it's certainly possible (depending on how dedicated a person is) to reverse engineer the use of those other devices as well. > and mouse interface is another thing entirely... let alone the special ASIC's > used in some models (many functions of which are obvious, fortunately!). > > If someone would tar up the Ultrix header files, I'd appreciate it... I > forgot to do this prior to erasing my Ultrix partition. :-) > *gasp* But Paul, isn't that a violation of the licensing agreement? ____________________________________________________________ "One unerring mark of the love of truth is not entertaining any propositions with greater assurance than the proofs it is built upon will warrant" -- John Locke, 1690 Trevor Schroeder tschroed@cheetah.wsc.edu ------------------------------------------------------------ From paul@suede.sw.oz.au Sun May 11 01:04:41 1997 Received-Date: Sun, 11 May 1997 01:04:40 +0200 (MET DST) Received: from staff.cs.su.OZ.AU by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA03295; Sun, 11 May 1997 01:01:16 +0200 (MET) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id XAA09176; Sat, 10 May 1997 23:04:16 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id XAA20867; Sat, 10 May 1997 23:04:15 GMT (SMI-8.6/1.34) (from paul for linux-mips@fnet.fr) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705102304.XAA20867@suede.sw.oz.au> Subject: Re: help with asm errors To: linux-mips@guadalquivir.fnet.fr Date: Sun, 11 May 1997 09:04:14 +1000 (EST) In-Reply-To: from "David Monro" at May 11, 97 00:24:52 am Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 2597 Lines: 66 David Monro wrote: > Background - I'm trying to compile Paul's decstation kernel for a Magnum > 3000. I'm using Paul's kernel because the 2.1.36 has too many r40000isms. > I'm using the include/asm/bitops.h from 2.1.36 (since the 2.1.1 bitops.h > fails for a big endian compile). I get most of the way there and then it > dies with the following error: > > mips-linux-gcc -D__KERNEL__ -I/amd/milawa/root/u2/staff/davidm/mips/li > nux-2.1.1.dec/include -Wall -Wstrict-prototypes -O2 -fomit-frame-point > er -G 0 -mno-abicalls -fno-pic -mcpu=r3000 -mips1 -c head.S -o head.o > head.S: Assembler messages: > head.S:665: Warning: Used $at without ".set noat" > head.S:705: Error: attempt to .org/.space backwards? (-53796) > head.S:705: Fatal error: Cannot write to output file. > make[1]: *** [head.o] Error 1 > > Line 665 is SAVE_ALL, Don't worry about this one for now... > and line 705 is the end of file. Not being intimately > familiar with mips asm, I currently don't have much clue here... help? The fatal error at 705 is saying that the 'origin' or asm equivalent of the PC is being moved backwards in the address space by either a .org or .space directive. The .org sets the present compile address, so: .org 0x80000000 would make the assembler compile any code that follows begining at that address. The assembler is having trouble with the series of .org directives at the end of head.S that look like this: .org 0x1000 EXPORT(swapper_pg_dir) .org 0x2000 EXPORT(empty_bad_page) etc. These should result in the 0x1000 being 'added' to the link address for this module (e.g. on the DECStation swapper_pg_dir is at 0x80031000, and empty_bad_page is at 0x80033000). It would seem that the assembler thinks that this address is 'backwards' from where it's up to, the only reason for which is that there is somehow more code in there than there was! This is hard to believe, because on my machine, the cpu_probe routine finishes about 0x770 bytes from the link address... Try compiling with the '.org' lines commented out, and then use objdump to dissassemble the module to see where cpu_probe ends... Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From paul@suede.sw.oz.au Sun May 11 01:15:58 1997 Received-Date: Sun, 11 May 1997 01:15:57 +0200 (MET DST) Received: from staff.cs.su.OZ.AU by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA03573; Sun, 11 May 1997 01:12:32 +0200 (MET) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id XAA09285; Sat, 10 May 1997 23:15:09 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id XAA20894; Sat, 10 May 1997 23:15:04 GMT (SMI-8.6/1.34) (from paul for linux-mips@fnet.fr) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705102315.XAA20894@suede.sw.oz.au> Subject: Re: DECstation stuff To: linux-mips@guadalquivir.fnet.fr Date: Sun, 11 May 1997 09:15:04 +1000 (EST) In-Reply-To: from "Trevor Schroeder" at May 10, 97 05:29:43 pm Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 1628 Lines: 35 Trevor wrote: > *nods* Right, that was kind of what I meant. But a lot of these guys have > other same(ish) hardware, just different offsets for functions. It's better > than nothing. And besides, using the same trick, it's certainly possible > (depending on how dedicated a person is) to reverse engineer the use of those > other devices as well. True, which is why we've been using this technique. I've been relying on the Mach3.0 source, as that has the most complete support for the various models. Others have been using OpenBSD/NetBSD/... > > If someone would tar up the Ultrix header files, I'd appreciate it... I > > forgot to do this prior to erasing my Ultrix partition. :-) > > *gasp* > But Paul, isn't that a violation of the licensing agreement? I have a valid Ultrix license, and I wasn't proposing copying those files for use in Linux/MIPS, just getting hold of that which my license allows anyway (without having to re-install Ultrix!!) Seriously though, we can't copy these files willy-nilly, and we won't. The Mach sources allow copying provided the copyright is attributed - presently I'm mostly using header files and little bits of code. Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From paul@suede.sw.oz.au Sun May 11 01:45:28 1997 Received-Date: Sun, 11 May 1997 01:45:27 +0200 (MET DST) Received: from staff.cs.su.OZ.AU by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA04205; Sun, 11 May 1997 01:42:02 +0200 (MET) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id XAA09613; Sat, 10 May 1997 23:45:08 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id XAA20963; Sat, 10 May 1997 23:45:07 GMT (SMI-8.6/1.34) (from paul for ) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705102345.XAA20963@suede.sw.oz.au> Subject: Re: 5100 DECsystem To: s100962@student.uq.edu.au Date: Sun, 11 May 1997 09:45:06 +1000 (EST) Cc: linux-mips@guadalquivir.fnet.fr (Linux MIPS mailing list) In-Reply-To: <199705101733.DAA20558@student.uq.edu.au> from "s100962@student.uq.edu.au" at May 11, 97 03:33:07 am Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 1337 Lines: 44 Hi Andrew, > I found some info on the 5100 in my DEC promo stuff for a few > years back.. inicdentally, I found there is another Q-Bus machine.. the > DECsystem 5500... Yeah, I think the 5400 and 5500 are Q-Bus. > On the 5100.. Stats here are : > CPU: R3000/R3010 > SPEC 16.2 > MIPS 27.3 > Clock speed 25 Mhz > Cache size 64 KB > 64 KB data > > Mem : 8MB - 128MB parity (probably same ram as 3100/2100 - 5000/1** > series) > > I/O Bus type: N/A I think the 5100 is a 3100 with an R3000... and a single slot that's used to allow more serial ports. > The box comes in the same case as my Micro-Vax 3100, which is how I found > the reference... Ahhh.... what are the Micro-Vaxen?? Are they MIPS-based, or really Vax microprocessor systems? > Limited info.. and certainly no tech stuff.. Thanks - I'm writing the code to assume a 3100 with R3000, so we'll see what happens! Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From davidm@fuzzbox.psrg.cs.usyd.edu.au Sun May 11 03:49:46 1997 Received-Date: Sun, 11 May 1997 03:49:45 +0200 (MET DST) Received: from fuzzbox.psrg.cs.usyd.edu.AU by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA06785; Sun, 11 May 1997 03:46:19 +0200 (MET) Received: from fuzzbox.psrg.cs.usyd.edu.au (davidm@fuzzbox.psrg.cs.usyd.edu.au [129.78.97.193]) by fuzzbox.psrg.cs.usyd.edu.au (8.8.4/8.8.4) with SMTP id LAA11776 for ; Sun, 11 May 1997 11:40:48 +1000 Date: Sun, 11 May 1997 11:40:48 +1000 (EST) From: David Monro Sender: davidm@fuzzbox.psrg.cs.usyd.edu.au Reply-To: David Monro Subject: Re: help with asm errors To: linux-mips@guadalquivir.fnet.fr In-Reply-To: <199705102304.XAA20867@suede.sw.oz.au> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-Length: 1922 Lines: 53 Paul wrote [...] > The assembler is having trouble with the series of .org directives at the > end of head.S that look like this: > > .org 0x1000 > EXPORT(swapper_pg_dir) > > .org 0x2000 > EXPORT(empty_bad_page) > > etc. Got it - in the head.s that I have, these are actually .org . + 0x1000 and then there is a #ifndef CONFIG_MIPS_DECSTATION around the .org for kernel_entry as follows: #ifndef CONFIG_MIPS_DECSTATION .org 0x8000 #else .org . + 0x8000 #endif /* CONFIG_MIPS_DECSTATION */ So of course I was getting the .org 0x8000 rather than . + 0x8000, and since the other entries used . +, I had gone over 0x8000 already. Changing them all back to just .org 0x?000 allows me to assemble head.S. I note they are just .org 0x?000 in the 2.1.36 version (which seems to have a radically different head.S anyway). However I now get a really weird error when linking: mips-linux-ld -N -G 0 -Ttext 0x80100000 arch/mips/kernel/head.o init/main.o in it/version.o \ arch/mips/mips1/mips.o arch/mips/kernel/kernel.o arch/mips/mm/mm.o kern el/kernel.o mm/mm.o fs/fs.o ipc/ipc.o net/network.a \ fs/filesystems.a \ drivers/block/block.a drivers/char/char.a drivers/net/net.a \ arch/mips/lib/lib.a /amd/milawa/root/u2/staff/davidm/mips/linux-2.1.1.dec/lib/lib.a arch/mips/lib/lib.a -o vmlinux mips-linux-ld: warning: cannot find entry symbol __start; defaulting to 0000000 080100000 mips-linux-ld: vmlinux: Not enough room for program headers (allocated 3, need 4) mips-linux-ld: final link failed: Bad value make: *** [vmlinux] Error 1 I'm not too worried about the 'can't find symbol start', but the other message is very odd. This is using binutils-2.8 and gcc-2.7.2.2 with gcc-2.7.2-3.diffs and float.h found on fnet, all compiled with --target=mips-linux. Anybody seen this one? David From paul@suede.sw.oz.au Sun May 11 06:03:59 1997 Received-Date: Sun, 11 May 1997 06:03:58 +0200 (MET DST) Received: from staff.cs.su.OZ.AU by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA10435; Sun, 11 May 1997 06:00:32 +0200 (MET) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id EAA12358; Sun, 11 May 1997 04:03:19 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id EAA21648; Sun, 11 May 1997 04:03:17 GMT (SMI-8.6/1.34) (from paul for linux-mips@fnet.fr) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705110403.EAA21648@suede.sw.oz.au> Subject: Re: help with asm errors To: linux-mips@guadalquivir.fnet.fr Date: Sun, 11 May 1997 14:03:16 +1000 (EST) In-Reply-To: from "David Monro" at May 11, 97 11:40:48 am Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 2648 Lines: 66 David wrote: > Got it - in the head.s that I have, these are actually > > .org . + 0x1000 > and then there is a #ifndef CONFIG_MIPS_DECSTATION around the .org for > kernel_entry as follows: > #ifndef CONFIG_MIPS_DECSTATION > .org 0x8000 > #else > .org . + 0x8000 > #endif /* CONFIG_MIPS_DECSTATION */ > > So of course I was getting the .org 0x8000 rather than . + 0x8000, and since > the other entries used . +, I had gone over 0x8000 already. Ahhh... you fell foul of a previous hack of mine! :-) > Changing them all back to just .org 0x?000 allows me to assemble head.S. > I note they are just .org 0x?000 in the 2.1.36 version (which seems to > have a radically different head.S anyway). Yes, so don't worry about it. The 2.1.14 tree is compiling fine like this. > However I now get a really weird error when linking: > mips-linux-ld -N -G 0 -Ttext 0x80100000 arch/mips/kernel/head.o init/main.o > init/version.o \ > arch/mips/mips1/mips.o arch/mips/kernel/kernel.o arch/mips/mm/mm.o > kernel/kernel.o mm/mm.o fs/fs.o ipc/ipc.o net/network.a \ > fs/filesystems.a \ > drivers/block/block.a drivers/char/char.a drivers/net/net.a \ > arch/mips/lib/lib.a > /amd/milawa/root/u2/staff/davidm/mips/linux-2.1.1.dec/lib/lib.a > arch/mips/lib/lib.a -o vmlinux > mips-linux-ld: warning: cannot find entry symbol __start; defaulting to > 0000000 080100000 Are you sure you want to link it as high as 0x80100000?? Even the DECStation only steals 192K of memory down low... > mips-linux-ld: vmlinux: Not enough room for program headers (allocated 3, > need 4) > mips-linux-ld: final link failed: Bad value > make: *** [vmlinux] Error 1 > > I'm not too worried about the 'can't find symbol start', Yes, this is probably just a problem in your ld.script. > but the other message is very odd. This is using binutils-2.8 and > gcc-2.7.2.2 with gcc-2.7.2-3.diffs and float.h found on fnet, all > compiled with --target=mips-linux. Anybody seen this one? No - what object format are you compiling to? I now leave all intermediate object files in elf and specify ecoff-littlemips to the linker as the final output format. Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From paul@suede.sw.oz.au Sun May 11 06:41:42 1997 Received-Date: Sun, 11 May 1997 06:41:41 +0200 (MET DST) Received: from staff.cs.su.OZ.AU by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA13598; Sun, 11 May 1997 06:38:16 +0200 (MET) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id EAA12661; Sun, 11 May 1997 04:41:13 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id EAA21700; Sun, 11 May 1997 04:41:10 GMT (SMI-8.6/1.34) (from paul for linux-mips@fnet.fr) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705110441.EAA21700@suede.sw.oz.au> Subject: Good news for hackers of MIPS Co. machines To: linux-mips@guadalquivir.fnet.fr (Linux MIPS mailing list) Date: Sun, 11 May 1997 14:41:09 +1000 (EST) Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 1025 Lines: 21 Hi folks, I've just downloaded the MK84 release of Mach 3.0, and have been tooling around in the source for the MIPS machines it supports. I have discovered that the 2100/3100 and DS5000/200 machines are near copies of MIPSen... the boot prom entry points are almost the same! The only difference I've found is the PUTCHAR call which is offset 12 on MIPSen and 13 on DECStations... This would seem to indicate that support for the MIPSen will follow quite quickly on the heels of the DECStations... especially given the progress that David Monro seems to be making. Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From s100962@student.uq.edu.au Sun May 11 10:05:29 1997 Received-Date: Sun, 11 May 1997 10:05:28 +0200 (MET DST) Received: from student.uq.edu.au by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA18641; Sun, 11 May 1997 10:02:03 +0200 (MET) Received: from localhost (s100962@localhost [127.0.0.1]) by student.uq.edu.au (8.8.5/8.8.5) with SMTP id SAA27578; Sun, 11 May 1997 18:05:22 +1000 (GMT+1000) Date: Sun, 11 May 1997 18:05:21 +1000 (GMT+1000) From: Emanuele John Gelsi To: Paul Antoine Cc: Linux MIPS mailing list Subject: Re: 5100 DECsystem In-Reply-To: <199705102345.XAA20963@suede.sw.oz.au> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 661 Lines: 20 On Sun, 11 May 1997, Paul Antoine wrote: > Hi Andrew, Hi.. > > > The box comes in the same case as my Micro-Vax 3100, which is how I found > > the reference... > > Ahhh.... what are the Micro-Vaxen?? Are they MIPS-based, or really Vax > microprocessor systems? No, they are true Vax based systems.. The only real difference between a MicroVax 3100 and a Vaxstation 3100 is the lack of a framebuffer.. > > Thanks - I'm writing the code to assume a 3100 with R3000, so we'll see > what happens! The 3100/2100 was actually listed as having a SCSI bus yet the 5100 was listed as N/A... Weird.. Andrew (Playing around with a UDB tonight that has a dead floppy) From davidm@fuzzbox.psrg.cs.usyd.edu.au Sun May 11 18:21:59 1997 Received-Date: Sun, 11 May 1997 18:21:58 +0200 (MET DST) Received: from fuzzbox.psrg.cs.usyd.edu.AU by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA28595; Sun, 11 May 1997 18:18:31 +0200 (MET) Received: from fuzzbox.psrg.cs.usyd.edu.au (davidm@fuzzbox.psrg.cs.usyd.edu.au [129.78.97.193]) by fuzzbox.psrg.cs.usyd.edu.au (8.8.4/8.8.4) with SMTP id CAA19834 for ; Mon, 12 May 1997 02:19:52 +1000 Date: Mon, 12 May 1997 02:19:52 +1000 (EST) From: David Monro Sender: davidm@fuzzbox.psrg.cs.usyd.edu.au Reply-To: David Monro Subject: 3230 - it links and loads - and faults... To: linux-mips@guadalquivir.fnet.fr Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-Length: 1358 Lines: 31 Well, I actually got a kernel for the magnum3000 to link, and got the machine to download it over the net and attempt to run it.... Firstly - thanks to Paul for lots of help and advice. Linking with -b elf32-bigmips -oformat ecoff-bigmips was most of the trick. Apart from that, creating a dummy file for a few undefined references (mem start and end etc) and commenting out a pmax_print was most of what was needed. I also copied the dec/ld.script and just changed the ouptut format and the origin, which probably isn't quite right.... Of course, it faults immediately with no other output - but hey, what else would I expect: EXCEPTION: Exception pc: 0x8010ed4c Cause register: 0x30001010 Status register: 0x8004 RAMBO Regs: Error=0x200100 Ch1Mode=0x400 Ch2Mode=0x60000047 Bad Vaddress: 0xfffffff9 The pc is in bi_TagFind, which actually seems to be right (kernel_entry calls bi_EarlySnarf calls bi_TagFind, which will fault on a random address since mips_memory_upper contains whatever happened to be an a0 when I booted it. I guess I had better write a pre-kernel_entry pice of code to sort this out). Well, I'm off to bed - its after 2am. Oh, btw - what is the code name for mangum 3000 machines? It seems to be jazz for the 4000. I need it for my machine-specific directory name :-) David From paul@suede.sw.oz.au Mon May 12 03:27:06 1997 Received-Date: Mon, 12 May 1997 03:27:05 +0200 (MET DST) Received: from staff.cs.su.OZ.AU by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA08217; Mon, 12 May 1997 03:23:39 +0200 (MET) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id BAA26098; Mon, 12 May 1997 01:26:36 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id BAA28557; Mon, 12 May 1997 01:26:36 GMT (SMI-8.6/1.34) (from paul for linux-mips@fnet.fr) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705120126.BAA28557@suede.sw.oz.au> Subject: Re: 3230 - it links and loads - and faults... To: linux-mips@guadalquivir.fnet.fr Date: Mon, 12 May 1997 11:26:35 +1000 (EST) In-Reply-To: from "David Monro" at May 12, 97 02:19:52 am Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 2944 Lines: 78 David Monro, very early this morning, wrote > Well, I actually got a kernel for the magnum3000 to link, and got > the machine to download it over the net and attempt to run it.... Yay!!! > Firstly - thanks to Paul for lots of help and advice. Linking with > -b elf32-bigmips -oformat ecoff-bigmips was most of the trick. Apart > from that, creating a dummy file for a few undefined references (mem > start and end etc) and commenting out a pmax_print was most of what > was needed. I also copied the dec/ld.script and just changed the > ouptut format and the origin, which probably isn't quite right.... Well, it's actually fine for the moment. > Of course, it faults immediately with no other output - but hey, > what else would I expect: > > EXCEPTION: > Exception pc: 0x8010ed4c > Cause register: 0x30001010 > Status register: 0x8004 > RAMBO Regs: Error=0x200100 Ch1Mode=0x400 Ch2Mode=0x60000047 > Bad Vaddress: 0xfffffff9 This is still pretty momentus! > The pc is in bi_TagFind, which actually seems to be right > (kernel_entry calls bi_EarlySnarf calls bi_TagFind, which will fault > on a random address since mips_memory_upper contains whatever > happened to be an a0 when I booted it. I guess I had better write a > pre-kernel_entry piece of code to sort this out). Indeed - in fact you should be able to use a lot of the DECStation code for this, particularly: mips/dec/ boot.S decstation.c (modified somewhat into magnum.c) bitags.c (needed to be able to add tags) mips/lib/ dec_promio.S dec_whichprom.c (don't know if you need this) This code's in the 2.1.14, and dec-promio.S file has all the new stuff I've integrated from Frieder et al for the DS2100/3100's. You should be able to use most of it, as the PROM-callbacks for the MIPSen are near identical. Frieder has also included memory-sizing code that you should be able to use, though I think the MIPSen may have a way of getting that info. from the PROMs... I'll probably be uploading a DEC 2.1.14 source snapshot for folks to play with later this week, prior to integrating the code into 2.1.36 and doing a formal release. > Well, I'm off to bed - its after 2am. And a well deserved rest it is too... > Oh, btw - what is the code name for magnum 3000 machines? It seems > to be jazz for the 4000. I need it for my machine-specific directory > name :-) I think 'magnum' would be fine, as jazz is used for the 4000. :-) Well done! Regards, Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From paul@suede.sw.oz.au Mon May 12 03:46:13 1997 Received-Date: Mon, 12 May 1997 03:46:12 +0200 (MET DST) Received: from staff.cs.su.OZ.AU by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA08612; Mon, 12 May 1997 03:42:45 +0200 (MET) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id BAA26412; Mon, 12 May 1997 01:45:46 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id BAA28956; Mon, 12 May 1997 01:45:45 GMT (SMI-8.6/1.34) (from paul for linux-mips@fnet.fr) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705120145.BAA28956@suede.sw.oz.au> Subject: New DECStation kernel binary to try... To: linux-mips@guadalquivir.fnet.fr (Linux MIPS mailing list) Date: Mon, 12 May 1997 11:45:44 +1000 (EST) Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 1187 Lines: 35 Hi folks, I have uploaded another test kernel binary for DECStation people to try: ftp://ftp.fnet.fr/linux-mips/ftp.softway.com.au/dec_vmlinux.2.1.14.test3.gz It should be at fnet shortly. Features of this wonder of co-operative engineering: Should now print on DS2100/3100's and maybe the DS5100 (cross fingers and hope, Stu :-) The previous test2 binary didn't work due to a *dumb* bug on my part... :-( Now distinguishes more types of DECStations, and has machine-specific setup code for each. May, in fact, show BogoMIPS on most machines... Will, in fact, die in the tlb load exception handler! (This is the current bug under investigation.) Please try it and report your findings on the mailing list. Regards, Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From allens@vivanet.com Mon May 12 05:36:59 1997 Received-Date: Mon, 12 May 1997 05:36:59 +0200 (MET DST) Received: from vivanet.com (mail.vivanet.com) by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA11829; Mon, 12 May 1997 05:33:34 +0200 (MET) Received: from shell10.vivanet.com (allens@shell10.vivanet.com [192.108.245.8]) by vivanet.com (8.7.5/8.7.3) with ESMTP id XAA27037 for ; Sun, 11 May 1997 23:43:55 -0400 (EDT) From: Stu Allen Received: (from allens@localhost) by shell10.vivanet.com (8.7.5/8.7.3) id XAA14358 for linux-mips@fnet.fr; Sun, 11 May 1997 23:43:08 -0400 (EDT) Message-Id: <199705120343.XAA14358@shell10.vivanet.com> Subject: Re: New DECStation kernel binary to try... To: linux-mips@guadalquivir.fnet.fr Date: Sun, 11 May 1997 23:43:08 -0400 (EDT) In-Reply-To: <199705120145.BAA28956@suede.sw.oz.au> from "Paul Antoine" at May 12, 97 11:45:44 am X-Mailer: ELM [version 2.4 PL25] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 905 Lines: 32 > > Should now print on DS2100/3100's and maybe the DS5100 (cross > fingers and hope, Stu :-) The previous test2 binary didn't > work due to a *dumb* bug on my part... :-( > "Yes oh Yes!!!!!!!!!" OK, that's corny, but it really _is_ what I said! Anyways. I get: "Found a PMAX/MIPS compatible boot PROM!" then "Linux/MIPS DECStation Boot" "Copyright (C) Paul M. Antoine 1995, 1996, 1997 and other, 1994, 1995, 1996, 1997" "Linux version 2.1.14 (paul@slither) (gcc version 2.6.3) #277 Sun May 11 23:06:47" and then it crashes. BUT ... this is obviously _great_ progress, from where I stand! Paul, did you say you'd have a source tree available soon? I think it's time for me to put that MIPS book to work! Thanks much Paul ... I know it's just a first step, but I can see some hope for this beast now! If you're ever in this part of the US (Upstate NY) dinner and drinks are on me! Stu From tschroed@cheetah.wsc.edu Mon May 12 09:01:57 1997 Received-Date: Mon, 12 May 1997 09:01:57 +0200 (MET DST) Received: from cheetah.wsc.edu by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA18643; Mon, 12 May 1997 08:58:32 +0200 (MET) Received: by cheetah.wsc.edu (5.65/DEC-Ultrix/4.4) id AA04263; Mon, 12 May 1997 02:01:35 -0500 Other-Stuff: localhost tschroed@cheetah.wsc.edu Trevor Schroeder tschroed Date: Mon, 12 May 1997 02:01:34 -0500 (CDT) From: Trevor Schroeder To: Linux/MIPS list Subject: Latest DECstation kernel Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 505 Lines: 10 Paul: Since I've got a DS5000/25, you know pretty much where it's at. But I just thought I'd say looking good so far (other than the TLB exceptions ;) ____________________________________________________________ "One unerring mark of the love of truth is not entertaining any propositions with greater assurance than the proofs it is built upon will warrant" -- John Locke, 1690 Trevor Schroeder tschroed@cheetah.wsc.edu ------------------------------------------------------------ From hkoerfg1@ford.com Mon May 12 10:23:01 1997 Received-Date: Mon, 12 May 1997 10:23:00 +0200 (MET DST) Received: from internet-mail2.ford.com by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA20944; Mon, 12 May 1997 10:19:33 +0200 (MET) Received: by internet-mail2.ford.com id AA02110 (InterLock SMTP Gateway 3.0 for linux-mips@fnet.fr); Mon, 12 May 1997 04:22:51 -0400 Message-Id: <199705120822.AA02110@internet-mail2.ford.com> Received: by internet-mail2.ford.com (Protected-side Proxy Mail Agent-1); Mon, 12 May 1997 04:22:51 -0400 Comments: Authenticated sender is From: "Harald Koerfgen" Organization: Ford Motor Company To: linux-mips@guadalquivir.fnet.fr Date: Mon, 12 May 1997 10:22:10 +0100 Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT Subject: Re: New DECStation kernel binary to try... Reply-To: hkoerfg1@ford.com Priority: normal In-Reply-To: <199705120145.BAA28956@suede.sw.oz.au> X-Mailer: Pegasus Mail for Windows (v2.53/R1) Content-Length: 855 Lines: 44 Hi all, Paul wrote: > I have uploaded another test kernel binary for DECStation people to > try: > > ftp://ftp.fnet.fr/linux-mips/ftp.softway.com.au/dec_vmlinux.2.1.14 > .test3.gz [snip] > Please try it and report your findings on the mailing list. > > Regards, > Paul This is what I got on a 5000/133: Found a REX compatible boot PROM! Linux/MIPS DECStation Boot Copyright (C) Paul M. Antoine 1995, 1996, 1997 and others, 1994, 1995, 1996, 1997 Linux version 2.1.14 (paul@slither) (gcc version 2.6.3) #277 Sun May 11 23:06:45 EST 1997 Found memory size: 32768kB I-Cache size: 64kB D-Cache size: 128kB System has firmware type: TCF0 This DECStation is a: DS5000/1x0 Eeeek! Don't know how to set one of these up! ?HLT: 0X0000000a#-1610600568 ^C >> If you need a hand at setting 5000/1xxs up, just give me the source :-). Regards, Harald From lahaye@europe.far.cea.fr Mon May 12 11:36:05 1997 Received-Date: Mon, 12 May 1997 11:36:05 +0200 (MET DST) Received: from nenuphar.saclay.cea.fr by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA23049; Mon, 12 May 1997 11:32:23 +0200 (MET) Received: from suisse.far.cea.fr by nenuphar.saclay.cea.fr (8.8.5/ CEANET-ROUTER-3.0) with SMTP id LAA06683 for ; Mon, 12 May 1997 11:35:38 +0200 (MET DST) Received: from europe.far.cea.fr by suisse.far.cea.fr (5.x/CEANET.2.0.1) id AA00561; Mon, 12 May 1997 11:34:44 +0200 Received: by europe.far.cea.fr (950413.SGI.8.6.12/PROTO-CEANET-3.0) id JAA06146 ; Mon, 12 May 1997 09:32:04 GMT Date: Mon, 12 May 1997 09:32:04 GMT Message-Id: <199705120932.JAA06146@europe.far.cea.fr> From: Lahaye Olivier To: linux-mips@guadalquivir.fnet.fr Subject: Lastest news? 10-Jun-96? Help Reply-To: lahaye@suisse.far.cea.fr X-Face: "kR8$DK9g02`&{R`Pc)>D?WVW5zc'c~Oy}i^22&\1&gEcL\<7V%7k-0Jr:U[L8egEgkkM>t T1L@%dWEa-pVV\EKSX`wJI$x|:Wum'4tE*+rV~\{ok[s5R0HGxB>P`M7,Xg s) Received: by suede.sw.oz.au id LAA00614; Mon, 12 May 1997 11:24:28 GMT (SMI-8.6/1.34) (from paul for linux-mips@fnet.fr) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705121124.LAA00614@suede.sw.oz.au> Subject: Re: New DECStation kernel binary to try... To: linux-mips@guadalquivir.fnet.fr Date: Mon, 12 May 1997 21:24:27 +1000 (EST) In-Reply-To: <199705120822.AA02110@internet-mail2.ford.com> from "Harald Koerfgen" at May 12, 97 10:22:10 am Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 1034 Lines: 31 Harald wrote: > This is what I got on a 5000/133: > This DECStation is a: DS5000/1x0 > Eeeek! Don't know how to set one of these up! > ?HLT: 0X0000000a#-1610600568 > ^C > >> > > If you need a hand at setting 5000/1xxs up, just give me the source :-). I'll do just that once I've got the tlb load exception happy. I'd also like to come up with a scheme that actually manages the different I/O (and hence interrupt) structures neatly, though I concede that may be impossible thanks to DEC engineering prowess :-) Any DS2100/3100 owners tried the kernel yet??? Regards, Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From paul@suede.sw.oz.au Mon May 12 13:29:13 1997 Received-Date: Mon, 12 May 1997 13:29:12 +0200 (MET DST) Received: from staff.cs.su.OZ.AU by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA26263; Mon, 12 May 1997 13:25:31 +0200 (MET) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id LAA03725; Mon, 12 May 1997 11:28:27 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id LAA00630; Mon, 12 May 1997 11:28:26 GMT (SMI-8.6/1.34) (from paul for ) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705121128.LAA00630@suede.sw.oz.au> Subject: Re: Lastest news? 10-Jun-96? Help To: linux-mips@guadalquivir.fnet.fr Date: Mon, 12 May 1997 21:28:26 +1000 (EST) Cc: olivier.lahaye@ramses.fdn.org In-Reply-To: <199705120932.JAA06146@europe.far.cea.fr> from "Lahaye Olivier" at May 12, 97 09:32:04 am Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 1231 Lines: 35 Olivier wrote: > I own a RC3330 MIPS workstation with 8Megs of RAM and a mochrome > console. > > Then I had a look at the linux-mips page http://lena.fnet.fr/ > What I'd like to know is: at this time: whaich systems are supported. > Will my MIPS RC3330 be able to boot milo? Maybe not milo, but something equivalent to it. David Monro has been working on a 3230, and has made some progress in just the last week. > The latest news are date from 10-Jun-96...... > http://lena.fnet.fr/linux-mips-news.html Hmmm... yes, the web pages remain under-maintained (we have made progress since then!!) Sounds like you need a web page, David. > Sorry for posting this message here, but I don't know where to find > this info. No problem - we're here to answer questions. Regards, Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From frist@lehr.chem.tu-berlin.de Mon May 12 13:47:29 1997 Received-Date: Mon, 12 May 1997 13:47:29 +0200 (MET DST) Received: from lehr.chem.TU-Berlin.DE by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA26779; Mon, 12 May 1997 13:44:03 +0200 (MET) Received: (from frist@localhost) by lehr.chem.tu-berlin.de (8.7.4/8.7.3) id NAA02857; Mon, 12 May 1997 13:47:21 +0200 Date: Mon, 12 May 1997 13:47:20 +0200 (MET DST) From: Frieder Streffer To: linux-mips@guadalquivir.fnet.fr Subject: Re: New DECStation kernel binary to try... In-Reply-To: <199705121124.LAA00614@suede.sw.oz.au> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 213 Lines: 15 On Mon, 12 May 1997, Paul Antoine wrote: > > Any DS2100/3100 owners tried the kernel yet??? > > Regards, > Paul Hi, I'll have a look in the evening (that about 6 hours from now). regards Frieder From engel@numerik.math.uni-siegen.de Mon May 12 15:34:25 1997 Received-Date: Mon, 12 May 1997 15:34:24 +0200 (MET DST) Received: from bernoulli.numerik.math.uni-siegen.de ([141.99.158.230]) by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA29967; Mon, 12 May 1997 15:30:56 +0200 (MET) Received: from hilbert.numerik.math.uni-siegen.de (hilbert.numerik.math.uni-siegen.de [141.99.112.8]) by bernoulli.numerik.math.uni-siegen.de (Mailhost) with SMTP id PAA02082 for ; Mon, 12 May 1997 15:27:39 +0200 (MET DST) From: Michael Engel Message-Id: <9705121334.AA26124@hilbert.numerik.math.uni-siegen.de> Received: by hilbert.numerik.math.uni-siegen.de id AA26124; Mon, 12 May 97 15:34:26 +0200 Subject: Re: New DECStation kernel binary to try... To: linux-mips@guadalquivir.fnet.fr Date: Mon, 12 May 1997 15:34:25 +0200 (MET DST) In-Reply-To: <199705121124.LAA00614@suede.sw.oz.au> from "Paul Antoine" at May 12, 97 09:24:27 pm X-Mailer: ELM [version 2.4 PL23] Content-Type: text Content-Length: 356 Lines: 22 Hi, > Any DS2100/3100 owners tried the kernel yet??? Just hooked a 3100 up again here ... The kernel crashes immediately after printing Linux version 2.1.14 ... EST 1997 with Excptn: Excptn pc: 0x80087754 Creg: 0x1c Sreg: 0x30080000 Sp: 0xa0ffbfa8 ... followed by a stack dump ... regards, Michael From paul@suede.sw.oz.au Tue May 13 09:18:33 1997 Received-Date: Tue, 13 May 1997 09:18:32 +0200 (MET DST) Received: from staff.cs.su.OZ.AU ([129.78.8.1]) by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA04300; Tue, 13 May 1997 09:15:04 +0200 (MET) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id HAA14095; Tue, 13 May 1997 07:17:53 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id HAA13433; Tue, 13 May 1997 07:17:52 GMT (SMI-8.6/1.34) (from paul for linux-mips@fnet.fr) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705130717.HAA13433@suede.sw.oz.au> Subject: Re: New DECStation kernel binary to try... To: linux-mips@guadalquivir.fnet.fr Date: Tue, 13 May 1997 17:17:51 +1000 (EST) In-Reply-To: <9705121334.AA26124@hilbert.numerik.math.uni-siegen.de> from "Michael Engel" at May 12, 97 03:34:25 pm Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 939 Lines: 31 Hi, > > Any DS2100/3100 owners tried the kernel yet??? > > Just hooked a 3100 up again here ... > The kernel crashes immediately after printing > > Linux version 2.1.14 ... EST 1997 > > with > > Excptn: > Excptn pc: 0x80087754 > Creg: 0x1c > Sreg: 0x30080000 > Sp: 0xa0ffbfa8 > > ... followed by a stack dump ... Seems like there may be another 'dumb' bug... as it should be getting the system ID at that point. I'll have a look. Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From sten.eriksson@udac.se Tue May 13 15:40:27 1997 Received-Date: Tue, 13 May 1997 15:40:27 +0200 (MET DST) Received: from solix.UDAC.SE by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA16093; Tue, 13 May 1997 15:36:53 +0200 (MET) Received: from caesar.udac.se (caesar.udac.se [193.44.79.10]) by solix.udac.se (8.8.5/8.7.3) with SMTP id PAA06819 for ; Tue, 13 May 1997 15:40:15 +0200 Received: from pc-77-22.UDAC.SE (pc-77-22.UDAC.SE [193.44.77.22]) by caesar.udac.se (8.7.3/NetworkC-1) with SMTP id PAA130315 for ; Tue, 13 May 1997 15:39:55 +0200 From: sten.eriksson@udac.se (Sten Eriksson) To: linux-mips@guadalquivir.fnet.fr Date: Tue, 13 May 1997 13:39:55 GMT Organization: UDAC / Datorhotellet Message-Id: <33796f0f.4304194@mail.networkc.net> X-Mailer: Forte Agent 1.0/32.390 Content-Length: 471 Lines: 8 Subscribe Sten.Eriksson@UDAC.SE (Sten Eriksson) ------------------------------------------------------------------------- Sten Eriksson Email: sten.eriksson@udac.se UDAC Telefon: 018 - 18 78 20 Box 174 GSM: 070 - 542 47 03 751 04 Uppsala Fax: 018 - 18 77 52 ------------------------------------------------------------------------- From linux@informatik.uni-koblenz.de Tue May 13 21:11:43 1997 Received-Date: Tue, 13 May 1997 21:11:42 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA24246; Tue, 13 May 1997 21:08:05 +0200 (MET) Received: from thoma (linux@thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.8.5/8.6.9) with SMTP id VAA21865 for ; Tue, 13 May 1997 21:11:21 +0200 (MEST) From: Systemkennung Linux Message-Id: <199705131911.VAA21865@informatik.uni-koblenz.de> Received: by thoma (SMI-8.6/KO-2.0) id VAA08146; Tue, 13 May 1997 21:11:17 +0200 Subject: Re: vmalloc help?? To: linux-mips@guadalquivir.fnet.fr Date: Tue, 13 May 1997 21:11:14 +0200 (MET DST) In-Reply-To: <199705101358.NAA19648@suede.sw.oz.au> from "Paul Antoine" at May 10, 97 11:58:45 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1445 Lines: 38 Hi, > I've discovered the latest point at which the DECStation kernel is failing: > namely in accessing KSEG2 after having vmalloc'd a hash table for the buffers. > > Can anyone with good knowledge of this help? It seems that vmalloc is > allocating an address of 0xc0000000 for the table, and that any access of > this hangs... Well, buffer_init() where vmalloc() is called for the first time is a first check of your R3000 specific mm stuff, especially the TLB exception handler. That first vmalloc on a Linux/MIPS system will return a pointer to the address 0xc0000000 aka KSEG2. The chunks returned by vmalloc are separated by at least one unmapped 4kb page. This helps to catch out off bounds accesses. With 99.9% probability your problem is in the TLB exception handler. Try some code like the following: x = vmalloc(PAGE_SIZE); *(unsigned int *)x = 42; and add code to the TLB exception handler (at KSEG0) to print out the content of the TLB's entries. What non-wired TLB entries do you have? Are they correct? .37 does no longer use the old scheme were a processes' page tables were also mapped at the address TLBMAP (0xe4000000), so you will have to rewrite the TLB exception handlers for the R3000. The good news is the current stuff makes a *lot* less of headache, specially on CPUs with virtual indexed caches like the R4k. Debugging these handlers is nasty and can easily lead you to undocumented MIPS ... Ralf From elias@scruznet.com Wed May 14 08:59:50 1997 Received-Date: Wed, 14 May 1997 08:59:48 +0200 (MET DST) Received: from [165.227.110.86] (elias.sj.scruznet.com [165.227.110.86]) by scruz.net (8.7.3/1.34) with SMTP id XAA10543 for ; Tue, 13 May 1997 23:59:40 -0700 (PDT) Message-Id: <199705140659.XAA10543@scruz.net> Subject: execution failing Date: Wed, 14 May 97 00:08:04 +0100 x-mailer: Claris Emailer 2.0, March 15, 1997 From: Elias Kesh To: "Mips List" Mime-Version: 1.0 Content-Type: text/plain; charset="US-ASCII" Content-Length: 557 Lines: 16 Question. I am trying to load and elf file (sh, the first file the kernel tries to execute) and it's failing. It seems as if it's trying to load it at 0x400000 which is a physical address. The function do_load_elf_binary in fs/binfmt_elf.c returns no error, but when it tries to pad zeros or execute it starting at 0x400000 logical it dies. Is there some mapping that has to go on that I have failed to set up ? I am using version 2.0.21 of the kernel on the Algorithmics P4032 board with a 5230 processor from QED. Thanks, Elias Elias@kesh.com From paul@suede.sw.oz.au Wed May 14 03:44:28 1997 Received-Date: Wed, 14 May 1997 03:44:27 +0200 (MET DST) Received: from staff.cs.su.OZ.AU ([129.78.8.1]) by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA03771; Wed, 14 May 1997 03:39:26 +0200 (MET) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id BAA28833; Wed, 14 May 1997 01:41:50 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id BAA21519; Wed, 14 May 1997 01:41:49 GMT (SMI-8.6/1.34) (from paul for linux-mips@fnet.fr) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705140141.BAA21519@suede.sw.oz.au> Subject: Re: vmalloc help (aka R3000 TLB help) To: linux-mips@guadalquivir.fnet.fr Date: Wed, 14 May 1997 11:41:48 +1000 (EST) In-Reply-To: <199705131911.VAA21865@informatik.uni-koblenz.de> from "Systemkennung Linux" at May 13, 97 09:11:14 pm Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 1768 Lines: 54 Ralf wrote: > Well, buffer_init() where vmalloc() is called for the first time is > a first check of your R3000 specific mm stuff, especially the TLB > exception handler. > > With 99.9% probability your problem is in the TLB exception handler. Yep, I've managed to narrow it down to that... > Try some code like the following: > > x = vmalloc(PAGE_SIZE); > *(unsigned int *)x = 42; This is exactly the sort of code in vmalloc that is failing. > and add code to the TLB exception handler (at KSEG0) to print out the > content of the TLB's entries. Sure. > What non-wired TLB entries do you have? Are they correct? Ummm... there are no wired entries. I've no idea what non-wired entries I have, nor whether they are correct. > .37 does no longer use the old scheme were a processes' page tables > were also mapped at the address TLBMAP (0xe4000000), so you will > have to rewrite the TLB exception handlers for the R3000. I see... > The good news is the current stuff makes a *lot* less of headache, > specially on CPUs with virtual indexed caches like the R4k. ...so the code for the R3K will be very different I assume. > Debugging these handlers is nasty and can easily lead you to > undocumented MIPS ... It is an almighty pain, but in other ways quite satisfying when it works... :-) Regards, Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From agrier@metro.grumblesmurf.net Wed May 14 05:14:15 1997 Received-Date: Wed, 14 May 1997 05:14:14 +0200 (MET DST) Received: from localhost by metro.grumblesmurf.net with smtp (/\o=/\ Smail3.1.29.1 #6) id m0wRUTQ-00000tC; Tue, 13 May 97 20:11 PDT Date: Tue, 13 May 1997 20:11:27 -0700 (PDT) From: "Aaron J. Grier" Reply-To: "Aaron J. Grier" To: linux-mips@fnet.fr Subject: Re: New DECStation kernel binary to try... In-Reply-To: <199705120343.XAA14358@shell10.vivanet.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 572 Lines: 22 On Sun, 11 May 1997, Stu Allen wrote: > "Linux version 2.1.14 (paul@slither) (gcc version 2.6.3) #277 Sun May 11 23:06:47 EST 1997" > > and then it crashes. Same here on my 3100. Excptn: Excptn pc: 0x80087754 Creg: 0x41c Sreg: 0x30080000 Sp: 0xa0ffbfa8 [stack dump snipped...] This is exciting! (: ---- The Finn / VLA | Aaron J. Grier | agrier@metro.grumblesmurf.net, DECBEE650DEE3A0C 1ED7F54D9E023CC4 | agrier@reed.edu, ...reed!vla!agrier "Not your ordinary poofy goof." | 0x0663D1A9 at a server near you. From linux@informatik.uni-koblenz.de Wed May 14 18:13:05 1997 Received-Date: Wed, 14 May 1997 18:12:39 +0200 (MET DST) Received: from thoma (linux@thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.8.5/8.6.9) with SMTP id SAA03325 for ; Wed, 14 May 1997 18:12:06 +0200 (MEST) From: Systemkennung Linux Message-Id: <199705141612.SAA03325@informatik.uni-koblenz.de> Received: by thoma (SMI-8.6/KO-2.0) id SAA11608; Wed, 14 May 1997 18:12:03 +0200 Subject: Re: execution failing To: linux-mips@fnet.fr Date: Wed, 14 May 1997 18:12:02 +0200 (MET DST) In-Reply-To: <199705140659.XAA10543@scruz.net> from "Elias Kesh" at May 14, 97 00:08:04 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1017 Lines: 23 Hi Elias, > Question. I am trying to load and elf file (sh, the first file the kernel > tries to execute) and it's failing. It seems as if it's trying to load it > at 0x400000 which is a physical address. The function do_load_elf_binary > in fs/binfmt_elf.c returns no error, but when it tries to pad zeros or > execute it starting at 0x400000 logical it dies. Is there some mapping > that has to go on that I have failed to set up ? I am using version > 2.0.21 of the kernel on the Algorithmics P4032 board with a 5230 > processor from QED. That function tries to access the userspace using clear_user(). Doing so it will also load TLB entries into the TLB. It might be that the pipeline of the 5230 is a bit different such that this fails. What exactly does happen? If you can catch the error - what's the TLB's contents after your problem happend? For testing purposes it is probably helpfull if you use a very small program like the assembler written hello world program available on FNet. Ralf From ryan@ravage.labs.gmu.edu Wed May 14 23:23:17 1997 Received-Date: Wed, 14 May 1997 23:23:16 +0200 (MET DST) Received: from ravage.labs.gmu.edu ([129.174.166.55]) by Utopia.EUnet.fr (5.65c8d/EUNET-4.2.3) via EUnet-France id AA07429; Wed, 14 May 1997 23:19:48 +0200 (MET) Received: from localhost (ryan@localhost) by ravage.labs.gmu.edu (8.8.3/8.8.3) with SMTP id RAA03962 for ; Wed, 14 May 1997 17:24:45 -0400 Date: Wed, 14 May 1997 17:24:45 -0400 (EDT) From: Ryan Rafferty To: linux-mips@guadalquivir.fnet.fr Subject: Shablamm! Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 362 Lines: 11 I am thinking about buying a Shablamm! Nitro VLB card. This is quite an unusual animal--a MIPS R4600 with 32 megs of (what I've been told) cache, and designed to run NT. Does anyone know much about this card? I saw mention of it on the homepage. I'm probably gonna buy it anyway, since it's such a unique and wierd thing =). Is mise le meas, Ryan Rafferty From linux@informatik.uni-koblenz.de Thu May 15 12:42:44 1997 Received-Date: Thu, 15 May 1997 12:42:42 +0200 (MET DST) Received: from thoma (linux@thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.8.5/8.6.9) with SMTP id BAA25701 for ; Thu, 15 May 1997 01:15:16 +0200 (MEST) From: Systemkennung Linux Message-Id: <199705142315.BAA25701@informatik.uni-koblenz.de> Received: by thoma (SMI-8.6/KO-2.0) id BAA12894; Thu, 15 May 1997 01:15:12 +0200 Subject: Re: Shablamm! To: linux-mips@fnet.fr Date: Thu, 15 May 1997 01:15:11 +0200 (MET DST) In-Reply-To: from "Ryan Rafferty" at May 14, 97 05:24:45 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 751 Lines: 21 Hi, > I am thinking about buying a Shablamm! Nitro VLB card. This is quite an > unusual animal--a MIPS R4600 with 32 megs of (what I've been told) cache, > and designed to run NT. > > Does anyone know much about this card? I saw mention of it on the > homepage. I'm probably gonna buy it anyway, since it's such a unique and > wierd thing =). We don't know too many technical details about the card. The card is only on the list on potential targets because the German retailer (IDT??? Can't remember for shure ...) once offered one to US for porting Linux to. Probably the card is an relatively easy target for porting NT to as it only uses it's host PC's peripherals. It's just that VLB machine have become rare in the meantime ... Ralf From elias@scruznet.com Thu May 15 12:35:24 1997 Received-Date: Thu, 15 May 1997 12:35:14 +0200 (MET DST) Received: from [165.227.110.86] (elias.sj.scruznet.com [165.227.110.86]) by scruz.net (8.7.3/1.34) with SMTP id BAA24076 for ; Thu, 15 May 1997 01:12:12 -0700 (PDT) Message-Id: <199705150812.BAA24076@scruz.net> Subject: Re: execution failing Date: Thu, 15 May 97 01:20:40 +0100 x-mailer: Claris Emailer 2.0, March 15, 1997 From: Elias Kesh To: Mime-Version: 1.0 Content-Type: text/plain; charset="US-ASCII" Content-Length: 1932 Lines: 47 Here is some more information. If we try and execute pad zero the machine locks up and we are not sure where yet. If we comment that out we then get a page fault when it tries to execute the code at 0x400000. The 5230 has 96 tlbs set up as two pairs of 48 (odd/even). Currently I have the tags set up for 32 tbl entries since we were using a 4300 before this. I have not verified that the tbls are in fact working but I imaging that that are in order to get this far. We turned on the code to dump the tlbs in the exception routine but we must not be getting there since they never get displayed. I'm not expert on virtual menory management so I am now reading up on how it is implemented on linux. We did try and write a very small binary which turns on the LEDs on the board and it behave exactly the same way as trying to execute sh. Any comments would be greatly appreciated. Thank, Elias Hi Elias, > Question. I am trying to load and elf file (sh, the first file the kernel > tries to execute) and it's failing. It seems as if it's trying to load it > at 0x400000 which is a physical address. The function do_load_elf_binary > in fs/binfmt_elf.c returns no error, but when it tries to pad zeros or > execute it starting at 0x400000 logical it dies. Is there some mapping > that has to go on that I have failed to set up ? I am using version > 2.0.21 of the kernel on the Algorithmics P4032 board with a 5230 > processor from QED. That function tries to access the userspace using clear_user(). Doing so it will also load TLB entries into the TLB. It might be that the pipeline of the 5230 is a bit different such that this fails. What exactly does happen? If you can catch the error - what's the TLB's contents after your problem happend? For testing purposes it is probably helpfull if you use a very small program like the assembler written hello world program available on FNet. Ralf From dom@algor.co.uk Thu May 15 12:36:31 1997 Received-Date: Thu, 15 May 1997 12:36:29 +0200 (MET DST) Received: from gladsmuir.algor.co.uk (dom@gladsmuir.algor.co.uk [193.117.190.129]) by moorgate.algor.co.uk (8.7.4/8.7.3) with ESMTP id IAA11930; Thu, 15 May 1997 08:11:28 +0100 (BST) From: Dom Sweetman Date: Thu, 15 May 1997 08:11:20 +0100 (BST) Message-Id: <191.199705150711@gladsmuir.algor.co.uk> To: linux-mips@fnet.fr Cc: linux-mips@guadalquivir.fnet.fr Subject: Re: Shablamm! In-Reply-To: References: Content-Length: 1658 Lines: 40 Ryan Rafferty (ryan@ravage.labs.gmu.edu) writes: > I am thinking about buying a Shablamm! Nitro VLB card. This is > quite an unusual animal--a MIPS R4600 with 32 megs of (what I've > been told) cache, and designed to run NT. Yes, it was one of the odder spin-offs of the Great MIPS/NT Hype. T talked to the guys who made it, but only ever saw a non-functional prototype. The memory uses unusual DRAMs (from RAMtron? something like that) which were described as "cache DRAM"; when you read a location the page around it (4Kbytes or so) is kept in an auxiliary static memory. All DRAMs do this, because it's by copying the data from the main array into the page buffer and back again that they get refreshed... But the cache DRAMs had the feature that the last-accessed page of data stayed in the static memory and would be returned quickly if the next access was to the same page. On other DRAMs (fast page, static column or EDO) you can do something similar but the memory controller has to make it happen, and there aren't enough control signals... More useful for the Shablamm!! (two exclamation marks, I seem to remember) was that the DRAM was fast at doing ordinary things. Performance should be pretty respectable. It plugs into a PC. I hope that's an ISA socket, because at that time it might have been VL bus. > Does anyone know much about this card? I saw mention of it on the > homepage. I'm probably gonna buy it anyway, since it's such a > unique and wierd thing =). It deserves a place in the museum... But do try to get hold of an original marketing flyer too, it's charming. Dominic Sweetman Algorithmics Ltd dom@algor.co.uk From rji@bristol.st.com Thu May 15 12:32:00 1997 Received-Date: Thu, 15 May 1997 12:31:55 +0200 (MET DST) Received: by hawkweed id LAA13936; Thu, 15 May 1997 11:31:42 +0100 (BST) From: "Richard Ingram" Message-Id: <9705151131.ZM20748@bristol.st.com> Date: Thu, 15 May 1997 11:31:33 +0100 In-Reply-To: Ryan Rafferty "Shablamm!" (May 14, 5:24pm) References: X-Mailer: Z-Mail (3.2.1 10apr95) To: ryan@ravage.labs.gmu.edu, linux-mips@fnet.fr Subject: Re: Shablamm! Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Length: 815 Lines: 33 Hi Ryan, Are there anymore of these ? Is it like the auction for the Magnums are there a number ? Its a Vesa local bus board - I remember hearing about them a while back. Hey hows your Magnum going ? Did you get a graphics card from somewhere ? Cheers, Rich. Magnum and Millenium owner ;-) (neither going yet though!). On May 14, 5:24pm, Ryan Rafferty wrote: > Subject: Shablamm! > > I am thinking about buying a Shablamm! Nitro VLB card. This is quite an > unusual animal--a MIPS R4600 with 32 megs of (what I've been told) cache, > and designed to run NT. > > Does anyone know much about this card? I saw mention of it on the > homepage. I'm probably gonna buy it anyway, since it's such a unique and > wierd thing =). > > Is mise le meas, > Ryan Rafferty > >-- End of excerpt from Ryan Rafferty -- From imp@village.org Thu May 15 16:43:16 1997 Received-Date: Thu, 15 May 1997 16:43:05 +0200 (MET DST) Received: from rover.village.org [127.0.0.1] by rover.village.org with esmtp (Exim 1.60 #1) id 0wS1k2-0000mu-00; Thu, 15 May 1997 08:42:50 -0600 To: linux-mips@fnet.fr Subject: Re: Shablamm! Cc: linux-mips@guadalquivir.fnet.fr In-reply-to: Your message of "Thu, 15 May 1997 08:11:20 BST." <191.199705150711@gladsmuir.algor.co.uk> References: <191.199705150711@gladsmuir.algor.co.uk> Date: Thu, 15 May 1997 08:42:49 -0600 From: Warner Losh Message-Id: Content-Length: 437 Lines: 10 In message <191.199705150711@gladsmuir.algor.co.uk> Dom Sweetman writes: : More useful for the Shablamm!! (two exclamation marks, I seem to : remember) was that the DRAM was fast at doing ordinary things. : Performance should be pretty respectable. Memory bandwidth and performance characteristics really matter a lot for MIPS processors. The better you can make your memory subsystem, generally the faster the system will be. Warner From imp@village.org Thu May 15 16:44:35 1997 Received-Date: Thu, 15 May 1997 16:44:21 +0200 (MET DST) Received: from rover.village.org [127.0.0.1] by rover.village.org with esmtp (Exim 1.60 #1) id 0wS1lR-0000nH-00; Thu, 15 May 1997 08:44:17 -0600 To: linux-mips@fnet.fr Subject: Re: Shablamm! In-reply-to: Your message of "Thu, 15 May 1997 01:15:11 +0200." <199705142315.BAA25701@informatik.uni-koblenz.de> References: <199705142315.BAA25701@informatik.uni-koblenz.de> Date: Thu, 15 May 1997 08:44:17 -0600 From: Warner Losh Message-Id: Content-Length: 326 Lines: 8 In message <199705142315.BAA25701@informatik.uni-koblenz.de> Systemkennung Linux writes: : It's just that VLB machine have become rare in the meantime ... Not too bad here in the states. You can fairly easily get a $75 motherboard with 486 generation of CPU that has VLB. Or you can pay the same price and get PCI. Warner From khp@dolphinics.no Thu May 15 17:02:08 1997 Received-Date: Thu, 15 May 1997 17:02:05 +0200 (MET DST) Received: by scimitar.dolphinics.no (4.1/SMI-4.1) id AA10481; Thu, 15 May 97 17:04:13 +0200 From: khp@dolphinics.no (Kai Harrekilde-Petersen) Message-Id: <9705151504.AA10481@scimitar.dolphinics.no> Subject: Re: Shablamm! In-Reply-To: from Warner Losh at "May 15, 97 08:44:17 am" To: linux-mips@fnet.fr Date: Thu, 15 May 1997 17:04:12 +0200 (MET DST) X-Mailer: ELM [version 2.4ME+ PL27 (25)] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 854 Lines: 19 Warner Losh writes: > In message <199705142315.BAA25701@informatik.uni-koblenz.de> > Systemkennung Linux writes: > : It's just that VLB machine have become rare in the meantime ... > > Not too bad here in the states. You can fairly easily get a $75 > motherboard with 486 generation of CPU that has VLB. Or you can pay > the same price and get PCI. Speaking of plug-in MIPS cards: Getting hold of one of Galileo's demo-boards with a R4600 + their GT64012(?) PCI bridge would be a neat thing. I remember seeing an errata-list (under NDA), but don't recall anything disastrous (but then I didn't look hard). Kai -- Kai Harrekilde-Petersen #include http://www.dolphinics.no/~khp/ Linux: the choice of a GNU generation "Argue for your limitations, and sure enough - they're yours" --Richard Bach. From davidm@fuzzbox.psrg.cs.usyd.edu.au Thu May 15 18:01:02 1997 Received-Date: Thu, 15 May 1997 18:00:45 +0200 (MET DST) Received: from fuzzbox.psrg.cs.usyd.edu.au (davidm@fuzzbox.psrg.cs.usyd.edu.au [129.78.97.193]) by fuzzbox.psrg.cs.usyd.edu.au (8.8.4/8.8.4) with SMTP id BAA12444 for ; Fri, 16 May 1997 01:58:13 +1000 Date: Fri, 16 May 1997 01:58:13 +1000 (EST) From: David Monro Sender: davidm@fuzzbox.psrg.cs.usyd.edu.au Reply-To: David Monro Subject: odd problems To: linux-mips@fnet.fr Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-Length: 2051 Lines: 40 I'm having trouble with the magnum port. Following the example of the DECstation port (I'm working with Paul's 2.1.1 tree) I decided to try moving the exception handlers down to 0x80000000 (since the kernel appears to only get booted if I link it at 0x81000000). Mind you I'm not entirely sure why this is done - Paul? In any case, I manage to move the code for exception 0 down using the following code fragment: la t0,except_vec0 # begining of exception code la t1,except_vec1 # end of exception code la t2,0x80000000 # where the code should live lw t3,(t0) # get first word 1: sw t3,(t2) # put it where it should go addiu t0,4 # increment both pointers addiu t2,4 lw t3,(t0) # will be in the delay slot bne t0,t1,1b Which works fine. However, the following almost identical code causes the machine to lock up (not generate an exception, just go silent). la t0,except_vec3 # begining of general exception la t1,end_except # end of general exception code la t2,0x80000080 # where the code should live 1: sw t3,(t2) # put it where it should go addiu t0,4 # increment both pointers addiu t2,4 lw t3,(t0) # will be in the delay slot bne t0,t1,1b The lock up occurs as soon as the sw t3,(t2) is executed (well at least if I put a 'j kernel_entry' just before the sw I get an exception in later code, but if I put it after it I never get anything more out of it). Given that I am about 8km away from the machine, the only way I have of unwedging it is to powercycle it remotely, which I don't like doing very often. Wish I could get it to reset from the serial line. Any ideas? David From ralf@Julia.DE Thu May 15 18:17:24 1997 Received-Date: Thu, 15 May 1997 18:17:02 +0200 (MET DST) Received: from kernel.panic.julia.de (kernel.panic.julia.de [194.221.49.153]) by alles.intern.julia.de (8.8.5/8.8.5) with ESMTP id RAA00497 for ; Thu, 15 May 1997 17:15:58 +0200 From: Ralf Baechle Received: (from ralf@localhost) by kernel.panic.julia.de (8.8.4/8.8.4) id SAA06883 for linux-mips@fnet.fr; Thu, 15 May 1997 18:11:46 +0200 Message-Id: <199705151611.SAA06883@kernel.panic.julia.de> Subject: Re: odd problems To: linux-mips@fnet.fr Date: Thu, 15 May 1997 18:11:46 +0200 (MET DST) In-Reply-To: from "David Monro" at May 16, 97 01:58:13 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1519 Lines: 31 > Which works fine. However, the following almost identical code causes the > machine to lock up (not generate an exception, just go silent). > > la t0,except_vec3 # begining of general exception > la t1,end_except # end of general exception code > la t2,0x80000080 # where the code should live > 1: sw t3,(t2) # put it where it should go During the first loop iteration t3 in uninitialized ... > addiu t0,4 # increment both pointers > addiu t2,4 > lw t3,(t0) # will be in the delay slot > bne t0,t1,1b > > The lock up occurs as soon as the sw t3,(t2) is executed (well at least if I > put a 'j kernel_entry' just before the sw I get an exception in later code, but > if I put it after it I never get anything more out of it). Given that I am > about 8km away from the machine, the only way I have of unwedging it is to > powercycle it remotely, which I don't like doing very often. Wish I could get > it to reset from the serial line. With a messed up exception handler you will never catch exceptions ... I assume you have disabled interrupts and ensured that the BEV flag in c0_status is cleared? Maybe somebody from the hardware fraction can give you some hints about how to reset the machine from remote? (One of the nice features of the RM200 - you can power it up/down and reset from the serial interface.) Ralf From mh@cs.tu-berlin.de Thu May 15 23:13:14 1997 Received-Date: Thu, 15 May 1997 23:13:12 +0200 (MET DST) Received: from nicky.cs.tu-berlin.de (mh@nicky.cs.tu-berlin.de [130.149.17.43]) by mail.cs.tu-berlin.de (8.8.5/8.8.5) with ESMTP id XAA18151 for ; Thu, 15 May 1997 23:13:03 +0200 (MET DST) From: Matthias Heidbrink Received: (from mh@localhost) by nicky.cs.tu-berlin.de (8.8.5/8.8.5) id XAA00595 for linux-mips@fnet.fr; Thu, 15 May 1997 23:12:59 +0200 (MET DST) Message-Id: <199705152112.XAA00595@nicky.cs.tu-berlin.de> Subject: Re: Shablamm! To: linux-mips@fnet.fr Date: Thu, 15 May 1997 23:12:43 +0200 (MET DST) In-Reply-To: <191.199705150711@gladsmuir.algor.co.uk> from "Dom Sweetman" at May 15, 97 08:11:20 am X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=US-ASCII Content-Length: 907 Lines: 24 Hi, > Yes, it was one of the odder spin-offs of the Great MIPS/NT Hype. T > talked to the guys who made it, but only ever saw a non-functional > prototype. I've seen it on the CeBit two or three years ago. They had a small corner at MIPS/SGI. There was a demonstration running that looked really impressing, but I don't know if it was real or only a kind of slideshow. I even took a data sheet with me, maybe I still have it somewhere. > > It plugs into a PC. I hope that's an ISA socket, because at that time > it might have been VL bus. I'm quite shure that it required VLB. But that's not so important, if I remember that right, it used the x86 PC only as an I/O subsystem and did everything else itself. And an old 486 with VLB should be cheap. Ciao, Matthias --- Matthias Heidbrink Helmstedter Str.20 10717 Berlin Tel. +49-30-8536361 E-Mail: mh@cs.tu-berlin.de Matthias_Heidbrink@b.maus.de From ralf@Julia.DE Thu May 15 23:21:47 1997 Received-Date: Thu, 15 May 1997 23:21:36 +0200 (MET DST) Received: from kernel.panic.julia.de (kernel.panic.julia.de [194.221.49.153]) by alles.intern.julia.de (8.8.5/8.8.5) with ESMTP id WAA04054 for ; Thu, 15 May 1997 22:20:31 +0200 From: Ralf Baechle Received: (from ralf@localhost) by kernel.panic.julia.de (8.8.4/8.8.4) id XAA07392 for linux-mips@fnet.fr; Thu, 15 May 1997 23:16:20 +0200 Message-Id: <199705152116.XAA07392@kernel.panic.julia.de> Subject: Re: Shablamm! To: linux-mips@fnet.fr Date: Thu, 15 May 1997 23:16:20 +0200 (MET DST) In-Reply-To: <199705152112.XAA00595@nicky.cs.tu-berlin.de> from "Matthias Heidbrink" at May 15, 97 11:12:43 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1251 Lines: 26 > > Hi, > > > Yes, it was one of the odder spin-offs of the Great MIPS/NT Hype. T > > talked to the guys who made it, but only ever saw a non-functional > > prototype. > > I've seen it on the CeBit two or three years ago. They had a small corner at MIPS/SGI. There was a demonstration running that looked really impressing, but I don't know if it was real or only a kind of slideshow. I even took a data sheet with me, maybe I still have it somewhere. > > > > It plugs into a PC. I hope that's an ISA socket, because at that time > > it might have been VL bus. > > I'm quite shure that it required VLB. But that's not so important, if I remember that right, it used the x86 PC only as an I/O subsystem and did everything else itself. And an old 486 with VLB should be cheap. Does anybody still remember the UltraP modules designed by SGI? Basically a modules that replaces the Pentium in it's sockets and provides two socket, one for the Pentium, one for the MIPS CPU. I wonder if these ever have been released to the market or what happend to them. Similar I found that Toshiba build a chipset that seems to be very closely related to the original MIPS Magnum 4000 chipset. I wonder if any machines based on that have been sold? Ralf From ralf@Julia.DE Thu May 15 23:24:21 1997 Received-Date: Thu, 15 May 1997 23:24:13 +0200 (MET DST) Received: from kernel.panic.julia.de (kernel.panic.julia.de [194.221.49.153]) by alles.intern.julia.de (8.8.5/8.8.5) with ESMTP id WAA04072 for ; Thu, 15 May 1997 22:23:04 +0200 From: Ralf Baechle Received: (from ralf@localhost) by kernel.panic.julia.de (8.8.4/8.8.4) id XAA07403 for linux-mips@fnet.fr; Thu, 15 May 1997 23:18:54 +0200 Message-Id: <199705152118.XAA07403@kernel.panic.julia.de> Subject: Re: Shablamm! To: linux-mips@fnet.fr Date: Thu, 15 May 1997 23:18:53 +0200 (MET DST) In-Reply-To: <191.199705150711@gladsmuir.algor.co.uk> from "Dom Sweetman" at May 15, 97 08:11:20 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 245 Lines: 9 Hi, > It deserves a place in the museum... But do try to get hold of an > original marketing flyer too, it's charming. Hmmm... as I remeber that flyer it was that dripping of marketing hype that for shure the paper isn't dry yet ... Ralf From ralf@Julia.DE Fri May 16 01:09:17 1997 Received-Date: Fri, 16 May 1997 01:09:14 +0200 (MET DST) Received: from kernel.panic.julia.de (kernel.panic.julia.de [194.221.49.153]) by alles.intern.julia.de (8.8.5/8.8.5) with ESMTP id AAA05315 for ; Fri, 16 May 1997 00:08:12 +0200 From: Ralf Baechle Received: (from ralf@localhost) by kernel.panic.julia.de (8.8.4/8.8.4) id BAA07581 for linux-mips@fnet.fr; Fri, 16 May 1997 01:04:02 +0200 Message-Id: <199705152304.BAA07581@kernel.panic.julia.de> Subject: Shamblamm!! for sale ... To: linux-mips@fnet.fr Date: Fri, 16 May 1997 01:04:02 +0200 (MET DST) MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 686 Lines: 24 Just in case somebody is interested - this is from comp.sys.mips: Newsgroups: comp.sys.mips Subject: FS: MIPS R4600, 32 Megs RAM - $200 From: MMower@Novell.com (Mathew Mower) Organization: Your Organization X-Newsreader: WinVN 0.99.9 (Beta 3) (x86 32bit) MIME-Version: 1.0 Content-Type: Text/Plain; charset=US-ASCII NNTP-Posting-Host: mmower2_pc.orem.novell.com Message-ID: <3379f6d1.0@gronk.provo.novell.com> Date: 14 May 97 17:30:57 GMT I have a Shablamm! card forsale. VL-bus host card works on a 486sx25 VL-bus system board. Card has a MIPS R4600 100Mhz, and 32 Megs of CACHE, That's CACHE RAM. Under WinNT, Mips cards becomes host CPU. $200 obo E-mail to: MMower@Novell.com From paul@suede.sw.oz.au Fri May 16 01:20:10 1997 Received-Date: Fri, 16 May 1997 01:20:07 +0200 (MET DST) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id XAA14586; Thu, 15 May 1997 23:19:33 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id XAA17512; Thu, 15 May 1997 23:19:31 GMT (SMI-8.6/1.34) (from paul for linux-mips@fnet.fr) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705152319.XAA17512@suede.sw.oz.au> Subject: Re: odd problems To: linux-mips@fnet.fr Date: Fri, 16 May 1997 09:19:30 +1000 (EST) In-Reply-To: from "David Monro" at May 16, 97 01:58:13 am Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 1899 Lines: 42 David Monro wrote: > I'm having trouble with the magnum port. Following the example of the > DECstation port (I'm working with Paul's 2.1.1 tree) I decided to try moving > the exception handlers down to 0x80000000 (since the kernel appears to only get > booted if I link it at 0x81000000). Mind you I'm not entirely sure why this is > done - Paul? Presumeably because the boot PROM's uses that space for 'stuff'... we can probably re-claim it once the console code is working independently of the boot PROMs. > Given that I am about 8km away from the machine, the only way I have > of unwedging it is to powercycle it remotely, which I don't like > doing very often. Wish I could get it to reset from the serial line. > > Any ideas? You could make an interface from the Carrier Detect line on the modem that closes the reset circuit, so that all you need do is disconnect and reconnect to reset (though this could become expensive wrt phone calls). The other way would be to have a small intermediary board with a pair of serial ports that would 'pass-through' most keystrokes to the Magnum console port, but could intercept a command on which it could close the reset circuit momentarily. A 386 running Linux is probably overkill, but would work. A 286 with a DOS program (yetch!) or any small embedded computer would do too... I have a 68000 based board that could do the trick, but there are plenty of other options I imagine. Regards, Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From romal@Informatik.Uni-Bremen.DE Fri May 16 08:33:04 1997 Received-Date: Fri, 16 May 1997 08:33:02 +0200 (MET DST) Received: by bettina.informatik.uni-bremen.de (8.8.3/3.10.94m) with SMTP from romal-2.home.informatik.uni-bremen.de [134.102.219.83] id IAA22308 for ; Fri, 16 May 1997 08:34:43 +0200 (MET DST) Received: by localhost with Microsoft MAPI; Fri, 16 May 1997 08:32:55 +0100 Message-ID: <01BC61D3.C0508F30@romal@informatik.uni-bremen.de> From: "Robert M. Albrecht" To: "'linux-mips@fnet.fr'" Subject: Your Linux-Port on a SNI RM200 Date: Fri, 16 May 1997 08:31:51 +0100 X-Mailer: Microsoft Internet E-mail/MAPI - 8.0.0.4025 Encoding: 28 TEXT Content-Length: 832 Lines: 28 Hello, I?ve got an used SNI RM200-120 :-) Sadly, the whole documentation and the key (for opening the case) is missing. It is only the box, a CD-ROM Sinix V5.42 and the keyboard. I have attached my VGA-Monitor and a PS/2-style mouse. According to the self-diagnosis this machine is equipped with 32MB RAM and a MIPS 4600. Built in s a harddisk (booting Sinix), a CD-Rom (looks like a toshiba) and a standard-cartridge-streamer (looks like tandberg).. I want to get Linux on this box and get rid of this ridiculous Sinix. I have gotten into the firmware-monitor, by guessing I tried: boot floppy()milo but the firmwares complains loading floppy()milo couldn?t determine fs type Coudln?t load floppy()milo Do you have any ideas how to boot these machine ? Do have know how to install Windows NT on this box ? cu romal From jung@ese-metz.fr Fri May 16 10:06:13 1997 Received-Date: Fri, 16 May 1997 10:06:10 +0200 (MET DST) Received: from eva.ese-metz.fr (eva.ese-metz.fr [192.70.65.55]) by esemetz.ese-metz.fr (8.7.1/8.7.2) with SMTP id KAA27526 for ; Fri, 16 May 1997 10:05:56 +0200 (MET DST) Received: by eva.ese-metz.fr with Microsoft Mail id <01BC61E0.C289CA70@eva.ese-metz.fr>; Fri, 16 May 1997 10:06:02 +0200 Message-ID: <01BC61E0.C289CA70@eva.ese-metz.fr> From: Philippe Jung To: "'linux-mips@fnet.fr'" Subject: RE: Your Linux-Port on a SNI RM200 Date: Fri, 16 May 1997 10:06:00 +0200 MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="---- =_NextPart_000_01BC61E0.C28B5110" Content-Length: 5522 Lines: 89 ------ =_NextPart_000_01BC61E0.C28B5110 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Hi, I have used a RM200 a year ago, so I hope what I say is correct : -) I think you have to change Firmware. Doing it, you will be able to use = floppy disk for booting. There is a floppy disk called sitont. In the = firmware monitor, you choose the option change firmware (I am not sure = of the name) and you insert the disk. Just another thing, if I am not mistaken, changing firmware require = settling processor in Big Endian thought it was in Little Endian for = Sinix. If you have a new RM, it is automaticly done, bt if your RM is = quite old, you'll have to change a jumper inside it. =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= Philippe JUNG (Philippe.Jung@Supelec.fr) 4 Place Edouard Branly, 57070 METZ, FRANCE http://www.ese-metz.fr/~jung/ ------ =_NextPart_000_01BC61E0.C28B5110 Content-Type: application/ms-tnef Content-Transfer-Encoding: base64 eJ8+IgIIAQaQCAAEAAAAAAABAAEAAQeQBgAIAAAA5AQAAAAAAADoAAEIgAcAGAAAAElQTS5NaWNy b3NvZnQgTWFpbC5Ob3RlADEIAQ2ABAACAAAAAgACAAEEkAYAlAEAAAEAAAAQAAAAAwAAMAIAAAAL AA8OAAAAAAIB/w8BAAAAQwAAAAAAAACBKx+kvqMQGZ1uAN0BD1QCAAAAAGxpbnV4LW1pcHNAZm5l dC5mcgBTTVRQAGxpbnV4LW1pcHNAZm5ldC5mcgAAHgACMAEAAAAFAAAAU01UUAAAAAAeAAMwAQAA ABMAAABsaW51eC1taXBzQGZuZXQuZnIAAAMAFQwBAAAAAwD+DwYAAAAeAAEwAQAAABUAAAAnbGlu dXgtbWlwc0BmbmV0LmZyJwAAAAACAQswAQAAABgAAABTTVRQOkxJTlVYLU1JUFNARk5FVC5GUgAD AAA5AAAAAAsAQDoBAAAAHgD2XwEAAAATAAAAbGludXgtbWlwc0BmbmV0LmZyAAACAfdfAQAAAEMA AAAAAAAAgSsfpL6jEBmdbgDdAQ9UAgAAAABsaW51eC1taXBzQGZuZXQuZnIAU01UUABsaW51eC1t aXBzQGZuZXQuZnIAAAMA/V8BAAAAAwD/XwAAAAACAfYPAQAAAAQAAAAAAAACuVIBBIABACMAAABS RTogWW91ciBMaW51eC1Qb3J0IG9uIGEgU05JIFJNMjAwAHsKAQWAAwAOAAAAzQcFABAACgAGAAAA BQD+AAEggAMADgAAAM0HBQAQAAkAOgAbAAUATAEBCYABACEAAABCOTZERTc2QUMxQ0REMDExQUY2 MDAwMjBBRkYxMTRGMwBEBwEDkAYAFAoAACEAAAALAAIAAQAAAAsAIwAAAAAAAwAmAAAAAAALACkA AAAAAAMALgAAAAAAAwA2AAAAAABAADkAUPIk/s9hvAEeAHAAAQAAACMAAABSRTogWW91ciBMaW51 eC1Qb3J0IG9uIGEgU05JIFJNMjAwAAACAXEAAQAAABYAAAABvGHP/hRq5227zcER0K9gACCv8RTz AAAeAB4MAQAAAAUAAABTTVRQAAAAAB4AHwwBAAAAEQAAAGp1bmdAZXNlLW1ldHouZnIAAAAAAwAG EFpM4k8DAAcQYAIAAB4ACBABAAAAZQAAAEhJLElIQVZFVVNFREFSTTIwMEFZRUFSQUdPLFNPSUhP UEVXSEFUSVNBWUlTQ09SUkVDVDotKUlUSElOS1lPVUhBVkVUT0NIQU5HRUZJUk1XQVJFRE9JTkdJ VCxZT1VXSUxMQkUAAAAAAgEJEAEAAADwBgAA7AYAALoOAABMWkZ1ZWvlCQMACgByY3BnMTI1cjIM YGMxAzABBwtgbpEOEDAzMw8WZmUPkZ42AfcCpANjAgBjaArACHNldALRcHJxMiUAACoKoW5vElAg MAsB0AHQNg+gMDUwNOcUIQHQFBA0fQdtAoMAUPcD1BH/EwtiE+EUUBOyGPQXFNAHExXkNhGOMjM4 RRdUIAdtIENFFeQ3Txp/FEAbrxy1eXIV5DlfEY4aUBYxHv8DgkcJ0Wu/AoMMASD/DlAiLwNzVAhw dyPUFjEhDTgaYSWfA4JC/QdAdA3gI9QlYRZsG3gHE/8dBhtwKv8etyyVIFUOMBZO/yHoLJQjiRph ME4lZiyUJud/HZEwTSiXLJQqJgKRCOY71QlvMDi/ZQ4wNTnqOwH/Or87yTnUO/I6Xz4vPe09b+c7 nznvEGAyOEO6RNFEj/9FmTnURcJEL0f/R71HP0Vv/Uk0OQ5QTIRN4UYDTeACglBzdHlsB5BoCeB0 hwAAE1AD8GRjdGwKscJcUDhhZGp1T1AFEHxnaAVCFjIMAQ9VGmBjhwnAUEADMHNuZXgXMJ8HsAWw AMACcxMQY3MPkFMDMFFwZGkqUHYTgERJARBhdSpAIFAKwGHBCcBhcGggRgIhTxNSMTARaS0PkDhS oWnXV/NQH1CDYgsgcglQWhKzFqBaEnc0QyEXAHAB0E9VMlGfUq9PQGJhFxBkjwIgVSFTw1eQIElt F3AXB5AAkAIgLVXROiDAMWAwT2JqFyBgMERh/nRgIFdEGmFX31jvWf9bD/9cF0+gW9ALgA4SVSEM MFykfw5QXS9ePykRXCATgF9aSf5uVbAXASBjVQAdkWHfYu/fY/9lCwhgUHALgGVPgFWwbmwBQGXf ZuIwcqAI0GLZCrB0OGdYXQYyHsBor8dexXMwakRS6XACIG7QmGUvVFaAAIBm6TjQ/2vVG5Fsf22P bp9lD1wfaB//dR92j3eRYEJgEmD5INB5P3tRL3LUOX0Pfh9ep4TQRPhvY3UHgAIwBdBWwDfhwQCx cmdsMTQdgYiy7nKJByHQiRZiiQIOowGSv3iAgwKCYIKgesEBgG5gkI8AYAnwVWCHwFxoeVbQ4Ghv dHo0MAJUQWaiLmUA8IfAT2BwPGBcdukIkHdrC4BkHsCPUgTw+QdAZTcwEA4AepJPsHxQ/5ClAhCN sAMxFyES8mEQX2CHC2BhEB0AOlxcUANgvVahbVbwAxAHkJNQTQ3g+QNgc28BgGBwASAN4I6gWlyV BkUAwAMQLl5wdN+IkBcQgqBTkXCieAFAkRJ9BJB5i6Ag0ZHRl9UI4XP2eJgCjPFuT7A40JakVfSe YwMgEvMAgAWQbHaI8P+bIQ5wVFCbIgGQACCbso+h/4gBAcGbIRbgD3AAAJswDND5AZAgLjfymxgO UJvSKkD/gvCcT51fnm8PwJswBYGgD9uhH6IvbB7AmzBsn8+kj32llSmenJCRo3+oX6WTYvwgKAKR qW+bYxpgpx+r3/+s763/m5AdkK9CnB+wr7G//56cG5CvT7TPtd+275uQIND/s8+5X7pvu3QK+QMw gl+Db8OE/1NFe0hpLAqFCoW8SSAW4FWxwqAJgCCl4NxSTQHQVBCl4HmRIAXAsVaQbywglLDFwm+P EBwgdxbgBUDF0HNheTogBAAgBaE40MFgXH5JYDAtKcT+dGgLgGv3xyAIYMXkdMfgFtEPcBOAy5Pw VFB3CsBlLoehZuFdyRB0x7DLggPwbAMgYt8TgAGgT4DMAsZBIA9AyDD2cMkAVXBzy2COIc5wkdH7 ZuHNMFSZwskSpeDPepBB/0+AU5AAkMwQAjDNMGtgyxG7z1HMxSAEYAMAzBByzdT/FtCR0BNx05LI MCpQAiDMNr3T1yjF0JPBE1AFQHMIcJXVwWbTg26TwGUpxoB/dtDLcwuAa7EFQNOSz+IufcT8SsKh 2QGNsJnByxNnb8ewBpDFwdeFbQQAAZBrPwnwx7DMQ81y09c40HF1/8zAE4AXEXCSzZAXcIfAX5HL BbHA0ULC4CBFdtAHMH/TcghgwvJVgMhgaTDg0kxfVYDBcBOA4VXQMlMLgGl+eNNB2FDLeKXgU9AH 4U2/3JHiIdFxcIADcSpRbM/B9QIgZcewYuIh2FDLgQXA38awyRLfUZMBBvBkzdMAEPvfUJhhIM5R y82l4MKQX2C/gOHZkcFA0UHTMMT8Pe0P9+4f7x0KhVDLMHjAz6ATgLBKVU5HuuDwli7bcDEPcEBT dY8QT4BjLvsDUMoWNFZQC2CVQOFAXnC+dQsRKhB3MebQx7A1i6Aji6AF0EVUWsewRlIcQU4dEAqF wwB0cDooLy939wAuB5BlLWcHgI3A8tEvfsKQD3AvBTfwAPiQAwAQEAAAAAADABEQAAAAAAMAgBD/ ////QAAHMEDWte/OYbwBQAAIMEDWte/OYbwBAwAAgAggBgAAAAAAwAAAAAAAAEYAAAAAEIUAAAAA AAALAAOACCAGAAAAAADAAAAAAAAARgAAAAADhQAAAAAAAAMABYAIIAYAAAAAAMAAAAAAAABGAAAA AFKFAAC3DQAAAwAJgAggBgAAAAAAwAAAAAAAAEYAAAAAAYUAAAAAAAAeABKACCAGAAAAAADAAAAA AAAARgAAAABUhQAAAQAAAAQAAAA4LjAACwAWgAggBgAAAAAAwAAAAAAAAEYAAAAADoUAAAAAAAAD ABeACCAGAAAAAADAAAAAAAAARgAAAAARhQAAAAAAAAMAGYAIIAYAAAAAAMAAAAAAAABGAAAAABiF AAAAAAAAHgAogAggBgAAAAAAwAAAAAAAAEYAAAAANoUAAAEAAAABAAAAAAAAAB4AKYAIIAYAAAAA AMAAAAAAAABGAAAAADeFAAABAAAAAQAAAAAAAAAeACqACCAGAAAAAADAAAAAAAAARgAAAAA4hQAA AQAAAAEAAAAAAAAAHgA9AAEAAAAFAAAAUkU6IAAAAAADAA00/TcAALRM ------ =_NextPart_000_01BC61E0.C28B5110-- From tim@cobaltmicro.com Fri May 16 11:01:28 1997 Received-Date: Fri, 16 May 1997 11:01:25 +0200 (MET DST) Received: from [205.179.228.26] by viavision.com with ESMTP (Apple Internet Mail Server 1.1.1); Fri, 16 May 1997 01:10:59 -0800 X-Sender: tim@mail.cobaltmicro.com (Unverified) Message-Id: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Fri, 16 May 1997 02:11:41 -0800 To: linux-mips@fnet.fr From: Timothy Stonis Subject: xxx called nonatomically from... Content-Length: 378 Lines: 13 Hi, I'm trying to get my interrupt structure straight on an Algorithmics P4032 board with a RM5230 from QED. The interrupts seem to be working fine, however, I am getting: gfp called nonatomically from interrupt 00000000 (and) kmalloc called... I saw a note in the sources that this has something to do with the caches. What exactly is going on here? Thanks. _Tim From linux@informatik.uni-koblenz.de Fri May 16 12:27:35 1997 Received-Date: Fri, 16 May 1997 12:26:13 +0200 (MET DST) Received: from thoma (linux@thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.8.5/8.6.9) with SMTP id MAA09679; Fri, 16 May 1997 12:25:32 +0200 (MEST) From: Systemkennung Linux Message-Id: <199705161025.MAA09679@informatik.uni-koblenz.de> Received: by thoma (SMI-8.6/KO-2.0) id MAA05507; Fri, 16 May 1997 12:25:31 +0200 Subject: Re: Your Linux-Port on a SNI RM200 To: linux-mips@fnet.fr Date: Fri, 16 May 1997 12:25:30 +0200 (MET DST) Cc: romal@informatik.uni-bremen.de In-Reply-To: <01BC61D3.C0508F30@romal@informatik.uni-bremen.de> from "Robert M. Albrecht" at May 16, 97 08:31:51 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1589 Lines: 45 Hi, > I?ve got an used SNI RM200-120 :-) > > Sadly, the whole documentation and the key (for opening the case) is > missing. It is only the box, a CD-ROM Sinix V5.42 and the keyboard. I have > attached my VGA-Monitor and a PS/2-style mouse. > > According to the self-diagnosis this machine is equipped with 32MB RAM and > a MIPS 4600. Built in s a harddisk (booting Sinix), a CD-Rom (looks like a > toshiba) and a standard-cartridge-streamer (looks like tandberg).. > > I want to get Linux on this box and get rid of this ridiculous Sinix. Who wonders - there is only one OS on this world and Milo is his prophet ;-) > I have gotten into the firmware-monitor, by guessing I tried: > > boot floppy()milo > > but the firmwares complains > > loading floppy()milo > couldn?t determine fs type > Coudln?t load floppy()milo > Do you have any ideas how to boot these machine ? Do have know how to > install Windows NT on this box ? You're running the big endian firmware for Sinix. That one is not supported. You will have to reconfigure that machine for Windows NT by reloading the EEPROM with NT ARC firmware. There are different types of RM200; Linux only supports the RM200C series machines which have PCI slots. Probably the big brother RM300C will work also. No idea about older EISA only machines. Try it, fry it ;-) As an aside - the entire RM port is currently being redone because SNI gave out the required technical information. Also the Lance32 driver Linux needs for that machine has been completely redone and will need to be ported to MIPS again. Ralf From davidm@fuzzbox.psrg.cs.usyd.edu.au Fri May 16 12:29:01 1997 Received-Date: Fri, 16 May 1997 12:28:56 +0200 (MET DST) Received: from fuzzbox.psrg.cs.usyd.edu.au (davidm@fuzzbox.psrg.cs.usyd.edu.au [129.78.97.193]) by fuzzbox.psrg.cs.usyd.edu.au (8.8.4/8.8.4) with SMTP id UAA22692 for ; Fri, 16 May 1997 20:28:17 +1000 Date: Fri, 16 May 1997 20:28:17 +1000 (EST) From: David Monro Sender: davidm@fuzzbox.psrg.cs.usyd.edu.au Reply-To: David Monro Subject: boot code stuff To: linux-mips@fnet.fr Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-Length: 281 Lines: 6 I remember people saying on the list a little while ago that the PROM on the DECstation 3100 (I think) was very similar to the MIPS 32xx one. Do I remember correctly? If so, could someone provide me with some sort of code for the DEC one so I can attack the MIPS properly? David From linux@informatik.uni-koblenz.de Fri May 16 12:39:19 1997 Received-Date: Fri, 16 May 1997 12:38:59 +0200 (MET DST) Received: from thoma (linux@thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.8.5/8.6.9) with SMTP id MAA11449; Fri, 16 May 1997 12:38:25 +0200 (MEST) From: Systemkennung Linux Message-Id: <199705161038.MAA11449@informatik.uni-koblenz.de> Received: by thoma (SMI-8.6/KO-2.0) id MAA05557; Fri, 16 May 1997 12:38:23 +0200 Subject: Re: xxx called nonatomically from... To: linux-mips@fnet.fr Date: Fri, 16 May 1997 12:38:22 +0200 (MET DST) Cc: tim@cobaltmicro.com In-Reply-To: from "Timothy Stonis" at May 16, 97 02:11:41 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 836 Lines: 21 Hi, > I'm trying to get my interrupt structure straight on an Algorithmics > P4032 board with a RM5230 from QED. The interrupts seem to be working fine, > however, I am getting: gfp called nonatomically from interrupt 00000000 > (and) > kmalloc called... > > I saw a note in the sources that this has something to do with the > caches. What exactly is going on here? Thanks. That is not directly related to a cache problem. In the past this message has been frequently caused by cache problems. The messages you're getting are from kmalloc when being called with another priority than GFP_ATOMIC. This is illegal because with other priorities the kernel might have to sleep for example when paging. Interrupts however may not sleep. There is a zillion other bugs that might cause this. Who is calling kmalloc? Ralf From paul@suede.sw.oz.au Fri May 16 13:02:10 1997 Received-Date: Fri, 16 May 1997 13:02:07 +0200 (MET DST) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id LAA23789; Fri, 16 May 1997 11:01:41 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id LAA26228; Fri, 16 May 1997 11:01:40 GMT (SMI-8.6/1.34) (from paul for linux-mips@fnet.fr) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705161101.LAA26228@suede.sw.oz.au> Subject: Re: boot code stuff To: linux-mips@fnet.fr Date: Fri, 16 May 1997 21:01:38 +1000 (EST) In-Reply-To: from "David Monro" at May 16, 97 08:28:17 pm Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 1384 Lines: 28 David Monro wrote: > I remember people saying on the list a little while ago that the PROM on > the DECstation 3100 (I think) was very similar to the MIPS 32xx one. Do > I remember correctly? If so, could someone provide me with some sort of > code for the DEC one so I can attack the MIPS properly? How about I upload a 2.1.14.2.dec snapshot from which you can extract some code? Or, you could add your big-endian patches and use that kernel instead. Send me the big-endian patches when you have them if you like, and I'll throw them in for Ralf to integrate... I'll let you know when you can grab the current snapshot! Other DECStation hackers can play too, but must promise not to whinge about crufty bits of daggy code... though you should all feel free to send patches/corrections. [N.B. The term 'daggy' is Australian and is a reference to the crusty bits of wool that hang from the back end of a sheep. :-] Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From davidm@fuzzbox.psrg.cs.usyd.edu.au Fri May 16 13:36:19 1997 Received-Date: Fri, 16 May 1997 13:36:14 +0200 (MET DST) Received: from fuzzbox.psrg.cs.usyd.edu.au (davidm@fuzzbox.psrg.cs.usyd.edu.au [129.78.97.193]) by fuzzbox.psrg.cs.usyd.edu.au (8.8.4/8.8.4) with SMTP id VAA23355 for ; Fri, 16 May 1997 21:34:51 +1000 Date: Fri, 16 May 1997 21:34:51 +1000 (EST) From: David Monro Sender: davidm@fuzzbox.psrg.cs.usyd.edu.au Reply-To: David Monro Subject: Re: boot code stuff To: linux-mips@fnet.fr In-Reply-To: <199705161101.LAA26228@suede.sw.oz.au> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-Length: 1964 Lines: 40 A 2.1.14 code snapshot would be great - I'm working off 2.1.1 at the moment. How hard would it be to make a patch against that rather than me clogging up your link with the whole 6MB (can't figure out why I only get 1.7k/s to your site - you are only about 6 hops away and most of them are multi-megabit. I'd do better getting you to batch it to me via mhs over the modem in our machine room you use for a newsfeed, 'cept I don't think it is set up properly for that :-)) David > David Monro wrote: > > > I remember people saying on the list a little while ago that the PROM on > > the DECstation 3100 (I think) was very similar to the MIPS 32xx one. Do > > I remember correctly? If so, could someone provide me with some sort of > > code for the DEC one so I can attack the MIPS properly? > > How about I upload a 2.1.14.2.dec snapshot from which you can extract some > code? Or, you could add your big-endian patches and use that kernel > instead. Send me the big-endian patches when you have them if you like, > and I'll throw them in for Ralf to integrate... > > I'll let you know when you can grab the current snapshot! Other DECStation > hackers can play too, but must promise not to whinge about crufty bits of > daggy code... though you should all feel free to send patches/corrections. > > [N.B. The term 'daggy' is Australian and is a reference to the crusty > bits of wool that hang from the back end of a sheep. :-] > > Paul > _______________________________________________________________________________ > Paul M. Antoine, Net: paul@sw.oz.au > Softway Pty Ltd WWW: www.softway.com.au > PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 > 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 > 9699 9174 > > "It is the lack of acceptance of diversity which threatens to > destroy society, NOT the free expression of it." - Me. From davidm@fuzzbox.psrg.cs.usyd.edu.au Fri May 16 13:36:11 1997 Received-Date: Fri, 16 May 1997 13:35:58 +0200 (MET DST) Received: from fuzzbox.psrg.cs.usyd.edu.au (davidm@fuzzbox.psrg.cs.usyd.edu.au [129.78.97.193]) by fuzzbox.psrg.cs.usyd.edu.au (8.8.4/8.8.4) with SMTP id VAA23362 for ; Fri, 16 May 1997 21:35:14 +1000 Date: Fri, 16 May 1997 21:35:14 +1000 (EST) From: David Monro Sender: davidm@fuzzbox.psrg.cs.usyd.edu.au Reply-To: David Monro Subject: hmmm... To: linux-mips@fnet.fr Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-Length: 566 Lines: 12 Turns out I actually own a DECstation 3100. I'm not sure how much use it is though - it may not work properly. Does anyone know what it means when it counts down to 3 and then says failure - press reset to continue? Can these handle a serial console? (somewhere I have an adapter cable for the stupid DEC serial port). If so, how do I make it use it - especially as I have no keyboard to plug into it at all as far as I know, so I can't set stuff in the environment or anything useful like that... or does it take the same keyboard as the 5100/2xx series? David From frist@lehr.chem.tu-berlin.de Fri May 16 13:51:43 1997 Received-Date: Fri, 16 May 1997 13:51:39 +0200 (MET DST) Received: (from frist@localhost) by lehr.chem.tu-berlin.de (8.7.4/8.7.3) id NAA00977; Fri, 16 May 1997 13:51:29 +0200 Date: Fri, 16 May 1997 13:51:29 +0200 (MET DST) From: Frieder Streffer To: linux-mips@fnet.fr Subject: Re: hmmm... In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 870 Lines: 28 On Fri, 16 May 1997, David Monro wrote: > Turns out I actually own a DECstation 3100. I'm not sure how much use it is > though - it may not work properly. Does anyone know what it means when it > counts down to 3 and then says failure - press reset to continue? > > Can these handle a serial console? (somewhere I have an adapter cable for the > stupid DEC serial port). If so, how do I make it use it - especially as I have > no keyboard to plug into it at all as far as I know, so I can't set stuff in Jep, no keyboard (and propably no mouse), that's your problem, that causes the self test to stop at ..3 as far as I remember. > the environment or anything useful like that... or does it take the same > keyboard as the 5100/2xx series? As far as I know the 3100 and the 5100/2xx have the same keybords (both LK201). regards, frieder > > David > > From engel@numerik.math.uni-siegen.de Fri May 16 13:48:14 1997 Received-Date: Fri, 16 May 1997 13:48:11 +0200 (MET DST) Received: from hilbert.numerik.math.uni-siegen.de (hilbert.numerik.math.uni-siegen.de [141.99.112.8]) by bernoulli.numerik.math.uni-siegen.de (Mailhost) with SMTP id NAA03414 for ; Fri, 16 May 1997 13:45:02 +0200 (MET DST) From: Michael Engel Message-Id: <9705161151.AA01231@hilbert.numerik.math.uni-siegen.de> Received: by hilbert.numerik.math.uni-siegen.de id AA01231; Fri, 16 May 97 13:51:55 +0200 Subject: Re: hmmm... To: linux-mips@fnet.fr Date: Fri, 16 May 1997 13:51:55 +0200 (MET DST) In-Reply-To: from "David Monro" at May 16, 97 09:35:14 pm X-Mailer: ELM [version 2.4 PL23] Content-Type: text Content-Length: 3552 Lines: 101 Hi, > > Turns out I actually own a DECstation 3100. I'm not sure how much use it is > though - it may not work properly. Does anyone know what it means when it > counts down to 3 and then says failure - press reset to continue? I found this in comp.unix.ultrix a while ago: ---------- Subject: DEC 3100 subtests, LEDs, pinouts DEC 3100 LED's, subtests, pinouts ------------------------------------------------------------ >From my first posting and from all your responses and questions, thought would update this to try and share this info. I have noticed postings recently about people having problems when booting a DEC 2100/3100, so I thought I would post this. ------------------------------------------------------------ This is a short description of the DEC subtests from the 2100/3100 manual. Subtest 7, 6, 5, 4 - No info listed in the manual, says to contact your DEC rep. from my notes Subtest 7 - Video SIM module Subtest 6 - VDAC PCC Subtest 5 - DZ Serial line Subtest 4 - System Module, i.e. motherboard, i.e. spare parts time from the manual Subtest 3 - Check keyboard & mouse Subtest 2 - Check Ethernet connections, make sure they are terminated properly if not on a network. Subtest 1 - Check SCSI bus connections ------------------------------------------------------------ I have had luck when a machine seemed dead or dying, by removing the cover and wiggling the memory modules and small video board. Seems if they set for a while, they must get a little corroded. ------------------------------------------------------------ On the back of the machine is a set of 8 LED's that show which test is being performed at boot. ff - set at power up or reset 7f - system module bf - disk & SCSI df - keyboard ef - memory f7 - mouse ------------------------------------------------------------ If your system comes up and you are at the '>>' prompt, there are some commands you can enter to do system tests. f arg - video test args: c displays color bars x displays white grid just press the space bar to terminate test Note that they are several other tests, some which are canned disk tests and destroy all data. So don't go poking keys to see what happens. Be careful what you ask for, you may get it. UPDATE - have had several requests for more on this, so will post it under something like: DEC 3100 Standalone Tests ------------------------------------------------------------ UPDATE - Someone asks for info on the 15 pin video port on the back. 1 Red 2 Ground 10 Green 11 Blue ------------------------------------------------------------ Hope this helps some of you, if so, I would appreciate a note from you. I am curious, how many are still using some of the 'older' DEC equipment. Mike Thompson --------- Looks like you have a problem with your keyboard and/or mouse ... > Can these handle a serial console? (somewhere I have an adapter cable for the > stupid DEC serial port). If so, how do I make it use it - especially as I have > no keyboard to plug into it at all as far as I know, so I can't set stuff in > the environment or anything useful like that... or does it take the same > keyboard as the 5100/2xx series? Yes, the standard LK201 keyboard and the "hockey puck" mouse (well, we prefer to call the thing rat ;-)) VS-XXX-AA work with the 3100. To use a serial console, you have to take the video RAM module out of the DS3100 and connect the console terminal to the printer port - 9600,8N1. regards, Michael Engel (engel@unix-ag.uni-siegen.de) From paul@suede.sw.oz.au Fri May 16 15:23:24 1997 Received-Date: Fri, 16 May 1997 15:23:11 +0200 (MET DST) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id NAA25672; Fri, 16 May 1997 13:22:43 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id NAA04776; Fri, 16 May 1997 13:22:41 GMT (SMI-8.6/1.34) (from paul for linux-mips@fnet.fr) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705161322.NAA04776@suede.sw.oz.au> Subject: Re: boot code stuff To: linux-mips@fnet.fr Date: Fri, 16 May 1997 23:22:40 +1000 (EST) In-Reply-To: from "David Monro" at May 16, 97 09:34:51 pm Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 1647 Lines: 40 David Monro wrote: > A 2.1.14 code snapshot would be great - I'm working off 2.1.1 at the moment. It's uploaded, and should appear at fnet within the hour. It's at: ftp://ftp.fnet.fr/linux-mips/ftp.softway.com.au/linux-2.1.14.2.dec.tar.gz > How hard would it be to make a patch against that rather than me clogging up > your link with the whole 6MB ...yuk... almost as big... well, not quite, but I don't have the 2.1.1.dec source tree on my filesystem anymore... > (can't figure out why I only get 1.7k/s to your site - you are only about > 6 hops away and most of them are multi-megabit. ...except the last one which is 28.8kbps... we promise to upgrade to ISDN Real Soon Now(tm). FTP'ing from France may prove quicker...! > I'd do better getting you to batch it to me via mhs over the modem in our > machine room you use for a newsfeed, 'cept I don't think it is set up > properly for that :-)) Well, I have no trouble doing a netftp via basser, so maybe it is set up right... don't know how to upload though...?? [For those confused by the last exchange, David is connected via the University of Sydney, which is just 500m from Softway, we get our newsfeed from them.] Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From paul@suede.sw.oz.au Fri May 16 15:25:06 1997 Received-Date: Fri, 16 May 1997 15:25:01 +0200 (MET DST) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id NAA25680; Fri, 16 May 1997 13:24:39 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id NAA04990; Fri, 16 May 1997 13:24:38 GMT (SMI-8.6/1.34) (from paul for linux-mips@fnet.fr) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705161324.NAA04990@suede.sw.oz.au> Subject: Re: hmmm... To: linux-mips@fnet.fr Date: Fri, 16 May 1997 23:24:37 +1000 (EST) In-Reply-To: <9705161151.AA01231@hilbert.numerik.math.uni-siegen.de> from "Michael Engel" at May 16, 97 01:51:55 pm Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 658 Lines: 18 Micheal Engel wrote: > I found this in comp.unix.ultrix a while ago: > > ---------- > Subject: DEC 3100 subtests, LEDs, pinouts Thanks Michael - I'll put it up on the DECStation web site for reference. Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From tschroed@cheetah.wsc.edu Fri May 16 15:27:15 1997 Received-Date: Fri, 16 May 1997 15:27:09 +0200 (MET DST) Received: by cheetah.wsc.edu (5.65/DEC-Ultrix/4.4) id AA13516; Fri, 16 May 1997 08:26:34 -0500 Other-Stuff: localhost tschroed@cheetah.wsc.edu Trevor Schroeder tschroed Date: Fri, 16 May 1997 08:26:33 -0500 (CDT) From: Trevor Schroeder To: linux-mips@fnet.fr Subject: Re: boot code stuff In-Reply-To: <199705161101.LAA26228@suede.sw.oz.au> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 784 Lines: 20 On Fri, 16 May 1997, Paul Antoine wrote: > How about I upload a 2.1.14.2.dec snapshot from which you can extract some > code? Or, you could add your big-endian patches and use that kernel Yes!! Please!! > [N.B. The term 'daggy' is Australian and is a reference to the crusty > bits of wool that hang from the back end of a sheep. :-] Could use the term natty too...It's a patois mangling of nasty...See also, Natty Dread ____________________________________________________________ "One unerring mark of the love of truth is not entertaining any propositions with greater assurance than the proofs it is built upon will warrant" -- John Locke, 1690 Trevor Schroeder tschroed@cheetah.wsc.edu ------------------------------------------------------------ From khp@imladris.dolphinics.no Fri May 16 15:31:54 1997 Received-Date: Fri, 16 May 1997 15:31:52 +0200 (MET DST) Received: from imladris.dolphinics.no (pcbob.dolphinics.no) by scimitar.dolphinics.no (4.1/SMI-4.1) id AA00668; Fri, 16 May 97 15:33:55 +0200 Received: (from khp@localhost) by imladris.dolphinics.no (8.7.5/8.7.3) id PAA03568 for linux-mips@fnet.fr; Fri, 16 May 1997 15:30:55 +0200 Message-Id: <199705161330.PAA03568@imladris.dolphinics.no> Subject: Re: boot code stuff In-Reply-To: <199705161322.NAA04776@suede.sw.oz.au> from Paul Antoine at "May 16, 97 11:22:40 pm" To: linux-mips@fnet.fr Date: Fri, 16 May 1997 15:30:54 +0200 (MET DST) From: Kai Harrekilde-Petersen X-Mailer: ELM [version 2.4ME+ PL27 (25)] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 689 Lines: 17 Paul Antoine writes: > David Monro wrote: > > I'd do better getting you to batch it to me via mhs over the modem in our > > machine room you use for a newsfeed, 'cept I don't think it is set up > > properly for that :-)) > > [For those confused by the last exchange, David is connected via the > University of Sydney, which is just 500m from Softway, we get our newsfeed > from them.] Walknet/TAPE seems to be a reasonable alternative :-) Kai -- Kai Harrekilde-Petersen #include http://www.dolphinics.no/~khp/ Linux: the choice of a GNU generation "Argue for your limitations, and sure enough - they're yours" --Richard Bach. From paul@suede.sw.oz.au Fri May 16 16:06:51 1997 Received-Date: Fri, 16 May 1997 16:06:45 +0200 (MET DST) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id OAA26309; Fri, 16 May 1997 14:06:13 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id OAA07191; Fri, 16 May 1997 14:06:12 GMT (SMI-8.6/1.34) (from paul for linux-mips@fnet.fr) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705161406.OAA07191@suede.sw.oz.au> Subject: Re: boot code stuff To: linux-mips@fnet.fr Date: Sat, 17 May 1997 00:06:11 +1000 (EST) In-Reply-To: <199705161330.PAA03568@imladris.dolphinics.no> from "Kai Harrekilde-Petersen" at May 16, 97 03:30:54 pm Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 660 Lines: 16 Kai writes: > Walknet/TAPE seems to be a reasonable alternative :-) You've got to be kidding! Walk?!?! Exercise!? That would mean leaving the keyboard for a moment... not to mention putting on shoes! :-) Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From paul@suede.sw.oz.au Fri May 16 16:10:25 1997 Received-Date: Fri, 16 May 1997 16:10:22 +0200 (MET DST) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id OAA26341; Fri, 16 May 1997 14:09:40 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id OAA07221; Fri, 16 May 1997 14:09:39 GMT (SMI-8.6/1.34) (from paul for linux-mips@fnet.fr) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705161409.OAA07221@suede.sw.oz.au> Subject: Re: boot code stuff To: linux-mips@fnet.fr Date: Sat, 17 May 1997 00:09:38 +1000 (EST) In-Reply-To: from "Trevor Schroeder" at May 16, 97 08:26:33 am Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 1144 Lines: 32 Trevor wrote: > > How about I upload a 2.1.14.2.dec snapshot from which you can extract some > > code? Or, you could add your big-endian patches and use that kernel > > Yes!! Please!! Settle! It should be up there. It seems to have one compilation problem: drivers/Makefile - add dec to ALL_SUB_DIRS ...so it will make depend correctly. Use this source at your peril... :-) > > [N.B. The term 'daggy' is Australian and is a reference to the crusty > > bits of wool that hang from the back end of a sheep. :-] > > Could use the term natty too...It's a patois mangling of nasty...See also, > Natty Dread Ohhhh... I always thought natty meant clever or useful (i.e. 'neat') Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From tschroed@cheetah.wsc.edu Fri May 16 16:21:09 1997 Received-Date: Fri, 16 May 1997 16:21:04 +0200 (MET DST) Received: by cheetah.wsc.edu (5.65/DEC-Ultrix/4.4) id AA13615; Fri, 16 May 1997 09:20:37 -0500 Other-Stuff: localhost tschroed@cheetah.wsc.edu Trevor Schroeder tschroed Date: Fri, 16 May 1997 09:20:37 -0500 (CDT) From: Trevor Schroeder To: Paul Antoine Cc: linux-mips@fnet.fr Subject: Re: boot code stuff In-Reply-To: <199705161409.OAA07221@suede.sw.oz.au> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 645 Lines: 15 On Sat, 17 May 1997, Paul Antoine wrote: > Ohhhh... I always thought natty meant clever or useful (i.e. 'neat') Well, I've also used/heard it used in the context of spiffy, neat looking, at least. ie, "ooooh...Segun (friend of mine who often dresses nattily), looking pretty natty today..." ____________________________________________________________ "One unerring mark of the love of truth is not entertaining any propositions with greater assurance than the proofs it is built upon will warrant" -- John Locke, 1690 Trevor Schroeder tschroed@cheetah.wsc.edu ------------------------------------------------------------ From rji@bristol.st.com Fri May 16 16:54:41 1997 Received-Date: Fri, 16 May 1997 16:54:36 +0200 (MET DST) Received: by hawkweed id PAA10933; Fri, 16 May 1997 15:54:39 +0100 (BST) From: "Richard Ingram" Message-Id: <9705161554.ZM4222@bristol.st.com> Date: Fri, 16 May 1997 15:54:31 +0100 In-Reply-To: paul@suede.sw.oz.au (Paul Antoine) "Re: hmmm..." (May 16, 11:24pm) References: <199705161324.NAA04990@suede.sw.oz.au> X-Mailer: Z-Mail (3.2.1 10apr95) To: linux-mips@fnet.fr Subject: Linux on MIPS RC3xxx boxes ? Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Length: 1148 Lines: 34 Hello, First off what is the difference between the Magnum 3000 and the RC3xxx sereis or ar they one and the same ? I have a Mips RC3xxx server box with 2*1.2gig and 1 exabyte streamer and will be getting another on monday and was wondering what status the port is at to these older Mips boxes (if one exists at all). Supposedly the box has RISCix on one of the HDs already but no memory or graphics card so I will be configuring it to be run off the seriel port. I was going to cannibalise one of the boxes for one of the HDs to go on my Magnum 4K box to try and get the serial kernel up in the near future (just need memory for this as well!). What sort of performance did these old RC3xxx boxes give ? Does anyone still use them ? I recall seeing in the past week that someone has got it booting up on a 3020 does this mean there is hope for my RC3xxx (I'll get the exact number next week). Thanks, Richard. Developer of Micro Development Tools @ SGS Thomson Microelectronics Ltd. PS My offer is stil open to support the WWW pages as I now have a PPro PC running Linux/NT at home and the modem/Inet account arrives in 10-14 days. -- From engel@numerik.math.uni-siegen.de Fri May 16 20:01:41 1997 Received-Date: Fri, 16 May 1997 20:01:40 +0200 (MET DST) Received: from hilbert.numerik.math.uni-siegen.de (hilbert.numerik.math.uni-siegen.de [141.99.112.8]) by bernoulli.numerik.math.uni-siegen.de (Mailhost) with SMTP id TAA03558 for ; Fri, 16 May 1997 19:58:42 +0200 (MET DST) From: Michael Engel Message-Id: <9705161805.AA01520@hilbert.numerik.math.uni-siegen.de> Received: by hilbert.numerik.math.uni-siegen.de id AA01520; Fri, 16 May 97 20:05:36 +0200 Subject: Re: boot code stuff To: linux-mips@fnet.fr Date: Fri, 16 May 1997 20:05:36 +0200 (MET DST) In-Reply-To: <199705161409.OAA07221@suede.sw.oz.au> from "Paul Antoine" at May 17, 97 00:09:38 am X-Mailer: ELM [version 2.4 PL23] Content-Type: text Content-Length: 520 Lines: 18 Paul writes: > > Settle! It should be up there. It seems to have one compilation problem: > > drivers/Makefile - add dec to ALL_SUB_DIRS > > ...so it will make depend correctly. > > Use this source at your peril... :-) > Well, it compiles fine! I can't get to the DS3100 to test it on at the moment, as the d*mn electronic key system won't let me into the room at university ... Have you already found (and fixed ;-)) the problem with the DS3100, Paul ? regards, Michael Engel (engel@unix-ag.uni-siegen.de) From paul@suede.sw.oz.au Sat May 17 00:28:29 1997 Received-Date: Sat, 17 May 1997 00:28:25 +0200 (MET DST) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id WAA02819; Fri, 16 May 1997 22:28:09 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id WAA09773; Fri, 16 May 1997 22:28:08 GMT (SMI-8.6/1.34) (from paul for linux-mips@fnet.fr) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705162228.WAA09773@suede.sw.oz.au> Subject: Re: boot code stuff To: linux-mips@fnet.fr Date: Sat, 17 May 1997 08:28:08 +1000 (EST) In-Reply-To: <9705161805.AA01520@hilbert.numerik.math.uni-siegen.de> from "Michael Engel" at May 16, 97 08:05:36 pm Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 868 Lines: 21 David Monro writes: > Well, it compiles fine! I can't get to the DS3100 to test it on at the moment, > as the d*mn electronic key system won't let me into the room at university ... Isn't that just typical?! > Have you already found (and fixed ;-)) the problem with the DS3100, Paul ? I haven't found the reason for it bombing after just a few messages, but it should be easy... Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From paul@suede.sw.oz.au Sat May 17 01:01:18 1997 Received-Date: Sat, 17 May 1997 01:01:15 +0200 (MET DST) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id XAA03191; Fri, 16 May 1997 23:00:56 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id XAA09922; Fri, 16 May 1997 23:00:55 GMT (SMI-8.6/1.34) (from paul for linux-mips@fnet.fr) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705162300.XAA09922@suede.sw.oz.au> Subject: Re: Linux on MIPS RC3xxx boxes ? To: linux-mips@fnet.fr Date: Sat, 17 May 1997 09:00:55 +1000 (EST) In-Reply-To: <9705161554.ZM4222@bristol.st.com> from "Richard Ingram" at May 16, 97 03:54:31 pm Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 2301 Lines: 53 Richard wrote: > First off what is the difference between the Magnum 3000 and the RC3xxx sereis > or ar they one and the same ? The same... model numbers versus marketing names :-) > I have a Mips RC3xxx server box with 2*1.2gig and 1 exabyte streamer and will > be getting another on monday and was wondering what status the port is at to > these older Mips boxes (if one exists at all). Lots of later MIPS based boxes were based on these designs, which presumeably were licensed by MIPS to people like DEC. This means that they have always been an implicit target, with several people (including myself) keen to proceed when time permitted. David Monro has recently joined us and is throwing his efforts at the Magnums... so progress has been better in the last few weeks. > Supposedly the box has RISCix on one of the HDs already but no memory or > graphics card so I will be configuring it to be run off the seriel port. Most people here who have them seem to be in the same position. > I was going to cannibalise one of the boxes for one of the HDs to go on my > Magnum 4K box to try and get the serial kernel up in the near future (just > need memory for this as well!). The 4K box needs 72-pin *parity* SIMMS, in sets of 4. > What sort of performance did these old RC3xxx boxes give ? Does anyone still > use them ? You'll get between 30 and 40 BogoMIPS out of them, which is fine for many uses. We still use them at Softway as workhorses running RISCyOS, but they desperately need a better supported OS. RISC/OS 4.51 is stable but hasn't got dynamic libraries (much like Ultrix for the DECs), and RISC/OS 5.01 is buggy but has dynamic libraries... > I recall seeing in the past week that someone has got it booting up on a > 3020 does this mean there is hope for my RC3xxx (I'll get the exact > number next week). Yep!! Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From paul@suede.sw.oz.au Sat May 17 10:28:05 1997 Received-Date: Sat, 17 May 1997 10:28:03 +0200 (MET DST) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id IAA09927; Sat, 17 May 1997 08:27:20 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id IAA11778; Sat, 17 May 1997 08:27:19 GMT (SMI-8.6/1.34) (from paul for linux-mips@fnet.fr) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705170827.IAA11778@suede.sw.oz.au> Subject: Re: boot code stuff To: linux-mips@fnet.fr Date: Sat, 17 May 1997 18:27:19 +1000 (EST) In-Reply-To: <9705161805.AA01520@hilbert.numerik.math.uni-siegen.de> from "Michael Engel" at May 16, 97 08:05:36 pm Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 1171 Lines: 33 Michael Engel writes: > Have you already found (and fixed ;-)) the problem with the DS3100, Paul ? Yep! At least, I think so... In pmax_get_memory_size() in decstation.c, add: extern long except_vec3_generic; to the definitions at the top, and: /* Copy down exception vector handler code */ memcpy((void *)(KSEG0 + 0x80), &except_vec3_generic, 0x80); just before the comment: /* Alter trap #7 (Data Bus Error) to our own routine */ In case you were wondering: I'd cleaned up some code in boot.S, and relied on code in traps.c to do the copying, forgetting that we needed it for the PMAX memory test. Don't worry about the state things are in after this, as it all gets tidied up in trap_init(). Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From paul@suede.sw.oz.au Sun May 18 14:21:01 1997 Received-Date: Sun, 18 May 1997 14:20:59 +0200 (MET DST) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id MAA25830; Sun, 18 May 1997 12:20:24 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id MAA19392; Sun, 18 May 1997 12:20:24 GMT (SMI-8.6/1.34) (from paul for linux-mips@fnet.fr) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705181220.MAA19392@suede.sw.oz.au> Subject: DS2100/3100/5100 owners: try this kernel To: linux-mips@fnet.fr (Linux MIPS mailing list) Date: Sun, 18 May 1997 22:20:23 +1000 (EST) Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 968 Lines: 24 Hi gang, I've had a frustrating weekend with the tlb exception handlers and the mm code in general (anyone ever thought of actually documenting this rubbish... errr... I mean 'code'?) In the meantime I have uploaded a revised kernel binary, that will hopefully correctly size memory on the DS2100/3100/5100 (Hi Stu!)... please tell me if it gets further than the last one. It is: ftp://ftp.fnet.fr/linux-mips/ftp.softway.com.au/dec_vmlinux.2.1.14.test4.gz ...and should be at fnet in an hour or so. Regards, Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From amitm@amitpc.iso.dec.com Sun May 18 15:36:36 1997 Received-Date: Sun, 18 May 1997 15:36:35 +0200 (MET DST) Received: from tavosf.iso.dec.com (tavosf.iso.dec.com [16.190.64.193]) by server21.digital.fr (8.7.5/8.7) with SMTP id PAA19316; Sun, 18 May 1997 15:42:32 +0200 (MET DST) Received: from amitpc by tavosf.iso.dec.com (5.65v3.2/1.1.10.5/24Mar97-0101PM) id AA28243; Sun, 18 May 1997 16:35:40 +0300 Received: by amitpc.iso.dec.com id (Debian Smail-3.2 1996-Jul-4 #2); Sun, 18 May 1997 16:34:26 +0300 (IDT) Sender: amitm@amitpc.iso.dec.com (Amit-Margalit) Message-Id: X-Mailer: XFMail 1.2-alpha [p0] on Linux Sender: amitm@amitpc.iso.dec.com Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 8bit Mime-Version: 1.0 Date: Sun, 18 May 1997 16:33:15 +0300 (IDT) Organization: Digital Equipment Corp. From: Amit Margalit To: Paul Antoine , Linux-MIPS Mailing List Subject: The new kernel... Content-Length: 780 Lines: 21 Paul, My DECstation 5000/150 said the following on the issue of the new kernel: Found a REX compatible boot prom and then it starts booting normally until it gets to the point where it says This DECstation is a 5100/1x0 Eeeek! I don't know how to set these up! What do you need in order that it will know how to set it up correctly? Can you write a short program to read the ROM and write it to a file? Amit ------------------------------------------------------------------------ Amit Margalit - Author of XBanner: http://physics.fullerton.edu/XBanner/ ------------------------------------------------------------------------ All information in the above mail message is personal, and does not imply on my employer. None of it should be attributed to my employer. From allens@vivanet.com Mon May 19 13:46:33 1997 Received-Date: Mon, 19 May 1997 13:46:32 +0200 (MET DST) Received: from shell10.vivanet.com (allens@shell10.vivanet.com [192.108.245.8]) by vivanet.com (8.7.5/8.7.3) with ESMTP id HAA03970 for ; Mon, 19 May 1997 07:53:36 -0400 (EDT) From: Stu Allen Received: (from allens@localhost) by shell10.vivanet.com (8.7.5/8.7.3) id HAA20995 for linux-mips@fnet.fr; Mon, 19 May 1997 07:52:49 -0400 (EDT) Message-Id: <199705191152.HAA20995@shell10.vivanet.com> Subject: Re: DS2100/3100/5100 owners: try this kernel To: linux-mips@fnet.fr Date: Mon, 19 May 1997 07:52:49 -0400 (EDT) In-Reply-To: <199705181220.MAA19392@suede.sw.oz.au> from "Paul Antoine" at May 18, 97 10:20:23 pm X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 706 Lines: 15 > > In the meantime I have uploaded a revised kernel binary, that will hopefully > correctly size memory on the DS2100/3100/5100 (Hi Stu!)... please tell me > if it gets further than the last one. It is: > Well, sadly it does not. It gets as far as the version banner, and then it blows up with a stack dump. Hmmm. I didn't think to check if it was the same or different as the last one. Please let me know if that would be useful information and I'll check tonight. What I really ought to do is scrounge up a cable so I can hook my laptop up as the console! Hopefully sometime this week I'll be able to build a new Linux (i386) box so I can start taking a look at the current source tree. Stu From khan@mogli.neuroma.uni-heidelberg.de Tue May 20 12:20:56 1997 Received-Date: Tue, 20 May 1997 12:20:54 +0200 (MET DST) Received: by mogli.neuroma.uni-heidelberg.de (940816.SGI.8.6.9/940406.SGI.AUTO) for linux-mips@fnet.fr id MAA10694; Tue, 20 May 1997 12:19:59 +0200 From: khan@mogli.neuroma.uni-heidelberg.de (Jens Beyer) Message-Id: <199705201019.MAA10694@mogli.neuroma.uni-heidelberg.de> Subject: subscribe To: linux-mips@fnet.fr Date: Tue, 20 May 1997 12:19:50 +0200 (MDT) X-Mailer: ELM [version 2.4 PL24 PGP6] Content-Type: text Content-Length: 171 Lines: 9 subscribe khan@mogli.neuroma.uni-heidelberg.de -- Nur tote Fische schwimmen mit dem Strom khan@mogli.neuroma.uni-heidelberg.de From rji@bristol.st.com Tue May 20 14:52:46 1997 Received-Date: Tue, 20 May 1997 14:52:44 +0200 (MET DST) Received: by hawkweed id NAA27435; Tue, 20 May 1997 13:52:46 +0100 (BST) From: "Richard Ingram" Message-Id: <9705201352.ZM28018@bristol.st.com> Date: Tue, 20 May 1997 13:52:39 +0100 In-Reply-To: paul@suede.sw.oz.au (Paul Antoine) "Re: Linux on MIPS RC3xxx boxes ?" (May 17, 9:00am) References: <199705162300.XAA09922@suede.sw.oz.au> X-Mailer: Z-Mail (3.2.1 10apr95) To: linux-mips@fnet.fr Subject: Re: Linux on MIPS RC3xxx boxes ? Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Length: 1083 Lines: 25 > Lots of later MIPS based boxes were based on these designs, which presumeably > were licensed by MIPS to people like DEC. This means that they have always > been an implicit target, with several people (including myself) keen to > proceed when time permitted. David Monro has recently joined us and is > throwing his efforts at the Magnums... so progress has been better in the > last few weeks. Ok got my second RC3330 today, both are memory less, they take 30 pin SIMMS ala Sun3/60 I believe ? What is the bank arrangement/min memory using 1MB SIMMS ? Any special jumpers to connect ? I also got a second 3/60 so one of them can be cannibalised to try and get a serial version of Linux up on the RC3330 (suppose the only way is via floppy at the moment ??). A general Linux/Mips query now, are any offboard Ethernet cards supported yet ? I have a number of DEC DEPCA? Ethernet cards (which have Linux drivers in the x86 Linux world) which I am using in my PC and would want to use thin ethernet on my Magnum 4000 and RC3330's ? Any hope ? Thanks for your time, Rich. -- From linux@informatik.uni-koblenz.de Tue May 20 17:22:47 1997 Received-Date: Tue, 20 May 1997 17:21:32 +0200 (MET DST) Received: from thoma (linux@thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.8.5/8.6.9) with SMTP id RAA20973 for ; Tue, 20 May 1997 17:21:12 +0200 (MEST) From: Systemkennung Linux Message-Id: <199705201521.RAA20973@informatik.uni-koblenz.de> Received: by thoma (SMI-8.6/KO-2.0) id RAA12273; Tue, 20 May 1997 17:21:11 +0200 Subject: Re: Linux on MIPS RC3xxx boxes ? To: linux-mips@fnet.fr Date: Tue, 20 May 1997 17:21:11 +0200 (MET DST) In-Reply-To: <9705201352.ZM28018@bristol.st.com> from "Richard Ingram" at May 20, 97 01:52:39 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 910 Lines: 22 Hi, > A general Linux/Mips query now, are any offboard Ethernet cards supported > yet? I have a number of DEC DEPCA? Ethernet cards (which have Linux > drivers in the x86 Linux world) which I am using in my PC and would want > to use thin ethernet on my Magnum 4000 and RC3330's ? Any hope ? Under the standard PC network cards we have three types: - PIO cards. All of these are supposed to work. I have been using a NE2000 clone for a long time. - Most shared memory cards should be working also. Not all because some drivers use the memory_* functions in the wrong way which works on Alpha and Intel but not MIPS. - DMA cards. All of these need some modifications to deal with the CPU's cache coherency. (It would be easy to implement some kind of generic kludge in kmalloc() and __get_free_pages(). For this class of drivers but I don't like that idea.) Ralf From ralf@informatik.uni-koblenz.de Wed May 21 04:55:42 1997 Received-Date: Wed, 21 May 1997 04:55:40 +0200 (MET DST) Received: from thoma (ralf@thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.8.5/8.6.9) with SMTP id EAA22462; Wed, 21 May 1997 04:51:41 +0200 (MEST) From: Ralf Baechle Message-Id: <199705210251.EAA22462@informatik.uni-koblenz.de> Received: by thoma (SMI-8.6/KO-2.0) id EAA13671; Wed, 21 May 1997 04:51:39 +0200 Subject: Binutils for IRIX To: linux-mips@fnet.fr Date: Wed, 21 May 1997 04:51:38 +0200 (MET DST) MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 709 Lines: 22 -----BEGIN PGP SIGNED MESSAGE----- Hi all, I'm uploading crosscompiler binaries of the GNU binutils 2.7 for little and big endian targets to ftp.fnet.fr and kernel.panic.julia.de. The binaries have been built on IRIX 6.2 for the o32 ABI. Ralf 85ec2b4e9e059c6ba0e0a1bcddc2a5ec mips-sgi-irix6.2/mips-linux/binutils-2.7-4.tar.gz 51ac173c805d82fba82c430f094787ff mips-sgi-irix6.2/mipsel-linux/binutils-2.7-4.tar.gz -----BEGIN PGP SIGNATURE----- Version: 2.6.2 iQCVAgUBM4JjN0ckbl6vezDBAQEGyAP/dGtICrAiuGfixDtFQaxj2U2Trt7chfEH rMMQhnRI7DQ8Oc8Qa97hrJVTmvLsHns8KiO4RI3am0af0z2MTFV2GCMR46f2+sR5 IwlreIDh+v0l8tm5cDDOV1xg1zI0fVMXnC7w20o7yeJn/SnM6WeA6Fpus/EsJoAa Q8VIAd5qS90= =YKhT -----END PGP SIGNATURE----- From paul@suede.sw.oz.au Wed May 21 04:55:41 1997 Received-Date: Wed, 21 May 1997 04:55:39 +0200 (MET DST) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id CAA04774; Wed, 21 May 1997 02:55:24 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id CAA19973; Wed, 21 May 1997 02:55:23 GMT (SMI-8.6/1.34) (from paul for linux-mips@fnet.fr) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705210255.CAA19973@suede.sw.oz.au> Subject: Re: Linux on MIPS RC3xxx boxes ? To: linux-mips@fnet.fr Date: Wed, 21 May 1997 12:55:22 +1000 (EST) In-Reply-To: <9705201352.ZM28018@bristol.st.com> from "Richard Ingram" at May 20, 97 01:52:39 pm Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 1063 Lines: 29 Rich wrote: > Ok got my second RC3330 today, both are memory less, they take 30 > pin SIMMS ala Sun3/60 I believe ? Yep, in sets of 8. > What is the bank arrangement/min memory using 1MB SIMMS ? Any > special jumpers to connect? No jumpers, just fill the correct banks in sequence. I think you fill the bank in front first, then the one in back, then the one in front again, and so on. > I also got a second 3/60 so one of them can be cannibalised to try > and get a serial version of Linux up on the RC3330 (suppose the only > way is via floppy at the moment ??). No - network boot only at the moment. Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From frist@lehr.chem.tu-berlin.de Wed May 21 09:03:53 1997 Received-Date: Wed, 21 May 1997 09:03:52 +0200 (MET DST) Received: (from frist@localhost) by lehr.chem.tu-berlin.de (8.7.4/8.7.3) id JAA04181; Wed, 21 May 1997 09:03:49 +0200 Date: Wed, 21 May 1997 09:03:49 +0200 (MET DST) From: Frieder Streffer To: linux-mips@fnet.fr Subject: Re: DS2100/3100/5100 owners: try this kernel In-Reply-To: <199705181220.MAA19392@suede.sw.oz.au> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 811 Lines: 28 On Sun, 18 May 1997, Paul Antoine wrote: > Hi gang, > > I've had a frustrating weekend with the tlb exception handlers and the mm > code in general (anyone ever thought of actually documenting this rubbish... > errr... I mean 'code'?) > > In the meantime I have uploaded a revised kernel binary, that will hopefully > correctly size memory on the DS2100/3100/5100 (Hi Stu!)... please tell me > if it gets further than the last one. It is: > > ftp://ftp.fnet.fr/linux-mips/ftp.softway.com.au/dec_vmlinux.2.1.14.test4.gz > > ...and should be at fnet in an hour or so. > > Regards, > Paul Hi Paul, I had a quick look at the source yesterday evening. I seams that th problem is in the memory sizing routine. I will write a analog to the rex routine until i'm alble to fix the bug. regards, Frieder From knobi@munich.sgi.com Wed May 21 09:53:33 1997 Received-Date: Wed, 21 May 1997 09:53:32 +0200 (MET DST) Received: from odin.corp.sgi.com (fddi-odin.corp.sgi.com [198.29.75.194]) by deliverator.sgi.com (950413.SGI.8.6.12/951211.SGI.AUTO) via ESMTP id AAA22188; Wed, 21 May 1997 00:52:02 -0700 env-from (knobi@munich.sgi.com) Received: from sgiger.munich.sgi.com by odin.corp.sgi.com via ESMTP (951211.SGI.8.6.12.PATCH1502/951211.SGI) id AAA23343; Wed, 21 May 1997 00:51:46 -0700 Received: from knobi.munich.sgi.com by sgiger.munich.sgi.com via ESMTP (951211.SGI.8.6.12.PATCH1502/940406.SGI) id JAA22540; Wed, 21 May 1997 09:51:19 +0200 Received: from knobi (localhost [127.0.0.1]) by knobi.munich.sgi.com (950413.SGI.8.6.12/951220.SGI.AUTOCF.knobi) via SMTP id JAA20009; Wed, 21 May 1997 09:51:06 +0200 Sender: knobi@munich.sgi.com Message-ID: <3382A96A.59E2@munich.sgi.com> Date: Wed, 21 May 1997 09:51:06 +0200 From: Martin Knoblauch Organization: Silicon Graphics GmbH, Am-Hochacker 3, D-85630 Grasbrunn X-Mailer: Mozilla 3.01SC-SGI (X11; I; IRIX 6.3 IP22) MIME-Version: 1.0 To: Ralf Baechle CC: linux-mips@fnet.fr Subject: Re: Binutils for IRIX References: <199705210251.EAA22462@informatik.uni-koblenz.de> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Length: 1462 Lines: 38 Ralf Baechle wrote: > > -----BEGIN PGP SIGNED MESSAGE----- > > Hi all, > > I'm uploading crosscompiler binaries of the GNU binutils 2.7 > for little and big endian targets to ftp.fnet.fr and > kernel.panic.julia.de. The binaries have been built on > IRIX 6.2 for the o32 ABI. > > Ralf > Ralf, due to problems with GNU support vs. IRIX 6.x, I would advise that one should configure all packages that deal with binary code for mips-sgi-irix-5.3 (--host, --target and --build). This makes life a bit easier. See this URL for a bit more of the gory details. The problems will hopefully be resolved in future versions of binutils and gcc. http://reality.sgi.com/knobi/gcc-2.7.2.x-on-irix-6.2-6.3 Built for 5.3, you should be able to put the stuff on a 5.3 build system as an additional benfit. Martin -- +---------------------------------+-----------------------------------+ |Martin Knoblauch | Silicon Graphics GmbH | |Manager Technical Marketing | Am Hochacker 3 - Technopark | |Silicon Graphics Computer Systems| D-85630 Grasbrunn-Neukeferloh, FRG| |---------------------------------| Phone: (+int) 89 46108-179 or -0 | |http://reality.sgi.com/knobi | Fax: (+int) 89 46107-179 | +---------------------------------+-----------------------------------+ |e-mail: | VM: 6-333-8197 | M/S: IDE-3150 | +---------------------------------------------------------------------+ From ralf@informatik.uni-koblenz.de Wed May 21 11:43:08 1997 Received-Date: Wed, 21 May 1997 11:42:22 +0200 (MET DST) Received: from thoma (ralf@thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.8.5/8.6.9) with SMTP id LAA13455; Wed, 21 May 1997 11:41:31 +0200 (MEST) From: Ralf Baechle Message-Id: <199705210941.LAA13455@informatik.uni-koblenz.de> Received: by thoma (SMI-8.6/KO-2.0) id LAA14335; Wed, 21 May 1997 11:41:30 +0200 Subject: Re: Binutils for IRIX To: knobi@munich.sgi.com (Martin Knoblauch) Date: Wed, 21 May 1997 11:41:29 +0200 (MET DST) Cc: ralf@informatik.uni-koblenz.de, linux-mips@fnet.fr In-Reply-To: <3382A96A.59E2@munich.sgi.com> from "Martin Knoblauch" at May 21, 97 09:51:06 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1166 Lines: 29 > Ralf, > > due to problems with GNU support vs. IRIX 6.x, I would advise > that one should configure all packages that deal with binary > code for mips-sgi-irix-5.3 (--host, --target and --build). > This makes life a bit easier. See this URL for a bit more > of the gory details. The problems will hopefully be resolved > in future versions of binutils and gcc. > > http://reality.sgi.com/knobi/gcc-2.7.2.x-on-irix-6.2-6.3 > > Built for 5.3, you should be able to put the stuff on a > 5.3 build system as an additional benfit. The way I've configured the tools the IRIX 6.2 binaries should also run under IRIX 5.3. The problems with GNU vs. IRIX is mostly only related to IRIX 6.x as target. But entire thing is a crosscompiler, so the only problem I did encounter was that GCC assumes on IRIX 6.x compile hosts that sizeof(long) == 8. I modified the GCC sources to build with all ABIs on IRIX. Just to be on the safe side - could anyone running a "real" IRIX 5.3 system test the binaries? Thanks. I've build GCC packages in the meantime also but I don't have time to upload them until the weekend - I'll be on the Linux Kongress for that long. Ralf From sayer@eclipse.nsp Wed May 21 19:38:08 1997 Received-Date: Wed, 21 May 1997 19:38:06 +0200 (MET DST) Received: from imaginet.fr by moria.imaginet.fr via ESMTP (950215.SGI.8.6.10/911001.SGI) for id TAA06732; Wed, 21 May 1997 19:38:24 +0200 Received: from eclipse.nsp (cyber132-3.paris.imaginet.fr [195.68.3.132]) by imaginet.fr (8.7.5/8.7.31) with ESMTP id TAA28225 for ; Wed, 21 May 1997 19:37:33 +0200 (METDST) Received: from eclipse (localhost [127.0.0.1]) by eclipse.nsp (8.8.4/8.8.4) with ESMTP id TAA01010 for ; Wed, 21 May 1997 19:37:27 +0200 Message-Id: <199705211737.TAA01010@eclipse.nsp> X-Mailer: exmh version 1.6.7 05/05/96 To: linux-mips@fnet.fr From: Nicolas.Sayer@inria.fr Reply-To: Nicolas.Sayer@inria.fr Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 21 May 1997 19:37:27 +0200 Sender: sayer@eclipse.nsp Content-Length: 31 Lines: 1 I want to subsribe to the list From cdemps@lsumc.edu Wed May 21 20:42:00 1997 Received-Date: Wed, 21 May 1997 20:41:58 +0200 (MET DST) Received: from [155.58.112.211] by sarah.ocs.lsumc.edu (post.office MTA v2.0 0813 ID# 0-11395) with SMTP id AAA108 for ; Wed, 21 May 1997 13:41:49 -0500 Message-ID: <33834355.7505@lsumc.edu> Date: Wed, 21 May 1997 13:47:49 -0500 From: cdemps@lsumc.edu (Dempsey, Charles) Reply-To: cdemps@lsumc.edu Organization: Dept Pharmacology, LSUMC-S X-Mailer: Mozilla 2.02 (Macintosh; I; PPC) MIME-Version: 1.0 To: linux-mips@fnet.fr Subject: Mips Magnum SCSI. Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Length: 294 Lines: 9 Does anyone know what version of the NCR53C9X SCSI chip is in the MIPS Magnum 4000 PC? Also, I looked inside my Magnum 4000 SC and I'm not finding an NCR chip. The only thing that looks like a controller chip is a Western Digital WD16C552-JT and an Emulex 2400119 FAS216. Thanks, Charles From ryan@ravage.labs.gmu.edu Wed May 21 21:23:14 1997 Received-Date: Wed, 21 May 1997 21:23:13 +0200 (MET DST) Received: from localhost (ryan@localhost) by ravage.labs.gmu.edu (8.8.3/8.8.3) with SMTP id PAA10273 for ; Wed, 21 May 1997 15:24:55 -0400 Date: Wed, 21 May 1997 15:24:55 -0400 (EDT) From: Ryan Rafferty To: linux-mips@fnet.fr Subject: Re: Magnum SCSI (fwd) Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 768 Lines: 24 Date: Wed, 21 May 1997 13:35:49 -0500 From: "Dempsey, Charles" Ryan Rafferty wrote: > > I posted a request for help with writing a SCSI driver for the NCR 53c90 > chipset in the Magnum, and someone replied that the Sun ESP driver has > support for that chipset. I looked at esp.h, and it does mention the > 53c90. Does anyone know if there's a reason that driver couldn't be made > to work with the MIPS port? Hmmm. Line 530 of esp.c in the linux-mips-2.0.21 kernel says case esp100: return "Sparc ESP100 (NCR53C90)"; That looks like a start. I wonder if the chip in the Magnum 4000 is the 53C90 or another rev. I looked in my 4000 SC and it's got a Western Digital chip. Ryan, please relay this message to linux-mips@fnet.fr. Charles From hub-fue!alpha.franken.de!tsbogend@rachael.franken.de Thu May 22 01:03:54 1997 Received-Date: Thu, 22 May 1997 01:03:53 +0200 (MET DST) Received: from hub-fue by rachael.franken.de with uucp (Smail3.1.29.1 #8) id m0wUKPP-000oYeC; Thu, 22 May 97 01:03 MET DST Received: by hub-fue.franken.de (Smail3.1.29.1 #35) id m0wUKPC-002OcJC; Thu, 22 May 97 01:02 MET DST Received: (from tsbogend@localhost) by alpha.franken.de (8.8.5/8.8.5) id AAA03029 for linux-mips@fnet.fr; Thu, 22 May 1997 00:53:41 +0200 From: Thomas Bogendoerfer Message-Id: <199705212253.AAA03029@alpha.franken.de> Subject: Re: Mips Magnum SCSI. In-Reply-To: <33834355.7505@lsumc.edu> from "Dempsey, Charles" at "May 21, 97 01:47:49 pm" To: linux-mips@fnet.fr Date: Thu, 22 May 1997 00:53:41 +0200 (MEST) X-Mailer: ELM [version 2.4ME+ PL27 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 475 Lines: 13 > Also, I looked inside my Magnum 4000 SC and I'm not finding an NCR chip. > The only thing that looks like a controller chip is a Western Digital > WD16C552-JT and an Emulex 2400119 FAS216. the Western chip is dual 16550 UART. The Emulex is a somehow improved NCR53C94. Thomas. -- See, you not only have to be a good coder to create a system like Linux, you have to be a sneaky bastard too ;-) [Linus Torvalds in <4rikft$7g5@linux.cs.Helsinki.FI>] From ryan@ravage.labs.gmu.edu Thu May 22 05:03:41 1997 Received-Date: Thu, 22 May 1997 05:03:40 +0200 (MET DST) Received: from localhost (ryan@localhost) by ravage.labs.gmu.edu (8.8.3/8.8.3) with SMTP id XAA16080 for ; Wed, 21 May 1997 23:05:22 -0400 Date: Wed, 21 May 1997 23:05:22 -0400 (EDT) From: Ryan Rafferty To: linux-mips@fnet.fr Subject: Shablamm In-Reply-To: <199705212253.AAA03029@alpha.franken.de> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 1165 Lines: 22 While I'm anticipating the arrival of the ShaBLAMM card I bought, my mind has been flying through ideas of how to make it useful. I'm sure NT will work fine, but Linux will probably take a bit of work to boot. The primary problem that comes to my mind is how to figure out which signals the 486 board has to send to the Nitro in order for it to become active. After that, I'll need to figure out how the board communicates with the bus and the host processor (some esoteric configurations spring to my mind--since I own a 486 VLB/PCI combo board I might try fitting the Nitro into the VLB slot on it and seeing if it can see the PCI bus... but that's a bit in the future =). I'm assuming the 486 remains active once the Nitro "takes over" the motherboard, and that the Nitro actually accesses the bus by sending commands to the 486--that would make to most sense from little I know, but it could be 100% wrong... Anyway, I apologize for wasting mailbox space with my mental wanderings, but I'd appreciate any ideas people might have to "sniff" the bus while booting NT in order to see what signals are being sent to the Nitro. Is mise le meas, Ryan Rafferty From dom@gladsmuir.algor.co.uk Thu May 22 08:43:05 1997 Received-Date: Thu, 22 May 1997 08:43:04 +0200 (MET DST) Received: from gladsmuir.algor.co.uk (dom@gladsmuir.algor.co.uk [193.117.190.129]) by embankment.algor.co.uk (8.8.5/8.8.5) with ESMTP id HAA16013; Thu, 22 May 1997 07:42:55 +0100 (BST) Received: (from dom@localhost) by gladsmuir.algor.co.uk (8.7.3/8.7.3) id HAA00213; Thu, 22 May 1997 07:43:55 +0100 (BST) Date: Thu, 22 May 1997 07:43:55 +0100 (BST) From: Dom Sweetman Message-Id: <199705220643.HAA00213@gladsmuir.algor.co.uk> To: linux-mips@fnet.fr Subject: Re: Shablamm In-Reply-To: References: <199705212253.AAA03029@alpha.franken.de> Content-Length: 1199 Lines: 32 Ryan, > While I'm anticipating the arrival of the ShaBLAMM card I bought, my mind > has been flying through ideas of how to make it useful. I'm sure NT will > work fine, but Linux will probably take a bit of work to boot. It's most likely hard-wired little-endian. > The primary problem that comes to my mind is how to figure out which > signals the 486 board has to send to the Nitro in order for it to become > active. After that, I'll need to figure out how the board communicates > with the bus and the host processor... Unreliable memories of unreliable conversations long ago... I'm pretty sure Shablamm!! is ROM-less. From power up it looks like a chunk of memory which the x86 host fills with a program and then toggles a register bit to unleash the MIPS. The idea was you got a program for the PC which became a sort of I/O server. I *think* they drove the video directly (assuming the video controller to be on VLB too) - so you may have to find a supported board; but the host processor ran the disk, keyboard, mouse etc. It will be interesting to see whether you get the NT software; if you do it should have all the host stuff. Good luck Dominic Sweetman dom@algor.co.uk From digitech!digitech.co.nz!arnim@atlantis.actrix.gen.nz Wed May 21 22:42:11 1997 Received-Date: Wed, 21 May 1997 22:42:08 +0200 (MET DST) Received: (from uudigi@localhost) by atlantis.actrix.gen.nz (4.3.7/4.3.5) id IAA20281 for linux-mips@fnet.fr; Thu, 22 May 1997 08:41:51 +1200 (NZST) X-Authentication-Warning: atlantis.actrix.gen.nz: uudigi set sender to digitech!digitech.co.nz!arnim using -f >Received: by digitech.co.nz (UUPC/extended 1.12r); Thu, 22 May 1997 08:23:43 +0100 Message-ID: <3383f47f.digitech@digitech.co.nz> Date: Thu, 22 May 1997 08:23:42 +0100 From: "Arnim Littek" Organization: Digi-Tech Ltd, Wellington, New Zealand Reply-To: "Arnim Littek" To: linux-mips@fnet.fr Subject: WD16C552 Content-Type: text Content-Length: 298 Lines: 7 I don't have the data sheet in front of me, but the Western Digital IC, the WD16C552, is almost certainly not going to be a SCSI controller... That part looks suspiciously like a UART/parallel port clone combo, ie. a takeoff on the NatSemi NS16C552. Arnim. -- From ryan@ravage.labs.gmu.edu Fri May 23 03:45:35 1997 Received-Date: Fri, 23 May 1997 03:45:34 +0200 (MET DST) Received: from localhost (ryan@localhost) by ravage.labs.gmu.edu (8.8.3/8.8.3) with SMTP id VAA23040 for ; Thu, 22 May 1997 21:47:28 -0400 Date: Thu, 22 May 1997 21:47:28 -0400 (EDT) From: Ryan Rafferty To: linux-mips@fnet.fr Subject: Working FS In-Reply-To: <199705212253.AAA03029@alpha.franken.de> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 416 Lines: 11 Of the people on this list, may I ask who has a functioning Linux/MIPS machine that can boot to a shell prompt (other than Ralf)? I'd like to compare versions of important packages like libc, fileutils, bash, getty, and the like, and if I could, inquire what upgrade path you took from the outdated filesystem image so that I can make a sort of mini-howto for those of us trying to jump in. Thanks, Ryan Rafferty From elias@scruznet.com Fri May 23 09:28:51 1997 Received-Date: Fri, 23 May 1997 09:28:50 +0200 (MET DST) Received: from [205.179.228.11] ([205.179.228.11]) by scruz.net (8.8.5/1.34) with SMTP id AAA02477 for ; Fri, 23 May 1997 00:28:46 -0700 (PDT) Message-Id: <199705230728.AAA02477@scruz.net> Subject: Compiling Date: Fri, 23 May 97 00:27:50 -0700 x-mailer: Claris Emailer 2.0, March 15, 1997 From: Elias Kesh To: "Mip" Mime-Version: 1.0 Content-Type: text/plain; charset="US-ASCII" Content-Length: 551 Lines: 16 A couple of quick questions about compiling. I have been trying to get Apache, qpopper and DNS to cross compile. The native compile for intel and PPC both work fine but the cross compile to mips generates more header errors than I can keep track of. I am making the assuption that I have installed the cross compiler (and/or includes) incorrectly even though it builds the kernel just fine. Has anyone else managed to get Apache to build and if so what configuration ? gcc 2.7.2-1, libc-951212, and binutils 2.6-1 Elias kesh Elias@kesh.com From ctrl@tatung.math.unibuc.ro Sun May 25 11:07:12 1997 Received-Date: Sun, 25 May 1997 11:07:08 +0200 (MET DST) Received: from sandwich (ctrl@localhost [127.0.0.1]) by sandwich.math.unibuc.ro (8.8.5/8.8.0) with SMTP id TAA13190 for ; Sat, 24 May 1997 19:38:04 +0300 Sender: ctrl@sandwich.math.unibuc.ro Message-ID: <3387196C.60B53911@tatung.math.unibuc.ro> Date: Sat, 24 May 1997 19:38:04 +0300 From: Cornel Ciocirlan Organization: Universitatea Bucuresti X-Mailer: Mozilla 3.01 (X11; I; Linux 2.0.0 i586) MIME-Version: 1.0 To: linux-mips@fnet.fr Subject: Hello to MIPS linux ! Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Length: 402 Lines: 11 Hello. I have a really old box here - A Control Data CD4330 based on MIPS R 3000. I want to know if there is any chance this machine will be supported (or if I can help to that). I am a student from Romania, and sysadmin at the Bucharest University. The machine I'm talking about in skylab.math.unibuc.ro, the operating system is EP/IX 1.4.3 - actually MIPS RISCos 4.0. Regards, Cornel Ciocirlan. From ralf@informatik.uni-koblenz.de Sun May 25 18:16:45 1997 Received-Date: Sun, 25 May 1997 18:16:41 +0200 (MET DST) Received: from thoma (ralf@thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.8.5/8.6.9) with SMTP id SAA24974; Sun, 25 May 1997 18:12:40 +0200 (MEST) From: Ralf Baechle Message-Id: <199705251612.SAA24974@informatik.uni-koblenz.de> Received: by thoma (SMI-8.6/KO-2.0) id SAA24302; Sun, 25 May 1997 18:12:39 +0200 Subject: Cross-GCC for IRIX 6.2 To: linux-mips@fnet.fr Date: Sun, 25 May 1997 18:12:38 +0200 (MET DST) MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 562 Lines: 21 -----BEGIN PGP SIGNED MESSAGE----- Hi, I've uploaded cross-compiler binaries for IRIX 6.2 to ftp.fnet.fr and kernel.panic.julia.de. Ralf d27482ba58b2543d4bda756b180da1d1 mips-linux/gcc-2.7.2-4.tar.gz ce3e8931cef63ef4c87e2c78acddff2b mipsel-linux/gcc-2.7.2-4.tar.gz -----BEGIN PGP SIGNATURE----- Version: 2.6.2 iQCVAgUBM4hk9Eckbl6vezDBAQEM6AP+I7SauhXfxrDxXedHht6uAGnIG06e1bbS /6xKlIJl0Hj689LWudZ5/+U4zsC8v/cy0RlKG/A3N/qN7BOViYQMTfB/yqlS1oLE LjAggWgQkQ6jiigCLNW6SSzO7lqtEBKOmUVmdYB4OgyLTcZKpIDIGfMExqsWYEm9 C/Tuz6fuaZ8= =K/nI -----END PGP SIGNATURE----- From linux@informatik.uni-koblenz.de Mon May 26 11:21:51 1997 Received-Date: Mon, 26 May 1997 11:21:37 +0200 (MET DST) Received: from ozzy (linux@ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.8.5/8.6.9) with SMTP id LAA01882 for ; Mon, 26 May 1997 11:20:43 +0200 (MEST) From: Systemkennung Linux Message-Id: <199705260920.LAA01882@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id LAA11560; Mon, 26 May 1997 11:20:43 +0200 Subject: Re: Compiling To: linux-mips@fnet.fr Date: Mon, 26 May 1997 11:20:42 +0200 (MET DST) In-Reply-To: <199705230728.AAA02477@scruz.net> from "Elias Kesh" at May 23, 97 00:27:50 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1191 Lines: 27 > A couple of quick questions about compiling. I have been trying to get > Apache, qpopper and DNS to cross compile. The native compile for intel > and PPC both work fine but the cross compile to mips generates more > header errors than I can keep track of. > I am making the assuption that I have installed the cross compiler > (and/or includes) incorrectly even though it builds the kernel just fine. > Has anyone else managed to get Apache to build and if so what > configuration ? Block port 80 ;-) > gcc 2.7.2-1, libc-951212, and binutils 2.6-1 Just to also send an answer to the list - you should really should upgrade to gcc 2.7.2-3 and binutils-2.7-3. Binutils 2.6 can not handle PIC code, so don't even dream about using it for userland stuff. I always wanted to fix two bugs before releasing the sources and binaries of GNU libc 2.0.2 to the net but never came around, so I'll release them with the bugs later this day: - with GNU libc 2.0 and newer make seems to loose file descriptors. When running big jobs it may run out of file descriptors. - the startup code passes a bad argv[0] pointer. This is the return of a bug fixed long time ago ... Ralf From linux@informatik.uni-koblenz.de Mon May 26 11:29:50 1997 Received-Date: Mon, 26 May 1997 11:29:39 +0200 (MET DST) Received: from ozzy (linux@ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.8.5/8.6.9) with SMTP id LAA02536 for ; Mon, 26 May 1997 11:29:13 +0200 (MEST) From: Systemkennung Linux Message-Id: <199705260929.LAA02536@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id LAA11649; Mon, 26 May 1997 11:29:13 +0200 Subject: Re: Magnum SCSI (fwd) To: linux-mips@fnet.fr Date: Mon, 26 May 1997 11:29:13 +0200 (MET DST) In-Reply-To: from "Ryan Rafferty" at May 21, 97 03:24:55 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 655 Lines: 19 Hi, > Hmmm. Line 530 of esp.c in the linux-mips-2.0.21 kernel says > > case esp100: > return "Sparc ESP100 (NCR53C90)"; ESP is Sparc language. MIPS calls the same chip ASC which stands for Advanced SCSI Controller. > That looks like a start. I wonder if the chip in the Magnum 4000 is the 53C90 or > another rev. I looked in my 4000 SC and it's got a Western Digital chip. In the various members of the Jazz family if so far seen 53C94 and FAS216 chips. We might have to deal with other chips also but using esp.c as base should cover almost all if not all of these cases, so probably all you need to take care of is the Jazz DMA engine. Ralf From paul@suede.sw.oz.au Tue May 27 00:48:28 1997 Received-Date: Tue, 27 May 1997 00:48:26 +0200 (MET DST) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id WAA12259; Mon, 26 May 1997 22:47:56 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id WAA20192; Mon, 26 May 1997 22:47:52 GMT (SMI-8.6/1.34) (from paul for linux-mips@fnet.fr) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705262247.WAA20192@suede.sw.oz.au> Subject: Re: Hello to MIPS linux ! To: linux-mips@fnet.fr Date: Tue, 27 May 1997 08:47:52 +1000 (EST) In-Reply-To: <3387196C.60B53911@tatung.math.unibuc.ro> from "Cornel Ciocirlan" at May 24, 97 07:38:04 pm Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 1103 Lines: 23 Cornel wrote: > I have a really old box here - A Control Data CD4330 based on MIPS R > 3000. I want to know if there is any chance this machine will be > supported (or if I can help to that). I am a student from Romania, > and sysadmin at the Bucharest University. The machine I'm talking > about in skylab.math.unibuc.ro, the operating system is EP/IX 1.4.3 > - actually MIPS RISCos 4.0. I think that Control Data re-sold MIPS machines (in much the same way Tandem resells SGI's now)... so it should just be a 3330 or 3230?? In which case the work by David Monro will help you - you probably need to convince David to put up a web page about it. Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From ralf@informatik.uni-koblenz.de Tue May 27 03:49:17 1997 Received-Date: Tue, 27 May 1997 03:49:14 +0200 (MET DST) Received: from thoma (ralf@thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.8.5/8.6.9) with SMTP id DAA13315 for ; Tue, 27 May 1997 03:48:58 +0200 (MEST) From: Ralf Baechle Message-Id: <199705270148.DAA13315@informatik.uni-koblenz.de> Received: by thoma (SMI-8.6/KO-2.0) id DAA02543; Tue, 27 May 1997 03:48:57 +0200 Subject: GNU libc sources To: linux-mips@fnet.fr Date: Tue, 27 May 1997 03:48:56 +0200 (MET DST) MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1036 Lines: 33 -----BEGIN PGP SIGNED MESSAGE----- Hi all, I'm just uploading the sources of GNU libc 2.0.1. As I've already announced before there are two capital bugs left in this version of the library: - argv[0] is set wrong by the startup code. - With this library GNU make will loose file descriptors. I should mention that due to various incompatible changes older libraries won't work to your satisfaction anymore with current kernels. All the RPM and tar binary packages needs binaries of the libraries built from this source. This source tree also contains the available add-ons including the DES crypt add-on. So take care if your one of these crypto-dictatorship countries ... Ralf 0e02adb25a35990b4e0e30720c7f8c5d glibc-2.0.1-1.tar.gz -----BEGIN PGP SIGNATURE----- Version: 2.6.2 iQCVAgUBM4o9hkckbl6vezDBAQGesAQAqiRZObTiSrNoevVbxQlgUjP7aeWgToPl TT4SBYkGJpwIiitZQJga7+SeHXloW1PoO449kK9/JkEoEZb0OPL/EGpE4gqr6Ptr Qybcj7aP1ECJdppmsDVe8wjOOQ6B8t97NWiuSuPQpd+I1D/bl594gXNNGjDPkREa ECQH2mKxHds= =clMl -----END PGP SIGNATURE----- From stuarta@foxln.com.au Tue May 27 04:01:13 1997 Received-Date: Tue, 27 May 1997 04:01:09 +0200 (MET DST) Received: from foxln.com.au(really [202.12.114.80]) by foxln.com.au via smail with esmtp id for ; Tue, 27 May 1997 12:00:33 +1000 (EST) (Smail-3.2 1996-Jul-4 #1 built 1997-May-19) Received: by foxln.com.au via send-mail with stdio id for linux-mips@fnet.fr; Tue, 27 May 1997 11:59:53 +1000 (EST) (Smail-3.2 1996-Jul-4 #2 built 1997-May-22) Message-Id: Date: Tue, 27 May 1997 11:59:53 +1000 (EST) From: stuarta@foxln.com.au (Stuart Auchterlonie) To: linux-mips@fnet.fr Subject: Re: GNU libc sources X-Sun-Charset: US-ASCII Content-Length: 296 Lines: 13 > > Hi all, > > I'm just uploading the sources of GNU libc 2.0.1. As I've already > announced before there are Ralf, Is there any reason why we aren't heading straight for glibc-2.0.3 which appears to be the latest version from GNU. It apparently has lots and lots of bugs fixed... Stuart From elias@scruznet.com Tue May 27 04:20:07 1997 Received-Date: Tue, 27 May 1997 04:20:05 +0200 (MET DST) Received: from [205.179.228.11] ([205.179.228.11]) by scruz.net (8.8.5/1.34) with SMTP id TAA21364 for ; Mon, 26 May 1997 19:20:03 -0700 (PDT) Message-Id: <199705270220.TAA21364@scruz.net> Subject: Re: GNU libc sources Date: Mon, 26 May 97 19:19:17 -0700 x-mailer: Claris Emailer 2.0, March 15, 1997 From: Elias Kesh To: Mime-Version: 1.0 Content-Type: text/plain; charset="US-ASCII" Content-Length: 456 Lines: 15 Great, I just rebuilt everything with the libc-961216.tar.gz and binutils 2.7-4 and Apache built just fine. If you tell me where you uploaded the glibc I will try it also. As far as the SCSI driver it seems that you need both the uncached access space and the cacheflush commands in a few key spots. I'm not sure why right now but as soon as I find out I will send you the file. Before the end of the week I should be done. Elias elias@kesh.com From linux@informatik.uni-koblenz.de Tue May 27 04:40:01 1997 Received-Date: Tue, 27 May 1997 04:39:59 +0200 (MET DST) Received: from thoma (linux@thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.8.5/8.6.9) with SMTP id EAA20136 for ; Tue, 27 May 1997 04:39:56 +0200 (MEST) From: Systemkennung Linux Message-Id: <199705270239.EAA20136@informatik.uni-koblenz.de> Received: by thoma (SMI-8.6/KO-2.0) id EAA02618; Tue, 27 May 1997 04:39:53 +0200 Subject: Re: GNU libc sources To: linux-mips@fnet.fr Date: Tue, 27 May 1997 04:39:53 +0200 (MET DST) In-Reply-To: from "Stuart Auchterlonie" at May 27, 97 11:59:53 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 562 Lines: 18 Hi, > > I'm just uploading the sources of GNU libc 2.0.1. As I've already > > announced before there are > > Ralf, > Is there any reason why we aren't heading straight for > glibc-2.0.3 which appears to be the latest version from GNU. > It apparently has lots and lots of bugs fixed... The reason is called time. I did the 2.0.1 port when 2.0.1 was brand new. I never found the time to track down the remaining bugs but updating is really essential now that I have modified struct termios for current kernels. Btw., 2.0.4 is about to be released. Ralf From linux@informatik.uni-koblenz.de Tue May 27 04:48:16 1997 Received-Date: Tue, 27 May 1997 04:48:15 +0200 (MET DST) Received: from thoma (linux@thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.8.5/8.6.9) with SMTP id EAA21322 for ; Tue, 27 May 1997 04:48:11 +0200 (MEST) From: Systemkennung Linux Message-Id: <199705270248.EAA21322@informatik.uni-koblenz.de> Received: by thoma (SMI-8.6/KO-2.0) id EAA02641; Tue, 27 May 1997 04:48:09 +0200 Subject: Re: GNU libc sources To: linux-mips@fnet.fr Date: Tue, 27 May 1997 04:48:08 +0200 (MET DST) In-Reply-To: <199705270220.TAA21364@scruz.net> from "Elias Kesh" at May 26, 97 07:19:17 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 959 Lines: 27 Hi, > Great, I just rebuilt everything with the libc-961216.tar.gz and binutils > 2.7-4 and Apache built just fine. If you tell me where you uploaded the > glibc I will try it also. ftp.fnet.fr and kernel.panic.julia.de. Both are however not yet online. If you're in a hurry try http://www.uni-koblenz.de/~linux/glibc-2.0.1-1.tar.gz and enjoy our lousy pipe. Hey, at that time it might even work ... > As far as the SCSI driver it seems that you need both the uncached access > space and the cacheflush commands in a few key spots. I'm not sure why > right now but as soon as I find out I will send you the file. Before the > end of the week I should be done. Great. I'm also working on the stuff. The NCR driver contains some bugs; probably more drivers are affected by this effect. The driver assumes that virt_to_bus(virt_to_bus(ptr)) == virt_to_bus(ptr) and as implemented on Intel and Alpha the Linux kernel tolerates that. Ralf From dom@gladsmuir.algor.co.uk Tue May 27 08:57:27 1997 Received-Date: Tue, 27 May 1997 08:57:26 +0200 (MET DST) Received: from gladsmuir.algor.co.uk (dom@gladsmuir.algor.co.uk [193.117.190.129]) by embankment.algor.co.uk (8.8.5/8.8.5) with ESMTP id HAA09413; Tue, 27 May 1997 07:57:21 +0100 (BST) Received: (from dom@localhost) by gladsmuir.algor.co.uk (8.7.3/8.7.3) id HAA00197; Tue, 27 May 1997 07:58:44 +0100 (BST) Date: Tue, 27 May 1997 07:58:44 +0100 (BST) From: Dom Sweetman Message-Id: <199705270658.HAA00197@gladsmuir.algor.co.uk> To: linux-mips@fnet.fr Subject: Re: GNU libc sources In-Reply-To: <199705270220.TAA21364@scruz.net> References: <199705270220.TAA21364@scruz.net> Content-Length: 1367 Lines: 34 Elias, > As far as the SCSI driver it seems that you need both the uncached access > space and the cacheflush commands in a few key spots. Uncached accesses and correct cache management (writeback and/or invalidation) are equivalent. If they act differently there are bugs in the driver code. I can guess where one of these might be, so check out the following. The 53C810 SCSI controller used on your P-4032 board is a very enthusiastic user of shared memory. As well as its buffers and buffer control structures, it uses shared memory to fetch "script" microcode and to maintain status words. It's the status word which is most commonly the problem. Some of the 53Cxx0 drivers in common circulation use a field in a per-controller software structure to maintain device status. But the structure also contains other software-only fields. If one cache line contains both the device status word (being updated by the SCSI controller) and commonly-used software variables then the system will be unreliable, because the CPU will often see a 'stale' copy of the status word which got brought into cache as a side-effect of a reference to another variable. You'll need to change the driver to make sure that any location read/written by the SCSI controller is safely fenced away from software-only fields. Dominic Sweetman Algorithmics Ltd dom@algor.co.uk From linux@informatik.uni-koblenz.de Tue May 27 15:37:44 1997 Received-Date: Tue, 27 May 1997 15:33:37 +0200 (MET DST) Received: from thoma (linux@thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.8.5/8.6.9) with SMTP id PAA03275 for ; Tue, 27 May 1997 15:32:45 +0200 (MEST) From: Systemkennung Linux Message-Id: <199705271332.PAA03275@informatik.uni-koblenz.de> Received: by thoma (SMI-8.6/KO-2.0) id PAA05748; Tue, 27 May 1997 15:32:44 +0200 Subject: Re: GNU libc sources To: linux-mips@fnet.fr Date: Tue, 27 May 1997 15:32:43 +0200 (MET DST) In-Reply-To: <199705270658.HAA00197@gladsmuir.algor.co.uk> from "Dom Sweetman" at May 27, 97 07:58:44 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 3875 Lines: 88 Dom, the general problem that we have under Linux is that the way the Linux device drivers for shared memory and DMA hardware on ISA/EISA/PCI bus are implemented is not really suitable for use on a MIPS system. The reason for this is easy; the Linux device drivers were originally written on for Intel where virtual and physical addresses are the same. Later on this model was extended for the Alpha; the device driver model now contains bus addresses which are the addresses under which the memory is accessible from a bus device. Still alot of device drivers get that wrong. Nothing however has so far been done to support DMA cache coherence in software because Intel and Alpha solve this in hardware invisible to the programmer. Fixing is easy in general and involves adding some calls to the cache flushing routines as well as some converting some addresses into KSEG1 addresses. I've already done this for several drivers; it's however not a very nice way to solve the problem because it's either unportable or the extra stuff needs to be wrapped by #ifdefs. Finally we have to deal with the more sophisticated DMA hardware available in some MIPS systems, in particular the Jazz machines. On these machines DMA bus addresses are being translated into virtual addresses by an onboard MMU. Note that this nukes the simple model of physical and bus address most drivers have. We currently cheat on these machines by setting up an 1:1 address translation table; more sophisticated drivers which have knowledge of the hardware like the floppy driver and the Sonic Ethernet driver install special mappings. (The 1:1 mapping is a bad thing; we don't use the capability to do scatter/ gather for all DMA devices. This is in particular interesting for modules like the floppy driver which is plagued by the fact that it cannot allocate DMA memory due fragmentation of the free page pool) Furthermore we have machines (Wreckstation Tyne) where DMA-able memory is allocated from a special pool. I'd like to solve all that by adding an hook to kmalloc(), __get_free_pages() and __free_pages() which will allow to plug architecture specific routines in. For some rare cases it might also make sense to have a portable way to flush caches. For that I suggest an hook similar to the other cache flushing functions. Ralf > > As far as the SCSI driver it seems that you need both the uncached access > > space and the cacheflush commands in a few key spots. > > Uncached accesses and correct cache management (writeback and/or > invalidation) are equivalent. If they act differently there are bugs > in the driver code. > > I can guess where one of these might be, so check out the following. > > The 53C810 SCSI controller used on your P-4032 board is a very > enthusiastic user of shared memory. As well as its buffers and buffer > control structures, it uses shared memory to fetch "script" microcode > and to maintain status words. > > It's the status word which is most commonly the problem. Some of the > 53Cxx0 drivers in common circulation use a field in a per-controller > software structure to maintain device status. But the structure also > contains other software-only fields. If one cache line contains both > the device status word (being updated by the SCSI controller) and > commonly-used software variables then the system will be unreliable, > because the CPU will often see a 'stale' copy of the status word which > got brought into cache as a side-effect of a reference to another > variable. > > You'll need to change the driver to make sure that any location > read/written by the SCSI controller is safely fenced away from > software-only fields. > > Dominic Sweetman > Algorithmics Ltd > dom@algor.co.uk > -- Ralf Baechle Internet: linux@informatik.uni-koblenz.de Fido: Ralf Baechle 2:245/5618.2 Mud: arrow@Mythos From dom@gladsmuir.algor.co.uk Tue May 27 16:23:46 1997 Received-Date: Tue, 27 May 1997 16:23:40 +0200 (MET DST) Received: from gladsmuir.algor.co.uk (dom@gladsmuir.algor.co.uk [193.117.190.129]) by embankment.algor.co.uk (8.8.5/8.8.5) with ESMTP id PAA00600; Tue, 27 May 1997 15:23:28 +0100 (BST) Received: (from dom@localhost) by gladsmuir.algor.co.uk (8.7.3/8.7.3) id PAA00477; Tue, 27 May 1997 15:24:57 +0100 (BST) Date: Tue, 27 May 1997 15:24:57 +0100 (BST) From: Dom Sweetman Message-Id: <199705271424.PAA00477@gladsmuir.algor.co.uk> To: linux-mips@fnet.fr Subject: Re: GNU libc sources In-Reply-To: <199705271332.PAA03275@informatik.uni-koblenz.de> References: <199705270658.HAA00197@gladsmuir.algor.co.uk> <199705271332.PAA03275@informatik.uni-koblenz.de> Content-Length: 2773 Lines: 73 Ralf wrote: > the general problem that we have under Linux is that the way the Linux > device drivers for shared memory and DMA hardware on ISA/EISA/PCI bus > are implemented is not really suitable for use on a MIPS system. > > The reason for this is easy; the Linux device drivers were originally > written on for Intel where virtual and physical addresses are the same. > ... > Nothing however has so far been done to support DMA cache coherence in > software because Intel and Alpha solve this in hardware invisible to > the programmer. We've notice (as gray-haired BSD folk at heart) that Linux sometimes is very PC-centric. But if you want Linux to be portable, you'd better work towards losing these two assumptions; and not just in 'ifdefs' either. Even Intel might ship a Merced one day. You *can* write a successful OS where virtual and physical are related in a more complex way; and you can cope with driver-visible caches. In fact, the MIPS approach has a lot to recommend it. As well as making the hardware simpler, it's often more efficient for the CPU to writeback or invalidate the cache - the hardware poking from the other side ends up shutting the CPU out of the cache while it does the job. It's an interesting question how the "mainstream" Linux camp would react to the idea that this is a portable coding practice question, which is everybody's problem. Many useful memory systems are a lot more complicated than a PC-clone forget-the-cache approach can support; it's not just MIPS. Random points: I can't see how anything you do at kmalloc() time is going to fix the problem. I decided a couple of manuals back that I'd give up using the term "cache flush" because I was never sure whether I (or anyone else) meant "invalidate" or "writeback" or both. > ... it might also make sense to have a portable way to flush caches. > For that I suggest an hook similar to the other cache flushing > functions. You really need two functions: writeback (addr, n) which on a vanilla MIPS machine would write back n bytes worth of locations from addr, and invalidate (addr, n) which would invalidate n bytes worth of locations from addr. Whether these are physical or virtual addresses doesn't matter much, so long as you decide which and don't change your mind! My preference would be to work in the kernel's normal address space for as long as possible, and provide a function for drivers wanting to calculate physical addresses... However, you wouldn't actually usually write those in drivers; you'd code them as: before_dma_out (addr, n) == writeback on MIPS after_dma_in (addr, n) == invalidate on MIPS On a PC clone, both "before..." and "after..." can be ifdef'd away. Dominic Sweetman Algorithmics Ltd dom@algor.co.uk From linux@informatik.uni-koblenz.de Tue May 27 18:02:51 1997 Received-Date: Tue, 27 May 1997 18:02:31 +0200 (MET DST) Received: from thoma (linux@thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.8.5/8.6.9) with SMTP id SAA29206 for ; Tue, 27 May 1997 18:02:03 +0200 (MEST) From: Systemkennung Linux Message-Id: <199705271602.SAA29206@informatik.uni-koblenz.de> Received: by thoma (SMI-8.6/KO-2.0) id SAA06637; Tue, 27 May 1997 18:02:02 +0200 Subject: Re: GNU libc sources To: linux-mips@fnet.fr Date: Tue, 27 May 1997 18:02:02 +0200 (MET DST) In-Reply-To: <199705271424.PAA00477@gladsmuir.algor.co.uk> from "Dom Sweetman" at May 27, 97 03:24:57 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 4666 Lines: 115 Hi, > We've notice (as gray-haired BSD folk at heart) that Linux sometimes > is very PC-centric. But if you want Linux to be portable, you'd > better work towards losing these two assumptions; and not just in > 'ifdefs' either. Even Intel might ship a Merced one day. You're right, Linux is still pretty PC-centric. But worse is that the people working on Linux are thinking PC centric. I'm a pessimist in that respect; that's why I'm trying to fix things invissible for device drivers. Being PC centric has so far not been a problem because m68k and Sparc where these problems do exist also use their own drivers and have their own, private solutions for the problem. > You *can* write a successful OS where virtual and physical are related > in a more complex way; and you can cope with driver-visible caches. > > In fact, the MIPS approach has a lot to recommend it. As well as > making the hardware simpler, it's often more efficient for the CPU to > writeback or invalidate the cache - the hardware poking from the other > side ends up shutting the CPU out of the cache while it does the job. > > It's an interesting question how the "mainstream" Linux camp would > react to the idea that this is a portable coding practice question, > which is everybody's problem. Many useful memory systems are a lot > more complicated than a PC-clone forget-the-cache approach can > support; it's not just MIPS. > > Random points: > > I can't see how anything you do at kmalloc() time is going to fix the > problem. Linux's kmalloc has a second flag argument where one can pass flags like GFP_DMA for DMA-able memory. As a simple fix kmalloc & Co. could then return cacheline aligned memory with a KSEG1 address. Simple fix for alot of drivers. In case of the Wreckstation Tyne we really need to tweak the memory allocation scheme - DMA-able memory comes from a different pool than the normal memory. Nice side effect - you never have DMA cache coherency problems on that machine. > I decided a couple of manuals back that I'd give up using the term > "cache flush" because I was never sure whether I (or anyone else) > meant "invalidate" or "writeback" or both. I agree. In the Linux kernel it does not matter in many cases whether one does write back the cache or just invalidate it, so the used term flush is ok, I think. Same for flush_page_to_ram, I think. > > ... it might also make sense to have a portable way to flush caches. > > For that I suggest an hook similar to the other cache flushing > > functions. > > You really need two functions: > > writeback (addr, n) > > which on a vanilla MIPS machine would write back n bytes worth of > locations from addr, and > > invalidate (addr, n) > See include/asm-mips/r4kcache.h: extern inline void flush_icache_line_indexed(unsigned long addr); extern inline void flush_dcache_line_indexed(unsigned long addr); extern inline void flush_scache_line_indexed(unsigned long addr); extern inline void flush_icache_line(unsigned long addr); extern inline void flush_dcache_line(unsigned long addr); extern inline void flush_scache_line(unsigned long addr); > these are physical or virtual addresses doesn't matter much, so long > as you decide which and don't change your mind! My preference would > be to work in the kernel's normal address space for as long as > possible, and provide a function for drivers wanting to calculate > physical addresses... > > However, you wouldn't actually usually write those in drivers; you'd > code them as: > > before_dma_out (addr, n) == writeback on MIPS > after_dma_in (addr, n) == invalidate on MIPS That's essentially my suggestion in my last posting about extending the already available cache flushing functionality: (from include/asm-mips/pgtable.h) /* Function pointers - all the MIPS hardware is sooo similar ... */ extern void (*flush_cache_all)(void); extern void (*flush_cache_mm)(struct mm_struct *mm); extern void (*flush_cache_range)(struct mm_struct *mm, unsigned long start, unsigned long end); extern void (*flush_cache_page)(struct vm_area_struct *vma, unsigned long page); extern void (*flush_cache_sigtramp)(unsigned long addr); and then new: extern void (*flush_cache_before_dma_out)(unsigned long start, unsigned long length); extern void (*flush_cache_after_dma_in)(unsigned long start, unsigned long length); The floppy driver already uses something similar, just abstracted out another way into the header files. > On a PC clone, both "before..." and "after..." can be ifdef'd away. Ralf From micha@Odalix.TeCNeT.DE Tue May 27 18:10:13 1997 Received-Date: Tue, 27 May 1997 18:10:11 +0200 (MET DST) Received: (from micha@localhost) by Odalix.TeCNeT.DE (8.8.5/8.8.5) id SAA04119 for linux-mips@fnet.fr; Tue, 27 May 1997 18:09:57 +0200 From: Michael Bruck Message-Id: <199705271609.SAA04119@Odalix.TeCNeT.DE> Subject: DECstation 5000/20 boot failure To: linux-mips@fnet.fr Date: Tue, 27 May 1997 18:09:57 +0200 (MET DST) Content-Type: text Content-Length: 693 Lines: 27 Hello all, I've just got the dec_vmlinux.2 image. Tried tftpboot. This image hangs after about to clear hash table here some lines that may be of interest: memsize is: 17ff000 vmalloc'd hash table, size is: fff4, addr is: c0000000 Page fault taken [swapper:1:80071d38]... couldn't find vma 0x0 Unable ... address c0000000, epc == 8005455c ... Status: 10000804 Cause : 00000800 ... Code : 00000000 00000000 3c04800d 0c014332 24844314 kfree of non-kmalloced memory: 800d2b2c, next= 00000000, order=0 800d2b1c 800d3030 Ok, that's it. If there is a way to get the whole screen over serial console I can supply more info. Micha From tschroed@cheetah.wsc.edu Tue May 27 18:25:51 1997 Received-Date: Tue, 27 May 1997 18:25:50 +0200 (MET DST) Received: by cheetah.wsc.edu (5.65/DEC-Ultrix/4.4) id AA28824; Tue, 27 May 1997 11:25:19 -0500 Other-Stuff: localhost tschroed@cheetah.wsc.edu Trevor Schroeder tschroed Date: Tue, 27 May 1997 11:25:18 -0500 (CDT) From: Trevor Schroeder To: Linux/MIPS list Subject: MkLinux Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 731 Lines: 13 Again, this may be a silly question with a simple answer...But has any thought been given to simply porting MkLinux to MIPS hardware. It should be much easier...But then maybe I'm missing something here, or maybe the whole point isn't 'easy' as much as 'oooh...neat...I know everything there is to know about the DECstation 2100' (I would be more inclined towards the latter). ____________________________________________________________ "One unerring mark of the love of truth is not entertaining any propositions with greater assurance than the proofs it is built upon will warrant" -- John Locke, 1690 Trevor Schroeder tschroed@cheetah.wsc.edu ------------------------------------------------------------ From linux@informatik.uni-koblenz.de Tue May 27 18:27:35 1997 Received-Date: Tue, 27 May 1997 18:27:24 +0200 (MET DST) Received: from thoma (linux@thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.8.5/8.6.9) with SMTP id SAA03270; Tue, 27 May 1997 18:25:47 +0200 (MEST) From: Systemkennung Linux Message-Id: <199705271625.SAA03270@informatik.uni-koblenz.de> Received: by thoma (SMI-8.6/KO-2.0) id SAA06720; Tue, 27 May 1997 18:25:46 +0200 Subject: Re: DECstation 5000/20 boot failure To: linux-mips@fnet.fr Date: Tue, 27 May 1997 18:25:45 +0200 (MET DST) In-Reply-To: <199705271609.SAA04119@Odalix.TeCNeT.DE> from "Michael Bruck" at May 27, 97 06:09:57 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 695 Lines: 23 Hi, > memsize is: 17ff000 > vmalloc'd hash table, size is: fff4, addr is: c0000000 > > Page fault taken [swapper:1:80071d38]... couldn't find vma 0x0 > Unable ... address c0000000, epc == 8005455c > ... > Status: 10000804 > Cause : 00000800 > ... > Code : 00000000 00000000 3c04800d 0c014332 24844314 > kfree of non-kmalloced memory: 800d2b2c, next= 00000000, order=0 > 800d2b1c > 800d3030 > > Ok, that's it. If there is a way to get the whole screen over > serial console I can supply more info. This is supposed to happen; the way the mm tables are handles has changed and the R3000 stuff still needs to be rewritten. Ralf From engel@numerik.math.uni-siegen.de Tue May 27 21:38:43 1997 Received-Date: Tue, 27 May 1997 21:38:42 +0200 (MET DST) Received: from hilbert.numerik.math.uni-siegen.de (hilbert.numerik.math.uni-siegen.de [141.99.112.8]) by bernoulli.numerik.math.uni-siegen.de (Mailhost) with SMTP id VAA01992 for ; Tue, 27 May 1997 21:35:16 +0200 (MET DST) From: Michael Engel Message-Id: <9705271942.AA09968@hilbert.numerik.math.uni-siegen.de> Received: by hilbert.numerik.math.uni-siegen.de id AA09968; Tue, 27 May 97 21:42:26 +0200 Subject: DECstation documentation ... To: linux-mips@fnet.fr Date: Tue, 27 May 1997 21:42:26 +0200 (MET DST) X-Mailer: ELM [version 2.4 PL23] Content-Type: text Content-Length: 645 Lines: 17 Hi to all you DECstation porters ... Some minutes ago, I browsed through the Mach sources for DECstations and found the names of some DEC specs for DECstations we would surely like to have ... To make the story short - I did a web search for the email address of the author of most of the files (his name is Alessandro Forin) and found his email address at CMU ! I mailed him just a minute ago - perhaps he still has some of the specs lying around. I also talked to John "Maddog" Hall at the Linux Kongress; he promised to look for the specs once more ... so, there's still hope ;) regards, Michael Engel (engel@unix-ag.uni-siegen.de) From chris@x.cyphercom.com Thu May 29 21:38:19 1997 Received-Date: Thu, 29 May 1997 21:38:17 +0200 (MET DST) Received: (from chris@localhost) by x.cyphercom.com (8.8.5/8.8.3) id TAA03186; Tue, 27 May 1997 19:04:39 -0400 Date: Tue, 27 May 1997 19:04:39 -0400 From: Chris Wedgwood Message-Id: <199705272304.TAA03186@x.cyphercom.com> To: linux-mips@fnet.fr Subject: Re: MkLinux Cc: Chris Wedgwood Content-Length: 456 Lines: 15 Trevor Schroeder wrote: Again, this may be a silly question with a simple answer...But has any thought been given to simply porting MkLinux to MIPS hardware. Probably, the speed would suck. I used to really like the idea of a micro-kernel coming from a text book, but reality is something quite different. I think DEC cheated a bit by placing everything in the same address space anyhow for this reason. (Or was this only for OSF/1?) -Chris From paul@suede.sw.oz.au Wed May 28 01:12:17 1997 Received-Date: Wed, 28 May 1997 01:12:14 +0200 (MET DST) Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id XAA18131; Tue, 27 May 1997 23:11:38 GMT (8.6.10/Unixware) (from paul@suede.sw.oz.au for ) Received: by suede.sw.oz.au id XAA26604; Tue, 27 May 1997 23:11:37 GMT (SMI-8.6/1.34) (from paul for linux-mips@fnet.fr) From: paul@suede.sw.oz.au (Paul Antoine) Message-Id: <199705272311.XAA26604@suede.sw.oz.au> Subject: Re: DECstation documentation ... To: linux-mips@fnet.fr Date: Wed, 28 May 1997 09:11:36 +1000 (EST) In-Reply-To: <9705271942.AA09968@hilbert.numerik.math.uni-siegen.de> from "Michael Engel" at May 27, 97 09:42:26 pm Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 1510 Lines: 38 Michael Engel wrote: > Hi to all you DECstation porters ... > > Some minutes ago, I browsed through the Mach sources for DECstations > and found the names of some DEC specs for DECstations we would > surely like to have ... Yep - Mach is very useful in that regard. I had previously passed these names on to Maddog to try to find within DEC... but he didn't have much luck last time (though I think he was insanely busy at the time too... :-) > To make the story short - I did a web search for the email address > of the author of most of the files (his name is Alessandro Forin) > and found his email address at CMU ! I mailed him just a minute ago > - perhaps he still has some of the specs lying around. Good work! > I also talked to John "Maddog" Hall at the Linux Kongress; he > promised to look for the specs once more ... so, there's still hope > ;) Indeed - Maddog has already spent quite a bit of time searching for these, so we should find some appropriate way to thank him methinx, particularly if he manages to get hold of them! Regards, Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 9698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 9699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From ctrl@skylab.math.unibuc.ro Wed May 28 10:49:27 1997 Received-Date: Wed, 28 May 1997 10:49:23 +0200 (MET DST) Received: (from ctrl@localhost) by skylab.math.unibuc.ro (8.8.5/8.8.0(beta-ctrl)) id OAA04690; Wed, 28 May 1997 14:08:37 +0300 (EET DST) Date: Wed, 28 May 1997 14:08:37 +0300 (EET DST) From: Cornel Ciocirlan To: linux-mips@fnet.fr Subject: Linux to R3000's with RISCos Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 636 Lines: 17 Hello! I have an old box here, Control Data CD4330. The OS is EP/IX 1.4.3, actually MIPS RISCos 4.0 or 4.52. I want to know if there is someone working on porting linux MIPS to it and how I can help. I guess the problem is it doesn't have a graphics card, it has a RISCos boot prom. I am a student at the Bucharest University, the machine I'm talking about is skylab.math.unibuc.ro. _/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_// Cornel Ciocirlan / UNIX sysadmin / Bucharest University / I'm ctrl@unibuc.ro. / ----------------------------/ From neumanns@uni-duesseldorf.de Wed May 28 21:25:58 1997 Received-Date: Wed, 28 May 1997 21:25:57 +0200 (MET DST) Received: from isdna39.extern.uni-duesseldorf.de by sirene.rz.uni-duesseldorf.de with SMTP (PP) id <17483-0@sirene.rz.uni-duesseldorf.de>; Wed, 28 May 1997 21:25:48 +0000 Received: from wintermute (neo@wintermute.huettenstudios.de [192.168.1.1]) by router.huettenstudios.de (8.7.5/8.7.3) with SMTP id VAA24403 for ; Wed, 28 May 1997 21:17:42 +0200 Sender: neo@router.huettenstudios.de Message-ID: <338C84DD.7E1E42B1@uni-duesseldorf.de> Date: Wed, 28 May 1997 21:17:49 +0200 From: Sven Neumann Organization: H"uttenstudios.de (TM) X-Mailer: Mozilla 3.01Gold (X11; I; Linux 2.0.29 i586) MIME-Version: 1.0 To: linux-mips@fnet.fr Subject: RS 6000 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Length: 558 Lines: 14 Hi, I've got a short question about the status of Linux/MIPS. A friend of mine got his hands on a machine that is said to have a RS 6000 cpu. It was fabricated by IBM (at least the label says so). Unfortunatley I don't have any more details about the computer now, but I would like to know if there's a chance that we might Linux working on that machine? I will now send him the instructions I got from the FAQ and Howto and tell him to try MILO. If you're interested I will send you any results and I hope I can give you more details later. Salut, Sven From tor@spacetec.no Wed May 28 21:37:13 1997 Received-Date: Wed, 28 May 1997 21:37:12 +0200 (MET DST) Received: (from tor@localhost) by pallas.spacetec.no (8.8.5/8.8.5) id VAA23583 for linux-mips@fnet.fr; Wed, 28 May 1997 21:38:34 +0200 Message-Id: <199705281938.VAA23583@pallas.spacetec.no> From: tor@spacetec.no (Tor Arntsen) Date: Wed, 28 May 1997 21:38:32 +0200 In-Reply-To: Sven Neumann "RS 6000" (May 28, 20:28) X-Mailer: Mail User's Shell (7.2.5 10/14/92) To: linux-mips@fnet.fr Subject: Re: RS 6000 Content-Length: 732 Lines: 16 On May 28, 20:28, Sven Neumann wrote: >I've got a short question about the status of Linux/MIPS. A friend of >mine got his hands on a machine that is said to have a RS 6000 cpu. It >was fabricated by IBM (at least the label says so). Unfortunatley I >don't have any more details about the computer now, but I would like to >know if there's a chance that we might Linux working on that machine? It *could* be that it's not really an RS 6000, but instead a PowerPC, and in that case it is a chance that it can (or will be able to) run Linux/ppc. But if it's an older IBM RT then it's no. In any case it's not a MIPS CPU, so Linux/MIPS is out of the question.. Tor (currently snowed over with an IBM/SP2 system, which is ~RS 6000) From tor@spacetec.no Wed May 28 21:41:11 1997 Received-Date: Wed, 28 May 1997 21:41:09 +0200 (MET DST) Received: (from tor@localhost) by pallas.spacetec.no (8.8.5/8.8.5) id VAA23590 for linux-mips@fnet.fr; Wed, 28 May 1997 21:43:10 +0200 Message-Id: <199705281943.VAA23590@pallas.spacetec.no> From: tor@spacetec.no (Tor Arntsen) Date: Wed, 28 May 1997 21:43:09 +0200 In-Reply-To: Sven Neumann "RS 6000" (May 28, 20:28) X-Mailer: Mail User's Shell (7.2.5 10/14/92) To: linux-mips@fnet.fr Subject: Re: RS 6000 Content-Length: 683 Lines: 15 On May 28, 20:28, Sven Neumann wrote: >I've got a short question about the status of Linux/MIPS. A friend of >mine got his hands on a machine that is said to have a RS 6000 cpu. It >was fabricated by IBM (at least the label says so). Unfortunatley I >don't have any more details about the computer now, but I would like to >know if there's a chance that we might Linux working on that machine? (I've already sent one response but you never know the sequence they arrive in..) I guess you're thinking about a MIPS R6000, but it's a different CPU. The IBM RS6000 is more like a PowerPC I believe -- at least I have IBM binaries that can execute on both. But MIPS it ain't.. Tor From tschroed@cheetah.wsc.edu Wed May 28 23:03:57 1997 Received-Date: Wed, 28 May 1997 23:03:54 +0200 (MET DST) Received: by cheetah.wsc.edu (5.65/DEC-Ultrix/4.4) id AA00698; Wed, 28 May 1997 16:03:47 -0500 Other-Stuff: localhost tschroed@cheetah.wsc.edu Trevor Schroeder tschroed Date: Wed, 28 May 1997 16:03:46 -0500 (CDT) From: Trevor Schroeder To: linux-mips@fnet.fr Subject: Re: RS 6000 In-Reply-To: <199705281938.VAA23583@pallas.spacetec.no> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 625 Lines: 16 On Wed, 28 May 1997, Tor Arntsen wrote: > But if it's an older IBM RT then it's no. > Just to clear up a bit of a misconception, the RT was not based on the POWER architecture. The RT processor was the IBM ROMP (16bit, among other things). Sorry about hte offtopicness ____________________________________________________________ "One unerring mark of the love of truth is not entertaining any propositions with greater assurance than the proofs it is built upon will warrant" -- John Locke, 1690 Trevor Schroeder tschroed@cheetah.wsc.edu ------------------------------------------------------------ From tor@spacetec.no Wed May 28 23:09:33 1997 Received-Date: Wed, 28 May 1997 23:09:32 +0200 (MET DST) Received: (from tor@localhost) by pallas.spacetec.no (8.8.5/8.8.5) id XAA23644 for linux-mips@fnet.fr; Wed, 28 May 1997 23:11:32 +0200 Message-Id: <199705282111.XAA23644@pallas.spacetec.no> From: tor@spacetec.no (Tor Arntsen) Date: Wed, 28 May 1997 23:11:30 +0200 In-Reply-To: Trevor Schroeder "Re: RS 6000" (May 28, 22:05) X-Mailer: Mail User's Shell (7.2.5 10/14/92) To: linux-mips@fnet.fr Subject: Re: RS 6000 Content-Length: 370 Lines: 11 On May 28, 22:05, Trevor Schroeder wrote: >Just to clear up a bit of a misconception, the RT was not based on the POWER >architecture. The RT processor was the IBM ROMP (16bit, among other things). Yes you're right. I was indeed thinking about the first box where IBM used the RS6000 chipset, which was later than the RT. >Sorry about hte offtopicness Me too.. Tor From srnm@email.msn.com Thu May 29 05:07:40 1997 Received-Date: Thu, 29 May 1997 05:07:39 +0200 (MET DST) From: srnm@email.msn.com Received: from UBIMSSMTP01 - 207.68.143.137 by email.msn.com with Microsoft SMTPSVC; Wed, 28 May 1997 20:06:15 -0700 Received: from frodo - 153.35.15.2 by email.msn.com with Microsoft SMTPSVC; Wed, 28 May 1997 20:06:12 -0700 To: Subject: help offered Date: Wed, 28 May 1997 23:02:57 -0400 X-MSMail-Priority: Normal X-Priority: 3 X-Mailer: Microsoft Internet Mail 4.70.1162 Message-ID: <0452f1206031d57UBIMSSMTP01@email.msn.com> Content-Length: 111 Lines: 10 Hello! Is the linux port to MIPS still an active project? Are you looking for help? thanks! Steven Marcus From linux@informatik.uni-koblenz.de Fri May 30 14:51:51 1997 Received-Date: Fri, 30 May 1997 14:51:43 +0200 (MET DST) Received: from thoma (linux@thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.8.5/8.6.9) with SMTP id OAA02115; Fri, 30 May 1997 14:49:08 +0200 (MEST) From: Systemkennung Linux Message-Id: <199705301249.OAA02115@informatik.uni-koblenz.de> Received: by thoma (SMI-8.6/KO-2.0) id OAA14390; Fri, 30 May 1997 14:49:33 +0200 Subject: Re: MkLinux To: linux-mips@fnet.fr Date: Fri, 30 May 1997 14:49:32 +0200 (MET DST) Cc: chris@cybernet.co.nz In-Reply-To: <199705272304.TAA03186@x.cyphercom.com> from "Chris Wedgwood" at May 27, 97 07:04:39 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 885 Lines: 22 Hi, > Again, this may be a silly question with a simple answer...But has any > thought been given to simply porting MkLinux to MIPS hardware. > > Probably, the speed would suck. I used to really like the idea of a > micro-kernel coming from a text book, but reality is something quite > different. If you do things right speed isn't that bad. The Microkernel approach gives you a nice and easy to understand OS structure. On the other side for things like a filesystem you get multithreading for free. In a microkernel OS you end up implementing some part of a monolithic kernel again ... Not to mentioned that all the microkernels I've seen violate their own strict structuring rules. > I think DEC cheated a bit by placing everything in the same address space > anyhow for this reason. (Or was this only for OSF/1?) The Evil Empire did the same for NT 4.0. Ralf From chris@x.cyphercom.com Sat May 31 03:36:06 1997 Received-Date: Sat, 31 May 1997 03:36:04 +0200 (MET DST) Received: (from chris@localhost) by x.cyphercom.com (8.8.5/8.8.3) id UAA03295; Fri, 30 May 1997 20:33:00 -0400 Date: Fri, 30 May 1997 20:33:00 -0400 From: Chris Wedgwood Message-Id: <199705310033.UAA03295@x.cyphercom.com> To: linux-mips@fnet.fr, linux@mailhost.uni-koblenz.de Subject: Re: MkLinux Cc: chris@cybernet.co.nz Content-Length: 1257 Lines: 30 If you do things right speed isn't that bad. The Microkernel approach gives you a nice and easy to understand OS structure. On the other side for things like a filesystem you get multithreading for free. Yes - but like any threading code you still have to put locks all over the place and hence may not achieve huge performance gains. In a microkernel OS you end up implementing some part of a monolithic kernel again... Not to mentioned that all the microkernels I've seen violate their own strict structuring rules. I've been wrong before, and I'll be wrong again. But I've yet to see any single microkernel offer all the advantages commonly found in textbooks. Digital Unix probably comes the closest IMO. (I like Digital Unix, so I'm biased). > I think DEC cheated a bit by placing everything in the same address > space anyhow for this reason. (Or was this only for OSF/1?) The Evil Empire did the same for NT 4.0. I though this was only for the GDI? I've used NT since the 3.1 days (hey, shoot me) and only now (4.0) are we lucky anough to get all sorts of previously unknown weird bugs which appear to be related to buggy video drivers.... that's not to say there weren't lots of bugs before. -Chris