From Gilles.Gravier@France.Sun.COM Thu Feb 1 02:06:48 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id CAA02761 for ; Thu, 1 Feb 1996 02:06:46 +0100 (MET) Received: from mercury.Sun.COM by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA18182; Thu, 1 Feb 1996 02:06:37 +0100 (MET) Received: from snail.Sun.COM by mercury.Sun.COM (Sun.COM) id RAA04697; Wed, 31 Jan 1996 17:06:40 -0800 Received: from France (isunfra) by snail.Sun.COM (4.1/SMI-4.1) id AA08263; Wed, 31 Jan 96 17:06:37 PST Received: from mygale.France.Sun.COM by France (4.1/SMI-4.1) id AA19809; Thu, 1 Feb 96 02:06:35 +0100 Received: from elektra.France.Sun.COM by mygale.France.Sun.COM (SMI-8.6/SMI-SVR4 (1/24/94)) id CAA10074; Thu, 1 Feb 1996 02:07:13 +0100 Received: by elektra.France.Sun.COM (SMI-8.6/SMI-SVR4 (1/24/94)) id CAA23932; Thu, 1 Feb 1996 02:06:33 +0100 Date: Thu, 1 Feb 1996 02:06:33 +0100 From: Gilles.Gravier@France.Sun.COM (Gilles Gravier - ITC) Message-Id: <199602010106.CAA23932@elektra.France.Sun.COM> To: linux-mips@fnet.fr Subject: Re: Linux/MIPS Cross Dev for PSX X-Sun-Charset: US-ASCII Content-Length: 719 Lines: 19 > From linux-mips-request@fnet.fr Thu Feb 1 02:00 MET 1996 > Resent-Date: Thu, 1 Feb 1996 01:57:57 +0100 (MET) > From: Jim Burnes > Subject: Linux/MIPS Cross Dev for PSX > > I'm the guy who is starting the Sony Playstation Hacking Group. Linux/SonyPlaystation... coool! :) Soon, your six year old kid will be typing "RM *" instead of shooting the invaders... He will be typing "WHO" to see a long-distance scan of the surrounding quadrants... He will type "ENV" to get is inventory when playing adventures... And you'll have to teach hime to type "SYNC" before he turns off the Playstation! > Its an R3000 CPU running at 33 MHZ with a hell of a 3d engine. What's the 3D chipset? Gilles. From imp@rover.village.org Thu Feb 1 03:36:16 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id DAA04905 for ; Thu, 1 Feb 1996 03:36:15 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA19164; Thu, 1 Feb 1996 03:35:54 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id TAA19798 for ; Wed, 31 Jan 1996 19:35:45 -0700 Message-Id: <199602010235.TAA19798@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Linux/MIPS Cross Dev for PSX In-Reply-To: Your message of Wed, 31 Jan 1996 18:49:34 Date: Wed, 31 Jan 1996 19:35:44 -0700 From: Warner Losh Content-Length: 1306 Lines: 33 : (1) How do I find out if the CPU is running big/little endian? By the endian bit :-) : (2) I'm really trying hard to figure out their executable format, but : it isnt anything that the "file" command under linux understands. Any : ideas? Yes. Look at the header that is on it. The FreeBSD file command calls things like "milo" a PDP-11 executable (at least in the 2.0R version). : (3) Once I find its exe format I think I can modify ld and related : utilities to write its format out. Anything tricky here? Relocation. Load addresses. Segmentation, etc. : Any help you can offer would be greatly appreciated. If I get the : target debugger running I'll give you guys a copy. It may come in : handy doing remote debugging to your target. Try to compile the binutils for MIPS (see http://www.village.org/villagers/imp/mips/build.html for a quick primer). Then use things like objdump to see if it works. Oh, you might want to also do things like try each of the known mips formats that binutils supports. It will take a while, and maybe a little bit of disk space, but you may be able to gain ground there. At a guess, it is some variation on COFF... Also, somewhere there is a disassembler knocking around. This might be useful in figuring out where the binary really begins. Warner From Stoned.Elipot@univ-evry.fr Thu Feb 1 09:02:45 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id JAA06736 for ; Thu, 1 Feb 1996 09:02:44 +0100 (MET) Received: from amiante.univ-evry.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA23424; Thu, 1 Feb 1996 09:02:37 +0100 (MET) Received: from bandsept.univ-evry.fr (bandsept.univ-evry.fr [192.93.214.19]) by amiante.univ-evry.fr with ESMTP (8.6.12/94.01.10); id IAA24178; Thu, 1 Feb 1996 08:14:52 +0100 Received: from localhost.univ-evry.fr (localhost.univ-evry.fr [127.0.0.1]) by bandsept.univ-evry.fr with SMTP (8.6.12/94.01.26); id JAA06162; Thu, 1 Feb 1996 09:01:21 +0100 Message-Id: <199602010801.JAA06162@bandsept.univ-evry.fr> To: linux-mips@fnet.fr Subject: Re: 1.3.58 on Magnum? In-Reply-To: Your message of "Sun, 31 Dec 1995 21:45:24 EST." <9602010245.AA23471@silver.sni.CA> X-It: bandsept (wrong OS, but right CPU) X-Mailer: MH 6.8.3, MH-E 5.0.2 Date: Thu, 01 Feb 1996 09:01:13 +0100 From: Stoned Elipot Content-Length: 1764 Lines: 39 Hi David, >>>>> "David" == "David Kerry" writes: [SNIP] David> linux-1.3.58.tar.gz source from fnet.fr. The build seems to David> make it most of the way through until it hits the magnum4000.S David> file upon which the assembler complains (err, this is David> configured for a Magnum 4000, elf compiler, extra elf compiler David> by the way): [SNIP] David> Any ideas? I presume it has something to do with elf vs a.out, David> but I have no idea where to look. Well the error message it quite different from the a,.out compiler I still use but I think the reason is the same: look at the end of magnum4000.S, there is some data defined and then a last function, without a .text before it ! Ralf already got patches including this so I believe it will be ok for 1.3.59... David> Oh, and another minor problem: the variable 'video_res_x' in David> drivers/char/vga.c is undefined. It seems that it was supposed David> to be an external, defined in drivers/char/console.c. Is that David> correct? I've patched it temporarily and it gets past it now, David> however I haven't attempted to boot a kernel yet... (See above David> :-) This is known to me and also included in the last patch I sent to Ralf: I used the same obvious fix :) BTW don't expect the 'core' 1.3.58 kernel to boot with milo 0.26 because of the NASP (see README in Milo's package or my previous announce). You can find my patches against 1.3.58 - if you can't wait for 1.3.59 :) - in ftp://ftp.fnet.fr:/linux-mips/private/Incoming/linux-1.3.58-1.3.58.1.diffs.gz BTW again, keyboard is broken since 1.2.11 for Magnum/Oily and I didn't yet found a solution, so... even with Ralf's advice in 1.3.58:driver/char/keyboard.c (search 'Stoned', thanks Ralf ;) Cheers, Stoned. From Stoned.Elipot@univ-evry.fr Thu Feb 1 09:13:58 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id JAA06821 for ; Thu, 1 Feb 1996 09:13:58 +0100 (MET) Received: from amiante.univ-evry.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA23656; Thu, 1 Feb 1996 09:13:51 +0100 (MET) Received: from bandsept.univ-evry.fr (bandsept.univ-evry.fr [192.93.214.19]) by amiante.univ-evry.fr with ESMTP (8.6.12/94.01.10); id IAA24202; Thu, 1 Feb 1996 08:26:06 +0100 Received: from localhost.univ-evry.fr (localhost.univ-evry.fr [127.0.0.1]) by bandsept.univ-evry.fr with SMTP (8.6.12/94.01.26); id JAA06195; Thu, 1 Feb 1996 09:12:37 +0100 Message-Id: <199602010812.JAA06195@bandsept.univ-evry.fr> To: linux-mips@fnet.fr Subject: Re: R3000 support? In-Reply-To: Your message of "Wed, 31 Jan 1996 17:11:35 GMT." X-It: bandsept (wrong OS, but right CPU) X-Mailer: MH 6.8.3, MH-E 5.0.2 Date: Thu, 01 Feb 1996 09:12:31 +0100 From: Stoned Elipot Content-Length: 1330 Lines: 33 Hi, >>>>> "Thomas" == Thomas Riemer writes: Thomas> A couple of questions: Has the R3000 code been integrated into Thomas> the 1.3.58 kernel yet? Not yet, but preliminary support - great work Paul - is on it's way to integration for 1.3.59 by Ralf. Thomas> Does anyone know where I should look to find out what external Thomas> timer is shipped with Decstation 2100s,3000s? [SNIP] Thomas> CP0, but the R3000 doesn't seem to have this stuff at all. Thomas> (The true frustrationg is that every place the R4000 is Thomas> mentioned about timer stuff... there is not a single "pointer" Thomas> or reference as to what to do if you want a timer interrupt on Thomas> the R3000.) Hum I know nothing about that, being a Rseries weenie... Ralf, Paul ? Thomas> Digital.com is sadly lacking here... DecStations happened Thomas> before there was really a net to speak of - the documentation Thomas> just doesn't seem to have made it onto the web -at least as Thomas> far as I can tell. Any other ideas as to places to look? Dig in ftp://gatekeeper.dec.com, if you find something usefull please report it to Luc (luc.beurton@fnet.fr): he will add it - with pleasure I guess - to our FTP doc tree. BTW look in ftp://ftp.fnet.fr/linux-mips/doc/DEC perhaps there is already something... Cheers, Stoned. From linux@jens1.whu-koblenz.de Thu Feb 1 11:25:08 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id LAA07455 for ; Thu, 1 Feb 1996 11:25:08 +0100 (MET) Received: from obelix.WHU-Koblenz.DE by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA25595; Thu, 1 Feb 1996 11:24:10 +0100 (MET) Received: from jens1.whu-koblenz.de (linux@jens1.WHU-Koblenz.DE [193.141.48.51]) by obelix.WHU-Koblenz.de (8.6.8.1/8.6.4) with ESMTP id LAA23624 for ; Thu, 1 Feb 1996 11:24:02 +0100 Received: (from linux@localhost) by jens1.whu-koblenz.de (8.6.12/8.6.9) id LAA09082 for linux-mips@fnet.fr; Thu, 1 Feb 1996 11:23:51 +0100 From: linux-user Message-Id: <199602011023.LAA09082@jens1.whu-koblenz.de> Subject: Re: Bogons in make config? To: linux-mips@fnet.fr Date: Thu, 1 Feb 1996 11:23:49 +0100 (MET) In-Reply-To: <199601312059.VAA05290@informatik.uni-koblenz.de> from "Systemkennung Linux" at Jan 31, 96 10:00:11 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: 364 Lines: 14 Hi, Systemkennung Linux wrote: > This is my problem - the connection of uni-koblenz.de to anything except .de > is NULL by policy. This might change this year, but I'm not supposed to > know about project "Pink Elephant" :-> Tell me more :)) Sounds interesting not to know about it :) btw: Do you have a modem and spare 35,- DM per month? Greetings, Jens From linux@mailhost.uni-koblenz.de Thu Feb 1 12:03:44 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id MAA07736 for ; Thu, 1 Feb 1996 12:03:43 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA26261; Thu, 1 Feb 1996 12:03:25 +0100 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with SMTP id MAA23384 for ; Thu, 1 Feb 1996 12:02:33 +0100 From: Systemkennung Linux Message-Id: <199602011102.MAA23384@informatik.uni-koblenz.de> Received: by ozzy (5.x/KO-2.0) id AA27502; Thu, 1 Feb 1996 12:02:46 +0100 Subject: Re: R3000 support? To: linux-mips@fnet.fr Date: Thu, 1 Feb 1996 12:02:45 +0100 (MET) In-Reply-To: from "Thomas Riemer" at Jan 31, 96 05:11:35 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1421 Lines: 34 Hi, > Has the R3000 code been integrated into the 1.3.58 kernel yet? Yes, I've got 1.3.59 with the R3000 stuff on my disk at home. The R3000 stuff seems to be a bit "magic"; means though there is no visible bug it somehow triggers another bug for the Acer making the keyboard and floppy partially unusable for Acer. I suspect this might be the same bug Stoned reported recently. > It goes on to define TCSR_IE.. codes to enable timer interrupts. > etc. It does not however give me any way knowing if this really is > the right thing to be poking. At least it sound like a reasonable place to poke around. > The R4000 apparently has hardware timer interrupts built into CP0, > but the R3000 doesn't seem to have this stuff at all. (The true frustrationg > is that every place the R4000 is mentioned about timer stuff... there > is not a single "pointer" or reference as to what to do if you want > a timer interrupt on the R3000.) Though not all members of the R4000 family have the c0_count register. This is also why Linux doesn't use it (yet). > Digital.com is sadly lacking here... DecStations happened before there > was really a net to speak of - the documentation just doesn't seem to > have made it onto the web -at least as far as I can tell. Any other ideas > as to places to look? There are NetBSD and MACH ports to DECstations. Probably the best reference to get without asking DEC. Ralf From hodgen@mailhost.uni-koblenz.de Thu Feb 1 12:34:28 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id MAA07858 for ; Thu, 1 Feb 1996 12:34:27 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA26819; Thu, 1 Feb 1996 12:34:08 +0100 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with SMTP id MAA25384 for ; Thu, 1 Feb 1996 12:33:12 +0100 From: Wayne Hodgen Message-Id: <199602011133.MAA25384@informatik.uni-koblenz.de> Received: by ozzy (5.x/KO-2.0) id AA27581; Thu, 1 Feb 1996 12:33:26 +0100 Subject: Re: Bogons in make config? To: linux-mips@fnet.fr Date: Thu, 1 Feb 1996 12:33:25 +0100 (MET) In-Reply-To: <199602011023.LAA09082@jens1.whu-koblenz.de> from "linux-user" at Feb 1, 96 11:23:49 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 779 Lines: 16 > > This is my problem - the connection of uni-koblenz.de to anything except .de > > is NULL by policy. This might change this year, but I'm not supposed to > > know about project "Pink Elephant" :-> > > Tell me more :)) Sounds interesting not to know about it :) > > btw: Do you have a modem and spare 35,- DM per month? I'm sorry, you both know too much already. I'm afraid I have to kill both of you immediatly. Come quietly, resistance is futile! ;) -- Wayne Hodgen | hodgen@informatik.uni-koblenz.de | #include From linux@mailhost.uni-koblenz.de Thu Feb 1 12:40:41 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id MAA07928 for ; Thu, 1 Feb 1996 12:40:41 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA26920; Thu, 1 Feb 1996 12:40:03 +0100 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with SMTP id MAA25745 for ; Thu, 1 Feb 1996 12:38:16 +0100 From: Systemkennung Linux Message-Id: <199602011138.MAA25745@informatik.uni-koblenz.de> Received: by ozzy (5.x/KO-2.0) id AA27598; Thu, 1 Feb 1996 12:38:29 +0100 Subject: Re: Bogons in make config? To: linux-mips@fnet.fr Date: Thu, 1 Feb 1996 12:38:28 +0100 (MET) In-Reply-To: <199602011023.LAA09082@jens1.whu-koblenz.de> from "linux-user" at Feb 1, 96 11:23:49 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 561 Lines: 17 Hi, > > This is my problem - the connection of uni-koblenz.de to anything except .de > > is NULL by policy. This might change this year, but I'm not supposed to > > know about project "Pink Elephant" :-> > > Tell me more :)) Sounds interesting not to know about it :) There must be a reason that 'I wanna become a provider' is a quite popular idea among students in Koblenz. But since you seem not be interested in "Pink Elephant" I won't tell you :-) > btw: Do you have a modem and spare 35,- DM per month? Guess we need a Anti-Spam Policy :-> Ralf From linux@mailhost.uni-koblenz.de Thu Feb 1 12:42:16 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id MAA07971 for ; Thu, 1 Feb 1996 12:42:15 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA26947; Thu, 1 Feb 1996 12:41:35 +0100 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with SMTP id MAA25896 for ; Thu, 1 Feb 1996 12:40:32 +0100 From: Systemkennung Linux Message-Id: <199602011140.MAA25896@informatik.uni-koblenz.de> Received: by ozzy (5.x/KO-2.0) id AA27607; Thu, 1 Feb 1996 12:40:46 +0100 Subject: Re: Bogons in make config? To: linux-mips@fnet.fr Date: Thu, 1 Feb 1996 12:40:45 +0100 (MET) In-Reply-To: <199602011133.MAA25384@informatik.uni-koblenz.de> from "Wayne Hodgen" at Feb 1, 96 12:33:25 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 590 Lines: 17 Hi, > > > This is my problem - the connection of uni-koblenz.de to anything except .de > > > is NULL by policy. This might change this year, but I'm not supposed to > > > know about project "Pink Elephant" :-> > > > > Tell me more :)) Sounds interesting not to know about it :) > > > > btw: Do you have a modem and spare 35,- DM per month? > > I'm sorry, you both know too much already. I'm afraid I have to kill both > of you immediatly. Come quietly, resistance is futile! ;) Hey, don't you know - we're talking about these famous railroad tickets sold under that name :-) Ralf From zaitcev@ithil.mcst.ru Thu Feb 1 16:33:26 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id QAA00194 for ; Thu, 1 Feb 1996 16:33:25 +0100 (MET) From: zaitcev@ithil.mcst.ru Received: from ithil.mcst.ru by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA01012; Thu, 1 Feb 1996 16:33:14 +0100 (MET) Received: by ithil.mcst.ru (SMI-8.6/SMI-SVR4) id SAA02142; Thu, 1 Feb 1996 18:36:10 +0300 Date: Thu, 1 Feb 1996 18:36:10 +0300 Message-Id: <199602011536.SAA02142@ithil.mcst.ru> To: linux-mips@fnet.fr Subject: Re: Bogons in make config? Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Md5: zJil6vHg2+f20LRg1tP2TQ== Content-Length: 485 Lines: 11 > Mabye we could setup a CVS tree for our stuff and use SUP to > distribute it (or rcvs for committers) like FreeBSD and NetBSD do. It > works out really well there. I do not know what rcvs is. A regular cvs allows you to set CVSROOT to host.name.dom:/home/cvs. But it works via rsh which is not allowed at out master site. So Linux/SPARC developers telnet to the machine and issue cvs commits manualy. R/O CVS may be cool, but I am pretty happy with often (weekly) snapshots. Pete From imp@rover.village.org Thu Feb 1 17:36:12 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id RAA00398 for ; Thu, 1 Feb 1996 17:35:59 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA01209; Thu, 1 Feb 1996 17:35:40 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id JAA22948 for ; Thu, 1 Feb 1996 09:35:33 -0700 Message-Id: <199602011635.JAA22948@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Bogons in make config? In-Reply-To: Your message of Thu, 01 Feb 1996 18:36:10 +0300 Date: Thu, 01 Feb 1996 09:35:33 -0700 From: Warner Losh Content-Length: 509 Lines: 14 : I do not know what rcvs is. rcvs is the old name for the remote capability of cvs. : A regular cvs allows you to set CVSROOT : to host.name.dom:/home/cvs. But it works via rsh which is not : allowed at out master site. So Linux/SPARC developers telnet to : the machine and issue cvs commits manualy. R/O CVS may be cool, : but I am pretty happy with often (weekly) snapshots. It also works with Kerberos, which should be relatively secure. I have it setup here for some other stuff and like it. Warner From zaitcev@ithil.mcst.ru Thu Feb 1 17:52:01 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id RAA00479 for ; Thu, 1 Feb 1996 17:52:01 +0100 (MET) From: zaitcev@ithil.mcst.ru Received: from chenas.inria.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA01912; Thu, 1 Feb 1996 17:51:52 +0100 (MET) Received: from ithil.mcst.ru by chenas.inria.fr (5.65c8d/92.02.29) via EUnet-France id AA04956; Thu, 1 Feb 1996 17:51:40 +0100 (MET) Received: by ithil.mcst.ru (SMI-8.6/SMI-SVR4) id TAA02534; Thu, 1 Feb 1996 19:46:23 +0300 Date: Thu, 1 Feb 1996 19:46:23 +0300 Message-Id: <199602011646.TAA02534@ithil.mcst.ru> To: linux-mips@fnet.fr Subject: Re: Bogons in make config? Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Md5: Q7NVKk9woeEMuYKazQMkhA== Content-Length: 337 Lines: 10 > : A regular cvs allows you to set CVSROOT > : to host.name.dom:/home/cvs. But it works via rsh which is not > : allowed at out master site. > > It also works with Kerberos, which should be relatively secure. I > have it setup here for some other stuff and like it. AFAIK, I cannot use Kerberos in Russia. Do I miss something? Pete From linux@jens1.whu-koblenz.de Thu Feb 1 18:01:57 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id SAA00751 for ; Thu, 1 Feb 1996 18:01:56 +0100 (MET) Received: from obelix.WHU-Koblenz.DE by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA02522; Thu, 1 Feb 1996 18:01:43 +0100 (MET) Received: from jens1.whu-koblenz.de (linux@jens1.WHU-Koblenz.DE [193.141.48.51]) by obelix.WHU-Koblenz.de (8.6.8.1/8.6.4) with ESMTP id SAA02065 for ; Thu, 1 Feb 1996 18:01:21 +0100 Received: (from linux@localhost) by jens1.whu-koblenz.de (8.6.12/8.6.9) id SAA09441 for linux-mips@fnet.fr; Thu, 1 Feb 1996 18:01:03 +0100 From: linux-user Message-Id: <199602011701.SAA09441@jens1.whu-koblenz.de> Subject: Predator is threatening! To: linux-mips@fnet.fr Date: Thu, 1 Feb 1996 18:01:02 +0100 (MET) In-Reply-To: <199602011133.MAA25384@informatik.uni-koblenz.de> from "Wayne Hodgen" at Feb 1, 96 12:33:25 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: 285 Lines: 15 Hi Predator :)) Wayne Hodgen wrote: > I'm sorry, you both know too much already. I'm afraid I have to kill both > of you immediatly. Come quietly, resistance is futile! ;) health ministers warn: "MUDding destroys you manners" grin bow predator sarcastically greetings, Jens From linux@jens1.whu-koblenz.de Thu Feb 1 18:03:57 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id SAA00774 for ; Thu, 1 Feb 1996 18:03:57 +0100 (MET) Received: from obelix.WHU-Koblenz.DE by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA02619; Thu, 1 Feb 1996 18:03:33 +0100 (MET) Received: from jens1.whu-koblenz.de (linux@jens1.WHU-Koblenz.DE [193.141.48.51]) by obelix.WHU-Koblenz.de (8.6.8.1/8.6.4) with ESMTP id SAA02402 for ; Thu, 1 Feb 1996 18:03:41 +0100 Received: (from linux@localhost) by jens1.whu-koblenz.de (8.6.12/8.6.9) id SAA09453 for linux-mips@fnet.fr; Thu, 1 Feb 1996 18:03:23 +0100 From: linux-user Message-Id: <199602011703.SAA09453@jens1.whu-koblenz.de> Subject: Re: Bogons in make config? To: linux-mips@fnet.fr Date: Thu, 1 Feb 1996 18:03:22 +0100 (MET) In-Reply-To: <199602011138.MAA25745@informatik.uni-koblenz.de> from "Systemkennung Linux" at Feb 1, 96 12:38:28 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: 440 Lines: 22 Hi, Systemkennung Linux wrote: > There must be a reason that 'I wanna become a provider' is a quite popular > idea among students in Koblenz. But since you seem not be interested in > "Pink Elephant" I won't tell you :-) Too late, I am one already.. > > btw: Do you have a modem and spare 35,- DM per month? > Guess we need a Anti-Spam Policy :-> Guilty... Greetings, Jens PS: Do you like a connection? It's up and running :) From linux@mailhost.uni-koblenz.de Thu Feb 1 18:08:42 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id SAA00834 for ; Thu, 1 Feb 1996 18:08:41 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA02824; Thu, 1 Feb 1996 18:08:15 +0100 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with SMTP id SAA03503 for ; Thu, 1 Feb 1996 18:07:23 +0100 From: Systemkennung Linux Message-Id: <199602011707.SAA03503@informatik.uni-koblenz.de> Received: by ozzy (5.x/KO-2.0) id AA28660; Thu, 1 Feb 1996 18:07:36 +0100 Subject: Re: Bogons in make config? To: linux-mips@fnet.fr Date: Thu, 1 Feb 1996 18:07:35 +0100 (MET) In-Reply-To: <199602011646.TAA02534@ithil.mcst.ru> from "zaitcev@ithil.mcst.ru" at Feb 1, 96 07:46:23 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 551 Lines: 16 Hi, > > : A regular cvs allows you to set CVSROOT > > : to host.name.dom:/home/cvs. But it works via rsh which is not > > : allowed at out master site. > > > > It also works with Kerberos, which should be relatively secure. I > > have it setup here for some other stuff and like it. > > AFAIK, I cannot use Kerberos in Russia. Do I miss something? There is a small difference between authentication and encryption. It is my understanding of the French/Russian law still lives using cryptographical algorithems for authentication legal. Ralf From imp@rover.village.org Thu Feb 1 19:43:15 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id TAA00141 for ; Thu, 1 Feb 1996 19:43:13 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA02195; Thu, 1 Feb 1996 19:42:27 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id LAA23259 for ; Thu, 1 Feb 1996 11:07:49 -0700 Message-Id: <199602011807.LAA23259@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Bogons in make config? In-Reply-To: Your message of Thu, 01 Feb 1996 18:07:35 +0100 Date: Thu, 01 Feb 1996 11:07:48 -0700 From: Warner Losh Content-Length: 1109 Lines: 25 : Hi, : : > > : A regular cvs allows you to set CVSROOT : > > : to host.name.dom:/home/cvs. But it works via rsh which is not : > > : allowed at out master site. : > > : > > It also works with Kerberos, which should be relatively secure. I : > > have it setup here for some other stuff and like it. : > : > AFAIK, I cannot use Kerberos in Russia. Do I miss something? : There is a small difference between authentication and encryption. It : is my understanding of the French/Russian law still lives using : cryptographical algorithems for authentication legal. It is my understanding as well that French and Russian law allow Kerberos for authentication, but forbid the use of encryption to hide data these governments are interested in (which, by definition, is everything). FreeBSD (and I think NetBSD) ship with a "sanitized" version of kerberos that does authentication, but not encryption. There is some software available from a server in south africa that supplies the missing encryption parts, but I don't have that address handy right now (I just grab eBones from ftp.freebsd.org). Warner From linux@mailhost.uni-koblenz.de Thu Feb 1 19:56:35 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id TAA00214 for ; Thu, 1 Feb 1996 19:56:34 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA02393; Thu, 1 Feb 1996 19:56:25 +0100 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with SMTP id TAA07974 for ; Thu, 1 Feb 1996 19:55:41 +0100 From: Systemkennung Linux Message-Id: <199602011855.TAA07974@informatik.uni-koblenz.de> Received: by ozzy (5.x/KO-2.0) id AA29006; Thu, 1 Feb 1996 19:55:53 +0100 Subject: Re: Bogons in make config? To: linux-mips@fnet.fr Date: Thu, 1 Feb 1996 19:55:52 +0100 (MET) In-Reply-To: <199602011807.LAA23259@rover.village.org> from "Warner Losh" at Feb 1, 96 11:07:48 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 889 Lines: 20 Hi, > It is my understanding as well that French and Russian law allow > Kerberos for authentication, but forbid the use of encryption to hide > data these governments are interested in (which, by definition, is > everything). FreeBSD (and I think NetBSD) ship with a "sanitized" > version of kerberos that does authentication, but not encryption. > There is some software available from a server in south africa that > supplies the missing encryption parts, but I don't have that address > handy right now (I just grab eBones from ftp.freebsd.org). There is some server in the domain hacktick.nl that is specialized on complete and uncensored cryptosoftware like cfs, pgp and so on. Wouldn't wonder if they also have Kerberos. I've got PGP for Linux/MIPS and wonder how we should best organize distribution of the binaries so that we have the least possible legal problems. Ralf From davem@caip.rutgers.edu Thu Feb 1 20:28:49 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id UAA00501 for ; Thu, 1 Feb 1996 20:28:48 +0100 (MET) Received: from caipfs.rutgers.edu by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA02860; Thu, 1 Feb 1996 20:28:40 +0100 (MET) Received: from huahaga.rutgers.edu (huahaga.rutgers.edu [128.6.155.53]) by caipfs.rutgers.edu (8.6.9+bestmx+oldruq+newsunq+grosshack/8.6.9) with ESMTP id OAA01069 for ; Thu, 1 Feb 1996 14:28:45 -0500 Received: (davem@localhost) by huahaga.rutgers.edu (8.6.9+bestmx+oldruq+newsunq+grosshack/8.6.9) id OAA22729; Thu, 1 Feb 1996 14:28:45 -0500 Date: Thu, 1 Feb 1996 14:28:45 -0500 Message-Id: <199602011928.OAA22729@huahaga.rutgers.edu> From: "David S. Miller" To: linux-mips@fnet.fr In-Reply-To: <199602011646.TAA02534@ithil.mcst.ru> (zaitcev@ithil.mcst.ru) Subject: Re: Bogons in make config? Content-Length: 856 Lines: 23 From: zaitcev@ithil.mcst.ru Date: Thu, 1 Feb 1996 19:46:23 +0300 > : A regular cvs allows you to set CVSROOT > : to host.name.dom:/home/cvs. But it works via rsh which is not > : allowed at out master site. > > It also works with Kerberos, which should be relatively secure. I > have it setup here for some other stuff and like it. AFAIK, I cannot use Kerberos in Russia. Do I miss something? All the kerberos limitations (if there are any, I am no expert in such things) are meaningless with the latest version of cvs which just hit the ftp sites. It has password capabilities built into the server which should help cases where kerberos/ssh is not an option. I will be looking into either ssh or cvs-passwd usage for the Sparc Linux developers to see which works the best. Later, David S. Miller davem@caip.rutgers.edu From paul@suite.sw.oz.au Fri Feb 2 00:54:12 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id AAA01737 for ; Fri, 2 Feb 1996 00:54:11 +0100 (MET) Received: from staff.cs.su.OZ.AU by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA06803; Fri, 2 Feb 1996 00:54:01 +0100 (MET) Received: from suite.sw.oz.au by staff.cs.su.OZ.AU (mail from paul for linux-mips@fnet.fr) with MHSnet (insertion MHSnet site: swallow.sw.oz.au); Fri, 02 Feb 1996 10:53:59 +1100 Received: from suite.sw.oz.au by swallow.sw.oz.au with SMTP id XAA00789; Thu, 1 Feb 1996 23:53:49 GMT (8.6.10/Unixware) (from paul@suite.sw.oz.au for ) Received: by suite.sw.oz.au id AA08456; Fri, 2 Feb 1996 10:53:40 +1100 (5.65c/1.34) (from paul@suite.sw.oz.au for linux-mips@fnet.fr) From: paul@suite.sw.oz.au (Paul Antoine) Message-Id: <199602012353.AA08456@suite.sw.oz.au> Subject: Re: Bogons in make config? To: linux-mips@fnet.fr Date: Fri, 2 Feb 1996 10:53:40 +1100 (EST) In-Reply-To: <199601312059.VAA05290@informatik.uni-koblenz.de> from "Systemkennung Linux" at Jan 31, 96 10:00:11 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: 810 Lines: 20 Hi folks, Ralf wrote; > This is my problem - the connection of uni-koblenz.de to anything except .de > is NULL by policy. This might change this year, but I'm not supposed to > know about project "Pink Elephant" :-> You guys have obviously been drinking toooo much beer! (Urban myth; Pink Elephant's are seen when withdrawing from alchohol addiction :-) 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 2 698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From paul@suite.sw.oz.au Fri Feb 2 00:59:54 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id AAA01787 for ; Fri, 2 Feb 1996 00:59:53 +0100 (MET) Received: from staff.cs.su.OZ.AU by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA07050; Fri, 2 Feb 1996 00:59:42 +0100 (MET) Received: from suite.sw.oz.au by staff.cs.su.OZ.AU (mail from paul for linux-mips@fnet.fr) with MHSnet (insertion MHSnet site: swallow.sw.oz.au); Fri, 02 Feb 1996 10:59:42 +1100 Received: from suite.sw.oz.au by swallow.sw.oz.au with SMTP id XAA01080; Thu, 1 Feb 1996 23:59:05 GMT (8.6.10/Unixware) (from paul@suite.sw.oz.au for ) Received: by suite.sw.oz.au id AA08504; Fri, 2 Feb 1996 10:58:46 +1100 (5.65c/1.34) (from paul@suite.sw.oz.au for linux-mips@fnet.fr) From: paul@suite.sw.oz.au (Paul Antoine) Message-Id: <199602012358.AA08504@suite.sw.oz.au> Subject: Re: R3000 support? To: linux-mips@fnet.fr Date: Fri, 2 Feb 1996 10:58:45 +1100 (EST) In-Reply-To: from "Thomas Riemer" at Jan 31, 96 05:11:35 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: 2191 Lines: 55 Hello people, > A couple of questions: > > Has the R3000 code been integrated into the 1.3.58 kernel yet? Patience my friend! Ralf's fingers are a blurr accross his keyboard as we speak (right, Ralf? :-) > Does anyone know where I should look to find out what external timer > is shipped with Decstation 2100s,3000s? > > The Ultrix header file mips/kn220.h has a #define: > #define TCSR PHYS_TO_K1(0x10084010) /* R3000 Interval Timer Register > (ITR) */ > > It goes on to define TCSR_IE.. codes to enable timer interrupts. > etc. It does not however give me any way knowing if this really is > the right thing to be poking. Yeah - frustrating isn't it. I suspect that the DECStation motherboard ASIC's implement a timer function, and that these addresses refer to the on-chip registers for accessing it (basically turning it on or off, as it is just a periodic ticker methinx). I will put the relevent MACH sources up on my ftp server for you to poke around in... > The R4000 apparently has hardware timer interrupts built into CP0, > but the R3000 doesn't seem to have this stuff at all. (The true frustrationg > is that every place the R4000 is mentioned about timer stuff... there > is not a single "pointer" or reference as to what to do if you want > a timer interrupt on the R3000.) Build one with support chips! > Digital.com is sadly lacking here... DecStations happened before there > was really a net to speak of - the documentation just doesn't seem to > have made it onto the web -at least as far as I can tell. Any other ideas > as to places to look? Yeah - I gather that the MACH team had access to internal DEC documents. Perhaps we could get Linus to convince DEC to release these docs... 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 2 698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From paul@suite.sw.oz.au Fri Feb 2 03:21:32 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id DAA04419 for ; Fri, 2 Feb 1996 03:21:31 +0100 (MET) Received: from staff.cs.su.OZ.AU by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA08667; Fri, 2 Feb 1996 03:21:20 +0100 (MET) Received: from suite.sw.oz.au by staff.cs.su.OZ.AU (mail from paul for linux-mips@fnet.fr) with MHSnet (insertion MHSnet site: swallow.sw.oz.au); Fri, 02 Feb 1996 13:21:25 +1100 Received: from suite.sw.oz.au by swallow.sw.oz.au with SMTP id CAA06949; Fri, 2 Feb 1996 02:21:05 GMT (8.6.10/Unixware) (from paul@suite.sw.oz.au for ) Received: by suite.sw.oz.au id AA10702; Fri, 2 Feb 1996 13:20:46 +1100 (5.65c/1.34) (from paul@suite.sw.oz.au for linux-mips@fnet.fr) From: paul@suite.sw.oz.au (Paul Antoine) Message-Id: <199602020220.AA10702@suite.sw.oz.au> Subject: Re: DECStation 5000 To: miguel@roxanne.nuclecu.unam.mx (Miguel de Icaza) Date: Fri, 2 Feb 1996 13:20:45 +1100 (EST) Cc: linux-mips@fnet.fr (Linux MIPS mailing list) In-Reply-To: <199601310532.XAA10313@roxanne.nuclecu.unam.mx> from "Miguel de Icaza" at Jan 30, 96 11:32:13 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: 1210 Lines: 35 Hi Miguel, (Miguel's away from his home account, so I'm cc'ing this to linux-mips, as it has some good news). > I;m still in the USA, so I can't code anything, but I was told by > the guys of SGI Mexico that they would be giving me a machine to > work on the linux/sgi port. That's great news! > Anyways, could you point me to the location where I could get the > Linux/MIPS files? (Sorry, I forgot where they are and I dont ahve > access to the account that is subscribed to linux-mips). The location is: ftp://ftp.fnet.fr/linux-mips The R3000 stuff for the DECStation has not been released as yet, but is expected soon. Andreas will probably want to talk to you about helping with the SGI port... if you haven't already spoken to him. 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 2 698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From miguel@roxanne.nuclecu.unam.mx Fri Feb 2 03:52:27 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id DAA04517 for ; Fri, 2 Feb 1996 03:52:26 +0100 (MET) Received: from roxanne.nuclecu.unam.mx by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA08943; Fri, 2 Feb 1996 03:52:15 +0100 (MET) Received: (from miguel@localhost) by roxanne.nuclecu.unam.mx (8.6.12/8.6.11) id UAA11423; Thu, 1 Feb 1996 20:50:17 -0600 Date: Thu, 1 Feb 1996 20:50:17 -0600 From: Miguel de Icaza Message-Id: <199602020250.UAA11423@roxanne.nuclecu.unam.mx> To: paul@suite.sw.oz.au Cc: linux-mips@fnet.fr In-Reply-To: <199602020220.AA10702@suite.sw.oz.au> (paul@suite.sw.oz.au) Subject: Re: DECStation 5000 X-Pecado: Tanto peca el que mata a la vaca como la vaca. Content-Length: 842 Lines: 27 Paul said: > > I;m still in the USA, so I can't code anything, but I was told by > > the guys of SGI Mexico that they would be giving me a machine to > > work on the linux/sgi port. > > That's great news! Oh yes. Thanks to Larry McVoy for sending a convincing mail message to the guy in SGI Mexico. > The R3000 stuff for the DECStation has not been released as yet, but > is expected soon. Ok, I will be waiting for it, in the meantime, I will continue my work with David Miller on his Sparc port. > Andreas will probably want to talk to you about helping with the SGI > port... if you haven't already spoken to him. I haven't talked to him yet, because the sgi port was more or less a surprise to me, the guy at sgi mexico called and told me if I was interested in getting the machine just before I left Mexico. Cheers, Miguel. From lm@neteng.engr.sgi.com Fri Feb 2 04:36:04 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id EAA04832 for ; Fri, 2 Feb 1996 04:36:03 +0100 (MET) Received: from sgi.sgi.com (SGI.COM) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA09360; Fri, 2 Feb 1996 04:35:55 +0100 (MET) Received: from neteng.engr.sgi.com by sgi.sgi.com via ESMTP (950405.SGI.8.6.12/910110.SGI) for <@sgi.engr.sgi.com:linux-mips@fnet.fr> id TAA21997; Thu, 1 Feb 1996 19:35:59 -0800 Received: from localhost by neteng.engr.sgi.com via SMTP (950413.SGI.8.6.12/940406.SGI.AUTO) for id TAA23670; Thu, 1 Feb 1996 19:35:59 -0800 Message-Id: <199602020335.TAA23670@neteng.engr.sgi.com> To: linux-mips@fnet.fr From: lm@neteng.engr.sgi.com (Larry McVoy) Subject: Re: DECStation 5000 Date: Thu, 01 Feb 1996 19:35:58 -0800 Sender: lm@neteng.engr.sgi.com Content-Length: 657 Lines: 14 I'm also getting Theo Deraadt an old R3K machine - he's going to get a OpenBSD port working on it along with some superstud in Sweden. He promised me that all the stuff he did would be released under both the BSD and the GPL copyright (i.e., two versions, same code, different copyrights) so that we can use it Linux. Theo is pretty studly. If he got a lot of strokes about how great of a job he was doing on the OpenBSD stuff, he might decide that our crowd is more fun than his crowd. A thought for the day. Miguel, let me know if the machine you got isn't a R4K. I want an R4K port. I'm also willing to try and get driver sources from IRIX. --lm From zaitcev@ithil.mcst.ru Fri Feb 2 09:01:31 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id JAA06257 for ; Fri, 2 Feb 1996 09:01:31 +0100 (MET) From: zaitcev@ithil.mcst.ru Received: from ithil.mcst.ru by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA12766; Fri, 2 Feb 1996 09:00:59 +0100 (MET) Received: by ithil.mcst.ru (SMI-8.6/SMI-SVR4) id LAA02875; Fri, 2 Feb 1996 11:03:48 +0300 Date: Fri, 2 Feb 1996 11:03:48 +0300 Message-Id: <199602020803.LAA02875@ithil.mcst.ru> To: linux-mips@fnet.fr Subject: Re: Bogons in make config? Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Md5: sgpuBUTViebQJaxmI/VEVA== Content-Length: 527 Lines: 13 > From: Warner Losh >[...] > It is my understanding as well that French and Russian law allow > Kerberos for authentication, but forbid the use of encryption to hide > data these governments are interested in (which, by definition, is > everything). How I can get my hands on Kerberos if it is forbidden for export? If someone in the .za stole it, can I download it and to tell 'I got it from outside of the USA, so I did not violated any code.'? This is a problem with american law, not with russian. Pete From imp@rover.village.org Fri Feb 2 15:01:23 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id PAA08104 for ; Fri, 2 Feb 1996 15:01:22 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA20201; Fri, 2 Feb 1996 15:00:57 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id HAA24806 for ; Fri, 2 Feb 1996 07:00:51 -0700 Message-Id: <199602021400.HAA24806@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Bogons in make config? In-Reply-To: Your message of Thu, 01 Feb 1996 19:55:52 +0100 Date: Fri, 02 Feb 1996 07:00:50 -0700 From: Warner Losh Content-Length: 547 Lines: 14 : I've got PGP for Linux/MIPS and wonder how we should best organize : distribution of the binaries so that we have the least possible : legal problems. Take PGP out :-(. Generally speaking, you'll find that any encryption software tends to have very nasty restrictions associated with it. Most people here in the states wouldn't touch it with a 10' pole. I'm not sure what the laws are in Germany on this, but it may save a lot of grief if you don't distribute it. You could distribute patches and have people build it themselves... Warner From imp@rover.village.org Fri Feb 2 15:03:22 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id PAA08138 for ; Fri, 2 Feb 1996 15:03:22 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA20225; Fri, 2 Feb 1996 15:02:53 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id HAA24824 for ; Fri, 2 Feb 1996 07:02:45 -0700 Message-Id: <199602021402.HAA24824@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Bogons in make config? In-Reply-To: Your message of Fri, 02 Feb 1996 11:03:48 +0300 Date: Fri, 02 Feb 1996 07:02:44 -0700 From: Warner Losh Content-Length: 581 Lines: 17 : How I can get my hands on Kerberos if it is forbidden for export? All of Kerberos isn't forbidden for export. Just those parts of it that are for encryption. Silly law, I know, but the law here none the less. : If someone in the .za stole it, can I download it and to tell : 'I got it from outside of the USA, so I did not violated any code.'? Well, someone in .za reimplemented the encryption parts... : This is a problem with american law, not with russian. Yes. The americans seem to think they can pass laws that apply to people outside of their borders :-(. Warner From linux@mailhost.uni-koblenz.de Fri Feb 2 17:28:24 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id RAA08896 for ; Fri, 2 Feb 1996 17:28:23 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA23242; Fri, 2 Feb 1996 17:28:02 +0100 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with SMTP id RAA22104 for ; Fri, 2 Feb 1996 17:27:19 +0100 From: Systemkennung Linux Message-Id: <199602021627.RAA22104@informatik.uni-koblenz.de> Received: by ozzy (5.x/KO-2.0) id AA29998; Fri, 2 Feb 1996 17:27:23 +0100 Subject: Re: R3000 support? To: linux-mips@fnet.fr Date: Fri, 2 Feb 1996 17:27:22 +0100 (MET) In-Reply-To: <199602012358.AA08504@suite.sw.oz.au> from "Paul Antoine" at Feb 2, 96 10:58:45 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 413 Lines: 11 Hi, > > Digital.com is sadly lacking here... DecStations happened before there > > was really a net to speak of - the documentation just doesn't seem to > > have made it onto the web -at least as far as I can tell. Any other ideas > > as to places to look? Some time ago someone post that Jim Paradis (head of Digital's Linux/Alpha team) offered help about getting the required technical information. Ralf From linux@mailhost.uni-koblenz.de Fri Feb 2 17:32:58 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id RAA08972 for ; Fri, 2 Feb 1996 17:32:58 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA23427; Fri, 2 Feb 1996 17:32:38 +0100 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with SMTP id RAA22518 for ; Fri, 2 Feb 1996 17:31:57 +0100 From: Systemkennung Linux Message-Id: <199602021631.RAA22518@informatik.uni-koblenz.de> Received: by ozzy (5.x/KO-2.0) id AA00015; Fri, 2 Feb 1996 17:32:00 +0100 Subject: Re: Bogons in make config? To: linux-mips@fnet.fr Date: Fri, 2 Feb 1996 17:32:00 +0100 (MET) In-Reply-To: <199602021400.HAA24806@rover.village.org> from "Warner Losh" at Feb 2, 96 07:00:50 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 501 Lines: 13 Hi, > Generally speaking, you'll find that any encryption software tends to > have very nasty restrictions associated with it. Most people here in > the states wouldn't touch it with a 10' pole. I'm not sure what the > laws are in Germany on this, but it may save a lot of grief if you > don't distribute it. You could distribute patches and have people > build it themselves... The German law is perfect in this direction - I can export all non classified cryptostuff. Well, still ... Ralf From imp@rover.village.org Fri Feb 2 23:58:32 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id XAA10795 for ; Fri, 2 Feb 1996 23:58:31 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA29421; Fri, 2 Feb 1996 23:58:10 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id PAA09638 for ; Fri, 2 Feb 1996 15:58:08 -0700 Message-Id: <199602022258.PAA09638@rover.village.org> To: linux-mips@fnet.fr Subject: New cross compiler building reference Date: Fri, 02 Feb 1996 15:58:08 -0700 From: Warner Losh Content-Length: 325 Lines: 11 I've updated the cross compiler builder page to include the right (I hope) links to fnet.fr and to reflect the cut over to 2.6 binutils and 2.7.2 gcc. Please check it out! http://www.village.org/pub/villagers/imp/mips/build.html Also, how do I upload this to fnet.fr for inclusion on that page? Thanks a bunch... Warner From linux@mailhost.uni-koblenz.de Sat Feb 3 20:06:57 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id UAA03124 for ; Sat, 3 Feb 1996 20:06:51 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA05654; Sat, 3 Feb 1996 20:06:45 +0100 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with SMTP id UAA24636 for ; Sat, 3 Feb 1996 20:06:09 +0100 From: Systemkennung Linux Message-Id: <199602031906.UAA24636@informatik.uni-koblenz.de> Received: by ozzy (5.x/KO-2.0) id AA11253; Sat, 3 Feb 1996 20:06:06 +0100 Subject: Re: New cross compiler building reference To: linux-mips@fnet.fr Date: Sat, 3 Feb 1996 20:06:05 +0100 (MET) In-Reply-To: <199602022258.PAA09638@rover.village.org> from "Warner Losh" at Feb 2, 96 03:58:08 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 513 Lines: 17 Hi, > I've updated the cross compiler builder page to include the right (I > hope) links to fnet.fr and to reflect the cut over to 2.6 binutils and > 2.7.2 gcc. Please check it out! Thanks, I'll check it out immediately. > http://www.village.org/pub/villagers/imp/mips/build.html > > Also, how do I upload this to fnet.fr for inclusion on that page? No uploads to ftp.fnet.fr by policy. You'll have to talk to Luc & Stoned about an account on lena.fnet.fr. For now I'll upload your page to lena. Ralf From imp@rover.village.org Sat Feb 3 22:20:55 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id WAA03808 for ; Sat, 3 Feb 1996 22:20:54 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA06897; Sat, 3 Feb 1996 22:20:49 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id OAA13493 for ; Sat, 3 Feb 1996 14:20:47 -0700 Message-Id: <199602032120.OAA13493@rover.village.org> To: linux-mips@fnet.fr Subject: Debugger support? Date: Sat, 03 Feb 1996 14:20:47 -0700 From: Warner Losh Content-Length: 1504 Lines: 33 This is with the 1.3.58.1 kernel and MILO 0.26 + Ralf's patches + my deskstation patches + cleanup compile patches. I'm hard wiring the memory addresses as 0x800100000 for the start of memory (rather than _end due to the hole in memory) and 0x82000000 as the end address (32M total) in the early parts of init so that the tagged reads work right (MILO doesn't seem to be setting these correctly, but that could be my imagination). Anyway, I finy my self in need of a debugger for the linux-mips kernel.... I seem to recall that there was some work on this, but I'm not sure how far it got along. I've also noticed that hooks for the debugger seem to be enabled after the point I'm having problems at :-(. I'm getting to the following line in init_swap_cache and hanging: memset(swap_cache, 0, swap_cache_size * sizeof (unsigned long)); I have no clue what swap_cache_size is, and printf doesn't work at this stage of the boot process (since the console driver isn't up yet). swap_cache_size should be 0x2000, which * 4 = 0x8000 which should take very little time to finish on a 50MHz machine. If I comment out this line, then the next problem is somewhere after the init_modules line in the kernel, but before the kernel prints anything, so I'm rather keen on figuring out what the deal is (most likely it is in the console_init routine, but don't hold me to that). Ideas? If all else fails, I can hack together a tiny routine that prints a number to the right locations in memory... Warner From linux@mailhost.uni-koblenz.de Sat Feb 3 22:53:07 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id WAA03908 for ; Sat, 3 Feb 1996 22:53:06 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA07156; Sat, 3 Feb 1996 22:53:00 +0100 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with SMTP id WAA29445 for ; Sat, 3 Feb 1996 22:52:25 +0100 From: Systemkennung Linux Message-Id: <199602032152.WAA29445@informatik.uni-koblenz.de> Received: by ozzy (5.x/KO-2.0) id AA11385; Sat, 3 Feb 1996 22:52:21 +0100 Subject: Re: Debugger support? To: linux-mips@fnet.fr Date: Sat, 3 Feb 1996 22:52:19 +0100 (MET) In-Reply-To: <199602032120.OAA13493@rover.village.org> from "Warner Losh" at Feb 3, 96 02:20:47 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 2179 Lines: 51 Hi, > This is with the 1.3.58.1 kernel and MILO 0.26 + Ralf's patches + my > deskstation patches + cleanup compile patches. I'm hard wiring the > memory addresses as 0x800100000 for the start of memory (rather than > _end due to the hole in memory) and 0x82000000 as the end address (32M > total) in the early parts of init so that the tagged reads work right > (MILO doesn't seem to be setting these correctly, but that could be my > imagination). The kernel makes the assumption that the memory from _end to end of the physical memory is available for use. This is obviously wrong for your machine. > Anyway, I finy my self in need of a debugger for the linux-mips > kernel.... I seem to recall that there was some work on this, but I'm > not sure how far it got along. I've also noticed that hooks for the > debugger seem to be enabled after the point I'm having problems at > :-(. The debugger support was written by Andy more than half a year ago. On my 8mb machine linking with debugging enabled takes more than 5 minutes, and it is still realativly slow on Andy's 32mb machine, so I never used it. No idea if it still useable ... > I'm getting to the following line in init_swap_cache and hanging: > > memset(swap_cache, 0, swap_cache_size * sizeof (unsigned long)); > > I have no clue what swap_cache_size is, and printf doesn't work at > this stage of the boot process (since the console driver isn't up > yet). swap_cache_size should be 0x2000, which * 4 = 0x8000 which Correct. > should take very little time to finish on a 50MHz machine. If I > comment out this line, then the next problem is somewhere after the > init_modules line in the kernel, but before the kernel prints > anything, so I'm rather keen on figuring out what the deal is (most > likely it is in the console_init routine, but don't hold me to that). > > Ideas? > If all else fails, I can hack together a tiny routine that prints a > number to the right locations in memory... Take a look at arch/mips/lib/console.c. It the console that I and Andy were using before we built support into the normal console.c. Tiny, incomplete and slow but has been extremly usefull. Ralf From imp@rover.village.org Sat Feb 3 23:12:47 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id XAA04178 for ; Sat, 3 Feb 1996 23:12:46 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA07389; Sat, 3 Feb 1996 23:12:40 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id PAA13617 for ; Sat, 3 Feb 1996 15:12:31 -0700 Message-Id: <199602032212.PAA13617@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Debugger support? In-Reply-To: Your message of Sat, 03 Feb 1996 22:52:19 +0100 Date: Sat, 03 Feb 1996 15:12:30 -0700 From: Warner Losh Content-Length: 527 Lines: 14 : The kernel makes the assumption that the memory from _end to end of : the physical memory is available for use. This is obviously wrong for : your machine. Yes. I've patched that on my machine to use KSEG0 + (1 << 20) as the start address, and KSEG0 + (32 << 20) as the end address. : Take a look at arch/mips/lib/console.c. It the console that I and Andy : were using before we built support into the normal console.c. Tiny, : incomplete and slow but has been extremly usefull. OK. I'll take a look at it... Warner From linux@mailhost.uni-koblenz.de Sun Feb 4 03:25:10 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id DAA07177 for ; Sun, 4 Feb 1996 03:25:09 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA10051; Sun, 4 Feb 1996 03:25:03 +0100 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with SMTP id DAA24255 for ; Sun, 4 Feb 1996 03:24:31 +0100 From: Systemkennung Linux Message-Id: <199602040224.DAA24255@informatik.uni-koblenz.de> Received: by ozzy (5.x/KO-2.0) id AA11580; Sun, 4 Feb 1996 03:24:25 +0100 Subject: Re: 1.3.58 on Magnum? To: linux-mips@fnet.fr Date: Sun, 4 Feb 1996 03:24:24 +0100 (MET) In-Reply-To: <9602010245.AA23471@silver.sni.CA> from "David Kerry" at Dec 31, 95 09:45:24 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 2705 Lines: 58 Hi David, Take a look at your mails headers: > Resent-Date: Thu, 1 Feb 1996 03:36:37 +0100 (MET) > Date: Sun, 31 Dec 1995 21:45:24 -0500 (Seems your clock is wrong by one month. That why my elm sorted your mail to the other ancient stuff ...) > Hello All, > > I'm in the process of catching up again with the current > progress (real busy at the other office :), and have run into a snag. > I've picked up the binutils-2.6-1.diffs and the gcc-2.7.2-1.diffs and > applied them to create a cross-development environment > (target=mipsel-linuxelf), and the linux-1.3.58.tar.gz source > from fnet.fr. So far good. The build seems to make it most of the way through > until it hits the magnum4000.S file upon which the assembler complains > (err, this is configured for a Magnum 4000, elf compiler, extra elf > compiler by the way): The ELF compiler may only be used for the user programs. There is some problem with the ELF assembler or linker that prevents to use these tools also for the kernel. Instead you should use the mipsel-linux tools for building the kernel. I know, it's annoying having to have these two flavours of the tools ... > mipsel-linuxelf-gcc -D__KERNEL__ -I/home/davidk/mips/kernel/v1.3/linux/include\ > -Wall -Wstrict-prototypes -O2 -fomit-frame-pointer -G0 -mno-abicalls\ > -fno-pic -D__R4000__ -mcpu=r4400 -mips2 -c -o jazzdma.o jazzdma.c > mipsel-linuxelf-gcc -D__KERNEL__ -I/home/davidk/mips/kernel/v1.3/linux/include\ > -Wall -Wstrict-prototypes -O2 -fomit-frame-pointer -G0 -mno-abicalls\ > -fno-pic -D__R4000__ -mcpu=r4400 -mips2 -c -o ksyms.o ksyms.c > /home/davidk/mips/kernel/v1.3/linux/include/asm/floppy.h:60: warning: `FDC2' defined but not used > mipsel-linuxelf-gcc -D__KERNEL__ -I/home/davidk/mips/kernel/v1.3/linux/include\ > -Wall -Wstrict-prototypes -O2 -fomit-frame-pointer -G0 -mno-abicalls\ > -fno-pic -D__R4000__ -mcpu=r4400 -mips2 -c magnum4000.S -o magnum4000.o > magnum4000.S: Assembler messages: > magnum4000.S:30: Error: Can not represent relocation in this object file format > make[1]: *** [magnum4000.o] Error 1 > make[1]: Leaving directory `/home/davidk/mips/kernel/v1.3/linux/arch/mips/kernel' > make: *** [linuxsubdirs] Error 2 > > Any ideas? I presume it has something to do with elf vs a.out, but > I have no idea where to look. Yes, it has something to do with ELF vs. a.out. magnum4000.S contains branch instructions to external targets. It is possible to do this with a.out, but not with ELF (This is no bug in the tools but in the ELF ...) Take a look at pica.S. The two files are that similar that we should only be using one, anyway. I leave that job to one of the Magnum/M700 hackers. Ralf From triemer@babbitt.bernstein.com Tue Jan 23 03:04:05 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id DAA16336 for ; Tue, 23 Jan 1996 03:04:03 +0100 (MET) Received: from babbitt.bernstein.com by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA18365; Tue, 23 Jan 1996 03:04:43 +0100 (MET) Received: (from triemer@localhost) by babbitt.bernstein.com (8.6.12/8.6.11) id WAA00128; Mon, 22 Jan 1996 22:36:47 -0500 Date: Mon, 22 Jan 1996 22:36:47 -30000 From: Thomas Riemer To: linux-mips@fnet.fr Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 150 Lines: 6 Please sign me up for linux-mips mailing list that you maintain. -Tom ---- Where theory and reality meet. ---- Thomas Riemer, triemer@wesleyan.edu From imp@rover.village.org Thu Feb 8 06:37:14 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id GAA10910 for ; Thu, 8 Feb 1996 06:37:09 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA19896; Thu, 8 Feb 1996 06:37:02 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id WAA12960 for ; Wed, 7 Feb 1996 22:37:00 -0700 Message-Id: <199602080537.WAA12960@rover.village.org> To: Linux/MIPS Mailing List Subject: Deskstation progress and questions... Date: Wed, 07 Feb 1996 22:36:59 -0700 From: Warner Losh Content-Length: 1747 Lines: 35 I've managed to get to my first "out_p" statement! I'm hanging there because mipsconfig.h defines PORT_BASE to be 0xe2000000. On the Dekstation, this needs to be 0xb0000000, so I've hacked mipsconfig.h and am rebuilding. Ideally, we'd have some way to do this that didn't require a recompile of the whole kernel, but it would also have to be fast, and I'm not sure that fast and configurable are both attainable here. For the moment, I'm just punting and putting the ifdef in... However, I couldn't help notice some comments in mipsconfig.h that I need to ask some questions about... Since this might later prove useful to other porters, I thought I'd ask them here... TLBMAP is defined to be 0xe4000000. Where does one find what this address is for one's machine? I don't think I have physical memory at 0xexxxxxxx. Is this a magic address, or can it be anywhere? Looking at the MIPS User's Manual seems to indicate to me that this is something other than TLB, which lives in the CP0 register space, so this must be something else... What, I'm not sure. Have I overlooked something silly in the MUM? TLB_ROOT is defined to be 0xe4390000. (or more accurrately, (TLBMAP + (TLBMAP >> (12 - 2)))). Once I find TLBMAP, then I'll have this, I'm sure... Just thought I'd confirm this. Grepping the sources, TLBMAP shows up in include/asm-mips/mipsconfig.h, arch/mips/kernel{head,r4xx0}.S From looking at the code there, and the MUM, it would appear that kseg3 is being setup to handle the TLB misses. Hmmm, looking at r4xx0.S it appears that it is reloading the TLB from the PTE array that lives at TLBMAP. I don't see where this gets created, however. What am I missing? Thanks for any help you might be able to render... Warner From imp@rover.village.org Thu Feb 8 07:58:15 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id HAA12073 for ; Thu, 8 Feb 1996 07:58:14 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA20592; Thu, 8 Feb 1996 07:57:49 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id XAA13920 for ; Wed, 7 Feb 1996 23:57:30 -0700 Message-Id: <199602080657.XAA13920@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Deskstation progress and questions... In-Reply-To: Your message of Wed, 07 Feb 1996 22:36:59 MST Date: Wed, 07 Feb 1996 23:57:30 -0700 From: Warner Losh Content-Length: 1744 Lines: 42 : I've managed to get to my first "out_p" statement! I'm hanging there : because mipsconfig.h defines PORT_BASE to be 0xe2000000. On the : Dekstation, this needs to be 0xb0000000, so I've hacked mipsconfig.h : and am rebuilding. Ideally, we'd have some way to do this that didn't : require a recompile of the whole kernel, but it would also have to be : fast, and I'm not sure that fast and configurable are both attainable : here. For the moment, I'm just punting and putting the ifdef in... I've been able to get to the following point in the kernel boot with *LOTS* of hacking that I'm not sure is correct Launching Kernel... (this is bold, the rest isn't) Console: 0 point font, 0 scans Console: colour VGA+ 80x50, 1 virutual console (max 63) Calibraring delay loop.. _ The _ is the blinking cursor. I do need to go in and fix the 80x50 number, but I doubt that is causing the apparent hang... Any ideas? My guess is a timer is set for a while in the future, and then it never expires, so the kernel just keeps counting... I could be wrong about that. Ralf, have you seen this before, say, on the Deskstation Tyne? I'm too tired to track it down now, but I think that I may have left out some code either in MILO or the early init that the tyne used... But at least I'm to the point in the kernel where I can get console output.... Anyway, I'm off to bed. Backup (I'm a wimp) run in 10 minutes, so I shouldn't lose everything this time. Warner P.S. The disk that was supposed to go into the Linux/MIPS box has died on me :-(. P.P.S. I'm wordsmithing the linux-mips-howto.html. Comments and/or suggestions are welcome. The latest draft can be found at http://www.village.org/villagers/imp/mips/linux-mips-howto.html. From linux@mailhost.uni-koblenz.de Thu Feb 8 11:29:52 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id LAA18949 for ; Thu, 8 Feb 1996 11:29:51 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA24229; Thu, 8 Feb 1996 11:29:41 +0100 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with SMTP id LAA06669 for ; Thu, 8 Feb 1996 11:28:27 +0100 From: Systemkennung Linux Message-Id: <199602081028.LAA06669@informatik.uni-koblenz.de> Received: by ozzy (5.x/KO-2.0) id AA06850; Thu, 8 Feb 1996 11:28:54 +0100 Subject: Re: Deskstation progress and questions... To: linux-mips@fnet.fr Date: Thu, 8 Feb 1996 11:28:53 +0100 (MET) In-Reply-To: <199602080537.WAA12960@rover.village.org> from "Warner Losh" at Feb 7, 96 10:36:59 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 4196 Lines: 83 Hi, > I've managed to get to my first "out_p" statement! I'm hanging there > because mipsconfig.h defines PORT_BASE to be 0xe2000000. On the > Dekstation, this needs to be 0xb0000000, so I've hacked mipsconfig.h > and am rebuilding. Ideally, we'd have some way to do this that didn't > require a recompile of the whole kernel, but it would also have to be > fast, and I'm not sure that fast and configurable are both attainable > here. For the moment, I'm just punting and putting the ifdef in... The whole in/out stuff originating from Linux's Intel history is a bit messy. The way I've coded this is optimized for maximum code speed at minimal code size and probably a bit overoptimized. (Andy'd says this is a confession ...) After all the in/out operations are *much* slower than the speed of the host bus. Speed is important for in/out in loops and there the load of the address is outside the loop anyway. For Deskstation Tyne, Magnum 4000 and Acer Pica the physical addresses of the ports are outside the 4gb address range of the 32 bit mode and therefore not accessible in the 32 bit mode. I therefore mapped them to address 0xe2000000. This address is an arbitrarily choosen free virtual address. Well, to be honest it is the Acer BIOS mapping but it's completly unimportant if whether to keep this mapping intact or not. > However, I couldn't help notice some comments in mipsconfig.h that I > need to ask some questions about... Since this might later prove > useful to other porters, I thought I'd ask them here... > > TLBMAP is defined to be 0xe4000000. Where does one find what this > address is for one's machine? I don't think I have physical memory at > 0xexxxxxxx. Is this a magic address, or can it be anywhere? Looking > at the MIPS User's Manual seems to indicate to me that this is > something other than TLB, which lives in the CP0 register space, so > this must be something else... What, I'm not sure. Have I overlooked > something silly in the MUM? TLBMAP is just an free virtual address range. The value 0xe4000000 has been choosen arbitrarily and can be changed as needed. This value is hardcoded for speed reasons. > TLB_ROOT is defined to be 0xe4390000. (or more accurrately, (TLBMAP + > (TLBMAP >> (12 - 2)))). Once I find TLBMAP, then I'll have this, I'm > sure... Just thought I'd confirm this. Yes. > Grepping the sources, TLBMAP shows up in > include/asm-mips/mipsconfig.h, arch/mips/kernel{head,r4xx0}.S From > looking at the code there, and the MUM, it would appear that kseg3 is > being setup to handle the TLB misses. Hmmm, looking at r4xx0.S it > appears that it is reloading the TLB from the PTE array that lives at > TLBMAP. I don't see where this gets created, however. What am I > missing? You're just about to find the wonders of Linux/MIPS memory management ... The big trick is that MIPS CPUs assume MMU data to be present in a normal array while Linux maintains this data in a tree. The Linux/MIPS 32 bit memory managment code maintains this two-level tree in KSEG1. Entry 0 in the TLB is reserved as pointer to the first level of this table. It maps the first level table (4k == PAGE_SIZE) to address TLB_ROOT. This table then contains entries which map the complete pagetable of 4mb size (for mapping the complete 4gb address space) to address TLBMAP. The trick is that the exception handler code has only to do the reload for tlbl/tlbs exceptions (and some handling of accessed/dirty bits, which is probably going away) and is optimized for maximum speed by moving all complexity from it into the C code of the memory management. This is important when the working set of memory exceeds the value number of free pages * page_size and therefore the tlb (which can essentilly be treated as cache) starts thrashing. At that point every instruction counts. The code is reading/writing the MMU table at both kseg1 and kseg3. Due to the problems with cache aliasing on the virtual indexed R4000 caches the pagetables have to be accessed uncached. This is a bit a problem for the performance and the best solution for it spells R10000 ... > Thanks for any help you might be able to render... Ralf From linux@mailhost.uni-koblenz.de Thu Feb 8 11:42:11 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id LAA19142 for ; Thu, 8 Feb 1996 11:42:10 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA24468; Thu, 8 Feb 1996 11:41:37 +0100 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with SMTP id LAA07385 for ; Thu, 8 Feb 1996 11:40:19 +0100 From: Systemkennung Linux Message-Id: <199602081040.LAA07385@informatik.uni-koblenz.de> Received: by ozzy (5.x/KO-2.0) id AA06885; Thu, 8 Feb 1996 11:40:47 +0100 Subject: Re: Deskstation progress and questions... To: linux-mips@fnet.fr Date: Thu, 8 Feb 1996 11:40:46 +0100 (MET) In-Reply-To: <199602080657.XAA13920@rover.village.org> from "Warner Losh" at Feb 7, 96 11:57:30 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 2213 Lines: 56 Hi, > : I've managed to get to my first "out_p" statement! I'm hanging there > : because mipsconfig.h defines PORT_BASE to be 0xe2000000. On the > : Dekstation, this needs to be 0xb0000000, so I've hacked mipsconfig.h > : and am rebuilding. Ideally, we'd have some way to do this that didn't > : require a recompile of the whole kernel, but it would also have to be > : fast, and I'm not sure that fast and configurable are both attainable > : here. For the moment, I'm just punting and putting the ifdef in... > > I've been able to get to the following point in the kernel boot with > *LOTS* of hacking that I'm not sure is correct > > Launching Kernel... (this is bold, the rest isn't) "Launching Kernel..." is printed by the ARC BIOS Write() call. Linux doesn't necessarily use the same stile/colour as the BIOS ... > Console: 0 point font, 0 scans > Console: colour VGA+ 80x50, 1 virutual console (max 63) > Calibraring delay loop.. _ > > The _ is the blinking cursor. > > I do need to go in and fix the 80x50 number, but I doubt that is > causing the apparent hang... > > Any ideas? My guess is a timer is set for a while in the future, and > then it never expires, so the kernel just keeps counting... I could > be wrong about that. Ralf, have you seen this before, say, on the > Deskstation Tyne? I'm too tired to track it down now, but I think > that I may have left out some code either in MILO or the early init > that the tyne used... You should try to test if the variable jiffies is getting counted up. If not your timer interrupt is broken. (The interrupts are enabled, aren't they?) Try something like while(1) printk("Jiffies == %d\n", jiffies); at the point where the kernel does the BogoMIPS benchmark. > But at least I'm to the point in the kernel where I can get console > output.... Whom you're telling. I was finally fed up of living without a cursor last night and hacked the missing four lines of code ... ed(1) users don't need a cursor ;-) > P.P.S. I'm wordsmithing the linux-mips-howto.html. Comments and/or > suggestions are welcome. The latest draft can be found at > http://www.village.org/villagers/imp/mips/linux-mips-howto.html. Great! Ralf From linux@mailhost.uni-koblenz.de Thu Feb 8 12:12:48 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id MAA19682 for ; Thu, 8 Feb 1996 12:12:46 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA24945; Thu, 8 Feb 1996 12:11:57 +0100 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with SMTP id MAA09255 for ; Thu, 8 Feb 1996 12:10:27 +0100 From: Systemkennung Linux Message-Id: <199602081110.MAA09255@informatik.uni-koblenz.de> Received: by ozzy (5.x/KO-2.0) id AA07022; Thu, 8 Feb 1996 12:10:50 +0100 Subject: Linux 1.3.59 To: linux-mips@fnet.fr Date: Thu, 8 Feb 1996 12:10:49 +0100 (MET) Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 764 Lines: 25 -----BEGIN PGP SIGNED MESSAGE----- Hi all, This is the long awaited Linux/MIPS 1.3.59 kernel. Linux 1.3.59 features Linus' patch 1.3.59, Stoned's stuff, Paul's R3000/DECstation stuff and various bits from me. linux-1.3.55-1.3.56.diffs.gz replaces the file on ftp.fnet.fr which was by accident the same as linux-1.3.54-1.3.55.diffs.gz. Ralf 86ccbe7522016fbad10d7c47d6315a54 linux-1.3.55-1.3.56.diffs.gz c8a91dfaeb828595b9a7a254662ccf61 linux-1.3.58-1.3.59.diffs.gz -----BEGIN PGP SIGNATURE----- Version: 2.6.2 iQCVAgUBMRnaN0ckbl6vezDBAQGG0AP6A30g2QExOn8MXhCAHsVgu/Y3VKwzxk8g bF/7xV33Ziz4DlWyK6/77c+5PrTo6mvKnRs+VHwd7UvScAnx0hvXOJ4xnVdAgozF HcDnb4JE4o6HzZwJ6iEqEH410Fytw4yy2YqaaEE+hOOi842yDQojr8H+F+keNgvw 8LFyJZjKaXc= =anml -----END PGP SIGNATURE----- From imp@rover.village.org Thu Feb 8 23:21:12 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id XAA11168 for ; Thu, 8 Feb 1996 23:21:10 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA05802; Thu, 8 Feb 1996 23:20:51 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id PAA15836 for ; Thu, 8 Feb 1996 15:20:18 -0700 Message-Id: <199602082220.PAA15836@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Deskstation progress and questions... In-Reply-To: Your message of Thu, 08 Feb 1996 11:40:46 +0100 Date: Thu, 08 Feb 1996 15:20:17 -0700 From: Warner Losh Content-Length: 942 Lines: 25 : "Launching Kernel..." is printed by the ARC BIOS Write() call. Linux : doesn't necessarily use the same stile/colour as the BIOS ... Yes. It does that on intel as well. My point was that I'm getting kernel output, rather than wondering why... : not your timer interrupt is broken. (The interrupts are enabled, aren't : they?) Try something like : : while(1) printk("Jiffies == %d\n", jiffies); : : at the point where the kernel does the BogoMIPS benchmark. That's a good suggestion. I'll double check to make sure the sti() line didn't get deleted by mistake, since I was adding and deleting the *(char *)0xa00a0000 = 'A'; type lines all over the place to see where things were hanging. I'll also see if jiffies is changing or not... Thanks for the pointers. Maybe the 1.3.60 kernel you release will have most of the Deskstation stuff in it! That would be too cool if I could get to the mounting / point by then. Warner From imp@rover.village.org Fri Feb 9 06:13:15 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id GAA18157 for ; Fri, 9 Feb 1996 06:13:14 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA11403; Fri, 9 Feb 1996 06:13:07 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id WAA17573 for ; Thu, 8 Feb 1996 22:13:08 -0700 Message-Id: <199602090513.WAA17573@rover.village.org> To: Linux/MIPS Mailing List Subject: No interrupts Date: Thu, 08 Feb 1996 22:13:07 -0700 From: Warner Losh Content-Length: 1191 Lines: 39 OK. I've let the machine run overnight (and it didn't ever finish). The problem is that it is stuck in the following kernel code: printk("Calibrating delay loop.. "); while (loops_per_sec <<= 1) { /* wait for "start of" clock tick */ ticks = jiffies; while (ticks == jiffies) printk("Jiffies = %d\r", jiffies ) /* nothing */; /* Go .. */ ticks = jiffies; __delay(loops_per_sec); ticks = jiffies - ticks; if (ticks) break; } Will just print Jiffies = 0 forever :-( I'm at a loss to understand why this is happening to me. The declaration looks correct to me: include/linux/sched.h: extern unsigned long volatile jiffies; so it likely is the case that I'm not taking interrupts that cause the jiffies count to be incremented. Any ideas on what to look for? I've gone ahead and added a sti() just before the while loop, so I know interrupts are enabled. I've not (yet) upgraded to 1.5.59, but a quick peek at the patches doesn't reveal anything that looked like it would help the situation. There must be something different between the 1.2.9ish kernel that I had working before and this one.... Anyway, I'd love to hear any ideas that people have... Warner From imp@rover.village.org Fri Feb 9 09:31:36 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id JAA20763 for ; Fri, 9 Feb 1996 09:31:35 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA13671; Fri, 9 Feb 1996 09:31:21 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id BAA22175 for ; Fri, 9 Feb 1996 01:31:20 -0700 Message-Id: <199602090831.BAA22175@rover.village.org> To: Linux/MIPS Mailing List Subject: Load address of 1.3.59 Date: Fri, 09 Feb 1996 01:31:19 -0700 From: Warner Losh Content-Length: 1229 Lines: 36 I just built the 1.3.59 kernel configured for the deskstation. It is loading at 0 rather than 0x80000000. It appears that on the DECstation the kernel gets loaded at 0x80030000, but on all the other machines the load address should be 0x80000000 (at least for the moment). As delivered, the 1.3.59 makefile will produce this only for when CONFIG_MIPS_JAZZ is enabled. Here's a quick patch that will always put it at 80000000 unless you are on a DECstation. Warner P.S. I'm typing DECstation because deskstation and decstation typed like that look too much alike... P.P.S. Yes, this is from CVS, but patch should grok it if you feed it the index line as well. If patch should fail, it should be easy enough to apply by hand :-). Index: arch/mips/Makefile =================================================================== RCS file: /home/imp/Repository/linux/arch/mips/Makefile,v retrieving revision 1.1.1.1 diff -u -r1.1.1.1 Makefile --- Makefile 1996/02/09 06:45:13 1.1.1.1 +++ Makefile 1996/02/09 08:25:13 @@ -106,8 +106,7 @@ # ifdef CONFIG_DECSTATION LINKFLAGS := $(LINKFLAGS) -T arch/mips/ld.script.dec -endif -ifdef CONFIG_MIPS_JAZZ +else LINKFLAGS := $(LINKFLAGS) -Ttext 0x80000000 endif From triemer@babbitt.bernstein.com Sat Feb 10 03:53:06 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id DAA04194 for ; Sat, 10 Feb 1996 03:53:05 +0100 (MET) Received: from babbitt.bernstein.com by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA28436; Sat, 10 Feb 1996 03:52:52 +0100 (MET) Received: (from triemer@localhost) by babbitt.bernstein.com (8.6.12/8.6.11) id QAA03771; Fri, 9 Feb 1996 16:53:43 -0500 Date: Fri, 9 Feb 1996 16:53:41 +0000 From: Thomas Riemer Subject: Re: No interrupts To: linux-mips@fnet.fr Cc: Linux/MIPS Mailing List In-Reply-To: <199602090513.WAA17573@rover.village.org> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 1551 Lines: 51 Because the interrupt timer is not enabled. So jiffies never changes - which is a real bummer. The culprit is a line of some dosisms that still remain. -Tom On Thu, 8 Feb 1996, Warner Losh wrote: > OK. I've let the machine run overnight (and it didn't ever finish). > The problem is that it is stuck in the following kernel code: > > printk("Calibrating delay loop.. "); > while (loops_per_sec <<= 1) { > /* wait for "start of" clock tick */ > ticks = jiffies; > while (ticks == jiffies) > printk("Jiffies = %d\r", jiffies ) /* nothing */; > /* Go .. */ > ticks = jiffies; > __delay(loops_per_sec); > ticks = jiffies - ticks; > if (ticks) > break; > } > > Will just print Jiffies = 0 forever :-( > > I'm at a loss to understand why this is happening to me. The > declaration looks correct to me: > > include/linux/sched.h: extern unsigned long volatile jiffies; > > so it likely is the case that I'm not taking interrupts that cause the > jiffies count to be incremented. > > Any ideas on what to look for? I've gone ahead and added a sti() just > before the while loop, so I know interrupts are enabled. > > I've not (yet) upgraded to 1.5.59, but a quick peek at the patches > doesn't reveal anything that looked like it would help the situation. > > There must be something different between the 1.2.9ish kernel that I > had working before and this one.... > > Anyway, I'd love to hear any ideas that people have... > > Warner > ---- Where theory and reality meet. ---- Thomas Riemer, triemer@wesleyan.edu From triemer@babbitt.bernstein.com Sat Feb 10 04:58:55 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id EAA04575 for ; Sat, 10 Feb 1996 04:58:55 +0100 (MET) Received: from babbitt.bernstein.com by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA29011; Sat, 10 Feb 1996 04:58:47 +0100 (MET) Received: (from triemer@localhost) by babbitt.bernstein.com (8.6.12/8.6.11) id RAA05705; Fri, 9 Feb 1996 17:59:43 -0500 Date: Fri, 9 Feb 1996 17:59:42 +0000 From: Thomas Riemer Subject: Re: Load address of 1.3.59 To: linux-mips@fnet.fr Cc: Linux/MIPS Mailing List In-Reply-To: <199602090831.BAA22175@rover.village.org> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 520 Lines: 20 So, I've gotten to the point where I'm able to compile... but not link the kernel for the DecStation R3000. The things that I'm missing are OFFSET_BOOTINFO_TLB_ENTRIES OFFSET_BOOTINFO_CPUTYPES OFFSET_BOOTINFO_MEMLOWER OFFSET_BOOTINFO_MEMUPPER OFFSET_BOOTINFO_CACHE_SIZE These used to be defined in include/asm-mips/bootinfo.h They were commented out. Any ideas what it should become? decstation.S seems to be the thing complaining. ---- Where theory and reality meet. ---- Thomas Riemer, triemer@wesleyan.edu From triemer@babbitt.bernstein.com Sat Feb 10 05:02:31 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id FAA04808 for ; Sat, 10 Feb 1996 05:02:31 +0100 (MET) Received: from babbitt.bernstein.com by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA29085; Sat, 10 Feb 1996 05:02:21 +0100 (MET) Received: (from triemer@localhost) by babbitt.bernstein.com (8.6.12/8.6.11) id SAA05718; Fri, 9 Feb 1996 18:03:19 -0500 Date: Fri, 9 Feb 1996 18:03:18 +0000 From: Thomas Riemer Subject: Re: No interrupts To: linux-mips@fnet.fr In-Reply-To: <199602100354.UAA25067@rover.village.org> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 1672 Lines: 50 linux-1.3.59/arch/mips/kernel/irq.c The function enable_irq The code is > outb(cach_21,0x21); After you figure all #defins and what not, it turns out that it pokes stuff in memory at the wrong place. -Tom On Fri, 9 Feb 1996, Warner Losh wrote: > : Because the interrupt timer is not enabled. > > Because the interrupt timer IRS is not being called. There is a > subtle difference there. > > : So jiffies never changes - which is a real bummer. > : The culprit is a line of some dosisms that still remain. > > Any ideas which one? > > Datapoint: Linux/MIPS 1.2.10ish booted to a much farther point, so I > kinda doubt it. Most likely some change between now and then broke > things :-(. I'm looking into that angle at the moment. I'm also > looking to see if the interrupt handlers is being called at all by > scribbling to video memory whenever it gets called.... > > Another datapoint: When I hit the numlock key, nothing happens. > Again, indication that the interrupts that should be happening aren't > calling the routines that one would expect, either due to a > dispatching bug, or due to an enabling bug. It is more general than > just the timer... > > Thanks for the ideas (especially to Drew who in his excited moments > can indeed rattle off 20 things that are so obvious to try that you > haven't even thought of them yet)... > > Ralf, any changes to the dispatch code that would might potentially > cause this? The tyne.S didn't compile when I started, so I'm thinking > there just might be some changes that require this... Just a thought. > > Warner > ---- Where theory and reality meet. ---- Thomas Riemer, triemer@wesleyan.edu From imp@rover.village.org Sat Feb 10 04:54:27 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id EAA04526 for ; Sat, 10 Feb 1996 04:54:25 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA28972; Sat, 10 Feb 1996 04:54:16 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id UAA25067 for ; Fri, 9 Feb 1996 20:54:18 -0700 Message-Id: <199602100354.UAA25067@rover.village.org> To: linux-mips@fnet.fr Subject: Re: No interrupts In-Reply-To: Your message of Fri, 09 Feb 1996 16:53:41 GMT Date: Fri, 09 Feb 1996 20:54:17 -0700 From: Warner Losh Content-Length: 1275 Lines: 31 : Because the interrupt timer is not enabled. Because the interrupt timer IRS is not being called. There is a subtle difference there. : So jiffies never changes - which is a real bummer. : The culprit is a line of some dosisms that still remain. Any ideas which one? Datapoint: Linux/MIPS 1.2.10ish booted to a much farther point, so I kinda doubt it. Most likely some change between now and then broke things :-(. I'm looking into that angle at the moment. I'm also looking to see if the interrupt handlers is being called at all by scribbling to video memory whenever it gets called.... Another datapoint: When I hit the numlock key, nothing happens. Again, indication that the interrupts that should be happening aren't calling the routines that one would expect, either due to a dispatching bug, or due to an enabling bug. It is more general than just the timer... Thanks for the ideas (especially to Drew who in his excited moments can indeed rattle off 20 things that are so obvious to try that you haven't even thought of them yet)... Ralf, any changes to the dispatch code that would might potentially cause this? The tyne.S didn't compile when I started, so I'm thinking there just might be some changes that require this... Just a thought. Warner From Stoned.Elipot@univ-evry.fr Sat Feb 10 14:27:27 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id OAA07455 for ; Sat, 10 Feb 1996 14:27:27 +0100 (MET) Received: from amiante.univ-evry.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA03974; Sat, 10 Feb 1996 14:27:20 +0100 (MET) Received: from bandsept.univ-evry.fr (bandsept.univ-evry.fr [192.93.214.19]) by amiante.univ-evry.fr with ESMTP (8.6.12/94.01.10); id NAA25368; Sat, 10 Feb 1996 13:39:41 +0100 Received: from localhost.univ-evry.fr (localhost.univ-evry.fr [127.0.0.1]) by bandsept.univ-evry.fr with SMTP (8.6.12/94.01.26); id OAA19330; Sat, 10 Feb 1996 14:26:05 +0100 Message-Id: <199602101326.OAA19330@bandsept.univ-evry.fr> To: linux-mips@fnet.fr Subject: DecStation and 1.3.59 [was: Re: Load address of 1.3.59] In-Reply-To: Your message of "Fri, 09 Feb 1996 17:59:42 GMT." X-It: bandsept (wrong OS, but right CPU) X-Mailer: MH 6.8.3, MH-E 5.0.2 Date: Sat, 10 Feb 1996 14:25:59 +0100 From: Stoned Elipot Content-Length: 822 Lines: 26 >So, I've gotten to the point where I'm able to compile... but not >link the kernel for the DecStation R3000. >The things that I'm missing are >OFFSET_BOOTINFO_TLB_ENTRIES [SNIP] >These used to be defined in include/asm-mips/bootinfo.h >They were commented out. >Any ideas what it should become? >decstation.S seems to be the thing complaining. Well,have you seen the comment added by Ralf at the top of decstation.S (under FIXME:) ? I believe Dec people have to put in dec_entry() the tags creation stuff that Milo does for ARC boxes. MACH does all hardware detection for Dec after the kernel is launched so I believe the same kind of job can be done in dec_entry(): it will create tags and then kernel_entry() and setup_arch() will read them... Not very optimized, any idea to improve this ? Cheers Stoned. From triemer@babbitt.bernstein.com Sun Feb 11 01:26:13 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id BAA10713 for ; Sun, 11 Feb 1996 01:26:12 +0100 (MET) Received: from babbitt.bernstein.com by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA09908; Sun, 11 Feb 1996 01:26:04 +0100 (MET) Received: (from triemer@localhost) by babbitt.bernstein.com (8.6.12/8.6.11) id OAA07792; Sat, 10 Feb 1996 14:27:21 -0500 Date: Sat, 10 Feb 1996 14:27:21 +0000 From: Thomas Riemer Subject: Re: DecStation and 1.3.59 [was: Re: Load address of 1.3.59] To: linux-mips@fnet.fr In-Reply-To: <199602101326.OAA19330@bandsept.univ-evry.fr> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 1273 Lines: 40 Sadly, I did see the FIXME.... but have no idea what NAPS refers to. My first guess is a replacement of the OFFSET_BOOTINFO_TLB_ENTRIES in decstation.S with mips_tlb_entries. That gets me a compiled version with the exception that it no longer boots - bummer, heh. -Tom On Sat, 10 Feb 1996, Stoned Elipot wrote: > > >So, I've gotten to the point where I'm able to compile... but not > >link the kernel for the DecStation R3000. > > >The things that I'm missing are > >OFFSET_BOOTINFO_TLB_ENTRIES > [SNIP] > > >These used to be defined in include/asm-mips/bootinfo.h > > >They were commented out. > >Any ideas what it should become? > >decstation.S seems to be the thing complaining. > > Well,have you seen the comment added by Ralf at the top of > decstation.S (under FIXME:) ? > > I believe Dec people have to put in dec_entry() the tags creation stuff > that Milo does for ARC boxes. MACH does all hardware detection for Dec > after the kernel is launched so I believe the same kind of job can be > done in dec_entry(): it will create tags and then kernel_entry() and > setup_arch() will read them... Not very optimized, any idea to improve > this ? > > Cheers Stoned. > > ---- Where theory and reality meet. ---- Thomas Riemer, triemer@wesleyan.edu From triemer@babbitt.bernstein.com Sun Feb 11 03:45:41 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id DAA12766 for ; Sun, 11 Feb 1996 03:45:41 +0100 (MET) Received: from babbitt.bernstein.com by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA10937; Sun, 11 Feb 1996 03:45:32 +0100 (MET) Received: (from triemer@localhost) by babbitt.bernstein.com (8.6.12/8.6.11) id QAA07955; Sat, 10 Feb 1996 16:46:52 -0500 Date: Sat, 10 Feb 1996 16:46:52 +0000 From: Thomas Riemer Subject: Re: DecStation and 1.3.59 [was: Re: Load address of 1.3.59] To: linux-mips@fnet.fr In-Reply-To: <199602110210.TAA02529@rover.village.org> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 806 Lines: 22 When I try to compile, I get two places where things don't go correctly: I suspect this code didn't make it when the merge from 1.3.58 to 1.3.59 1. include/asm-mips/stackframe.h It comes up and says "Implement SAVE_ALL and RESTORE_ALL!" 2. it seems like decstation.S has an entry called dec_entry that used to be called by head.S (well, in the version that didn't have source released) - at least that is what I'm lead to believe by the comments in head.S and decstation.S. Am I to believe that milo now supports the DecStation? Nothing in the milo-0.26 source makes me believe that. If someone has the old version of head.S before it was merged to 1.3.59, would you please send it to me? -Tom ---- Where theory and reality meet. ---- Thomas Riemer, triemer@wesleyan.edu From triemer@babbitt.bernstein.com Sun Feb 11 04:51:24 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id EAA13109 for ; Sun, 11 Feb 1996 04:51:23 +0100 (MET) Received: from babbitt.bernstein.com by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA11420; Sun, 11 Feb 1996 04:51:14 +0100 (MET) Received: (from triemer@localhost) by babbitt.bernstein.com (8.6.12/8.6.11) id RAA08148; Sat, 10 Feb 1996 17:52:36 -0500 Date: Sat, 10 Feb 1996 17:52:35 +0000 From: Thomas Riemer Subject: Re: DecStation and 1.3.59 [was: Re: Load address of 1.3.59] To: linux-mips@fnet.fr In-Reply-To: Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 352 Lines: 16 I'm again look at decstation.S This time I'm actually beginning to try and figure out what it does: So what does the following line do? la a0, dec_signon "la" is not an op code that shows up in either Mips Risc Architecture or the MIPs IV instruction set manual ---- Where theory and reality meet. ---- Thomas Riemer, triemer@wesleyan.edu From triemer@babbitt.bernstein.com Sun Feb 11 06:25:35 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id GAA13721 for ; Sun, 11 Feb 1996 06:25:34 +0100 (MET) Received: from babbitt.bernstein.com by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA12746; Sun, 11 Feb 1996 06:25:26 +0100 (MET) Received: (from triemer@localhost) by babbitt.bernstein.com (8.6.12/8.6.11) id TAA08405; Sat, 10 Feb 1996 19:26:48 -0500 Date: Sat, 10 Feb 1996 19:26:47 +0000 From: Thomas Riemer To: linuxmips Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 478 Lines: 15 Where is the start of the kernel... I thought it was in mips/arch/kernel/head.S (kernel_entry) The reason I'm confused is that mips/arch/kernel/decstation.S (dec_entry) jumps to kernel_entry.... implying that it should run first.... and understandably, it probably should, since all it is, is something that updates variables. So what is the real entry point, if not one of the these two. -Tom ---- Where theory and reality meet. ---- Thomas Riemer, triemer@wesleyan.edu From imp@rover.village.org Sun Feb 11 03:10:56 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id DAA12667 for ; Sun, 11 Feb 1996 03:10:55 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA10673; Sun, 11 Feb 1996 03:10:38 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id TAA02529 for ; Sat, 10 Feb 1996 19:10:38 -0700 Message-Id: <199602110210.TAA02529@rover.village.org> To: linux-mips@fnet.fr Subject: Re: DecStation and 1.3.59 [was: Re: Load address of 1.3.59] In-Reply-To: Your message of Sat, 10 Feb 1996 14:27:21 GMT Date: Sat, 10 Feb 1996 19:10:37 -0700 From: Warner Losh Content-Length: 429 Lines: 12 : Sadly, I did see the FIXME.... but have no idea what NAPS refers to. NAPS is Stone's new system for passing parameters from MILO to Linux using tagged memory (from high memory down). : My first guess is a replacement of the OFFSET_BOOTINFO_TLB_ENTRIES in : decstation.S with mips_tlb_entries. That gets me a compiled version : with the exception that it no longer boots - bummer, heh. Yes. That is a good guess... Warner From imp@rover.village.org Sun Feb 11 05:46:02 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id FAA13427 for ; Sun, 11 Feb 1996 05:46:02 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA12446; Sun, 11 Feb 1996 05:45:48 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id VAA02865 for ; Sat, 10 Feb 1996 21:45:51 -0700 Message-Id: <199602110445.VAA02865@rover.village.org> To: linux-mips@fnet.fr Subject: Re: DecStation and 1.3.59 [was: Re: Load address of 1.3.59] In-Reply-To: Your message of Sat, 10 Feb 1996 17:52:35 GMT Date: Sat, 10 Feb 1996 21:45:51 -0700 From: Warner Losh Content-Length: 650 Lines: 20 : la a0, dec_signon : : "la" is not an op code that shows up in either Mips Risc Architecture : or the MIPs IV instruction set manual LA == Load Address. It is a pseudo-op in gas. It literally translates to: lui a0, %hi(dec_signon) addiu a0, a0, %lo(dec_signon) Which is effectively loading the address into a0. I just discovered this because I was looking at a disassembled version of rpc44.o to make sure that it mached rpc44.S (which it did, except for things like this). And now that I recall, I think there was email about milo's disassembler that stated that it had been modified to merge these two instructions into one. Warner From linux@mailhost.uni-koblenz.de Sun Feb 11 10:03:54 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id KAA14855 for ; Sun, 11 Feb 1996 10:03:53 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA14853; Sun, 11 Feb 1996 10:03:45 +0100 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with ESMTP id KAA15116 for ; Sun, 11 Feb 1996 10:02:35 +0100 From: Systemkennung Linux Message-Id: <199602110902.KAA15116@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id KAA07376; Sun, 11 Feb 1996 10:02:58 +0100 Subject: Re: DecStation and 1.3.59 [was: Re: Load address of 1.3.59] To: linux-mips@fnet.fr Date: Sun, 11 Feb 1996 10:02:58 +0100 (MET) In-Reply-To: <199602110445.VAA02865@rover.village.org> from "Warner Losh" at Feb 10, 96 09:45:51 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1140 Lines: 35 Hi, > : la a0, dec_signon > : > : "la" is not an op code that shows up in either Mips Risc Architecture > : or the MIPs IV instruction set manual > > LA == Load Address. It is a pseudo-op in gas. It literally > translates to: > > lui a0, %hi(dec_signon) > addiu a0, a0, %lo(dec_signon) > > Which is effectively loading the address into a0. > > I just discovered this because I was looking at a disassembled version > of rpc44.o to make sure that it mached rpc44.S (which it did, except > for things like this). And now that I recall, I think there was email > about milo's disassembler that stated that it had been modified to > merge these two instructions into one. This is not the only macro instruction in MIPS assembler; in fact most of the instructions are macro instructions that expand sometimes into five instructions or more. Many of them are described in the appendix of the Kane. Ralf Name : MIPS RISC Architecture Author : Gerry Kane Publisher: Prentice Hall, Englewood Cliffs, N.J. 07632 ISBN : 0-12-58479-4 The book contains another ISBN: 0-13-584293-X, so be carefull. From linux@mailhost.uni-koblenz.de Sun Feb 11 15:41:10 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id PAA16402 for ; Sun, 11 Feb 1996 15:41:10 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA17173; Sun, 11 Feb 1996 15:41:01 +0100 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with ESMTP id PAA21888 for ; Sun, 11 Feb 1996 15:39:52 +0100 From: Systemkennung Linux Message-Id: <199602111439.PAA21888@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id PAA07597; Sun, 11 Feb 1996 15:40:13 +0100 Subject: Re: Unidentified subject! To: linux-mips@fnet.fr Date: Sun, 11 Feb 1996 15:40:13 +0100 (MET) In-Reply-To: from "Thomas Riemer" at Feb 10, 96 07:26:47 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 704 Lines: 19 Hi Tom, > Where is the start of the kernel... > > I thought it was in mips/arch/kernel/head.S (kernel_entry) > > The reason I'm confused is that mips/arch/kernel/decstation.S (dec_entry) > jumps to kernel_entry.... implying that it should run first.... > and understandably, it probably should, since all it is, is something > that updates variables. > > So what is the real entry point, if not one of the these two. kernel_entry in head.S is the real entry point for all machines except DECstations which use dec_entry. The trick is just that some additional setup has to be done for the DECstations before they can use the normal startup routine. Not very nice and will be cleaned up. Ralf From linux@mailhost.uni-koblenz.de Sun Feb 11 15:45:39 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id PAA16450 for ; Sun, 11 Feb 1996 15:45:39 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA17202; Sun, 11 Feb 1996 15:45:30 +0100 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with ESMTP id PAA22118 for ; Sun, 11 Feb 1996 15:44:20 +0100 From: Systemkennung Linux Message-Id: <199602111444.PAA22118@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id PAA07608; Sun, 11 Feb 1996 15:44:42 +0100 Subject: Re: DecStation and 1.3.59 [was: Re: Load address of 1.3.59] To: linux-mips@fnet.fr Date: Sun, 11 Feb 1996 15:44:42 +0100 (MET) In-Reply-To: from "Thomas Riemer" at Feb 10, 96 04:46:52 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 974 Lines: 29 Hi, > When I try to compile, I get two places where things don't go correctly: > I suspect this code didn't make it when the merge from 1.3.58 to 1.3.59 > > 1. include/asm-mips/stackframe.h > It comes up and says "Implement SAVE_ALL and RESTORE_ALL!" Ooops, you cought me :-) > 2. it seems like decstation.S has an entry called dec_entry that > used to be called by head.S (well, in the version that didn't > have source released) - at least that is what I'm lead to believe > by the comments in head.S and decstation.S. At the end of the dec_entry routine in decstation.S kernel_entry in head.S is called. > Am I to believe that milo now supports the DecStation? > Nothing in the milo-0.26 source makes me believe that. Milo still supports only Arc machines. > If someone has the old version of head.S before it was merged to 1.3.59, > would you please send it to me? There is still a full tree of Linux/MIPS 1.3.58 on the ftp sites. Ralf From linux@mailhost.uni-koblenz.de Sun Feb 11 15:47:44 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id PAA16496 for ; Sun, 11 Feb 1996 15:47:43 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA17212; Sun, 11 Feb 1996 15:47:33 +0100 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with ESMTP id PAA22158 for ; Sun, 11 Feb 1996 15:46:24 +0100 From: Systemkennung Linux Message-Id: <199602111446.PAA22158@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id PAA07618; Sun, 11 Feb 1996 15:46:46 +0100 Subject: Re: DecStation and 1.3.59 [was: Re: Load address of 1.3.59] To: linux-mips@fnet.fr Date: Sun, 11 Feb 1996 15:46:45 +0100 (MET) In-Reply-To: from "Thomas Riemer" at Feb 10, 96 02:27:21 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 362 Lines: 10 Hi, > My first guess is a replacement of the OFFSET_BOOTINFO_TLB_ENTRIES in > decstation.S with mips_tlb_entries. That gets me a compiled version > with the exception that it no longer boots - bummer, heh. There is still not every bit of Paul's DECstation/R3000 stuff in the distribution kernel so don't yet expect the kernel to build/run perfectly. Ralf From linux@mailhost.uni-koblenz.de Sun Feb 11 15:49:17 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id PAA16541 for ; Sun, 11 Feb 1996 15:49:17 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA17248; Sun, 11 Feb 1996 15:49:07 +0100 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with ESMTP id PAA22195 for ; Sun, 11 Feb 1996 15:47:58 +0100 From: Systemkennung Linux Message-Id: <199602111447.PAA22195@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id PAA07626; Sun, 11 Feb 1996 15:48:20 +0100 Subject: Re: DecStation and 1.3.59 [was: Re: Load address of 1.3.59] To: linux-mips@fnet.fr Date: Sun, 11 Feb 1996 15:48:20 +0100 (MET) In-Reply-To: <199602101326.OAA19330@bandsept.univ-evry.fr> from "Stoned Elipot" at Feb 10, 96 02:25:59 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 477 Lines: 13 Hi, > I believe Dec people have to put in dec_entry() the tags creation stuff > that Milo does for ARC boxes. MACH does all hardware detection for Dec > after the kernel is launched so I believe the same kind of job can be > done in dec_entry(): it will create tags and then kernel_entry() and > setup_arch() will read them... Not very optimized, any idea to improve > this ? Don't think we need to optimize this; it's also the more or less most logic way to it so. Ralf From linux@mailhost.uni-koblenz.de Sun Feb 11 16:11:42 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id QAA16827 for ; Sun, 11 Feb 1996 16:11:40 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA17492; Sun, 11 Feb 1996 16:11:26 +0100 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with ESMTP id QAA22661 for ; Sun, 11 Feb 1996 16:10:18 +0100 From: Systemkennung Linux Message-Id: <199602111510.QAA22661@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id QAA07659; Sun, 11 Feb 1996 16:10:39 +0100 Subject: Re: No interrupts To: linux-mips@fnet.fr Date: Sun, 11 Feb 1996 16:10:39 +0100 (MET) In-Reply-To: <199602100354.UAA25067@rover.village.org> from "Warner Losh" at Feb 9, 96 08:54:17 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1985 Lines: 48 Hi, > : Because the interrupt timer is not enabled. > > Because the interrupt timer IRS is not being called. There is a > subtle difference there. > > : So jiffies never changes - which is a real bummer. > : The culprit is a line of some dosisms that still remain. > > Any ideas which one? > > Datapoint: Linux/MIPS 1.2.10ish booted to a much farther point, so I > kinda doubt it. Most likely some change between now and then broke > things :-(. I'm looking into that angle at the moment. I'm also > looking to see if the interrupt handlers is being called at all by > scribbling to video memory whenever it gets called.... > > Another datapoint: When I hit the numlock key, nothing happens. > Again, indication that the interrupts that should be happening aren't > calling the routines that one would expect, either due to a > dispatching bug, or due to an enabling bug. It is more general than > just the timer... > Ralf, any changes to the dispatch code that would might potentially > cause this? The tyne.S didn't compile when I started, so I'm thinking > there just might be some changes that require this... Just a thought. Try to use the interrupt handler from pica.S (ll_isa_irq). I started with the Tyne irq handler for the Acer. The Tyne handler was buggy; so I repaired it for the Acer. The two handler are essentially the same. In fact the Acer could perfectly run with the Tyne handler. The reason that they're not the same is that the Jazz chipset offers a better way (Port accesses are incredible slooow) to do the i486-like interrupt acknowledge cycle: Acer: ll_isa_irq: lw t1,JAZZ_EISA_IRQ_ACK lui s0,%hi(PORT_BASE) andi t0,t1,8 # which pic? Tyne: lui s0,%hi(PORT_BASE) li t1,0x0f sb t1,%lo(PORT_BASE+0x20)(s0) # poll command lb t1,%lo(PORT_BASE+0x20)(s0) # read result Ralf From linux@mailhost.uni-koblenz.de Sun Feb 11 16:19:38 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id QAA16877 for ; Sun, 11 Feb 1996 16:19:38 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA17526; Sun, 11 Feb 1996 16:19:29 +0100 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with ESMTP id QAA22794 for ; Sun, 11 Feb 1996 16:18:19 +0100 From: Systemkennung Linux Message-Id: <199602111518.QAA22794@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id QAA07672; Sun, 11 Feb 1996 16:18:41 +0100 Subject: Re: No interrupts To: linux-mips@fnet.fr Date: Sun, 11 Feb 1996 16:18:41 +0100 (MET) In-Reply-To: from "Thomas Riemer" at Feb 9, 96 06:03:18 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 319 Lines: 15 Hi, > linux-1.3.59/arch/mips/kernel/irq.c > The function enable_irq > > The code is > > outb(cach_21,0x21); > > After you figure all #defins and what not, it turns out that > it pokes stuff in memory at the wrong place. I Warner's machine is really that PC-like as we believe this outb should be correct. Ralf From imp@rover.village.org Sun Feb 11 21:56:22 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id VAA18321 for ; Sun, 11 Feb 1996 21:56:20 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA19526; Sun, 11 Feb 1996 21:56:09 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id LAA03919 for ; Sun, 11 Feb 1996 11:52:59 -0700 Message-Id: <199602111852.LAA03919@rover.village.org> To: linux-mips@fnet.fr Subject: MILO Date: Sun, 11 Feb 1996 11:52:59 -0700 From: Warner Losh Content-Length: 2714 Lines: 51 I've been playing around with MILO the last day or so to get my mind off the Interrupt problem that I've been having... I've noticed there is a lot of fluff in milo now. Lots of commands for dumping registers, etc. These are wonderfully useful (in fact, I want to add to them, see below). However, they have grown to about 14k out of 50k. If they were eliminated, that would make MILO load about 30% faster than it does now. On my machine with its disk drive, this could be a win, so it is likely a win for others. I'm thinking of hacking milo in the following way: By default, milo will be as small as we can make it. There will be options that you can turn on, however, that will bloat milo so that it can be more useful. The defaults don't matter to me, so we can turn this around. However, I'd also like to add "ls" and "cp" to milo, plus the ability to do decompression and grok ext2fs file systems. The reason for this is so that we can repair disks from MILO and not need to build a new kernel. At least that's the theory, it may turn out to be too much of a PITA to do all of this, but I do want to do the decompression and cp and ls at least for FS that support this (eg FAT and ISO-9660 and maybe TFTP for cp). The decompression is for kernels and for fonts for the console, etc. Why do I want to do this? Well, I discovered this weekend that this really old disk that I've had laying around is actually not useless. The UltraStor 34F controller in my PC wouldn't recognize it at all. Since I'm looking for a disk for the Deskstation (the one that I had borrowed from it bought the farm recently), I thought "Heck, there is no way this will work, but I'll give it a shot none the less." And it worked. Now I want to put MILO on the hard disk. I could just hook it up to another machine and do a copy. However, I thoguht to myself, "hmm, people are going to need to do this for their machines" so I thought "Hmmm, this would make a nice side trip while I ponder the state of the interrupts on my machine." Maybe not the best reason do to something... However, after the fact I realized we'll have to be able to do things like this under ARC BIOS for the install anyway, so I thought now might not be a bad time to look into the difficulty of providing cp and maybe ls for MILO. Actually, MILO is a good core, but we'd need something like SASH that was on Solbourne machines (and maybe Suns). If we did this right, then we could layer the bootstrapping part of the OS on top of SASH. Something that would potentially partition the disks, format the partitions, copy milo, vmlinux, et al onto the machine and rebooting into the real install process like NT does. Comments? Warner From linux@mailhost.uni-koblenz.de Sun Feb 11 22:40:03 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id WAA18642 for ; Sun, 11 Feb 1996 22:40:02 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA20149; Sun, 11 Feb 1996 22:39:48 +0100 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with ESMTP id WAA03193 for ; Sun, 11 Feb 1996 22:38:42 +0100 From: Systemkennung Linux Message-Id: <199602112138.WAA03193@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id WAA07997; Sun, 11 Feb 1996 22:39:01 +0100 Subject: Re: MILO To: linux-mips@fnet.fr Date: Sun, 11 Feb 1996 22:39:00 +0100 (MET) In-Reply-To: <199602111852.LAA03919@rover.village.org> from "Warner Losh" at Feb 11, 96 11:52:59 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 3856 Lines: 70 Hi, > I've been playing around with MILO the last day or so to get my mind > off the Interrupt problem that I've been having... Welcome to the true kernel headaches :-) > I've noticed there is a lot of fluff in milo now. Lots of commands > for dumping registers, etc. These are wonderfully useful (in fact, I > want to add to them, see below). However, they have grown to about > 14k out of 50k. If they were eliminated, that would make MILO load > about 30% faster than it does now. On my machine with its disk drive, > this could be a win, so it is likely a win for others. I'm thinking > of hacking milo in the following way: I'm also not very lucky with the implementation of these commands. After building Pandora is a symlink to Milo; definately not the real thing ... > By default, milo will be as small as we can make it. There will be > options that you can turn on, however, that will bloat milo so that > it can be more useful. The defaults don't matter to me, so we can > turn this around. However, I'd also like to add "ls" and "cp" to > milo, plus the ability to do decompression and grok ext2fs file > systems. The reason for this is so that we can repair disks from MILO > and not need to build a new kernel. At least that's the theory, it > may turn out to be too much of a PITA to do all of this, but I do want > to do the decompression and cp and ls at least for FS that support > this (eg FAT and ISO-9660 and maybe TFTP for cp). The decompression > is for kernels and for fonts for the console, etc. > > Why do I want to do this? > > Well, I discovered this weekend that this really old disk that I've > had laying around is actually not useless. The UltraStor 34F > controller in my PC wouldn't recognize it at all. Since I'm looking > for a disk for the Deskstation (the one that I had borrowed from it > bought the farm recently), I thought "Heck, there is no way this will > work, but I'll give it a shot none the less." And it worked. Now I > want to put MILO on the hard disk. I could just hook it up to another > machine and do a copy. However, I thoguht to myself, "hmm, people are > going to need to do this for their machines" so I thought "Hmmm, this > would make a nice side trip while I ponder the state of the interrupts > on my machine." Maybe not the best reason do to something... > However, after the fact I realized we'll have to be able to do things > like this under ARC BIOS for the install anyway, so I thought now > might not be a bad time to look into the difficulty of providing cp > and maybe ls for MILO. > > Actually, MILO is a good core, but we'd need something like SASH that > was on Solbourne machines (and maybe Suns). If we did this right, > then we could layer the bootstrapping part of the OS on top of SASH. > Something that would potentially partition the disks, format the > partitions, copy milo, vmlinux, et al onto the machine and rebooting > into the real install process like NT does. You're right; I've tried some days ago whether Arc drives may be accessed similar to UNIX block devices - et voila - it works though not documented. So partitioning under ARC can really work (withhin the narrow limits set by some extremly buggy ARC implementations like the Magnum ...) What is missing is a port of termcap to Arc; then we could use programs like cfdisk for installation. Another program that might be interesting to be ported to Arc is an sash type shell with lots of builtin programs that recently has been published. Think it's name was mash. The sad fact that every Arc BIOS I've yet seen will turn writing a complex Arc program into a walk through a minefield. Even worse; Arc is bugged by design. There are all functions required to access and manipulate files - except a function to delete files. Btw. - the mines are (C) Microsoft ... Ralf From paul@suite.sw.oz.au Mon Feb 12 00:13:07 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id AAA19297 for ; Mon, 12 Feb 1996 00:13:06 +0100 (MET) Received: from staff.cs.su.OZ.AU by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA21402; Mon, 12 Feb 1996 00:12:56 +0100 (MET) Received: from suite.sw.oz.au by staff.cs.su.OZ.AU (mail from paul for linux-mips@fnet.fr) with MHSnet (insertion MHSnet site: swallow.sw.oz.au); Mon, 12 Feb 1996 10:12:56 +1100 Received: from suite.sw.oz.au by swallow.sw.oz.au with SMTP id XAA26921; Sun, 11 Feb 1996 23:12:39 GMT (8.6.10/Unixware) (from paul@suite.sw.oz.au for ) Received: by suite.sw.oz.au id AA03930; Mon, 12 Feb 1996 10:12:25 +1100 (5.65c/1.34) (from paul@suite.sw.oz.au for linux-mips@fnet.fr) From: paul@suite.sw.oz.au (Paul Antoine) Message-Id: <199602112312.AA03930@suite.sw.oz.au> Subject: Re: Unidentified subject! To: linux-mips@fnet.fr Date: Mon, 12 Feb 1996 10:12:25 +1100 (EST) In-Reply-To: <199602111439.PAA21888@informatik.uni-koblenz.de> from "Systemkennung Linux" at Feb 11, 96 03:40:13 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: 1351 Lines: 31 Ralf wrote, in response to Tom's question > > Where is the start of the kernel... > > > > I thought it was in mips/arch/kernel/head.S (kernel_entry) > > > > The reason I'm confused is that mips/arch/kernel/decstation.S (dec_entry) > > jumps to kernel_entry.... implying that it should run first.... > > and understandably, it probably should, since all it is, is something > > that updates variables. > > > > So what is the real entry point, if not one of the these two. > > kernel_entry in head.S is the real entry point for all machines except > DECstations which use dec_entry. The trick is just that some additional > setup has to be done for the DECstations before they can use the normal > startup routine. Not very nice and will be cleaned up. Yeah - though I think the change in entry point is handled reasonably cleanly by using the link script provided in arch/mips/ld.script.dec 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 2 698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From paul@suite.sw.oz.au Mon Feb 12 00:16:37 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id AAA19341 for ; Mon, 12 Feb 1996 00:16:36 +0100 (MET) Received: from staff.cs.su.OZ.AU by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA21419; Mon, 12 Feb 1996 00:16:25 +0100 (MET) Received: from suite.sw.oz.au by staff.cs.su.OZ.AU (mail from paul for linux-mips@fnet.fr) with MHSnet (insertion MHSnet site: swallow.sw.oz.au); Mon, 12 Feb 1996 10:16:29 +1100 Received: from suite.sw.oz.au by swallow.sw.oz.au with SMTP id XAA27041; Sun, 11 Feb 1996 23:16:11 GMT (8.6.10/Unixware) (from paul@suite.sw.oz.au for ) Received: by suite.sw.oz.au id AA04092; Mon, 12 Feb 1996 10:16:03 +1100 (5.65c/1.34) (from paul@suite.sw.oz.au for linux-mips@fnet.fr) From: paul@suite.sw.oz.au (Paul Antoine) Message-Id: <199602112316.AA04092@suite.sw.oz.au> Subject: Re: DecStation and 1.3.59 [was: Re: Load address of 1.3.59] To: linux-mips@fnet.fr Date: Mon, 12 Feb 1996 10:16:01 +1100 (EST) In-Reply-To: <199602101326.OAA19330@bandsept.univ-evry.fr> from "Stoned Elipot" at Feb 10, 96 02:25:59 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: 1666 Lines: 43 Stoned wrote, in response to Tom's question: > >So, I've gotten to the point where I'm able to compile... but not > >link the kernel for the DecStation R3000. > > >The things that I'm missing are > >OFFSET_BOOTINFO_TLB_ENTRIES > [SNIP] > > >These used to be defined in include/asm-mips/bootinfo.h > > >They were commented out. > >Any ideas what it should become? > >decstation.S seems to be the thing complaining. > > Well,have you seen the comment added by Ralf at the top of > decstation.S (under FIXME:) ? > > I believe Dec people have to put in dec_entry() the tags creation stuff > that Milo does for ARC boxes. MACH does all hardware detection for Dec > after the kernel is launched so I believe the same kind of job can be > done in dec_entry(): it will create tags and then kernel_entry() and > setup_arch() will read them... Not very optimized, any idea to improve > this ? The whole of dec_entry needs some optimization, so for the moment this solution is fine. My current thinking on dec_entry is that we need a dec kernel loader program, that can use either disk or network (making use of the appropriate boot prom calls) to load the plain kernel... I'll see if I can hack something up... 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 2 698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From linux@mailhost.uni-koblenz.de Mon Feb 12 00:34:02 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id AAA19426 for ; Mon, 12 Feb 1996 00:34:02 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA21617; Mon, 12 Feb 1996 00:33:43 +0100 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with ESMTP id AAA06022 for ; Mon, 12 Feb 1996 00:32:38 +0100 From: Systemkennung Linux Message-Id: <199602112332.AAA06022@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id AAA08060; Mon, 12 Feb 1996 00:32:57 +0100 Subject: Re: Unidentified subject! To: linux-mips@fnet.fr Date: Mon, 12 Feb 1996 00:32:57 +0100 (MET) In-Reply-To: <199602112312.AA03930@suite.sw.oz.au> from "Paul Antoine" at Feb 12, 96 10:12:25 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 305 Lines: 10 Hi, > Yeah - though I think the change in entry point is handled reasonably > cleanly by using the link script provided in arch/mips/ld.script.dec I intend to rewrite almost the whole thing into C at sometime. This will make machine specific changes like for the DECstation definately easier. Ralf From imp@rover.village.org Mon Feb 12 02:10:38 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id CAA20141 for ; Mon, 12 Feb 1996 02:10:35 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA22380; Mon, 12 Feb 1996 02:10:00 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id SAA04697 for ; Sun, 11 Feb 1996 18:10:02 -0700 Message-Id: <199602120110.SAA04697@rover.village.org> To: linux-mips@fnet.fr Subject: Re: MILO In-Reply-To: Your message of Sun, 11 Feb 1996 22:39:00 +0100 Date: Sun, 11 Feb 1996 18:10:02 -0700 From: Warner Losh Content-Length: 2735 Lines: 59 : Welcome to the true kernel headaches :-) Yes. These are harder to cure than caffeine headaches :-) : You're right; I've tried some days ago whether Arc drives may be : accessed similar to UNIX block devices - et voila - it works though : not documented. I would expect it to work because they have to be able to format the disk somehow. And there is only one arcinst program, not a dti one and a tyne one, etc. : So partitioning under ARC can really work (withhin the narrow limits set : by some extremly buggy ARC implementations like the Magnum ...) Yes. I wonder how arcinst even runs on that system :-(. : What is missing is a port of termcap to Arc; then we could use programs : like cfdisk for installation. Another program that might be interesting : to be ported to Arc is an sash type shell with lots of builtin programs : that recently has been published. Think it's name was mash. I like it. There is a song over hear called the Monster Mash that is old enough everyone on the planet should have heard by now. : The sad fact that every Arc BIOS I've yet seen will turn writing a complex : Arc program into a walk through a minefield. Even worse; Arc is bugged by : design. There are all functions required to access and manipulate files - : except a function to delete files. Btw. - the mines are (C) Microsoft ... So rm becomes much harder :-). All of this talk about install reminds me... I got a S.u.S.E. CD offer in my email (for the ParcPlace OI giveway, no doubt) about the time that Ralf put out root-0.00.tar.gz. I filled it out form they had sent and dropped the "Yah, I'm working on Linux/MIPS and am thinking of putting together a distribution for it, would you put that on your CD?" Just to see what would happen (and at the time I was thinking about this). It turns out they are interested! So if we have a BOGUS or Debian distribution or something like them, then we can be on their CD. I hope that I've not overcommitted us, or stepped on anybody's toes with this. I think it would be way cool to install Linux/MIPS off of a CD like this... Then again, I know someone with a CD-R that can make me one off CDs from a tape image if it comes down to it... So that should explain my thinking about these things now... It just might be reason enough to get a ZIP or Syquest drive so I can see if my BIOS will grok ISO-9660 images on it.... Or even just another hard disk that I can format via mkisofs :-). For the moment, I'm going to finish up the cp and see if I can reduce by boot times by 5-10 seconds. I know that doesn't sound like much, but every bit helps... Once I have that, I'll go ahead and try all the interrupt hacks that Ralf and others have suggested... Warner From linux@jens1.whu-koblenz.de Mon Feb 12 02:26:37 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id CAA21524 for ; Mon, 12 Feb 1996 02:26:36 +0100 (MET) Received: from obelix.WHU-Koblenz.DE by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA22550; Mon, 12 Feb 1996 02:26:25 +0100 (MET) Received: from jens1.whu-koblenz.de (linux@jens1.WHU-Koblenz.DE [193.141.48.51]) by obelix.WHU-Koblenz.de (8.6.8.1/8.6.4) with ESMTP id CAA24345 for ; Mon, 12 Feb 1996 02:26:28 +0100 Received: (from linux@localhost) by jens1.whu-koblenz.de (8.6.12/8.6.9) id CAA00563 for linux-mips@fnet.fr; Mon, 12 Feb 1996 02:25:57 +0100 From: linux-user Message-Id: <199602120125.CAA00563@jens1.whu-koblenz.de> Subject: Re: MILO To: linux-mips@fnet.fr Date: Mon, 12 Feb 1996 02:25:56 +0100 (MET) In-Reply-To: <199602120110.SAA04697@rover.village.org> from "Warner Losh" at Feb 11, 96 06:10:02 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: 1230 Lines: 29 Hi Warner, Warner Losh wrote: > All of this talk about install reminds me... I got a S.u.S.E. CD > offer in my email (for the ParcPlace OI giveway, no doubt) about the > time that Ralf put out root-0.00.tar.gz. I filled it out form they > had sent and dropped the "Yah, I'm working on Linux/MIPS and am > thinking of putting together a distribution for it, would you put that > on your CD?" Just to see what would happen (and at the time I was > thinking about this). It turns out they are interested! So if we > have a BOGUS or Debian distribution or something like them, then we > can be on their CD. I hope that I've not overcommitted us, or stepped > on anybody's toes with this. I think it would be way cool to install > Linux/MIPS off of a CD like this... Then again, I know someone with a > CD-R that can make me one off CDs from a tape image if it comes down > to it... Have a look on the CD. Linux/MIPS should already be there, at least parts of it. I talked to them a while ago and they changed the mirror source for linux-mips to uni-mainz. So all the good work should be on this CD :))) An integration into their YAST (jens panics and seeks a manhole) wouldn't be _that_ bad, would it? Greetings, Jens From linux@mailhost.uni-koblenz.de Mon Feb 12 02:38:27 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id CAA21596 for ; Mon, 12 Feb 1996 02:38:26 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA22637; Mon, 12 Feb 1996 02:38:15 +0100 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with ESMTP id CAA25800 for ; Mon, 12 Feb 1996 02:37:10 +0100 From: Systemkennung Linux Message-Id: <199602120137.CAA25800@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id CAA08385; Mon, 12 Feb 1996 02:37:29 +0100 Subject: Re: MILO To: linux-mips@fnet.fr Date: Mon, 12 Feb 1996 02:37:28 +0100 (MET) In-Reply-To: <199602120110.SAA04697@rover.village.org> from "Warner Losh" at Feb 11, 96 06:10:02 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 3329 Lines: 76 Hi, > : You're right; I've tried some days ago whether Arc drives may be > : accessed similar to UNIX block devices - et voila - it works though > : not documented. > > I would expect it to work because they have to be able to format the > disk somehow. And there is only one arcinst program, not a dti one > and a tyne one, etc. > > : So partitioning under ARC can really work (withhin the narrow limits set > : by some extremly buggy ARC implementations like the Magnum ...) > > Yes. I wonder how arcinst even runs on that system :-(. Because both BIOS and installer are MS and fit each other like lock and key ... > : What is missing is a port of termcap to Arc; then we could use programs > : like cfdisk for installation. Another program that might be interesting > : to be ported to Arc is an sash type shell with lots of builtin programs > : that recently has been published. Think it's name was mash. > > I like it. There is a song over hear called the Monster Mash that is > old enough everyone on the planet should have heard by now. ??? > : The sad fact that every Arc BIOS I've yet seen will turn writing a complex > : Arc program into a walk through a minefield. Even worse; Arc is bugged by > : design. There are all functions required to access and manipulate files - > : except a function to delete files. Btw. - the mines are (C) Microsoft ... > > So rm becomes much harder :-). > > All of this talk about install reminds me... I got a S.u.S.E. CD > offer in my email (for the ParcPlace OI giveway, no doubt) about the > time that Ralf put out root-0.00.tar.gz. I filled it out form they > had sent and dropped the "Yah, I'm working on Linux/MIPS and am > thinking of putting together a distribution for it, would you put that > on your CD?" Just to see what would happen (and at the time I was > thinking about this). It turns out they are interested! So if we > have a BOGUS or Debian distribution or something like them, then we > can be on their CD. I hope that I've not overcommitted us, or stepped > on anybody's toes with this. I think it would be way cool to install > Linux/MIPS off of a CD like this... Then again, I know someone with a > CD-R that can make me one off CDs from a tape image if it comes down > to it... Of course the Suse people are interested; I've met them several times and know one (Florian la Roche) pretty good as he is a friend of a friend :-) We've also talked to Lunetix, another Linux distribution vendor here in Germany and they're also interested just as Caldera is. You see: I did my homework :-) Probably we'll be taking the same path as most other distributions. Debian as free distribution and then RedHat as commercial distribution which will lead us in direction to Caldera. For now I suggest that just create tar packages until we have something better. > For the moment, I'm going to finish up the cp and see if I can reduce > by boot times by 5-10 seconds. I know that doesn't sound like much, > but every bit helps... At the time where I still was booting ramdisk images I was more interested in shrinking the loader because Milo + kernel + diskimage were about a HD in size ... > Once I have that, I'll go ahead and try all the interrupt hacks that > Ralf and others have suggested... Get coffee and Aspirin first :-) Ralf From linux@mailhost.uni-koblenz.de Mon Feb 12 02:40:15 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id CAA21661 for ; Mon, 12 Feb 1996 02:40:15 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA22672; Mon, 12 Feb 1996 02:40:04 +0100 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with ESMTP id CAA25836 for ; Mon, 12 Feb 1996 02:38:59 +0100 From: Systemkennung Linux Message-Id: <199602120138.CAA25836@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id CAA08403; Mon, 12 Feb 1996 02:39:17 +0100 Subject: Linux 1.3.60 To: linux-mips@fnet.fr Date: Mon, 12 Feb 1996 02:39:17 +0100 (MET) Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 517 Lines: 20 -----BEGIN PGP SIGNED MESSAGE----- Hi, I've uploaded linux-1.3.59-1.3.60.diffs.gz to lena.fnet.fr. It's almost 3am so thistime a short announcement ... Ralf 34de6429aac68099785590f0f7156dc2 linux-1.3.59-1.3.60.diffs.gz -----BEGIN PGP SIGNATURE----- Version: 2.6.2 iQCVAgUBMR6aQkckbl6vezDBAQH2TAP8DRWvTvMKXedY2DMv1CH4qjC4OZGllBCF /+qIdISJie8t5lVKgxVBlR6goHdQVP7HIwt4vhrVTF2V6F4O6un3PvHbXbsLiT31 XYIfbcTtRPv/IEEGP/8eAy2Uca5/EQloC42Y02EbX9muN/wHg+MrMlnI7Y5qd1Lr hOc6ik7BVqI= =8NsD -----END PGP SIGNATURE----- From imp@rover.village.org Mon Feb 12 06:43:36 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id GAA22893 for ; Mon, 12 Feb 1996 06:43:35 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA25330; Mon, 12 Feb 1996 06:43:24 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id WAA05058 for ; Sun, 11 Feb 1996 22:43:28 -0700 Message-Id: <199602120543.WAA05058@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Linux 1.3.60 In-Reply-To: Your message of Sun, 11 Feb 1996 21:34:01 Date: Sun, 11 Feb 1996 22:43:27 -0700 From: Warner Losh Content-Length: 288 Lines: 8 : I presume lena.fnet.fr is different from ftp.fnet.fr - that it is some : sort of staging area before general release... Ralf uploads it to lena, then a cron job fires up and copies it into /linux-mips/private/Incoming, then Stoned or Luc moves it to the right place on ftp... Warner From imp@rover.village.org Mon Feb 12 06:54:52 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id GAA22954 for ; Mon, 12 Feb 1996 06:54:51 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA25399; Mon, 12 Feb 1996 06:54:33 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id WAA05080 for ; Sun, 11 Feb 1996 22:54:30 -0700 Message-Id: <199602120554.WAA05080@rover.village.org> To: linux-mips@fnet.fr Subject: Re: MILO In-Reply-To: Your message of Mon, 12 Feb 1996 02:25:56 +0100 Date: Sun, 11 Feb 1996 22:54:29 -0700 From: Warner Losh Content-Length: 1503 Lines: 34 linux-user (jens) jens> So all the good work should be on this CD :))) jens> An integration into their YAST (jens panics and seeks a manhole) jens> wouldn't be _that_ bad, would it? Don't know, but I agree with Ralf that we're not (yet) that stage... The problem I'm dancing around right now is how do I get Linux onto the system so all the cool install programs can run.... For the moment, cp will be all that I need. Other things are needed over time like a way to partition the disk, a way to format the disk (both FAT and ext2fs), a way to get enough of Linux onto the system to boot, etc. etc. etc. I do have a crude version of cp working now in milo, so I can copy milo to my hard disk and change the parameters around in my boot menu... I'll coordinate with Stoned to get these patches into MILO 0.27 or 0.28 depending on what the state of his world is at right now. Ralf> Probably we'll be taking the same path as most other Ralf> distributions. Debian as free distribution and then RedHat as Ralf> commercial distribution which will lead us in direction to Ralf> Caldera. Ralf> For now I suggest that just create tar packages until we have Ralf> something better. That sounds good for the moment. At some point in the future we'll need to have a complete install system, but for now a compressed kernel, milo and a floppy or two would do the job. We might also be able to assume that the target system has NT running on it already (or ultrix) and leverage off of that fact. Warner From imp@rover.village.org Mon Feb 12 07:01:24 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id HAA23187 for ; Mon, 12 Feb 1996 07:01:24 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA25464; Mon, 12 Feb 1996 07:01:04 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id XAA05172 for ; Sun, 11 Feb 1996 23:00:59 -0700 Message-Id: <199602120600.XAA05172@rover.village.org> To: Linux/MIPS Mailing List Subject: Alpha Date: Sun, 11 Feb 1996 23:00:58 -0700 From: Warner Losh Content-Length: 126 Lines: 4 Does anybody know where to find the linux stuff for the Alpha? Maybe they have been down the ARC BIOS path before... Warner From beurton@nil.fnet.fr Mon Feb 12 10:55:22 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id KAA24185 for ; Mon, 12 Feb 1996 10:55:22 +0100 (MET) Received: from nil.fnet.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA28865; Mon, 12 Feb 1996 10:55:13 +0100 (MET) Received: from nil.fnet.fr (localhost [127.0.0.1]) by nil.fnet.fr (8.7.3/8.7.3) with ESMTP id KAA27259 for ; Mon, 12 Feb 1996 10:55:30 +0100 (MET) Message-Id: <199602120955.KAA27259@nil.fnet.fr> X-Mailer: exmh version 1.6 4/21/95 To: linux-mips@fnet.fr Subject: Re: Linux 1.3.60 In-Reply-To: Your message of "Mon, 12 Feb 1996 02:39:17 MET." <199602120138.CAA25836@informatik.uni-koblenz.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 12 Feb 1996 10:55:30 +0100 From: Luc Beurton Content-Length: 283 Lines: 13 > Hi, > > I've uploaded linux-1.3.59-1.3.60.diffs.gz to lena.fnet.fr. It's almost > 3am so thistime a short announcement ... Good night :) > Ralf > > 34de6429aac68099785590f0f7156dc2 linux-1.3.59-1.3.60.diffs.gz Well, it's on the ftp.fnet.fr (src/kernel/v1.3). Cheers Luc. From linux@mailhost.uni-koblenz.de Mon Feb 12 12:23:08 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id MAA24777 for ; Mon, 12 Feb 1996 12:23:07 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA00690; Mon, 12 Feb 1996 12:22:59 +0100 (MET) Received: from lars.uni-koblenz.de (linux@lars.uni-koblenz.de [141.26.6.189]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with ESMTP id MAA16688 for ; Mon, 12 Feb 1996 12:21:46 +0100 From: Systemkennung Linux Message-Id: <199602121121.MAA16688@informatik.uni-koblenz.de> Received: by lars.uni-koblenz.de (8.6.12/KO-2.0) id MAA00795; Mon, 12 Feb 1996 12:52:23 +0100 Subject: Re: MILO To: linux-mips@fnet.fr Date: Mon, 12 Feb 1996 12:52:22 +0100 (MET) In-Reply-To: <199602120554.WAA05080@rover.village.org> from "Warner Losh" at Feb 11, 96 10:54:29 pm X-Mailer: ELM [version 2.4 PL24] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 421 Lines: 12 Hi, > That sounds good for the moment. At some point in the future we'll > need to have a complete install system, but for now a compressed > kernel, milo and a floppy or two would do the job. We might also be > able to assume that the target system has NT running on it already (or > ultrix) and leverage off of that fact. No; I've never used NT on any MIPS box and even couldn't install it on my machine. Ralf From Stoned.Elipot@univ-evry.fr Mon Feb 12 13:13:52 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id NAA25357 for ; Mon, 12 Feb 1996 13:13:51 +0100 (MET) Received: from amiante.univ-evry.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA01590; Mon, 12 Feb 1996 13:13:43 +0100 (MET) Received: from bandsept.univ-evry.fr (bandsept.univ-evry.fr [192.93.214.19]) by amiante.univ-evry.fr with ESMTP (8.6.12/94.01.10); id MAA27956; Mon, 12 Feb 1996 12:26:00 +0100 Received: from localhost.univ-evry.fr (localhost.univ-evry.fr [127.0.0.1]) by bandsept.univ-evry.fr with SMTP (8.6.12/94.01.26); id NAA21058; Mon, 12 Feb 1996 13:12:21 +0100 Message-Id: <199602121212.NAA21058@bandsept.univ-evry.fr> To: linux-mips@fnet.fr Subject: Re: MILO + linux 1.3.59 on Magnum/Oily In-Reply-To: Your message of "Sun, 11 Feb 1996 18:10:02 MST." <199602120110.SAA04697@rover.village.org> X-It: bandsept (wrong OS, but right CPU) X-Mailer: MH 6.8.3, MH-E 5.0.2 Date: Mon, 12 Feb 1996 13:12:14 +0100 From: Stoned Elipot Content-Length: 1275 Lines: 35 Hi all, >>>>> "Warner" == Warner Losh writes: Warner> Yes. I wonder how arcinst even runs on that system :-(. Olivetti got sd.exe, a self-test program: it's really huge, have a kind of multi-tasking core (all the test can run concurently), and well it works :) [SNIP] Warner> So rm becomes much harder :-). I was once thinking about ARC fileutils... BTW sometimes I put milo on an hard disk, sure it's faster ! And well, the keyboard works again with kernel 1.3.59 on my Oily, it turns out that in keyboard_initialize(), after the KBD_RESET command I have to do *two* kbd_wait_for_input() before the I got the keyboard's 'OK power-on reset complete' (aka the keyboard returns 0xAA): first I got 0xff and then the 0xAA, strange, insn't it (no matter the setting of #if 0/1 - Stoned refs - in kbd_init)? Then mount() of root FS on floppy succeed, but I fail to launch /bin/sh... So the point is, Magnum/Oily are nearly up on 1.3.xx kernel. Hum,... nobody had tried Andy's sonic proto-driver since long ? It have to be hack a little to compile on 1.3.xx: I'm on it... Oh yes, one more thing: I end up using binary of GNU libc for mipsel 'cause I wasn't able to compile it: Ralf could you give us a little roadbook on this ? Thanks. Cheers, Stoned. From Stoned.Elipot@univ-evry.fr Mon Feb 12 13:18:35 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id NAA25404 for ; Mon, 12 Feb 1996 13:18:35 +0100 (MET) Received: from amiante.univ-evry.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA01760; Mon, 12 Feb 1996 13:18:32 +0100 (MET) Received: from bandsept.univ-evry.fr (bandsept.univ-evry.fr [192.93.214.19]) by amiante.univ-evry.fr with ESMTP (8.6.12/94.01.10); id MAA27971; Mon, 12 Feb 1996 12:30:50 +0100 Received: from localhost.univ-evry.fr (localhost.univ-evry.fr [127.0.0.1]) by bandsept.univ-evry.fr with SMTP (8.6.12/94.01.26); id NAA21077; Mon, 12 Feb 1996 13:17:13 +0100 Message-Id: <199602121217.NAA21077@bandsept.univ-evry.fr> To: linux-mips@fnet.fr Subject: Re: MILO In-Reply-To: Your message of "Sun, 11 Feb 1996 22:54:29 MST." <199602120554.WAA05080@rover.village.org> X-It: bandsept (wrong OS, but right CPU) X-Mailer: MH 6.8.3, MH-E 5.0.2 Date: Mon, 12 Feb 1996 13:17:06 +0100 From: Stoned Elipot Content-Length: 602 Lines: 14 >>>>> "Warner" == Warner Losh writes: [SNIP] Warner> I do have a crude version of cp working now in milo, so I can Warner> copy milo to my hard disk and change the parameters around in Warner> my boot menu... I'll coordinate with Stoned to get these Warner> patches into MILO 0.27 or 0.28 depending on what the state of Warner> his world is at right now. I'm ready to put out a 0.27 release: it got your patches and Ralf's latest one he sent to the list. Perhaps we can design a new Milo/Pandora scheme and throw in Pandora your 'fileutils' for a brand new 0.28 ? Cheers, Stoned. From linux@mailhost.uni-koblenz.de Mon Feb 12 12:53:13 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id MAA25059 for ; Mon, 12 Feb 1996 12:53:13 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA01157; Mon, 12 Feb 1996 12:52:21 +0100 (MET) Received: from lars.uni-koblenz.de (linux@lars.uni-koblenz.de [141.26.6.189]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with ESMTP id MAA18912 for ; Mon, 12 Feb 1996 12:51:12 +0100 From: Systemkennung Linux Message-Id: <199602121151.MAA18912@informatik.uni-koblenz.de> Received: by lars.uni-koblenz.de (8.6.12/KO-2.0) id NAA00908; Mon, 12 Feb 1996 13:21:49 +0100 Subject: Mail loss To: linux-mips@fnet.fr Date: Mon, 12 Feb 1996 13:21:48 +0100 (MET) X-Mailer: ELM [version 2.4 PL24] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 364 Lines: 10 Hi all, being subscribed to the mailinglist on two addresses found that sometimes mail to me gets lost. I've got not idea what the reason for this might be; the cause might be at fnet, local or something else ... So just in case I don't answer you email please mail me again and/or cc to ralf@waldorf-gmbh.de. And now get ready for Linux/MIPS 1.3.62! Ralf From linux@mailhost.uni-koblenz.de Mon Feb 12 13:42:21 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id NAA25533 for ; Mon, 12 Feb 1996 13:42:20 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA02229; Mon, 12 Feb 1996 13:42:14 +0100 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with ESMTP id NAA21825 for ; Mon, 12 Feb 1996 13:40:42 +0100 From: Systemkennung Linux Message-Id: <199602121240.NAA21825@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id NAA09465; Mon, 12 Feb 1996 13:40:58 +0100 Subject: Re: MILO + linux 1.3.59 on Magnum/Oily To: linux-mips@fnet.fr Date: Mon, 12 Feb 1996 13:40:57 +0100 (MET) In-Reply-To: <199602121212.NAA21058@bandsept.univ-evry.fr> from "Stoned Elipot" at Feb 12, 96 01:12:14 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1990 Lines: 51 Hi all, > [SNIP] > Warner> So rm becomes much harder :-). > > I was once thinking about ARC fileutils... BTW sometimes I put milo on > an hard disk, sure it's faster ! How about booting Milo & vmlinux from one disk and a compressed ramdisk image from the other? I think we can assume at least 8mb RAM for every MIPS machine so this should be usable and provide the usual Linux environment for installation. > And well, the keyboard works again with kernel 1.3.59 on my Oily, it > turns out that in keyboard_initialize(), after the KBD_RESET command I > have to do *two* kbd_wait_for_input() before the I got the keyboard's > 'OK power-on reset complete' (aka the keyboard returns 0xAA): first I > got 0xff and then the 0xAA, strange, insn't it Thanks; in the meantime we've installed Wayne's Olli here at the Uni. His machine showed no keyboard problems at all. It might be interesting if changing the keyboard affects the problem? The bigger problem is that the Olli console is a memory pig and screen output seems to start at some random line after launching the kernel. > (no matter the setting > of #if 0/1 - Stoned refs - in kbd_init)? OK, then I'll remove that code. > Then mount() of root FS on floppy succeed, but I fail to launch > /bin/sh... > > So the point is, Magnum/Oily are nearly up on 1.3.xx > kernel. Hum,... nobody had tried Andy's sonic proto-driver since long > ? It have to be hack a little to compile on 1.3.xx: I'm on it... There were some changes to the driver iterface; I wouldn't really bet that the Sonic driver provides still as much "functionality" as it did for 1.2 ;-) > Oh yes, one more thing: I end up using binary of GNU libc for mipsel > 'cause I wasn't able to compile it: Ralf could you give us a little > roadbook on this ? Thanks. I'm just updating from libc 960120 to yesterday's snapshot. I'll make a new set of diffs and write a little howto. Just have some days patience. Ralf From imp@rover.village.org Wed Feb 14 07:35:15 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id HAA05412 for ; Wed, 14 Feb 1996 07:35:10 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA11525; Wed, 14 Feb 1996 07:34:52 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id XAA03281 for ; Tue, 13 Feb 1996 23:34:43 -0700 Message-Id: <199602140634.XAA03281@rover.village.org> To: Linux/MIPS Mailing List Subject: Slower BogoMIPS rating on 1.3.62 than on 1.2.10 :-) Date: Tue, 13 Feb 1996 23:34:34 -0700 From: Warner Losh Content-Length: 3279 Lines: 70 OK. I'm really upset now. 1.2.10.4 told me that I had a BogoMIPS rating of 50.03. Now, I go and boot 1.3.62 and I have only 49.87. Who, exactly, is going to pick up the over .16 BogoMIP? Tell me that, why don't you? BogoMIPS don't grow on tress, you know. If I wanted only 49.87 BogoMIPS, I'd be running NT... :-) :-) :-) :-) :-) :-) :-) :-) :-) :-) :-) :-) :-) :-) :-) :-) :-) :-) :-) Or, in other words, 1.3.62 seems to work for me, for a suitible definition of work. Here's the boot sequence. Seems like a thing or two is amiss: Launching Kernel... (bold font, rest lighter) Console: 0 point font, 0 scans Console: colour VGA+ 80x50, 1 virtual console (max 63) Calibrating delay loop.. ok - 49.87 BogoMIPS Memory: 31280k/32768k available (424k kernel code, 964k data) Swansea University Computer Society NET3.033 for Linux 1.3.50 NET3: Unix domain sockets 0.10 BETA for Linux NET3.033. Checking for 'wait' instruction... unavailable. Linux version 1.3.62 (imp@rover.village.org) (gcc version 2.7.2) #29 Tue Feb 13 22:36:40 MST 1996 initialize_kbd: reset kbd failed, no ACK. Floppy drive(s): fd0 is 1.44M Started kswapd v 1.1.1.2 floppy0: no floppy controllers found VFS: Insert root floppy and press ENTER _ Well, as you might have guessed from the above messages, it hangs with the blinking underscore. The keyboard is dead. But after a time, the screen saver comes on :-) I had to transcribe the above quickly... This leads me to some questions... 1) Why can't it find the keyboard? Another interrupt thing? I rebooted a second time (to see if I could find my missing bogomips) and found that I got a *flood* of keyboard error messages. 2) Why did it find a floopy (fd0) and then say it couldn't find a controller? 3) Shouldn't I have a WAIT instruction? Hmmm... Not in the MUM, so maybe not. Anyway, this is with an almost a bog standard 1.3.62 distribution. This is good news, I think. The diffs between what I built and 1.3.62 are about 262 lines long. The diffs are in the ./Makefile (gmake and awk rather than make and gawk), the load address chanage that I sent a while ago to this list in arch/mips/Makefile, a couple of lines of debug that incremented the upper righthand corner of the screen on each interrupt in arch/mips/kernel/rpc44.S, spurious_interrupt removed from rpc44.S and tyne.S, a horrible kludge in init/main.c in calibrate_delay (outb(0,0x21)), a hack to scripts/Config to use a local bash and a horrible video kludge vga.c (which is replicated code). Time to take my kludges out one at a time and see if that helps my situation. Anyway, this is likely trivia for most people, but good places to look if you are adding a new machine type. Why does it now work? Obviously, it is because Ralf is a God. It may have also been faulty hardware on my end. I noticed that the system BIOS thought LPT lived at IRQ 0 for some reasoon. Changed that to 7, but the old kernels wouldn't boot right, the "second" interrupt problem that I talked about before, but the new ones booted as above. I now have a new slew of problems to solve, plus some interesting integration issues to solve that Ralf raised. Plus all the other cool stuff I wanted to work on :-). I'm much happier now about my MIPS box... Warner From Stoned.Elipot@univ-evry.fr Wed Feb 14 12:22:49 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id MAA07035 for ; Wed, 14 Feb 1996 12:22:48 +0100 (MET) Received: from amiante.univ-evry.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA16676; Wed, 14 Feb 1996 12:22:43 +0100 (MET) Received: from bandsept.univ-evry.fr (bandsept.univ-evry.fr [192.93.214.19]) by amiante.univ-evry.fr with ESMTP (8.6.12/94.01.10); id LAA05255; Wed, 14 Feb 1996 11:35:03 +0100 Received: from localhost.univ-evry.fr (localhost.univ-evry.fr [127.0.0.1]) by bandsept.univ-evry.fr with SMTP (8.6.12/94.01.26); id MAA01421; Wed, 14 Feb 1996 12:21:25 +0100 Message-Id: <199602141121.MAA01421@bandsept.univ-evry.fr> To: linux-mips@fnet.fr Subject: Re: Slower BogoMIPS rating on 1.3.62 than on 1.2.10 :-) In-Reply-To: Your message of "Tue, 13 Feb 1996 23:34:34 MST." <199602140634.XAA03281@rover.village.org> X-It: bandsept (wrong OS, but right CPU) X-Mailer: MH 6.8.3, MH-E 5.0.2 Date: Wed, 14 Feb 1996 12:21:18 +0100 From: Stoned Elipot Content-Length: 1006 Lines: 32 Hi all, >>>>> "Warner" == Warner Losh writes: [SNIP] Warner> 1) Why can't it find the keyboard? Another interrupt Warner> thing? I rebooted a second time (to see if I could find my Warner> missing bogomips) and found that I got a *flood* of keyboard Warner> error messages. Warner> 2) Why did it find a floopy (fd0) and then Warner> say it couldn't find a controller? I got these too, the solution is: increase the value of TIMEOUT_CONST in keyboard.c, it's used by kbd_wait_for_input(), after I increased it, keyboard initialization went ok and floppy too !! Warner> 3) Shouldn't I have a WAIT Warner> instruction? Hmmm... Not in the MUM, so maybe not. I don't think so. [SNIP] Warner> each interrupt in arch/mips/kernel/rpc44.S, spurious_interrupt Warner> removed from rpc44.S and tyne.S, a horrible kludge in Idem for magnum.c and spurious_interrupt() for my Oily. [SNIP] Warner> Why does it now work? Obviously, it is because Ralf is a God. Yes ! ;) Cheers, Stoned. From 74155@foo.hva.nl Wed Feb 14 13:13:39 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id NAA07353 for ; Wed, 14 Feb 1996 13:13:28 +0100 (MET) Received: from gatekeeper.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA17484; Wed, 14 Feb 1996 13:13:19 +0100 (MET) Received: from foo-dns.foo.hva.nl (foo-dns.foo.hva.nl [145.92.99.250]) by gatekeeper.waldorf-gmbh.de (8.6.12/8.6.9) with SMTP id NAA10084 for ; Wed, 14 Feb 1996 13:05:18 +0100 Received: from foobee.foo.hva.nl by foo-dns.foo.hva.nl with smtp (Smail3.1.29.1 #4) id m0tmftH-0003UyC; Wed, 14 Feb 96 13:00 MET Received: from FOO_STD/SpoolDir by foobee.foo.hva.nl (Mercury 1.21); 14 Feb 96 12:47:46 GMT+0200 Received: from SpoolDir by FOO_STD (Mercury 1.21); 14 Feb 96 12:47:27 GMT+0200 From: "Ivo Spanjersberg" <74155@foo.hva.nl> Organization: Hogeschool van Amsterdam, FOO To: linux@waldorf-gmbh.de Date: Wed, 14 Feb 1996 12:47:25 +0100 Subject: Linux/MIPS Priority: normal X-Mailer: Pegasus Mail/Windows (v1.22) Message-Id: <949EE8107E@foobee.foo.hva.nl> Content-Length: 224 Lines: 12 Dear Sir, Could you please send me information on Linux/MIPS? Also, if possible, could you please tell me where I can get more information on the MIPS in general. I only have e-mail. Yours sincerely, Ivo Spanjersberg From linux@mailhost.uni-koblenz.de Wed Feb 14 14:35:06 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id OAA07874 for ; Wed, 14 Feb 1996 14:35:05 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA18666; Wed, 14 Feb 1996 14:34:30 +0100 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with ESMTP id OAA02080 for ; Wed, 14 Feb 1996 14:32:23 +0100 From: Systemkennung Linux Message-Id: <199602141332.OAA02080@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id OAA14562; Wed, 14 Feb 1996 14:12:33 +0100 Subject: Re: Slower BogoMIPS rating on 1.3.62 than on 1.2.10 :-) To: linux-mips@fnet.fr Date: Wed, 14 Feb 1996 14:12:32 +0100 (MET) In-Reply-To: <199602140634.XAA03281@rover.village.org> from "Warner Losh" at Feb 13, 96 11:34:34 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 4257 Lines: 98 Hi, > OK. I'm really upset now. 1.2.10.4 told me that I had a BogoMIPS > rating of 50.03. Now, I go and boot 1.3.62 and I have only 49.87. > Who, exactly, is going to pick up the over .16 BogoMIP? Tell me that, > why don't you? BogoMIPS don't grow on tress, you know. If I wanted > only 49.87 BogoMIPS, I'd be running NT... > > :-) :-) :-) :-) :-) :-) :-) :-) :-) :-) :-) :-) :-) :-) :-) :-) :-) :-) :-) I suggest to replace the BogoMIPS benchmark in the kernel by an assignment. It's more efficient, anyway ;-) > Or, in other words, 1.3.62 seems to work for me, for a suitible > definition of work. Here's the boot sequence. Seems like a thing or > two is amiss: > > Launching Kernel... (bold font, rest lighter) > Console: 0 point font, 0 scans > Console: colour VGA+ 80x50, 1 virtual console (max 63) > Calibrating delay loop.. ok - 49.87 BogoMIPS > Memory: 31280k/32768k available (424k kernel code, 964k data) > Swansea University Computer Society NET3.033 for Linux 1.3.50 > NET3: Unix domain sockets 0.10 BETA for Linux NET3.033. > Checking for 'wait' instruction... unavailable. Correct for R4000. > Linux version 1.3.62 (imp@rover.village.org) (gcc version 2.7.2) #29 Tue Feb 13 22:36:40 MST 1996 > initialize_kbd: reset kbd failed, no ACK. > Floppy drive(s): fd0 is 1.44M > Started kswapd v 1.1.1.2 > floppy0: no floppy controllers found > VFS: Insert root floppy and press ENTER > _ > > Well, as you might have guessed from the above messages, it hangs with > the blinking underscore. The keyboard is dead. But after a time, the > screen saver comes on :-) I had to transcribe the above quickly... > > This leads me to some questions... > > 1) Why can't it find the keyboard? Another interrupt thing? > I rebooted a second time (to see if I could find my missing > bogomips) and found that I got a *flood* of keyboard error > messages. > 2) Why did it find a floopy (fd0) and then say it couldn't > find a controller? It kind a assumes the existence of a floppy. The kernel really tries to detect the type of floppy controller this might fail just as the keyboard stuff if either your ports aren't accessible or interrupts don't work. > 3) Shouldn't I have a WAIT instruction? Hmmm... Not in the > MUM, so maybe not. No, R4000/R4400 don't have one. Most other R4xx0 like the R4200, R4300i, R4600, R4650, R4700 have a wait insn. > Anyway, this is with an almost a bog standard 1.3.62 distribution. > This is good news, I think. The diffs between what I built and 1.3.62 > are about 262 lines long. The diffs are in the ./Makefile (gmake and > awk rather than make and gawk), the load address chanage that I sent a > while ago to this list in arch/mips/Makefile, a couple of lines of > debug that incremented the upper righthand corner of the screen on > each interrupt in arch/mips/kernel/rpc44.S, spurious_interrupt removed > from rpc44.S and tyne.S, a horrible kludge in init/main.c in > calibrate_delay (outb(0,0x21)), a hack to scripts/Config to use a > local bash and a horrible video kludge vga.c (which is replicated > code). Time to take my kludges out one at a time and see if that > helps my situation. Anyway, this is likely trivia for most people, > but good places to look if you are adding a new machine type. > > Why does it now work? Obviously, it is because Ralf is a God. It may Think I'll have to print out this mail and stick it to the wall :->>> If I only could resurrect my car's electronic ignition that easily ... > have also been faulty hardware on my end. I noticed that the system > BIOS thought LPT lived at IRQ 0 for some reasoon. Changed that to 7, > but the old kernels wouldn't boot right, the "second" interrupt > problem that I talked about before, but the new ones booted as above. > > I now have a new slew of problems to solve, plus some interesting > integration issues to solve that Ralf raised. Plus all the other cool > stuff I wanted to work on :-). > > I'm much happier now about my MIPS box... Congratulations - well done! After some learning last night I tried to compile sendspam^H^H^H^Hmail for Linux/MIPS and voila it worked almost out of the box. I just had to add the two missing syscalls statfs and fstatfs to glibc. Ralf From imp@rover.village.org Thu Feb 15 17:45:01 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id RAA00586 for ; Thu, 15 Feb 1996 17:44:43 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA14748; Thu, 15 Feb 1996 17:44:17 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id JAA07809 for ; Thu, 15 Feb 1996 09:44:15 -0700 Message-Id: <199602151644.JAA07809@rover.village.org> To: linux-mips@fnet.fr Subject: Re: DecStation and 1.3.59 [was: Re: Load address of 1.3.59] In-Reply-To: Your message of Sun, 11 Feb 1996 10:02:58 +0100 Date: Thu, 15 Feb 1996 09:44:15 -0700 From: Warner Losh Content-Length: 530 Lines: 13 : This is not the only macro instruction in MIPS assembler; in fact most of : the instructions are macro instructions that expand sometimes into five : instructions or more. Many of them are described in the appendix of : the Kane. Ralf, Just noticed this in my inbox... Anyway, could you please give those of us who have to order books like this a 3-4 line summary of what is contained in Kane's book? Would it be useful to me in my work to get the Deskstation up and running, or will the MUM that I have be enough. Warner From imp@rover.village.org Thu Feb 15 17:47:32 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id RAA00623 for ; Thu, 15 Feb 1996 17:47:30 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA14785; Thu, 15 Feb 1996 17:47:20 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id JAA07843 for ; Thu, 15 Feb 1996 09:47:21 -0700 Message-Id: <199602151647.JAA07843@rover.village.org> To: linux-mips@fnet.fr Subject: Re: DecStation and 1.3.59 [was: Re: Load address of 1.3.59] In-Reply-To: Your message of Sun, 11 Feb 1996 10:02:58 +0100 Date: Thu, 15 Feb 1996 09:47:20 -0700 From: Warner Losh Content-Length: 169 Lines: 6 : ISBN : 0-12-58479-4 : The book contains another ISBN: 0-13-584293-X, so The first ISBN you gave is too short. You need 11 characters in it... Warner From linux@mailhost.uni-koblenz.de Thu Feb 15 18:03:36 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id SAA00891 for ; Thu, 15 Feb 1996 18:03:33 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA15103; Thu, 15 Feb 1996 18:03:10 +0100 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with ESMTP id SAA06321 for ; Thu, 15 Feb 1996 18:00:51 +0100 From: Systemkennung Linux Message-Id: <199602151700.SAA06321@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id SAA21732; Thu, 15 Feb 1996 18:01:41 +0100 Subject: Re: DecStation and 1.3.59 [was: Re: Load address of 1.3.59] To: linux-mips@fnet.fr Date: Thu, 15 Feb 1996 18:01:40 +0100 (MET) In-Reply-To: <199602151644.JAA07809@rover.village.org> from "Warner Losh" at Feb 15, 96 09:44:15 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1546 Lines: 44 Hi, > : This is not the only macro instruction in MIPS assembler; in fact most of > : the instructions are macro instructions that expand sometimes into five > : instructions or more. Many of them are described in the appendix of > : the Kane. > > Ralf, > Just noticed this in my inbox... Anyway, could you please > give those of us who have to order books like this a 3-4 line summary > of what is contained in Kane's book? Would it be useful to me in my > work to get the Deskstation up and running, or will the MUM that I > have be enough. [FAQ on] What Literature about MIPS CPUs is available? The book MIPS RISC Architecture from Gerry Kane is something like a bible for MIPS programmers. It covers all CPUs upto the R4000 family. The book also contains much other usefull information like the MIPS calling sequence, a list of macro instructions that expand into multiple machine instructions, examples of multiprecission arithmetic and more: Name MIPS RISC Architecture Author Gerry Kane Publisher Prentice Hall, Englewood Cliffs, N.J. 07632 ISBN 0-12-58479-4 There's another ISBN mentioned in the book (0-13-584293-X), so be careful. There is some additional literature about available as Postscript file on SGI's FTP Server. These books contain much more details about the instruction set and hardware but don't cover software aspects very good. Another advantage is of course that these files are for free. [FAQ off] Ralf From linux@mailhost.uni-koblenz.de Thu Feb 15 18:04:18 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id SAA00920 for ; Thu, 15 Feb 1996 18:04:16 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA15126; Thu, 15 Feb 1996 18:04:02 +0100 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with ESMTP id SAA06364 for ; Thu, 15 Feb 1996 18:02:07 +0100 From: Systemkennung Linux Message-Id: <199602151702.SAA06364@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id SAA21751; Thu, 15 Feb 1996 18:03:02 +0100 Subject: Re: DecStation and 1.3.59 [was: Re: Load address of 1.3.59] To: linux-mips@fnet.fr Date: Thu, 15 Feb 1996 18:03:01 +0100 (MET) In-Reply-To: <199602151647.JAA07843@rover.village.org> from "Warner Losh" at Feb 15, 96 09:47:20 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 248 Lines: 10 Hi, > : ISBN : 0-12-58479-4 > : The book contains another ISBN: 0-13-584293-X, so > > The first ISBN you gave is too short. You need 11 characters in it... It's what is printed on the book so Prentice Hall has bugged it. Ralf From imp@rover.village.org Thu Feb 15 18:33:15 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id SAA01077 for ; Thu, 15 Feb 1996 18:33:14 +0100 (MET) Received: from chenas.inria.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA15710; Thu, 15 Feb 1996 18:33:08 +0100 (MET) Received: from rover.village.org by chenas.inria.fr (5.65c8d/92.02.29) via EUnet-France id AA01591; Thu, 15 Feb 1996 18:33:07 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id KAA08188 for ; Thu, 15 Feb 1996 10:27:54 -0700 Message-Id: <199602151727.KAA08188@rover.village.org> To: linux-mips@fnet.fr Subject: Re: DecStation and 1.3.59 [was: Re: Load address of 1.3.59] In-Reply-To: Your message of Thu, 15 Feb 1996 18:01:40 +0100 Date: Thu, 15 Feb 1996 10:27:53 -0700 From: Warner Losh Content-Length: 55 Lines: 6 : [FAQ on] : [FAQ off] Thanks Ralf... Warner From imp@rover.village.org Fri Feb 16 07:48:25 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id HAA06667 for ; Fri, 16 Feb 1996 07:48:24 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA25782; Fri, 16 Feb 1996 07:48:11 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id XAA10180 for ; Thu, 15 Feb 1996 23:48:06 -0700 Message-Id: <199602160648.XAA10180@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Slower BogoMIPS rating on 1.3.62 than on 1.2.10 :-) In-Reply-To: Your message of Wed, 14 Feb 1996 12:21:18 +0100 Date: Thu, 15 Feb 1996 23:48:05 -0700 From: Warner Losh Content-Length: 1157 Lines: 23 : I got these too, the solution is: increase the value of TIMEOUT_CONST : in keyboard.c, it's used by kbd_wait_for_input(), after I increased : it, keyboard initialization went ok and floppy too !! I wish it had been that easy. I increased TIMEOUT_CONST by a factor of 10 and it didn't help. I keep getting 0xff on reads from the keyboard port. Something is messed up with my interrupts, I think, or maybe there are bugs in the keyboard code that I'm tripping over, but I think the former is more likely. I tried enabling more interrupts in the IM field of the status register, and that didn't help either. Since it was none of the easy and trivial fixes, it must mean I'll have to get some time when I have a brain and hack rpc44.S.... Thanks for the suggestions. I feared it was going to be hard, and it is turning out to be not easy... Warner P.S. If you are into CVS and are using CVS to track kernel changes (or if you want to be a guinee pig), please check out http://www.village.org/villagers/imp and follow the CVS link (or any of them if you want other things) to a draft howto on how to do this sort of thing that I wrote this evening. From linux@mailhost.uni-koblenz.de Sat Feb 17 02:52:24 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id CAA25435 for ; Sat, 17 Feb 1996 02:52:23 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA14659; Sat, 17 Feb 1996 02:52:14 +0100 (MET) Received: from brecht (brecht.uni-koblenz.de [141.26.4.62]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with SMTP id CAA00519 for ; Sat, 17 Feb 1996 02:50:32 +0100 From: Systemkennung Linux Message-Id: <199602170150.CAA00519@informatik.uni-koblenz.de> Received: by brecht (5.x/KO-2.0) id AA06272; Sat, 17 Feb 1996 02:52:32 +0100 Subject: Re: Slower BogoMIPS rating on 1.3.62 than on 1.2.10 :-) To: linux-mips@fnet.fr Date: Sat, 17 Feb 1996 02:52:30 +0100 (MET) In-Reply-To: <199602160648.XAA10180@rover.village.org> from "Warner Losh" at Feb 15, 96 11:48:05 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 824 Lines: 17 Hi, > I wish it had been that easy. I increased TIMEOUT_CONST by a factor > of 10 and it didn't help. I keep getting 0xff on reads from the > keyboard port. Something is messed up with my interrupts, I think, or > maybe there are bugs in the keyboard code that I'm tripping over, but > I think the former is more likely. I tried enabling more interrupts > in the IM field of the status register, and that didn't help either. > Since it was none of the easy and trivial fixes, it must mean I'll > have to get some time when I have a brain and hack rpc44.S.... Your board is based on a PC chipset which means than keyboard controller access via ports is dead slow. On the Jazz chipset all this stuff is implemented in a different way and therefore MUCH faster. That's why we needed to increase TIMEOUT_CONST. Ralf From imp@rover.village.org Sat Feb 17 05:40:16 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id FAA27802 for ; Sat, 17 Feb 1996 05:40:14 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA16834; Sat, 17 Feb 1996 05:40:05 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id VAA13252 for ; Fri, 16 Feb 1996 21:40:05 -0700 Message-Id: <199602170440.VAA13252@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Slower BogoMIPS rating on 1.3.62 than on 1.2.10 :-) In-Reply-To: Your message of Sat, 17 Feb 1996 02:52:30 +0100 Date: Fri, 16 Feb 1996 21:40:04 -0700 From: Warner Losh Content-Length: 523 Lines: 11 : Your board is based on a PC chipset which means than keyboard controller : access via ports is dead slow. On the Jazz chipset all this stuff is : implemented in a different way and therefore MUCH faster. That's why we : needed to increase TIMEOUT_CONST. I'm not sure I understand what this means. Does this mean that I too need to increase the TIMEOUT_CONST and I should try 100 or 1000x the original value, or does this mean "I'm not surprised that increasing it didn't help conentrate on those interrupts"? Warner From imp@rover.village.org Sat Feb 17 06:47:37 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id GAA28705 for ; Sat, 17 Feb 1996 06:47:36 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA17355; Sat, 17 Feb 1996 06:47:25 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id WAA13756 for ; Fri, 16 Feb 1996 22:47:26 -0700 Message-Id: <199602170547.WAA13756@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Slower BogoMIPS rating on 1.3.62 than on 1.2.10 :-) In-Reply-To: Your message of Wed, 14 Feb 1996 14:12:32 +0100 Date: Fri, 16 Feb 1996 22:47:26 -0700 From: Warner Losh Content-Length: 1052 Lines: 21 OK. I fixed a minor mismatch between rpc44.S and pica.S. They were almost, but not quite, identical. It wouldn't have impacted the keyboard driver (since it only impacted IRQs larger than 7), nor would it have impacted the floppy problem. I've also upped the timeout to be insanely large (1000 x the original value). The silly thing waits for 50 seconds now in the boot process... Neither one of these items helped much. After disabling the error message for scan code 0xff, I get one keyboard: unrecognized scancode (70) - ignored, message and then it finds FD0:. Claims that it is a 8272A. I then get more unrecognized scancode messages (63 and 7f) when I'm prompted for the root floopy. At this point, my keyboard never works. NumLock doesn't work at any point in the boot process once Linux is launched. Anyway, I'm seeing what the deal is here. Obviously Linux is doing something that my keyboard doesn't like..... Time to go off and debug it (although it is tempting to hack the insert root floppy code to just keep going). Warner From imp@rover.village.org Sat Feb 17 07:40:26 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id HAA29514 for ; Sat, 17 Feb 1996 07:40:25 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA17791; Sat, 17 Feb 1996 07:40:13 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id XAA14292 for ; Fri, 16 Feb 1996 23:40:17 -0700 Message-Id: <199602170640.XAA14292@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Slower BogoMIPS rating on 1.3.62 than on 1.2.10 :-) In-Reply-To: Your message of Wed, 14 Feb 1996 14:12:32 +0100 Date: Fri, 16 Feb 1996 23:40:17 -0700 From: Warner Losh Content-Length: 1390 Lines: 28 I disabled the initialize_kbd code in keyboard.c. It was making my keyboard go *NUTS*. I was getting all kinds of odd keyboard interrupts when this happened. So, I tried disabling this code and I can now hit return to continue (kinda) the boot process (it dies when trying to read the floppy for reasons unknown). Once I disabled that code, the keyboard started interrupting normally again. I added debugs to print the raw keycodes as they were coming out of they interrupt driver... When I press Return, I get a scancode of 5a back from the keyboard. When it is released, I get a f0 followed by a 5a from the keyboard, but it looks like the driver is expecting ca only and isn't quite sure what to do with the f0 5a. Call me crazy, but this doesn't sound quite right for keyscans. A quick peek at my BSD keymap (which is slightly hacked, but not seriously) shows that 5a (which is 90 decimal) should be rctrl. The <- key is 0x66 (98) which should be fkey55 per the map. Odd. Very odd. Why do I get this oddness, I wonder? The same keyboard works great on my FreeBSD box. The keyboard has a A X switch on it, which is in the A position. Anyway, sounds like I have some hacking to do to the keyboard driver. Looks like it is almost, but not quite, what is expected. I can come up with tables and new code, but I really want to understand this off the wall behavior. Warner From imp@rover.village.org Sun Feb 18 02:28:19 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id CAA16279 for ; Sun, 18 Feb 1996 02:28:16 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA27509; Sun, 18 Feb 1996 02:27:50 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id SAA16269 for ; Sat, 17 Feb 1996 18:27:46 -0700 Message-Id: <199602180127.SAA16269@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Slower BogoMIPS rating on 1.3.62 than on 1.2.10 :-) In-Reply-To: Your message of Fri, 16 Feb 1996 23:40:17 MST Date: Sat, 17 Feb 1996 18:27:46 -0700 From: Warner Losh Content-Length: 2998 Lines: 62 : Anyway, sounds like I have some hacking to do to the keyboard driver. Trivia for the keyboard hackers out there (Ralf likely knew this already, since I stole his code): PC keyboards have at least two scan modes. Mode 1 you get a scan code when the key goes down, repeats of it per the autorepeat stuff, and then an 0xf0 followed by the key that is coming up. In Mode 2 you when key k goes down, you get scan k, repeats, and then a k + 0x80 when it comes back up. Mode 1 and Mode 2 scan codes are completely different. Linux and X (I think) presume Mode 2. The Deskstation starts out in Mode 1. Putting it in Mode 2 either by using the FreeBSD source (which sends 0x4d to to keyboard port) or the Linux stuff on the #if 0 of the pica (which sends 41) or by the method used in initalize_kbd (which sends a 65) gets things bascially working for me. For reasons unknown, initialize_kbd really hurts my keyboard's brains, so I've disabled it when you are running on a Deskstation. With this disable, I get a single bogus scan code, but I might be able to "eat" it before interrupts are enabled. I now get to the VFS: Insert root floppy and press ENTER stage. However, the read never seems to complete. Sound familiar to anybody? If not, then I'll take a look into this and see what is up. Actually, now that I'm this far, it is time to drop back 10 and do some of the cleanup work that I need to do: o Get rid of ifdefs in slots.h and mipsconfig.h through proper TLB initialization. o Load the kernel at 0x80100000 (KSEG0 + 1M) on the deskstation to get around the 640K limitations of the motherboard (0x800a0000-0x800fffff is where the ISA bus is memory mapped). Make sure that the low memory isn't wasted, especially on ACER machines. Alternatively, figure out how to have a hole in the image... o Hack MILO to: + More minimal in verbage on boot (old patch integration) + Compressed kernels + Get the commands to actually work, rather than always boot linux + Get the command args from the ARC BIOS to work on the Deskstation. + Patches to allow a minimalistic MILO + Proper memory stuff on the Deskstation: SIMM detection, etc. o Look for all the XXX fixme places in the kernel that I put in and fix them. o Take a look at the Kernel Debugging via Serial port stuff that Andy did and see if I can get them working. o Get rid of the single spurious bogus scan code o Figure out the FDC driver problems o Get the Bustech SCSI driver working o Investigate removing some of the if (mips_machtype == xxx) code by better use of "virtual" functions (aka function pointers in structs) to allow for easier machine integration and more unified kernels (long term goal). o continue work on various howtos that I have out there, especially the word smithing of the linux-mips howto. More progress. Life is good. Must drink microbrews in honor of progress and Drew's birthday... Warner From linux@mailhost.uni-koblenz.de Mon Feb 19 11:29:49 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id LAA15936 for ; Mon, 19 Feb 1996 11:29:42 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA19109; Mon, 19 Feb 1996 11:29:23 +0100 (MET) Received: from flake.uni-koblenz.de (linux@flake.uni-koblenz.de [141.26.4.37]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with SMTP id LAA22668 for ; Mon, 19 Feb 1996 11:27:50 +0100 From: Systemkennung Linux Message-Id: <199602191027.LAA22668@informatik.uni-koblenz.de> Received: by flake.uni-koblenz.de (4.1/KO-2.0) id AA28628; Mon, 19 Feb 96 11:28:28 +0100 Subject: Re: Slower BogoMIPS rating on 1.3.62 than on 1.2.10 :-) To: linux-mips@fnet.fr Date: Mon, 19 Feb 1996 11:28:28 +0100 (MET) In-Reply-To: <199602170440.VAA13252@rover.village.org> from "Warner Losh" at Feb 16, 96 09:40:04 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 596 Lines: 15 Hi, > : Your board is based on a PC chipset which means than keyboard controller > : access via ports is dead slow. On the Jazz chipset all this stuff is > : implemented in a different way and therefore MUCH faster. That's why we > : needed to increase TIMEOUT_CONST. > > I'm not sure I understand what this means. Does this mean that I too > need to increase the TIMEOUT_CONST and I should try 100 or 1000x the > original value, or does this mean "I'm not surprised that increasing > it didn't help conentrate on those interrupts"? I don't think increasing will help in your case. Ralf From linux@mailhost.uni-koblenz.de Mon Feb 19 12:04:03 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id MAA16558 for ; Mon, 19 Feb 1996 12:04:02 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA19678; Mon, 19 Feb 1996 12:03:51 +0100 (MET) Received: from flake.uni-koblenz.de (linux@flake.uni-koblenz.de [141.26.4.37]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with SMTP id MAA23515 for ; Mon, 19 Feb 1996 12:02:28 +0100 From: Systemkennung Linux Message-Id: <199602191102.MAA23515@informatik.uni-koblenz.de> Received: by flake.uni-koblenz.de (4.1/KO-2.0) id AA28699; Mon, 19 Feb 96 12:03:06 +0100 Subject: Re: Slower BogoMIPS rating on 1.3.62 than on 1.2.10 :-) To: linux-mips@fnet.fr Date: Mon, 19 Feb 1996 12:03:05 +0100 (MET) In-Reply-To: <199602180127.SAA16269@rover.village.org> from "Warner Losh" at Feb 17, 96 06:27:46 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 813 Lines: 20 Hi, > o Load the kernel at 0x80100000 (KSEG0 + 1M) on the > deskstation to get around the 640K limitations of the > motherboard (0x800a0000-0x800fffff is where the ISA bus is > memory mapped). Make sure that the low memory isn't > wasted, especially on ACER machines. Alternatively, figure > out how to have a hole in the image... How about: load the kernel to 0x80100000, copy the exception handlers down to KSEG0, declare KSEG0+sizeof(lowmemstuff) as memory base address and all the kernel code+data+bss to be memory in use. Sounds easier than hacking complex things for the loader just to cut the kernel into two pieces. > More progress. Life is good. Must drink microbrews in honor of > progress and Drew's birthday... A glass of Cyber-Champagne on Drew's birthday :-) Ralf From Stoned.Elipot@univ-evry.fr Mon Feb 19 12:32:03 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id MAA17102 for ; Mon, 19 Feb 1996 12:32:02 +0100 (MET) Received: from amiante.univ-evry.fr ([192.93.214.6]) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA20284; Mon, 19 Feb 1996 12:31:51 +0100 (MET) Received: from bandsept.univ-evry.fr (bandsept.univ-evry.fr [192.93.214.19]) by amiante.univ-evry.fr with ESMTP (8.6.12/94.01.10); id LAA18930; Mon, 19 Feb 1996 11:44:19 +0100 Received: from localhost.univ-evry.fr (localhost.univ-evry.fr [127.0.0.1]) by bandsept.univ-evry.fr with SMTP (8.6.12/94.01.26); id MAA06842; Mon, 19 Feb 1996 12:30:35 +0100 Message-Id: <199602191130.MAA06842@bandsept.univ-evry.fr> To: linux-mips@fnet.fr Subject: Re: Slower BogoMIPS rating on 1.3.62 than on 1.2.10 :-) In-Reply-To: Your message of "Sat, 17 Feb 1996 18:27:46 MST." <199602180127.SAA16269@rover.village.org> X-It: bandsept (wrong OS, but right CPU) X-Mailer: MH 6.8.3, MH-E 5.0.2 Date: Mon, 19 Feb 1996 12:30:29 +0100 From: Stoned Elipot Content-Length: 297 Lines: 12 Hi all, >>>>> "Warner" == Warner Losh writes: [SNIP] Warner> I now get to the VFS: Insert root floppy and press ENTER Warner> stage. However, the read never seems to complete. Sound Warner> familiar to anybody? Yep, it seems I got the same problem with the Oily. Stoned. From imp@rover.village.org Mon Feb 19 16:10:18 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id QAA20360 for ; Mon, 19 Feb 1996 16:10:17 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA24832; Mon, 19 Feb 1996 16:10:03 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id IAA20015 for ; Mon, 19 Feb 1996 08:10:04 -0700 Message-Id: <199602191510.IAA20015@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Slower BogoMIPS rating on 1.3.62 than on 1.2.10 :-) In-Reply-To: Your message of Mon, 19 Feb 1996 12:30:29 +0100 Date: Mon, 19 Feb 1996 08:10:04 -0700 From: Warner Losh Content-Length: 275 Lines: 9 : Yep, it seems I got the same problem with the Oily. So the floppy drive light comes on, but the read never seems to complete. I can't hear any track to track movement going on like it was reading in the whole disk. Had anybody been able to read off of floppies? Warner From cayman!uupsi!ksala!silver!davidk%harvard.harvard.edu%cicerone.uunet.ca Mon Feb 19 16:17:41 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id QAA20462 for ; Mon, 19 Feb 1996 16:17:40 +0100 (MET) Received: from cicerone.uunet.ca by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA24965; Mon, 19 Feb 1996 16:17:29 +0100 (MET) Received: from sni.ca ([192.75.236.203]) by cicerone.uunet.ca with SMTP id <189578-1>; Mon, 19 Feb 1996 10:17:31 -0500 Received: from silver.sni.ca by sni.ca (5.61/smail2.5/07-11-92) id AA27151; Mon, 19 Feb 96 10:25:29 -0500 Received: by silver.sni.CA (5.61/smail2.5/02-09-92) id AA26363; Mon, 19 Feb 96 10:25:23 -0500 From: "David Kerry" Message-Id: <9602191525.AA26363@silver.sni.CA> Subject: Re: Slower BogoMIPS rating on 1.3.62 than on 1.2.10 :-) To: linux-mips@fnet.fr Date: Mon, 19 Feb 1996 10:25:23 -0500 In-Reply-To: <199602191510.IAA20015@rover.village.org> from "Warner Losh" at Feb 19, 96 08:10:04 am X-Mailer: ELM [version 2.4 PL24] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT Content-Length: 720 Lines: 23 > > : Yep, it seems I got the same problem with the Oily. > > So the floppy drive light comes on, but the read never seems to > complete. I can't hear any track to track movement going on like it > was reading in the whole disk. > > Had anybody been able to read off of floppies? > > Warner > No luck here either, although it does seem to get something off of it before hanging - I get ext2_fs errors, but only sometimes. On another note, how long is the average boot via floppy supposed to take using milo 0.26? On my magnum, it takes ~5-6 minutes to read the kernel image. I have an older milo somewhere that used to boot it at full floppy speed (ie: ~30 seconds or less). -- David Kerry (davidk@sni.ca) From linux@mailhost.uni-koblenz.de Mon Feb 19 17:22:24 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id RAA21398 for ; Mon, 19 Feb 1996 17:22:23 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA26274; Mon, 19 Feb 1996 17:22:09 +0100 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with ESMTP id RAA03598 for ; Mon, 19 Feb 1996 17:20:01 +0100 From: Systemkennung Linux Message-Id: <199602191620.RAA03598@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id RAA25441; Mon, 19 Feb 1996 17:20:33 +0100 Subject: Re: Slower BogoMIPS rating on 1.3.62 than on 1.2.10 :-) To: linux-mips@fnet.fr Date: Mon, 19 Feb 1996 17:20:33 +0100 (MET) In-Reply-To: <199602191510.IAA20015@rover.village.org> from "Warner Losh" at Feb 19, 96 08:10:04 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 389 Lines: 14 Hi, > : Yep, it seems I got the same problem with the Oily. > > So the floppy drive light comes on, but the read never seems to > complete. I can't hear any track to track movement going on like it > was reading in the whole disk. > > Had anybody been able to read off of floppies? Reading works perfectly for me; there is a problem that sometimes corrupts disks on writing. Ralf From linux@mailhost.uni-koblenz.de Mon Feb 19 17:31:55 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id RAA21538 for ; Mon, 19 Feb 1996 17:31:55 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA26479; Mon, 19 Feb 1996 17:31:33 +0100 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with ESMTP id RAA03985 for ; Mon, 19 Feb 1996 17:30:09 +0100 From: Systemkennung Linux Message-Id: <199602191630.RAA03985@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id RAA25462; Mon, 19 Feb 1996 17:30:41 +0100 Subject: Re: Slower BogoMIPS rating on 1.3.62 than on 1.2.10 :-) To: linux-mips@fnet.fr Date: Mon, 19 Feb 1996 17:30:41 +0100 (MET) In-Reply-To: <9602191525.AA26363@silver.sni.CA> from "David Kerry" at Feb 19, 96 10:25:23 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 644 Lines: 19 Hi, > No luck here either, although it does seem to get something > off of it before hanging - I get ext2_fs errors, but only sometimes. > > On another note, how long is the average boot via floppy > supposed to take using milo 0.26? On my magnum, it takes > ~5-6 minutes to read the kernel image. I have an older milo > somewhere that used to boot it at full floppy speed (ie: ~30 seconds > or less). Milo contains some workarounds for Magnum specific ARC BIOS bug that make it crawling. Andy once told me had'd have speed up that thing so I currently don't see the reason why Milo 0.26 is crawling on your machine. Andy??? Ralf From lm@neteng.engr.sgi.com Mon Feb 19 17:39:56 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id RAA21656 for ; Mon, 19 Feb 1996 17:39:55 +0100 (MET) Received: from sgi.sgi.com (SGI.COM) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA26632; Mon, 19 Feb 1996 17:39:40 +0100 (MET) Received: from neteng.engr.sgi.com by sgi.sgi.com via ESMTP (950405.SGI.8.6.12/910110.SGI) for <@sgi.engr.sgi.com:linux-mips@fnet.fr> id IAA28867; Mon, 19 Feb 1996 08:39:44 -0800 Received: from localhost by neteng.engr.sgi.com via SMTP (950413.SGI.8.6.12/940406.SGI.AUTO) for id IAA11243; Mon, 19 Feb 1996 08:39:42 -0800 Message-Id: <199602191639.IAA11243@neteng.engr.sgi.com> To: linux-mips@fnet.fr From: lm@neteng.engr.sgi.com (Larry McVoy) Subject: Re: Slower BogoMIPS rating on 1.3.62 than on 1.2.10 :-) Date: Mon, 19 Feb 1996 08:39:42 -0800 Sender: lm@neteng.engr.sgi.com Content-Length: 59 Lines: 1 Whoever is sending that encapsulated mail, please stop it. From imp@rover.village.org Mon Feb 19 22:53:13 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id WAA25823 for ; Mon, 19 Feb 1996 22:53:12 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA01067; Mon, 19 Feb 1996 22:52:58 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id OAA20807 for ; Mon, 19 Feb 1996 14:53:01 -0700 Message-Id: <199602192153.OAA20807@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Slower BogoMIPS rating on 1.3.62 than on 1.2.10 :-) In-Reply-To: Your message of Mon, 19 Feb 1996 17:30:41 +0100 Date: Mon, 19 Feb 1996 14:53:01 -0700 From: Warner Losh Content-Length: 1191 Lines: 24 : Milo contains some workarounds for Magnum specific ARC BIOS bug that : make it crawling. Andy once told me had'd have speed up that thing : so I currently don't see the reason why Milo 0.26 is crawling on your : machine. : : Andy??? I'm not Andy.... However, he went in and hacked it so that the entire file was read in, and then things copied down into locore. He did this by requesting large (maybe even just one) reads to happen. At the time he said that it should reduce the amount of time spend reading the kernel by an order of magnitude on stupid, brain-dead floppy drives. The reason I recall this is that I had some patches that conflicted with his patches. They were on the disk that died, otherwise I'd try to merge the fastboot patches back into the milo. On my machine, to give a datapoint, it takes about 30-45 seconds for the machine to come up from poweron or reset. It then takes about 5-7 seconds to load milo, and about 30 to load vmlinux (which is about 650k right now for my config). I vaguely recall that Andy's fixes made this happen in < 15 seconds, but I'm not positive, nor do I have a copy of them any more :-(. I really hate flakey disks... Warner From imp@rover.village.org Mon Feb 19 22:56:39 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id WAA25894 for ; Mon, 19 Feb 1996 22:56:38 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA01109; Mon, 19 Feb 1996 22:56:25 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id OAA20837 for ; Mon, 19 Feb 1996 14:56:29 -0700 Message-Id: <199602192156.OAA20837@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Slower BogoMIPS rating on 1.3.62 than on 1.2.10 :-) In-Reply-To: Your message of Mon, 19 Feb 1996 17:20:33 +0100 Date: Mon, 19 Feb 1996 14:56:29 -0700 From: Warner Losh Content-Length: 469 Lines: 12 : Reading works perfectly for me; there is a problem that sometimes corrupts : disks on writing. OK... I'll poke into the floppy driver a bit and see what the deal with it is then. Just for yucks, can you try to boot a kernel that is expecting a root on floppy that only has ext2fs compiled in, but the floppy that you feed it is fat? I just realized that is what I'm doing, but it should at least give me an error message about bad magic numbers and such. Warner From imp@rover.village.org Mon Feb 19 23:00:43 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id XAA26175 for ; Mon, 19 Feb 1996 23:00:42 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA01228; Mon, 19 Feb 1996 23:00:30 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id PAA20889 for ; Mon, 19 Feb 1996 15:00:31 -0700 Message-Id: <199602192200.PAA20889@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Slower BogoMIPS rating on 1.3.62 than on 1.2.10 :-) In-Reply-To: Your message of Mon, 19 Feb 1996 12:03:05 +0100 Date: Mon, 19 Feb 1996 15:00:31 -0700 From: Warner Losh Content-Length: 557 Lines: 11 : How about: load the kernel to 0x80100000, copy the exception handlers down : to KSEG0, declare KSEG0+sizeof(lowmemstuff) as memory base address and all : the kernel code+data+bss to be memory in use. Sounds easier than hacking : complex things for the loader just to cut the kernel into two pieces. OK. That's what I had in mind. I was going to copy the stuff down in head.S, since it isn't much to move and we want the interrupt handlers in shortly. I may need help on the latter part, but I think I know who to ask about it locally... :-). Warner From paul@suite.sw.oz.au Mon Feb 19 23:17:52 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id XAA26394 for ; Mon, 19 Feb 1996 23:17:52 +0100 (MET) Received: from staff.cs.su.OZ.AU by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA01392; Mon, 19 Feb 1996 23:17:36 +0100 (MET) Received: from suite.sw.oz.au by staff.cs.su.OZ.AU (mail from paul for linux-mips@fnet.fr) with MHSnet (insertion MHSnet site: swallow.sw.oz.au); Tue, 20 Feb 1996 09:17:33 +1100 Received: from suite.sw.oz.au by swallow.sw.oz.au with SMTP id WAA22391; Mon, 19 Feb 1996 22:17:19 GMT (8.6.10/Unixware) (from paul@suite.sw.oz.au for ) Received: by suite.sw.oz.au id AA03315; Tue, 20 Feb 1996 09:17:08 +1100 (5.65c/1.34) (from paul@suite.sw.oz.au for linux-mips@fnet.fr) From: paul@suite.sw.oz.au (Paul Antoine) Message-Id: <199602192217.AA03315@suite.sw.oz.au> Subject: Re: Slower BogoMIPS rating on 1.3.62 than on 1.2.10 :-) To: linux-mips@fnet.fr Date: Tue, 20 Feb 1996 09:17:08 +1100 (EST) In-Reply-To: <199602192200.PAA20889@rover.village.org> from "Warner Losh" at Feb 19, 96 03:00:31 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: 1183 Lines: 23 > : How about: load the kernel to 0x80100000, copy the exception handlers down > : to KSEG0, declare KSEG0+sizeof(lowmemstuff) as memory base address and all > : the kernel code+data+bss to be memory in use. Sounds easier than hacking > : complex things for the loader just to cut the kernel into two pieces. > > OK. That's what I had in mind. I was going to copy the stuff down in > head.S, since it isn't much to move and we want the interrupt handlers > in shortly. I know this is not difficult code, but if you want some code for this, look in decstation.S... you may also want to create a deskstation.S and use a link script as I have done for the DEC, rather than have this kind of machine-specific code appear in head.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 2 698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From mtidwell@netcom.com Tue Feb 20 01:31:40 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id BAA28273 for ; Tue, 20 Feb 1996 01:31:39 +0100 (MET) From: mtidwell@netcom.com Received: from netcom.netcom.com by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA03417; Tue, 20 Feb 1996 01:31:28 +0100 (MET) Received: by netcom.netcom.com (8.6.12/Netcom) id PAA16368; Mon, 19 Feb 1996 15:26:21 -0800 Date: Mon, 19 Feb 1996 15:26:21 -0800 Message-Id: <199602192326.PAA16368@netcom.netcom.com> To: linux-mips@fnet.fr X-Url: mailto:linux-mips@fnet.fr X-Mailer: Lynx, Version 2-4-2 X-Personal_Name: montgomery farrar tidwell Subject: mailto:linux-mips@fnet.fr Content-Length: 287 Lines: 12 Howdy, i have an RC3230 that i REALLY want to put Linux on! Please tell me how i can help in porting Linux to the RC3230! i have 16yrs programming experience, 10yrs C, 5 yrs Unix. i don't claim to be a C or Unix god, but i'm good and highly motivated. \\//_ From imp@rover.village.org Tue Feb 20 02:21:51 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id CAA00506 for ; Tue, 20 Feb 1996 02:21:49 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA03943; Tue, 20 Feb 1996 02:21:35 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id SAA21444 for ; Mon, 19 Feb 1996 18:21:40 -0700 Message-Id: <199602200121.SAA21444@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Slower BogoMIPS rating on 1.3.62 than on 1.2.10 :-) In-Reply-To: Your message of Tue, 20 Feb 1996 09:17:08 +1100 Date: Mon, 19 Feb 1996 18:21:40 -0700 From: Warner Losh Content-Length: 1082 Lines: 21 : I know this is not difficult code, but if you want some code for this, : look in decstation.S... you may also want to create a deskstation.S : and use a link script as I have done for the DEC, rather than have : this kind of machine-specific code appear in head.S I rather hoped that all ARC BIOS machines would copy down the interrupt hanlders and load at KSEG0 + 1M. I really don't want to create a link script, but I'll take a look at the stuff to make sure that this is what I really want. :-) I certainly will do it in such a way that it won't break the DECstation. A stated goal early on was to have one kernel that could boot on all of the supported machines. I don't know if we can hit that goal with all of the machines, but I rather hoped the ARC BIOS based machines would be cool with that. I'd rather not branch off the deskstation to that large degree, since it is bascially the same as the other machines once a few sillinesses have been eliminated. And yes, the current code needs to be cleaned up by me so that some of the sillier ifdefs go away... Warner From imp@rover.village.org Tue Feb 20 03:01:46 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id DAA01116 for ; Tue, 20 Feb 1996 03:01:46 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA04315; Tue, 20 Feb 1996 03:01:28 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id TAA21828 for ; Mon, 19 Feb 1996 19:01:32 -0700 Message-Id: <199602200201.TAA21828@rover.village.org> To: Linux/MIPS Mailing List Subject: Floppies Date: Mon, 19 Feb 1996 19:01:32 -0700 From: Warner Losh Content-Length: 909 Lines: 34 OK. I feel silly. One line change and the floppy read completes (I get a kernel panic, but that is because I don't have fat compiled into the kernel). Time to create a ext2fs disk with /bin/sh on it... Here's the diff. It won't help people that don't have TYNE or RPC44 machines. After reading it, I think you'll understand why I feel silly. Warner Index: tyne-c.c =================================================================== RCS file: /home/imp/Repository/linux/arch/mips/kernel/tyne-c.c,v retrieving revision 1.2 diff -c -r1.2 tyne-c.c *** tyne-c.c 1996/02/09 07:14:59 1.2 --- tyne-c.c 1996/02/20 01:49:59 *************** *** 32,38 **** #include /* KLUDGE TO GET IT TO COMPILE */ ! #define FLOPPY_DMA 0 #define FLOPPY_IRQ 5 /* --- 32,38 ---- #include /* KLUDGE TO GET IT TO COMPILE */ ! #define FLOPPY_DMA 2 #define FLOPPY_IRQ 5 /* From imp@rover.village.org Tue Feb 20 05:25:32 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id FAA03004 for ; Tue, 20 Feb 1996 05:25:32 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA05582; Tue, 20 Feb 1996 05:25:15 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id VAA22054 for ; Mon, 19 Feb 1996 21:25:20 -0700 Message-Id: <199602200425.VAA22054@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Floppies In-Reply-To: Your message of Mon, 19 Feb 1996 19:01:32 MST Date: Mon, 19 Feb 1996 21:25:19 -0700 From: Warner Losh Content-Length: 66 Lines: 5 : #define FLOPPY_IRQ 5 I know this should be 6, not 5. Warner From linux@mailhost.uni-koblenz.de Tue Feb 20 09:59:37 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id JAA06686 for ; Tue, 20 Feb 1996 09:59:37 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA09337; Tue, 20 Feb 1996 09:59:23 +0100 (MET) Received: from flake.uni-koblenz.de (linux@flake.uni-koblenz.de [141.26.4.37]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with SMTP id JAA18445 for ; Tue, 20 Feb 1996 09:58:06 +0100 From: Systemkennung Linux Message-Id: <199602200858.JAA18445@informatik.uni-koblenz.de> Received: by flake.uni-koblenz.de (4.1/KO-2.0) id AA00651; Tue, 20 Feb 96 09:58:38 +0100 Subject: Re: Slower BogoMIPS rating on 1.3.62 than on 1.2.10 :-) To: linux-mips@fnet.fr Date: Tue, 20 Feb 1996 09:58:37 +0100 (MET) In-Reply-To: <199602192156.OAA20837@rover.village.org> from "Warner Losh" at Feb 19, 96 02:56:29 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 341 Lines: 11 Hi, > Just for yucks, can you try to boot a kernel that is expecting a root > on floppy that only has ext2fs compiled in, but the floppy that you > feed it is fat? I just realized that is what I'm doing, but it should > at least give me an error message about bad magic numbers and such. PANIC: Cannot mount root filesystem ... Ralf From linux@mailhost.uni-koblenz.de Tue Feb 20 10:15:38 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id KAA07098 for ; Tue, 20 Feb 1996 10:15:37 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA09610; Tue, 20 Feb 1996 10:15:24 +0100 (MET) Received: from flake.uni-koblenz.de (linux@flake.uni-koblenz.de [141.26.4.37]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with SMTP id KAA19188 for ; Tue, 20 Feb 1996 10:14:06 +0100 From: Systemkennung Linux Message-Id: <199602200914.KAA19188@informatik.uni-koblenz.de> Received: by flake.uni-koblenz.de (4.1/KO-2.0) id AA00768; Tue, 20 Feb 96 10:14:38 +0100 Subject: Re: Slower BogoMIPS rating on 1.3.62 than on 1.2.10 :-) To: linux-mips@fnet.fr Date: Tue, 20 Feb 1996 10:14:37 +0100 (MET) In-Reply-To: <199602200121.SAA21444@rover.village.org> from "Warner Losh" at Feb 19, 96 06:21:40 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1824 Lines: 35 Hi, > : I know this is not difficult code, but if you want some code for this, > : look in decstation.S... you may also want to create a deskstation.S > : and use a link script as I have done for the DEC, rather than have > : this kind of machine-specific code appear in head.S > > I rather hoped that all ARC BIOS machines would copy down the > interrupt hanlders and load at KSEG0 + 1M. I really don't want to > create a link script, but I'll take a look at the stuff to make sure > that this is what I really want. :-) I certainly will do it in such a > way that it won't break the DECstation. > > A stated goal early on was to have one kernel that could boot on all > of the supported machines. I don't know if we can hit that goal with > all of the machines, but I rather hoped the ARC BIOS based machines > would be cool with that. I'd rather not branch off the deskstation to > that large degree, since it is bascially the same as the other > machines once a few sillinesses have been eliminated. And yes, the > current code needs to be cleaned up by me so that some of the sillier > ifdefs go away... At some point we might have to code a more flexible loading system anyway. ARC systems only guarantee to have 4k (or was it even less) of memory at physical address 0x0; the rest of their memory might as well be at some different address. When we want to have a universal kernel for all R4000 machines we'll have to write a real ELF loader anyway. I'm currently tweaking the R4000 stuff from mips2 to mips3/mips4 to build a new execution model for user mode stuff. R3000 binaries (all the currently distributed user stuff) will run unchanged but using 64 bit CPU features will be possible; some disabled code in the kernel can be reenabled and other code can run again with interrupts enabled. Ralf From imp@rover.village.org Tue Feb 20 19:13:08 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id TAA22165 for ; Tue, 20 Feb 1996 19:13:06 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA18284; Tue, 20 Feb 1996 19:12:42 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id LAA24173 for ; Tue, 20 Feb 1996 11:12:45 -0700 Message-Id: <199602201812.LAA24173@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Slower BogoMIPS rating on 1.3.62 than on 1.2.10 :-) In-Reply-To: Your message of Tue, 20 Feb 1996 10:14:37 +0100 Date: Tue, 20 Feb 1996 11:12:45 -0700 From: Warner Losh Content-Length: 2203 Lines: 43 : At some point we might have to code a more flexible loading system anyway. : ARC systems only guarantee to have 4k (or was it even less) of memory at : physical address 0x0; the rest of their memory might as well be at some : different address. When we want to have a universal kernel for all R4000 : machines we'll have to write a real ELF loader anyway. I'm not sure that I follow this. If the memory on these machines is scattered all over the place, then how will ELF help? I know that ELF cna have segments and the like, and that those segments can be loaded aat different locations in memory. What I don't understand is how the loader will know enough to load the right things in the right locations. It seems to me that MILO would need to know what to load and where and that the "what" and "where" are more complicated than today. I'm looking at, btw, using some NetBSD tools called elf2ecoff and elf2aout that will hopefully enable us to have one compiler (the ELF one) and then use these tools to post process the results for MILO and maybe the kernel. This should save a lot of baggage that would otherwise be required to build the whole system. : I'm currently tweaking the R4000 stuff from mips2 to mips3/mips4 to : build a new execution model for user mode stuff. R3000 binaries : (all the currently distributed user stuff) will run unchanged but : using 64 bit CPU features will be possible; some disabled code in : the kernel can be reenabled and other code can run again with : interrupts enabled. So does this mean I shouldn't try to do the kernel copy stuff or that I should. I'm confused. I guess I'm back to not doing this and trying to build a root disk and configure more of the kernel (like the SCSI code and BUSTEC drivers). BTW, in trying to build a root disk, I can't help but notice that we have no shared libc. What's the status on this for Linux/MIPS? From the OpenBSD list (or was that the pmax port list), I know that GNU libc doesn't support ld.so on the mips and that was the biggest block to getting shared libraries there. A minimal root disk has to be something like /bin/sh, /sbin/mount*, and tar or cp to get the OS onto the hard disk. Warner From linux@mailhost.uni-koblenz.de Wed Feb 21 16:18:48 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id QAA16331 for ; Wed, 21 Feb 1996 16:18:46 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA05733; Wed, 21 Feb 1996 16:18:28 +0100 (MET) Received: from flake.uni-koblenz.de (linux@flake.uni-koblenz.de [141.26.4.37]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with SMTP id QAA11518 for ; Wed, 21 Feb 1996 16:16:46 +0100 From: Systemkennung Linux Message-Id: <199602211516.QAA11518@informatik.uni-koblenz.de> Received: by flake.uni-koblenz.de (4.1/KO-2.0) id AA07270; Wed, 21 Feb 96 16:17:09 +0100 Subject: Re: Slower BogoMIPS rating on 1.3.62 than on 1.2.10 :-) To: linux-mips@fnet.fr Date: Wed, 21 Feb 1996 16:17:08 +0100 (MET) In-Reply-To: <199602201812.LAA24173@rover.village.org> from "Warner Losh" at Feb 20, 96 11:12:45 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 590 Lines: 14 Hi, > BTW, in trying to build a root disk, I can't help but notice that we > have no shared libc. What's the status on this for Linux/MIPS? From > the OpenBSD list (or was that the pmax port list), I know that GNU > libc doesn't support ld.so on the mips and that was the biggest block > to getting shared libraries there. ld.so is a minor problem compared to the linker; someone from Japan is working on this problem since quite some time. His stuff is however still not usable to build real shared library images. people with IRIX acess might try to build shared objects ... Ralf From abhinav@teil.soft.net Wed Feb 21 04:11:12 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id EAA26227 for ; Wed, 21 Feb 1996 04:11:10 +0100 (MET) Received: from teil.soft.net ([164.164.10.2]) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA24822; Wed, 21 Feb 1996 04:10:51 +0100 (MET) Received: by teil.soft.net (920330.SGI/920502.SGI.JF) for linux-mips@fnet.fr id AA14793; Wed, 21 Feb 96 08:35:32 -0800 Date: Wed, 21 Feb 1996 08:22:08 -0800 (PST) From: Abhinav Jawadekar Subject: A few questions about Linux MIPS To: linux-mips@fnet.fr Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 1351 Lines: 27 We are looking at the possibility of running Linux on cards based on MIPS R3000/R4000/R8000/R10000 processors, designed by our organisation. Since these cards have their own architecture, could you please tell us about the steps we will be required to take in order to do this? We are a full-fledged h/w-s/w developement organisation and hence we just want to know the complexity of the problem and the direction to proceed. Thank you very much. Abhinav Jawadekar / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / Abhinav M. Jawadekar, / / Sr. Engineer R&D. / / / / Office: / / Tata Elxsi (India) Ltd., 123, Richmond Road, Bangalore 560025, India. / / Phone: 91-80-8452014, 8452015, 8452016, 8452185. Fax: 91-80-8452019. / / / / Residence: / / 151, Udani Layout, Cambridge Road, Ulsoor, Bangalore 560008, India. / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / From imp@rover.village.org Thu Feb 22 07:58:08 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id HAA00258 for ; Thu, 22 Feb 1996 07:58:07 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA19059; Thu, 22 Feb 1996 07:57:50 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id XAA05620 for ; Wed, 21 Feb 1996 23:57:56 -0700 Message-Id: <199602220657.XAA05620@rover.village.org> To: linux-mips@fnet.fr Subject: MIPS machines, a longish list Date: Wed, 21 Feb 1996 23:57:56 -0700 From: Warner Losh Content-Length: 723 Lines: 27 Here's something I found surfing through altavista. It is a list of all machines that had been tested and ran NT, version unknown. A few more than MILO groks :-) Warner ACER ARC1 AcerFormula Carerra R4000™ DESKStation Technology Evolution E4400 RISC PC DESKStation Technology Evolution R4000 RISC PC DESKStation Technology Tyne v4633x RISC PC MIPS ArcSystem Magnum PC-50 MIPS Millenium PC-50 NEC Image RISCstation NeTpower FASTseries MP Server 200MHz NeTpower FASTseries SMP Server 200MHz NeTpower NeTstation 100 NeTpower NeTstation 200 NeTpower NeTstation 300 NeTpower Server 1000 Olivetti PWS4000 Progen R4400-50/100 Shuttle RiscPC 4475 Siemens-Nixdorf RM 200 16 Siemens-Nixdorf RM 400 16 UniMicro RISCStation 4400NT From imp@rover.village.org Fri Feb 23 18:39:46 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id SAA00398 for ; Fri, 23 Feb 1996 18:39:45 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA19770; Fri, 23 Feb 1996 18:39:23 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id KAA12510 for ; Fri, 23 Feb 1996 10:39:19 -0700 Message-Id: <199602231739.KAA12510@rover.village.org> To: linux-mips@fnet.fr Subject: TLB entries Date: Fri, 23 Feb 1996 10:39:19 -0700 From: Warner Losh Content-Length: 345 Lines: 12 Quick question on TLB entries. When loading EntryHi and EntryLo[01], are the PFN numbers shifted to reflect the size of the page mask register? That is, would the PFN be different if I was creating a 4K page or a 16M page with the same base address? >From my readings of the MUM, I think the answer is "yes." but wanted to make sure. Warner From herrod@ncube.stanford.edu Fri Feb 23 18:56:25 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id SAA00770 for ; Fri, 23 Feb 1996 18:56:20 +0100 (MET) Received: from ncube (ncube.Stanford.EDU) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA20007; Fri, 23 Feb 1996 18:56:04 +0100 (MET) Received: (from herrod@localhost) by ncube (940816.SGI.8.6.9/8.6.6) id JAA10980; Fri, 23 Feb 1996 09:56:13 -0800 Date: Fri, 23 Feb 1996 09:56:13 -0800 Message-Id: <199602231756.JAA10980@ncube> From: Steve Herrod X-Url: http://www-flash.stanford.edu/~herrod To: linux-mips@fnet.fr Subject: linux mips mailing list Content-Length: 340 Lines: 11 I am interested in working on the linux/mips port and would like to be added to any developer's mailing list that exists. Thanks, Steve Herrod ========================================================================== Steve Herrod Stanford University herrod@cs.stanford.edu (415)725-1776 http://www-flash.stanford.edu/~herrod/ From imp@rover.village.org Sat Feb 24 01:54:45 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id BAA06636 for ; Sat, 24 Feb 1996 01:54:45 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA24858; Sat, 24 Feb 1996 01:54:12 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id RAA13991 for ; Fri, 23 Feb 1996 17:54:11 -0700 Message-Id: <199602240054.RAA13991@rover.village.org> To: linux-mips@fnet.fr Subject: Re: TLB entries In-Reply-To: Your message of Fri, 23 Feb 1996 10:39:19 MST Date: Fri, 23 Feb 1996 17:54:11 -0700 From: Warner Losh Content-Length: 299 Lines: 8 : When loading EntryHi and EntryLo[01], are the PFN numbers shifted to : reflect the size of the page mask register? No. I read the MUM wrong. Good reference is milo's lib/dumptlb.c routine. A must read if you are writing TLB entries in head.S. Much clearer than the MUM on this point. Wanrer From linux@mailhost.uni-koblenz.de Sat Feb 24 03:47:37 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id DAA09868 for ; Sat, 24 Feb 1996 03:47:36 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA25743; Sat, 24 Feb 1996 03:47:18 +0100 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with ESMTP id DAA12382 for ; Sat, 24 Feb 1996 03:46:30 +0100 From: Systemkennung Linux Message-Id: <199602240246.DAA12382@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id DAA00085; Sat, 24 Feb 1996 03:46:28 +0100 Subject: Re: TLB entries To: linux-mips@fnet.fr Date: Sat, 24 Feb 1996 03:46:27 +0100 (MET) In-Reply-To: <199602231739.KAA12510@rover.village.org> from "Warner Losh" at Feb 23, 96 10:39:19 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 836 Lines: 23 Hi, > Quick question on TLB entries. > > When loading EntryHi and EntryLo[01], are the PFN numbers shifted to > reflect the size of the page mask register? > > That is, would the PFN be different if I was creating a 4K page or a > 16M page with the same base address? > > >From my readings of the MUM, I think the answer is "yes." but wanted > to make sure. I'm not shure about how to understand your question; there is however one example for a working tlb entry != 4k already in the Linux code. Look in head.S for the map0 table for the Acer or Tyne; these map the 64k port space to the virtual address 0xe2000000. Hope this helps. I think I'm slowly getting the 64bit stuff under control. More things than I had originally planed need to be patched, eg vsprintf which assumed that long is the biggest type to print. Ralf From imp@rover.village.org Sat Feb 24 05:43:03 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id FAA11455 for ; Sat, 24 Feb 1996 05:43:02 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA27693; Sat, 24 Feb 1996 05:42:30 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id VAA14914 for ; Fri, 23 Feb 1996 21:42:30 -0700 Message-Id: <199602240442.VAA14914@rover.village.org> To: linux-mips@fnet.fr Subject: Re: TLB entries In-Reply-To: Your message of Sat, 24 Feb 1996 03:46:27 +0100 Date: Fri, 23 Feb 1996 21:42:30 -0700 From: Warner Losh Content-Length: 1041 Lines: 21 : I'm not shure about how to understand your question; there is however : one example for a working tlb entry != 4k already in the Linux code. : Look in head.S for the map0 table for the Acer or Tyne; these map : the 64k port space to the virtual address 0xe2000000. Hope this helps. My question was "Do I need to adjust the PFN for the TLB entry if I change the page size?" The answer is no. I took a look at those examples, but wasn't sure where they were supposed to be mapping to, so I couldn't sanity check my understanding of the TLB stuff against them. The milo dumptlb stuff, however, has enough info to help me sort it out. : I think I'm slowly getting the 64bit stuff under control. More things : than I had originally planed need to be patched, eg vsprintf which : assumed that long is the biggest type to print. This sounds cool. What is it that you are doing? It sounds like you are making it possible to run Linux/MIPS in 64 bit mode somehow and have 64 bit (and 32 bit, right) executables. Is that correct? Warner From triemer@babbitt.bernstein.com Mon Feb 12 03:33:27 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id DAA21980 for ; Mon, 12 Feb 1996 03:33:27 +0100 (MET) Received: from babbitt.bernstein.com by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA23080; Mon, 12 Feb 1996 03:33:17 +0100 (MET) Received: (from triemer@localhost) by babbitt.bernstein.com (8.6.12/8.6.11) id VAA01014; Sun, 11 Feb 1996 21:34:01 -0500 Date: Sun, 11 Feb 1996 21:34:01 -30000 From: Thomas Riemer Subject: Re: Linux 1.3.60 To: linux-mips@fnet.fr In-Reply-To: <199602120138.CAA25836@informatik.uni-koblenz.de> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 816 Lines: 32 I presume lena.fnet.fr is different from ftp.fnet.fr - that it is some sort of staging area before general release... -Tom On Mon, 12 Feb 1996, Systemkennung Linux wrote: > -----BEGIN PGP SIGNED MESSAGE----- > > Hi, > > I've uploaded linux-1.3.59-1.3.60.diffs.gz to lena.fnet.fr. It's almost > 3am so thistime a short announcement ... > > Ralf > > 34de6429aac68099785590f0f7156dc2 linux-1.3.59-1.3.60.diffs.gz > > -----BEGIN PGP SIGNATURE----- > Version: 2.6.2 > > iQCVAgUBMR6aQkckbl6vezDBAQH2TAP8DRWvTvMKXedY2DMv1CH4qjC4OZGllBCF > /+qIdISJie8t5lVKgxVBlR6goHdQVP7HIwt4vhrVTF2V6F4O6un3PvHbXbsLiT31 > XYIfbcTtRPv/IEEGP/8eAy2Uca5/EQloC42Y02EbX9muN/wHg+MrMlnI7Y5qd1Lr > hOc6ik7BVqI= > =8NsD > -----END PGP SIGNATURE----- > ---- Where theory and reality meet. ---- Thomas Riemer, triemer@wesleyan.edu From linux@mailhost.uni-koblenz.de Sat Feb 24 19:59:07 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id TAA05475 for ; Sat, 24 Feb 1996 19:58:54 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA03707; Sat, 24 Feb 1996 19:58:34 +0100 (MET) Received: from flake.uni-koblenz.de (linux@flake.uni-koblenz.de [141.26.4.37]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with SMTP id TAA06512 for ; Sat, 24 Feb 1996 19:57:46 +0100 From: Systemkennung Linux Message-Id: <199602241857.TAA06512@informatik.uni-koblenz.de> Received: by flake.uni-koblenz.de (4.1/KO-2.0) id AA20019; Sat, 24 Feb 96 19:57:45 +0100 Subject: Re: TLB entries To: linux-mips@fnet.fr Date: Sat, 24 Feb 1996 19:57:45 +0100 (MET) In-Reply-To: <199602240442.VAA14914@rover.village.org> from "Warner Losh" at Feb 23, 96 09:42:30 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 990 Lines: 22 Hi, > : I think I'm slowly getting the 64bit stuff under control. More things > : than I had originally planed need to be patched, eg vsprintf which > : assumed that long is the biggest type to print. > > This sounds cool. What is it that you are doing? It sounds like you > are making it possible to run Linux/MIPS in 64 bit mode somehow and > have 64 bit (and 32 bit, right) executables. Is that correct? Yes. The kernel/user mode stuff of 1.3.62 is running in the 32bit mode. This means that addresses outside of the 4G address range need to be mapped into the address range using one of the scarce tlb entries. Running in 64 bit mode this is still possible but no longer required - you can directly access the new addresses. Other advantages are: some code does no longer have to run with interrupts disabled; User code may take advantage of 64 bit instructions. This isn't a real 64 bit kernel yet which would mean that we'd also need to have a 64 bit address space. Ralf From afritz@starlink.com Mon Feb 26 06:56:28 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.3/8.7.3) with SMTP id GAA06303 for ; Mon, 26 Feb 1996 06:56:23 +0100 (MET) Received: from orion.starlink.com by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA23851; Mon, 26 Feb 1996 06:56:05 +0100 (MET) Received: from afritz.phx.primenet.com (pm1-10.starlink.com [206.26.42.234]) by orion.starlink.com (8.6.9/8.6.9) with SMTP id WAA08886 for ; Sun, 25 Feb 1996 22:47:16 -0700 Message-Id: <199602260547.WAA08886@orion.starlink.com> X-Sender: afritz@starlink.com X-Mailer: Windows Eudora Pro Version 2.1.2 Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Sun, 25 Feb 1996 22:41:43 -0700 To: linux-mips@fnet.fr From: Adam Fritzler Content-Length: 1099 Lines: 25 Dear Linux/MIPS Developers: I have an old DECstation 5000/20 (a.k.a. Personal DECstation) laying here and would like to run Linux on it...is the DECstation Linux/MIPS running/working yet? I was also wandering if Linux/MIPS would work if I was to purchase the R4000 upgrade daughterboard (from DEC) for it? Thanx, Adam Fritzler afritz@starlink.com PS: Does anyone know of a way to clear/bypass a DECstation's "console" password?? (I would have to do this to boot anything other than it's Ultrix 4.2a OS! :( And, would upgrading to the R4000 processor (and the associated ROM) make it ARC compliant? -- I really don't know much about RISC machines :) I would also be willing to do what ever I could to help your effort, although I know very little about C/C++ programming and probably couldn't be much help. -------------------------------------- - Adam Fritzler - - afritz@starlink.com - -------------------------------------- From paul@suite.sw.oz.au Tue Feb 27 00:04:36 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.4/8.7.3) with SMTP id AAA23891 for ; Tue, 27 Feb 1996 00:04:34 +0100 (MET) Received: from staff.cs.su.OZ.AU ([129.78.8.1]) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA14261; Tue, 27 Feb 1996 00:04:11 +0100 (MET) Received: from suite.sw.oz.au by swallow.sw.oz.au with SMTP id XAA10703; Mon, 26 Feb 1996 23:04:04 GMT (8.6.10/Unixware) (from paul@suite.sw.oz.au for ) Received: by suite.sw.oz.au id AA03561; Tue, 27 Feb 1996 09:53:01 +1100 (5.65c/1.34) (from paul@suite.sw.oz.au for linux-mips@fnet.fr) From: paul@suite.sw.oz.au (Paul Antoine) Message-Id: <199602262253.AA03561@suite.sw.oz.au> Subject: Re: linux on decstation To: A.Schroeter@dkfz-heidelberg.de (Andre Schroeter) Date: Tue, 27 Feb 1996 09:53:00 +1100 (EST) Cc: linux-mips@fnet.fr (Linux MIPS mailing list) In-Reply-To: <9602261601.ZM19955@borneo.inet.dkfz.heidelberg.de> from "Andre Schroeter" at Feb 26, 96 04:01:35 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: 2026 Lines: 52 Andre, > i have found some references to linux on mips processors and > decstations. please tellm me more about the exact model of > decstations that is usable for linux. The port to the DECStations is in its early stages: we can boot through the first few layers of kernel initialisation, but none of the interrupt code or device drivers are complete. > we have some labeled as 3100 5000 and 5000/150. We have people on the linux-mips mailing list that are contributing to the porting effort such that between us we have at least one of each of these different systems. It is hoped that we will support them all. > how can one find out more about the exact moddel ? It depends upon what you want to know. > is there a difference between a decstation and a personal decstation ? Yep - a personal decstation is smaller :-) Seriously though, for the models you mentioned: DECStation 3100 R2000 CPU @ 16Mhz, no turbochannel DECStation 5000/2x R3000 CPU @ 20/25 Mhz, 3xturbochannel slots DECStation 5000/1xx,2xx R3000 CPU, lots of turbochannel slots There are more specs, but support for the 5000 range is relatively simple, as they all share a common architecture. The 2100's & 3100's are a little simpler in their architectures, but use similar chipsets for peripherals etc. so we expect to be able to support them quite easily. Please consider joining the mailing list, particularly if you can help write the kernel code or test it. Otherwise, stay tuned to comp.os.linux.announce for news about the release of DECStation support later this year. 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 2 698 2322 Level 2, 79 Myrtle St, Chippendale, NSW 2008, Australia Fax: +61 2 699 9174 "It is the lack of acceptance of diversity which threatens to destroy society, NOT the free expression of it." - Me. From upi@gorba.lpt.fi Tue Feb 27 14:06:54 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.4/8.7.3) with SMTP id OAA06555 for ; Tue, 27 Feb 1996 14:06:53 +0100 (MET) Received: from gorba.lpt.fi by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA00254; Tue, 27 Feb 1996 14:06:33 +0100 (MET) Received: (from upi@localhost) by gorba.lpt.fi (8.6.12/8.6.9) id PAA02067; Tue, 27 Feb 1996 15:06:45 +0200 Date: Tue, 27 Feb 1996 15:06:42 +0200 (EET) From: Pasi Pirhonen To: linux-mips@fnet.fr Subject: Project - machines Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=ISO-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Content-Length: 587 Lines: 25 Hi, I am working on University of Helsinki, Lahti Research and Training Centre and i have currently one DS3100 just "eating power and running Ultrix" I'd like to contribute my freetime to DS-project=B4and do what i can. Basically i don't know anything about kernel-hacking, but i can help in state when it's possible to install and even run something. I can run cross-compiler on SunOS or Linux/386 if needed. So, what is next step ? BTW. I have meiled to vgers.rutgers.edu to subcribe linux-mips, but no answer yet... TIA -- Pasi.Pirhonen@helsinki.fi http://www.lpt.fi/~upi/ From linux@mailhost.uni-koblenz.de Tue Feb 27 15:27:44 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.4/8.7.3) with SMTP id PAA10029 for ; Tue, 27 Feb 1996 15:27:43 +0100 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA01701; Tue, 27 Feb 1996 15:25:10 +0100 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.6.9/8.6.9) with ESMTP id PAA12409 for ; Tue, 27 Feb 1996 15:23:17 +0100 From: Systemkennung Linux Message-Id: <199602271423.PAA12409@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id PAA03031; Tue, 27 Feb 1996 15:22:45 +0100 Subject: Re: Project - machines To: linux-mips@fnet.fr Date: Tue, 27 Feb 1996 15:22:44 +0100 (MET) In-Reply-To: from "Pasi Pirhonen" at Feb 27, 96 03:06:42 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1845 Lines: 46 Hi, > I am working on University of Helsinki, Lahti Research and Training Centre > and i have currently one DS3100 just "eating power and running Ultrix" > > I'd like to contribute my freetime to DS-project=B4and do what i can. > Basically i don't know anything about kernel-hacking, but i can help > in state when it's possible to install and even run something. Kernel hacking is just like other programming. You make a mistake - it crashes ... > I can run cross-compiler on SunOS or Linux/386 if needed. Good; these operating systems give nice crossdevelopment systems. > So, what is next step ? I'm the main developer of Linux/MIPS. Since I'm working on a Acer using the R4000 I can only help you with general thing about Linux, Linux/MIPS and the MIPS specific stuff. The R3000 and DECstation ports are being done by Paul Antoine (paul@suite.sw.oz.au). I suggest that you contact him about details of the R3000/DECstation. For everything else I'll be happy to answer your questions. Technically you should probably start to setup a crosscompiling environment. There is some documentation about this available on http://lena.fnet.fr/, but this doc isn't perfect yet. This step can be a bit nasty ... If you say me for OS you need crosscompiler binaries, I might be able to provide some for you. Linux binaries are already available on ftp.fnet.fr:/linux-mips/. > BTW. I have meiled to vgers.rutgers.edu to subcribe linux-mips, > but no answer yet... Yes, Vger is slow and not used by the MIPS developers. Instead the developers are using the mailing list you've just mailed to. Forgive the curious question to you - since some time I'm receiving much more mail from people interested in Linux/MIPS and I'm interested to find what made that massive increase. Was it Webpages that made you ask? Happy hacking, Ralf From imp@rover.village.org Thu Feb 29 07:05:48 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.4/8.7.3) with SMTP id HAA08312 for ; Thu, 29 Feb 1996 07:05:47 +0100 (MET) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA07479; Thu, 29 Feb 1996 07:05:21 +0100 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.6.11/8.6.6) with SMTP id XAA01495 for ; Wed, 28 Feb 1996 23:05:15 -0700 Message-Id: <199602290605.XAA01495@rover.village.org> To: Linux/MIPS Mailing List Subject: Floppy problems Date: Wed, 28 Feb 1996 23:05:15 -0700 From: Warner Losh Content-Length: 486 Lines: 13 Any idea how to track this one down... I added a checksum to the kernel to checksum the blocks that were coming in off the floppy. The first block checksums correctly, and comes from track 0, head 0, sector 3. The second block checksums to the exact same thing as the first one (this is track 2, head 1, sector 11 or the 101st (1 relative) secort) according to the interrupt handler, but should checksum to something different according to my simple checksum progam. Ideas? Warner