From hannon@Crissy.Stanford.EDU Thu May 2 10:42:27 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 KAA13809 for ; Thu, 2 May 1996 10:42:26 +0200 (MET DST) Received: from crissy.Stanford.EDU by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA09035; Thu, 2 May 1996 10:42:02 +0200 (MET) Received: by Crissy.Stanford.EDU (8.6.8/25-LAGUNITA-eef) id BAA03702; Thu, 2 May 1996 01:44:58 -0700 From: Steph Hannon Message-Id: <199605020844.BAA03702@Crissy.Stanford.EDU> Subject: Linux/MIPS compiling question To: linux-mips@fnet.fr Date: Thu, 2 May 96 1:44:58 PDT X-Mailer: ELM [version 2.3 PL11] Content-Length: 946 Lines: 25 Hi all. I am a little confused about the cross development tools available for Linux/MIPS. Given that it can be compiled into four different formats (a.out, ELF/ big, little endian) are there different versions of gcc/bin utilities necessary for different outputs? When I retrieved gcc/bin utilities from ftp://ftp.mcc.ac.uk/pub/linux/MIPS, I got a mips-linux-gcc executable. But I want to compile my code for little endian and it seems to be looking for a mipsel-linux-gcc executable. The FAQ says "The current version are gcc 2.7.2 and binutils 2.6. There are patches required to both of these packages in the "src" subdirectory in the above locations." I am not sure if the versions of gcc and binutils I pulled off that site are already patched or if I need to do some more work I am sure this is probably documented somewhere but I haven't found the place yet! Any help would be mighty appreciated! Thanks in advance, Stephanie From herrod@ncube.stanford.edu Thu May 2 10:46:31 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 KAA13865 for ; Thu, 2 May 1996 10:46:30 +0200 (MET DST) Received: from ncube (ncube.Stanford.EDU) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA09093; Thu, 2 May 1996 10:46:05 +0200 (MET) Received: (from herrod@localhost) by ncube (940816.SGI.8.6.9/8.6.6) id BAA23029; Thu, 2 May 1996 01:46:18 -0700 Date: Thu, 2 May 1996 01:46:18 -0700 Message-Id: <199605020846.BAA23029@ncube> From: Steve Herrod To: linux-mips@fnet.fr Subject: Re: Linux/MIPS compiling question References: <199605020844.BAA03702@Crissy.Stanford.EDU> Content-Length: 265 Lines: 8 Hey Steph, Sorry. I didn't realize that you were getting caught up with the compilation tools. Why don't we sit down together sometime and take a look at it. If you want to wait until after the meeting we can do that or we can meet earlier... your choice. -Steve From tor@spacetec.no Thu May 2 11:45:11 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 LAA14534 for ; Thu, 2 May 1996 11:45:10 +0200 (MET DST) Received: from pallas.spacetec.no by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA10226; Thu, 2 May 1996 11:44:48 +0200 (MET) Received: (from tor@localhost) by pallas.spacetec.no (8.7.1/8.7.1) id LAA26193 for linux-mips@fnet.fr; Thu, 2 May 1996 11:50:28 +0200 Message-Id: <199605020950.LAA26193@pallas.spacetec.no> From: tor@spacetec.no (Tor Arntsen) Date: Thu, 2 May 1996 11:50:27 +0200 X-Mailer: Mail User's Shell (7.2.5 10/14/92) To: linux-mips@fnet.fr Subject: m700-10 success Content-Length: 709 Lines: 14 Just a quick note to tell that the 1.3.96 kernel (+ Linus hint about adding to skbuff.h, fixed in .97) boots and runs fine on my Olivetti. I never got interrupts working on 1.3.62, so this is progress. (with 1.3.62 I could send, but not receive, with NE2000 as well as 3c503) I don't have any docs on the m700-10 except for the inst manual, so I'm in a difficult situtation for trying to fix such things myself. :-( My 3c503 ethernet board works fine now, in PIO mode. There's still something wrong with the shared mem detection, I'll come back to that when I get time again. The nfsmounts works fine, things look good, and the console is really bad, as Ralf indicated.. :-) Good work! Tor From Stoned.Elipot@univ-evry.fr Thu May 2 12:31:15 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 MAA15111 for ; Thu, 2 May 1996 12:31:15 +0200 (MET DST) Received: from amiante.univ-evry.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA11561; Thu, 2 May 1996 12:30:54 +0200 (MET) Received: from bandsept.univ-evry.fr (bandsept.univ-evry.fr [192.93.214.19]) by amiante.univ-evry.fr with ESMTP (8.6.12/96.04.09/louis-blanc); id MAA01322; Thu, 2 May 1996 12:31:11 +0200 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 MAA23449; Thu, 2 May 1996 12:30:29 +0200 Message-Id: <199605021030.MAA23449@bandsept.univ-evry.fr> To: linux-mips@fnet.fr Subject: Re: m700-10 success In-Reply-To: Your message of "Thu, 02 May 1996 11:50:27 +0200." <199605020950.LAA26193@pallas.spacetec.no> X-It: bandsept (wrong OS, but right CPU) X-Mailer: MH 6.8.3, MH-E 5.0.2 Date: Thu, 02 May 1996 12:30:23 +0200 From: Stoned Elipot Content-Length: 703 Lines: 19 Hi Tor, >>>>> "Tor" == Tor Arntsen writes: Tor> Just a quick note to tell that the 1.3.96 kernel (+ Linus hint Tor> about adding to skbuff.h, fixed in .97) boots and [SNIP] Congrats Tor, I wish my Oily run as well, could you please send me your kernel's .config file ? And I guess you're a perfect candidat for the M700 tech manual ? I'll had two copies from Olivetti France months ago, Andy got one. I'll try to xerox it and send it over you. And please could you tell me more about the corss-compiler env you use to compile binaries (which lib, which sources, or which binaries from Ralf...) as I wasn't able to make ash (home compilation) run. Thanks. Stoned. From tor@spacetec.no Thu May 2 13:23:40 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 NAA15802 for ; Thu, 2 May 1996 13:23:27 +0200 (MET DST) Received: from pallas.spacetec.no by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA13247; Thu, 2 May 1996 13:23:03 +0200 (MET) Received: (from tor@localhost) by pallas.spacetec.no (8.7.1/8.7.1) id NAA26284 for linux-mips@fnet.fr; Thu, 2 May 1996 13:28:49 +0200 Message-Id: <199605021128.NAA26284@pallas.spacetec.no> From: tor@spacetec.no (Tor Arntsen) Date: Thu, 2 May 1996 13:28:48 +0200 In-Reply-To: Stoned Elipot "Re: m700-10 success" (May 2, 11:30) X-Mailer: Mail User's Shell (7.2.5 10/14/92) To: linux-mips@fnet.fr Subject: Re: m700-10 success Content-Length: 1288 Lines: 41 On May 2, 11:30, Stoned Elipot wrote: >Congrats Tor, I wish my Oily run as well, could you please send me >your kernel's .config file ? Here it comes, through |grep =y to make it smaller: CONFIG_MIPS_MAGNUM_4000=y CONFIG_MIPS_JAZZ=y CONFIG_CPU_R4X00=y CONFIG_TLB_SHUTDOWN=y CONFIG_BINFMT_ELF=y CONFIG_CPU_LITTLE_ENDIAN=y CONFIG_NET=y CONFIG_BLK_DEV_FD=y CONFIG_INET=y CONFIG_SKB_LARGE=y CONFIG_NETDEVICES=y CONFIG_NET_VENDOR_3COM=y CONFIG_EL2=y I have a 3c503 (NB! a couple of tricks needed) CONFIG_NET_ISA=y CONFIG_NE2000=y and a NE2000 which I keep putting in and out. CONFIG_EXT2_FS=y CONFIG_PROC_FS=y CONFIG_NFS_FS=y CONFIG_ROOT_NFS=y CONFIG_RNFS_RARP=y >And I guess you're a perfect candidat for the M700 tech manual ? I'll >had two copies from Olivetti France months ago, Andy got one. I'll try >to xerox it and send it over you. Absolutely great! I'll get in touch with you later in private email when I get time (I'm working now, this is lunchtime.. the canteen closes in five minutes, argh..) >And please could you tell me more about the corss-compiler env you use >to compile binaries (which lib, which sources, or which binaries from >Ralf...) as I wasn't able to make ash (home compilation) run. Thanks. I'll come back to this as well, this evening I hope. Tor From linux@mailhost.uni-koblenz.de Thu May 2 14:37:14 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 OAA16663 for ; Thu, 2 May 1996 14:37:13 +0200 (MET DST) Received: from informatik.uni-koblenz.de ([141.26.4.1]) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA15279; Thu, 2 May 1996 14:28:50 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id OAA04454 for ; Thu, 2 May 1996 14:28:25 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605021228.OAA04454@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id OAA02028; Thu, 2 May 1996 14:26:35 +0200 Subject: Re: m700-10 success To: linux-mips@fnet.fr Date: Thu, 2 May 1996 14:26:34 +0200 (MET DST) In-Reply-To: <199605020950.LAA26193@pallas.spacetec.no> from "Tor Arntsen" at May 2, 96 11:50:27 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1527 Lines: 31 Hi, > Just a quick note to tell that the 1.3.96 kernel (+ Linus hint about adding > to skbuff.h, fixed in .97) boots and runs fine on my > Olivetti. I never got interrupts working on 1.3.62, so this is progress. > (with 1.3.62 I could send, but not receive, with NE2000 as well as 3c503) > I don't have any docs on the m700-10 except for the inst manual, so I'm in > a difficult situtation for trying to fix such things myself. :-( Until 1.2.63 or so I still had the EISA interrupts disabled just because this part of the kernel was *completly* untested on Magnum/M700. Probably this was your problem. Anyway, networking is still broken; the problem is a weired combination between a bug in my checksum code and erroneous code generated by GCC. > My 3c503 ethernet board works fine now, in PIO mode. There's still something > wrong with the shared mem detection, I'll come back to that when I get time > again. The nfsmounts works fine, things look good, and the console is really > bad, as Ralf indicated.. :-) Question: I had a M700 that shows a line of trash at the top of the screen after some time - even when just idleing around. Did you see this problem, too? What CPU version are you using? Milo prints this and you can do a cat /proc/cpuinfo to get this information. All that now is still missing is an updated version of libc which I've built yesterday for big (untested) and little endian. The versions that I've published can no longer be used with current kernel header files. Ralf From linux@mailhost.uni-koblenz.de Thu May 2 14:36:56 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 OAA16633 for ; Thu, 2 May 1996 14:36:56 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA15416; Thu, 2 May 1996 14:35:54 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id OAA04704 for ; Thu, 2 May 1996 14:35:14 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605021235.OAA04704@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id OAA02046; Thu, 2 May 1996 14:33:24 +0200 Subject: Re: m700-10 success To: linux-mips@fnet.fr Date: Thu, 2 May 1996 14:33:24 +0200 (MET DST) In-Reply-To: <199605021030.MAA23449@bandsept.univ-evry.fr> from "Stoned Elipot" at May 2, 96 12:30:23 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 932 Lines: 21 Hi, > And please could you tell me more about the corss-compiler env you use > to compile binaries (which lib, which sources, or which binaries from > Ralf...) as I wasn't able to make ash (home compilation) run. Thanks. Give me another day to give the new libc a little bit more testing. Things look good but I've added almost three month of libc development to my sources and adapted my libc port to the changed header files setup in the kernel so I want to try to compile something bigger than "Hello, world!" type test programs first. Again just as I've already written in an aside to Tor's mail - the current kernel header files cannot be used in combination with the yet published libc include files. I however can't see a reason why you can't run any version of sh on your machine. If you have a combination of kernel/libc header files that make it possible for you to *compile* something it should run also. Ralf From Stoned.Elipot@univ-evry.fr Thu May 2 14:54:31 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 OAA16860 for ; Thu, 2 May 1996 14:54:30 +0200 (MET DST) Received: from amiante.univ-evry.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA15855; Thu, 2 May 1996 14:54:10 +0200 (MET) Received: from bandsept.univ-evry.fr (bandsept.univ-evry.fr [192.93.214.19]) by amiante.univ-evry.fr with ESMTP (8.6.12/96.04.09/louis-blanc); id OAA01774; Thu, 2 May 1996 14:54:23 +0200 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 OAA23581; Thu, 2 May 1996 14:53:39 +0200 Message-Id: <199605021253.OAA23581@bandsept.univ-evry.fr> To: linux-mips@fnet.fr Subject: Re: m700-10 success In-Reply-To: Your message of "Thu, 02 May 1996 14:26:34 +0200." <199605021228.OAA04454@informatik.uni-koblenz.de> X-It: bandsept (wrong OS, but right CPU) X-Mailer: MH 6.8.3, MH-E 5.0.2 Date: Thu, 02 May 1996 14:53:30 +0200 From: Stoned Elipot Content-Length: 312 Lines: 8 >>>>> "Systemkennung" == Systemkennung Linux writes: Systemkennung> Question: I had a M700 that shows a line of trash at Systemkennung> the top of the screen after some time - even when just Systemkennung> idleing around. Did you see this problem, too? I got it too. Stoned. From imp@rover.village.org Thu May 2 16:06:03 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 QAA18317 for ; Thu, 2 May 1996 16:05:58 +0200 (MET DST) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA24191; Thu, 2 May 1996 16:05:33 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.7.5/8.6.6) with SMTP id IAA03744 for ; Thu, 2 May 1996 08:05:38 -0600 (MDT) Message-Id: <199605021405.IAA03744@rover.village.org> To: linux-mips@fnet.fr Subject: Re: m700-10 success In-Reply-To: Your message of Thu, 02 May 1996 14:26:34 +0200 Date: Thu, 02 May 1996 08:05:38 -0600 From: Warner Losh Content-Length: 650 Lines: 15 : > Olivetti. I never got interrupts working on 1.3.62, so this is progress. Hmmm, 1.3.62 was the last release that I was getting interrupts :-(. They certainly don't work in .96 for me. Time to go back and make sure I can rebuild 1.3.62. The disk problems I had recently hit my source base harder than I had thought so the diff process has become non-trivial. : All that now is still missing is an updated version of libc which I've built : yesterday for big (untested) and little endian. The versions that I've : published can no longer be used with current kernel header files. I'd love to see a new libc that I can build ash with. Warner From linux@mailhost.uni-koblenz.de Thu May 2 17:34:33 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 RAA19422 for ; Thu, 2 May 1996 17:34:32 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA17849; Thu, 2 May 1996 17:31:23 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id RAA12132 for ; Thu, 2 May 1996 17:31:05 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605021531.RAA12132@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id RAA02459; Thu, 2 May 1996 17:29:15 +0200 Subject: Magnum 4000 config disk To: linux-mips@fnet.fr Date: Thu, 2 May 1996 17:29:15 +0200 (MET DST) Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 463 Lines: 13 I just received a message from some SGI lawyer. They'll allow us to make the Magnum 4000 configuration disk, that is required to reconfigure the machine into little endian byte order, available via ftp under certain - very liberal - conditions. Think this is a good message for Magnum owners who do not longer have this disk. All I do need now to make this disk available via ftp is a disk image... Anyone out there who could give me one? Thanks, Ralf From velis@fcaglp.fcaglp.unlp.edu.ar Thu May 2 17:42:52 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 RAA19515 for ; Thu, 2 May 1996 17:42:51 +0200 (MET DST) Received: from fcaglp.fcaglp.unlp.edu.ar ([163.10.4.1]) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA19889; Thu, 2 May 1996 17:42:22 +0200 (MET) Received: by fcaglp.fcaglp.unlp.edu.ar (1.38.193.4/16.2) id AA05216; Thu, 2 May 1996 12:42:09 -0300 Message-Id: <9605021542.AA05216@fcaglp.fcaglp.unlp.edu.ar> From: velis@fcaglp.fcaglp.unlp.edu.ar (Danilo Velis) Subject: Linux + Silicon Indy R4600PC? To: linux-mips@fnet.fr Date: Thu, 2 May 96 12:42:09 -0300 Mailer: Elm [revision: 70.85] Content-Length: 521 Lines: 19 Hello!. Could you please let me know whether Linux-MIPS can be installed in a Silicon Graphics Indy R4600PC, 133MHz system? Any relevant information would be very much appreciated. Thank-you. Danilo Velis. -------------------------------------------------------------------- Department of Applied Geophysics National University of La Plata, Paseo del Bosque s/n La Plata 1900 Argentina Ph +54 21-217308 Fax +54 21-211761 E-Mail velis@fcaglp.edu.ar --------------------------------------------------------------------( From hannon@Crissy.Stanford.EDU Thu May 2 17:52:18 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 RAA19620 for ; Thu, 2 May 1996 17:52:18 +0200 (MET DST) Received: from crissy.Stanford.EDU by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA21368; Thu, 2 May 1996 17:51:50 +0200 (MET) Received: by Crissy.Stanford.EDU (8.6.8/25-LAGUNITA-eef) id IAA04009; Thu, 2 May 1996 08:54:46 -0700 From: Steph Hannon Message-Id: <199605021554.IAA04009@Crissy.Stanford.EDU> Subject: Re: Linux/MIPS compiling question To: linux-mips@fnet.fr Date: Thu, 2 May 96 8:54:46 PDT In-Reply-To: <199605020846.BAA23029@ncube>; from "Steve Herrod" at May 2, 96 1:46 am X-Mailer: ELM [version 2.3 PL11] Content-Length: 534 Lines: 23 Hi Steve, I didn't think I would get caught up in them, but I did, and I figured that I might as well ask the experts. I am going to come into the lab at about 9:30 am, so if you are around I would appreciate some help. Thanks, Steph > > Hey Steph, > > Sorry. I didn't realize that you were getting caught up with the > compilation tools. Why don't we sit down together sometime and take a > look at it. If you want to wait until after the meeting we can do that > or we can meet earlier... your choice. > > -Steve > > From paul@suite.sw.oz.au Fri May 3 00:37:04 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 AAA24305 for ; Fri, 3 May 1996 00:37:03 +0200 (MET DST) 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 AA15112; Fri, 3 May 1996 00:36:38 +0200 (MET) Received: from suite.sw.oz.au by swallow.sw.oz.au with SMTP id WAA25951; Thu, 2 May 1996 22:36:22 GMT (8.6.10/Unixware) (from paul@suite.sw.oz.au for ) Received: by suite.sw.oz.au id AA26562; Fri, 3 May 1996 08:36:16 +1000 (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: <199605022236.AA26562@suite.sw.oz.au> Subject: Re: Decstation port of Linux... (fwd) To: linux-mips@fnet.fr (Linux MIPS mailing list) Date: Fri, 3 May 1996 08:36:16 +1000 (EST) Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 1679 Lines: 44 Hi folks, I'm back from my Euro-american jaunt and have received some good news re the DECStations from Jon "maddog" Hall at DEC: > From: "Jon 'maddog' Hall, USG Senior Leader" > > Paul, > > I have sent a message to the Vice President in charge of workstations > outlining the need for documentation, and asking him to both point me to > the engineers in charge of this information, and to authorize them to > release any and all hardware information to you. > > In the meantime, you might also look at the code for netBSD. netBSD > works on a variety of DECstation platforms, and you might be able to > glean drivers, bootcode, etc. from them. > > I will let you know when I hear from the VP. If you don't hear from me > by next Tuesday, ping me again. > > md > -- > ========================================================================= > Jon "maddog" Hall > Officer - Linux International Senior Leader > Mailstop ZK03-2/U15 UNIX Software Group > Digital Equipment Corporation Internet: maddog@zk3.dec.com > 110 Spit Brook Rd. Voice: 603.881.1341 > Nashua, N.H. 03062-2698 Fax: 603.881.6059 So there you go - within a week or two we should have full doco... 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 May 3 01:02:22 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 BAA24705 for ; Fri, 3 May 1996 01:02:22 +0200 (MET DST) Received: from staff.cs.su.OZ.AU by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA15702; Fri, 3 May 1996 01:01:53 +0200 (MET) Received: from suite.sw.oz.au by swallow.sw.oz.au with SMTP id XAA26514; Thu, 2 May 1996 23:01:43 GMT (8.6.10/Unixware) (from paul@suite.sw.oz.au for ) Received: by suite.sw.oz.au id AA26702; Fri, 3 May 1996 09:01:39 +1000 (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: <199605022301.AA26702@suite.sw.oz.au> Subject: Re: Linux boot question To: linux-mips@fnet.fr Date: Fri, 3 May 1996 09:01:39 +1000 (EST) In-Reply-To: <199604261140.NAA04055@informatik.uni-koblenz.de> from "Systemkennung Linux" at Apr 26, 96 01:39:07 pm Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 1005 Lines: 27 Hi folks, Andy writes: > (yeah, I know, I was *very* silent in the past months -- sorry!) Wow, he still lives! > > Let's start the race, but be careful with your fingers! You need at > > least three to hit CTRL-ALT-DEL ;-) > > I'd like a machine with the reset button on a two meter cable much more. > One can even use this without fingers... I have one for my DECStation! It hangs just under my monitor, where, with a simple push, all bogus assembly language stops running :-) My DECStation now lives at my feet where it belongs - DOWN boy! :-) 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 andy@soft-n-hard.de Fri May 3 08:04:37 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 IAA00851 for ; Fri, 3 May 1996 08:04:36 +0200 (MET DST) Received: from aw.bnc.net by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA18463; Fri, 3 May 1996 08:04:09 +0200 (MET) Received: from newton.soft-n-hard.de (andy@newton.soft-n-hard.de [194.163.66.33]) by aw.bnc.net (8.6.12/8.6.12) with ESMTP id IAA17971 for ; Fri, 3 May 1996 08:08:40 +0200 Received: (from andy@localhost) by newton.soft-n-hard.de (8.6.12/8.6.9) id IAA00501; Fri, 3 May 1996 08:02:16 +0200 Date: Fri, 3 May 1996 08:02:16 +0200 From: Andreas Busse Message-Id: <199605030602.IAA00501@newton.soft-n-hard.de> To: linux-mips@fnet.fr Subject: Re: Magnum 4000 config disk In-Reply-To: <199605021531.RAA12132@informatik.uni-koblenz.de> X-Mailer: [XMailTool v3.1.2b] Content-Length: 594 Lines: 20 Hi all. > > All I do need now to make this disk available via ftp is a disk > image... Anyone out there who could give me one? > Yes, I guess I have one. But, isn't it on the FNET ftp site? I'm not sure, but if my memory serves right, it was uploaded to FNET last year... Cheers, Andy ----------------------------------------------------------- Andreas Busse | andy@soft-n-hard.de Soft N Hard GbR | Phone: +49 2636-970105 Im Hufen Boden 16, D-53498 Waldorf | Fax: +49 2636-970106 ----------------------------------------------------------- From linux@mailhost.uni-koblenz.de Fri May 3 15:50:35 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 PAA05954 for ; Fri, 3 May 1996 15:50:34 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA06415; Fri, 3 May 1996 15:49:27 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id PAA02662 for ; Fri, 3 May 1996 15:47:58 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605031347.PAA02662@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id PAA03316; Fri, 3 May 1996 15:46:08 +0200 Subject: GNU libc 960501 binaries To: linux-mips@fnet.fr Date: Fri, 3 May 1996 15:46:07 +0200 (MET DST) Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 850 Lines: 28 -----BEGIN PGP SIGNED MESSAGE----- Hi, I've uploaded GNU libc binaries built from snapshot 960501 to ftp.fnet.fr. The archive glibc-960501.tar.gz contains little endian libraries, documentation and the executables built by GNU libc. For native use you can just unpack the archive into the root of your Linux/MIPS box. If you want to use the binaries for crosscompilation you'll have to install the binaries in different places, so be carefull or you'll mipsify your system ;-) Happy hacking, Ralf 414cf01e0a5f6a78007dcd1a31b228e3 glibc-960501.tar.gz -----BEGIN PGP SIGNATURE----- Version: 2.6.2 iQCVAgUBMYoOHUckbl6vezDBAQEWQQP/aAzultR/3nbMsf6GUdhMfNXJrQpCI4Ym eVZHeZ5cgXlMfBqTFITeNfkROUZ9oLYkwi36aXF685I7x2yasR3DqCLNY3zvxinZ MVScj7fV8RQGMGi91MCGfFM+OCy5VPJPeVTfYq3gN5Flq/BODN1O8iyy+mWli+fB dWZH2ljCPX8= =kQba -----END PGP SIGNATURE----- From linux@mailhost.uni-koblenz.de Fri May 3 15:50:19 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 PAA05930 for ; Fri, 3 May 1996 15:50:18 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA06418; Fri, 3 May 1996 15:49:34 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id PAA02714 for ; Fri, 3 May 1996 15:49:38 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605031349.PAA02714@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id PAA03324; Fri, 3 May 1996 15:47:48 +0200 Subject: Re: Magnum 4000 config disk To: linux-mips@fnet.fr Date: Fri, 3 May 1996 15:47:48 +0200 (MET DST) In-Reply-To: <199605030602.IAA00501@newton.soft-n-hard.de> from "Andreas Busse" at May 3, 96 08:02:16 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 385 Lines: 13 Hi all. > > All I do need now to make this disk available via ftp is a disk > > image... Anyone out there who could give me one? > > Yes, I guess I have one. But, isn't it on the FNET ftp site? > I'm not sure, but if my memory serves right, it was uploaded > to FNET last year... Well, *if* at all it's in the private area and that's where it no longer needs to reside. Ralf From stig@spodbox.linux.org.uk Fri May 3 16:22:42 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 QAA06446 for ; Fri, 3 May 1996 16:22:40 +0200 (MET DST) Received: from gatekeeper.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA07409; Fri, 3 May 1996 16:22:13 +0200 (MET) Received: from spodbox.linux.org.uk (stig@[193.62.1.250]) by gatekeeper.waldorf-gmbh.de (8.6.12/8.6.9) with ESMTP id QAA17386 for ; Fri, 3 May 1996 16:25:07 +0200 Received: (from stig@localhost) by spodbox.linux.org.uk (8.6.12/8.6.9) id PAA01808 for linux@waldorf-gmbh.de; Fri, 3 May 1996 15:21:03 +0100 Date: Fri, 3 May 1996 15:21:03 +0100 From: David Priestley Message-Id: <199605031421.PAA01808@spodbox.linux.org.uk> To: linux@waldorf-gmbh.de Subject: MIPS-Linux Content-Length: 316 Lines: 10 Hi, I've just been given 2 Prime EXL 7000 machines which have in them: CPU: MIPS R3000A processor chip Rev: 3.0 FPU: MIPS R3010A VLSI Floating point Chip Rev: 3.0 Total real memory = 41943040 Have Linux been ported to this spec machine yet? If not - want a couple of machines to port it onto? David Priestley From imp@rover.village.org Fri May 3 16:25:30 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 QAA06496 for ; Fri, 3 May 1996 16:25:27 +0200 (MET DST) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA07472; Fri, 3 May 1996 16:25:01 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.7.5/8.6.6) with SMTP id IAA09777 for ; Fri, 3 May 1996 08:25:17 -0600 (MDT) Message-Id: <199605031425.IAA09777@rover.village.org> To: linux-mips@fnet.fr Subject: Re: GNU libc 960501 binaries In-Reply-To: Your message of Fri, 03 May 1996 15:46:07 +0200 Date: Fri, 03 May 1996 08:25:16 -0600 From: Warner Losh Content-Length: 243 Lines: 7 : 414cf01e0a5f6a78007dcd1a31b228e3 glibc-960501.tar.gz Just grabbed this and its md5 was cool. Given the network problems earlier this week, this is some good news. Now, to figure out how to update my cross compilation www page... Warner From imp@rover.village.org Fri May 3 16:31:43 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 QAA06589 for ; Fri, 3 May 1996 16:31:42 +0200 (MET DST) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA07648; Fri, 3 May 1996 16:31:13 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.7.5/8.6.6) with SMTP id IAA09838 for ; Fri, 3 May 1996 08:31:28 -0600 (MDT) Message-Id: <199605031431.IAA09838@rover.village.org> To: linux-mips@fnet.fr Subject: Re: MIPS-Linux In-Reply-To: Your message of Fri, 03 May 1996 15:21:03 BST Date: Fri, 03 May 1996 08:31:28 -0600 From: Warner Losh Content-Length: 472 Lines: 16 : CPU: MIPS R3000A processor chip Rev: 3.0 : FPU: MIPS R3010A VLSI Floating point Chip Rev: 3.0 : Total real memory = 41943040 The DECstation port should be a good basis for this port.... : Have Linux been ported to this spec machine yet? No. Not yet. However, unless you know the boot sequence and have docs on the hardware (or can get them) it is likely to be hard. : If not - want a couple of machines to port it onto? I always want more machines :-) Warner From linux@mailhost.uni-koblenz.de Fri May 3 17:04:28 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 RAA07055 for ; Fri, 3 May 1996 17:04:28 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA11501; Fri, 3 May 1996 17:03:14 +0200 (MET) Received: from marvin..uni-koblenz (marvin.uni-koblenz.de [141.26.4.6]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id RAA05716 for ; Fri, 3 May 1996 17:02:51 +0200 (MET DST) Received: by marvin..uni-koblenz (SMI-8.6/SMI-SVR4) id RAA06822; Fri, 3 May 1996 17:00:37 +0200 From: linux@mailhost.uni-koblenz.de (Linux Support) Message-Id: <199605031500.RAA06822@marvin..uni-koblenz> Subject: Re: GNU libc 960501 binaries To: linux-mips@fnet.fr Date: Fri, 3 May 1996 17:00:36 +0200 (MET DST) In-Reply-To: <199605031425.IAA09777@rover.village.org> from "Warner Losh" at May 3, 96 08:25:16 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1209 Lines: 26 Hi, > : 414cf01e0a5f6a78007dcd1a31b228e3 glibc-960501.tar.gz > > Just grabbed this and its md5 was cool. Given the network problems > earlier this week, this is some good news. Now, to figure out how to > update my cross compilation www page... The whole trick is to make the traffic not cross the Ebone which is currently in a close to unusable state 24h/day. That's one reason why one never has to many accounts ... Please add comment to the cross compilation page about the tar to use to unpack the package. Demand paged MIPS binaries contain huge areas of zeros. Thats why I pack my archives using the -S (or --sparse) option. In case of this glibc archive the difference is about 230kb, for root-0.01.tar.gz even 8mb which otherwise would be used just store zeros on the disk ... Unfortunately there are tar versions out there that break when unpacking such an archive. Among them are the versions of tar delivered with SunOS 4 and 5 which break with the message "Bad directory checksum" or even just silently. Use GNU tar and it'll work. Please note that this comment applies to all my binary archive, not just the two mentioned. Ralf "I'll never visit a dentist again in my life ..." From linux@mailhost.uni-koblenz.de Fri May 3 17:17:26 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 RAA07200 for ; Fri, 3 May 1996 17:17:26 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA11935; Fri, 3 May 1996 17:16:58 +0200 (MET) Received: from marvin..uni-koblenz (marvin.uni-koblenz.de [141.26.4.6]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id RAA06134; Fri, 3 May 1996 17:15:58 +0200 (MET DST) Received: by marvin..uni-koblenz (SMI-8.6/SMI-SVR4) id RAA06841; Fri, 3 May 1996 17:13:43 +0200 From: linux@mailhost.uni-koblenz.de (Linux Support) Message-Id: <199605031513.RAA06841@marvin..uni-koblenz> Subject: Re: MIPS-Linux To: linux-mips@fnet.fr Date: Fri, 3 May 1996 17:13:43 +0200 (MET DST) Cc: stig@spodbox.linux.org.uk In-Reply-To: <199605031421.PAA01808@spodbox.linux.org.uk> from "David Priestley" at May 3, 96 03:21:03 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 552 Lines: 16 Hi, > I've just been given 2 Prime EXL 7000 machines which have in them: > CPU: MIPS R3000A processor chip Rev: 3.0 > FPU: MIPS R3010A VLSI Floating point Chip Rev: 3.0 > Total real memory = 41943040 > > Have Linux been ported to this spec machine yet? > If not - want a couple of machines to port it onto? If I'me informed right this prime is OEM version of a Mips Computer Systems, Inc system. So this port might not be to difficult. But anyway, maybe you or some SGI/Mips guy on this list knows more about this machine than me??? Ralf From linux@mailhost.uni-koblenz.de Fri May 3 20:11:05 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 UAA09227 for ; Fri, 3 May 1996 20:11:02 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA17393; Fri, 3 May 1996 20:10:17 +0200 (MET) Received: from marvin..uni-koblenz (marvin.uni-koblenz.de [141.26.4.6]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id UAA10734 for ; Fri, 3 May 1996 20:10:31 +0200 (MET DST) Received: by marvin..uni-koblenz (SMI-8.6/SMI-SVR4) id UAA06960; Fri, 3 May 1996 20:08:18 +0200 From: linux@mailhost.uni-koblenz.de (Linux Support) Message-Id: <199605031808.UAA06960@marvin..uni-koblenz> Subject: Linux/MIPS job offer To: linux-mips@fnet.fr Date: Fri, 3 May 1996 20:08:18 +0200 (MET DST) Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 950 Lines: 29 Hi all, even far before Linux/MIPS has reached release quality, here is the first related job offer! I'm forwarding this to the list from Eugen Schenfeld. Ralf A system programmer is needed for a Linux/MIPS port to a Massivelly Parallel Processing new architecture. The intention is to create a MP MIPS version based on the MIPS 4300 CPUs, and other aspects of a distribited shared memory architecture. The work location is at the NEC Research Institute, in Princeton, NJ USA, for one year. A good Ph.D. student (hacker) that would like to combine this work as part of his/hers Ph.D. thesis is also encouraged to consider. For more info. please contact Dr. Eugen Schenfeld at eugen@research.nj.nec.com or (609)951-2742. Thanks very much, Eugen Eugen Schenfeld NEC Research Institute 4 Independence Way Princeton, NJ 08540 phone: 609 951 2742 fax: 609 951 2482 email: eugen@research.nj.nec.com (Inet) From imp@rover.village.org Fri May 3 21:36:51 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 VAA10201 for ; Fri, 3 May 1996 21:36:44 +0200 (MET DST) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA20788; Fri, 3 May 1996 21:36:13 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.7.5/8.6.6) with SMTP id NAA10600 for ; Fri, 3 May 1996 13:36:27 -0600 (MDT) Message-Id: <199605031936.NAA10600@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Linux/MIPS job offer In-Reply-To: Your message of Fri, 03 May 1996 20:08:18 +0200 Date: Fri, 03 May 1996 13:36:27 -0600 From: Warner Losh Content-Length: 513 Lines: 13 : The work location is at the NEC Research Institute, in Princeton, NJ : USA, for one year. I've talked to Eugen, and it sounds like a great place to work and would be a lot of fun. Unfortunately, they really seem to want someone that is full time that can go on site for that time. Alas, my present situation wouldn't let me do that :-(. Looks like Linux/MIPS is starting to get some momentum. Witness the SGI announcement of a few days ago. Won't be long and the flood gates of jobs will open :-) Warner From imp@rover.village.org Fri May 3 21:38:56 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 VAA10241 for ; Fri, 3 May 1996 21:38:49 +0200 (MET DST) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA20874; Fri, 3 May 1996 21:38:17 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.7.5/8.6.6) with SMTP id NAA10625 for ; Fri, 3 May 1996 13:38:34 -0600 (MDT) Message-Id: <199605031938.NAA10625@rover.village.org> To: linux-mips@fnet.fr Subject: Re: GNU libc 960501 binaries In-Reply-To: Your message of Fri, 03 May 1996 17:00:36 +0200 Date: Fri, 03 May 1996 13:38:34 -0600 From: Warner Losh Content-Length: 323 Lines: 13 : Please add comment to the cross compilation page about the tar to use : to unpack the package. ... : Use GNU tar and it'll work. OK. I'll add that to the list. : Please note that this comment applies to all my binary archive, not just : the two mentioned. Does that apply to the source distributions as well? Warner From linux@mailhost.uni-koblenz.de Fri May 3 22:00:48 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 WAA10621 for ; Fri, 3 May 1996 22:00:41 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA21638; Fri, 3 May 1996 22:00:09 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id WAA13186 for ; Fri, 3 May 1996 22:00:24 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605032000.WAA13186@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id VAA07426; Fri, 3 May 1996 21:58:31 +0200 Subject: Re: Linux/MIPS job offer To: linux-mips@fnet.fr Date: Fri, 3 May 1996 21:58:29 +0200 (MET DST) In-Reply-To: <199605031936.NAA10600@rover.village.org> from "Warner Losh" at May 3, 96 01:36:27 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 318 Lines: 12 Hi, > Looks like Linux/MIPS is starting to get some momentum. Witness the > SGI announcement of a few days ago. Won't be long and the flood gates ^^^^^ > of jobs will open :-) ^^^^ I don't think these two guys will show interest ... Ralf From linux@mailhost.uni-koblenz.de Fri May 3 22:04:34 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 WAA10677 for ; Fri, 3 May 1996 22:04:27 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA21752; Fri, 3 May 1996 22:03:52 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id WAA13258 for ; Fri, 3 May 1996 22:04:08 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605032004.WAA13258@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id WAA07441; Fri, 3 May 1996 22:02:18 +0200 Subject: Re: GNU libc 960501 binaries To: linux-mips@fnet.fr Date: Fri, 3 May 1996 22:02:18 +0200 (MET DST) In-Reply-To: <199605031938.NAA10625@rover.village.org> from "Warner Losh" at May 3, 96 01:38:34 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 436 Lines: 18 Hi, > : Please add comment to the cross compilation page about the tar to use > : to unpack the package. > ... > : Use GNU tar and it'll work. > > OK. I'll add that to the list. > > : Please note that this comment applies to all my binary archive, not just > : the two mentioned. > > Does that apply to the source distributions as well? No; the sources don't contain hole at all so this option doesn't make sense at all. Ralf From imp@rover.village.org Fri May 3 22:15:54 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 WAA10811 for ; Fri, 3 May 1996 22:15:53 +0200 (MET DST) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA22169; Fri, 3 May 1996 22:15:27 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.7.5/8.6.6) with SMTP id OAA10756 for ; Fri, 3 May 1996 14:15:39 -0600 (MDT) Message-Id: <199605032015.OAA10756@rover.village.org> To: linux-mips@fnet.fr Subject: Re: GNU libc 960501 binaries In-Reply-To: Your message of Fri, 03 May 1996 22:02:18 +0200 Date: Fri, 03 May 1996 14:15:38 -0600 From: Warner Losh Content-Length: 227 Lines: 7 : No; the sources don't contain hole at all so this option doesn't make sense : at all. Didn't think so, but wanted to make sure. I know I've never had a problem, but I wasn't sure that I had ever used a non-gnu tar. Warner From linux@mailhost.uni-koblenz.de Fri May 3 22:28:25 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 WAA10934 for ; Fri, 3 May 1996 22:28:16 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA22648; Fri, 3 May 1996 22:27:36 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id WAA14078 for ; Fri, 3 May 1996 22:27:49 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605032027.WAA14078@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id WAA07494; Fri, 3 May 1996 22:25:59 +0200 Subject: FAT alignment problem To: linux-mips@fnet.fr Date: Fri, 3 May 1996 22:25:58 +0200 (MET DST) Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 2811 Lines: 73 -----BEGIN PGP SIGNED MESSAGE----- Hi, the following patch fixes an alignment problem in the FAT filesystem that makes the kernel panic with an ADEL exception otherwise. Ralf - --- linux-1.3.96.orig/fs/fat/inode.c Sun Apr 21 10:11:13 1996 +++ linux-1.3.96/fs/fat/inode.c Fri May 3 21:44:05 1996 @@ -3,6 +3,8 @@ * * Written 1992,1993 by Werner Almesberger * VFAT extensions by Gordon Chaffee, merged with msdos fs by Henrik Storner + * + * 3 May 1996 Fixed alignment problems for RISC architectures. */ #define __NO_VERSION__ @@ -242,7 +244,13 @@ #define ROUND_TO_MULTIPLE(n,m) ((n) && (m) ? (n)+(m)-1-((n)-1)%(m) : 0) /* don't divide by zero */ - - logical_sector_size = CF_LE_W(*(unsigned short *) &b->sector_size); + /* + * There are some misaligned 16 bit members in the msdos bootblock + * that we decode manualy without the CF_LE_W() macro. This avoids + * problems on architecture with restrictive alignment rules like + * MIPS or Alpha -- Ralf + */ + logical_sector_size = b->sector_size[0] | (b->sector_size[1] << 8); sector_mult = logical_sector_size >> SECTOR_BITS; MSDOS_SB(sb)->cluster_size = b->cluster_size*sector_mult; MSDOS_SB(sb)->fats = b->fats; @@ -250,13 +258,13 @@ MSDOS_SB(sb)->fat_length = CF_LE_W(b->fat_length)*sector_mult; MSDOS_SB(sb)->dir_start = (CF_LE_W(b->reserved)+b->fats*CF_LE_W( b->fat_length))*sector_mult; - - MSDOS_SB(sb)->dir_entries = CF_LE_W(*((unsigned short *) &b->dir_entries - - )); + MSDOS_SB(sb)->dir_entries = (b->dir_entries[0] | + (b->dir_entries[1] << 8)); MSDOS_SB(sb)->data_start = MSDOS_SB(sb)->dir_start+ROUND_TO_MULTIPLE(( MSDOS_SB(sb)->dir_entries << MSDOS_DIR_BITS) >> SECTOR_BITS, sector_mult); - - data_sectors = (CF_LE_W(*((unsigned short *) &b->sectors)) ? - - CF_LE_W(*((unsigned short *) &b->sectors)) : + data_sectors = ((b->sectors[0] | (b->sectors[1] << 8)) ? + (b->sectors[0] | (b->sectors[1] << 8)) : CF_LE_L(b->total_sect))*sector_mult-MSDOS_SB(sb)->data_start; error = !b->cluster_size || !sector_mult; if (!error) { @@ -293,7 +301,7 @@ MSDOS_SB(sb)->fats,MSDOS_SB(sb)->fat_start,MSDOS_SB(sb)->fat_length, MSDOS_SB(sb)->dir_start,MSDOS_SB(sb)->dir_entries, MSDOS_SB(sb)->data_start, - - CF_LE_W(*(unsigned short *) &b->sectors), + (b->sectors[0] | (b->sectors[1] << 8)), (unsigned long)b->total_sect,logical_sector_size); printk ("Transaction block size = %d\n",blksize); } -----BEGIN PGP SIGNATURE----- Version: 2.6.2 iQCVAgUBMYpr1Eckbl6vezDBAQEqbAP/WIscK4B0pcogF91AxHJ9l9KWtW0LOczO +C4jcT3EevabEhtqK/USRLaxuNIrf1iT69LWU3V9iPLjpkLvXOY4P9khdlzKFF4y LiLmdflBINMJIpQDK81L4xAUwA5wZRaF6oMNIXPvqkQ+eNJW9TvWlYHgjjkZ2l+Q HWHJzH7r3w4= =Ztzr -----END PGP SIGNATURE----- From Frederic.Roussel@eng.efi.com Fri May 3 22:54:52 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 WAA11170 for ; Fri, 3 May 1996 22:54:38 +0200 (MET DST) Received: from brussels.efi.com by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA23478; Fri, 3 May 1996 22:54:05 +0200 (MET) Received: from quimper.efi.com ([198.93.202.20]) by brussels.efi.com (post.office MTA v1.9.3 ID# 0-12429) with SMTP id AAA17396 for ; Fri, 3 May 1996 13:53:52 -0700 Received: by quimper.efi.com (940816.SGI.8.6.9) id NAA00876; Fri, 3 May 1996 13:53:45 -0700 From: "Frederic Roussel" Message-Id: <199605032053.NAA00876@quimper.efi.com> Subject: MILO on NEC RISCstation 2200 To: linux-mips@fnet.fr Date: Fri, 3 May 1996 13:53:45 -0700 (PDT) X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 760 Lines: 30 Hello, I've just tried MILO on an NEC RISCstation 2200. Well, milo-0.21 started Ok on that machine, printing: ARC Vendor ID : [NEC W&S] ARC ProductID : [00004c81823d0072] (ethernet addr. related) ARC System ID : [NEC RA-94] Saying: Processor type : [UNKNOWN I04 V6.0] And: Secondary cache size : [unknown] (which is quite sensible as L2 cache is not installed on that machine) milo-0.26 has been leading to an `TlbS exception occured' message therefore entering the low level monitor of the NEC ARC BIOS. milo printed its version, copyright string, the 3 ARC Id's and failed before trying to guess on what HW it's currently running. Hope it may be somehow useful! -- Frederic.R.R.Roussel -- Frederic.Roussel@eng.efi.com From paul@suite.sw.oz.au Sat May 4 02:13:55 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 CAA13602 for ; Sat, 4 May 1996 02:13:50 +0200 (MET DST) 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 AA29348; Sat, 4 May 1996 02:13:15 +0200 (MET) Received: from suite.sw.oz.au by swallow.sw.oz.au with SMTP id AAA02152; Sat, 4 May 1996 00:13:15 GMT (8.6.10/Unixware) (from paul@suite.sw.oz.au for ) Received: by suite.sw.oz.au id AA07473; Sat, 4 May 1996 10:13:08 +1000 (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: <199605040013.AA07473@suite.sw.oz.au> Subject: Attn all DECStation contributors To: linux-mips@fnet.fr (Linux MIPS mailing list) Date: Sat, 4 May 1996 10:13:07 +1000 (EST) Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 1370 Lines: 33 I have put a new test kernel on my ftp site, which I would like all DECStation contributors to try: ftp://softway.com.au/pub/linux-mips/dec_vmlinux.1.3.62.test.gz (It'd be nice if fnet could mirror it.) It ought to correctly determine the amount of RAM, and the type and firmware revision of you system. Please let me know if it does not correctly identify your system - tell me the system id number, and the precise configuration. Please note: DS2100/3100 owners may find it doesn't do anything... Owners of R4000 CPU upgrades need to tell me their system id's! It is my first cut at the code required to cope with Stoned's new argument passing scheme that just happens to also try to correctly identify DEC hardware. With luck I will have it correctly identifying hardware and passing that to the kernel any day now. Then all I have to do is to try to keep up with Ralf's kernel patches...!! 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 zaitcev@lab.ipmce.su Sat May 4 13:45:10 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 NAA22474 for ; Sat, 4 May 1996 13:45:09 +0200 (MET DST) Received: from ipmce.su (purgator-ppp0.ipmce.su) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA22304; Sat, 4 May 1996 13:44:45 +0200 (MET) Received: from tretyak.sun.mcst.ru by ipmce.su (8.6.9/SMI-RJMc) id PAA24542; Sat, 4 May 1996 15:40:07 +0400 Received: by tretyak.sun.mcst.ru (5.x/SMI-SVR4) id AA02613; Sat, 4 May 1996 15:47:35 +0400 Received: from terra.lab.ipmce.su(192.124.171.107) by tretyak.sun.mcst.ru via smap (V1.3) id sma002611; Sat May 4 15:47:12 1996 Received: from jamica.lab.ipmce.su by lab.ipmce.su (4.1/SMI-4.1) id AA03214; Sat, 4 May 96 15:45:16 MSD Received: from mellorn.lab.sun.mcst.ru by jamica.lab.ipmce.su (SMI-8.6/SMI-SVR4) id PAA02729; Sat, 4 May 1996 15:45:16 +0400 Received: by mellorn.lab.sun.mcst.ru (SMI-8.6/SMI-SVR4) id PAA00417; Sat, 4 May 1996 15:45:14 +0400 Date: Sat, 4 May 1996 15:45:14 +0400 From: zaitcev@lab.ipmce.su (Pete A. Zaitcev) Message-Id: <199605041145.PAA00417@mellorn.lab.sun.mcst.ru> To: linux-mips@fnet.fr Subject: Re: Linux/MIPS job offer Cc: linux@mailhost.uni-koblenz.de X-Sun-Charset: US-ASCII Content-Length: 315 Lines: 10 >: The work location is at the NEC Research Institute, in Princeton, NJ >: USA, for one year. > >Looks like Linux/MIPS is starting to get some momentum. Witness the >SGI announcement of a few days ago. Won't be long and the flood gates >of jobs will open :-) The question is, will they make an H-1B visas? Pete From linux@mailhost.uni-koblenz.de Sat May 4 15:11:01 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 PAA23538 for ; Sat, 4 May 1996 15:10:58 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA25087; Sat, 4 May 1996 15:10:32 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id PAA16156 for ; Sat, 4 May 1996 15:10:47 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605041310.PAA16156@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id PAA07734; Sat, 4 May 1996 15:08:55 +0200 Subject: Re: Attn all DECStation contributors To: linux-mips@fnet.fr Date: Sat, 4 May 1996 15:08:55 +0200 (MET DST) In-Reply-To: <199605040013.AA07473@suite.sw.oz.au> from "Paul Antoine" at May 4, 96 10:13:07 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 235 Lines: 9 Hi Paul, > Then all I have to do is to try to keep up with Ralf's kernel > patches...!! You can relax again... I've catched up with Linus (hopefully still current ...) 1.3.97, so there is just one patch on hold for you :-) Ralf From linux@mailhost.uni-koblenz.de Sat May 4 15:29:09 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 PAA23697 for ; Sat, 4 May 1996 15:29:09 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA25805; Sat, 4 May 1996 15:28:41 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id PAA16554; Sat, 4 May 1996 15:28:51 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605041328.PAA16554@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id PAA07773; Sat, 4 May 1996 15:26:59 +0200 Subject: Re: MILO on NEC RISCstation 2200 To: linux-mips@fnet.fr Date: Sat, 4 May 1996 15:26:59 +0200 (MET DST) Cc: Frederic.Roussel@eng.efi.com In-Reply-To: <199605032053.NAA00876@quimper.efi.com> from "Frederic Roussel" at May 3, 96 01:53:45 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1318 Lines: 39 Hi, > I've just tried MILO on an NEC RISCstation 2200. > > Well, milo-0.21 started Ok on that machine, printing: > > ARC Vendor ID : [NEC W&S] > ARC ProductID : [00004c81823d0072] (ethernet addr. related) > ARC System ID : [NEC RA-94] > > Saying: > > Processor type : [UNKNOWN I04 V6.0] Interesting. Can you please exactly tell me what CPU your machine has? > And: > > Secondary cache size : [unknown] > > (which is quite sensible as L2 cache is not installed on that machine) Milo detect only (via the ARC BIOS' limitied capabilites) CPU controlled second level cache, not PC-style chipset controlled. The later type doesn't need special support, so Milo doesn't really care about it. > milo-0.26 has been leading to an `TlbS exception occured' message > therefore entering the low level monitor of the NEC ARC BIOS. > milo printed its version, copyright string, the 3 ARC Id's and failed > before trying to guess on what HW it's currently running. Hm... Not unusual. The ARC BIOSes that I've tried yet are *that* buggy that they are barely useable. I bet they've just been developed to the point where they can be used to install & boot WNT & related tools. Would you run some special hardwaretest program, if I send you one to find out more about your machine? Ralf From imp@rover.village.org Sat May 4 17:03:34 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 RAA24812 for ; Sat, 4 May 1996 17:03:31 +0200 (MET DST) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA28761; Sat, 4 May 1996 17:03:04 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.7.5/8.6.6) with SMTP id JAA15851; Sat, 4 May 1996 09:03:21 -0600 (MDT) Message-Id: <199605041503.JAA15851@rover.village.org> To: linux-mips@fnet.fr Subject: Re: MILO on NEC RISCstation 2200 Cc: Frederic.Roussel@eng.efi.com In-Reply-To: Your message of Sat, 04 May 1996 15:26:59 +0200 Date: Sat, 04 May 1996 09:03:21 -0600 From: Warner Losh Content-Length: 1286 Lines: 25 : Milo detect only (via the ARC BIOS' limitied capabilites) CPU controlled : second level cache, not PC-style chipset controlled. The later type : doesn't need special support, so Milo doesn't really care about it. This is good information to add to some doc somewhere. Took me a while to realize that while I did have 512k of PC-Style Cache, I didn't really have 512K of MIPS-Style L2 Cache. : > milo-0.26 has been leading to an `TlbS exception occured' message : > therefore entering the low level monitor of the NEC ARC BIOS. : > milo printed its version, copyright string, the 3 ARC Id's and failed : > before trying to guess on what HW it's currently running. : : Hm... Not unusual. The ARC BIOSes that I've tried yet are *that* : buggy that they are barely useable. I bet they've just been developed : to the point where they can be used to install & boot WNT & related tools. I've seen this problem (or a similar one) when I was booting MILO at the wrong address. I'll bet the BIOSes were developed to the point they could install and boot WNT 3.1. I've seen rumblings in the news groups about NT needing new ROMs from time to time when new releases came out due to bugs in the ROMs. I don't recall the exact details, but was around the 3.51 beta time frame. Warner From linux@mailhost.uni-koblenz.de Sat May 4 20:24:32 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 UAA26951 for ; Sat, 4 May 1996 20:24:32 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA05552; Sat, 4 May 1996 20:24:00 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id UAA23310 for ; Sat, 4 May 1996 20:24:17 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605041824.UAA23310@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id UAA08318; Sat, 4 May 1996 20:22:25 +0200 Subject: Re: MILO on NEC RISCstation 2200 To: linux-mips@fnet.fr Date: Sat, 4 May 1996 20:22:24 +0200 (MET DST) In-Reply-To: <199605041503.JAA15851@rover.village.org> from "Warner Losh" at May 4, 96 09:03:21 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1010 Lines: 20 Hi, > I've seen this problem (or a similar one) when I was booting MILO at > the wrong address. I'll bet the BIOSes were developed to the point > they could install and boot WNT 3.1. I've seen rumblings in the news > groups about NT needing new ROMs from time to time when new releases > came out due to bugs in the ROMs. I don't recall the exact details, > but was around the 3.51 beta time frame. You can't even find out exactly how much RAM is installed ... Oh, do I have to add that the Magnum/M700 BIOSes are (C) M$ ... There is also no exactly defined address for where to load an executable to. The address that I'm using for Milo is the address of the NT installer but there are other addresses in use. On my Tyne I never could use the network card installation tools becaue at the load address i had no RAM installed ... Have to experiment with relocatable code when the next public release of the binutils is out - I really try to avoid using snapshots which could already do that. Ralf From imp@rover.village.org Sat May 4 22:59:51 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 WAA28647 for ; Sat, 4 May 1996 22:59:51 +0200 (MET DST) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA10255; Sat, 4 May 1996 22:59:08 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.7.5/8.6.6) with SMTP id OAA16810 for ; Sat, 4 May 1996 14:59:17 -0600 (MDT) Message-Id: <199605042059.OAA16810@rover.village.org> To: linux-mips@fnet.fr Subject: Re: MILO on NEC RISCstation 2200 In-Reply-To: Your message of Sat, 04 May 1996 20:22:24 +0200 Date: Sat, 04 May 1996 14:59:17 -0600 From: Warner Losh Content-Length: 799 Lines: 21 : You can't even find out exactly how much RAM is installed ... I know. I wish I knew how NT and/or my BIOS found this out (it tells me when I boot), but I can't. : There is also no exactly defined address for where to load an executable : to. The address that I'm using for Milo is the address of the NT : installer but there are other addresses in use. I was wondering where that address came from... : Have to experiment with relocatable code when : the next public release of the binutils is out - I really try to avoid : using snapshots which could already do that. Yes, but how will that help you? Or does the BIOS load it to some random address, but pic code will render what that address is a moot point. That would be very very very useful in doing the install tools :-) Warner From linux@mailhost.uni-koblenz.de Sun May 5 02:21:58 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 CAA01529 for ; Sun, 5 May 1996 02:21:56 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA17727; Sun, 5 May 1996 02:20:59 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id CAA11073 for ; Sun, 5 May 1996 02:21:03 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605050021.CAA11073@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id CAA08919; Sun, 5 May 1996 02:19:11 +0200 Subject: Re: MILO on NEC RISCstation 2200 To: linux-mips@fnet.fr Date: Sun, 5 May 1996 02:19:10 +0200 (MET DST) In-Reply-To: <199605042059.OAA16810@rover.village.org> from "Warner Losh" at May 4, 96 02:59:17 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 729 Lines: 21 Hi, > : You can't even find out exactly how much RAM is installed ... > > I know. I wish I knew how NT and/or my BIOS found this out (it tells > me when I boot), but I can't. Probably some heuristic. When you add up all RAM reported by the BIOS you get less than is really installed. Except on a Tyne which reports 256MB to much ... > Yes, but how will that help you? Or does the BIOS load it to some > random address, but pic code will render what that address is a moot > point. That would be very very very useful in doing the install tools > :-) The "Portable bootloader standard" contains some gossip about relocation. Maybe they're talking about PIC. Too much to ignore it and not enough to use it ... Ralf From imp@rover.village.org Sun May 5 06:57:57 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 GAA05674 for ; Sun, 5 May 1996 06:57:56 +0200 (MET DST) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA00752; Sun, 5 May 1996 06:57:31 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.7.5/8.6.6) with SMTP id WAA20089 for ; Sat, 4 May 1996 22:57:49 -0600 (MDT) Message-Id: <199605050457.WAA20089@rover.village.org> To: linux-mips@fnet.fr Subject: Re: MILO on NEC RISCstation 2200 In-Reply-To: Your message of Sun, 05 May 1996 02:19:10 +0200 Date: Sat, 04 May 1996 22:57:47 -0600 From: Warner Losh Content-Length: 907 Lines: 22 : Probably some heuristic. When you add up all RAM reported by the BIOS you : get less than is really installed. Except on a Tyne which reports 256MB : to much ... Hmmm, if the BIOS is gone when the kernel is started, how do I get that information out of the BIOS? Don't I need that to find out where the tags start? I notice the PICA and Jazz based machines check a register that the R4400PC doesn't seem to have :-(. : > Yes, but how will that help you? Or does the BIOS load it to some : > random address, but pic code will render what that address is a moot : > point. That would be very very very useful in doing the install tools : > :-) : : The "Portable bootloader standard" contains some gossip about relocation. : Maybe they're talking about PIC. Too much to ignore it and not enough to : use it ... Oh good. Sure wish that standard wasn't (C) Microsoft. I'd want a copy :-(. Warner From linux@mailhost.uni-koblenz.de Sun May 5 14:00:10 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 OAA10700 for ; Sun, 5 May 1996 14:00:10 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA23678; Sun, 5 May 1996 13:59:44 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id OAA22412 for ; Sun, 5 May 1996 14:00:00 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605051200.OAA22412@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id NAA09093; Sun, 5 May 1996 13:58:08 +0200 Subject: Re: MILO on NEC RISCstation 2200 To: linux-mips@fnet.fr Date: Sun, 5 May 1996 13:58:08 +0200 (MET DST) In-Reply-To: <199605050457.WAA20089@rover.village.org> from "Warner Losh" at May 4, 96 10:57:47 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1233 Lines: 28 Hi, > : Probably some heuristic. When you add up all RAM reported by the BIOS you > : get less than is really installed. Except on a Tyne which reports 256MB > : to much ... > > Hmmm, if the BIOS is gone when the kernel is started, how do I get > that information out of the BIOS? Don't I need that to find out where > the tags start? I notice the PICA and Jazz based machines check a > register that the R4400PC doesn't seem to have :-(. A reasonable heuristic is to query the ARC BIOS for the RAM, filter out obvious wrong answers (like on the Tyne), the round up the end address of the last chunk returned to the next multiple of two. Or, if your machine doesn't throw exceptions for accessing non-existing memory, do test by writing date to the memory. These two methods will of course fail for machines with multiple, non continuously mapped chunks of memory - the ARC standard and portable bootloader standard allow them - but I didn't encounter one of them yet. Well, except that 640k-1024k hole in your machine. Completly unrelated - the new libc 960501 is significantly better than the old. I've just built a usable bash and do a native recompile of the lib to get rid of some crosscompiltion artefacts. Ralf From dfrick@dial.eunet.ch Mon May 6 01:13:03 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 BAA26981 for ; Mon, 6 May 1996 01:13:01 +0200 (MET DST) Received: from chsun.eunet.ch by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA19783; Mon, 6 May 1996 01:12:37 +0200 (MET) Received: from miston.any.net by chsun.eunet.ch (8.6.10/1.34) id BAA10551; Mon, 6 May 1996 01:12:43 +0200 Received: (from dfrick@localhost) by miston.any.net (8.6.12/8.6.12) id BAA03252; Mon, 6 May 1996 01:16:16 +0200 Date: Mon, 6 May 1996 01:16:16 +0200 Message-Id: <199605052316.BAA03252@miston.any.net> From: Didier Frick To: linux-mips@fnet.fr Subject: Linux-mips on R3000 Content-Length: 1214 Lines: 29 Hello, I'm currently considering using Linux-mips on an embedded system based on a R3000-compatible CPU. (IDT R3052E) I have read the FAQ, and I still have some questions: - The target hardware design relies on the fact that the CPU is in big-endian mode. The FAQ says that in this case "all kernel endian dependencies must be removed." Are there many such dependencies ? With the size of the kernel, I guess that's hard to say. I understand there's a port of Linux to the 68000, which is big-endian. Does it run into the problem ? - If I need only serial, timer and ramdisk drivers (loading software from ROM), what are the chances that the kernel in its current state would function correctly in this environment ? - By the way, the serial port would be used as a console. The current console code is heavily hardware dependent, assuming the use of a PC display adapter. Is a serial console hack available somewhere ? - Once (and if) this basic configuration is up and running, drivers for additional hardware interfaces could be developed as kernel modules. If there is a chance to get the system running on this particular target, maybe I could attempt to do the job. Thanks for any feedback, Didier From hodgen@mailhost.uni-koblenz.de Mon May 6 12:25:39 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 MAA00276 for ; Mon, 6 May 1996 12:25:30 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA09740; Mon, 6 May 1996 12:24:46 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id MAA02434 for ; Mon, 6 May 1996 12:23:17 +0200 (MET DST) From: Wayne Hodgen Message-Id: <199605061023.MAA02434@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id MAA09799; Mon, 6 May 1996 12:21:24 +0200 Subject: Procps binary package To: linux-mips@fnet.fr Date: Mon, 6 May 1996 12:21:24 +0200 (MET DST) Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1373 Lines: 26 Hi all, I got sick of traps using "ps aux" (and of listening to Ralf telling me to use "ps al" cos it still worked) so I got (some) of procps to compile (native on the Olli :) If anyone is interested I'll get Ralf to upload the archive (just now he says no-one has complained so no-one must need/want it ;) Make my day, prove him wrong >-) I was on holiday for a week or so and I'm catching up on the list. I saw several people posting about success with the Olli's, I had mine booting over the net 2 or 3 weeks ago. I put a 500MB IDE disk in it and after Ralf killed the unaligned problems this brought to light, now run from that: much better :) The console is still the worst problem though. I spoke to ralf and he's arranged to borrow the Olli docs from Andy (including a doc for the G384 I hope! I WANT A CURSOR!). Once I have this I'll start work on a new console. The main reason for the Olli crashing at the moment is memory accesses by the console screwing things up. Once we have a new console, the Olli's should be pretty stable... Wayne -- Wayne Hodgen | hodgen@informatik.uni-koblenz.de | #include From paul@suite.sw.oz.au Mon May 6 14:19:58 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 OAA06814 for ; Mon, 6 May 1996 14:19:55 +0200 (MET DST) 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 AA13222; Mon, 6 May 1996 14:19:13 +0200 (MET) Received: from suite.sw.oz.au by swallow.sw.oz.au with SMTP id MAA08474; Mon, 6 May 1996 12:14:38 GMT (8.6.10/Unixware) (from paul@suite.sw.oz.au for ) Received: by suite.sw.oz.au id AA03213; Mon, 6 May 1996 22:14:32 +1000 (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: <199605061214.AA03213@suite.sw.oz.au> Subject: NAPS now integrated to DECStation code... issues! To: linux-mips@fnet.fr (Linux MIPS mailing list) Date: Mon, 6 May 1996 22:14:32 +1000 (EST) Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 4055 Lines: 118 Hi folks, In email, Ralf says... > Hmm, I've fixed a whole bunch of bugs in my newest kernels so I really > suggest to use them. The kernel code has been reorganised compared with > 1.3.62 and this might - at least to a certain degree - make your life > easier though adding your patches into my kernel will be slightly more > work. Head.S is still not really reorganized, but we really should do > this with your machine and the SGI stuff to come in mind. This needs doing, but I also found other issues as I tried to integrate the new bitags stuff into the DECStation code... I needed to add bitags.c from Milo (I used .26), so that I could add tags prior to launching the kernel. This code will be generally usefull for any kernel that can be loaded via tftp (unless we write a tftp-capable version of Milo :-) or in any system with a primitive boot loader. Some bitags functions (for reading tags) are included in setup.c and should probably be in a separate file as this is a little clearer if nothing else... A lot of code still assumes gross amounts of pc-ism. The time.c file is one of them... the DEC RTC is PC-compatible, but is addressed as a flat set of registers... we need to modify the functions in the feature structure to allow reading and writing via functions instead of the half-baked macro/function approach currently used in CMOS macros. The tags for machine hierarchy still need some work. I propose the following: mach_group: DECStation ARC others... And that tag_machtype then be a different list of machines depending upon machgroup: DECStation: UNKNOWN 0 DS3100 1 DS5000/200 2 DS5000/100 3 DS5000/2x 7 ARC UNKNOWN 0 DESKSTATION_RPC44 1 DESKSTATION_TYNE 2 ACER_PICA_61 3 MIPS_MAGNUM_4000 4 OLIVETTI_M700 4 MIPS UNKNOWN 0 etc. I say this because it then lets me use the standard DEC machine type codes, and is notionally a better hierarchy. Where do I find the definition of the data in the drive_info structure to figure out what to put in it (apart from the zero I had to use to get the kernel to boot any further). As for the interrupt and DMA code... yuk!!! Now one thing I've noticed is that lots of these issues ought to have been covered in the SPARC port... but the current OS releases seem to be incomplete with regard to SPARC code. Is the SPARC source tree currently separate from the main Linux tree?? Should I get it to see what they've done?? The good news from all of the above is that I think the DEC code is a lot cleaner than it was, and now does the following correctly: Gets the system ID from the boot prom, and uses that to set: Machine type CPU type & revision Boot prom revision level Gets the available RAM size from the boot prom (rather than simply assuming 8MB :-) It also needs to be able to: Get default command line from boot prom environment variable Get default boot device from boot prom environment variable Get default console device from boot prom environment variable Lots of model-specific configuration... ...and to store them in DEC-specific tags. This is pretty easy though, as the new config code is now in C rather than assembler. All I need to do now is integrate all this into a 1.3.97 kernel and release it all! I also would like to hear from other owners of DECStations that have tried my test kernel. Please note that the pub/linux-mips directory is now mirrored as: private/Incoming/ftp.softway.com.au at fnet, so please use the server nearest/fastest to you... 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 Stoned.Elipot@univ-evry.fr Mon May 6 14:54:33 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 OAA12815 for ; Mon, 6 May 1996 14:54:32 +0200 (MET DST) Received: from amiante.univ-evry.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA14217; Mon, 6 May 1996 14:54:06 +0200 (MET) Received: from bandsept.univ-evry.fr (bandsept.univ-evry.fr [192.93.214.19]) by amiante.univ-evry.fr with ESMTP (8.6.12/96.04.09/louis-blanc); id OAA11450; Mon, 6 May 1996 14:54:27 +0200 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 OAA03422; Mon, 6 May 1996 14:53:41 +0200 Message-Id: <199605061253.OAA03422@bandsept.univ-evry.fr> To: linux-mips@fnet.fr Subject: Re: NAPS now integrated to DECStation code... issues! In-Reply-To: Your message of "Mon, 06 May 1996 22:14:32 +1000." <199605061214.AA03213@suite.sw.oz.au> X-It: bandsept (wrong OS, but right CPU) X-Mailer: MH 6.8.3, MH-E 5.0.2 Date: Mon, 06 May 1996 14:53:33 +0200 From: Stoned Elipot Content-Length: 1308 Lines: 42 Hi all, >>>>> "Paul" == Paul Antoine writes: [SNIP] Paul> The tags for machine hierarchy still need some work. I Paul> propose the following: Paul> mach_group: DECStation ARC others... Paul> And that tag_machtype then be a different list of machines Paul> depending upon machgroup: Paul> DECStation: UNKNOWN 0 DS3100 1 DS5000/200 2 DS5000/100 3 DS5000/2x 7 Paul> ARC UNKNOWN 0 DESKSTATION_RPC44 1 DESKSTATION_TYNE 2 ACER_PICA_61 3 MIPS_MAGNUM_4000 4 OLIVETTI_M700 4 Paul> MIPS UNKNOWN 0 Paul> etc. I say this because it then lets me use the standard DEC Paul> machine type codes, and is notionally a better hierarchy. This implies some tuning in various kernel files (asm-mips/floppy.h) but I think this is great. I'll try to integrate this in the patchs I'll send to Ralf tomorrow or the day after tomorrow (Yeah, I know I said I would release Milo 0.27 before the end of... last week. But I slept spleepy this week-end :) Paul, could you tell me more about the exact names/values of the cpp defs I guess you'd put in bootinfo.h ? Cheers, Stoned. From linux@mailhost.uni-koblenz.de Mon May 6 15:50:30 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 PAA16902 for ; Mon, 6 May 1996 15:50:29 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA15345; Mon, 6 May 1996 15:49:24 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id PAA02199 for ; Mon, 6 May 1996 15:48:58 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605061348.PAA02199@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id PAA10260; Mon, 6 May 1996 15:47:05 +0200 Subject: Re: NAPS now integrated to DECStation code... issues! To: linux-mips@fnet.fr Date: Mon, 6 May 1996 15:47:04 +0200 (MET DST) In-Reply-To: <199605061253.OAA03422@bandsept.univ-evry.fr> from "Stoned Elipot" at May 6, 96 02:53:33 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 721 Lines: 19 Hi, > Paul> etc. I say this because it then lets me use the standard DEC > Paul> machine type codes, and is notionally a better hierarchy. > > This implies some tuning in various kernel files (asm-mips/floppy.h) > but I think this is great. This file will probably end up as one sigle dispatcher into machine specific code - we just have far too much different hardware to support. > I'll try to integrate this in the patchs I'll send to Ralf tomorrow or > the day after tomorrow (Yeah, I know I said I would release Milo 0.27 > before the end of... last week. But I slept spleepy this week-end :) (Grin) This is what I promised/threatend to Linus also :-) People will love to get another big patch ... Ralf From linux@mailhost.uni-koblenz.de Mon May 6 16:16:21 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 QAA17313 for ; Mon, 6 May 1996 16:16:20 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA15938; Mon, 6 May 1996 16:14:44 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id QAA03349 for ; Mon, 6 May 1996 16:15:01 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605061415.QAA03349@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id QAA10294; Mon, 6 May 1996 16:13:09 +0200 Subject: Re: NAPS now integrated to DECStation code... issues! To: linux-mips@fnet.fr Date: Mon, 6 May 1996 16:13:08 +0200 (MET DST) In-Reply-To: <199605061214.AA03213@suite.sw.oz.au> from "Paul Antoine" at May 6, 96 10:14:32 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 3442 Lines: 87 Hi all, > > Hmm, I've fixed a whole bunch of bugs in my newest kernels so I really > > suggest to use them. The kernel code has been reorganised compared with > > 1.3.62 and this might - at least to a certain degree - make your life > > easier though adding your patches into my kernel will be slightly more > > work. Head.S is still not really reorganized, but we really should do > > this with your machine and the SGI stuff to come in mind. > > This needs doing, but I also found other issues as I tried to integrate > the new bitags stuff into the DECStation code... Well, so I herewith declare the discussion about the directory structure to be opened. How about organizing the tree as follows: arch/mips/kernel generic part of the kernel. arch/mips/lib arch/mips/mm arch/mips/mips1 ISA dependent code arch/mips/mips2 arch/mips/mips3 arch/mips/mips4 arch/mips/decstation arch/mips/decstation/boot arch/mips/decstation/mips1 machine and cpu specific code arch/mips/decstation/mips3 arch/mips/deskstation arch/mips/magnum arch/mips/magnum/boot arch/mips/magnum/mips1 special magnum & cpu specific code arch/mips/magnum/mips3 As you see there are two levels of directories with CPU specific code. We might search them via VPATH in Makefiles; code in the machine & model specific dirs will be found first and therefore has higher precedence than the generic code. Opinions? > Some bitags functions (for reading tags) are included in setup.c > and should probably be in a separate file as this is a little > clearer if nothing else... > A lot of code still assumes gross amounts of pc-ism. The time.c > file is one of them... the DEC RTC is PC-compatible, but is > addressed as a flat set of registers... we need to modify the > functions in the feature structure to allow reading and writing > via functions instead of the half-baked macro/function approach > currently used in CMOS macros. I'd like to burry the feature structure. The structure is already almost one screen page of code and it will get at least three or four times as big; there will also be aditional CPU dependencies and submodel dependencies. We should go back to plain and boring variables pointing to function or so. Given a reasonable organisation this will be better managable than that stoneage time 68k-ism struct feature. > The tags for machine hierarchy still need some work. I propose > the following: Ok, I can live with that. > Now one thing I've noticed is that lots of these issues ought to have > been covered in the SPARC port... but the current OS releases seem to > be incomplete with regard to SPARC code. Is the SPARC source tree > currently separate from the main Linux tree?? Should I get it to see > what they've done?? At least a large part of the Sparc kernel is now in Linus' tree and therefore also in the MIPS tree. > It also needs to be able to: > > Get default command line from boot prom environment variable > Get default boot device from boot prom environment variable > Get default console device from boot prom environment variable > Lots of model-specific configuration... > > ...and to store them in DEC-specific tags. This is pretty easy though, > as the new config code is now in C rather than assembler. Yes, we're getting ol'... > All I need to do now is integrate all this into a 1.3.97 kernel and > release it all! Ralf From wayne@Julia.DE Mon May 6 16:21:47 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 QAA17389 for ; Mon, 6 May 1996 16:21:46 +0200 (MET DST) Received: from alles.intern.julia.de ([194.221.49.2]) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA16173; Mon, 6 May 1996 16:21:04 +0200 (MET) Received: from bryant.intern.julia.de (wayne@bryant.intern.julia.de [194.221.49.34]) by alles.intern.julia.de (8.7.5/8.7.3) with ESMTP id QAA04149 for ; Mon, 6 May 1996 16:18:30 +0200 From: Wayne Hodgen Received: (from wayne@localhost) by bryant.intern.julia.de (8.7.5/8.7.3) id QAA10354 for linux-mips@fnet.fr; Mon, 6 May 1996 16:24:26 +0200 Message-Id: <199605061424.QAA10354@bryant.intern.julia.de> Subject: Procps binary package To: linux-mips@fnet.fr Date: Mon, 6 May 1996 16:24:26 +0200 (MET DST) X-Mailer: ELM [version 2.4 PL24 PGP6] Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1417 Lines: 26 Hi all, (repost due to maillost.uni-koblenz.de :( ) I got sick of traps using "ps aux" (and of listening to Ralf telling me to use "ps al" cos it still worked) so I got (some) of procps to compile (native on the Olli :) If anyone is interested I'll get Ralf to upload the archive (just now he says no-one has complained so no-one must need/want it ;) Make my day, prove him wrong >-) I was on holiday for a week or so and I'm catching up on the list. I saw several people posting about success with the Olli's, I had mine booting over the net 2 or 3 weeks ago. I put a 500MB IDE disk in it and after Ralf killed the unaligned problems this brought to light, now run from that: much better :) The console is still the worst problem though. I spoke to ralf and he's arranged to borrow the Olli docs from Andy (including a doc for the G384 I hope! I WANT A CURSOR!). Once I have this I'll start work on a new console. The main reason for the Olli crashing at the moment is memory accesses by the console screwing things up. Once we have a new console, the Olli's should be pretty stable... Wayne -- Wayne Hodgen | hodgen@informatik.uni-koblenz.de | #include From Stoned.Elipot@univ-evry.fr Mon May 6 16:28:29 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 QAA17561 for ; Mon, 6 May 1996 16:28:28 +0200 (MET DST) Received: from amiante.univ-evry.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA16328; Mon, 6 May 1996 16:28:04 +0200 (MET) Received: from bandsept.univ-evry.fr (bandsept.univ-evry.fr [192.93.214.19]) by amiante.univ-evry.fr with ESMTP (8.6.12/96.04.09/louis-blanc); id QAA11945; Mon, 6 May 1996 16:28:26 +0200 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 QAA03517; Mon, 6 May 1996 16:27:41 +0200 Message-Id: <199605061427.QAA03517@bandsept.univ-evry.fr> To: linux-mips@fnet.fr Subject: Re: NAPS now integrated to DECStation code... issues! In-Reply-To: Your message of "Mon, 06 May 1996 16:13:08 +0200." <199605061415.QAA03349@informatik.uni-koblenz.de> X-It: bandsept (wrong OS, but right CPU) X-Mailer: MH 6.8.3, MH-E 5.0.2 Date: Mon, 06 May 1996 16:27:33 +0200 From: Stoned Elipot Content-Length: 147 Lines: 4 >>>>> "Ralf" == Systemkennung Linux writes: Ralf> Opinions? arch/mips/magnum should be arch/mips/jazz, isn't it ? From miguel@roxanne.nuclecu.unam.mx Mon May 6 16:53:52 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 QAA17807 for ; Mon, 6 May 1996 16:53:52 +0200 (MET DST) Received: from roxanne.nuclecu.unam.mx by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA17028; Mon, 6 May 1996 16:53:26 +0200 (MET) Received: (from miguel@localhost) by roxanne.nuclecu.unam.mx (8.6.12/8.6.11) id JAA06182; Mon, 6 May 1996 09:52:47 -0500 Date: Mon, 6 May 1996 09:52:47 -0500 From: Miguel de Icaza Message-Id: <199605061452.JAA06182@roxanne.nuclecu.unam.mx> To: linux-mips@fnet.fr Cc: linux-mips@fnet.fr In-Reply-To: <199605061214.AA03213@suite.sw.oz.au> (paul@suite.sw.oz.au) Subject: Re: NAPS now integrated to DECStation code... issues! Mime-Version: 1.0 (generated by tm-edit 7.52) Content-Type: text/plain; charset=US-ASCII Content-Length: 597 Lines: 13 > Now one thing I've noticed is that lots of these issues ought to have > been covered in the SPARC port... but the current OS releases seem to > be incomplete with regard to SPARC code. Is the SPARC source tree > currently separate from the main Linux tree?? Should I get it to see > what they've done?? Linus tree is close to the Sparc tree. We have a couple of changes to the main kernel that have not been yet merged. You can look at them in vger.rutgers.edu:/pub/linux/Sparc/patches (we have patches relative to our last tree and relative to Linus latest kernel at the time). Miguel. From imp@rover.village.org Mon May 6 18:34:34 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 SAA00136 for ; Mon, 6 May 1996 18:34:31 +0200 (MET DST) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA19816; Mon, 6 May 1996 18:33:57 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.7.5/8.6.6) with SMTP id KAA00994 for ; Mon, 6 May 1996 10:33:44 -0600 (MDT) Message-Id: <199605061633.KAA00994@rover.village.org> To: linux-mips@fnet.fr Subject: Re: MILO on NEC RISCstation 2200 In-Reply-To: Your message of Sun, 05 May 1996 13:58:08 +0200 Date: Mon, 06 May 1996 10:33:43 -0600 From: Warner Losh Content-Length: 614 Lines: 15 : A reasonable heuristic is to query the ARC BIOS for the RAM, filter out : obvious wrong answers (like on the Tyne), the round up the end address of : the last chunk returned to the next multiple of two. Hmmm. think you could pass along some BIOS querry magic for those that might not have the docs to the BIOS :-) : Completly unrelated - the new libc 960501 is significantly better than : the old. I've just built a usable bash and do a native recompile of : the lib to get rid of some crosscompiltion artefacts. Cool. Your system is sounding stable if it is able to do that. libc isn't small :-) Warner From linux@mailhost.uni-koblenz.de Tue May 7 00:43:10 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 AAA05227 for ; Tue, 7 May 1996 00:43:09 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA01492; Tue, 7 May 1996 00:42:36 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id AAA18828 for ; Tue, 7 May 1996 00:42:50 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605062242.AAA18828@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id AAA10614; Tue, 7 May 1996 00:40:58 +0200 Subject: Re: NAPS now integrated to DECStation code... issues! To: linux-mips@fnet.fr Date: Tue, 7 May 1996 00:40:58 +0200 (MET DST) In-Reply-To: <199605061427.QAA03517@bandsept.univ-evry.fr> from "Stoned Elipot" at May 6, 96 04:27:33 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 184 Lines: 10 Hi, > >>>>> "Ralf" == Systemkennung Linux writes: > Ralf> Opinions? > > arch/mips/magnum should be arch/mips/jazz, isn't it ? Of course ... Ralf From munira@Crissy.Stanford.EDU Tue May 7 00:38:53 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 AAA05151 for ; Tue, 7 May 1996 00:38:52 +0200 (MET DST) Received: from crissy.Stanford.EDU by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA01354; Tue, 7 May 1996 00:38:23 +0200 (MET) Received: by Crissy.Stanford.EDU (8.6.8/25-LAGUNITA-eef) id PAA08029; Mon, 6 May 1996 15:42:36 -0700 From: Munira Rahemtulla Message-Id: <199605062242.PAA08029@Crissy.Stanford.EDU> Subject: Re: Linux/SGI coming To: linux-mips@fnet.fr Date: Mon, 6 May 96 15:42:36 PDT In-Reply-To: <199604271725.KAA23439@neteng.engr.sgi.com>; from "Larry McVoy" at Apr 27, 96 10:25 am X-Mailer: ELM [version 2.3 PL11] Content-Length: 580 Lines: 18 Hi, I'm working on a port of linux-mips to a hardware simulator that is currently running IRIX and I wanted to use whatever has been done for the SGI port as a starting point for my port. Could anyone tell me what the status of the SGI port is right now. How different is the source tree for the SGI from the source tree of the rest of linux-mips? Also, I was wondering if an ftp server to the sgi-linux account has been setup yet and if so, what the address is. Is the source tree for the SGI port available there? Thanks for your help. Munira munira@crissy.stanford.edu From lm@neteng.engr.sgi.com Tue May 7 00:56:40 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 AAA05351 for ; Tue, 7 May 1996 00:56:39 +0200 (MET DST) Received: from sgigate.SGI.COM by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA01865; Tue, 7 May 1996 00:56:11 +0200 (MET) Received: from cthulhu.engr.sgi.com by sgigate.sgi.com via ESMTP (950911.SGI.8.6.12.PATCH825/940406a.SGI) for <@sgigate.sgi.com:linux-mips@fnet.fr> id PAA11965; Mon, 6 May 1996 15:56:32 -0700 Received: from neteng.engr.sgi.com (neteng.engr.sgi.com [192.26.80.10]) by cthulhu.engr.sgi.com (950413.SGI.8.6.12/960327.SGI.AUTOCF) via ESMTP id PAA12716 for <@cthulhu.engr.sgi.com:linux-mips@fnet.fr>; Mon, 6 May 1996 15:56:31 -0700 Received: from localhost (lm@localhost) by neteng.engr.sgi.com (950413.SGI.8.6.12/960327.SGI.AUTOCF) via SMTP id PAA04905 for ; Mon, 6 May 1996 15:56:30 -0700 Message-Id: <199605062256.PAA04905@neteng.engr.sgi.com> To: linux-mips@fnet.fr From: lm@gate1-neteng.engr.sgi.com (Larry McVoy) Subject: Re: Linux/SGI coming Date: Mon, 06 May 1996 15:56:30 -0700 Sender: lm@neteng.engr.sgi.com Content-Length: 116 Lines: 4 The SGI port is non-existant right now. You are about 3 months early. Check back at the end of the summer. --lm From linux@mailhost.uni-koblenz.de Tue May 7 01:07:31 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 BAA05662 for ; Tue, 7 May 1996 01:07:30 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA02140; Tue, 7 May 1996 01:06:45 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id BAA22326 for ; Tue, 7 May 1996 01:07:02 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605062307.BAA22326@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id BAA10638; Tue, 7 May 1996 01:05:11 +0200 Subject: Re: MILO on NEC RISCstation 2200 To: linux-mips@fnet.fr Date: Tue, 7 May 1996 01:05:10 +0200 (MET DST) In-Reply-To: <199605061633.KAA00994@rover.village.org> from "Warner Losh" at May 6, 96 10:33:43 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 882 Lines: 21 Hi, > : Completly unrelated - the new libc 960501 is significantly better than > : the old. I've just built a usable bash and do a native recompile of > : the lib to get rid of some crosscompiltion artefacts. > > Cool. Your system is sounding stable if it is able to do that. libc > isn't small :-) You think this is *big*? I've already stripped all local symbols out of the archive. Otherwise libc.a would be far bigger. Indeed my system is relativly useable now. Crash reason #1 during the last time was my IDE disk, which I already had thrown away because of too many read errors ... I just had digged it out again because I don't own no other IDE disk. The whole thing is however still not 100% reliable. The stupid thing with the remaining bugs that crash the kernel is, that they rarely leave a useable information for debugging - but I'm working on 'em. Ralf From imp@rover.village.org Tue May 7 05:58:23 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 FAA10140 for ; Tue, 7 May 1996 05:58:22 +0200 (MET DST) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA14254; Tue, 7 May 1996 05:57:47 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.7.5/8.6.6) with SMTP id VAA03463 for ; Mon, 6 May 1996 21:57:58 -0600 (MDT) Message-Id: <199605070357.VAA03463@rover.village.org> To: linux-mips@fnet.fr Subject: Re: MILO on NEC RISCstation 2200 In-Reply-To: Your message of Tue, 07 May 1996 01:05:10 +0200 Date: Mon, 06 May 1996 21:57:58 -0600 From: Warner Losh Content-Length: 949 Lines: 20 : You think this is *big*? I've already stripped all local symbols out of : the archive. Otherwise libc.a would be far bigger. I was thinking more in terms of how many forks/execs of gcc you'd need to run thought, and the amount of the memory that they used might even push you into swapping. : Indeed my system is relativly useable now. Crash reason #1 during the : last time was my IDE disk, which I already had thrown away because of : too many read errors ... I just had digged it out again because I don't : own no other IDE disk. The whole thing is however still not 100% : reliable. The stupid thing with the remaining bugs that crash the kernel : is, that they rarely leave a useable information for debugging - but I'm : working on 'em. Gotta get one of them there SCSI drives :-). What's an IDE drive doing on that fast MIPS box :-) I recently scammed the use of a 320M hard drive, but haven't put anything on it yet.... Warner From paul@suite.sw.oz.au Tue May 7 13:22:10 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 NAA16817 for ; Tue, 7 May 1996 13:22:09 +0200 (MET DST) Received: from staff.cs.su.OZ.AU by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA24681; Tue, 7 May 1996 13:21:38 +0200 (MET) Received: from suite.sw.oz.au by swallow.sw.oz.au with SMTP id LAA01517; Tue, 7 May 1996 11:21:36 GMT (8.6.10/Unixware) (from paul@suite.sw.oz.au for ) Received: by suite.sw.oz.au id AA13653; Tue, 7 May 1996 21:21:33 +1000 (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: <199605071121.AA13653@suite.sw.oz.au> Subject: Re: NAPS now integrated to DECStation code... issues! To: linux-mips@fnet.fr Date: Tue, 7 May 1996 21:21:32 +1000 (EST) In-Reply-To: <199605061452.JAA06182@roxanne.nuclecu.unam.mx> from "Miguel de Icaza" at May 6, 96 09:52:47 am Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 1833 Lines: 38 Miguel, > > Now one thing I've noticed is that lots of these issues ought to have > > been covered in the SPARC port... but the current OS releases seem to > > be incomplete with regard to SPARC code. Is the SPARC source tree > > currently separate from the main Linux tree?? Should I get it to see > > what they've done?? > > Linus tree is close to the Sparc tree. We have a couple of changes to > the main kernel that have not been yet merged. You can look at them > in vger.rutgers.edu:/pub/linux/Sparc/patches (we have patches relative > to our last tree and relative to Linus latest kernel at the time). Yeah - I just patched my development environment machine up to the 1.3.98 level, and noticed that plenty of the SPARC code has changed. I see that the sbus is now catered for, and that the serial, frame buffer and keyboard drivers are now under drivers/sbus/char. This is the particular code I was interested in because the DECStation is going to need something similar for the TURBOChannel devices. I also wanted to know that someone else had done a Z8530 async. serial driver etc... (call me lazy, or at least just wanting some consistancy amongst the architectures :-) I gather your offer of modifying sunlance.c into a DEC version is still on offer?? (We still have lots of interrupt and DMA code to do before you can do that though... I think.) 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 Tue May 7 13:35:14 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 NAA16950 for ; Tue, 7 May 1996 13:35:13 +0200 (MET DST) Received: from staff.cs.su.OZ.AU by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA24837; Tue, 7 May 1996 13:34:44 +0200 (MET) Received: from suite.sw.oz.au by swallow.sw.oz.au with SMTP id LAA01865; Tue, 7 May 1996 11:34:53 GMT (8.6.10/Unixware) (from paul@suite.sw.oz.au for ) Received: by suite.sw.oz.au id AA13699; Tue, 7 May 1996 21:34:51 +1000 (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: <199605071134.AA13699@suite.sw.oz.au> Subject: Re: The great namespace debate (was NAPS in DECStation) To: linux-mips@fnet.fr Date: Tue, 7 May 1996 21:34:50 +1000 (EST) In-Reply-To: <199605061415.QAA03349@informatik.uni-koblenz.de> from "Systemkennung Linux" at May 6, 96 04:13:08 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: 2544 Lines: 75 Hi folks, Ralf wrote: > Well, so I herewith declare the discussion about the directory structure > to be opened. How about organizing the tree as follows: > > arch/mips/kernel generic part of the kernel. > arch/mips/lib > arch/mips/mm > arch/mips/mips1 ISA dependent code > arch/mips/mips2 > arch/mips/mips3 > arch/mips/mips4 > arch/mips/decstation > arch/mips/decstation/boot > arch/mips/decstation/mips1 machine and cpu specific code > arch/mips/decstation/mips3 > arch/mips/deskstation > arch/mips/magnum > arch/mips/magnum/boot > arch/mips/magnum/mips1 special magnum & cpu specific code > arch/mips/magnum/mips3 > > As you see there are two levels of directories with CPU specific code. We > might search them via VPATH in Makefiles; code in the machine & model > specific dirs will be found first and therefore has higher precedence than > the generic code. > > Opinions? Hmmm... I hope we don't need another level to the hierarchy... e.g.: arch/mips/dec/kernel arch/mips/dec/pmax arch/mips/dec/kmin arch/mips/arc/kernel arch/mips/arc/magnum arch/mips/arc/olivetti arch/mips/mips/kernel arch/mips/mips/magnum ...I'm not sure but my brain isn't functioning well enough to say specifically why... and it therefore may be the completely unnecessary fabrication of a tired mind. > I'd like to burry the feature structure. The structure is already almost > one screen page of code and it will get at least three or four times as > big; there will also be aditional CPU dependencies and submodel dependencies. > We should go back to plain and boring variables pointing to function or so. > Given a reasonable organisation this will be better managable than that > stoneage time 68k-ism struct feature. O.K. I can live with that :-) > > as the new config code is now in C rather than assembler. > > Yes, we're getting ol'... Hmmpfhhh... speak for yourself! Call me old if you wish - I just feel it's silly to code in assembler just for the pain of it... :-) Paul _______________________________________________________________________________ Paul M. Antoine, Net: paul@sw.oz.au Softway Pty Ltd WWW: www.softway.com.au PO Box 305, Strawberry Hills, NSW 2012, Australia Tel: +61 2 698 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 Tue May 7 13:57:12 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 NAA17147 for ; Tue, 7 May 1996 13:57:12 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA25304; Tue, 7 May 1996 13:56:37 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id NAA06630 for ; Tue, 7 May 1996 13:56:37 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605071156.NAA06630@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id NAA11515; Tue, 7 May 1996 13:54:43 +0200 Subject: Re: MILO on NEC RISCstation 2200 To: linux-mips@fnet.fr Date: Tue, 7 May 1996 13:54:43 +0200 (MET DST) In-Reply-To: <199605070357.VAA03463@rover.village.org> from "Warner Losh" at May 6, 96 09:57:58 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 481 Lines: 12 Hi, > Gotta get one of them there SCSI drives :-). What's an IDE drive > doing on that fast MIPS box :-) I recently scammed the use of a 320M > hard drive, but haven't put anything on it yet.... Oh, first IDE is working and second a IDE drive IS usefull when you write SCSI stuff. Mess up the SCSI subsystem but IDE usually still continues to work. Besides - if Mike Neuffer is at the Uni today I'll fix SCSI & DPT. Should hopefully work tomorrow (famous last ...) Ralf From linux@mailhost.uni-koblenz.de Tue May 7 14:04:59 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 OAA17410 for ; Tue, 7 May 1996 14:04:58 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA25440; Tue, 7 May 1996 14:03:01 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id NAA06667 for ; Tue, 7 May 1996 13:59:28 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605071159.NAA06667@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id NAA11526; Tue, 7 May 1996 13:57:35 +0200 Subject: Re: The great namespace debate (was NAPS in DECStation) To: linux-mips@fnet.fr Date: Tue, 7 May 1996 13:57:34 +0200 (MET DST) In-Reply-To: <199605071134.AA13699@suite.sw.oz.au> from "Paul Antoine" at May 7, 96 09:34:50 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 214 Lines: 9 Hi, > Call me old if you wish - I just feel it's silly to code in assembler > just for the pain of it... :-) People that do something just for the pain usually go to some other "whip-me-type" place ;-) Ralf From miguel@roxanne.nuclecu.unam.mx Tue May 7 18:04:03 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 SAA23490 for ; Tue, 7 May 1996 18:04:02 +0200 (MET DST) Received: from roxanne.nuclecu.unam.mx by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA00744; Tue, 7 May 1996 18:03:26 +0200 (MET) Received: (from miguel@localhost) by roxanne.nuclecu.unam.mx (8.6.12/8.6.11) id LAA15464; Tue, 7 May 1996 11:02:20 -0500 Date: Tue, 7 May 1996 11:02:20 -0500 From: Miguel de Icaza Message-Id: <199605071602.LAA15464@roxanne.nuclecu.unam.mx> To: linux-mips@fnet.fr Cc: linux-mips@fnet.fr In-Reply-To: <199605071121.AA13653@suite.sw.oz.au> (paul@suite.sw.oz.au) Subject: Re: NAPS now integrated to DECStation code... issues! Mime-Version: 1.0 (generated by tm-edit 7.52) Content-Type: text/plain; charset=US-ASCII Content-Length: 618 Lines: 14 > I see that the sbus is now catered for, and that the serial, frame > buffer and keyboard drivers are now under drivers/sbus/char. This is > the particular code I was interested in because the DECStation is going > to need something similar for the TURBOChannel devices. I wonder how devices work on the DECstation, I guess I have to go and take a look at the NetBSD stuff :-) > I also wanted to know that someone else had done a Z8530 > async. serial driver etc... (call me lazy, or at least just wanting > some consistancy amongst the architectures :-) Let's say it's consistency with other architectures. From imp@rover.village.org Tue May 7 18:46:14 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 SAA23958 for ; Tue, 7 May 1996 18:46:13 +0200 (MET DST) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA01934; Tue, 7 May 1996 18:46:09 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.7.5/8.6.6) with SMTP id KAA05738 for ; Tue, 7 May 1996 10:46:04 -0600 (MDT) Message-Id: <199605071646.KAA05738@rover.village.org> To: linux-mips@fnet.fr Subject: Re: NAPS now integrated to DECStation code... issues! In-Reply-To: Your message of Tue, 07 May 1996 11:02:20 CDT Date: Tue, 07 May 1996 10:46:03 -0600 From: Warner Losh Content-Length: 413 Lines: 10 : I wonder how devices work on the DECstation, I guess I have to go and : take a look at the NetBSD stuff :-) I'm not sure the pmax NetBSD port compiles these days. I've had problems with their PICA port. The OpenBSD PICA port seems to be at least to the point it compiles (I don't have a pica to test it on), but I've not tried the pmax port in OpenBSD. Either should be good for reference, however. Warner From linux@mailhost.uni-koblenz.de Tue May 7 19:15:25 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 TAA24400 for ; Tue, 7 May 1996 19:15:25 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA02642; Tue, 7 May 1996 19:15:13 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id TAA20686 for ; Tue, 7 May 1996 19:14:52 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605071714.TAA20686@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id TAA16090; Tue, 7 May 1996 19:12:59 +0200 Subject: Re: NAPS now integrated to DECStation code... issues! To: linux-mips@fnet.fr Date: Tue, 7 May 1996 19:12:58 +0200 (MET DST) In-Reply-To: <199605071646.KAA05738@rover.village.org> from "Warner Losh" at May 7, 96 10:46:03 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 578 Lines: 15 Hi, > : I wonder how devices work on the DECstation, I guess I have to go and > : take a look at the NetBSD stuff :-) > > I'm not sure the pmax NetBSD port compiles these days. I've had > problems with their PICA port. The OpenBSD PICA port seems to be at > least to the point it compiles (I don't have a pica to test it on), > but I've not tried the pmax port in OpenBSD. Either should be good > for reference, however. NetBSD has been ported to PICA, not OpenBSD. But were have a person on this list who should be able to give an authoritative answer (Hi Per!) Ralf From linux@mailhost.uni-koblenz.de Tue May 7 19:59:06 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 TAA24768 for ; Tue, 7 May 1996 19:59:05 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA03321; Tue, 7 May 1996 19:58:52 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id TAA21840 for ; Tue, 7 May 1996 19:58:48 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605071758.TAA21840@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id TAA16158; Tue, 7 May 1996 19:56:54 +0200 Subject: More shells (announcement) To: linux-mips@fnet.fr Date: Tue, 7 May 1996 19:56:54 +0200 (MET DST) Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 631 Lines: 22 -----BEGIN PGP SIGNED MESSAGE----- Hi all, here is the help for all the people that didn't like ash superior command line editing capabilities. As usual you can get the binaries from ftp.fnet.fr:/linux-mips/ and it's mirrors. Ralf 21bc47c9b03d8f3e026e2b2421fa181e bash-1.14.2.tar.gz d320bfee70b110f1601dc42eafc8731c tcsh-6.06.tar.gz -----BEGIN PGP SIGNATURE----- Version: 2.6.2 iQCVAgUBMY+O40ckbl6vezDBAQFyrgP+NcLjORaHbVljGeUka3OKqaWMm9q6tXiw 3F8oUW2Rr4DBXjEhEoa/LPxqd8ZWbz4k1AAwiNn6JNsIukH+L/0NfuSBobzavbEG VwOuFuFzQ3YxE91v1rlFfayonciKigEyqOxvEagcIHmh2GzI29sieq7UqDRGVUEt 62CuuVIiU2M= =HG/W -----END PGP SIGNATURE----- From afritz@starlink.com Wed May 8 08:47:53 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 IAA05265 for ; Wed, 8 May 1996 08:47:51 +0200 (MET DST) Received: from orion.starlink.com by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA14561; Wed, 8 May 1996 08:47:47 +0200 (MET) Received: from afritz.starlink.com ([206.26.42.254]) by orion.starlink.com (8.6.9/8.6.9) with SMTP id XAA06385 for ; Tue, 7 May 1996 23:38:57 -0700 Message-Id: <199605080638.XAA06385@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: Tue, 07 May 1996 11:50:56 -0700 To: linux-mips@fnet.fr From: Adam Fritzler Subject: SGI UltraP... Content-Length: 308 Lines: 8 This really has no relavence, but I was reading that press release that your site links to about SGI's UltraP Pentium-Replacement boards. That press release mentioned no one actually making these boards. Did anyone ever produce them? (They sounded interesting :) Thanx, Adam Fritzler afritz@starlink.com From linux@mailhost.uni-koblenz.de Tue May 7 21:13:34 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 VAA25775 for ; Tue, 7 May 1996 21:13:34 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA04530; Tue, 7 May 1996 21:13:07 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id VAA23691 for ; Tue, 7 May 1996 21:13:05 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605071913.VAA23691@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id VAA16265; Tue, 7 May 1996 21:11:11 +0200 Subject: GCC crosscompiler problem on Alpha To: linux-mips@fnet.fr Date: Tue, 7 May 1996 21:11:11 +0200 (MET DST) Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 542 Lines: 14 Hi all, someone reported to me problems with his crosscompiler installation running on Linux/Alpha. The problem is related to the type long being 64 bit on Alphas and results in GCC generating bad code which can most easily be recognices by huge (bigger than 2 << 32) constants in the code. lbu $3,color_table+4294967266($3) or comparing the code. Has anyone else has such problems? Does this problem also appear when builing the crosscompiler binaries on other operating systems (eg using the 64bit ABI on IRIX 6?) Ralf From imp@rover.village.org Tue May 7 21:22:09 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 VAA25942 for ; Tue, 7 May 1996 21:22:09 +0200 (MET DST) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA04639; Tue, 7 May 1996 21:22:04 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.7.5/8.6.6) with SMTP id NAA06193 for ; Tue, 7 May 1996 13:22:01 -0600 (MDT) Message-Id: <199605071922.NAA06193@rover.village.org> To: linux-mips@fnet.fr Subject: Re: NAPS now integrated to DECStation code... issues! In-Reply-To: Your message of Tue, 07 May 1996 19:12:58 +0200 Date: Tue, 07 May 1996 13:22:01 -0600 From: Warner Losh Content-Length: 694 Lines: 17 : NetBSD has been ported to PICA, not OpenBSD. But were have a person : on this list who should be able to give an authoritative answer (Hi Per!) Hmmm, Per checked in changes to the OpenBSD tree and sent me mail when he was done a while back. I think it is a case that both of them have been ported. I just updated this morning to NetBSD and there are a few pica rleated changes there too, so maybe the config/compile problems I've had are gone. I've been able to build bsd kernels for the pica in the OpenBSD tree that I've been grabbing, so I'm fairly sure that this is working there. Well, it is working for picas, but not for, say, Deskstations :-) Per, have I lost my mind? Warner From imp@rover.village.org Tue May 7 21:25:00 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 VAA26000 for ; Tue, 7 May 1996 21:25:00 +0200 (MET DST) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA04660; Tue, 7 May 1996 21:24:53 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.7.5/8.6.6) with SMTP id NAA06219 for ; Tue, 7 May 1996 13:24:50 -0600 (MDT) Message-Id: <199605071924.NAA06219@rover.village.org> To: linux-mips@fnet.fr Subject: Re: More shells (announcement) In-Reply-To: Your message of Tue, 07 May 1996 19:56:54 +0200 Date: Tue, 07 May 1996 13:24:50 -0600 From: Warner Losh Content-Length: 238 Lines: 7 : 21bc47c9b03d8f3e026e2b2421fa181e bash-1.14.2.tar.gz : d320bfee70b110f1601dc42eafc8731c tcsh-6.06.tar.gz Bless you. Now I can use the one true shell. I'll not start a flame war by saying which one of these is that shell :-) Warner From tor@spacetec.no Wed May 8 00:50:44 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 AAA28293 for ; Wed, 8 May 1996 00:50:42 +0200 (MET DST) Received: from pallas.spacetec.no by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA07822; Wed, 8 May 1996 00:50:31 +0200 (MET) Received: (from tor@localhost) by pallas.spacetec.no (8.7.1/8.7.1) id AAA29293 for linux-mips@fnet.fr; Wed, 8 May 1996 00:56:46 +0200 Message-Id: <199605072256.AAA29293@pallas.spacetec.no> From: tor@spacetec.no (Tor Arntsen) Date: Wed, 8 May 1996 00:56:44 +0200 In-Reply-To: tor (Tor Arntsen) "m700-10 success" (May 2, 10:50) X-Mailer: Mail User's Shell (7.2.5 10/14/92) To: linux-mips@fnet.fr Subject: Re: m700-10 success Content-Length: 2735 Lines: 50 I just got a few minutes free so I can follow up on my own mail a week ago: On May 2, 10:50, I wrote about my Olivetti & 1.3.96: >My 3c503 ethernet board works fine now, in PIO mode. There's still something >wrong with the shared mem detection, I'll come back to that when I get time First, I had to patch io.h a little (mismatched parentheses) in order to be able to compile 3c503.c. --- include/asm-mips/io.h.orig Sun Mar 24 23:45:49 1996 +++ include/asm-mips/io.h Wed May 1 14:00:01 1996 @@ -79,13 +79,13 @@ * differently. On the x86 architecture, we just read/write the * memory location directly. */ -#define readb(addr) (*(volatile unsigned char *) ((isa_slot_offset + (unsigned long)addr)) -#define readw(addr) (*(volatile unsigned short *) ((isa_slot_offset + (unsigned long)addr)) -#define readl(addr) (*(volatile unsigned int *) ((isa_slot_offset + (unsigned long)addr)) +#define readb(addr) (*(volatile unsigned char *) ((isa_slot_offset + (unsigned long)addr))) +#define readw(addr) (*(volatile unsigned short *) ((isa_slot_offset + (unsigned long)addr))) +#define readl(addr) (*(volatile unsigned int *) ((isa_slot_offset + (unsigned long)addr))) -#define writeb(b,addr) ((*(volatile unsigned char *) ((isa_slot_offset + (unsigned long)addr)) = (b)) -#define writew(b,addr) ((*(volatile unsigned short *) ((isa_slot_offset + (unsigned long)addr)) = (b)) -#define writel(b,addr) ((*(volatile unsigned int *) ((isa_slot_offset + (unsigned long)addr)) = (b)) +#define writeb(b,addr) ((*(volatile unsigned char *) ((isa_slot_offset + (unsigned long)addr)) = (b))) +#define writew(b,addr) ((*(volatile unsigned short *) ((isa_slot_offset + (unsigned long)addr)) = (b))) +#define writel(b,addr) ((*(volatile unsigned int *) ((isa_slot_offset + (unsigned long)addr)) = (b))) #define memset_io(a,b,c) memset((void *)(isa_slot_offset + (unsigned long)a),(b),(c)) #define memcpy_fromio(a,b,c) memcpy((a),(void *)(isa_slot_offset + (unsigned long)b),(c)) (If that's they way it's meant to be or not is not completely clear.. see below) Secondly, when the 3c503.c probe code tries to use readb(0xddffe) it fails with a kernel error (right now I can't provide the exact error, but I can re- create the error and write it down later if needed). So what I had to do to get up and running was to disable the part of the code in 3c503.c (el2_probe) that probes for shared memory. This is because the 3c503 code tries shared memory probing first before it falls back to ioport probing. The code piece I'm talking about is the one that starts with for (addr = addrs; *addr; addr++) { int i; unsigned int base_bits = readb(*addr); Leaving out that things (ie ioport & irq probing) went fine. -Tor From tor@spacetec.no Wed May 8 01:08:12 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 BAA28649 for ; Wed, 8 May 1996 01:08:12 +0200 (MET DST) Received: from pallas.spacetec.no by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA08108; Wed, 8 May 1996 01:08:06 +0200 (MET) Received: (from tor@localhost) by pallas.spacetec.no (8.7.1/8.7.1) id BAA29311 for linux-mips@fnet.fr; Wed, 8 May 1996 01:14:27 +0200 Message-Id: <199605072314.BAA29311@pallas.spacetec.no> From: tor@spacetec.no (Tor Arntsen) Date: Wed, 8 May 1996 01:14:22 +0200 In-Reply-To: Stoned Elipot "Re: m700-10 success" (May 2, 11:30) X-Mailer: Mail User's Shell (7.2.5 10/14/92) To: linux-mips@fnet.fr Subject: Re: m700-10 success Content-Length: 1643 Lines: 36 (More late followup. Sorry about the late answer, too busy at work right now, should be better later I hope) On May 2, 11:30, Stoned Elipot wrote: >And please could you tell me more about the corss-compiler env you use >to compile binaries (which lib, which sources, or which binaries from >Ralf...) as I wasn't able to make ash (home compilation) run. Thanks. I'm using the mips-sgi-irix5/mipsel-linux (for compiling the kernel) and the mips-sgi-irixt/mipsel-linuxelf (for userland) cross compiler tools, the same ones as are now at ftp.fnet.fr in the crossdev directory. The libc I used (I have compiled just a little) was the libc-951212 mipsel-linuxelf version, although the header files were include-960120. I have no idea where I got those from.. it doesn't matter much now anyway, as Ralf has provided a new libc. The header files I installed in several places: /usr/include (as seen by the oli), and also in (just a link) /usr/mipsel-linuxelf/include just for good measure. I also put them in /usr/local/mipsel-linuxelf/include directory (as seen by the cross-compiling host). Then it's important to fix up the include/linux and include/asm symbolic links everywhere, no use having them to point to Ralf's home directory :-) With this installed I could compile relatively easy, just by specifying gcc -b mipsel-linuxelf on the SGI (my cross-compiling host). For binaries I use root-0.01.tar.gz from ftp.fnet.fr, I have shuffled a bit around with some stuff (particularly the rc scripts and /etc things) to get the network and nfsboot/mount scenario that I want, but this is not really of much use for anyone else. -Tor From bigbug.franken.de!tsbogend@hub-fue.franken.de Wed May 8 01:48:42 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 BAA28983 for ; Wed, 8 May 1996 01:48:41 +0200 (MET DST) Received: from rachael.franken.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA08596; Wed, 8 May 1996 01:48:38 +0200 (MET) Received: from hub-fue.franken.de by rachael.franken.de with smtp (Smail3.1.29.1 #8) id m0uGwUf-000oKiC; Wed, 8 May 96 01:48 MET DST Received: by hub-fue.franken.de (Smail3.1.29.1 #35) id m0uGwUY-0005smC; Wed, 8 May 96 01:48 MET DST Received: by bigbug.franken.de (8.6.12/tbs) id BAA01934 for linux-mips@fnet.fr; Wed, 8 May 1996 01:46:27 +0200 From: Thomas Bogendoerfer Message-Id: <199605072346.BAA01934@bigbug.franken.de> Subject: Re: GCC crosscompiler problem on Alpha To: linux-mips@fnet.fr Date: Wed, 8 May 1996 01:46:26 +0200 (MET DST) In-Reply-To: <199605071913.VAA23691@informatik.uni-koblenz.de> from "Systemkennung Linux" at May 7, 96 09:11:11 pm Reply-To: tsbogend@bigbug.franken.de X-Mailer: ELM [version 2.4 PL24 ME8b] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 2142 Lines: 98 > someone reported to me problems with his crosscompiler installation running that's me.-) > on Linux/Alpha. The problem is related to the type long being 64 bit on > Alphas and results in GCC generating bad code which can most easily be > recognices by huge (bigger than 2 << 32) constants in the code. that's the obvious case, but the code is much worse. Here is a example: mm/memory.c compiled on a 386: .ent oom oom: .frame $sp,48,$31 # vars= 0, regs= 2/0, args= 32, extra= 0 .mask 0x80010000,-8 .fmask 0x00000000,0 lw $5,current_set subu $sp,$sp,48 sd $16,32($sp) move $16,$4 la $4,$LC2 sd $31,40($sp) .set noreorder .set nomacro jal printk addu $5,$5,498 .set macro .set reorder lw $2,940($16) li $4,0x00000009 # 9 move $5,$16 sw $0,268($2) lw $2,16($16) li $3,-257 # 0xfffffeff li $6,0x00000001 # 1 and $2,$2,$3 .set noreorder .set nomacro jal send_sig sw $2,16($5) .set macro .set reorder [....] and now on the alpha: oom: .frame $sp,48,$31 # vars= 0, regs= 2/0, args= 32, extra= 0 .mask 0x80010000,-8 .fmask 0x00000000,0 lw $5,current_set subu $sp,$sp,48 sd $16,32($sp) move $16,$4 la $4,$LC2 sd $31,40($sp) .set noreorder .set nomacro jal printk addu $5,$5,498 .set macro .set reorder lw $2,940($16) li $3,0x00000000ffff0000 # 4294901760 ori $3,$3,0xfeff li $4,0x0000000000000009 # 9 sw $0,200($2) lw $2,16($16) move $5,$16 li $6,0x0000000000000001 # 1 and $2,$2,$3 .set noreorder .set nomacro jal send_sig sw $2,16($5) .set macro .set reorder [...] I debugged gcc one day, and it looks like there are lot of conversion HOST_WIDE_INT -> 32bit unsigned int -> 32bit signed int. And because on the alpha HOST_WIDE_INT is 64bit wide, the strange code above is generated. Today I have done a quick hack, and compiled the cross compiler with #define HOST_WIDE_INT int (which is 32bit wide). Now the code looks much better, but there are still minor differences. Any other hints, what I can do ? Thomas. -- That process _deserves_ to die ("My name is Linus Torvalds, prepare to die"). [Linus Torvalds on linux-kernel] From linux@mailhost.uni-koblenz.de Wed May 8 06:19:07 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 GAA03697 for ; Wed, 8 May 1996 06:19:06 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA12933; Wed, 8 May 1996 06:19:01 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id GAA18891 for ; Wed, 8 May 1996 06:19:00 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605080419.GAA18891@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id GAA18152; Wed, 8 May 1996 06:17:06 +0200 Subject: Re: m700-10 success To: linux-mips@fnet.fr Date: Wed, 8 May 1996 06:17:06 +0200 (MET DST) In-Reply-To: <199605072314.BAA29311@pallas.spacetec.no> from "Tor Arntsen" at May 8, 96 01:14:22 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1922 Lines: 42 Hi, > Then it's important to fix up the include/linux and include/asm symbolic > links everywhere, no use having them to point to Ralf's home directory :-) This is just the urgent request to give me an account on every Linux/MIPS machine ... > For binaries I use root-0.01.tar.gz from ftp.fnet.fr, I have shuffled a bit > around with some stuff (particularly the rc scripts and /etc things) to get > the network and nfsboot/mount scenario that I want, but this is not really > of much use for anyone else. Well, this archive was clearly just a snapshot of my development source tree as it was at that point. Looks like a mess but still usefull. I'll try to replace the userland in root-0.01.tar.gz with lots of smaller archives as a first step before we have a real Linux distribution. Well, distributions are a special thing with Linux/MIPS anyway. We're about to endup with two complete binary trees - one little and one big endian tree. I think this might be a real problem with some FTP servers mirroring Linux/MIPS. Two complete packages for one not that popular architecture could affect the spreading negativly. Being little endian many of use don't have the advantage of conforming to the ABI, exchanging binaries gets difficult. For this reasons I'm really thinking hard about making a bi-endian kernel. This is however Pandora's box and I'd like to keep it closed ... Some time ago I've already done some successfull experiments with big endian usercode on my Acer. This however was a tiny program; real world apps would require us to change large parts of the kernel. In particular ioctl's will be fun. The only thing that accounts positivly for us is that enforced by Intel's "maladia segmenti" almost every useraccess goes through get_user() and put_user() which'd be very usefull hooks for making Linux bi-endian. So there is a decission we have to make. Ralf PS: Little endian sucks. From linux@mailhost.uni-koblenz.de Wed May 8 06:28:41 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 GAA03797 for ; Wed, 8 May 1996 06:28:40 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA13105; Wed, 8 May 1996 06:28:36 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id GAA19088 for ; Wed, 8 May 1996 06:28:34 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605080428.GAA19088@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id GAA18168; Wed, 8 May 1996 06:26:41 +0200 Subject: Re: m700-10 success To: linux-mips@fnet.fr Date: Wed, 8 May 1996 06:26:40 +0200 (MET DST) In-Reply-To: <199605072256.AAA29293@pallas.spacetec.no> from "Tor Arntsen" at May 8, 96 00:56:44 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1681 Lines: 42 Hi, > On May 2, 10:50, I wrote about my Olivetti & 1.3.96: > >My 3c503 ethernet board works fine now, in PIO mode. There's still something > >wrong with the shared mem detection, I'll come back to that when I get time > > First, I had to patch io.h a little (mismatched parentheses) in order to be > able to compile 3c503.c. Ok, patch is in. There was some bug in the interrupt routine that made the interrupt detection for (E)ISA cards sometimes fail. Change your interrupt routine in arch/mips/kernel/entry.S to like as follows: /* * Don't return & unblock the pic */ LEAF(bad_interrupt) .set reorder lw t0,%lo(intr_count)(s3) .set noreorder subu t0,1 j return sw t0,%lo(intr_count)(s3) # delay slot END(bad_interrupt) The old routine did unblock the PIC chips and therefore wipe out the interrupt's traces that the interrupt detection uses. This was the cause for the strange boot messages when using IDE disk. > So what I had to do to get up and running was to disable the part of the code > in 3c503.c (el2_probe) that probes for shared memory. This is because the > 3c503 code tries shared memory probing first before it falls back to ioport > probing. The code piece I'm talking about is the one that starts with This is a bit weired because the address the the readl() accesses should be correctly mapped in the TLB. Could you send me your crash messages, please? Have to find one of these 3c503 at the university. I'm living inNE2000 & Sonic-land ... Ralf From ai.kharkov.ua!sny@citynet.kharkov.ua Wed May 8 09:33:35 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 JAA05846 for ; Wed, 8 May 1996 09:33:35 +0200 (MET DST) Received: from xs4all.freenet.kiev.ua by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA15133; Wed, 8 May 1996 09:33:27 +0200 (MET) Received: from citynet.kharkov.ua (root@[194.44.6.252]) by xs4all.freenet.kiev.ua (8.6.13/sol2.4xla) with ESMTP id KAA22536 for ; Wed, 8 May 1996 10:29:02 +0300 Received: by citynet.kharkov.ua id KAA02159; (8.6.10/vak/1.8r) Wed, 8 May 1996 10:15:42 +0300 Received: from alpha.ai.kharkov.ua(194.44.6.247) by citynet.kharkov.ua via smap (V1.3) id sma002152; Wed May 8 10:15:14 1996 Received: by alpha.ai.kharkov.ua; (5.65/1.1.8.2/30Jan96-1124AM) id AA15457; Wed, 8 May 1996 10:20:38 +0200 Received: from localhost(127.0.0.1) by alpha.ai.kharkov.ua via smap (V1.3) id sma015451; Wed May 8 10:20:28 1996 Sender: sny@ai.kharkov.ua Message-Id: <3190594B.41C6@ai.kharkov.ua> Date: Wed, 08 May 1996 10:20:27 +0200 From: Sergei Yemelyantev Organization: KHAI X-Mailer: Mozilla 2.01 (X11; I; OSF1 V3.2 alpha) Mime-Version: 1.0 To: linux-mips@fnet.fr Subject: I have Mips Box X-Url: http://lena.fnet.fr/ Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Length: 225 Lines: 9 Dear Sir ! I,m writing to you from Ukraine (former USSR rebublic). I have a good deal of VaxStation 3100 and VaxStation 4000. Can I run Linux MIPS on these computers ? Yours sincerely S.N. Yemeliantsev From linux@mailhost.uni-koblenz.de Wed May 8 16:13:24 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 QAA11133 for ; Wed, 8 May 1996 16:13:24 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA26466; Wed, 8 May 1996 16:13:15 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id QAA09483 for ; Wed, 8 May 1996 16:13:07 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605081413.QAA09483@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id QAA22176; Wed, 8 May 1996 16:11:14 +0200 Subject: The bad news ... To: linux-mips@fnet.fr Date: Wed, 8 May 1996 16:11:13 +0200 (MET DST) Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 250 Lines: 8 is that I now have not one SCSI driver running. The good news is that I now have two of 'em running :-) So far EATA DMA and EATA PIO are running. I'll now swap the hostadapters of my Acer and my 486 and make that Adaptec 1542 running ... Ralf From imp@rover.village.org Wed May 8 16:31:26 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 QAA11455 for ; Wed, 8 May 1996 16:31:25 +0200 (MET DST) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA26745; Wed, 8 May 1996 16:31:15 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.7.5/8.6.6) with SMTP id IAA10551 for ; Wed, 8 May 1996 08:31:11 -0600 (MDT) Message-Id: <199605081431.IAA10551@rover.village.org> To: linux-mips@fnet.fr Subject: Re: m700-10 success In-Reply-To: Your message of Wed, 08 May 1996 06:17:06 +0200 Date: Wed, 08 May 1996 08:31:11 -0600 From: Warner Losh Content-Length: 2119 Lines: 41 : Well, distributions are a special thing with Linux/MIPS anyway. We're about : to endup with two complete binary trees - one little and one big endian : tree. I think this might be a real problem with some FTP servers mirroring : Linux/MIPS. Two complete packages for one not that popular architecture : could affect the spreading negativly. Being little endian many of use : don't have the advantage of conforming to the ABI, exchanging binaries gets : difficult. Maybe, maybe not. I think that it will really be a drop in the bucket for most FTP sites. If we have to go through a while with having dual byte gendered binaries, I think that is reasonable. : For this reasons I'm really thinking hard about making a bi-endian kernel. : This is however Pandora's box and I'd like to keep it closed ... Some time : ago I've already done some successfull experiments with big endian usercode : on my Acer. This however was a tiny program; real world apps would : require us to change large parts of the kernel. In particular ioctl's : will be fun. The only thing that accounts positivly for us is that : enforced by Intel's "maladia segmenti" almost every useraccess goes through : get_user() and put_user() which'd be very usefull hooks for making Linux : bi-endian. : : So there is a decission we have to make. I'd vote a little bit for stability. If this sort of change can be made such that it doesn't break anything that is already working, great. It sounds somewhat largish to me, with some sort of special code needed for the driver's ioctl routines in every driver, which doesn't sound happy to me. I know that interrupts on the Deskstation and such broke when I upgraded from 1.3.62 to 1.3.96, so I'm a little gun shy from that still. I've just now got back to the point I can start to walk forward in the releases to see which one might have caused the problem. So this colors my perspective a little. Since I now have a hard disk that I can install from my intel box, I'd love to have something more up to date than root-0.1. I'd be happy to help build this if that would help... Warner From imp@rover.village.org Wed May 8 16:32:51 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 QAA11500 for ; Wed, 8 May 1996 16:32:50 +0200 (MET DST) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA26785; Wed, 8 May 1996 16:32:40 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.7.5/8.6.6) with SMTP id IAA10590 for ; Wed, 8 May 1996 08:32:36 -0600 (MDT) Message-Id: <199605081432.IAA10590@rover.village.org> To: linux-mips@fnet.fr Subject: Re: The bad news ... In-Reply-To: Your message of Wed, 08 May 1996 16:11:13 +0200 Date: Wed, 08 May 1996 08:32:36 -0600 From: Warner Losh Content-Length: 340 Lines: 9 : The good news is that I now have two of 'em running :-) So : far EATA DMA and EATA PIO are running. I'll now swap : the hostadapters of my Acer and my 486 and make that Adaptec 1542 : running ... Cool! How hard was this. What did you need to do. I'll be facing the same problem with the BusTech driver here in a little bit. Warner From andy@soft-n-hard.de Wed May 8 18:20:06 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 SAA13680 for ; Wed, 8 May 1996 18:20:05 +0200 (MET DST) Received: from aw.bnc.net by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA28888; Wed, 8 May 1996 18:19:57 +0200 (MET) Received: from newton.soft-n-hard.de (andy@newton.soft-n-hard.de [194.163.66.33]) by aw.bnc.net (8.6.12/8.6.12) with ESMTP id SAA07543 for ; Wed, 8 May 1996 18:24:10 +0200 Received: (from andy@localhost) by newton.soft-n-hard.de (8.6.12/8.6.9) id SAA05428; Wed, 8 May 1996 18:17:30 +0200 Date: Wed, 8 May 1996 18:17:30 +0200 From: Andreas Busse Message-Id: <199605081617.SAA05428@newton.soft-n-hard.de> To: linux-mips@fnet.fr Subject: Re: The bad news ... In-Reply-To: <199605081413.QAA09483@informatik.uni-koblenz.de> X-Mailer: [XMailTool v3.1.2b] Content-Length: 494 Lines: 15 ...and the good news: I received a tiny little box with the "SNI" logo on it today. Didn't opened it yet, but I guess it's a RM200... Well, Michael, when do you like to come over? Cheers, Andy ----------------------------------------------------------- Andreas Busse | andy@soft-n-hard.de Soft N Hard GbR | Phone: +49 2636-970105 Im Hufen Boden 16, D-53498 Waldorf | Fax: +49 2636-970106 ----------------------------------------------------------- From bigbug.franken.de!tsbogend@hub-fue.franken.de Wed May 8 18:36:00 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 SAA13945 for ; Wed, 8 May 1996 18:35:58 +0200 (MET DST) Received: from rachael.franken.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA29138; Wed, 8 May 1996 18:35:45 +0200 (MET) Received: from hub-fue.franken.de by rachael.franken.de with smtp (Smail3.1.29.1 #8) id m0uHCDE-000oLDC; Wed, 8 May 96 18:35 MET DST Received: by hub-fue.franken.de (Smail3.1.29.1 #35) id m0uHCDC-0005tBC; Wed, 8 May 96 18:35 MET DST Received: by bigbug.franken.de (8.6.12/tbs) id SAA04536 for linux-mips@fnet.fr; Wed, 8 May 1996 18:33:26 +0200 From: Thomas Bogendoerfer Message-Id: <199605081633.SAA04536@bigbug.franken.de> Subject: Re: GCC crosscompiler problem on Alpha To: linux-mips@fnet.fr Date: Wed, 8 May 1996 18:33:24 +0200 (MET DST) In-Reply-To: <199605072346.BAA01934@bigbug.franken.de> from "Thomas Bogendoerfer" at May 8, 96 01:46:26 am Reply-To: tsbogend@bigbug.franken.de X-Mailer: ELM [version 2.4 PL24 ME8b] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 3557 Lines: 107 > Today I have done a quick hack, and compiled the cross compiler > with #define HOST_WIDE_INT int (which is 32bit wide). Now the > code looks much better, but there are still minor differences. Update: I found the reason for the minor differences. I only rebuilt cc1 and not cpp. So the test for (~0UL) == 0xffffffff in asm-mips/types.h fails and I ended up with a u64 which was only 32bit wide:-( Next thing which hit me, is a bug in gas. On a 64bit machine it translates dli 0x100000000 to nothing. I first wondered why nobody else had this problem before, but I realized that on a 32bit platform bignums are used for numbers wider than 32bit. Finally there are some problems in the .S files with loading of pointers. Example: arch/mips/lib/csum.S: #include #include #include /* * Compute kernel code checksum to check kernel code against corruption */ LEAF(csum) jal sys_cacheflush move t8,ra # delay slot li t0,KSEG1 la t1,final li t2,KSEG1 on my 386 this will result in: 00000000 jal 00000000 ... 00000008 move $t8,$ra 0000000c lui $t0,40960 00000010 lui $t1,0 00000014 daddiu $t1,$t1,0 00000018 lui $t2,40960 as you can see t0 is not loaded with 0xa0000000 but with 0xffffffffa0000000. The alpha mipsel-linux-as generates following code: 0000000000000000 jal 0000000000000000 ... 0000000000000008 move $t8,$ra 000000000000000c li $t0,40960 0000000000000010 dsll $t0,$t0,0x10 0000000000000014 lui $t1,0 0000000000000018 daddiu $t1,$t1,0 000000000000001c li $t2,40960 0000000000000020 dsll $t2,$t2,0x10 To get the wanted result I had to modify csum.S a little bit: #include #include #include /* * Compute kernel code checksum to check kernel code against corruption */ LEAF(csum) jal sys_cacheflush move t8,ra # delay slot li t0,(KSEG1 | 0xffffffff00000000) la t1,final li t2,(KSEG1 | 0xffffffff00000000) There are serveral places in arch/mips where the same problem occurs. For me it looks like a assembler bug and a problem with some kernel pointer in Linux/Mips. But now the good news: Console: colour MIPS-G364 128x48, 1 virtual console (max 63) Calibrating delay loop.. ok - 49.87 BogoMIPS VDMA: R4030 DMA pagetables initialized. Memory: 31384k/32764k available (608k kernel code, 772k data) Swansea University Computer Society NET3.034 for Linux 1.3.77 NET3: Unix domain sockets 0.12 for Linux NET3.033. Swansea University Computer Society TCP/IP for NET3.034 IP Protocols: ICMP, UDP, TCP Checking for 'wait' instruction... unavailable. Linux version 1.3.91 (root@alpha.franken.de) (gcc version 2.7.2) #38 Wed May 8 18:06:09 MET DST 1996 Floppy drive(s): fd0 is 1.44M Started kswapd v 1.4.2.2 FDC 0 is a post-1991 82077 ne.c:v1.10 9/23/94 Donald Becker (becker@cesdis.gsfc.nasa.gov) NE*000 ethercard probe at 0x300: 00 80 28 80 2b 70 eth0: NE2000 found at 0x300, using IRQ 10. Sending BOOTP requests..... OK Root-NFS: Got BOOTP answer from 193.175.24.66, my address is 193.175.24.69 Root-NFS: Got file handle for /tftpboot/193.175.24.69 via RPC VFS: Mounted root (nfs filesystem). Thomas. -- That process _deserves_ to die ("My name is Linus Torvalds, prepare to die"). [Linus Torvalds on linux-kernel] From linux@mailhost.uni-koblenz.de Wed May 8 21:19:03 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 VAA17233 for ; Wed, 8 May 1996 21:19:02 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA01487; Wed, 8 May 1996 21:18:54 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id VAA20247 for ; Wed, 8 May 1996 21:18:28 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605081918.VAA20247@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id VAA01595; Wed, 8 May 1996 21:16:34 +0200 Subject: Re: The bad news ... To: linux-mips@fnet.fr Date: Wed, 8 May 1996 21:16:33 +0200 (MET DST) In-Reply-To: <199605081617.SAA05428@newton.soft-n-hard.de> from "Andreas Busse" at May 8, 96 06:17:30 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 263 Lines: 10 Hi, > I received a tiny little box with the "SNI" logo on it today. > Didn't opened it yet, but I guess it's a RM200... > Well, Michael, when do you like to come over? This was the right thing to say at the wrong time. We'll come at let's say 3am ... Ralf From linux@mailhost.uni-koblenz.de Wed May 8 21:25:29 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 VAA17363 for ; Wed, 8 May 1996 21:25:27 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA01575; Wed, 8 May 1996 21:25:19 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id VAA20464 for ; Wed, 8 May 1996 21:25:10 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605081925.VAA20464@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id VAA01605; Wed, 8 May 1996 21:23:16 +0200 Subject: Re: GCC crosscompiler problem on Alpha To: linux-mips@fnet.fr Date: Wed, 8 May 1996 21:23:15 +0200 (MET DST) In-Reply-To: <199605081633.SAA04536@bigbug.franken.de> from "Thomas Bogendoerfer" at May 8, 96 06:33:24 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 3616 Lines: 102 Hi, > > Today I have done a quick hack, and compiled the cross compiler > > with #define HOST_WIDE_INT int (which is 32bit wide). Now the > > code looks much better, but there are still minor differences. > > Update: I found the reason for the minor differences. I only > rebuilt cc1 and not cpp. So the test for (~0UL) == 0xffffffff > in asm-mips/types.h fails and I ended up with a u64 which was > only 32bit wide:-( Ok, I'll remove that thing. This test is historic and no longer worth a Penny. > Next thing which hit me, is a bug in gas. On a 64bit machine it > translates dli 0x100000000 to nothing. I first wondered why nobody > else had this problem before, but I realized that on a 32bit > platform bignums are used for numbers wider than 32bit. Interesting; acutally GAS doesn't know dli yet at all and should complain. One of a couple of bugs that prevent us from really going 64bit. > LEAF(csum) > jal sys_cacheflush > move t8,ra # delay slot > li t0,KSEG1 > la t1,final > li t2,KSEG1 > > on my 386 this will result in: > > 00000000 jal 00000000 > ... > 00000008 move $t8,$ra > 0000000c lui $t0,40960 > 00000010 lui $t1,0 > 00000014 daddiu $t1,$t1,0 > 00000018 lui $t2,40960 > > as you can see t0 is not loaded with 0xa0000000 but with 0xffffffffa0000000. > The alpha mipsel-linux-as generates following code: > > 0000000000000000 jal 0000000000000000 > ... > 0000000000000008 move $t8,$ra > 000000000000000c li $t0,40960 > 0000000000000010 dsll $t0,$t0,0x10 > 0000000000000014 lui $t1,0 > 0000000000000018 daddiu $t1,$t1,0 > 000000000000001c li $t2,40960 > 0000000000000020 dsll $t2,$t2,0x10 > > To get the wanted result I had to modify csum.S a little bit: > > #include > #include > #include > > /* > * Compute kernel code checksum to check kernel code against corruption > */ > LEAF(csum) > jal sys_cacheflush > move t8,ra # delay slot > li t0,(KSEG1 | 0xffffffff00000000) > la t1,final > li t2,(KSEG1 | 0xffffffff00000000) > > > There are serveral places in arch/mips where the same problem occurs. > > For me it looks like a assembler bug and a problem with some kernel > pointer in Linux/Mips. This is indeed a bug in the assembler. > But now the good news: > > Console: colour MIPS-G364 128x48, 1 virtual console (max 63) > Calibrating delay loop.. ok - 49.87 BogoMIPS > VDMA: R4030 DMA pagetables initialized. > Memory: 31384k/32764k available (608k kernel code, 772k data) > Swansea University Computer Society NET3.034 for Linux 1.3.77 > NET3: Unix domain sockets 0.12 for Linux NET3.033. > Swansea University Computer Society TCP/IP for NET3.034 > IP Protocols: ICMP, UDP, TCP > Checking for 'wait' instruction... unavailable. > Linux version 1.3.91 (root@alpha.franken.de) (gcc version 2.7.2) #38 Wed May 8 18:06:09 MET DST 1996 > Floppy drive(s): fd0 is 1.44M > Started kswapd v 1.4.2.2 > FDC 0 is a post-1991 82077 > ne.c:v1.10 9/23/94 Donald Becker (becker@cesdis.gsfc.nasa.gov) > NE*000 ethercard probe at 0x300: 00 80 28 80 2b 70 > eth0: NE2000 found at 0x300, using IRQ 10. > Sending BOOTP requests..... OK > Root-NFS: Got BOOTP answer from 193.175.24.66, my address is 193.175.24.69 > Root-NFS: Got file handle for /tftpboot/193.175.24.69 via RPC > VFS: Mounted root (nfs filesystem). Next satisfied customer :-) Ralf From linux@mailhost.uni-koblenz.de Wed May 8 21:35:24 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 VAA17558 for ; Wed, 8 May 1996 21:35:23 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA01700; Wed, 8 May 1996 21:34:50 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id VAA20727 for ; Wed, 8 May 1996 21:34:48 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605081934.VAA20727@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id VAA01718; Wed, 8 May 1996 21:32:53 +0200 Subject: Re: m700-10 success To: linux-mips@fnet.fr Date: Wed, 8 May 1996 21:32:51 +0200 (MET DST) In-Reply-To: <199605081431.IAA10551@rover.village.org> from "Warner Losh" at May 8, 96 08:31:11 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 691 Lines: 22 Hi, > Since I now have a hard disk that I can install from my intel box, > I'd love to have something more up to date than root-0.1. I'd be > happy to help build this if that would help... Yes, root-0.01 will be obsolete faster than I want to. I've fixed a bug in the libc/kernel signal handlers. The problem was that some routines assumed that sizeof(sigset_t) == 8 while it really should be 16 because (remember that we're sys-five'ish!) of sigset_t defined as: typedef sigset_t { int __sigbits[4]; } The fixed kernel now overwrites a little bit too much memory when it copies a sigset_t to usermem. Yikes. Ok, I think I'll cheat a bit to keep the user stuff at live ... Ralf From hodgen@mailhost.uni-koblenz.de Wed May 8 21:46:31 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 VAA17767 for ; Wed, 8 May 1996 21:46:30 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA01852; Wed, 8 May 1996 21:46:23 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id VAA21047 for ; Wed, 8 May 1996 21:46:21 +0200 (MET DST) From: Wayne Hodgen Message-Id: <199605081946.VAA21047@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id VAA01739; Wed, 8 May 1996 21:44:27 +0200 Subject: Magnum != Olivetti To: linux-mips@fnet.fr Date: Wed, 8 May 1996 21:44:26 +0200 (MET DST) Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 831 Lines: 17 Hi, a small "oops" here. I wanted to make sure serial was working before starting on a new console (so I can see what's crashing) but I had no luck getting it working. Ralf brought the Oli dox that Andy gave him and I found that the serial baud generator for the Oli's is a PC part (1.8432 MHz). The code that Andy wrote uses an 8MHz part in the magnum :( I changed the JAZZ_BASE_BAUD define to the PC norm, et viola! she works. I guess that means we need some sort if "I'm an Oli" define somewhere... Wayne -- Wayne Hodgen | hodgen@informatik.uni-koblenz.de | #include From linux@mailhost.uni-koblenz.de Wed May 8 21:57:01 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 VAA17975 for ; Wed, 8 May 1996 21:57:00 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA01998; Wed, 8 May 1996 21:56:52 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id VAA21254 for ; Wed, 8 May 1996 21:56:44 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605081956.VAA21254@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id VAA01747; Wed, 8 May 1996 21:54:50 +0200 Subject: Re: The bad news ... To: linux-mips@fnet.fr Date: Wed, 8 May 1996 21:54:49 +0200 (MET DST) In-Reply-To: <199605081432.IAA10590@rover.village.org> from "Warner Losh" at May 8, 96 08:32:36 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1633 Lines: 36 Hi, > : The good news is that I now have two of 'em running :-) So > : far EATA DMA and EATA PIO are running. I'll now swap > : the hostadapters of my Acer and my 486 and make that Adaptec 1542 > : running ... > > Cool! How hard was this. What did you need to do. I'll be facing > the same problem with the BusTech driver here in a little bit. Well, basically this was pretty easy. The PIO driver tried directly to access the memory of the VGA card for scrolling up a line and printing the word "reset". I threw that shit away and it worked. The DMA driver was a bit harder. A virt_to_phys call was missing, a zero pointer was being dereferenced and the primary caches weren't correctly being flushed and the interrupt code was broken. That bad thing is that the SCSI code isn't that broken as I already knew. It is a complete mess. For a low-mem MIPS like my 8mb machine sometimes after booting the machine and adding 48mb of swap (while just init, five /sbin/getty, one ash and fsck are running) not enough resources are left to continue for the SCSI code. Ok, I hear you say why does this stupid guy have just 8mb of RAM in box (Because I don't print my money myself ...) - if he' have more RAM this wouldn't happen. Wrong. Completly wrong. The problem is that the SCSI code (and the network code and ...) run into heavy problems when the physical memory is fragmented enough. When this happens usually at some hundred kilobytes memory are free plus some megs in the buffercache (not dirty buffers!). And now you know with what I spend the rest of time with; so there is still no 1542 support yet... Ralf From linux@mailhost.uni-koblenz.de Wed May 8 22:00:56 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 WAA18254 for ; Wed, 8 May 1996 22:00:55 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA02063; Wed, 8 May 1996 22:00:47 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id WAA21568 for ; Wed, 8 May 1996 22:00:46 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605082000.WAA21568@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id VAA01779; Wed, 8 May 1996 21:58:52 +0200 Subject: Re: Magnum != Olivetti To: linux-mips@fnet.fr Date: Wed, 8 May 1996 21:58:51 +0200 (MET DST) In-Reply-To: <199605081946.VAA21047@informatik.uni-koblenz.de> from "Wayne Hodgen" at May 8, 96 09:44:26 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 632 Lines: 16 Hi, > a small "oops" here. I wanted to make sure serial was working before > starting on a new console (so I can see what's crashing) but I had no > luck getting it working. Ralf brought the Oli dox that Andy gave him and > I found that the serial baud generator for the Oli's is a PC part (1.8432 > MHz). The code that Andy wrote uses an 8MHz part in the magnum :( > > I changed the JAZZ_BASE_BAUD define to the PC norm, et viola! she works. > I guess that means we need some sort if "I'm an Oli" define somewhere... Which Stoned has eleminated with deadly precission ... Ralf (Grin, it's good to know that I was right :-> From info%mmb.com@mmb.com Wed May 8 22:42:13 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 WAA18969 for ; Wed, 8 May 1996 22:42:11 +0200 (MET DST) Received: from gatekeeper.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA02625; Wed, 8 May 1996 22:40:56 +0200 (MET) Received: from teamate.mmb.com (teamate.mmb.com [192.215.168.2]) by gatekeeper.waldorf-gmbh.de (8.6.12/8.6.9) with SMTP id WAA14004 for ; Wed, 8 May 1996 22:42:20 +0200 Date: Wed, 8 May 1996 22:42:20 +0200 Received: from tm1.mmb.com by teamate.mmb.com id aa16522; 8 May 96 13:18 PDT To: linux@waldorf-gmbh.de Subject: TEAMate Web/BBS Server for Linux From: info@mmb.com Message-Id: <9605081318.aa16522@teamate.mmb.com> Content-Length: 5962 Lines: 154 MMB TEAMate Newsletter - May 7, 1996 TEAMate is the only server product for UNIX that combines both a Web and BBS interface in a single product. Users may access a TEAMate server with the Web browser of their choice, a session based GUI client for Windows or Mac, a VT100 terminal or even by sending a mail message with an included query. All information is dynamically indexed for immediate retrieval and the server generates HTML so you dont have to. Since TEAMate also works with our proprietary, session based client, you can provide your users with a client program customized exactly to your specifications, with your logo, look and feel, etc., that will operate across the Internet via TCP/IP or even over normal dial up service without a TCP/IP connection. You receive an unlimited distribution license for our client when you license the server. Visit our site to run TEAMate and to get the latest at http://mmb.com Internet World 96 Show This was a great show! Everyone who was anyone in the Internet world was there. Our booth was in a great location and we got over 1000 entries in our contest (see below). The show was so big that the exhibition company set up a tent to house extra exhibits. The bad news was that the weather in San Jose was unseasonable (warm) and the temperature in the tent was over 100. The show next year will be in Los Angeles at the LA Convention Center. We showed our test version of our Java client for TEAMate and the response was excellent. A number of people commented that it was one of the few Java applications that really did something. Well have it on our Web Site soon. We expect a production release of our Java client during the summer. Linux $495 We believe that UNIX is a much stronger platform for commercial applications than NT and to show our continued support for UNIX as a platform at the show we announced that we would be supporting the Linux version of UNIX with a special price of $495. This includes ALL of the TEAMate modules and an unlimited user license. A maintenance and support contract is available for $495 per year. We hope that by supporting this free version of UNIX we will increase the overall use of UNIX and promote UNIX as the best platform for multi-user applications. We will begin shipping the Linux version of TEAMate in June. Netscape Server TEAMate Plug-In We are now testing the next release of TEAMate that will run as a server plug-in to a Netscape server. All of the features of TEAMate will be available including access via our customizable GUI client for Windows and Macintosh. If youre already a Netscape server user, you should take a look at this plug-in to add forums, classified ads, reference libraries, event calendars, user authoring and much more to your existing server. Try TEAMate out at http://mmb.com TEAMate Phrase Search & User Defined Fields The latest release adds a flexible phrase search to the TEAMate search. In addition this release supports user defined fields. This means that if you need to define an additional field to be associated with an entry, you can easily do so. A good example of this would be a classified ad where you needed to add a field for the price of the item. With the new capability you can add the numeric field for price and your users will be able to search from a certain price to a certain price to locate exactly the information they need. This capability is integral to TEAMate and does not require any add-on software. Sierra Solutions E-mail Product You may have noticed that we are now answering our phones MMB Sierra. This is because we have joined with Sierra Solutions to help them with the marketing and distribution of their MailRoom product. MailRoom is a complete off-line electronic mail product that works with MCI mail. Sierra is also testing a new version that will work with Internet mail via a POP-3 server. You can get a complete working copy of their software from their Web site at http://www.sierrasol.com/sierrasol Contest Winners As many of you know we held a contest in conjunction with the Internet World show in San Jose. The first prize, a complete TEAMate system for the computer of their choice was won by Bob Britt of United Press International. The Eddie Bauer duffel bag was won by Dale Rickman of Hughes Aircraft. Training Class Our next one week training class is scheduled for July 15th. The cost is $1800. Please call Aisling or Lisa at (800) 832-6022 for reservations. Demo Site Access You can access our demonstration site in a variety of ways. With a Web browser you can go to http://mmb.com. You can retrieve a copy of our GUI client software from our FTP site (use the login of anonymous). You can telnet to teamate.mmb.com and use the login name of public. You can send an e-mail message to query@mmb.com. You can use a Gopher client to teamate.mmb.com. If you have a PC with a terminal program such as Procomm you can call (310) 318-5302. Use VT100 emulation and when prompted for a login use public. Our site is a complete functioning version of our software and includes as content, complete information about our products. You can prototype your own information right on our demo site. MMB Development Corporation Contact Information We are located in Manhattan Beach, California. Our address is 904 Manhattan Avenue, Manhattan Beach, CA 90266. Our phone numbers are (800) 832-6022 or (310) 318-1322. Our fax number is (310) 318-2162. Our e- mail address is info@mmb.com. "TEAMate Software to Help People Work Together" MMB Development Corporation MOSAIC URL:http://teamate.mmb.com 904 Manhattan Ave. TELNET teamate.mmb.com (login: public) Manhattan Beach, CA 90266 GOPHER teamate.mmb.com (800) 832-6022 or (310) 318-1322 DIALUP (310) 318-5302 (login: public) FAX: (310) 318-2162 EMAIL QUERY query@mmb.com From linux@mailhost.uni-koblenz.de Thu May 9 00:37:41 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 AAA21400 for ; Thu, 9 May 1996 00:37:40 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA04663; Thu, 9 May 1996 00:36:02 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id AAA26755 for ; Thu, 9 May 1996 00:35:51 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605082235.AAA26755@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id AAA02188; Thu, 9 May 1996 00:33:56 +0200 Subject: Re: SGI UltraP... To: linux-mips@fnet.fr Date: Thu, 9 May 1996 00:33:56 +0200 (MET DST) In-Reply-To: <199605080638.XAA06385@orion.starlink.com> from "Adam Fritzler" at May 7, 96 11:50:56 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 697 Lines: 16 Hi, > This really has no relavence, but I was reading that press release that your > site links to about SGI's UltraP Pentium-Replacement boards. That press > release mentioned no one actually making these boards. Did anyone ever > produce them? (They sounded interesting :) Yes, they indeed sounded interesting; I even think someone once offered me such a thing once for porting Linux to ... But anyway - that thingy isn't supported yet (and probably has never been really been marketed; I had to tell some SGI official that this thing exists at all ...) and I have to admit that the UltraP module nowadays has lost a lot of it's atractivity due to the availability of i686 boards. Ralf From digitech!digitech.co.nz!arnim@atlantis.actrix.gen.nz Thu May 9 06:36:30 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 GAA29916 for ; Thu, 9 May 1996 06:36:29 +0200 (MET DST) Received: from atlantis.actrix.gen.nz by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA11108; Thu, 9 May 1996 06:36:21 +0200 (MET) Received: (from uucp@localhost) by atlantis.actrix.gen.nz (8.6.11/8.6.9) id QAA06860 for linux-mips@fnet.fr; Thu, 9 May 1996 16:36:09 +1200 >Received: by digitech.co.nz (UUPC/extended 1.12r); Thu, 09 May 1996 16:12:15 +1200 Message-Id: <3191628f.digitech@digitech.co.nz> Date: Thu, 09 May 1996 16:12:13 +1200 From: "Arnim Littek" Organization: Digi-Tech Ltd, Wellington, New Zealand Reply-To: "Arnim Littek" To: linux-mips@fnet.fr Subject: Re: m700-10 success Content-Type: text Content-Length: 202 Lines: 11 On Wed, 8 May 1996 06:20:31 +0200 (MET DST), linux-mips@fnet.fr wrote: > PS: Little endian sucks. I'd be interested in your quantifiable reasons behind that, if you don't mind... Thanks, Arnim. -- From andy@soft-n-hard.de Thu May 9 09:07:01 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 JAA03144 for ; Thu, 9 May 1996 09:07:00 +0200 (MET DST) Received: from relay2.EUnet.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA14009; Thu, 9 May 1996 09:06:56 +0200 (MET) Received: from aw.bnc.net by relay2.eunet.fr (5.65c8d/96.05.03) via EUnet-France id AA12874; Thu, 9 May 1996 09:06:53 +0200 (MET) Received: from newton.soft-n-hard.de (andy@newton.soft-n-hard.de [194.163.66.33]) by aw.bnc.net (8.6.12/8.6.12) with ESMTP id IAA09782 for ; Thu, 9 May 1996 08:49:47 +0200 Received: (from andy@localhost) by newton.soft-n-hard.de (8.6.12/8.6.9) id IAA26687; Thu, 9 May 1996 08:43:13 +0200 Date: Thu, 9 May 1996 08:43:13 +0200 From: Andreas Busse Message-Id: <199605090643.IAA26687@newton.soft-n-hard.de> To: linux-mips@fnet.fr Subject: Re: The bad news ... In-Reply-To: <199605081918.VAA20247@informatik.uni-koblenz.de> X-Mailer: [XMailTool v3.1.2b] Content-Length: 750 Lines: 21 Hi again, > This was the right thing to say at the wrong time. We'll > come at let's say 3am ... Ok, I'll try to be here then, but please give me a call before. BTW, for the sake of the contract with SNI: I'd like to create an account for Mike here, so that the at least the initial conversation with SNI can be done with my company's name. Just to avoid confusion. Michael, can you please mail me your preferred /etc/passwd entry? Thanks, Andy ----------------------------------------------------------- Andreas Busse | andy@soft-n-hard.de Soft N Hard GbR | Phone: +49 2636-970105 Im Hufen Boden 16, D-53498 Waldorf | Fax: +49 2636-970106 ----------------------------------------------------------- From Stoned.Elipot@univ-evry.fr Thu May 9 08:53:12 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 IAA02664 for ; Thu, 9 May 1996 08:53:11 +0200 (MET DST) Received: from amiante.univ-evry.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA13611; Thu, 9 May 1996 08:53:07 +0200 (MET) Received: from bandsept.univ-evry.fr (bandsept.univ-evry.fr [192.93.214.19]) by amiante.univ-evry.fr with ESMTP (8.6.12/96.04.09/louis-blanc); id IAA19359; Thu, 9 May 1996 08:53:09 +0200 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 IAA02791; Thu, 9 May 1996 08:52:26 +0200 Message-Id: <199605090652.IAA02791@bandsept.univ-evry.fr> To: linux-mips@fnet.fr Subject: Re: Magnum != Olivetti In-Reply-To: Your message of "Wed, 08 May 1996 21:44:26 +0200." <199605081946.VAA21047@informatik.uni-koblenz.de> X-It: bandsept (wrong OS, but right CPU) X-Mailer: MH 6.8.3, MH-E 5.0.2 Date: Thu, 09 May 1996 08:52:19 +0200 From: Stoned Elipot Content-Length: 1523 Lines: 34 Hi all, >>>>> "Wayne" == Wayne Hodgen writes: [SNIP] Wayne> I changed the JAZZ_BASE_BAUD define to the PC norm, et viola! Wayne> she works. I guess that means we need some sort if "I'm an Wayne> Oli" define somewhere... Ah,... Well I was finally right from the "beginning", I/we had put from place to place OLIVETTI... stuff for now with just the same values as for the MAGNUM 4000, time to make the things clearer ! Regarding this subject I got a little kernel patch for kernel 1.3.96 wich changes a little bit the mips_machgroup mips_machtype handling (the harder for me was to fix wire_mapping in head.S: it now use a two level array scheme for getting the wired TLB entries) but I think it's working, there is also a quick and dirty hack in proc.c for reporting the correct family group and model, someone would problably find a better way). Unfortunatelly the motherboard of the box holding the disks on which this patch is, died last night at 11h: for who knows it, bickel.ppp.univ-evry.fr is dead, houin... !! And I finnally used with success my developpment Milo 0.27 which binary image is now only 28k big. I was about to add a ethernet and IDE card in my oily in order to reach the level of reliability other people like Tor or Warner/Ralf had achivied with this box when this motherboard's trip to "hardware cemetary" occured :( Cheers, Stoned. PS/ Another motherboard to stick on the walls of the flat ! Luc, a few others and I'll have as much as you in this state :) From Stoned.Elipot@univ-evry.fr Thu May 9 09:03:52 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 JAA03033 for ; Thu, 9 May 1996 09:03:51 +0200 (MET DST) Received: from amiante.univ-evry.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA13978; Thu, 9 May 1996 09:03:47 +0200 (MET) Received: from bandsept.univ-evry.fr (bandsept.univ-evry.fr [192.93.214.19]) by amiante.univ-evry.fr with ESMTP (8.6.12/96.04.09/louis-blanc); id JAA19398; Thu, 9 May 1996 09:03:48 +0200 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 JAA02811; Thu, 9 May 1996 09:03:09 +0200 Message-Id: <199605090703.JAA02811@bandsept.univ-evry.fr> To: linux-mips@fnet.fr Subject: Re: Magnum != Olivetti In-Reply-To: Your message of "Wed, 08 May 1996 21:58:51 +0200." <199605082000.WAA21568@informatik.uni-koblenz.de> X-It: bandsept (wrong OS, but right CPU) X-Mailer: MH 6.8.3, MH-E 5.0.2 Date: Thu, 09 May 1996 09:03:03 +0200 From: Stoned Elipot Content-Length: 1277 Lines: 36 Hi, >>>>> "Ralf" == Systemkennung Linux writes: [SNIP] Ralf> Which Stoned has eleminated with deadly precission ... Hum,.. as I said, not really just change the value of MACH_MIPS_OLIVETTI :) Ralf> (Grin, it's good to know that I was right :-> Was it you ?? :):) Another subject: I had split init_IRQ() in arch/mips/kernel/irq.c in init_IRQ_jazz(), init_IRQ_dec(), init_IRQ_arc(). Ah yes in my attempt to re-organize the mips_machgroup stuff I made a group called ARC with for now hold Deskstation boxes, and the is also a Jazz boxes group To be clearer: each group is a single bit im mips_machgroup. mips_machgroup: 0000000000000000000000000000000 group unknown 0000000000000000000000000000001 group ARC 0000000000000000000000000000010 group DEC 0000000000000000000000000000100 group Jazz ??????????????????????????????? group SGI ? ??????????????????????????????? group SNI ? (if nedded) And then each models in each group use in turn 0,1,2,3,... for mips_machtype. So for Dec we'll be able to identify all this 3mxx, pmax,... boxes with their original/Digital model number. Comments ? Ralf, as soon I would be able to make bickel running again, I'll send you all this. Cheers, Stoned. From hodgen@mailhost.uni-koblenz.de Thu May 9 10:20:08 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 KAA04582 for ; Thu, 9 May 1996 10:20:07 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA01744; Thu, 9 May 1996 10:18:17 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id KAA23501 for ; Thu, 9 May 1996 10:17:54 +0200 (MET DST) From: Wayne Hodgen Message-Id: <199605090817.KAA23501@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id KAA02719; Thu, 9 May 1996 10:15:59 +0200 Subject: Re: Magnum != Olivetti To: linux-mips@fnet.fr Date: Thu, 9 May 1996 10:15:58 +0200 (MET DST) In-Reply-To: <199605090703.JAA02811@bandsept.univ-evry.fr> from "Stoned Elipot" at May 9, 96 09:03:03 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1483 Lines: 34 Hi, > >>>>> "Ralf" == Systemkennung Linux writes: > [SNIP] > Ralf> Which Stoned has eleminated with deadly precission ... > > Hum,.. as I said, not really just change the value of MACH_MIPS_OLIVETTI :) Hmmm, so is there a way to tell them apart? Or could you point me at the right place to look/change? At the moment I have #if 0 round the magnum bit which isn't very nice. > Ralf, as soon I would be able to make bickel running again, I'll send > you all this. Thats good to hear, I need the parameter passing in milo 27 :) Then I can boot from HD without nasty hacks... On another note: with the correct serial clock generator in, I can send stuff to another machine (eg cat /etc/passwd > /dev/ttyS0). But, it never closes the serial connection, it gets stuck in an interrupt loop (I switched the int debugging back on) for ever amen :( You get the file once right, then with 1 empty line after each correct one, then 2 empty lines etc. Very, very wierd stuff. I really hate this sort of shit, you know, you set out to fix one thing and end up fixing 5 others before you can get to the one you started out to fix. Ciao -- Wayne Hodgen | hodgen@informatik.uni-koblenz.de | #include From tor@spacetec.no Thu May 9 10:29:21 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 KAA04810 for ; Thu, 9 May 1996 10:29:20 +0200 (MET DST) Received: from pallas.spacetec.no by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA02026; Thu, 9 May 1996 10:29:10 +0200 (MET) Received: (from tor@localhost) by pallas.spacetec.no (8.7.1/8.7.1) id KAA30761 for linux-mips@fnet.fr; Thu, 9 May 1996 10:35:39 +0200 Message-Id: <199605090835.KAA30761@pallas.spacetec.no> From: tor@spacetec.no (Tor Arntsen) Date: Thu, 9 May 1996 10:35:38 +0200 In-Reply-To: "Arnim Littek" "Re: m700-10 success" (May 9, 5:35) X-Mailer: Mail User's Shell (7.2.5 10/14/92) To: linux-mips@fnet.fr Subject: Re: m700-10 success Content-Length: 2444 Lines: 54 On May 9, 5:35, "Arnim Littek" wrote: >On Wed, 8 May 1996 06:20:31 +0200 (MET DST), linux-mips@fnet.fr wrote: >> PS: Little endian sucks. > > >I'd be interested in your quantifiable reasons behind that, if you >don't mind... We should avoid too much on this I think, however I agree with the 'le sucks' opinion. The only feature I know about for little endian is the following: PROGRAM PJX0 INTEGER*4 J11 INTEGER*4 N J11 = 24 CALL YE_OLDE_CODE (J11,N) WRITE (*,*) 'IF THIS WORKS IT IS PROBABLY A VAX',N END SUBROUTINE YE_OLDE_CODE (I0,I1) INTEGER*2 I0 INTEGER*2 I1 I1 = I2 * 9 END And as I really consider that a bug, not a feature, not much is left.. You could (in the past at least) find a lot of this among VMS programmers, and it's a way of keeping huge amounts of old programs and code alive for a little longer, in that even if you write all your new progs with 32 bit ints you can still use the old old libs from 20 years back (with 16 bit ints) without having to rewrite them. Of course this will fail in unpredictable (and silent) ways when you call the old libs with values allowed for I4 but not for I2. A lot of times I've worked with people complaining about their code not working on this **$! computer we're porting their code to.. and then had to explain to them that the **$! comp is using something called Big Endian and it's really a nice comp and that their programs are actually buggy :-) And people get so confused when they try to convert between BE and LE, because they don't realise that it applies to *integers* stored into memory only, and not byte sequences in general (a struct with bitfields on a big endian host will usually not have to be byte-swapped before it's inter- changed with a LE host, for example). And even LE computers are usually BE inside registers. This applies to e.g. DMA registers, and registers on IO boards too -- this gives the byte-swappers no end of confusion. On the other hand, BE is BE is BE all the time.. I would really like to switch my Olivetti M700-10 over to big endian at some stage, if that's at all technically possible. If for no other reason than that SGI and Sun are big endian and I want to interchange data and do other interesting things. (And then I have my really old bitstring functions laying around somewhere, that's at least easier to write with BE) Cheers folks, Tor From tor@spacetec.no Thu May 9 10:33:05 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 KAA04920 for ; Thu, 9 May 1996 10:33:05 +0200 (MET DST) Received: from pallas.spacetec.no by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA02178; Thu, 9 May 1996 10:33:01 +0200 (MET) Received: (from tor@localhost) by pallas.spacetec.no (8.7.1/8.7.1) id KAA30769 for linux-mips@fnet.fr; Thu, 9 May 1996 10:39:37 +0200 Message-Id: <199605090839.KAA30769@pallas.spacetec.no> From: tor@spacetec.no (Tor Arntsen) Date: Thu, 9 May 1996 10:39:36 +0200 In-Reply-To: tor@spacetec.no (Tor Arntsen) "Re: m700-10 success" (May 9, 9:28) X-Mailer: Mail User's Shell (7.2.5 10/14/92) To: linux-mips@fnet.fr Subject: Re: m700-10 success Content-Length: 153 Lines: 3 Ok, it was a little bug in that program, but that wouldn't be noticed either when you executed it so it fits fine with the general point anyway :-) -Tor From Stoned.Elipot@univ-evry.fr Thu May 9 10:42:25 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 KAA05093 for ; Thu, 9 May 1996 10:42:24 +0200 (MET DST) Received: from amiante.univ-evry.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA02522; Thu, 9 May 1996 10:42:21 +0200 (MET) Received: from bandsept.univ-evry.fr (bandsept.univ-evry.fr [192.93.214.19]) by amiante.univ-evry.fr with ESMTP (8.6.12/96.04.09/louis-blanc); id KAA20073; Thu, 9 May 1996 10:42:18 +0200 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 KAA03045; Thu, 9 May 1996 10:41:38 +0200 Message-Id: <199605090841.KAA03045@bandsept.univ-evry.fr> To: linux-mips@fnet.fr Subject: Re: Magnum != Olivetti In-Reply-To: Your message of "Thu, 09 May 1996 10:15:58 +0200." <199605090817.KAA23501@informatik.uni-koblenz.de> X-It: bandsept (wrong OS, but right CPU) X-Mailer: MH 6.8.3, MH-E 5.0.2 Date: Thu, 09 May 1996 10:41:32 +0200 From: Stoned Elipot Content-Length: 1859 Lines: 40 >Hmmm, so is there a way to tell them apart? Or could you point me at the >right place to look/change? At the moment I have #if 0 round the magnum >bit which isn't very nice. The idea is Magnum 4000 and Olivetti belong to mips_machgroup = MACH_GROUP_JAZZ in my fixed scheme, but mips_machtype is different for Magnum and Oily (as for kernel 1.3.96 they are the same). So we could just test for mips_machgroup = MACH_GROUP_JAZZ for features supported/working the same way for both machine (and note that Acer Pica 61 is also in group JAZZ) and then for things like the serial driver just make a futher test for mips_machtype= OILY or MAGNUM. This can be done quickly by changing the value for MIPS_MACH_OLIVETTI in asm-mips/bootinfo.h and changing the tag_machtype entry in the default tag list array in Milo's identify.c. For now the magnum and the Oily goes throught the same flow of execution in Milo, so the solution is: make two default array of tags for the two boxes (this will increase the .data section of Milo :( ). Or perhaps we could design a way of chaining these tag arrays in identify.c. So we could have this layout: array for Magnum -> +----- | magnum specific tags | PTR +-- |---- `-----------------+ array for Oily -> +---------- | | oily specific tags | | PTR | +--|-------- | `-----\ /-----------+ array for Oily and Magnum -> +---- | tags for Oily and Magums +---- I'll try to implement this tonight. Hum, perhaps just having two independent arrays with redundant infos is simpler :) Have a look at the comments in Milo's identify.c you'll get the picture. Stoned. From dom@algor.co.uk Thu May 9 11:04:45 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 LAA05662 for ; Thu, 9 May 1996 11:04:44 +0200 (MET DST) Received: from moorgate.algor.co.uk by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA04139; Thu, 9 May 1996 11:04:37 +0200 (MET) Received: from bank.algor.co.uk (bank.algor.co.uk [193.117.190.11]) by moorgate.algor.co.uk (8.7.4/8.7.3) with SMTP id KAA04530 for ; Thu, 9 May 1996 10:04:32 +0100 (BST) From: Dominic Sweetman Date: Thu, 9 May 96 10:04:31 +0100 Message-Id: <25255.9605090904@bank.algor.co.uk> To: linux-mips@fnet.fr Subject: Re: m700-10 success In-Reply-To: <199605090835.KAA30761@pallas.spacetec.no> References: <199605090835.KAA30761@pallas.spacetec.no> Content-Length: 2863 Lines: 64 > We should avoid too much on this I think, however I agree with the > 'le sucks' opinion. Scientific manner, chaps! LE and BE are two different *arbitrary* choices about how you represent bigger-than-byte integers in a byte-organised memory. Neither one is right or wrong; I find LE easier because I began my computing career on DEC and got into low-level code on Intel CPUs. Tor finds BE natural, I suspect, because he started off on Motorola? Arguably DEC and IBM (whose bit numbers increase the same way as their byte addresses) are righter than Motorola (whose bit numbers go down as byte addresses go up). It would be nice if everyone did the same, of course - one less thing to go wrong. MIPS does both because the founders all used BE Sun workstations, and made the Stanford MIPS that way, but their first customer (Prime) was LE. > And people get so confused when they try to convert between BE and > LE, because they don't realise that it applies to *integers* stored > into memory only, and not byte sequences in general (a struct with > bitfields on a big endian host will usually not have to be > byte-swapped before it's inter- changed with a LE host, for > example). There are two different problems. The software problem (as above) afflicts most bigger-than-byte quantities - but it's easy to convert once you know what the quantities are. I don't think bitfield structs are very safe at all; the resulting memory layouts are explicitly compiler- and architecture-dependent. But there's a related hardware problem. Computers are of course wired up with wider-than-byte buses; on a 32-bit bus you have to decide which byte lane is used for bytes whose address is 0 mod 4. Hardware engineers (bless 'em) just wire bit 0 to bit 0 through to bit 31 to 31. Do that with a Motorola CPU and a PCI bus, and all the bytes *will* get scrambled (unix becomes xinu, etc). > And even LE computers are usually BE inside registers. There's no byte addresses inside the registers of any machine I can think of, so there ain't no endianness at all. > I would really like to switch my Olivetti M700-10 over to big endian > at some stage, if that's at all technically possible. Don't even think about it! PS: the "change endianness in user mode" feature was a late addition to the R3000 chip, by disgruntled MIPS engineers who wanted to be able to run DECstation binaries on MIPS Unix boxes. The system software to make the hardware feature usable was never included in any MIPS OS, and I think has never been used by anyone. Regards, Dominic Sweetman phone: +44 171 700 3301 Algorithmics Ltd home: +44 171 226 0032 3 Drayton Park fax: +44 171 700 3400 London N5 1NU email: dom@algor.co.uk ENGLAND. www: http://www.algor.co.uk ftp: ftp.algor.co.uk From tor@spacetec.no Thu May 9 13:05:41 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 NAA13728 for ; Thu, 9 May 1996 13:05:40 +0200 (MET DST) Received: from pallas.spacetec.no by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA13228; Thu, 9 May 1996 13:05:34 +0200 (MET) Received: (from tor@localhost) by pallas.spacetec.no (8.7.1/8.7.1) id NAA30905 for linux-mips@fnet.fr; Thu, 9 May 1996 13:12:10 +0200 Message-Id: <199605091112.NAA30905@pallas.spacetec.no> From: tor@spacetec.no (Tor Arntsen) Date: Thu, 9 May 1996 13:12:09 +0200 In-Reply-To: Dominic Sweetman "Re: m700-10 success" (May 9, 10:04) X-Mailer: Mail User's Shell (7.2.5 10/14/92) To: linux-mips@fnet.fr Subject: Re: m700-10 success Content-Length: 2590 Lines: 51 On May 9, 10:04, Dominic Sweetman wrote: >Scientific manner, chaps! LE and BE are two different *arbitrary* >choices about how you represent bigger-than-byte integers in a >byte-organised memory. Neither one is right or wrong; I find LE >easier because I began my computing career on DEC and got into >low-level code on Intel CPUs. Tor finds BE natural, I suspect, >because he started off on Motorola? Well, microprosessors weren't much used when I started, and I don't know even today what an Alpha LSI used for byte endianity.. however, when I began to do things where I had to know about such things I was working with VAX and another mini with Motorola endianity, at roughly the same time. >Arguably DEC and IBM (whose bit numbers increase the same way as their >byte addresses) are righter than Motorola (whose bit numbers go down >as byte addresses go up). It would be nice if everyone did the same, >of course - one less thing to go wrong. IMHO it doesn't matter much which way the bits are counted, however I prefer that the sequence is linear also when crossing byte boundaries (e.g. bit 7 sits beside bit 8 is more important than which comes 'first'). One reason is that I worked with satellite data where we had to write code to extract say 47 bits from a particular window inside a set of 111 bits. For me it was always easier to code that on big endian computers. There are others though who argue that LE is easier, although I never really got it :-) This is no big deal though, and I've been using my Linux/Intel PC since 1992 without suffering too much :-) >> I would really like to switch my Olivetti M700-10 over to big endian >> at some stage, if that's at all technically possible. > >Don't even think about it! The real reason I'm interested in BE for Linux/MIPS is as I said that Sun and SGI use it. So are you saying that it can't be done for Jazz boxes? That's bad news :-( >PS: the "change endianness in user mode" feature was a late addition > to the R3000 chip, by disgruntled MIPS engineers who wanted to be > able to run DECstation binaries on MIPS Unix boxes. The system > software to make the hardware feature usable was never included in > any MIPS OS, and I think has never been used by anyone. Interesting! I didn't know it could be done in user mode for MIPS, I thought it had to be done no later than during boot (I seem to remember a pin on the chip for this?). I know PPC can be switched at any time, one task can run BE and the other LE -- at the same time even. At least that's what I've been told. Cheers, -Tor From linux@mailhost.uni-koblenz.de Thu May 9 16:24:51 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 QAA18218 for ; Thu, 9 May 1996 16:24:50 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA05708; Thu, 9 May 1996 16:21:00 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id QAA12794 for ; Thu, 9 May 1996 16:20:55 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605091420.QAA12794@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id QAA03576; Thu, 9 May 1996 16:19:00 +0200 Subject: Re: m700-10 success To: linux-mips@fnet.fr Date: Thu, 9 May 1996 16:18:59 +0200 (MET DST) In-Reply-To: <199605091112.NAA30905@pallas.spacetec.no> from "Tor Arntsen" at May 9, 96 01:12:09 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1563 Lines: 35 Hi, > >Don't even think about it! Why? > The real reason I'm interested in BE for Linux/MIPS is as I said that Sun > and SGI use it. So are you saying that it can't be done for Jazz boxes? > That's bad news :-( Jazz box isn't Jazz box. The M700 has the endian switching hardware. I've recently tried to boot the configuration disk from the Magnum on the M700 but the Oily didn' accept the disk. I suspect the boot disk was the problem - opposed to my Acer the M700 does still have the endian switch for the kernel mode. > Interesting! I didn't know it could be done in user mode for MIPS, I thought > it had to be done no later than during boot (I seem to remember a pin on the > chip for this?). I know PPC can be switched at any time, one task can run BE > and the other LE -- at the same time even. At least that's what I've been > told. You can switch the reverse endian bit on a MIPS at any time but the kernel byte order is fixed and configured at boot time. There are some exceptions to this conncept. The oldest R2000 CPU don't work properly, the R8000 is (remember the R8000 is an SGI product, not a real MIPS chip) big endian only and I think there was also something special with the R6000. This as far as memory serves ... Byteorder switching for the kernel mode requires a motherboard that can deal with both byteorders. The trick how to switch the byteorder for the kernel is that a serial PROM supplies configuration information to the CPU - among that the byteorder information. So this PROM has to be changed also. Ralf From linux@mailhost.uni-koblenz.de Thu May 9 16:31:05 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 QAA18402 for ; Thu, 9 May 1996 16:31:03 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA05976; Thu, 9 May 1996 16:30:45 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id QAA13159 for ; Thu, 9 May 1996 16:30:28 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605091430.QAA13159@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id QAA03599; Thu, 9 May 1996 16:28:33 +0200 Subject: Re: m700-10 success To: linux-mips@fnet.fr Date: Thu, 9 May 1996 16:28:33 +0200 (MET DST) In-Reply-To: <3191628f.digitech@digitech.co.nz> from "Arnim Littek" at May 9, 96 04:12:13 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 704 Lines: 19 Hi, > On Wed, 8 May 1996 06:20:31 +0200 (MET DST), linux-mips@fnet.fr wrote: > > PS: Little endian sucks. > > I'd be interested in your quantifiable reasons behind that, if you > don't mind... Big endian machines are far more aggressive against software problem like interchanging pointers to integer objects of different size. On a big endian machine such problems are quite visible while on a little endian machine the problem is pretty invisible. And of course because we intended to create only one new UNIX/Linux flavour, not two. Having two doesn't even buy us something - one of both is superfluous. With respect to existing Unices for MIPSes I say this is the little endian flavour. Ralf From imp@rover.village.org Thu May 9 16:29:40 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 QAA18321 for ; Thu, 9 May 1996 16:29:40 +0200 (MET DST) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA05911; Thu, 9 May 1996 16:29:29 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.7.5/8.6.6) with SMTP id IAA17013 for ; Thu, 9 May 1996 08:29:22 -0600 (MDT) Message-Id: <199605091429.IAA17013@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Magnum != Olivetti In-Reply-To: Your message of Thu, 09 May 1996 08:52:19 +0200 Date: Thu, 09 May 1996 08:29:21 -0600 From: Warner Losh Content-Length: 837 Lines: 19 : And I finnally used with success my developpment Milo 0.27 which : binary image is now only 28k big. Cool. You might be able to get another 1-2k out of that if that is the stripped size. Does this mean that 0.27 is about ready to go out the door, or was that also on the disk that died. : I was about to add a ethernet and IDE card in my oily in order to : reach the level of reliability other people like Tor or Warner/Ralf : had achivied with this box when this motherboard's trip to "hardware : cemetary" occured :( Ummm, I'm not sure I'd call my system reliable just yet. I've still not had a # prompt on it (but I haven't tried a tftp root either). I'm still trying to track down what changed between .62 and .96 to cause my machine to stop servicing interrupts... It is a slow process, but I am making progress. Warner From imp@rover.village.org Thu May 9 16:31:48 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 QAA18429 for ; Thu, 9 May 1996 16:31:47 +0200 (MET DST) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA05986; Thu, 9 May 1996 16:31:39 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.7.5/8.6.6) with SMTP id IAA17035 for ; Thu, 9 May 1996 08:31:34 -0600 (MDT) Message-Id: <199605091431.IAA17035@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Magnum != Olivetti In-Reply-To: Your message of Thu, 09 May 1996 09:03:03 +0200 Date: Thu, 09 May 1996 08:31:34 -0600 From: Warner Losh Content-Length: 311 Lines: 8 : I had split init_IRQ() in arch/mips/kernel/irq.c in : init_IRQ_jazz(), init_IRQ_dec(), init_IRQ_arc(). Ah yes in my attempt : to re-organize the mips_machgroup stuff I made a group called ARC with : for now hold Deskstation boxes, and the is also a Jazz boxes group That sounds reasonable to me... Warner From imp@rover.village.org Thu May 9 16:37:22 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 QAA18564 for ; Thu, 9 May 1996 16:37:21 +0200 (MET DST) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA06086; Thu, 9 May 1996 16:37:09 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.7.5/8.6.6) with SMTP id IAA17077 for ; Thu, 9 May 1996 08:36:44 -0600 (MDT) Message-Id: <199605091436.IAA17077@rover.village.org> To: linux-mips@fnet.fr Subject: Re: m700-10 success In-Reply-To: Your message of Thu, 09 May 1996 13:12:09 +0200 Date: Thu, 09 May 1996 08:36:44 -0600 From: Warner Losh Content-Length: 608 Lines: 13 : Interesting! I didn't know it could be done in user mode for MIPS, I thought : it had to be done no later than during boot (I seem to remember a pin on the : chip for this?). That's why Ralf keeps talking about the bi-endian kernel. There is a bit in the PSW that says "I'm running in the opposite byte order now" that just needs to be toggled, and a few other kernel<-> changes. Well, and some library changes (htons is a noop on BE, but requires code on LE). You wouldn't believe the bugs you get when you are running in one endian and using numbers from the other endian once in a while... Warner From Stoned.Elipot@univ-evry.fr Thu May 9 17:00:05 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 RAA19008 for ; Thu, 9 May 1996 17:00:04 +0200 (MET DST) Received: from amiante.univ-evry.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA06633; Thu, 9 May 1996 17:00:00 +0200 (MET) Received: from bandsept.univ-evry.fr (bandsept.univ-evry.fr [192.93.214.19]) by amiante.univ-evry.fr with ESMTP (8.6.12/96.04.09/louis-blanc); id QAA22648; Thu, 9 May 1996 16:59:58 +0200 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 QAA03724; Thu, 9 May 1996 16:59:13 +0200 Message-Id: <199605091459.QAA03724@bandsept.univ-evry.fr> To: linux-mips@fnet.fr Subject: Re: Magnum != Olivetti In-Reply-To: Your message of "Thu, 09 May 1996 08:29:21 MDT." <199605091429.IAA17013@rover.village.org> X-It: bandsept (wrong OS, but right CPU) X-Mailer: MH 6.8.3, MH-E 5.0.2 Date: Thu, 09 May 1996 16:59:04 +0200 From: Stoned Elipot Content-Length: 375 Lines: 11 >>>>> "Warner" == Warner Losh writes: [SNIP] Warner> Cool. You might be able to get another 1-2k out of that if Warner> that is the stripped size. Does this mean that 0.27 is about Warner> ready to go out the door, or was that also on the disk that Warner> died. As far as I know the disks are fine just the motherboard, and that's enought :( Stoned. From linux@mailhost.uni-koblenz.de Thu May 9 17:25:21 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 RAA19757 for ; Thu, 9 May 1996 17:25:20 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA07282; Thu, 9 May 1996 17:21:37 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id RAA16833 for ; Thu, 9 May 1996 17:21:12 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605091521.RAA16833@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id RAA03977; Thu, 9 May 1996 17:19:17 +0200 Subject: Re: m700-10 success To: linux-mips@fnet.fr Date: Thu, 9 May 1996 17:19:17 +0200 (MET DST) In-Reply-To: <199605091436.IAA17077@rover.village.org> from "Warner Losh" at May 9, 96 08:36:44 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 964 Lines: 22 Hi, > : Interesting! I didn't know it could be done in user mode for MIPS, I thought > : it had to be done no later than during boot (I seem to remember a pin on the > : chip for this?). > > That's why Ralf keeps talking about the bi-endian kernel. There is a > bit in the PSW that says "I'm running in the opposite byte order now" > that just needs to be toggled, and a few other kernel<-> changes. > Well, and some library changes (htons is a noop on BE, but requires > code on LE). You wouldn't believe the bugs you get when you are > running in one endian and using numbers from the other endian once in > a while... GNU libc is _very_ in clean with respect to byte order problems. I found a single byte order problem yet which was crosscompiling related. If you tried to compile the lib for a big endian machine the crosscompiler host's byteorder was actually used which of course fails on a Intel box ... The big problem is in the kernel. Ralf From ralf@Julia.DE Thu May 9 17:52:03 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 RAA20487 for ; Thu, 9 May 1996 17:52:00 +0200 (MET DST) Received: from alles.intern.julia.de (loehnberg1.core.julia.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA07878; Thu, 9 May 1996 17:50:14 +0200 (MET) Received: from bryant.intern.julia.de (ralf@bryant.intern.julia.de [194.221.49.34]) by alles.intern.julia.de (8.7.5/8.7.3) with ESMTP id RAA08593 for ; Thu, 9 May 1996 17:47:56 +0200 From: Ralf Baechle Received: (from ralf@localhost) by bryant.intern.julia.de (8.7.5/8.7.3) id RAA29069 for linux-mips@fnet.fr; Thu, 9 May 1996 17:52:50 +0200 Message-Id: <199605091552.RAA29069@bryant.intern.julia.de> Subject: Re: TEAMate Web/BBS Server for Linux To: linux-mips@fnet.fr Date: Thu, 9 May 1996 17:52:50 +0200 (MET DST) In-Reply-To: <9605081318.aa16522@teamate.mmb.com> from "info@mmb.com" at May 8, 96 10:42:20 pm X-Mailer: ELM [version 2.4 PL24 PGP6] Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 6276 Lines: 156 > > > MMB TEAMate Newsletter - May 7, 1996 > > TEAMate is the only server product for UNIX that > combines both a Web and BBS interface in a single > product. Users may access a TEAMate server with the > Web browser of their choice, a session based GUI client > for Windows or Mac, a VT100 terminal or even by > sending a mail message with an included query. All > information is dynamically indexed for immediate > retrieval and the server generates HTML so you dont > have to. > > Since TEAMate also works with our proprietary, session > based client, you can provide your users with a client > program customized exactly to your specifications, with > your logo, look and feel, etc., that will operate across the > Internet via TCP/IP or even over normal dial up service > without a TCP/IP connection. You receive an unlimited > distribution license for our client when you license the > server. > > Visit our site to run TEAMate and to get the latest at > http://mmb.com > > Internet World 96 Show > > This was a great show! Everyone who was anyone in the > Internet world was there. Our booth was in a great > location and we got over 1000 entries in our contest (see > below). The show was so big that the exhibition > company set up a tent to house extra exhibits. The bad > news was that the weather in San Jose was unseasonable > (warm) and the temperature in the tent was over 100. > The show next year will be in Los Angeles at the LA > Convention Center. > > We showed our test version of our Java client for > TEAMate and the response was excellent. A number of > people commented that it was one of the few Java > applications that really did something. Well have it on > our Web Site soon. We expect a production release of > our Java client during the summer. > > Linux $495 > > We believe that UNIX is a much stronger platform for > commercial applications than NT and to show our > continued support for UNIX as a platform at the show we > announced that we would be supporting the Linux > version of UNIX with a special price of $495. This > includes ALL of the TEAMate modules and an unlimited > user license. A maintenance and support contract is > available for $495 per year. We hope that by supporting > this free version of UNIX we will increase the overall use > of UNIX and promote UNIX as the best platform for > multi-user applications. We will begin shipping the > Linux version of TEAMate in June. > > Netscape Server TEAMate Plug-In > > We are now testing the next release of TEAMate that will > run as a server plug-in to a Netscape server. All of the > features of TEAMate will be available including access > via our customizable GUI client for Windows and > Macintosh. If youre already a Netscape server user, you > should take a look at this plug-in to add forums, > classified ads, reference libraries, event calendars, user > authoring and much more to your existing server. Try > TEAMate out at http://mmb.com > > TEAMate Phrase Search & User Defined Fields > > The latest release adds a flexible phrase search to the > TEAMate search. In addition this release supports user > defined fields. This means that if you need to define an > additional field to be associated with an entry, you can > easily do so. A good example of this would be a > classified ad where you needed to add a field for the price > of the item. With the new capability you can add the > numeric field for price and your users will be able to > search from a certain price to a certain price to locate > exactly the information they need. This capability is > integral to TEAMate and does not require any add-on > software. > > Sierra Solutions E-mail Product > > You may have noticed that we are now answering our > phones MMB Sierra. This is because we have joined > with Sierra Solutions to help them with the marketing > and distribution of their MailRoom product. MailRoom > is a complete off-line electronic mail product that works > with MCI mail. Sierra is also testing a new version that > will work with Internet mail via a POP-3 server. You > can get a complete working copy of their software from > their Web site at http://www.sierrasol.com/sierrasol > > Contest Winners > > As many of you know we held a contest in conjunction > with the Internet World show in San Jose. The first > prize, a complete TEAMate system for the computer of > their choice was won by Bob Britt of United Press > International. The Eddie Bauer duffel bag was won by > Dale Rickman of Hughes Aircraft. > > Training Class > > Our next one week training class is scheduled for July > 15th. The cost is $1800. Please call Aisling or Lisa at > (800) 832-6022 for reservations. > > Demo Site Access > > You can access our demonstration site in a variety of > ways. With a Web browser you can go to > http://mmb.com. You can retrieve a copy of our GUI > client software from our FTP site (use the login of > anonymous). You can telnet to teamate.mmb.com and > use the login name of public. You can send an e-mail > message to query@mmb.com. You can use a Gopher > client to teamate.mmb.com. If you have a PC with a > terminal program such as Procomm you can call (310) > 318-5302. Use VT100 emulation and when prompted for > a login use public. > > Our site is a complete functioning version of our software > and includes as content, complete information about our > products. You can prototype your own information right > on our demo site. > > MMB Development Corporation Contact Information > > We are located in Manhattan Beach, California. Our > address is 904 Manhattan Avenue, Manhattan Beach, CA > 90266. Our phone numbers are (800) 832-6022 or (310) > 318-1322. Our fax number is (310) 318-2162. Our e- > mail address is info@mmb.com. > > > > > > "TEAMate Software to Help People Work Together" > > MMB Development Corporation MOSAIC URL:http://teamate.mmb.com > 904 Manhattan Ave. TELNET teamate.mmb.com (login: public) > Manhattan Beach, CA 90266 GOPHER teamate.mmb.com > (800) 832-6022 or (310) 318-1322 DIALUP (310) 318-5302 (login: public) > FAX: (310) 318-2162 EMAIL QUERY query@mmb.com > > > > From bigbug.franken.de!tsbogend@hub-fue.franken.de Thu May 9 18:42:06 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 SAA21566 for ; Thu, 9 May 1996 18:42:05 +0200 (MET DST) Received: from rachael.franken.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA08768; Thu, 9 May 1996 18:42:00 +0200 (MET) Received: from hub-fue.franken.de by rachael.franken.de with smtp (Smail3.1.29.1 #8) id m0uHYmT-000oLbC; Thu, 9 May 96 18:41 MET DST Received: by hub-fue.franken.de (Smail3.1.29.1 #35) id m0uHYmN-0005W2C; Thu, 9 May 96 18:41 MET DST Received: by bigbug.franken.de (8.6.12/tbs) id SAA00198 for linux-mips@fnet.fr; Thu, 9 May 1996 18:39:06 +0200 From: Thomas Bogendoerfer Message-Id: <199605091639.SAA00198@bigbug.franken.de> Subject: Re: GCC crosscompiler problem on Alpha To: linux-mips@fnet.fr Date: Thu, 9 May 1996 18:39:06 +0200 (MET DST) In-Reply-To: <199605081925.VAA20464@informatik.uni-koblenz.de> from "Systemkennung Linux" at May 8, 96 09:23:15 pm Reply-To: tsbogend@bigbug.franken.de X-Mailer: ELM [version 2.4 PL24 ME8b] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 853 Lines: 25 > Interesting; acutally GAS doesn't know dli yet at all and should complain. > One of a couple of bugs that prevent us from really going 64bit. ?? Gas from binutils must know it (and it does). Also gcc 2.7.2 produces dli statements. Look at this (it's from the 386 gcc, but my alpha gcc produces the same): 1008 0800 01001224 li $18,0x00000001 # 1 1009 0804 0000143C la $20,task 1009 00009466 1010 080c 01001324 dli $19,0x100000000 1010 3C981300 1011 0814 2D180000 move $3,$0 this is from mm/vmscan.c function swap_out. > This is indeed a bug in the assembler. which one ? alpha or 386 assembler ? Thomas. -- That process _deserves_ to die ("My name is Linus Torvalds, prepare to die"). [Linus Torvalds on linux-kernel] From tor@spacetec.no Thu May 9 23:05:49 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 XAA26968 for ; Thu, 9 May 1996 23:05:41 +0200 (MET DST) Received: from pallas.spacetec.no by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA12954; Thu, 9 May 1996 23:05:37 +0200 (MET) Received: (from tor@localhost) by pallas.spacetec.no (8.7.1/8.7.1) id VAA31481 for linux-mips@fnet.fr; Thu, 9 May 1996 21:01:27 +0200 Message-Id: <199605091901.VAA31481@pallas.spacetec.no> From: tor@spacetec.no (Tor Arntsen) Date: Thu, 9 May 1996 21:01:26 +0200 In-Reply-To: Dom Sweetman "Re: m700-10 success" (May 9, 21:34) X-Mailer: Mail User's Shell (7.2.5 10/14/92) To: linux-mips@fnet.fr Subject: BE/LE (Was Re: m700-10 success) Content-Length: 1988 Lines: 41 Thanks, the info is appreciated. Just a couple of comments: On May 9, 21:34, Dom Sweetman wrote: >o Workstations/computers from SGI and their "friends" Pyramid and > Siemens-Nixdorf. All big-endian, probably not adaptable. SGI's > success means that this is much the largest group of machines. I believe Larry McVoy mentioned in passing (on the sparc list) that SGI *could* do LE (or 'that other byte order' as he said :-) if necessary, although he certainly didn't mean bi-endian I think. [..] >Where should Linux be? LE is more like a x86, so will cause less >porting trouble. BE is more like SGI, but if you're planning on >running SGI binaries there's an awful lot of work to do - starting >with Unix SVR4-style shared libraries and running up through the SGI >3D graphics system. If you're not planning to run SGI binaries, who >cares? I have lots of software that I have carefully written (or made sure others have written) in a way portable enough to let me compile it without problems under Linux. I nfs-mount the whole disk with hundreds of megs of code and start the build, and it works fine. The problem comes when I execute the code: I want to share the data too, and I can't do that today without writing code to painfully byteswap field by field inside all the structs.. and I know this is a real pain to do, because I have already written the code for it for the (relatively small) part of the data which is sent out of the system (must be VAX-readable). (I use SGIs not for the - otherwise nice - graphics, but for (satellite) data processing purposes) >As for emulating Ultrix over Linux, or MIPS/ABI over Linux... sounds >pretty yukky to me. But with lots of effort it might let you build LE >and BE Linux kernels, but run the same binaries on both. Any heroes >out there? Bi-endian is not an issue for me, I would be perfectly happy with a BE-only Linux/MIPS box, but of course a LE Linux/MIPS box is better than no Linux/MIPS! Cheers, -Tor From linux@mailhost.uni-koblenz.de Thu May 9 21:22:03 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 VAA24781 for ; Thu, 9 May 1996 21:22:03 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA11263; Thu, 9 May 1996 21:21:53 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id VAA27538 for ; Thu, 9 May 1996 21:21:51 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605091921.VAA27538@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id VAA04228; Thu, 9 May 1996 21:19:55 +0200 Subject: Re: TEAMate Web/BBS Server for Linux To: linux-mips@fnet.fr Date: Thu, 9 May 1996 21:19:54 +0200 (MET DST) In-Reply-To: <199605091552.RAA29069@bryant.intern.julia.de> from "Ralf Baechle" at May 9, 96 05:52:50 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 696 Lines: 17 Hi all, > > MMB TEAMate Newsletter - May 7, 1996 sorry for sending that damn spam again. I received a dozen copies of it yesterday from almost every mailinglist and on almost every account that carries the word "linux" in his name. Happy as I was about this friendly offer when I had some more copies of this wonderful advert this morning I felt that I had to mail them how delighted I am about their incredibly offer and hit the wrong button. Ralf (Now waiting for small stones, big stones, pointy stones, gravel ...) PS: I've got now the RM200 sittin here on my desk and yes, so far the thing looks good! I just wonder why SNI included a copy of NT with their delivery ... From linux@mailhost.uni-koblenz.de Thu May 9 21:30:21 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 VAA24955 for ; Thu, 9 May 1996 21:30:20 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA11414; Thu, 9 May 1996 21:30:07 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id VAA27757 for ; Thu, 9 May 1996 21:29:59 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605091929.VAA27757@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id VAA04351; Thu, 9 May 1996 21:28:03 +0200 Subject: Re: GCC crosscompiler problem on Alpha To: linux-mips@fnet.fr Date: Thu, 9 May 1996 21:28:02 +0200 (MET DST) In-Reply-To: <199605091639.SAA00198@bigbug.franken.de> from "Thomas Bogendoerfer" at May 9, 96 06:39:06 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 825 Lines: 21 Hi, > > Interesting; acutally GAS doesn't know dli yet at all and should complain. > > One of a couple of bugs that prevent us from really going 64bit. > > ?? Gas from binutils must know it (and it does). Also gcc 2.7.2 > produces dli statements. Look at this (it's from the 386 gcc, but > my alpha gcc produces the same): > > 1008 0800 01001224 li $18,0x00000001 # 1 > 1009 0804 0000143C la $20,task > 1009 00009466 > 1010 080c 01001324 dli $19,0x100000000 > 1010 3C981300 > 1011 0814 2D180000 move $3,$0 Sorry, I was wrong. Earlier versions of the GNU assembler didn't have this instruction. This problem was of no big importance anyway so I never checked again if someone has implemented it in the meantime or not. Ralf From dom@algor.co.uk Thu May 9 22:34:48 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 WAA26218 for ; Thu, 9 May 1996 22:34:46 +0200 (MET DST) Received: from moorgate.algor.co.uk by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA12520; Thu, 9 May 1996 22:34:40 +0200 (MET) Received: from gladsmuir.algor.co.uk (dom@gladsmuir.algor.co.uk [193.117.190.129]) by moorgate.algor.co.uk (8.7.4/8.7.3) with SMTP id VAA13110; Thu, 9 May 1996 21:34:35 +0100 (BST) From: Dom Sweetman Date: Thu, 9 May 1996 21:33:17 +0100 Message-Id: <439.199605092033@gladsmuir.algor.co.uk> To: linux-mips@fnet.fr Subject: Re: m700-10 success In-Reply-To: <199605091112.NAA30905@pallas.spacetec.no> References: <199605091112.NAA30905@pallas.spacetec.no> Reply-To: dom@algor.co.uk Content-Length: 5403 Lines: 116 > >PS: the "change endianness in user mode" feature was a late addition > > to the R3000 chip, by disgruntled MIPS engineers who wanted to be > > able to run DECstation binaries on MIPS Unix boxes. The system > > software to make the hardware feature usable was never included in > > any MIPS OS, and I think has never been used by anyone. > Interesting! I didn't know it could be done in user mode for MIPS, I > thought it had to be done no later than during boot (I seem to > remember a pin on the chip for this?). I know PPC can be switched > at any time, one task can run BE and the other LE -- at the same > time even. At least that's what I've been told. Perhaps it's worth explaining a bit more. Pretty much all MIPS CPUs ever have been configurable at startup for endianness, in some way or another. Choose big or little. The only difference in the CPU's operation is what byte lanes get used when you do partial word loads and stores. But the system around the CPU is often designed for one endianness. It's committed in the following ways: 1. The CPU bus interface logic has to be able to figure out what byte lanes are active during partial-word transfers. MIPS CPUs don't have straightforward per-byte "enable" signals (like x86's do); they give you a width code and a byte address. So to work out where the data is, you need to know whether the CPU is being big or little. 2. The CPU's endianness imposes an implicit byte order on the bytes travelling along the different byte-lanes of the CPU bus; for BE CPUs the low-addressed-byte is the one on D24-31, while for LE CPUs the low-addressed byte is on D0-7. So the hardware had better fit in, or anything outside the CPU which has a byte order (a diskette drive or network interface which does DMA, for instance) had better fit in. To be able to work either way around the system *must* attend to (1); (2), if wrong, can be painfully fixed around in software. There are three groups of MIPS machines you're likely to meet up with: o Workstations/computers from SGI and their "friends" Pyramid and Siemens-Nixdorf. All big-endian, probably not adaptable. SGI's success means that this is much the largest group of machines. o DECstations and Sony workstations. All little-endian, probably not adaptable. o PC's built during the great Windows/NT-on-MIPS scam. All little-endian, usually not adaptable. But very, very cheap. [Advertisement mode on: I've never met a MIPS board which can work either way around, except the ones Algorithmics make.] It used to be true that most 'embedded' MIPS applications seemed to use the chip BE. There's still a large majority of applications that way, but recently a number of really big-volume applications have been using little-endian. Where should Linux be? LE is more like a x86, so will cause less porting trouble. BE is more like SGI, but if you're planning on running SGI binaries there's an awful lot of work to do - starting with Unix SVR4-style shared libraries and running up through the SGI 3D graphics system. If you're not planning to run SGI binaries, who cares? Doing only LE would rule out an SGI port; doing only BE means those of you with Pica's and Deskstation PCs will need to throw them away. Better stick with both. So now, back to the 'reverse endianness in user mode' bit. This got added to the status register (MIPS for the CPU control register) just in time to appear on the R3000 CPU. What it does is make a BE CPU act LE when it's in user mode (and vice versa); and was intended to be implemented on a per-process basis in Unix, to allow a BE operating system (ie MIPS Unix System V) to run LE binaries (ie DECstation Ultrix binaries). This shouldn't work (because partial-word transfers in this mode will confuse the bus interface logic); but it's OK because user programs usually only read/write cacheable locations, so only transfer data in whole-word or bigger units. The software results are pretty nasty, because the BE view of LE data is consistent only for aligned objects of CPU-bus-width size - doublewords on R4x00, words on R3000. *Everything* which passes between kernel and application needs to be swapped byte-within-[word or doubleword] (to restore consistent byte addressing) and then swapped again in a data dependent way (to resolve the difference between LE and BE ideas of shorts, integers, floats). As someone said, the fact that Linux is living on a crummy architecture (x86) where the kernel can't read from user-space pointers may help localise the problem. In practice, the software job of emulating enough of the Ultrix system call interface over SystemV was too big for MIPS - the idea was a symptom of the megalomania which ultimately led the company to the edge of bankruptcy (and into SGI's, fortunately friendly, arms). As for emulating Ultrix over Linux, or MIPS/ABI over Linux... sounds pretty yukky to me. But with lots of effort it might let you build LE and BE Linux kernels, but run the same binaries on both. Any heroes out there? -- Regards, Dominic Sweetman phone: +44 171 700 3301 Algorithmics Ltd home: +44 171 226 0032 3 Drayton Park fax: +44 171 700 3400 London N5 1NU email: dom@algor.co.uk ENGLAND. www: http://www.algor.co.uk ftp: ftp.algor.co.uk From andy@soft-n-hard.de Fri May 10 11:20:52 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 LAA01119 for ; Fri, 10 May 1996 11:20:46 +0200 (MET DST) Received: from aw.bnc.net by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA20566; Fri, 10 May 1996 08:02:14 +0200 (MET) Received: from newton.soft-n-hard.de (andy@newton.soft-n-hard.de [194.163.66.33]) by aw.bnc.net (8.6.12/8.6.12) with ESMTP id IAA13583 for ; Fri, 10 May 1996 08:06:42 +0200 Received: (from andy@localhost) by newton.soft-n-hard.de (8.6.12/8.6.9) id IAA20889; Fri, 10 May 1996 08:00:05 +0200 Date: Fri, 10 May 1996 08:00:05 +0200 From: Andreas Busse Message-Id: <199605100600.IAA20889@newton.soft-n-hard.de> To: linux-mips@fnet.fr Subject: Re: m700-10 success In-Reply-To: <439.199605092033@gladsmuir.algor.co.uk> X-Mailer: [XMailTool v3.1.2b] Content-Length: 1354 Lines: 35 Hi all, > > There are three groups of MIPS machines you're likely to meet up with: > > o Workstations/computers from SGI and their "friends" Pyramid and > Siemens-Nixdorf. All big-endian, probably not adaptable. SGI's > success means that this is much the largest group of machines. that's not true: The box we received from SNI runs WinNT (LE) as well as Sinix (BE). So it must be reconfigurable. > > o PC's built during the great Windows/NT-on-MIPS scam. All > little-endian, usually not adaptable. But very, very cheap. > > [Advertisement mode on: I've never met a MIPS board which can work > either way around, except the ones Algorithmics make.] Also, not true :-) Both Magnum and Olivetti (boxes actually designed for WinNT) can run in BE mode. At least for the Magnum I can confirm that since I've used a Magnum with Mips Risc/OS (BE) for years. SGI or Mips, resp. itself was working on a Bi-endian kernel (afaik) but after lots of nasty problems this project was stopped. Well, that's what I heard... Cheers, Andy ----------------------------------------------------------- Andreas Busse | andy@soft-n-hard.de Soft N Hard GbR | Phone: +49 2636-970105 Im Hufen Boden 16, D-53498 Waldorf | Fax: +49 2636-970106 ----------------------------------------------------------- From linux@mailhost.uni-koblenz.de Fri May 10 12:43:41 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 MAA02903 for ; Fri, 10 May 1996 12:43:39 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA25246; Fri, 10 May 1996 12:36:42 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id MAA06792 for ; Fri, 10 May 1996 12:34:48 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605101034.MAA06792@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id MAA09990; Fri, 10 May 1996 12:32:51 +0200 Subject: Re: m700-10 success To: linux-mips@fnet.fr Date: Fri, 10 May 1996 12:32:50 +0200 (MET DST) In-Reply-To: <439.199605092033@gladsmuir.algor.co.uk> from "Dom Sweetman" at May 9, 96 09:33:17 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 2803 Lines: 66 Hi, > There are three groups of MIPS machines you're likely to meet up with: > > o Workstations/computers from SGI and their "friends" Pyramid and > Siemens-Nixdorf. All big-endian, probably not adaptable. SGI's > success means that this is much the largest group of machines. At least the SNI RM200 is adaptable. Changing is a option in the firmware. If you stop the process of reprogramming the EEPROM you have to replace it physically; the process also takes about 5 minutes and involves fiddling with floppies. So this is not an option for Linux :-( > o DECstations and Sony workstations. All little-endian, probably not > adaptable. Add Acer PICA to this list. > o PC's built during the great Windows/NT-on-MIPS scam. All > little-endian, usually not adaptable. But very, very cheap. Deskstation Type, Deskstation rPC44. Not adaptable because based on a chipset for Intel PCs. > [Advertisement mode on: I've never met a MIPS board which can work > either way around, except the ones Algorithmics make.] Admited, the description of your boards look nice! > Where should Linux be? LE is more like a x86, so will cause less > porting trouble. BE is more like SGI, but if you're planning on > running SGI binaries there's an awful lot of work to do - starting > with Unix SVR4-style shared libraries and running up through the SGI > 3D graphics system. If you're not planning to run SGI binaries, who > cares? The bad thing is that the MIPS ABI to which IRIX5 is relativly close is big endian. This cuts the little endian machine off from the supply with shrink wrapped software for MIPS systems. > Doing only LE would rule out an SGI port; doing only BE means those of > you with Pica's and Deskstation PCs will need to throw them away. > Better stick with both. If you look through the code you'll find that there is quite some extra code for support of bigendian machines. > In practice, the software job of emulating enough of the Ultrix system > call interface over SystemV was too big for MIPS - the idea was a > symptom of the megalomania which ultimately led the company to the > edge of bankruptcy (and into SGI's, fortunately friendly, arms). Look through the system call table of Linux/MIPS and you'll find "similarities" with Mips' RISC/os and the IRIX extensions. The hooks are there but of course not all implemented. > As for emulating Ultrix over Linux, or MIPS/ABI over Linux... sounds > pretty yukky to me. But with lots of effort it might let you build LE > and BE Linux kernels, but run the same binaries on both. Any heroes > out there? I've already run big endian software on my "home edition of Linux/MIPS". It works but of course only with pretty trivial software - my program was assembler written "Bye, bye world!" program. Ralf From linux@mailhost.uni-koblenz.de Fri May 10 12:50:03 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 MAA03048 for ; Fri, 10 May 1996 12:50:02 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA25625; Fri, 10 May 1996 12:49:47 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id MAA07273 for ; Fri, 10 May 1996 12:49:30 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605101049.MAA07273@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id MAA10023; Fri, 10 May 1996 12:47:33 +0200 Subject: RM200 To: linux-mips@fnet.fr Date: Fri, 10 May 1996 12:47:33 +0200 (MET DST) Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 822 Lines: 19 Hi all, yesterday a played a bit with that SNI delivered RM200. Well, it's nice and well configured box except that it has no reset button on the outside of the box. Instead there is a reset and a "debug interrupt" button on the motherboard itself. Not good because NT crashed half a dozen times during installation ... The ARC BIOS however has a nice killerbug. It doesn't load the executable to the specified load address - it loads it to address + 8 bytes. Yikes. So after just playing around with that thing for some minutes I've already found the first capital bug. At least my theory of ARC BIOS == buggy and only useable for NT installation is proofed again. The M$ stuff is running on this BIOS because it's PIC (ARC doesn't require this) while Milo isn't. So I'll have to make Milo PIC now ... Ralf From dom@algor.co.uk Fri May 10 13:08:56 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 NAA03543 for ; Fri, 10 May 1996 13:08:55 +0200 (MET DST) Received: from relay2.EUnet.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA26068; Fri, 10 May 1996 13:08:52 +0200 (MET) Received: from moorgate.algor.co.uk by relay2.eunet.fr (5.65c8d/96.05.03) via EUnet-France id AA21125; Fri, 10 May 1996 13:06:25 +0200 (MET) Received: from bank.algor.co.uk (bank.algor.co.uk [193.117.190.11]) by moorgate.algor.co.uk (8.7.4/8.7.3) with SMTP id MAA01321; Fri, 10 May 1996 12:05:05 +0100 (BST) From: Dominic Sweetman Date: Fri, 10 May 96 12:05:04 +0100 Message-Id: <26115.9605101105@bank.algor.co.uk> To: linux-mips@fnet.fr Subject: Re: BE/LE (Was Re: m700-10 success) In-Reply-To: <199605091901.VAA31481@pallas.spacetec.no> References: <199605091901.VAA31481@pallas.spacetec.no> Content-Length: 2415 Lines: 49 > On May 9, 21:34, Dom Sweetman wrote: > >o Workstations/computers from SGI and their "friends" Pyramid and > > Siemens-Nixdorf. All big-endian, probably not adaptable. SGI's > > success means that this is much the largest group of machines. > > I believe Larry McVoy mentioned in passing (on the sparc list) that SGI > *could* do LE (or 'that other byte order' as he said :-) if necessary, > although he certainly didn't mean bi-endian I think. Yes, Andreas pointed that out too - and now I remember that SGI were going to provide Windows/NT as an alternative OS, or on really hallucinated days as a *replacement OS*. > I have lots of software that I have carefully written (or made sure > others have written) in a way portable enough to let me compile it > without problems under Linux. I nfs-mount the whole disk with > hundreds of megs of code and start the build, and it works fine. > The problem comes when I execute the code: I want to share the data > too, and I can't do that today without writing code to painfully > byteswap field by field inside all the structs.. and I know this is > a real pain to do, because I have already written the code for it > for the (relatively small) part of the data which is sent out of the > system (must be VAX-readable). The C standard doesn't allow you to assume that the memory image of data structures is portable between compilers, let alone between architectures. Endianness is one problem - but so are different numeric representations, size of data types and alignment rules. For most purposes it's better to export/import data in 'printable text' forms; but if this is really too slow then there are tools like Sun's XDR. [Do you remember XDR, invented somewhere in the early-mid 80's? it had tools which would auto-generate modules to dump/load arbitrary C data structures to a file-like stream of bytes. And it went with a remote-call thing too.] But if the $10M-per-year project which upgrades Word for Windows can't afford to do anything other than brain-dump binary data, it seems a bit unkind to expect unpaid volunteers to do so... Regards, Dominic Sweetman phone: +44 171 700 3301 Algorithmics Ltd home: +44 171 226 0032 3 Drayton Park fax: +44 171 700 3400 London N5 1NU email: dom@algor.co.uk ENGLAND. www: http://www.algor.co.uk ftp: ftp.algor.co.uk From tor@spacetec.no Fri May 10 15:25:55 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 PAA06488 for ; Fri, 10 May 1996 15:25:54 +0200 (MET DST) Received: from relay2.EUnet.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA00234; Fri, 10 May 1996 15:25:50 +0200 (MET) Received: from pallas.spacetec.no by relay2.eunet.fr (5.65c8d/96.05.03) via EUnet-France id AA28993; Fri, 10 May 1996 15:25:26 +0200 (MET) Received: (from tor@localhost) by pallas.spacetec.no (8.7.1/8.7.1) id NAA31928 for linux-mips@fnet.fr; Fri, 10 May 1996 13:06:32 +0200 Message-Id: <199605101106.NAA31928@pallas.spacetec.no> From: tor@spacetec.no (Tor Arntsen) Date: Fri, 10 May 1996 13:06:31 +0200 In-Reply-To: Systemkennung Linux "Re: m700-10 success" (May 8, 5:27) X-Mailer: Mail User's Shell (7.2.5 10/14/92) To: linux-mips@fnet.fr Subject: Re: m700-10 success Content-Length: 685 Lines: 19 On May 8, 5:27, Systemkennung Linux wrote: [I wrote:] >> So what I had to do to get up and running was to disable the part of the code >> in 3c503.c (el2_probe) that probes for shared memory. This is because the >> 3c503 code tries shared memory probing first before it falls back to ioport >> probing. The code piece I'm talking about is the one that starts with > >This is a bit weired because the address the the readl() accesses should >be correctly mapped in the TLB. Could you send me your crash messages, >please? Ok, rebuilt with original 3c503.c: Got dbe at ffffffff80093010. Unable to handle kernel paging request at virtual address 00000000, \ epc == 00000000 -Tor From imp@rover.village.org Fri May 10 16:01:45 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 QAA07433 for ; Fri, 10 May 1996 16:01:35 +0200 (MET DST) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA01113; Fri, 10 May 1996 16:01:20 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.7.5/8.6.6) with SMTP id IAA10930 for ; Fri, 10 May 1996 08:01:05 -0600 (MDT) Message-Id: <199605101401.IAA10930@rover.village.org> To: linux-mips@fnet.fr Subject: Progress Date: Fri, 10 May 1996 08:01:04 -0600 From: Warner Losh Content-Length: 1791 Lines: 52 OK. It isn't much, but I've managed to get up through .64 to book correctly. I have all the sources in CVS now and am bringing the versions in one at a time. .96 definitely doesn't get interrupts on my machine, even with the extra patches I've not sent off to Ralf yet that were in .62 to make the keyboard happy. Warner P.S. Here ore those same patches. Notice that I need to just enable scan mode 2 and my keyboard is golden. The entire reset sequence in initialize_kbd() breaks my keyboard. Also, this is relative to .65, not .96 so you may have to apply it by hand. I was torn between checking for just the Deskstation and not doing the init there, or doing the init for only the PICA and the Magnum. The latter won out. Index: drivers/char/keyboard.c =================================================================== RCS file: /home/imp/Repository/linux/drivers/char/keyboard.c,v retrieving revision 1.1.1.3 diff -u -r1.1.1.3 keyboard.c --- keyboard.c 1996/05/05 07:29:22 1.1.1.3 +++ keyboard.c 1996/05/10 04:44:31 @@ -1270,8 +1270,17 @@ /* * These two machines address their keyboard via the normal * port address range. + * + * Also enable Scan Mode 2. */ request_region(0x60,16,"kbd"); + kb_wait(); + kbd_outb(0x60, 0x64); /* 60 == PS/2 MODE ?? */ + kb_wait(); + kbd_outb(0x41, 0x60); /* 4d:same as freebsd, 41:KCC+EKI */ + kb_wait(); + if (!send_data(0xf0) || !send_data(0x02)) + printk("Scanmode 2 change failed\n"); } if (mips_machgroup == MACH_GROUP_JAZZ) { /* @@ -1282,7 +1291,9 @@ } #endif /* defined (__mips__) */ #ifdef INIT_KBD - initialize_kbd(); + if (mips_machtype == MACH_ACER_PICA_61 || + mips_machtype == MACH_MIPS_MAGNUM_4000) + initialize_kbd(); #endif mark_bh(KEYBOARD_BH); enable_bh(KEYBOARD_BH); From linux@mailhost.uni-koblenz.de Fri May 10 22:53:21 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id WAA15870 for ; Fri, 10 May 1996 22:53:21 +0200 (MET DST) Received: from relay2.EUnet.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA17174; Fri, 10 May 1996 22:53:18 +0200 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by relay2.eunet.fr (5.65c8d/96.05.03) via EUnet-France id AA26338; Fri, 10 May 1996 22:01:31 +0200 (MET) Received: from thoma..uni-koblenz (thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id VAA09224 for ; Fri, 10 May 1996 21:59:37 +0200 (MET DST) Received: by thoma..uni-koblenz (5.x/SMI-SVR4) id AA02796; Fri, 10 May 1996 21:57:43 +0200 From: linux@mailhost.uni-koblenz.de (Linux Support) Message-Id: <9605101957.AA02796@thoma..uni-koblenz> Subject: Re: m700-10 success To: linux-mips@fnet.fr Date: Fri, 10 May 1996 21:57:43 +0200 (MET DST) In-Reply-To: <199605101106.NAA31928@pallas.spacetec.no> from "Tor Arntsen" at May 10, 96 01:06:31 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1554 Lines: 39 > > On May 8, 5:27, Systemkennung Linux wrote: > [I wrote:] > >> So what I had to do to get up and running was to disable the part of the code > >> in 3c503.c (el2_probe) that probes for shared memory. This is because the > >> 3c503 code tries shared memory probing first before it falls back to ioport > >> probing. The code piece I'm talking about is the one that starts with > > > >This is a bit weired because the address the the readl() accesses should > >be correctly mapped in the TLB. Could you send me your crash messages, > >please? > > Ok, rebuilt with original 3c503.c: > > Got dbe at ffffffff80093010. Ouch, a data bus error. This says that the kernel tries to access non- existant memory but the motherboard refuses to serve such requests. Could you send me the output of the disassembler of this kernel image? (Don't disassemble arch/mips/boot/zImage which has only four symbols left, instead use ./vmlinux) About the following command should send me what I need: mipsel-linux-objdump -d --reloc --start-address=0x80092f00 \ --stop-address=0x80093100 vmlinux > Unable to handle kernel paging request at virtual address 00000000, \ > epc == 00000000 This one is bad; it look as if the kernel tried to restore a corrupted stackframe and then jumped to address zero which is unmappped. The CPU then takes a double TLB exception and you get your message. Pretty impossible to say why exactly this happend, but probably because due to the dbe exception the kernel tried to throw away process 0 ... Happy hacking, Ralf From linux@mailhost.uni-koblenz.de Fri May 10 22:53:02 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id WAA15830 for ; Fri, 10 May 1996 22:53:01 +0200 (MET DST) Received: from relay2.EUnet.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA17013; Fri, 10 May 1996 22:52:58 +0200 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by relay2.eunet.fr (5.65c8d/96.05.03) via EUnet-France id AA26457; Fri, 10 May 1996 22:04:18 +0200 (MET) Received: from thoma..uni-koblenz (thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id WAA09492 for ; Fri, 10 May 1996 22:02:49 +0200 (MET DST) Received: by thoma..uni-koblenz (5.x/SMI-SVR4) id AA02808; Fri, 10 May 1996 22:00:55 +0200 From: linux@mailhost.uni-koblenz.de (Linux Support) Message-Id: <9605102000.AA02808@thoma..uni-koblenz> Subject: Re: BE/LE (Was Re: m700-10 success) To: linux-mips@fnet.fr Date: Fri, 10 May 1996 22:00:55 +0200 (MET DST) In-Reply-To: <26115.9605101105@bank.algor.co.uk> from "Dominic Sweetman" at May 10, 96 12:05:04 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 430 Lines: 13 Hi Dom, > But if the $10M-per-year project which upgrades Word for Windows can't > afford to do anything other than brain-dump binary data, it seems a > bit unkind to expect unpaid volunteers to do so... If this'd be a competition I'd say the Linux/*BSD/GNU/ etc. projects are the clear winner. Optimized for maximum results with minimum money put in while "results" is not measured in units of money. Just my $0.02, Ralf From linux@mailhost.uni-koblenz.de Fri May 10 22:53:04 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id WAA15836 for ; Fri, 10 May 1996 22:53:04 +0200 (MET DST) Received: from relay2.EUnet.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA17041; Fri, 10 May 1996 22:53:01 +0200 (MET) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by relay2.eunet.fr (5.65c8d/96.05.03) via EUnet-France id AA26638; Fri, 10 May 1996 22:07:01 +0200 (MET) Received: from flake.uni-koblenz.de (linux@flake.uni-koblenz.de [141.26.4.37]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id WAA09639 for ; Fri, 10 May 1996 22:05:43 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605102005.WAA09639@informatik.uni-koblenz.de> Received: by flake.uni-koblenz.de (4.1/KO-2.0) id AA07443; Fri, 10 May 96 22:04:32 +0200 Subject: Re: Progress To: linux-mips@fnet.fr Date: Fri, 10 May 1996 22:04:31 +0200 (MET DST) In-Reply-To: <199605101401.IAA10930@rover.village.org> from "Warner Losh" at May 10, 96 08:01:04 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 556 Lines: 15 Hi all, > Warner > > P.S. Here ore those same patches. Notice that I need to just enable > scan mode 2 and my keyboard is golden. The entire reset sequence in > initialize_kbd() breaks my keyboard. Also, this is relative to .65, > not .96 so you may have to apply it by hand. I was torn between > checking for just the Deskstation and not doing the init there, or > doing the init for only the PICA and the Magnum. The latter won out. The Jazz machines doesn't need this though there is some trange problem with the keyboard very rarely. Ralf From paul@suite.sw.oz.au Sat May 11 08:49:52 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id IAA29236 for ; Sat, 11 May 1996 08:49:50 +0200 (MET DST) Received: from staff.cs.su.OZ.AU by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA02812; Sat, 11 May 1996 08:49:45 +0200 (MET) Received: from suite.sw.oz.au by swallow.sw.oz.au with SMTP id GAA02154; Sat, 11 May 1996 06:49:29 GMT (8.6.10/Unixware) (from paul@suite.sw.oz.au for ) Received: by suite.sw.oz.au id AA20926; Sat, 11 May 1996 16:49:19 +1000 (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: <199605110649.AA20926@suite.sw.oz.au> Subject: Re: Magnum != Olivetti To: linux-mips@fnet.fr Date: Sat, 11 May 1996 16:49:19 +1000 (EST) In-Reply-To: <199605090703.JAA02811@bandsept.univ-evry.fr> from "Stoned Elipot" at May 9, 96 09:03:03 am Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 1568 Lines: 38 Stoned wrote: > Another subject: > I had split init_IRQ() in arch/mips/kernel/irq.c in > init_IRQ_jazz(), init_IRQ_dec(), init_IRQ_arc(). Ah yes in my attempt > to re-organize the mips_machgroup stuff I made a group called ARC with > for now hold Deskstation boxes, and the is also a Jazz boxes group > To be clearer: each group is a single bit im mips_machgroup. > > mips_machgroup: > 0000000000000000000000000000000 group unknown > 0000000000000000000000000000001 group ARC > 0000000000000000000000000000010 group DEC > 0000000000000000000000000000100 group Jazz > ??????????????????????????????? group SGI ? > ??????????????????????????????? group SNI ? (if nedded) > > And then each models in each group use in turn 0,1,2,3,... for mips_machtype. > So for Dec we'll be able to identify all this 3mxx, pmax,... boxes > with their original/Digital model number. > > Comments ? It always surprises me when my suggestions are taken so seriously :-) Sounds fine - thanks! When you've got these changes in, I'll begin to re-integrate the DEC goo 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 Sat May 11 08:55:09 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id IAA29366 for ; Sat, 11 May 1996 08:55:08 +0200 (MET DST) Received: from staff.cs.su.OZ.AU by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA02923; Sat, 11 May 1996 08:55:03 +0200 (MET) Received: from suite.sw.oz.au by swallow.sw.oz.au with SMTP id GAA02282; Sat, 11 May 1996 06:54:52 GMT (8.6.10/Unixware) (from paul@suite.sw.oz.au for ) Received: by suite.sw.oz.au id AA20955; Sat, 11 May 1996 16:54:51 +1000 (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: <199605110654.AA20955@suite.sw.oz.au> Subject: Re: Magnum != Olivetti To: linux-mips@fnet.fr Date: Sat, 11 May 1996 16:54:50 +1000 (EST) In-Reply-To: <199605090841.KAA03045@bandsept.univ-evry.fr> from "Stoned Elipot" at May 9, 96 10:41:32 am Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 1134 Lines: 22 Stoned wrote: > The idea is Magnum 4000 and Olivetti belong to mips_machgroup = MACH_GROUP_JAZZ > in my fixed scheme, but mips_machtype is different for Magnum and Oily (as > for kernel 1.3.96 they are the same). So we could just test for mips_machgroup > = MACH_GROUP_JAZZ for features supported/working the same way for both machine > (and note that Acer Pica 61 is also in group JAZZ) and then for things like > the serial driver just make a futher test for mips_machtype= OILY or MAGNUM. It's interesting that my request for hierarchical machine definitions should prove so useful to others so quickly... or is this 'hack' simply a sad statement about Olivetti in general? :-) 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 Stoned.Elipot@univ-evry.fr Sat May 11 13:30:18 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id NAA04972 for ; Sat, 11 May 1996 13:30:17 +0200 (MET DST) Received: from amiante.univ-evry.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA05890; Sat, 11 May 1996 13:30:14 +0200 (MET) Received: from bandsept.univ-evry.fr (bandsept.univ-evry.fr [192.93.214.19]) by amiante.univ-evry.fr with ESMTP (8.6.12/96.04.09/louis-blanc); id NAA02393; Sat, 11 May 1996 13:30:14 +0200 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 NAA06735; Sat, 11 May 1996 13:29:32 +0200 Message-Id: <199605111129.NAA06735@bandsept.univ-evry.fr> To: linux-mips@fnet.fr Subject: Re: Magnum != Olivetti In-Reply-To: Your message of "Sat, 11 May 1996 16:54:50 +1000." <199605110654.AA20955@suite.sw.oz.au> X-It: bandsept (wrong OS, but right CPU) X-Mailer: MH 6.8.3, MH-E 5.0.2 Date: Sat, 11 May 1996 13:29:22 +0200 From: Stoned Elipot Content-Length: 532 Lines: 14 Hi Paul, Hi all, >>>>> "Paul" == Paul Antoine writes: [SNIP] Paul> It's interesting that my request for hierarchical machine Paul> definitions should prove so useful to others so quickly... or is Paul> this 'hack' simply a sad statement about Olivetti in general? Paul> :-) I should say that your request make Luc and I think a little bit more about what can be done with mips_machgroup which was in the first place wished by Andy in order to reduce some boring tests in the floppy driver :) Cheers, Stoned. From linux@mailhost.uni-koblenz.de Sat May 11 15:46:10 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id PAA07578 for ; Sat, 11 May 1996 15:46:09 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA07306; Sat, 11 May 1996 15:46:02 +0200 (MET) Received: from thoma..uni-koblenz (thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id PAA13675 for ; Sat, 11 May 1996 15:45:56 +0200 (MET DST) Received: by thoma..uni-koblenz (5.x/SMI-SVR4) id AA03299; Sat, 11 May 1996 15:44:03 +0200 From: linux@mailhost.uni-koblenz.de (Linux Support) Message-Id: <9605111344.AA03299@thoma..uni-koblenz> Subject: Re: Magnum != Olivetti To: linux-mips@fnet.fr Date: Sat, 11 May 1996 15:44:02 +0200 (MET DST) In-Reply-To: <199605110654.AA20955@suite.sw.oz.au> from "Paul Antoine" at May 11, 96 04:54:50 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 884 Lines: 19 Hi all, > It's interesting that my request for hierarchical machine definitions > should prove so useful to others so quickly... or is this 'hack' simply a > sad statement about Olivetti in general? :-) (We always knew that is useful - but someone how really needed that stuff had to implement it ...) The same is probably true for the Siemens Nixdorf machines. They're ranging from uniprocessor desktop machines upto 1024 processor machines. I've built a workaround into Milo that makes Milo useable even without PIC code (which again would require using a binutils snapshot ...) and I'm just building the first kernel for this thing. Overall the machine look closer to the standard PC architecture than all other jet supported machines; I think both Ethernet and SCSI chips are already supported by Linux, so we should have relatibly fast progress with this machine. Ralf From imp@rover.village.org Sat May 11 16:03:11 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id QAA08065 for ; Sat, 11 May 1996 16:03:10 +0200 (MET DST) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA07522; Sat, 11 May 1996 16:03:06 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.7.5/8.6.6) with SMTP id IAA23493 for ; Sat, 11 May 1996 08:03:04 -0600 (MDT) Message-Id: <199605111403.IAA23493@rover.village.org> To: linux-mips@fnet.fr Subject: Deskstation info In-Reply-To: Your message of Sat, 11 May 1996 15:44:02 +0200 Date: Sat, 11 May 1996 08:03:04 -0600 From: Warner Losh Content-Length: 640 Lines: 14 : so we should have relatibly fast progress with this machine. Go figure a change to the keyboard driver would fix linux's ability to mount the floppies on my machine. 1.3.65 paniced when I tried to mount root because it wasn't reading in the blocks correctly. 1.3.66 now mounts my root file system, but I'm missing a few things on the disk I have that keep me from getting to a # prompt. I'm off trying to build a sane root floppy now. If this works out as well as I think it will, then there is a very good chance that I'll have all of this on hard disk before the end of the weekend! Maybe even self-hosting! I'm excited! Warner From imp@rover.village.org Sat May 11 17:33:02 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id RAA09782 for ; Sat, 11 May 1996 17:33:01 +0200 (MET DST) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA08724; Sat, 11 May 1996 17:32:54 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.7.5/8.6.6) with SMTP id JAA01529 for ; Sat, 11 May 1996 09:32:50 -0600 (MDT) Message-Id: <199605111532.JAA01529@rover.village.org> Subject: Re: Deskstation info To: linux-mips@fnet.fr In-Reply-To: Your message of Sat, 11 May 1996 08:03:04 MDT Date: Sat, 11 May 1996 09:32:48 -0600 From: Warner Losh Content-Length: 762 Lines: 17 : Go figure a change to the keyboard driver would fix linux's ability to : mount the floppies on my machine. 1.3.65 paniced when I tried to : mount root because it wasn't reading in the blocks correctly. 1.3.66 : now mounts my root file system, but I'm missing a few things on the : disk I have that keep me from getting to a # prompt. I'm off trying : to build a sane root floppy now. Well, 1.3.66 gets farther, but still has problems after the first couple of successful reads, so the bug is still there. Bummer. At least I've backported ext2fs from FreeBSD-current to -stable so I can create root disks now. Oh well, back to the big march to 1.3.96. I had hoped for a while that 1.3.66 would have been the ticket for me to get the sh prompt. Warner From thompson@vulcan.edsi.org Sat May 11 20:30:59 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id UAA13282 for ; Sat, 11 May 1996 20:30:56 +0200 (MET DST) Received: from SISTER_RAY.EDSI.ORG (port3.edsi.org) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA11376; Sat, 11 May 1996 20:30:31 +0200 (MET) Received: from sister_ray (sister_ray [127.0.0.1]) by SISTER_RAY.EDSI.ORG (8.7.1/8.7.1) with SMTP id NAA01013 for ; Sat, 11 May 1996 13:30:24 -0500 Message-Id: <199605111830.NAA01013@SISTER_RAY.EDSI.ORG> Date: Sat, 11 May 96 13:30:28 -0500 Sender: thompson@SISTER_RAY.EDSI.ORG From: Paul Thompson X-Mailer: Mozilla 1.12 (X11; I; Linux 1.2.13 i486) Mime-Version: 1.0 To: linux-mips@fnet.fr Subject: MIPS X-Url: http://www.uni-koblenz.de/~ralf/linux-mips.html Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=us-ascii Content-Length: 409 Lines: 11 Greetings I have a salvaged-from-the-dumpster Decsystem5400 in perfect working order running ultrix 4.2a that I would be just tickled to get Linux running on. I do not, however, have any experience in C programming, although I have no trouble with the various cookbook style patches I have performed on my 486 running linux. Please keep me in mind, I would be happy to serve as a alpha/beta tester. Paul From imp@rover.village.org Sat May 11 22:17:39 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id WAA15417 for ; Sat, 11 May 1996 22:17:39 +0200 (MET DST) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA12488; Sat, 11 May 1996 22:17:32 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.7.5/8.6.6) with SMTP id OAA19745 for ; Sat, 11 May 1996 14:17:30 -0600 (MDT) Message-Id: <199605112017.OAA19745@rover.village.org> To: linux-mips@fnet.fr Subject: Bonehead linux question Date: Sat, 11 May 1996 14:17:30 -0600 From: Warner Losh Content-Length: 147 Lines: 5 Is there any way to say make XXX and have it behave just like make config using .config for all of your answers, and N for all new things? Warner From davem@caip.rutgers.edu Sat May 11 22:21:57 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id WAA15528 for ; Sat, 11 May 1996 22:21:56 +0200 (MET DST) Received: from caipfs.rutgers.edu by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA12543; Sat, 11 May 1996 22:21:53 +0200 (MET) Received: from huahaga.rutgers.edu (huahaga.rutgers.edu [128.6.155.53]) by caipfs.rutgers.edu (8.6.9+bestmx+oldruq+newsunq+grosshack/8.6.9) with ESMTP id QAA05507 for ; Sat, 11 May 1996 16:21:48 -0400 Received: (davem@localhost) by huahaga.rutgers.edu (8.6.9+bestmx+oldruq+newsunq+grosshack/8.6.9) id QAA23491; Sat, 11 May 1996 16:21:47 -0400 Date: Sat, 11 May 1996 16:21:47 -0400 Message-Id: <199605112021.QAA23491@huahaga.rutgers.edu> From: "David S. Miller" To: linux-mips@fnet.fr In-Reply-To: <199605112017.OAA19745@rover.village.org> (message from Warner Losh on Sat, 11 May 1996 14:17:30 -0600) Subject: Re: Bonehead linux question Content-Length: 364 Lines: 13 Date: Sat, 11 May 1996 14:17:30 -0600 From: Warner Losh Is there any way to say make XXX and have it behave just like make config using .config for all of your answers, and N for all new things? I think you want to have a look at the 'oldconfig' target in the toplevel makefile ;-) Later, David S. Miller davem@caip.rutgers.edu From linux@mailhost.uni-koblenz.de Sun May 12 00:15:23 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id AAA17905 for ; Sun, 12 May 1996 00:15:22 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA14298; Sun, 12 May 1996 00:12:44 +0200 (MET) Received: from thoma..uni-koblenz (thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id AAA25522 for ; Sun, 12 May 1996 00:12:40 +0200 (MET DST) Received: by thoma..uni-koblenz (5.x/SMI-SVR4) id AA03596; Sun, 12 May 1996 00:10:47 +0200 From: linux@mailhost.uni-koblenz.de (Linux Support) Message-Id: <9605112210.AA03596@thoma..uni-koblenz> Subject: Re: Deskstation info To: linux-mips@fnet.fr Date: Sun, 12 May 1996 00:10:46 +0200 (MET DST) In-Reply-To: <199605111532.JAA01529@rover.village.org> from "Warner Losh" at May 11, 96 09:32:48 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1629 Lines: 36 Hi, > : Go figure a change to the keyboard driver would fix linux's ability to > : mount the floppies on my machine. 1.3.65 paniced when I tried to > : mount root because it wasn't reading in the blocks correctly. 1.3.66 > : now mounts my root file system, but I'm missing a few things on the > : disk I have that keep me from getting to a # prompt. I'm off trying > : to build a sane root floppy now. > > Well, 1.3.66 gets farther, but still has problems after the first > couple of successful reads, so the bug is still there. Bummer. At > least I've backported ext2fs from FreeBSD-current to -stable so I can > create root disks now. > > Oh well, back to the big march to 1.3.96. I had hoped for a while > that 1.3.66 would have been the ticket for me to get the sh prompt. Oh well, without docs it's always a bit difficult. I just got the RM200 console up and running; I'll now have to find out how the whole interrupt thing works on this machine. Just couldn't wait for docs to arrive. The interesting thing in my work on the RM200 for you is that the machine as far as I know yet is pretty similar to the rPC44. PC style memory layout with hole from 640kb-1mb, all hardware accessible without using wired mappings - just in all aspect a PC with a RISC CPU. So stay tuned for my next patch. I've also made the big code cleanup and moved the board dependand stuff into four new subdirs. I'll just have to check if the whole mess still compiles for the Acer. But at least the thing looks much cleaner now. Hm... My private IP subnet will soon run out of IPs for my MIPS boxes ... Happy hacking, Ralf From linux@mailhost.uni-koblenz.de Sun May 12 01:49:15 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id BAA19651 for ; Sun, 12 May 1996 01:49:15 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA15426; Sun, 12 May 1996 01:48:55 +0200 (MET) Received: from thoma..uni-koblenz (thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id BAA29626 for ; Sun, 12 May 1996 01:48:37 +0200 (MET DST) Received: by thoma..uni-koblenz (5.x/SMI-SVR4) id AA03708; Sun, 12 May 1996 01:46:42 +0200 From: linux@mailhost.uni-koblenz.de (Linux Support) Message-Id: <9605112346.AA03708@thoma..uni-koblenz> Subject: Re: Deskstation info To: linux-mips@fnet.fr Date: Sun, 12 May 1996 01:46:42 +0200 (MET DST) In-Reply-To: <9605112210.AA03596@thoma..uni-koblenz> from "Linux Support" at May 12, 96 00:10:46 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 446 Lines: 11 Hi, > Oh well, without docs it's always a bit difficult. I just got the RM200 > console up and running; I'll now have to find out how the whole interrupt > thing works on this machine. Just couldn't wait for docs to arrive. Small update about the RM200 stuff: the box reaches 133.12 BogoMIPS. Seems IDT worked on the branch taken penatly for v2.0 of the R4600. So far this is the highes scoring MIPS box that I've tested myself :-) Ralf From paul@suite.sw.oz.au Sun May 12 03:04:44 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id DAA22741 for ; Sun, 12 May 1996 03:04:43 +0200 (MET DST) Received: from staff.cs.su.OZ.AU by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA16036; Sun, 12 May 1996 03:04:35 +0200 (MET) Received: from suite.sw.oz.au by swallow.sw.oz.au with SMTP id BAA21485; Sun, 12 May 1996 01:04:22 GMT (8.6.10/Unixware) (from paul@suite.sw.oz.au for ) Received: by suite.sw.oz.au id AA23979; Sun, 12 May 1996 11:04:20 +1000 (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: <199605120104.AA23979@suite.sw.oz.au> Subject: Another new DECStation kernel... To: linux-mips@fnet.fr (Linux MIPS mailing list) Date: Sun, 12 May 1996 11:04:19 +1000 (EST) Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 2382 Lines: 53 Hi folks, I've been playing with the 1.3.62 kernel for the DECStations some more, and have uploaded the latest version to my ftp site. This should be mirrored on fnet within a day of this message (right, Luc?) and has the same name as before: ftp://softway.com.au/pub/linux-mips/dec_vmlinux.1.3.62.test.gz Please try it out, as it does quite a bit more than the previous version, including accessing the clock correctly (at least on DS5000/2x's). I am trying to track down a continuing problem with the production of DEC-compatible ecoff binaries. It appears that gas may be the culprit... as whatever gas produces, the other binutils will happily create. I still can't see why there are two different types of ecoff-littlemips binaries though... e.g.: vmlinux: mipsel executable not stripped - version 2.11 vmlinux.2.6.2: mipsel executable not stripped ...the version 2.11's are the only ones the DECStation will load. In the meantime, anyone wanting to compile DECStation code is advised to build a version of binutils 2.6.x with the target as mips-dec-ultrix: ./configure --target=mips-dec-utlrix and to make the standard release of linux/MIPS gcc use these binutils (i.e. specifically as) to produce .o's until I figure out what's really happening. The other standard mips-linuxelf-* binutils (ar, nm, ld etc.) will then happily deal with the resultant .o's and produce archives and link objects in the DEC-compatible format provided you tell ld to expect ecoff-littlemips input and produce it as output: ld -b ecoff-littlemips -oformat ecoff-littlemips Ralf: I was going to wait a little bit until the latest round of kernel hacks had settled before re-integrating the DECStation goo. Could you code-freeze at some convenient point so I can grab a snapshot of the full tree and re-integrate... this should probably make our jobs a little easier, as I have plenty of patches to sort out. 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 dom@algor.co.uk Sun May 12 10:49:42 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id KAA01816 for ; Sun, 12 May 1996 10:49:42 +0200 (MET DST) Received: from moorgate.algor.co.uk by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA21473; Sun, 12 May 1996 10:49:35 +0200 (MET) Received: from gladsmuir.algor.co.uk (dom@gladsmuir.algor.co.uk [193.117.190.129]) by moorgate.algor.co.uk (8.7.4/8.7.3) with SMTP id JAA19497; Sun, 12 May 1996 09:49:31 +0100 (BST) From: Dom Sweetman Date: Sun, 12 May 1996 09:47:03 +0100 Message-Id: <169.199605120847@gladsmuir.algor.co.uk> To: linux-mips@fnet.fr Subject: Re: Deskstation info In-Reply-To: <9605112346.AA03708@thoma..uni-koblenz> References: <9605112210.AA03596@thoma..uni-koblenz> <9605112346.AA03708@thoma..uni-koblenz> Reply-To: dom@algor.co.uk Content-Length: 1468 Lines: 36 > Small update about the RM200 stuff: the box reaches 133.12 BogoMIPS. > Seems IDT worked on the branch taken penatly for v2.0 of the R4600. Like the original R3000 (but not the R4000 and R4400) the R4600 has a 5-stage pipeline, and always executes the instruction which follows any taken branch (the "branch delay slot"). So long as the branch delay slot instruction does useful work (and it usually can on the R4x00) there's no branch penalty for any R4600 CPU. The R4600 was a complete redesign of the R4000-without-secondary-cache made faster by: o Changing the 8-stage R4000/4400 pipeline back to 5 stages; o Bigger, cleverer (2x16Kbytes, 2-way set associative) caches; o On a data cache miss the R4600 restarts as soon as it gets the data it wants; the R4000/4400 waits for the whole line. It really did work. "Version 2.0" which Ralf referred to might be the R4700, an updated pin-compatible design. But I think the only tweaks in the R4700 are to the floating point unit. The real speed-up comes with the R5000, which all reports indicate is pretty neat. > So far this is the highes scoring MIPS box that I've tested myself > :-) There's an interesting point. Potential MIPS users often ask us what indication we can give them of performance for MIPS vs other architectures. Does anyone out there on planet Linux have some big-program "benchmarks" which give any leads to MIPS vs x86 system performance? Dominic Sweetman dom@algor.co.uk From linux@mailhost.uni-koblenz.de Sun May 12 13:38:49 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id NAA05185 for ; Sun, 12 May 1996 13:38:48 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA23084; Sun, 12 May 1996 13:38:42 +0200 (MET) Received: from thoma..uni-koblenz (thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id NAA21959 for ; Sun, 12 May 1996 13:38:40 +0200 (MET DST) Received: by thoma..uni-koblenz (5.x/SMI-SVR4) id AA04000; Sun, 12 May 1996 13:36:47 +0200 From: linux@mailhost.uni-koblenz.de (Linux Support) Message-Id: <9605121136.AA04000@thoma..uni-koblenz> Subject: Re: Deskstation info To: linux-mips@fnet.fr Date: Sun, 12 May 1996 13:36:46 +0200 (MET DST) In-Reply-To: <169.199605120847@gladsmuir.algor.co.uk> from "Dom Sweetman" at May 12, 96 09:47:03 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 2566 Lines: 60 Hi, > Like the original R3000 (but not the R4000 and R4400) the R4600 has a > 5-stage pipeline, and always executes the instruction which follows > any taken branch (the "branch delay slot"). So long as the branch > delay slot instruction does useful work (and it usually can on the > R4x00) there's no branch penalty for any R4600 CPU. > > The R4600 was a complete redesign of the R4000-without-secondary-cache > made faster by: > > o Changing the 8-stage R4000/4400 pipeline back to 5 stages; > o Bigger, cleverer (2x16Kbytes, 2-way set associative) caches; > o On a data cache miss the R4600 restarts as soon as it gets the data > it wants; the R4000/4400 waits for the whole line. > > It really did work. > > "Version 2.0" which Ralf referred to might be the R4700, an updated > pin-compatible design. But I think the only tweaks in the R4700 are > to the floating point unit. The real speed-up comes with the R5000, > which all reports indicate is pretty neat. No, it's definately a R4600. > > So far this is the highes scoring MIPS box that I've tested myself > > :-) > > There's an interesting point. Potential MIPS users often ask us what > indication we can give them of performance for MIPS vs other > architectures. Does anyone out there on planet Linux have some > big-program "benchmarks" which give any leads to MIPS vs x86 > system performance? Oh, I was talking about BogoMIPS which are - just as the name says - bogus. They're not a real benchmark, but a meassure of how fast .set noreorder 1: bnez reg,1b subu reg,1 is executed. This is internally used in the kernel for short delays. The loop is competly executed in the primay cache. The interesting is now that the first MIPS machine I ported Linux to - a Deskstation Tyne with a R4600/133MHz - needed three cycles to execute this loop while this machine seems to execute the loop in *one* cycle. From my under- standing this cannot have been caused by something outside the CPU, so it looks as if there were changes to the CPU. There are two versions of the R4600; my Tyne had version 1.0 as far as I remember while the SNI box has version 2.0. The 2.0 chip has some bugfixes; most important is probably the way it accesses primary caches. This bug can be worked around by disabeling interrupts during cacheflush or (untested ...) flushing way B before way A of the cache. What type of benchmark would you account as real world benchmark? If you need "realworld" benchmarks probably some guy at MTI might supply you more representative data than I could. Ralf From linux@mailhost.uni-koblenz.de Sun May 12 13:46:42 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id NAA05343 for ; Sun, 12 May 1996 13:46:41 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA23184; Sun, 12 May 1996 13:46:36 +0200 (MET) Received: from thoma..uni-koblenz (thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id NAA22108 for ; Sun, 12 May 1996 13:46:35 +0200 (MET DST) Received: by thoma..uni-koblenz (5.x/SMI-SVR4) id AA04012; Sun, 12 May 1996 13:44:42 +0200 From: linux@mailhost.uni-koblenz.de (Linux Support) Message-Id: <9605121144.AA04012@thoma..uni-koblenz> Subject: Re: Another new DECStation kernel... To: linux-mips@fnet.fr Date: Sun, 12 May 1996 13:44:41 +0200 (MET DST) In-Reply-To: <199605120104.AA23979@suite.sw.oz.au> from "Paul Antoine" at May 12, 96 11:04:19 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 779 Lines: 18 Hi, > Ralf: I was going to wait a little bit until the latest round of > kernel hacks had settled before re-integrating the DECStation goo. > Could you code-freeze at some convenient point so I can grab a > snapshot of the full tree and re-integrate... this should probably > make our jobs a little easier, as I have plenty of patches to sort out. ok, I have to admit that this latest round of kernel hacks is not very much the code you'll like. It sorts out some things into extra directories and will that way definately break your patches. The new setup however looks much cleaner and it'll take me quite a while to bring everything into a consistent state again. At the moment I wouldn't guaratee anything but that it's working for a RM200 ... So stay tuned, Ralf From linux@mailhost.uni-koblenz.de Sun May 12 18:28:26 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id SAA10884 for ; Sun, 12 May 1996 18:28:25 +0200 (MET DST) Received: from informatik.uni-koblenz.de ([141.26.4.1]) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA26020; Sun, 12 May 1996 18:28:18 +0200 (MET) Received: from thoma..uni-koblenz (thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id SAA28761 for ; Sun, 12 May 1996 18:27:44 +0200 (MET DST) Received: by thoma..uni-koblenz (5.x/SMI-SVR4) id AA04243; Sun, 12 May 1996 18:25:50 +0200 From: linux@mailhost.uni-koblenz.de (Linux Support) Message-Id: <9605121625.AA04243@thoma..uni-koblenz> Subject: RM200 success! To: linux-mips@fnet.fr Date: Sun, 12 May 1996 18:25:49 +0200 (MET DST) Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 141 Lines: 6 Hi all, I've just booted the RM200 successfully via NFS. Time for porting Linux/MIPS to RM200 so far: a bit more than two days :-> Ralf From paul@suite.sw.oz.au Sun May 12 23:26:59 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id XAA16778 for ; Sun, 12 May 1996 23:26:58 +0200 (MET DST) Received: from staff.cs.su.OZ.AU by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA00232; Sun, 12 May 1996 23:26:48 +0200 (MET) Received: from suite.sw.oz.au by swallow.sw.oz.au with SMTP id VAA11507; Sun, 12 May 1996 21:26:37 GMT (8.6.10/Unixware) (from paul@suite.sw.oz.au for ) Received: by suite.sw.oz.au id AA27074; Mon, 13 May 1996 07:26:33 +1000 (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: <199605122126.AA27074@suite.sw.oz.au> Subject: Re: RM200 success! To: linux-mips@fnet.fr Date: Mon, 13 May 1996 07:26:33 +1000 (EST) In-Reply-To: <9605121625.AA04243@thoma..uni-koblenz> from "Linux Support" at May 12, 96 06:25:49 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: 647 Lines: 16 Hi Ralf, > I've just booted the RM200 successfully via NFS. Time for porting > Linux/MIPS to RM200 so far: a bit more than two days :-> This is great - but you really must get some sleep!! :-) 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 May 13 00:20:21 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id AAA17991 for ; Mon, 13 May 1996 00:20:20 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA00990; Mon, 13 May 1996 00:19:15 +0200 (MET) Received: from thoma..uni-koblenz (thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id AAA08300 for ; Mon, 13 May 1996 00:19:02 +0200 (MET DST) Received: by thoma..uni-koblenz (5.x/SMI-SVR4) id AA04435; Mon, 13 May 1996 00:17:07 +0200 From: linux@mailhost.uni-koblenz.de (Linux Support) Message-Id: <9605122217.AA04435@thoma..uni-koblenz> Subject: Re: RM200 success! To: linux-mips@fnet.fr Date: Mon, 13 May 1996 00:17:07 +0200 (MET DST) In-Reply-To: <199605122126.AA27074@suite.sw.oz.au> from "Paul Antoine" at May 13, 96 07:26:33 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 350 Lines: 11 Hi Paul, > > I've just booted the RM200 successfully via NFS. Time for porting > > Linux/MIPS to RM200 so far: a bit more than two days :-> > > This is great - but you really must get some sleep!! :-) I had lots of sleep. Sometimes I'm happy that my 486 is slow like hell. Gives you lots of time for the unnecessary things of live ... Ralf From imp@rover.village.org Mon May 13 00:49:39 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id AAA18570 for ; Mon, 13 May 1996 00:49:38 +0200 (MET DST) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA01274; Mon, 13 May 1996 00:49:32 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.7.5/8.6.6) with SMTP id QAA07935 for ; Sun, 12 May 1996 16:49:31 -0600 (MDT) Message-Id: <199605122249.QAA07935@rover.village.org> To: linux-mips@fnet.fr Subject: Booting 1.3.96, kinda Date: Sun, 12 May 1996 16:49:29 -0600 From: Warner Losh Content-Length: 1146 Lines: 22 OK. I've managed to bring 1.3.96 up on the Deskstation to the same level that I've been able to get previous releases up to (namely the mounting of root on my floppy failing). I'll be sending Ralf the changes that I have. On disturbing trend in all of this is that my 1.3.62 kernel was 600273 bytes (as reported by size), while my 1.3.96 kernel (exact same config) is 626081 bytes. All the revs (almost) inbetween that I compiled were bigger than the previous ones. This bodes not well me thinks. At least the 640k limit is history for me, so I'm much less worried now than I was before. Oh well, time to see if I can get the BusLogic driver working so I can have SCSI. Maybe the floppy is just busted and the hard disk will be better (I fear this is wishful thinking on my part because I think there are DMA problems with the kernel on my machine that I've not yet fixed). Also noticing some oddities in the output of Ctrl-ScrollLock, namely there is a swapper process whose pid keeps getting bigger. It is a child of swapper process of pid 1. Maybe nothing to worry about, so I won't until I get the DMA issues ironed out. Warner From linux@mailhost.uni-koblenz.de Mon May 13 04:58:50 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id EAA24984 for ; Mon, 13 May 1996 04:58:48 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA03580; Mon, 13 May 1996 04:58:38 +0200 (MET) Received: from thoma..uni-koblenz (thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id EAA25267 for ; Mon, 13 May 1996 04:58:29 +0200 (MET DST) Received: by thoma..uni-koblenz (5.x/SMI-SVR4) id AA04619; Mon, 13 May 1996 04:56:35 +0200 From: linux@mailhost.uni-koblenz.de (Linux Support) Message-Id: <9605130256.AA04619@thoma..uni-koblenz> Subject: Re: Booting 1.3.96, kinda To: linux-mips@fnet.fr Date: Mon, 13 May 1996 04:56:34 +0200 (MET DST) In-Reply-To: <199605122249.QAA07935@rover.village.org> from "Warner Losh" at May 12, 96 04:49:29 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1687 Lines: 37 Hi Warner, > OK. I've managed to bring 1.3.96 up on the Deskstation to the same > level that I've been able to get previous releases up to (namely the > mounting of root on my floppy failing). I'll be sending Ralf the > changes that I have. Ok, I'll try to integrate them into my source tree though this is now a bit more difficult ... > On disturbing trend in all of this is that my 1.3.62 kernel was 600273 > bytes (as reported by size), while my 1.3.96 kernel (exact same > config) is 626081 bytes. All the revs (almost) inbetween that I > compiled were bigger than the previous ones. This bodes not well me > thinks. At least the 640k limit is history for me, so I'm much less > worried now than I was before. The workaround for the 640k problem (Bill is watching you ...) is buggy in your release. When freeing the memory I've freed ALL memory from 0-640kb - even the copy of the exception handlers. This should have affected you yet ... > Oh well, time to see if I can get the BusLogic driver working so I can > have SCSI. Maybe the floppy is just busted and the hard disk will be > better (I fear this is wishful thinking on my part because I think > there are DMA problems with the kernel on my machine that I've not yet > fixed). Also noticing some oddities in the output of Ctrl-ScrollLock, > namely there is a swapper process whose pid keeps getting bigger. It > is a child of swapper process of pid 1. Maybe nothing to worry about, > so I won't until I get the DMA issues ironed out. Your kernel is trying to start /bin/sh again and again but fails each time trying ... You might really try to use NFS root; it's extremly usefull for development. Ralf From imp@rover.village.org Mon May 13 05:16:49 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id FAA25492 for ; Mon, 13 May 1996 05:16:47 +0200 (MET DST) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA03774; Mon, 13 May 1996 05:16:39 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.7.5/8.6.6) with SMTP id VAA00676 for ; Sun, 12 May 1996 21:16:39 -0600 (MDT) Message-Id: <199605130316.VAA00676@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Booting 1.3.96, kinda In-Reply-To: Your message of Mon, 13 May 1996 04:56:34 +0200 Date: Sun, 12 May 1996 21:16:39 -0600 From: Warner Losh Content-Length: 8401 Lines: 245 : Ok, I'll try to integrate them into my source tree though this is now : a bit more difficult ... After a mild panic attack re fs corruption, I have the patches ready. The exerimental EXT2FS feature of FreeBSD (when used in -stable rather than the more supported -current) seems to cause some mild file system corruptions sometimes for reasons unknown. I've backed out these changes :-(. : The workaround for the 640k problem (Bill is watching you ...) is buggy : in your release. When freeing the memory I've freed ALL memory from : 0-640kb - even the copy of the exception handlers. This should have : affected you yet ... Yes. It bit me! Right after it printed the banner for the linux version. I've fixed it, I believe. Also, it would free the first page of the ISA bus as well. : Your kernel is trying to start /bin/sh again and again but fails each : time trying ... Yep. That's what I figure. I think it is trying three times and can never find it any of those times. :-) Or trying three different programs (/bin/init /sbin/init and /bin/sh would be my guess). : You might really try to use NFS root; it's extremly usefull for development. I know. My machines are in the DMZ for the village's network, so I can't run NFS on them. I'm working on a scam to let me do that, however.... Warner P.S. Here are my patches. They add comments, fix bugs and cleanup a little of the deskstation support. Some have been posted before. The CONFIG_PROC_FS one is to allow me to compile a kernel w/o procfs. scsi driver still has problems, but I'm about to start compiling it in, so I've not fixed them. These diffs are relative to 1.3.96 from the ftp.fnet.fr server. If need be, I can still identify what, exactly, I was doing for each one, so feel free to ask. diff -r -u linux-96/arch/mips/Makefile linux/arch/mips/Makefile --- linux-96/arch/mips/Makefile Sun May 5 17:39:55 1996 +++ linux/arch/mips/Makefile Sat May 11 23:36:46 1996 @@ -100,6 +100,10 @@ # The DECStation boot PROM loads kernels at 0x80030000. # Jazz machines load the kernel at 0x80000000. # +# Deskstation machines can load their kernels anywhere, but if loaded at +# 0x80000000 they are subject to the familiar 640k limit. We cope by +# loading them at 1M. +# ifdef CONFIG_DECSTATION LINKFLAGS := $(LINKFLAGS) -T arch/mips/ld.script.dec else diff -r -u linux-96/arch/mips/kernel/irq.c linux/arch/mips/kernel/irq.c --- linux-96/arch/mips/kernel/irq.c Sun May 5 07:07:39 1996 +++ linux/arch/mips/kernel/irq.c Sat May 11 17:44:29 1996 @@ -372,22 +372,10 @@ setup_x86_irq(2, &irq2); break; case MACH_DESKSTATION_TYNE: - /* set the clock to 100 Hz */ - outb_p(0x34,0x43); /* binary, mode 2, LSB/MSB, ch 0 */ - outb_p(LATCH & 0xff , 0x40); /* LSB */ - outb(LATCH >> 8 , 0x40); /* MSB */ - request_region(0x20,0x20, "pic1"); - request_region(0xa0,0x20, "pic2"); - setup_x86_irq(2, &irq2); - - if (request_irq(2, no_action, SA_INTERRUPT, "cascade", NULL)) - printk("Unable to get IRQ2 for cascade\n"); - break; case MACH_DESKSTATION_RPC44: - /* - * For the moment just steal the TYNE support. In the - * future, we need to consider merging the two -- imp - */ + /* Assume that Tyne and rPC44 are the same here. */ + set_cp0_status(ST0_IM, + IE_IRQ4 | IE_IRQ3 | IE_IRQ2 | IE_IRQ1); /* set the clock to 100 Hz */ outb_p(0x34, 0x43); /* binary, mode 2, LSB/MSB, ch 0 */ outb_p(LATCH & 0xff , 0x40); /* LSB */ @@ -395,9 +383,6 @@ request_region(0x20,0x20, "pic1"); request_region(0xa0,0x20, "pic2"); setup_x86_irq(2, &irq2); - set_cp0_status(ST0_IM, IE_IRQ4 | IE_IRQ3 | IE_IRQ2 | IE_IRQ1); - if (request_irq(2, no_action, SA_INTERRUPT, "cascade", NULL)) - printk("Unable to get IRQ2 for cascade\n"); break; case MACH_DECSTATION: /* FIXME: should set up the clock as per above? */ diff -r -u linux-96/arch/mips/kernel/rpc44-c.c linux/arch/mips/kernel/rpc44-c.c --- linux-96/arch/mips/kernel/rpc44-c.c Sun May 5 00:16:36 1996 +++ linux/arch/mips/kernel/rpc44-c.c Thu May 9 14:48:21 1996 @@ -8,6 +8,4 @@ * Copyright (C) 1994, 1995 by Ralf Baechle */ #define DS_TYPE deskstation_rpc44_ -#define DS_MACHINE_TYPE MACH_DESKSTATION_RPC44 -#define DS_MACH_STR "Rpc44" #include "tyne-c.c" diff -r -u linux-96/arch/mips/kernel/setup.c linux/arch/mips/kernel/setup.c --- linux-96/arch/mips/kernel/setup.c Sun May 5 19:21:23 1996 +++ linux/arch/mips/kernel/setup.c Sun May 12 15:36:26 1996 @@ -412,15 +412,7 @@ if (atag) root_mountflags |= MS_RDONLY; - /* - * On the deskstation rpc44, for the moment, only the kernel is - * below 1M. xxx imp fixme MILO should set these, but doesn't - * seem to be. - */ - if (mips_machtype == MACH_DESKSTATION_RPC44) - memory_start = (unsigned long) KSEG0 + (1 << 20); - else - memory_start = (unsigned long) &_end; + memory_start = (unsigned long) &_end; atag = bi_TagFind(tag_command_line); if (atag) diff -r -u linux-96/arch/mips/kernel/tyne-c.c linux/arch/mips/kernel/tyne-c.c --- linux-96/arch/mips/kernel/tyne-c.c Sun May 5 00:23:44 1996 +++ linux/arch/mips/kernel/tyne-c.c Thu May 9 15:10:26 1996 @@ -12,12 +12,6 @@ #ifndef DS_TYPE #define DS_TYPE deskstation_tyne_ #endif -#ifndef DS_MACHINE_TYPE -#define DS_MACHINE_TYPE MACH_DESKSTATION_TYNE -#endif -#ifndef DS_MACH_STR -#define DS_MACH_STR "Tyne" -#endif #define MKNAME3(a,b) b##a #define MKNAME2(a,b) MKNAME3(a,b) diff -r -u linux-96/arch/mips/mips3/cache.S linux/arch/mips/mips3/cache.S --- linux-96/arch/mips/mips3/cache.S Sun May 5 18:50:55 1996 +++ linux/arch/mips/mips3/cache.S Thu May 9 16:22:36 1996 @@ -7,6 +7,8 @@ * * Copyright (C) 1994, 1995, 1996 by Ralf Baechle and Andreas Busse */ +#include + #include #include #include @@ -45,12 +47,14 @@ beqz t1,do_icache li t0,KSEG0 # delay slot +#ifdef CONFIG_PROC_FS /* * Some profiling ... */ lw v0,dflushes addiu v0,1 sw v0,dflushes +#endif /* * Writeback data cache, even lines */ @@ -106,12 +110,14 @@ do_icache: andi t1,a2,ICACHE beqz t1,done +#ifdef CONFIG_PROC_FS /* * Again some profiling ... */ lw v0,iflushes addiu v0,1 sw v0,iflushes +#endif /* * Flush instruction cache, even lines */ diff -r -u linux-96/arch/mips/mm/init.c linux/arch/mips/mm/init.c --- linux-96/arch/mips/mm/init.c Sun May 5 18:14:23 1996 +++ linux/arch/mips/mm/init.c Sun May 12 16:39:47 1996 @@ -267,18 +267,22 @@ /* * For the rPC44 we've reserved some memory too much. Free the - * memory from PAGE_SIZE to PAGE_OFFSET + 0xa0000 again. + * memory from PAGE_SIZE to PAGE_OFFSET + 0xa0000 again. Don't + * free anything between 0xa0000 and 0xf0000 because that's where + * hardware lives. Also, maybe other PC-like machines need this + * too. */ if (mips_machtype == MACH_DESKSTATION_RPC44) { - tmp = PAGE_OFFSET; - while (tmp <= PAGE_OFFSET + 0xa000) { + tmp = PAGE_OFFSET + PAGE_SIZE; + while (tmp < PAGE_OFFSET + 0xa000) { clear_bit(PG_reserved, &mem_map[MAP_NR(tmp)].flags); tmp += PAGE_SIZE; } } #ifdef CONFIG_DESKSTATION_TYNE - deskstation_tyne_dma_init(); + if (mips_machtype == MACH_DESKSTATION_TYNE) + deskstation_tyne_dma_init(); #endif for (tmp = PAGE_OFFSET ; tmp < high_memory ; tmp += PAGE_SIZE) { /* diff -r -u linux-96/drivers/char/keyboard.c linux/drivers/char/keyboard.c --- linux-96/drivers/char/keyboard.c Sun May 5 18:18:45 1996 +++ linux/drivers/char/keyboard.c Sat May 11 23:35:42 1996 @@ -1288,9 +1288,18 @@ /* * These two machines address their keyboard via the normal * port address range. + * + * Also enable Scan Mode 2. */ request_irq(KEYBOARD_IRQ, keyboard_interrupt, 0, "keyboard", NULL); request_region(0x60,16,"kbd"); + kb_wait(); + kbd_outb(0x60, 0x64); /* 60 == PS/2 MODE ?? */ + kb_wait(); + kbd_outb(0x41, 0x60); /* 4d:same as freebsd, 41:KCC+EKI */ + kb_wait(); + if (!send_data(0xf0) || !send_data(0x02)) + printk("Scanmode 2 change failed\n"); } if (mips_machgroup == MACH_GROUP_JAZZ) { /* @@ -1301,7 +1310,9 @@ } #endif /* defined (__mips__) */ #ifdef INIT_KBD - initialize_kbd(); + if (mips_machtype == MACH_ACER_PICA_61 || + mips_machtype == MACH_MIPS_MAGNUM_4000) + initialize_kbd(); #endif init_bh(KEYBOARD_BH, kbd_bh); mark_bh(KEYBOARD_BH); From andy@soft-n-hard.de Mon May 13 08:16:52 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id IAA29028 for ; Mon, 13 May 1996 08:16:52 +0200 (MET DST) Received: from aw.bnc.net by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA06118; Mon, 13 May 1996 08:16:45 +0200 (MET) Received: from newton.soft-n-hard.de (andy@newton.soft-n-hard.de [194.163.66.33]) by aw.bnc.net (8.6.12/8.6.12) with ESMTP id IAA25112 for ; Mon, 13 May 1996 08:21:16 +0200 Received: (from andy@localhost) by newton.soft-n-hard.de (8.6.12/8.6.9) id IAA26173; Mon, 13 May 1996 08:14:36 +0200 Date: Mon, 13 May 1996 08:14:36 +0200 From: Andreas Busse Message-Id: <199605130614.IAA26173@newton.soft-n-hard.de> To: linux-mips@fnet.fr Subject: Re: RM200 success! In-Reply-To: <9605121625.AA04243@thoma..uni-koblenz> X-Mailer: [XMailTool v3.1.2b] Content-Length: 482 Lines: 16 > I've just booted the RM200 successfully via NFS. Time for porting > Linux/MIPS to RM200 so far: a bit more than two days :-> Can't believe it... Congrats, Ralf! Cheers, Andy ----------------------------------------------------------- Andreas Busse | andy@soft-n-hard.de Soft N Hard GbR | Phone: +49 2636-970105 Im Hufen Boden 16, D-53498 Waldorf | Fax: +49 2636-970106 ----------------------------------------------------------- From Stoned.Elipot@univ-evry.fr Mon May 13 10:03:58 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id KAA01288 for ; Mon, 13 May 1996 10:03:57 +0200 (MET DST) Received: from amiante.univ-evry.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA07663; Mon, 13 May 1996 10:03:53 +0200 (MET) Received: from bandsept.univ-evry.fr (bandsept.univ-evry.fr [192.93.214.19]) by amiante.univ-evry.fr with ESMTP (8.6.12/96.04.09/louis-blanc); id KAA05482; Mon, 13 May 1996 10:03:54 +0200 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 KAA09202; Mon, 13 May 1996 10:03:12 +0200 Message-Id: <199605130803.KAA09202@bandsept.univ-evry.fr> To: linux-mips@fnet.fr Subject: Re: RM200 success! In-Reply-To: Your message of "Mon, 13 May 1996 08:14:36 +0200." <199605130614.IAA26173@newton.soft-n-hard.de> X-It: bandsept (wrong OS, but right CPU) X-Mailer: MH 6.8.3, MH-E 5.0.2 Date: Mon, 13 May 1996 10:03:06 +0200 From: Stoned Elipot Content-Length: 528 Lines: 16 Hi all, >>>>> "Andreas" == Andreas Busse writes: [SNIP] Andreas> Can't believe it... Congrats, Ralf! Yep, what a score ! Hum,... I have a bunch of patch against 1.3.96, I understand that it will be difficult for you to integrate them in your source tree, so do you prefer that I'll wait till your next release Ralf? These are mostly concercing the machtype and machgroup stuff we were chatting about last week but they require milo 0.27 which is not out yet... (few more days please:). Cheers, Stoned. From dom@algor.co.uk Mon May 13 10:06:54 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id KAA01385 for ; Mon, 13 May 1996 10:06:53 +0200 (MET DST) Received: from moorgate.algor.co.uk by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA07681; Mon, 13 May 1996 10:06:40 +0200 (MET) Received: from gladsmuir.algor.co.uk (dom@gladsmuir.algor.co.uk [193.117.190.129]) by moorgate.algor.co.uk (8.7.4/8.7.3) with SMTP id JAA28367; Mon, 13 May 1996 09:06:15 +0100 (BST) From: Dom Sweetman Date: Mon, 13 May 1996 09:03:17 +0100 Message-Id: <173.199605130803@gladsmuir.algor.co.uk> To: linux-mips@fnet.fr Subject: Re: Deskstation info In-Reply-To: <9605121136.AA04000@thoma..uni-koblenz> References: <169.199605120847@gladsmuir.algor.co.uk> <9605121136.AA04000@thoma..uni-koblenz> Reply-To: dom@algor.co.uk Content-Length: 2052 Lines: 49 > [discussing speed of RM200 vs other R4600 machines, and BogoIPS:] > .set noreorder > 1: bnez reg,1b > subu reg,1 > > ... the first MIPS machine I ported Linux to - a Deskstation Tyne > with a R4600/133MHz - needed three cycles to execute this loop while > this machine seems to execute the loop in *one* cycle. Bizarre. It should take 2 cycles/loop on *any* R4600, and 4 cycles/loop on an R4000/4400. I'd put money on it *not* taking one cycle/loop - only an R10000 can do that! There are some unlikely reasons I can think of for the difference (eg the loop spans a page boundary and is in mapped space, when it will repeatedly have to do a micro-tlb refill - but I'm sure you're in unmapped space). > There are two versions of the R4600; my Tyne had version 1.0 as far as > I remember while the SNI box has version 2.0. The 2.0 chip has some > bugfixes; most important is probably the way it accesses primary caches. > This bug can be worked around by disabeling interrupts during cacheflush > or (untested ...) flushing way B before way A of the cache. I don't recall anything on the published Rev 1 buglist which would affect this. Now you've got me intrigued... I'd better go look this up! > What type of benchmark would you account as real world benchmark? Any modest set of reasonably large applications which can be compiled without change on a number of different platforms, and run and timed off test data. Perl/groff/emacs running scripts might be good. I hoped this might have been so straightforward between two Linux machines that someone had just done an informal run-off. > If you need "realworld" benchmarks probably some guy at MTI might supply > you more representative data than I could. MTI have commercial interests here, so it's not quite fair to expect them to be objective. > I've just booted the RM200 successfully via NFS. Time for porting ? Linux/MIPS to RM200 so far: a bit more than two days :-> Congratulations! we'd better try and find you one of our boards... Dominic Sweetman dom@algor.co.uk From hodgen@mailhost.uni-koblenz.de Mon May 13 11:21:39 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id LAA02812 for ; Mon, 13 May 1996 11:21:38 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA00435; Mon, 13 May 1996 11:21:19 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id KAA04951 for ; Mon, 13 May 1996 10:58:22 +0200 (MET DST) From: Wayne Hodgen Message-Id: <199605130858.KAA04951@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id KAA00705; Mon, 13 May 1996 10:56:21 +0200 Subject: Re: Deskstation info To: linux-mips@fnet.fr Date: Mon, 13 May 1996 10:56:20 +0200 (MET DST) In-Reply-To: <9605112210.AA03596@thoma..uni-koblenz> from "Linux Support" at May 12, 96 00:10:46 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 774 Lines: 18 Hi, > I've also made the big code cleanup and moved the board dependand stuff > into four new subdirs. I'll just have to check if the whole mess still > compiles for the Acer. But at least the thing looks much cleaner now. Sounds good, when do we get it? > Hm... My private IP subnet will soon run out of IPs for my MIPS boxes ... I can make it easier for you! Bring back my NE2000 and you won't have enough cards for the net, let alone IP's ;) -- Wayne Hodgen | hodgen@informatik.uni-koblenz.de | #include From linux@mailhost.uni-koblenz.de Mon May 13 11:46:28 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id LAA03269 for ; Mon, 13 May 1996 11:46:27 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA00952; Mon, 13 May 1996 11:42:32 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id LAA06872 for ; Mon, 13 May 1996 11:42:20 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605130942.LAA06872@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id LAA00946; Mon, 13 May 1996 11:40:18 +0200 Subject: Re: Deskstation info To: linux-mips@fnet.fr Date: Mon, 13 May 1996 11:40:17 +0200 (MET DST) In-Reply-To: <173.199605130803@gladsmuir.algor.co.uk> from "Dom Sweetman" at May 13, 96 09:03:17 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1273 Lines: 27 Hi, > Bizarre. It should take 2 cycles/loop on *any* R4600, and 4 > cycles/loop on an R4000/4400. I'd put money on it *not* taking one > cycle/loop - only an R10000 can do that! There are some unlikely > reasons I can think of for the difference (eg the loop spans a page > boundary and is in mapped space, when it will repeatedly have to do a > micro-tlb refill - but I'm sure you're in unmapped space). Yes, you're right; this effect is bizzare and cannot be explained with some pagefill effect because it is running in KSEG0. If this machine wouldn't be that PC-like I'd say the timers are running at a different clock. So I'll check this timer... > > There are two versions of the R4600; my Tyne had version 1.0 as far as > > I remember while the SNI box has version 2.0. The 2.0 chip has some > > bugfixes; most important is probably the way it accesses primary caches. > > This bug can be worked around by disabeling interrupts during cacheflush > > or (untested ...) flushing way B before way A of the cache. > > I don't recall anything on the published Rev 1 buglist which would > affect this. Now you've got me intrigued... I'd better go look this up! I wouldn't wonder if thisone isn't documented. Write me if you don't find it in the docs. Ralf From linux@mailhost.uni-koblenz.de Mon May 13 11:47:19 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id LAA03311 for ; Mon, 13 May 1996 11:47:18 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA00994; Mon, 13 May 1996 11:46:17 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id LAA06998 for ; Mon, 13 May 1996 11:46:08 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605130946.LAA06998@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id LAA00990; Mon, 13 May 1996 11:44:06 +0200 Subject: Re: RM200 success! To: linux-mips@fnet.fr Date: Mon, 13 May 1996 11:44:05 +0200 (MET DST) In-Reply-To: <199605130803.KAA09202@bandsept.univ-evry.fr> from "Stoned Elipot" at May 13, 96 10:03:06 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 329 Lines: 10 Hi, > These are mostly concercing the machtype and machgroup stuff we were > chatting about last week but they require milo 0.27 which is not out > yet... (few more days please:). Ok, I also had to make some patches to Milo to compensate the SNI ARC BIOS bug and recognize the machine; I'll send them to you for 0.27. Ralf From pefo@enea.se Mon May 13 13:43:08 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id NAA07971 for ; Mon, 13 May 1996 13:43:07 +0200 (MET DST) Received: from outside.enea.se (gateway.enea.se) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA03471; Mon, 13 May 1996 13:43:02 +0200 (MET) Received: by outside.enea.se (8.6.9/8.6.9) id NAA25493 for ; Mon, 13 May 1996 13:44:15 +0200 Received: by outside via smap (V1.3) id sma025489; Mon May 13 13:44:05 1996 Received: (from pefo@localhost) by enea.se (8.7.1/8.7.1) id NAA29298 for linux-mips@fnet.fr; Mon, 13 May 1996 13:42:35 +0200 (MET DST) From: Per Fogelstrom Message-Id: <199605131142.NAA29298@gordon.enea.se> Subject: Re: Deskstation info To: linux-mips@fnet.fr Date: Mon, 13 May 1996 13:42:34 +0200 (MET DST) In-Reply-To: <199605130942.LAA06872@informatik.uni-koblenz.de> from "Systemkennung Linux" at May 13, 96 11:40:17 am X-Mailer: ELM [version 2.4 PL23] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 574 Lines: 14 > > Yes, you're right; this effect is bizzare and cannot be explained with some > pagefill effect because it is running in KSEG0. If this machine wouldn't be > that PC-like I'd say the timers are running at a different clock. So I'll > check this timer... > Easy.. get the count register before and after the loop and check how many clock ticks it took, divide by number of loops... btw, mips shared lib stuff now runs really good on OpenBSD. the X11 bin dir shrinked to about 20% of what it was before. :-) now, if i had a big endian machine besides the pica... :-) From linux@mailhost.uni-koblenz.de Mon May 13 16:25:59 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id QAA11236 for ; Mon, 13 May 1996 16:25:58 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA02775; Mon, 13 May 1996 16:21:23 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id QAA06623 for ; Mon, 13 May 1996 16:19:15 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605131419.QAA06623@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id QAA04435; Mon, 13 May 1996 16:17:16 +0200 Subject: Re: Deskstation info To: linux-mips@fnet.fr Date: Mon, 13 May 1996 16:17:15 +0200 (MET DST) In-Reply-To: <199605131142.NAA29298@gordon.enea.se> from "Per Fogelstrom" at May 13, 96 01:42:34 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1934 Lines: 42 Hi, > Easy.. get the count register before and after the loop and check how > many clock ticks it took, divide by number of loops... Well, I have to correct myself. I didn't think of the fact that the BogoMIPS printk() accounts for the fact of the loop having two instructions. So my PICA's results and the SNI's results are correct but the ones that I on the Tyne are strange. Oh well, I did only the first steps with MIPS CPUs and the Linux port on that thing ... > btw, mips shared lib stuff now runs really good on OpenBSD. the X11 bin > dir shrinked to about 20% of what it was before. :-) Who wonders. I know that the stuff will shrink without end. But now that I have a 32m MIPS box the static binaries don't hurt that extremly anymore. I could have done the shared library stuff quite some time ago but I want to avoid working with a binutils snapshot. You know how these guys from the FSF and Cygnus think about publishing snapshots. You now how all these people'd like to get these binaries ... I don't track the BSD flavourism very closely, so why did you switch to OpenBSD? > now, if i had a big endian machine besides the pica... :-) I have one now; just the documentation for the big endian firmware from SNI is missing. Well, all documentation is missing for this machine; what I did is just based on reverse engineering and looking at the output of the configuration tools of NT ... Finally all the code that I wrote for the Tyne and rPC helped me alot so that doing the job in just two days was possible. It's just anoying that switching the SNI box between byteorders takes five minutes and is a dangerous procedure. A power failure and the firmware EEPROM has to be replaced physically. Time to brew something faster - but this is more for the hardware guys ... Anyway, the fun thing with the SNI box is that the SNI guys probably believe that I didn't unpack the machine yet or so :-> Ralf From pefo@enea.se Mon May 13 18:01:45 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id SAA13198 for ; Mon, 13 May 1996 18:01:44 +0200 (MET DST) Received: from outside.enea.se by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA04952; Mon, 13 May 1996 18:01:40 +0200 (MET) Received: by outside.enea.se (8.6.9/8.6.9) id SAA09754 for ; Mon, 13 May 1996 18:02:51 +0200 Received: by outside via smap (V1.3) id sma009751; Mon May 13 18:02:23 1996 Received: (from pefo@localhost) by enea.se (8.7.1/8.7.1) id SAA16304 for linux-mips@fnet.fr; Mon, 13 May 1996 18:00:53 +0200 (MET DST) From: Per Fogelstrom Message-Id: <199605131600.SAA16304@gordon.enea.se> Subject: Re: Deskstation info To: linux-mips@fnet.fr Date: Mon, 13 May 1996 18:00:52 +0200 (MET DST) In-Reply-To: <199605131419.QAA06623@informatik.uni-koblenz.de> from "Systemkennung Linux" at May 13, 96 04:17:15 pm X-Mailer: ELM [version 2.4 PL23] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 1420 Lines: 32 > > > btw, mips shared lib stuff now runs really good on OpenBSD. the X11 bin > > dir shrinked to about 20% of what it was before. :-) > > Who wonders. I know that the stuff will shrink without end. But now that > I have a 32m MIPS box the static binaries don't hurt that extremly anymore. Heh! When you start to port X you will change your mind! I also have 32Mb in my pica system but there is also disk space. If a program needs 1M disk or 100kb really matters when you start to have 100's of them. > I could have done the shared library stuff quite some time ago but I want > to avoid working with a binutils snapshot. You know how these guys from > the FSF and Cygnus think about publishing snapshots. You now how all these > people'd like to get these binaries ... > Sure you could have. But to what avail? Shared libs are not important until you can rebuild the system native. I've been able to do that for about a year now so shared libs was a natural next step. (after X11 and some other stuff). The snapshots are not perfect, but according to all the patches you've done to get 2.6.0 to work like you want it, "relesed" versions aren't either. However, having it working on OpenBSD helped to get several bugs out so when it's finally released it will be a more stable release. > I don't track the BSD flavourism very closely, so why did you switch to > OpenBSD? > I like the people. Have fun! Per From linux@mailhost.uni-koblenz.de Mon May 13 19:11:13 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id TAA01007 for ; Mon, 13 May 1996 19:11:10 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA06244; Mon, 13 May 1996 19:10:49 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.4/8.6.9) with SMTP id TAA13116 for ; Mon, 13 May 1996 19:10:35 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605131710.TAA13116@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id TAA06802; Mon, 13 May 1996 19:08:36 +0200 Subject: Re: Deskstation info To: linux-mips@fnet.fr Date: Mon, 13 May 1996 19:08:36 +0200 (MET DST) In-Reply-To: <199605131600.SAA16304@gordon.enea.se> from "Per Fogelstrom" at May 13, 96 06:00:52 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 861 Lines: 17 Hi, > Sure you could have. But to what avail? Shared libs are not important until > you can rebuild the system native. I've been able to do that for about a year > now so shared libs was a natural next step. (after X11 and some other stuff). > The snapshots are not perfect, but according to all the patches you've done > to get 2.6.0 to work like you want it, "relesed" versions aren't either. > However, having it working on OpenBSD helped to get several bugs out so when > it's finally released it will be a more stable release. Almost every userland binary that I have has been built native so shared libs ARE one of the next logic steps. Shared libs will also help to spread improvements in the libc much faster throughout the userland and thats a property of shared libc in which I'm very interested. Oh, and in making my uploads faster ;-) Ralf From imp@rover.village.org Tue May 14 17:39:25 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id RAA28990 for ; Tue, 14 May 1996 17:39:20 +0200 (MET DST) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA03345; Tue, 14 May 1996 17:38:59 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.7.5/8.6.6) with SMTP id JAA11279 for ; Tue, 14 May 1996 09:38:47 -0600 (MDT) Message-Id: <199605141538.JAA11279@rover.village.org> To: linux-mips@fnet.fr Subject: Floppy help??? Date: Tue, 14 May 1996 09:38:47 -0600 From: Warner Losh Content-Length: 1179 Lines: 27 OK. Now that I have something to hack on, I'm hacking on it :-) On boot the kernel will mount a floppy to be the root floppy. When it does this for an ext2fs file system, it will read track 0, head 0, sector 3 (aka t0h0s0) first. It will then read t0h0s5 and then s11. Once it has done this it pronounces the file system mounted. It then goes on to read s13 a couple of times. Then it reads t3h1s1, or should I say tries to read it. The checksum on that block is exactly identical to the checksum on t0h0s13, which I know from my FreeBSD machine it isn't. Basically, it looks like it is reading track 0 w/o any problems, both with and without the track buffering code enabled. However, when it goes to read another track the first sector read from that track seems to be bad (this is with the track caching code disabled), but additional sectors appear to be reading fine (their checksum is different). Now my question is why? Does anybody know enough about the floppy driver to tell me why this is happening? Or point me in the right direction? Also, does anybody else who has taken a look at the floopy driver find it to be a really twisted maze of code? Warner From tor@spacetec.no Tue May 14 20:55:31 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id UAA03234 for ; Tue, 14 May 1996 20:55:31 +0200 (MET DST) Received: from pallas.spacetec.no by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA06980; Tue, 14 May 1996 20:55:25 +0200 (MET) Received: (from tor@localhost) by pallas.spacetec.no (8.7.1/8.7.1) id SAA01556 for linux-mips@fnet.fr; Tue, 14 May 1996 18:51:51 +0200 Message-Id: <199605141651.SAA01556@pallas.spacetec.no> From: tor@spacetec.no (Tor Arntsen) Date: Tue, 14 May 1996 18:51:51 +0200 In-Reply-To: Wayne Hodgen "Olivetti console - eureka!" (May 14, 19:37) X-Mailer: Mail User's Shell (7.2.5 10/14/92) To: linux-mips@fnet.fr Subject: Re: Olivetti console - eureka! Content-Length: 258 Lines: 7 >Subject: Re: Olivetti console - eureka! Great! Virtual beers are on me (non-virtual if I were there). I'm dying to test it, however I'll just have to wait some more days before I can put my hands on it - still overloaded with job work here.. Cheers, -Tor From hodgen@mailhost.uni-koblenz.de Tue May 14 20:37:51 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id UAA02755 for ; Tue, 14 May 1996 20:37:48 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA06785; Tue, 14 May 1996 20:36:57 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.5/8.6.9) with SMTP id UAA10586 for ; Tue, 14 May 1996 20:36:42 +0200 (MET DST) From: Wayne Hodgen Message-Id: <199605141836.UAA10586@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id UAA16740; Tue, 14 May 1996 20:34:47 +0200 Subject: Olivetti console - eureka! To: linux-mips@fnet.fr Date: Tue, 14 May 1996 20:34:46 +0200 (MET DST) Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 22819 Lines: 376 Hi, well, I did it. I took the tga.c in drivers/char and used it as a template for g364.c. I removed all the CONFIG_MIPS_MAGNUM_4000 hacks from vga.c and console.c. The result is a white on black (linux standard) 128x48 console (still with the same font as before ...) which works 4 times faster than the old AND DOESN'T CRASH! Virtual consoles work now as the new console uses buffers to fake normal text screens. At the moment it will only do white/black. I'll do colours soon along with a cursor soon but I'll give the patch to Ralf and wait for his next working kernel before I do any more work. For those who can't wait, diffs are below (against 1.3.97). I'll leave 6 VC's scrolling all night to see if it blows up. Wayne, feeling pleased with himself :) PS: Thanks to Ralf for his assistance. ==== cut ==== begin 644 linux-1.3.97.diffs.gz M'XL("+?1F#$"`VQI;G5X+3$N,RXY-RYD:69F_N_O[,KP(`QVA42^F!R M6B>&1ZO]>'9V9G9VMF-12M2A^Y'T+'MXIVH[C9VC@QW'M:YW#;?=W>U;`V_W MW+@QJ=4S9T`+OB^KJBI44.G+T"3GQCTA!T0[:.X=-6M[1#LZVB\KBA+VEM*7 M[C!X^(C4ZTU=FSS\^C51]_:K6HTH\-<^>?VZ3#Z\__CSVP\G[RY+S5?DN\KD MURVB.M1Q^X9/S+9#J=JS?+]GLC>5B6EW+%HF9=6B'9.2TXN/;]^_:WT^.[_X MS-+^_*ZFE0:E#LZ;C,Z[(:/*R6R7/XCYQ^^H5TS`%\:-@=TG;Z`VB" MNVMXGMF_@G\19^!;CNT1J`S_=]_ZRV"?[+"G6RU6H7V=>%UGV.N0*Q,*HY9M MPC_OB=\U)R7N$,*@9D_EW6]>T-=TWPQ M'F(^MJWSDP\?+D[+2H#_;/8=WR3O?GA#;H)".ZR$:U;L0FJ0X+E1#0:N`ZP% M,/&&@X'C^I.'/GV^>/O^PUF96)3\09Y]-_OQ,_**/+M_1OX\9N-LEPDA%FL8 M^<0+-($2%O6AP4/[49&MRQ_?O_U"ZDM'O>-:WTP7NKIKN+N'=]K^#H7R9\9C M(>2)D5^(Y1/]TAR0^CZI-9J-6K/!1V_O\=@_\3R,_D^&38A&:C#1:_`?/']0 M"T:_JM>.#HA2J];8X*NUNYI>%?B!2$0B4@9)19!T]L=B)'WJQS225D-_2+^= M+D72Z"V21(KW?!)E%@^9Q!C%AY1DLM#LB#+CGOS7XC)U8=91E)^B2"J'7$$J M"B!I=$DK64]=1-(NZ:H8^A/Y&>_ZKD=8WR591R.5*5G/1?]Z]/8E'812,0)2 M7`,4TM9H54"0S+%NB1A+%BE>3W%-55)GV#S6Q;5NZO$B=3ED6#UU.618VW4Y M)!6>'4)(23VDR/S4162()'+QNOEDF<5&ADT[44V`AEAG2[2+PLMD*J?Y1[,[ M)"V4,,5VN5P2FL7B6I"0[HVV3%)>(QIB(PAI:_K3NN!:D51.*NK1D>*])-GS MR,^HGI,POS>6B;)N^7H4MA:+SW=))!76!"1E")7S9L>"I,+^^=7VSN)KT;IL M[176HR4VZI/S/&7FW>H49ERY2 MET/&,D:;9I6+6]!+9IS0#HXN)Q6%_)]+:BRY?Z0+>\R2C8W1(^@,V=[K M$?)`2OJ3(^\?2>J*DG%!5'B,:(@O*&PG5(_@R8]Y9V3QC(MU;W<#O`>2FK\N M9\?%C*2IOEW(]XM1&1/]G0>3G:GM2\45DQ1)O@Y$DR$_Y M=1/7]PV.3TYWW5S+.908VH[R,^%(T2C(*&]?7YP_RL^BK^^ZW#[LJKN!6?$> M(#_S+S^3/:T3WSF4#""1GP7Q?Q:OGLC/HO.3HOZ9C[58\@P.W1"D>,Q)+&^/ MLA>)\C.J!DCE-$"A,SA"T:="YZ3$3X(([:HC0R(@Q2-S]!`D%2F3"I\16U)F MV-'-K)0IWO;(_2DY1I+CGK%3P\7+E5303#AZ!.F=%TT`68=,WB!+2EQZ;W*9 MXA&8430!U%01B'=AX?@KY;83DDB12Z-1PY`R<>F:0D?/;H":0%A*U-42N]=Q97N)^$XLE MCN]T'MKO&>GY0B&+DK6FH#X!R3N`$LODH*,%C:+0XDN!GWKZ_(RXC$E%E)_YU#_UV)B<*^\!VN\Y MV9=9LRV#]CO*S[5*,!WE)\K/(OKGHXP[ZI^(+/QN-:[:R,^-Y"=&S>7ZI))D MAE@\*X3(C.[;%23N M'VV2_"P>$OF9_?C/Q.[7SH&D14L*^9EG).J?N=,_:4&L>LM;$8A\5ZD0`\A.1Z%_*$C(LC&,NFH*&I%Y8 M=-NO_G3>`UH52.BU_/:E,*2>?R3R4R063H"?^M/\7((,6^0E8XW"IE#DMJ>& M1/\G[F^B_Q/WWW,3Z83[1R@_-PA)0_*M+;F`-*X<;B@_LRH5H]Q.F\6VH_S, M2'ZP]=VO'3D/).8/P?TC1"(_45?,=YXHY"^1_/-A;G)$QHE2=ST$<;/*)%X&+^4:_^G'M57B<@X M>GZ3Y6<2]YG2!)!)W`^;V#VVB=VWB_(SWER(B-PX^9G$'=,QWP\;95][,3)* M]"G&52(2D0E:IA2URMSG=4=M#9$YO1,Y!YD<NVGBF">UWS/]9 M\/6]*+>M8?PG\A/YF3ED473:L`5[_4BA^)`E8Q1VY&PUI/C;)5N$ZWM2>0\0 MN8YS4KB^)R"3PVQ8/.DI@D3684XGM-]19T!^XOX1RD_,4XHY17'_*,,[39@S M/$OR$^_?1&3VXC^SG2D4\W\FRB6,2DHA*@GYB3F=,#X9D<@ZS/^).44Q_R>R M+GOR0G1F#B^8XBK,49 M8#)-0,YG,5-]E)Y'JRE%BC-<0G)R9#LLW/^/*^XOXF[K]+<2F->83YP3`_6*P>R)CS M9F/^9)2?&5W?D9\;S,_$;K/2X_42(S\W6'XFL6I3Y"?R$WU6F']^+?XEBO81 MVN]HOV^"?83\+!(_BVQ*Y/JA< MF9(W?5`\7QPKEY(8HUANC8ERYX*>?>3FG1G!_4W,/X\^`>0G\A/W-Y&?N+[C M^3B,_RRZ_$1^8GPRGB_&_(J9U!G0ED']$V4=<@G7]SS&K4E*VF3WRM>7BT9R MKR>)/.28B3&M_+2)W9:>@7IB_N2"K-IIR*4H#,%(T6+FKXL9F6\[CLJU"%?M M[,N0(LNZ-&YGP/QUF#\$[?=",!G/'Q58J\3\G^B?1WZBSQ_7]P2M3P*TQ-*1%XD@]@TB,=$*K M/*55&T_2H=639:L'^9F^/R0O>^49N/\(3[7CV>HT,S'F8%<(D7GEIYZ4[IT# M^TAR'JT+N3B?@.0:)^[=2FS5%L_Y$-;V*%Q*+&)P?3D?D)\K\7/56*/5HF1I M]'J&B7`!+E&TCW#_'??T<\[Y)9*6QA'_*9X!23*F5+R>DFU'&V&]R/PQ)+79 M@;[TM/;?<:<)XT,*Q$_<:4+_?.VI+*EYD39"OB"]*G"X)_-G`X4B3BQOS>F-\[']FP+ M5D_V^4F3XJ=DF:N^'?DI%U>Y)$/'HOX4NA=8Z,3*\L$/0R[9J4<9@LA\S+BP MV9&!&PIP-/$D2%S["$([]7G9`4<)EG\)AC(9)5B.)!C*9)2T3T;B(>N02Z@) M%&+&(3(I38#*23#)J#F:,#+*"4KTNN?)DZ_+G1H6/_E%$T:*UU.R[*LBP^96!0W%=;G4\N6F2E)FQ7D"OE4Z4H1 M1#2!J*15ZRF92Y8*EZG'X0<3EXIZ^LC$)*TN+&DCK[#BWH,53E#2=6K^XOL( MR=Z%3>-8980\Y&N(%%W!4QI9JUR#+SW?*U>470S)\W$YR(@2)^9=6WX..ENL8D<*9EG/`3YJ`-4'E/#S)6BBQ[)E* MYI%(M\P8M"":@O\3(U[B\!;2!'R5N"NTF?%+42(T)$_XKH",P5>)#,EK_))0 MYNJX(H17U:BC1`PB0[*ZPNHB_(R2)TI?D?,9*!.1*42&(S+[2)P=B$0D(A&) M2$0B$I&(1"0B$5G\K(GBOO0HD7A"D3G9N^EXU3@KH9L45KOQ%CF_#I]J$ON; M-%[.1[E?6RB6.*?1DMF.Q).,XM:?]NQ)9EH60H9=-Y?3&X@0&37.2I*?0K<6 MALDEH1L&HR"%]F$E[S_2D?-KW55/(EHRL;=G\7ZN3;Z_.*X8";H.2PJ1J^=U M1TDKDN,EYKR%63PIG^SI>XP/B2T:+3[_4N2LB>*K87PY&\4]9C2M,4+Y&36K MIYXMOTUBOJ#(?EJ4M-GR1"$R@7P"&Y3Q;'U:913_?`Z0FRR7)+-AQ[*#(WEJ M0Q?VL:21GU82&65]S^P8Y73_/8I])+ZSK,>!%*]G8F^/S$_<54<+);*%$K;3 M))2G5&!W=85Z%MZ"CCD_6&IWP:0KO5%^1LUTMP*3$],_J;#^285UQ3200BU* M0O^4S`9#=I6L?(3_7:Q_E/CXY%J]1S//,I)]M"AE(C)Y[T$:.5J3R"6+_,Q@3E%=V#9<+<-AE%BC53/( M15WKH6?UZK$6VO63]H-NK!\Z]?$U5O M5`^(`C\/R>O796+:T-PR*1.+=DQ*3B\^OGW_KG7^_M-EZ_SDW<=?SEMZK58K MJQ]:%V]^NB3**T)M__!.V]]QRLK#A]>-?9U]\GP1;OHEYG](Y;O*Z#6GOY]^ M./GA['*K>K_%*Z<=Z:QVVM%>]8A7[Z$X_]K@1?4\LTR@('NJNI>_?&S!/R\O M/IR5U:EGOK%GR#?3,UI7/<.^836._Q\T3)D[;L3VG9^ZT MEXSF!"-"G0F8C_W)P(7Q)C6MJ1\UZPT1[CP4,":?IL-SS;U:4Y\BS]X!&Q[^ MDXT.V2:G3L_8QC.8V82-$O.&`%<1* M^6ST*'ECF.TN3#W#[I`3N^.:AD?>##T@AQ+R2'46STOXU;BW3?*CT[DV[1W^ M8E:G7SS358$JEFUVR)79ZQ'/&=J=*K'-6^*9/F\Q])SO.O"5^9^A:;=-CQ6X^UTH=N> MC_FH/KK8SJ3_F)#`WWE`EV,P0`ZV/`!X?I#HT>,#@R0Y_$G M:W=FK5YC?SC-?"C(L:\I(/EU\_M)Z7GS\TG;9U!M?XSM6M==CN6CM50$FVR*V61H>]8U8R2; MN-`GMO_'G\=E=?(Q[^86]%_K#CX&&I)KUQAT6WTV4U^1VG&)S/V!"OGN,)C& M'&NUR4@@>+QKR^IN,!4YK8G!7VVTV63V'?AU])`W?FJ'@]^^_^W\K$F\GL/^ MW++/H"S/-WPHW[)[K&>^.5:GK%Y[;;^R6IK[ M@C+20(-+K(96]>LQJW:)?PJM9MU$%%)IOWQYL'5 MJ%BO:L?$>ODP:/";HFP1]M(`\94AOCX@?H/?)HC2=@4JJ7S=@M=N5W@%X#M6 ML](_[`=\R5>6F6%B7\%_S_GRIK)NA$8\'CJ&8#QF0_8$3X(!([-DNCS[_/[D M0^OL],<+M@S/BP^0M>Z(2[9A] M%*S&HZ]`*K;N^0A04GFH^!9OM.,!@BU<(/L4NBX;*3ZV_W#-A,PU/6#$A*EL1+=)!]Y8F?ULBU2"0I79 M&F_[`7'F2O`D2JCXBK:UH)0V+&4^*^=*]558"K9)9:ZWOO^>U+>":51B\[]4 M`M")L?A/=^[/SL.?T9-?NE8@[[\.@=PP^VIW6JVNWX'8!#D"$L`T;5A@R6WW MGC@#'S2!OYC:,"F`S<_2;9?IVQ5>^?%\X[^HZO%H\G783&13SYN9=N/V+FCD ML5S)HWXR7<,S6UQVO'P)$AU64S+_S>CU9$R/OQF%)L/@=9EZ,D^(T:=+*;&H M%$^JE#$M&.4/]AO5.DSV`TVO@MK'.%\FC^4T:;L5SIPQ[\N\FYZ8[>JKI5.: M?0U3<3QK)[_.3LR[JP7R`%95T+&_0`[4\F`9S;+:'_5;;Z0W[-JS74Y^QCO*@&;O;DH^P MURA3S\!R!_6QG5&51I-5!;7P:GC-U98?[?,4Y;-0G MJV^TN_[]`);D5^3\Y/3'!4JSNO"]S,50TR>K^"RO.:%'7=(R?']^LG%E=*^A M5^M'H(WJ!]!=>S/K$Z]@QVQ;_2W^*VA"L+#[HSZ:6JQ"6_IHVHSD86E*4?-! M-YO3UK:W8+8$(CC"F&*@1T'S^H0:#S3YMP*#*8%-F;!HPCW4_F$B30BS>-;]\[!B!G.+5632B&E"S0[Y3%L6D#ZN M5HQ[Y7%CQMSGS(]43T6DFLKC6BHSE52FZJ@\4/.!F(R6S-H`JGF..\NX@7-C M!@Y)>,/(Z*Z,M+EZK:[QE50[:N@CXPT>>5S35V`#]+SCF>_X\@_?@(KH/6'* MS2G002G,W)\P9/J7WZ<5OX6/CK3`I[\;UR_0.CAD1HT)ZKP]I]D?\U;/6(3C MA_IFOS565JY@>1[WFP[]ML^]SQK3A_EL[QH]&F@+P1^^B!\>!UZ3:]>\#[2? M;SX?AC_&H_2G^OW`\'RS=6M8_EA+!&,?QI-Y#2W;Z$U&M$HZH,CT3,/=XKH: M]V#^:OE=9^@3GUE-4&%XL@K&]S4TF'QKOP@L*3KL]8A#03$'COD6_,+<8]1R MP<#B!3+C2>'*W`(GW@)77REX$\R&1^2?Z[HJ>?/AY.//7("3^O;<:&P_8AMK MVI-:#_1""U[+AN[AQ3T'K#_&PRD?$/^,$6.DZ3]%W&5T3YS4=8`&KU*6?1F9 MUF-/1;VN:J&OUO8F39D33X"'F]6IU3>NZ"^,U?IMA!T5OC83,6#R/[*0P%?ND/RDV$3HK$-FEH-_B-, MT0@TCJJF$:56Y<)G['1]RWR7S.]X#L8'.0]D`'?US3EK&9H_P;SM[^T^R+YW MC7U]RCUKN!V/L-V'^9+8G@7(#O[T10\J[OL6.3^`;YBQP^SMKN$Q`Y`[4C]? MG!/')B#9F+_&)&RAXX_Z#N"^F01&FSBW]L[(YSOEG9YR6@/;_V:JTL@7^8QW M$'34L[+ZS['X8/.MQV4C'0!$ACE`/CU$R\=X_/3L!MK_M_>L76TC67XVOZ(Z M;215?PWC*P*M/?.,DWX@E[5 MNLLBO.'DAKKFACN@E+V6SO3OPF^#?,;%.F:8?5<;8BE'O%GI9N\:9Z M-K<^^O%/^,OJ0-EU>WLP=NP`EJ'QA'P.283G_>'M!%!T"X0=0]^&=OQ(?\=; M0HS'#C5HHV'4-<=SD*!;PBGQQG8OWK(&$C2<`I1_OJYL[S MI.^U9LOHCBU\U*V.WAH_0@?XR?C!PR@ZYOCVQL^IK;5:9GU"T5_F^.FVV^-7 MP<3-YD7^PH%/L#C))SAO%XUZCA M&F\_O/!_!IESS%?W5+:U[E"!K;>$%WI\!1#KK%[=X84;1=?7R*4S2[\W8/96 M7]'@7S<+7%K\I52U'*\U?UVN9211`4(,>[+#D=8OP\?=8LP^S9KN!J M]QW`>,`]N=9UL>*GVD7U0GA^&??]'@K?;Q!FX![V^X?[,&>GM>OJ>Q!/?'YG MAN[":UU#PM1Z#G3[]NKJ5E3@\(?!QF[/+LXNWP?*R\HA;PXX)2H22EE50&4\ MO;I\=:W0/U=?_LGM^>50.E>H/33]4WM]NK]^_,J`@RO MQMNSV]I'[L;_'4I57HHC`[RUV>XNCA"4("!E)Z['OA(8V6GUO/*U=E.Y^'B. MZRUF$^B`K_39Y6WU.L<-G\'"L]^AYX_5&K2#Q7N![LLJ)YV@C4WLW]_`+K;` MN\<6;*VM,Y!6M#L31)$MT*XZKQSX\`AZ%/912%5<*-[*?F"8R`MJ ME=-KF,<)+]X+#'-/#'-W\C"YEE2[>O>.U\8Q#LK>75V?5&L<0I3N!_;G_+?: MNT^7M9LOMR71HO/J9[&VO-]!::7V&;`<.K]P"_=]A:>U MFZ^7)_[27?^(83FJM7?G7VL?*[=N>=E7?O6Q=G%U6LVY1X?Z@\.;D^M+?I[([**M>5M["_GR^?>\5EG.^_8,N_2=@?^0$?*R M#YW`U^7#P(Z=?/CM!C1B?Y4#(`+^*@=E/+2]#A=G`G5YAU M&=&A0"FO]6;S<$RCE0"=*2N!.N7A.A=7ES6/E&&-/4')L,J885]7;ZJ^YO<. M!V3O"UK'!G2O+[+>]8R6LP62%5<(VEI'N]>1$7*2[V@=Q^[+CY=7M]4C3K># ME4"J1)&_Y"D?KF`&;``T0&14O#9`Z2"+@035=QFS7]F!\N%?#+@!]R"K!X?\ M$WY0#S_\:WN(-_B4^)S8^]%8G5'@KRP']&=\M$[1Q[PX/T.1N>93:WS\E4O3 M7*TI_N#W*^"%Y/<'2)1N):"BI?G^)],&A_G>A!_<9N_S[`\ANHACF-1& MJ/]YV!5T/5Q7+^#T`8]\^^D]EZL&[!N^U+BF)/#9%=O0Y]CM.:?HV#WAW@<> MAY7G;?-HK:)7?,-E>P%0:*#\F-\HHN7;#:40W]A/;]BK_U5>N4;Q0-MN+6[X MYK\*(S@:Q-%&5]C:\C71>94?:8!_A2H_^L9%=Y3N`4,;J&M@XU*KF(;K5X6S M5S,ZAF-H+0,HXAOE.'"NT=:X4433)$KE''+(./GHLX;Z9=N"W[0EUL2S]0CVN/![Y.:$]`"?E6[_0IB&K#W-_F6*+"T4/4%MOB"@^0LW>D!!@=&]6.`MK6:+W9KR+QM-IM0Z&V5 MNP\%+.RU&JQC.NP.%'70/_5&";B8@]XVVT3#/18^H(K++>UB`W[X^.0'Y"_< M6L(1CC4MLPT3UI^@/F&_9]/_C\>C$NTB=A<\S$A7\.Y;V9+<_`$!UO'N,4Q86=^&2NX M55CA/]Q?HC3Y#P_V<8#A_3G^^_[H=^&2J'>?-X<[]DM%)?;2-Y<2EPHX51D] MY>ZA\V_1C>[X=P@4Y,#R^WQPP8`,L0LNK?IIM"M.KD866I`@7L?3CEU''H<7 M2"Q@INPOK)-_>S>*PW@GA*Y[&#L<9EC"S3P7L^S`!TMG(%BYXMV=7M=Z;OP\ M+L?G]Q6&YDB,6P?U)E`3C<&LV!;ZTTJ.?]C-83![TT2M'R/F(UQ3$8I/7CAN%K!_W(P'\"#0W_[=+8 MJ20V1+"K[S#;XU>X)+A4D+8-@@6&R/G6,+_XQ:TY;LAY_PZX8#Q*C+M%@65K MG7L=0S<##&GR)F\.]5WT1BVUU4QJKT>W.-CI6-#A3<2EFHX_@?&/19'^]L:) M">ZQ?X=.8/0;`PD7-O<`'7,-U'YJ+\+EV5_',RG_A-*\7Z<5A$N$T0N,,!@& MQ_(0^A+&Q;_9=27!OX27%6@FMZ#_8?S)Y4"OMTTO%,[2&W_\]6<_QE9$#/5+ M[ZV.6WHX6GC7ZD$AMTJB4#E"'UU:!RHF_!]20J&+(K^'5>_?B2IYM'&8+HKZ MNCUPB+C.#L;E'U#.;9W'A`MY#X0_E",:9D??9E]`[?SM\NI+P1V$YJ`G#4$: MI@OOND%0Q-GVM.D=5X`5^XFTM=[6NL/Q04#:09[`[<%#Q/PG'[?;AV>X2S`; MH_E<`W:@;>))_@?[7+T^>_<5-K1RRHZ\O[Y12W M>A`3_YJ`B?X?=RBP.MN-K;_#7RYQ_>.Q6/R3!R'ZZ==Q/(T)%@[8Y;.&%\#9!?C([&[R=[%V\'@X$U&%&&W.7)L\V@VW4J50*! ME0^NX/?V!H+J6'YF`SCE0%!>GN7=F:$UR$,$$7*N<`N"X!MU#-95ON\J?;+N M8T)`/(8N>Y;XO(MC3%/],1;$RK@N*?CA?059NXP1Y MQ@"]ML,,-HC+/Z(=2G;"H_D\&;B$D31/NO73A,-:$\%_\9S9$@9F8S7C;L+Q MQ3K-^[9F/X+4SO\;]IQ#3S7\/N:\EU"UA47F14,B=QWTEIFH*(;.;R!YY7"< M07/Q006.J?[=P=)- M6YF+&2QPL1N*5Y_PAB-R0QC!J9UH)2RU$PO>I^?83%Z3B"Z] M:N%";9M`"6<$P0*(=+@MP,X5<,OK!T)N]]6C_3$AM\K//$:^B(2V)/$/01(D M08:!5&4@U>`_XR'52?_X(=72S']"]ZY.A52CSR@DI/S*)]'FZD$FL4?Q08;$ M9*G3$>7$3?QM?)N*--:I1#]E(=5PD'-010E(-3JE#3E.18;23EFJ&-:3\#-> M_JY$X.\AL4Z-U&;(<8[[;:3W*0M$5#$"I+P$*"6MJ24)0C*$=5/(6+*0\N.4 MEU1#R@SKAW5Q\4TE7D@E'.2L<2KA(&?-70D'J4J?#BG(D'+(*N.G(D-#0D*. MYYL3VUQMR%G'3E824&=H9U.DBY6GR6HXR3^:WA%20YDEV$ZG2U*G6%X*DI*] M29=)RFJDSM`1I*0U9;(LN%!(-1Q55*)#RJ]2R)4G_(QJ.9EE]Z8VB=9-YT>S M>+'\>0\)J4I+`B%IB!K.FAT+I"IMGY_/=Q;?C!:E:\_!CZ;HJ!//^RP=(3%( MJ7&&Y,5)S%U*9DABE:3FKBY+DU*3AXSLK9[CQ"T7,J2G7@D'&K9M6+J]! M3SEQ4AX<)1Q5E+)_3AEQ2/^1(FTQ2S8V1HD@,Z3;UR-E@0QI3X[L/PHI*X:, M"U*E]TB=80N:Y0E5(ECR8_:,C#]QL?IVU\!Z$%+R5\+I<3%#JDOM7*1H&,TOOBXOR)?JXZ?U?"^6'G]0:FQ7I`^)E]^IGL;9WX[J&D M`)+P3/;/#BD'=PU#6!E(\YB:7W*+Y(HI]1)4`UG`0H M=0='*OI4ZIZ4_$T0*:\Z84@$2/G('&4&I"K3IBI]1VQ*F[.N;J:E3?FY1U[/ MD'L4YS;E(S"C2`(DJ1(D M068]ZY=*D&E>3[K_O@1(HB$$&2KR5HUN\P_99@IZ)PQ)M:8?4G]7I//"S8*, MG-1"[YUE)>XWL5CB^&[GD?Z> MDI5?*"')OA0A6Z:4#SHQ2/D(]N6.,^:W(8@? M4:[.U8@^)?H93D-9QIMT\5G(H^2)(HUO"?BI+!\_(^=T6L@X0ZX2\?>Y[M3' M3.ND[D&'C#J>`T/B>NEC&:>#[(I+Y7$IHXI$/[,I?RJQ87*FK`>DOV?$+[-@ M78;T=Z*?"Z5@"M%/HI^K:)^/LN\D?Q+DRGNKB6L3?JXE?E+47*9O*H7,$$MW MA0@RI9X[\AN2)I4.NI1&3SWAT@K'AV3`4DJ:5.8P69[L2V1?RH:O MG&+=EY61C_"3\'-)^$GR9Y:RUL2B'ZW4C0#"3X(D^U*:(&>%<0Q%4Z@S4B^, M>^U7F9SW0"U))/2:_OK2+$@E^Y"$GS*Q.EC%GY&B<2C^*5,VS^5J+9* M@HQCY=>9?B;QGJF:`&02[\,F]HYM8N_M$OV,-QF8WX>-XM<> M#QDE^I3B*@F2(!/43%62*C.?UYVD-8+,Z)O(&:7QAU+$WQ.@R;-T6+KI*0-)6$WR3(],5_ICM3*.7_3!27*"II"5%)A)^4TXGBDPF2L([R?U). M4N(OU/\/-D_*5,HY?^<$W(9[\O$C"'9BQA4"#\I`I/N M=ZP"+TX!)JL)T/DT9JJ/LO*D]:3D]N@L:\P42"D-124/3G;ES_@@Y5_FE8(, MF9=#RJ<_14^C7)WK$TN\+K%&"XA/3HR&I!L_X\O[2OY-\K^'PJ5EG"/*#T;Y MP6*U0,:<-YOR)Q/]3"E_)_Q<8_Q,[#4K)5XK,>'G&M//)+BV2OA)^$DV*\H_ MOQ#[DDKZ$>GOI+^O@WY$^+E*^+EZ^A'A)]'/N.AG$I26XNO6.+YN:?(GX2?A MYVI(`H2?JX2?I,NL7Z$WX2?J8T*FFY62CGN^VX*I&BI+^G\'[Q](Z,D'^3\L^338#P MD_"3_)N$G\3?Z7X@K&2?F35X1K+X,N1<$0BA1=S?QU M,4-F6X]3P\V(N';Z:<@JT[IEO,Y`^>LH?PCI[RN!R73_:(6E2LK_2?9YPD^R M^1-_3U`KEW\C()93K!+]7/'[<2'IDB)]KV>.5XFE-!156NM1PUEC9AF9)%Z- M46?,2!Y222$D13J15KXDKDTWZ4CK2;/60_BY?'M(5GSE*7C_B&ZUT]WJ969B MS(!7B""SBI]*4K)W!O2CD.=H49#C\PF$Y''RUJW$N+9\SH=9G4%HUCOA/^0Q((6-* MY<<9/^R6M)_WXJ2:%GR';G+=WPL]P<953,G2,6T^I=X&E M;JQ,W_Q9D%,\]41#"#(;)V[6Z4C!"P6TFW03)"X_@I2G/BL><*)@V:=@1).) M@F6(@A%-)DH[,1*/L(YPB22!E3AQ!)F4)*"&HV`AH^;4A"&CW*`DJWN6+/E* MN%O#\C>_U(0AY<<9?$AND?&8V)7G(!=_G"ADEJV8$,N:\1E+G75T^I!JN37FLBP\R ML2AN55J>7UJVS%11VK1`SI%/59TK@DA-("IIWG&&S"6K2K>IQ&$'DZ>*RO(A M$Z.TBC2EC2,J_-+\/.EJN M8X*4SK2<`?Q4$]`FU'`6GF0UE%A\IB'S2"RWS1BD('4)]D^*>(G#6J@F8*LD MK]!ZQB]%B=`(><-W#L@8;)6$(5F-7Y+*7!U7A/"\$G64B$'"D+1R6$4&/Z/D MB5+FQ/D4M$F02X@,)\CT0]+I($B")$B")$B")$B")$B")$B"7/VLB?*V]"B1 M>%*1.>E[Z7C>."NIEQ3F>_&6<'X1-M4D_)MJO#@?Y7UMJ5CBC$9+ICL2+V04 MMS+9LABKILMLD9@N*;*<'VQI M;\$LEWH3_8R:Z6X.3$Y,_E2EY4]56E9>K",'*U)Y)(E_$QA3E%%6C><+\-AE%BC>3/(1*)O:25_IZ)\A%W`66RG5,-Q<(,G[\C!S_.6_L*^$G01+D(F7OR&=3 M)>DW0W&J:O08L_@@E0B0=/\H*4D@I`8M=3@>'(&.%W&@8S2;;ZEF7K&5T>M^WRMM[VS^K MVZ9EW.\T+..;;MD[]0?-VK'UEEYW#+.S_1``G0BUL;6U%:+1W!>]P2ZT9\8. M6;E\I.P?[:JL_///!QO%8E&NQ]QM3^=-E/>AYM'K_:/7NZ*)7W]E6V6E7%)9 M$?_S6F&__KK!_KW!;JO,H*.QM%*#>:K*$WC0Z,>-.%NCC[>%.[J+R__'11VU<4)8\MW#[H M[/W>P3ZK:U;#/F*VUM:99C/MSORF;_/6-HKZ=T>'$7PSC0:[!^#:7CH9NUMMZN06'[&(>W M4]@HLH([#;-CFRV=V75+USL,X$SKF6GUNF[;"#6`-&Q6Z)A.@6D,=TFK0WL[ MFN-8QEW/T5E=;[5@";\9=?V8-,[;-%VOT-8L=Z"65I.3VMQ$!AY MB>D&-&_QQLPF[^KSR2N;W6GU1P-6Q79,2R\QT^)%+^P'K6$^N:-YX:W"$[2@ M^X;)6X,%<_2.`SL.98]Z%S8"MM_QT*)I(4+<]9I-W1HL&4=K7"N^3+Z)?M'% M\C@Z_-.$X6CLPH`U&S\MZ$8#,`V:8$\P04,,:75;95K+ M-H\0WM)AB'!Q& MVX$!U&':0&YT<;)@4C4^RZ>A(\6^:2W?L?*C'8-0XD'?@TVBS\Y'Z'D^S#4E]B&8Z3!;U]^ M@4-@PPCY7',Y).]36_1(Z)2&!_5_;!1_()D,HLX06B`2P?HW1G!H'+X$.1*V M/LR1?+P&Y\0V$,<0&SJFU88U1.SW2+5+G*79_#?=UF`9M,[C=GT*U_6#R3!Z M/WSN]J$'I]!BY4.FO$9.OZ_(_O(:?G_R*?WV!_ M,SKU5J^ALU]X\SNP.DWC?OOA[U"X];X>>6D>EW[>'92J1V4\^P__QDN M'^'VL$DS6O$/0K/;.W>FZ1B=ILD'X>X?F;BN52NXF`0]M8WC]2]J2V453VB6I0LWRT6_;MWP'?OP-W_X9W MK]WFBS;\V3!QM59D4W$9U->E\BXKPG\.Q#J(H;"&5@=R][]<-\L`") M*"_X.3]A'4;GN>4*@P8(8O>6UGVHMN:R=7M4.]G][FP/J^-OYI]/W579K,I0;D.=YDA`@\`-*E2T0 M>9!@NF/#Q8'FV(1UR6UL0;/P#]]$$(RFV&?0`&U$$RT^UM=B8DSY:I(1=_ MT%!&<$"U06;1JS^`(.3"BI8TV^ZU=2'D@;35L]W?`X/`9GA-$]!Q*`GN-?<[/<8UZ-AZ535OO MH/9GVWFQ&EVN!FYOCS8"A\)MY/7QH$BL`OM\=EJ]JMU^_5@52X%(Z(/R)!6` MY.OW#81V_F$(ALM`P^*-#X:?FCX`(``N;*_=L?%T7%W>WIS]3[7V.Z]0/:PG&]@'*$X'B):'!J=EXY(#/\LX=*@0:_W:/DOB4D;$YD8*4% MQ8&E$;5P>4#3S;GGEY^&W(;`D:MK.`IB22ZN3JN()6H>C]<9:A2`]("'9L>L M/U@FH)\[T'_P8Y5#A7ITU6[.KVYO/@+=9$78@CO`,.68G^^]O=U2&<[WGK+K M4OI<;D1"RB/[,4"M$.T*";:&I`3$,)83I8%U!U$,*KX9%MV^U;$R*%M_U'N6 MA:3G3VP`93-8@$`+PW^+%HNC0\#=PD4;(#H>GAR(]E#?'5J1/1?"6BYV0J;GY_9=?RBB!AH`%*15()^HS8MT/%;'N^ZJW[I/G+@P!8Q>: E[V\1B$W3UIW@6O9'E=DERP$%!LU/,&B@'>XJ_C\A2%E$X:$%`)/G ` end ==== cut ==== -- Wayne Hodgen | hodgen@informatik.uni-koblenz.de | #include From Stoned.Elipot@univ-evry.fr Wed May 15 10:41:02 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id KAA20824 for ; Wed, 15 May 1996 10:41:01 +0200 (MET DST) Received: from amiante.univ-evry.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA16270; Wed, 15 May 1996 10:40:59 +0200 (MET) Received: from bandsept.univ-evry.fr (bandsept.univ-evry.fr [192.93.214.19]) by amiante.univ-evry.fr with ESMTP (8.6.12/96.04.09/louis-blanc); id KAA15461; Wed, 15 May 1996 10:40:59 +0200 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 KAA01698; Wed, 15 May 1996 10:40:55 +0200 Message-Id: <199605150840.KAA01698@bandsept.univ-evry.fr> To: linux-mips@fnet.fr Subject: mm problem with Olivetti [Was: Olivetti console - eureka! ] In-Reply-To: Your message of "Tue, 14 May 1996 20:34:46 +0200." <199605141836.UAA10586@informatik.uni-koblenz.de> X-It: bandsept (wrong OS, but right CPU) X-Mailer: MH 6.8.3, MH-E 5.0.2 Date: Wed, 15 May 1996 10:40:47 +0200 From: Stoned Elipot Content-Length: 1097 Lines: 23 Hi all, >>>>> "Wayne" == Wayne Hodgen writes: [SNIP] Wayne> result is a white on black (linux standard) 128x48 console Wayne> (still with the same font as before ...) which works 4 times Wayne> faster than the old AND DOESN'T CRASH! Virtual consoles work [SNIP] Great, good work Wayne. From my side I still have problems with the Olivetti made a plain separeted machine from the magnum (but sharing a lot of code both via mips_machtype= MACH_JAZZ and in few places with mips_machtype == MACH_OLIVETTI_M700 || mips_machtype == MACH_MIPS_MAGNUM_4000 when these last two boxes differ from he PICA): I'd broken the mm part of the kernel but I can't find any Magnum specific part of this that I should share with the "new machine" (aka Olivetti). Ralf could you tell me more please ? Is there any place where I should have a triple-look for machine dependent (and not only CPU dependent) code regarding the mm ? Once I'll work out this minor problem, I'll be ready to release Milo 0.27 and kernel patches with new machgroup and machtype handling... Cheers, Stoned. From linux@mailhost.uni-koblenz.de Wed May 15 12:17:22 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id MAA25434 for ; Wed, 15 May 1996 12:17:21 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA18337; Wed, 15 May 1996 12:16:53 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.5/8.6.9) with SMTP id MAA21232 for ; Wed, 15 May 1996 12:16:19 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605151016.MAA21232@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id MAA17619; Wed, 15 May 1996 12:14:24 +0200 Subject: Re: mm problem with Olivetti [Was: Olivetti console - eureka! ] To: linux-mips@fnet.fr Date: Wed, 15 May 1996 12:14:24 +0200 (MET DST) In-Reply-To: <199605150840.KAA01698@bandsept.univ-evry.fr> from "Stoned Elipot" at May 15, 96 10:40:47 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 724 Lines: 16 Hi, > Great, good work Wayne. From my side I still have problems with the > Olivetti made a plain separeted machine from the magnum (but sharing a > lot of code both via mips_machtype= MACH_JAZZ and in few places with > mips_machtype == MACH_OLIVETTI_M700 || mips_machtype == > MACH_MIPS_MAGNUM_4000 when these last two boxes differ from he PICA): > I'd broken the mm part of the kernel but I can't find any Magnum > specific part of this that I should share with the "new machine" (aka > Olivetti). Ralf could you tell me more please ? Is there any place > where I should have a triple-look for machine dependent (and not only > CPU dependent) code regarding the mm ? Could you please go a bit more into detail? Ralf From hannon@Crissy.Stanford.EDU Thu May 16 21:54:29 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id VAA20005 for ; Thu, 16 May 1996 21:54:27 +0200 (MET DST) Received: from crissy.Stanford.EDU by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA25827; Thu, 16 May 1996 21:54:18 +0200 (MET) Received: by Crissy.Stanford.EDU (8.6.8/25-LAGUNITA-eef) id NAA19590; Thu, 16 May 1996 13:01:14 -0700 From: Steph Hannon Message-Id: <199605162001.NAA19590@Crissy.Stanford.EDU> Subject: Libc questions for development with SGI compiler To: linux-mips@fnet.fr Date: Thu, 16 May 96 13:01:13 PDT In-Reply-To: <199605041328.PAA16554@informatik.uni-koblenz.de>; from "Systemkennung Linux" at May 4, 96 3:26 pm X-Mailer: ELM [version 2.3 PL11] Content-Length: 423 Lines: 9 I am trying to use the newly developed compiler for SGI. I am a little confused about which libraries I should be using. I initially thought that libc-4.6.27.tar.gz was the correct set but it doesn't contain stdlib.h, file.h and many others. I know I am missing something basic, but I am new to this game :) Can someone tell me what libraries I should be using and where I can get them. Thank you in advance, Steph From linux@mailhost.uni-koblenz.de Thu May 16 22:17:59 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id WAA20601 for ; Thu, 16 May 1996 22:17:58 +0200 (MET DST) Received: from informatik.uni-koblenz.de ([141.26.4.1]) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA26067; Thu, 16 May 1996 22:17:47 +0200 (MET) Received: from brian.uni-koblenz.de (linux@brian.uni-koblenz.de [141.26.4.22]) by informatik.uni-koblenz.de (8.7.5/8.6.9) with SMTP id WAA03928 for ; Thu, 16 May 1996 22:17:09 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605162017.WAA03928@informatik.uni-koblenz.de> Received: by brian.uni-koblenz.de (4.1/KO-2.0) id AA15542; Thu, 16 May 96 22:16:06 +0200 Subject: Re: Libc questions for development with SGI compiler To: linux-mips@fnet.fr Date: Thu, 16 May 1996 22:16:06 +0200 (MET DST) In-Reply-To: <199605162001.NAA19590@Crissy.Stanford.EDU> from "Steph Hannon" at May 16, 96 01:01:13 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1137 Lines: 24 Hi, > I am trying to use the newly developed compiler for SGI. I am a little confused about which libraries I should be using. I initially thought that > > libc-4.6.27.tar.gz Most of Linux itself and related programs has been developed using GCC. If you don't intend to rewrite _large_ parts of the code use GCC 2.7.2 plus my patches from ftp.fnet.fr. SGI's C compiler may be superior in many aspects but it donesn't have the goodies that GCC has ... The same applies for as/ld/objdump/ranlib etc. Use the versions built with from bintuils-2.6.tar.gz plus my patches from ftp.fnet.fr or get fun ... > was the correct set but it doesn't contain stdlib.h, file.h and many others. I know I am missing something basic, but I am new to this game :) Can someone tell me what libraries I should be using and where I can get them. stdlib.h is part of the include file archive which is a separate archive named include-4.6.27.tar.gz. Some of the include files are generated during the installation of GCC in /lib/gcc-lib///include/. Ralf PS: It is recommended to press the return key every 80 chars ... From paul@suite.sw.oz.au Sat May 18 06:16:37 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id GAA06930 for ; Sat, 18 May 1996 06:16:35 +0200 (MET DST) Received: from staff.cs.su.OZ.AU by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA12695; Sat, 18 May 1996 06:16:18 +0200 (MET) Received: from suite.sw.oz.au by swallow.sw.oz.au with SMTP id EAA23692; Sat, 18 May 1996 04:16:17 GMT (8.6.10/Unixware) (from paul@suite.sw.oz.au for ) Received: by suite.sw.oz.au id AA29005; Sat, 18 May 1996 14:15:58 +1000 (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: <199605180415.AA29005@suite.sw.oz.au> Subject: Re: Olivetti console - eureka! To: linux-mips@fnet.fr Date: Sat, 18 May 1996 14:15:58 +1000 (EST) In-Reply-To: <199605141836.UAA10586@informatik.uni-koblenz.de> from "Wayne Hodgen" at May 14, 96 08:34:46 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: 1498 Lines: 35 Hi folks, Wayne wrote: > well, I did it. I took the tga.c in drivers/char and used it as a template > for g364.c. I removed all the CONFIG_MIPS_MAGNUM_4000 hacks from vga.c > and console.c. YAY! The hacks to vga.c were horrible... and now I can use g364.c as a template for the DECStation console! I love it when you guys reduce the amount of work to be done.. :-) > The result is a white on black (linux standard) 128x48 > console (still with the same font as before ...) which works 4 times > faster than the old AND DOESN'T CRASH! Virtual consoles work now as the > new console uses buffers to fake normal text screens. At the moment it > will only do white/black. I'll do colours soon along with a cursor soon > but I'll give the patch to Ralf and wait for his next working kernel before > I do any more work. For those who can't wait, diffs are below (against > 1.3.97). I'll leave 6 VC's scrolling all night to see if it blows up. Famous last words... :-) How did it go? > Wayne, feeling pleased with himself :) And so you should! 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 hodgen@mailhost.uni-koblenz.de Mon May 20 10:19:28 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id KAA11228 for ; Mon, 20 May 1996 10:19:27 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA17746; Mon, 20 May 1996 10:18:46 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.5/8.6.9) with SMTP id KAA19360 for ; Mon, 20 May 1996 10:18:11 +0200 (MET DST) From: Wayne Hodgen Message-Id: <199605200818.KAA19360@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id JAA23645; Mon, 20 May 1996 09:20:33 +0200 Subject: Re: Olivetti console - eureka! To: linux-mips@fnet.fr Date: Mon, 20 May 1996 09:20:33 +0200 (MET DST) In-Reply-To: <199605180415.AA29005@suite.sw.oz.au> from "Paul Antoine" at May 18, 96 02:15:58 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1365 Lines: 30 Hi, > YAY! The hacks to vga.c were horrible... and now I can use g364.c as > a template for the DECStation console! I love it when you guys > reduce the amount of work to be done.. :-) *snicker* Yeah, I was hoping someone else would do it too ;) > > 1.3.97). I'll leave 6 VC's scrolling all night to see if it blows up. > > Famous last words... :-) How did it go? Fine, the console seems solid now. I left it going for over 17 hours doing ls -lR in 6 VC's (load 6+ poor thing) and to Ralfs bitter disappointment nothing happened ;) I haven't had a kernel panic since then... (touch wood). I have had some wierd problems with "ri exception in delay slot" which seems to come and go though. I spoke to Ralf yesterday and he says he's fixed some nasty bugs over the weekend and will clean up his 98 tree for release RSN (yeah - we've heard that before right :) On another note: I fished out a set of X11R5 sources (they still have the mips hardware support) and I'll have a look see. Maybe I can get X11 for the g364 pretty soon ... Ciao -- Wayne Hodgen | hodgen@informatik.uni-koblenz.de | #include From Stoned.Elipot@univ-evry.fr Mon May 20 10:37:07 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id KAA11564 for ; Mon, 20 May 1996 10:37:07 +0200 (MET DST) Received: from amiante.univ-evry.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA18056; Mon, 20 May 1996 10:36:50 +0200 (MET) Received: from bandsept.univ-evry.fr (bandsept.univ-evry.fr [192.93.214.19]) by amiante.univ-evry.fr with ESMTP (8.6.12/96.04.09/louis-blanc); id KAA27923; Mon, 20 May 1996 10:37:05 +0200 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 KAA10175; Mon, 20 May 1996 10:37:02 +0200 Message-Id: <199605200837.KAA10175@bandsept.univ-evry.fr> To: linux-mips@fnet.fr Subject: Re: Olivetti console - eureka! In-Reply-To: Your message of "Mon, 20 May 1996 09:20:33 +0200." <199605200818.KAA19360@informatik.uni-koblenz.de> X-It: bandsept (wrong OS, but right CPU) X-Mailer: MH 6.8.3, MH-E 5.0.2 Date: Mon, 20 May 1996 10:36:55 +0200 From: Stoned Elipot Content-Length: 1005 Lines: 22 Hi all, >>>>> "Wayne" == Wayne Hodgen writes: [SNIP] Wayne> problems with "ri exception in delay slot" which seems to come Wayne> and go though. I spoke to Ralf yesterday and he says he's fixed Wayne> some nasty bugs over the weekend and will clean up his 98 tree Wayne> for release RSN (yeah - we've heard that before right :) Hum Hum,... I really need to send as quickly as possible my kernel patches to Ralf with all this mahtype stuff but first I have to polish milo 0.27 'cause Milo 0.26 will be useless with these kernel patches. And well we (Luc and I) booted my Oily with an NFS rootfs, use gcc natively, telnet,... Sorry Wayne we hadn't yet tested your g364.c, but cat /proc/cpuinfo tell us it's an Olivetti M700 that was one thing we wanted - I think ;) - and /proc/cmdline can hold whatever you want (if passed to Milo of course) up to CL_SIZE which is for now 80. Ok now let's try to finish milo 0.27 this night after a long day of work :( Cheers, Stoned. From linux@mailhost.uni-koblenz.de Mon May 20 13:07:52 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id NAA14778 for ; Mon, 20 May 1996 13:07:51 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA20988; Mon, 20 May 1996 13:05:07 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.5/8.6.9) with SMTP id NAA26535 for ; Mon, 20 May 1996 13:00:02 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605201100.NAA26535@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id MAA24132; Mon, 20 May 1996 12:59:13 +0200 Subject: Re: Olivetti console - eureka! To: linux-mips@fnet.fr Date: Mon, 20 May 1996 12:59:13 +0200 (MET DST) In-Reply-To: <199605180415.AA29005@suite.sw.oz.au> from "Paul Antoine" at May 18, 96 02:15:58 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 457 Lines: 12 Hi folks, > YAY! The hacks to vga.c were horrible... and now I can use g364.c as > a template for the DECStation console! I love it when you guys > reduce the amount of work to be done.. :-) Ha! Don't be happy to early - last night I got 1.3.98 on the SNI machine not just booting but also useable (undocumented R4600 "features"); so I made the 1.3.98 snapshot which turns large parts of arch/mips/ upside down. Have fun with your patches ... Ralf From linux@mailhost.uni-koblenz.de Mon May 20 13:06:11 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id NAA14737 for ; Mon, 20 May 1996 13:06:10 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA20957; Mon, 20 May 1996 13:03:08 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.5/8.6.9) with SMTP id NAA26625 for ; Mon, 20 May 1996 13:01:26 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605201101.NAA26625@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id NAA24142; Mon, 20 May 1996 13:00:38 +0200 Subject: Re: Olivetti console - eureka! To: linux-mips@fnet.fr Date: Mon, 20 May 1996 13:00:37 +0200 (MET DST) In-Reply-To: <199605200818.KAA19360@informatik.uni-koblenz.de> from "Wayne Hodgen" at May 20, 96 09:20:33 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 217 Lines: 9 Hi, > On another note: I fished out a set of X11R5 sources (they still have the > mips hardware support) and I'll have a look see. Maybe I can get X11 for > the g364 pretty soon ... Moriturus me salutat :-) Ralf From linux@mailhost.uni-koblenz.de Mon May 20 13:19:27 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id NAA15034 for ; Mon, 20 May 1996 13:19:26 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA21084; Mon, 20 May 1996 13:09:22 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.5/8.6.9) with SMTP id NAA26848 for ; Mon, 20 May 1996 13:07:47 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605201107.NAA26848@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id NAA24159; Mon, 20 May 1996 13:06:59 +0200 Subject: Re: Olivetti console - eureka! To: linux-mips@fnet.fr Date: Mon, 20 May 1996 13:06:58 +0200 (MET DST) In-Reply-To: <199605200837.KAA10175@bandsept.univ-evry.fr> from "Stoned Elipot" at May 20, 96 10:36:55 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1034 Lines: 21 Hi, > Hum Hum,... I really need to send as quickly as possible my kernel > patches to Ralf with all this mahtype stuff but first I have to polish > milo 0.27 'cause Milo 0.26 will be useless with these kernel > patches. And well we (Luc and I) booted my Oily with an NFS rootfs, > use gcc natively, telnet,... Sorry Wayne we hadn't yet tested your > g364.c, but cat /proc/cpuinfo tell us it's an Olivetti M700 that was > one thing we wanted - I think ;) - and /proc/cmdline can hold > whatever you want (if passed to Milo of course) up to CL_SIZE which is > for now 80. I'm thinking about if we should add something like a bootrc for milo with all relevant information for bootstrapping. Linux kills the ARC BIOS completly (gaining between several hundere kb to some megs RAM dependand from the ARC BIOS implementation ...) and we can therefore not write to the ARC BIOS variables directly. This means we have to change them directly from the BIOS which is very unhandy. (But this comfort is worth that much memory ...) Ralf From Stoned.Elipot@univ-evry.fr Mon May 20 16:47:51 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id QAA19108 for ; Mon, 20 May 1996 16:47:50 +0200 (MET DST) Received: from amiante.univ-evry.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA25861; Mon, 20 May 1996 16:47:32 +0200 (MET) Received: from bandsept.univ-evry.fr (bandsept.univ-evry.fr [192.93.214.19]) by amiante.univ-evry.fr with ESMTP (8.6.12/96.04.09/louis-blanc); id QAA29404; Mon, 20 May 1996 16:47:45 +0200 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 QAA10670; Mon, 20 May 1996 16:47:39 +0200 Message-Id: <199605201447.QAA10670@bandsept.univ-evry.fr> To: linux-mips@fnet.fr Subject: Re: Olivetti console - eureka! In-Reply-To: Your message of "Mon, 20 May 1996 13:06:58 +0200." <199605201107.NAA26848@informatik.uni-koblenz.de> X-It: bandsept (wrong OS, but right CPU) X-Mailer: MH 6.8.3, MH-E 5.0.2 Date: Mon, 20 May 1996 16:47:30 +0200 From: Stoned Elipot Content-Length: 1112 Lines: 25 Hi all, >>>>> "Ralf" == Systemkennung Linux writes: [SNIP] Ralf> I'm thinking about if we should add something like a bootrc for Ralf> milo with all relevant information for bootstrapping. Once ago Luc and I already thought of such a Milo helper file with among other the kind of information which is actually harcoded in identify.c. So the idea would be: Milo try very 'lightly' to detect the box's model and then read revelant info from a file. Ralf> Linux Ralf> kills the ARC BIOS completly (gaining between several hundere kb Ralf> to some megs RAM dependand from the ARC BIOS implementation ...) Ralf> and we can therefore not write to the ARC BIOS variables Ralf> directly. This means we have to change them directly from the Ralf> BIOS which is very unhandy. (But this comfort is worth that Ralf> much memory ...) There I don't get you, what do you mean by 'change them directly from the BIOS', the kernel never touch anything belonging to the BIOS, isn't it ? It just gather infos Milo passed to it, so why do you want to write something in the BIOS var ? Stoned. From linux@mailhost.uni-koblenz.de Mon May 20 18:19:34 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id SAA21031 for ; Mon, 20 May 1996 18:19:33 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA27928; Mon, 20 May 1996 18:18:53 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.5/8.6.9) with SMTP id SAA14546 for ; Mon, 20 May 1996 18:18:36 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605201618.SAA14546@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id SAA29418; Mon, 20 May 1996 18:17:48 +0200 Subject: Re: Olivetti console - eureka! To: linux-mips@fnet.fr Date: Mon, 20 May 1996 18:17:48 +0200 (MET DST) In-Reply-To: <199605201447.QAA10670@bandsept.univ-evry.fr> from "Stoned Elipot" at May 20, 96 04:47:30 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1220 Lines: 27 Hi, > Ralf> Linux > Ralf> kills the ARC BIOS completly (gaining between several hundere kb > Ralf> to some megs RAM dependand from the ARC BIOS implementation ...) > Ralf> and we can therefore not write to the ARC BIOS variables > Ralf> directly. This means we have to change them directly from the > Ralf> BIOS which is very unhandy. (But this comfort is worth that > Ralf> much memory ...) > > There I don't get you, what do you mean by 'change them directly from > the BIOS', the kernel never touch anything belonging to the BIOS, > isn't it ? It just gather infos Milo passed to it, so why do you want > to write something in the BIOS var ? I'm not talking about UNIX enviroment variables, I'm talking about ARC environment variables via which Milo get's his parameters passed when it isn't run interactivly. By killing the BIOS we loose the capability to manipulate these variables and therefore need something else to replace this capabilities and this is the main reason why I suggested a "milo.rc". Another thing we might try is to save the ARC BIOS memory stuff on disk during the startup, a bit a strange procedure for a UNIX system, isn't it? I prefer a textfile over a userinterface anyway. Ralf From linux@mailhost.uni-koblenz.de Mon May 20 18:45:00 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id SAA21641 for ; Mon, 20 May 1996 18:44:59 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA28621; Mon, 20 May 1996 18:44:23 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.5/8.6.9) with SMTP id SAA15613 for ; Mon, 20 May 1996 18:43:54 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605201643.SAA15613@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id SAA29457; Mon, 20 May 1996 18:43:06 +0200 Subject: Kernel patches To: linux-mips@fnet.fr Date: Mon, 20 May 1996 18:43:06 +0200 (MET DST) Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1474 Lines: 42 -----BEGIN PGP SIGNED MESSAGE----- Hi all, I've uploaded two new kernel patches to ftp.fnet.fr; 1.3.98 will be of special interest because it does a big cleanup in the sources. As I alrready have posted the board specific sources are now in extra subdirectories in arch/mips/. 1.3.98 now also can handle the bigger SysV typedef struct { int __sigbits[4]; } sigset_t; correctly. Due to extra fields that already had been reserved in struct sigaction for this extension most userland binaries have survived this change; the only exception should be binaries that we're compiled using the old sigset_t defined in directly. Linux/MIPS 1.3.98 also features support for SNI RM200 PCI, some changes to the low level memory management code to adapt it to the "real world R4600" and last, not least, Wayne's G364 console for Magnum 4000/Olivetti M700. Still without cursor & colour but more important without crashes ... For booting Linux/MIPS on a RM200 you'll need a tuned Milo version 0.26; I've sent my patches to Stoned. Ralf d4afc1918986674b9982f33f1bd73902 linux-1.3.96-1.3.97.diffs.gz 23b0f1e8f2c9eb9ec744a9a51af3171c linux-1.3.97-1.3.98.diffs.gz -----BEGIN PGP SIGNATURE----- Version: 2.6.2 iQCVAgUBMaChF0ckbl6vezDBAQGmMwP+PTWIHivpkv0+/5WaSupZsCPTJfi9oCDI qvrvFkki9XfIXu1QqAkbLR4Z8DE2+WWnSfuGKInwhWmnEGLmZCUgudiag6N5KeYq JiK4wJ5gh3aGLL8pn8Iwb+WCqoqG50OB+5Yy+jthZtVhKFQh/9yAE1tk4TnKuTzn e6FO5fQ22KY= =DCaP -----END PGP SIGNATURE----- From Stoned.Elipot@univ-evry.fr Mon May 20 19:15:58 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id TAA22357 for ; Mon, 20 May 1996 19:15:58 +0200 (MET DST) Received: from amiante.univ-evry.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA29048; Mon, 20 May 1996 19:15:40 +0200 (MET) Received: from bandsept.univ-evry.fr (bandsept.univ-evry.fr [192.93.214.19]) by amiante.univ-evry.fr with ESMTP (8.6.12/96.04.09/louis-blanc); id TAA29983; Mon, 20 May 1996 19:15:54 +0200 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 TAA11026; Mon, 20 May 1996 19:15:49 +0200 Message-Id: <199605201715.TAA11026@bandsept.univ-evry.fr> To: linux-mips@fnet.fr Subject: Re: Kernel patches In-Reply-To: Your message of "Mon, 20 May 1996 18:43:06 +0200." <199605201643.SAA15613@informatik.uni-koblenz.de> X-It: bandsept (wrong OS, but right CPU) X-Mailer: MH 6.8.3, MH-E 5.0.2 Date: Mon, 20 May 1996 19:15:42 +0200 From: Stoned Elipot Content-Length: 611 Lines: 15 >>>>> "Ralf" == Systemkennung Linux writes: [SNIP] Ralf> For booting Linux/MIPS on a RM200 you'll need a tuned Milo Ralf> version 0.26; I've sent my patches to Stoned. Ok ok it means I should finish it really quickly :). Oh and you will need a patch on kernel (Ralf you got me, I'll put back what I'd done in 1.3.96 into 1.3.98,... I hoped you would have to do it, you were faster than me ;) in order to use Milo 0.27. Ought it's getting complicated, perhaps we should find a way of putting back milo in kernel source tree, somewhere under arch/mips/{arc|jazz}/boot ? Stoned. From imp@rover.village.org Mon May 20 20:41:26 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id UAA24006 for ; Mon, 20 May 1996 20:41:25 +0200 (MET DST) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA00520; Mon, 20 May 1996 20:40:25 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.7.5/8.6.6) with SMTP id MAA23176 for ; Mon, 20 May 1996 12:40:30 -0600 (MDT) Message-Id: <199605201840.MAA23176@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Kernel patches In-Reply-To: Your message of Mon, 20 May 1996 19:15:42 +0200 Date: Mon, 20 May 1996 12:40:30 -0600 From: Warner Losh Content-Length: 828 Lines: 19 : Ought it's getting complicated, perhaps we should find a way of : putting back milo in kernel source tree, somewhere under : arch/mips/{arc|jazz}/boot ? There already is a very old version of MILO in the arch/mips/boot directory. One that I had to create a dummy makefile for because I don't have all the right header files to do the kernel build otherwise. However, there is one minor problem with this idea. There are currently (as of .26 anyway) parts of MILO that cannot be distributed as source and are binary only (although some of the information in the binary only code is also in the sources to OpenBSD). It may be tricky to arrange for things of this nature to be in the tree. Have fun at the pre-2.0 conference to all those that are lucky enough to be going. Say hi to Drew for me if you see him :-) Warner From paul@suite.sw.oz.au Mon May 20 22:42:40 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id WAA26528 for ; Mon, 20 May 1996 22:42:40 +0200 (MET DST) Received: from staff.cs.su.OZ.AU by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA02182; Mon, 20 May 1996 22:42:20 +0200 (MET) Received: from suite.sw.oz.au by swallow.sw.oz.au with SMTP id UAA02416; Mon, 20 May 1996 20:42:21 GMT (8.6.10/Unixware) (from paul@suite.sw.oz.au for ) Received: by suite.sw.oz.au id AA22178; Tue, 21 May 1996 06:42:19 +1000 (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: <199605202042.AA22178@suite.sw.oz.au> Subject: Re: Kernel patches To: linux-mips@fnet.fr Date: Tue, 21 May 1996 06:42:18 +1000 (EST) In-Reply-To: <199605201643.SAA15613@informatik.uni-koblenz.de> from "Systemkennung Linux" at May 20, 96 06:43:06 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: 954 Lines: 21 Ralf said: > Linux/MIPS 1.3.98 also features support for SNI RM200 PCI, some > changes to the low level memory management code to adapt it to the > "real world R4600" and last, not least, Wayne's G364 console for Magnum > 4000/Olivetti M700. Still without cursor & colour but more important > without crashes ... I must say that you've done a great job to get all this together. Now, can I bother you to put up a full 1.3.98 source tree snapshot...? It'll save me patching up a kernel all the way. 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 May 20 22:46:46 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id WAA26617 for ; Mon, 20 May 1996 22:46:45 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA02213; Mon, 20 May 1996 22:46:10 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.5/8.6.9) with SMTP id WAA24417 for ; Mon, 20 May 1996 22:45:48 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605202045.WAA24417@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id WAA29690; Mon, 20 May 1996 22:45:00 +0200 Subject: Re: Kernel patches To: linux-mips@fnet.fr Date: Mon, 20 May 1996 22:44:59 +0200 (MET DST) In-Reply-To: <199605201840.MAA23176@rover.village.org> from "Warner Losh" at May 20, 96 12:40:30 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 751 Lines: 20 Hi, > However, there is one minor problem with this idea. There are > currently (as of .26 anyway) parts of MILO that cannot be distributed > as source and are binary only (although some of the information in the > binary only code is also in the sources to OpenBSD). It may be tricky > to arrange for things of this nature to be in the tree. I think I have to new sources for ARC documentation: - SNI will hopefully help us - A part of the information is contained in the M700 technical manual > Have fun at the pre-2.0 conference to all those that are lucky enough > to be going. Say hi to Drew for me if you see him :-) I'll be there and I'd be happy to meet as many people from this list as possible. Who else will be there? Ralf From linux@mailhost.uni-koblenz.de Mon May 20 22:48:55 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id WAA26686 for ; Mon, 20 May 1996 22:48:54 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA02232; Mon, 20 May 1996 22:47:49 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.5/8.6.9) with SMTP id WAA24466 for ; Mon, 20 May 1996 22:47:29 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605202047.WAA24466@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id WAA29699; Mon, 20 May 1996 22:46:41 +0200 Subject: Re: Kernel patches To: linux-mips@fnet.fr Date: Mon, 20 May 1996 22:46:41 +0200 (MET DST) In-Reply-To: <199605201715.TAA11026@bandsept.univ-evry.fr> from "Stoned Elipot" at May 20, 96 07:15:42 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 286 Lines: 10 Hi, > Ought it's getting complicated, perhaps we should find a way of > putting back milo in kernel source tree, somewhere under > arch/mips/{arc|jazz}/boot ? Not a good idea; Linus will probably reject this patch (or why are loadlin & Lilo no parts of the kernel sources)? Ralf From Stoned.Elipot@univ-evry.fr Mon May 20 22:53:46 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id WAA26943 for ; Mon, 20 May 1996 22:53:45 +0200 (MET DST) Received: from amiante.univ-evry.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA02365; Mon, 20 May 1996 22:53:28 +0200 (MET) Received: from bandsept.univ-evry.fr (bandsept.univ-evry.fr [192.93.214.19]) by amiante.univ-evry.fr with ESMTP (8.6.12/96.04.09/louis-blanc); id WAA00412; Mon, 20 May 1996 22:53:42 +0200 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 WAA11272; Mon, 20 May 1996 22:53:40 +0200 Message-Id: <199605202053.WAA11272@bandsept.univ-evry.fr> To: linux-mips@fnet.fr Subject: Re: Kernel patches In-Reply-To: Your message of "Mon, 20 May 1996 22:46:41 +0200." <199605202047.WAA24466@informatik.uni-koblenz.de> X-It: bandsept (wrong OS, but right CPU) X-Mailer: MH 6.8.3, MH-E 5.0.2 Date: Mon, 20 May 1996 22:53:33 +0200 From: Stoned Elipot Content-Length: 215 Lines: 8 >Not a good idea; Linus will probably reject this patch (or why >are loadlin & Lilo no parts of the kernel sources)? So we should better ripe off from the kernel sources the milo's ones, shouldn't we ? Stoned. From linux@mailhost.uni-koblenz.de Mon May 20 23:03:16 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id XAA27312 for ; Mon, 20 May 1996 23:03:15 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA02516; Mon, 20 May 1996 23:02:55 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.5/8.6.9) with SMTP id XAA24845 for ; Mon, 20 May 1996 23:02:37 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605202102.XAA24845@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id XAA29724; Mon, 20 May 1996 23:01:50 +0200 Subject: Re: Kernel patches To: linux-mips@fnet.fr Date: Mon, 20 May 1996 23:01:49 +0200 (MET DST) In-Reply-To: <199605202042.AA22178@suite.sw.oz.au> from "Paul Antoine" at May 21, 96 06:42:18 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 461 Lines: 12 Hi, > I must say that you've done a great job to get all this together. Now, > can I bother you to put up a full 1.3.98 source tree snapshot...? It'll > save me patching up a kernel all the way. Sorry, not until after the Linux Kongress. Fate has hit me hard twice and I'm happy that I can still be at the Kongress and hold my speech about Linux ports (except 68k which will be proceeded by Roman Hodek). I'll also show Linux running on the RM200. Ralf From linux@mailhost.uni-koblenz.de Tue May 21 04:28:12 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id EAA05476 for ; Tue, 21 May 1996 04:28:10 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA07105; Tue, 21 May 1996 04:27:51 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.5/8.6.9) with SMTP id EAA15691 for ; Tue, 21 May 1996 04:27:38 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605210227.EAA15691@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id EAA29938; Tue, 21 May 1996 04:26:49 +0200 Subject: Cascade interrupt To: linux-mips@fnet.fr Date: Tue, 21 May 1996 04:26:48 +0200 (MET DST) Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 359 Lines: 9 Hi all, while playing around with the RM200 I had to find out that some of the "RISC PC" class machines use the PICs in cascade mode while others don't. This is completly opposed the the standard Intel PC design where the PICs are always in cascade mode. So just in case if you wonder why your machine never gets high interrupts (>= 8) check this. Ralf From imp@rover.village.org Tue May 21 07:18:37 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id HAA09017 for ; Tue, 21 May 1996 07:18:35 +0200 (MET DST) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA09984; Tue, 21 May 1996 07:18:12 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.7.5/8.6.6) with SMTP id XAA26645 for ; Mon, 20 May 1996 23:18:24 -0600 (MDT) Message-Id: <199605210518.XAA26645@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Cascade interrupt In-Reply-To: Your message of Tue, 21 May 1996 04:26:48 +0200 Date: Mon, 20 May 1996 23:18:23 -0600 From: Warner Losh Content-Length: 904 Lines: 18 : while playing around with the RM200 I had to find out that some of the : "RISC PC" class machines use the PICs in cascade mode while others don't. : This is completly opposed the the standard Intel PC design where the PICs : are always in cascade mode. So just in case if you wonder why your : machine never gets high interrupts (>= 8) check this. Hmmmm. Interesting. Was it here that I reported that I never got high interrupts. Just grumpy low ones that wish they could aford the good stuff to get high? If the PICs are in cascade mode, what mode are they in? And should I force them to be in cascade mode? Looks like it is time to get a good data sheet on the PICs. BTW, anybody know a good source for my opti chip set data sheets? After seeing the one for the Dallas Semi DS1488 in my machine, I want them for more of my chips. That is assuming that Alta-vista can't help me.... Warner From hodgen@mailhost.uni-koblenz.de Tue May 21 09:57:04 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id JAA12020 for ; Tue, 21 May 1996 09:57:04 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA12258; Tue, 21 May 1996 09:55:16 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.5/8.6.9) with SMTP id JAA24247 for ; Tue, 21 May 1996 09:54:35 +0200 (MET DST) From: Wayne Hodgen Message-Id: <199605210754.JAA24247@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id JAA00178; Tue, 21 May 1996 09:53:45 +0200 Subject: Re: Kernel patches To: linux-mips@fnet.fr Date: Tue, 21 May 1996 09:53:45 +0200 (MET DST) In-Reply-To: <199605202102.XAA24845@informatik.uni-koblenz.de> from "Systemkennung Linux" at May 20, 96 11:01:49 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 753 Lines: 17 Hi, > > I must say that you've done a great job to get all this together. Now, > > can I bother you to put up a full 1.3.98 source tree snapshot...? It'll > > save me patching up a kernel all the way. > > Sorry, not until after the Linux Kongress. Fate has hit me hard twice Hmm, if someone gives me ftp access and a path to put it in, I can upload it (I have the full source tree that was used to generate the patches :) Bye -- Wayne Hodgen | hodgen@informatik.uni-koblenz.de | #include From paul@suite.sw.oz.au Tue May 21 13:25:54 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id NAA16269 for ; Tue, 21 May 1996 13:25:53 +0200 (MET DST) Received: from staff.cs.su.OZ.AU by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA16849; Tue, 21 May 1996 13:25:26 +0200 (MET) Received: from suite.sw.oz.au by swallow.sw.oz.au with SMTP id JAA19980; Tue, 21 May 1996 09:04:11 GMT (8.6.10/Unixware) (from paul@suite.sw.oz.au for ) Received: by suite.sw.oz.au id AA00155; Tue, 21 May 1996 19:04:09 +1000 (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: <199605210904.AA00155@suite.sw.oz.au> Subject: Re: Kernel patches To: linux-mips@fnet.fr Date: Tue, 21 May 1996 19:04:08 +1000 (EST) In-Reply-To: <199605202102.XAA24845@informatik.uni-koblenz.de> from "Systemkennung Linux" at May 20, 96 11:01:49 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: 1161 Lines: 26 > > I must say that you've done a great job to get all this together. Now, > > can I bother you to put up a full 1.3.98 source tree snapshot...? It'll > > save me patching up a kernel all the way. > > Sorry, not until after the Linux Kongress. Fate has hit me hard twice No problem - I just fetched the source and patches. I now have a 1.3.98 kernel but it will not build for R3000. gas seems to be complaining that gcc and/or the inline code in system.h is emitting R4K instructions. > and I'm happy that I can still be at the Kongress and hold my speech > about Linux ports (except 68k which will be proceeded by Roman Hodek). > I'll also show Linux running on the RM200. Have a great time! 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 tor@spacetec.no Tue May 21 11:19:35 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id LAA13772 for ; Tue, 21 May 1996 11:19:34 +0200 (MET DST) Received: from pallas.spacetec.no by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA14159; Tue, 21 May 1996 11:19:14 +0200 (MET) Received: (from tor@localhost) by pallas.spacetec.no (8.7.1/8.7.1) id LAA03195 for linux-mips@fnet.fr; Tue, 21 May 1996 11:15:58 +0200 Message-Id: <199605210915.LAA03195@pallas.spacetec.no> From: tor@spacetec.no (Tor Arntsen) Date: Tue, 21 May 1996 11:15:58 +0200 In-Reply-To: Systemkennung Linux "Re: Kernel patches" (May 20, 21:45) X-Mailer: Mail User's Shell (7.2.5 10/14/92) To: linux-mips@fnet.fr Subject: Re: Kernel patches Content-Length: 427 Lines: 11 On May 20, 21:45, Systemkennung Linux wrote: >> Have fun at the pre-2.0 conference to all those that are lucky enough >> to be going. Say hi to Drew for me if you see him :-) > >I'll be there and I'd be happy to meet as many people from this list >as possible. Who else will be there? It's clear to me that I'm currently suffering from not having time to read news or anything.. where's that conference? (and when :-) -Tor From ehsjony@ehs.ericsson.se Tue May 21 12:53:02 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id MAA15513 for ; Tue, 21 May 1996 12:52:57 +0200 (MET DST) Received: from glacier.wise.edt.ericsson.se (glacier-ext.wise.edt.ericsson.se) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA16211; Tue, 21 May 1996 12:52:34 +0200 (MET) Received: from ehs (ehs.ericsson.se [130.100.157.2]) by glacier.wise.edt.ericsson.se (8.7.5/8.7.3/glacier-0.9) with SMTP id MAA07622 for ; Tue, 21 May 1996 12:52:45 +0200 (MET DST) Received: from vkpc131 (vkpc128.ericsson.se) by ehs (4.1/SMI-4.1-LME1.6) id AA17003; Tue, 21 May 96 12:52:44 +0200 Message-Id: <31A1AECA.5917@ehs.ericsson.se> Date: Tue, 21 May 1996 12:53:46 +0100 From: Jonas Nygren Organization: Ericsson Hewlett-Packard Telecommunications AB X-Mailer: Mozilla 2.01Gold (WinNT; I) Mime-Version: 1.0 To: linux-mips@fnet.fr Subject: Interested in Linux for DEC3100 X-Url: http://lena.fnet.fr/ Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Length: 914 Lines: 24 I have an old DEC3100 with a disk and tape (Dec RK RZ something) which I would like to use as backup server for my Linux 586 box. I see two alternatives: 1) keep old ultrix (3.1 I believe) and connect it to Linux via ethernet: ftp, rlogin/telnet and preferably NFS. 2) get Linux running on the DEC3100. To really use it I would have to have drivers for the DEC disc and tape. They are SCSI but may have their peculiarities. Do you have any suggestions? I would like to do it in a two step mode: 1 and then 2. Do you know of anybody doing 1 and can give me a reference - I need to identify type of ethernet plus driver for Linux 586 that will fit together with the DEC3100. When I get a connection I could be interested to do some kernel hacking to get Linux on the DEC. Though, my experience is limited and so is my freetime. At least I should be able to do some testing. Best regards Jonas Nygren From linux@mailhost.uni-koblenz.de Tue May 21 14:27:03 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id OAA17696 for ; Tue, 21 May 1996 14:27:02 +0200 (MET DST) Received: from relay2.EUnet.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA18007; Tue, 21 May 1996 14:26:43 +0200 (MET) Received: from informatik.uni-koblenz.de ([141.26.4.1]) by relay2.eunet.fr (5.65c8d/96.05.03) via EUnet-France id AA07026; Tue, 21 May 1996 14:20:34 +0200 (MET) Received: from thoma (thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.7.5/8.6.9) with SMTP id OAA07347 for ; Tue, 21 May 1996 14:15:20 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605211215.OAA07347@informatik.uni-koblenz.de> Received: by thoma (SMI-8.6/KO-2.0) id OAA11211; Tue, 21 May 1996 14:15:16 +0200 Subject: Re: Cascade interrupt To: linux-mips@fnet.fr Date: Tue, 21 May 1996 14:15:15 +0200 (MET DST) In-Reply-To: <199605210518.XAA26645@rover.village.org> from "Warner Losh" at May 20, 96 11:18:23 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 2413 Lines: 54 Hi, > : while playing around with the RM200 I had to find out that some of the > : "RISC PC" class machines use the PICs in cascade mode while others don't. > : This is completly opposed the the standard Intel PC design where the PICs > : are always in cascade mode. So just in case if you wonder why your > : machine never gets high interrupts (>= 8) check this. > > Hmmmm. Interesting. Was it here that I reported that I never got > high interrupts. Just grumpy low ones that wish they could aford the > good stuff to get high? > > If the PICs are in cascade mode, what mode are they in? And should I > force them to be in cascade mode? I'd assume the firmware initializes them correctly. Big trick with this cascade mode is that the interrupt output of PIC 2 is wired to interrupt input 2 of PIC 1 and then some cascade bit is set via software. I've lost my data sheets when I left at Waldorf, so everything is just from memory ... What you could try to check if you get the cadcade interrupt #2 instead of a high interrupt. Or simply try the following change: lui s0,%hi(PORT_BASE) li a0,0x0f sb a0,%lo(PORT_BASE+0x20)(s0) # poll command lb a0,%lo(PORT_BASE+0x20)(s0) # read result bgtz a0,poll_second andi a0,7 beq a0,2,poll_second # cascade? li s1,1 # delay slot ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ Insert these two lines /* * Acknowledge first pic */ When PICs are in cascade mode and the handler sees and INT #2 it assumes that it really received an high interrupt and polls the second PIC. This change fixed it for the RM200. Looks like it is time to get a good > data sheet on the PICs. BTW, anybody know a good source for my opti > chip set data sheets? After seeing the one for the Dallas Semi DS1488 > in my machine, I want them for more of my chips. That is assuming > that Alta-vista can't help me.... Hmm, I've never tried to get data-sheets via Alta-vista. But If you want to try, the PIC's are from Intel type number 8259. It might be difficult to get data sheets for these chips because nowadays the PICs are integrated into either chipsets or even the CPU. Ralf From linux@mailhost.uni-koblenz.de Tue May 21 14:19:57 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id OAA17535 for ; Tue, 21 May 1996 14:19:56 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA17752; Tue, 21 May 1996 14:18:34 +0200 (MET) Received: from thoma (thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.7.5/8.6.9) with SMTP id OAA07441 for ; Tue, 21 May 1996 14:17:37 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605211217.OAA07441@informatik.uni-koblenz.de> Received: by thoma (SMI-8.6/KO-2.0) id OAA11231; Tue, 21 May 1996 14:17:34 +0200 Subject: Re: Kernel patches To: linux-mips@fnet.fr Date: Tue, 21 May 1996 14:17:34 +0200 (MET DST) In-Reply-To: <199605210754.JAA24247@informatik.uni-koblenz.de> from "Wayne Hodgen" at May 21, 96 09:53:45 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 574 Lines: 15 Hi, > > > I must say that you've done a great job to get all this together. Now, > > > can I bother you to put up a full 1.3.98 source tree snapshot...? It'll > > > save me patching up a kernel all the way. > > > > Sorry, not until after the Linux Kongress. Fate has hit me hard twice > > Hmm, if someone gives me ftp access and a path to put it in, I can upload > it (I have the full source tree that was used to generate the patches :) Small correction: you have the dirty .orig source tree that was used to generate the patch thus generating a reject ;-) Ralf From hodgen@mailhost.uni-koblenz.de Tue May 21 14:38:40 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id OAA18202 for ; Tue, 21 May 1996 14:38:39 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA18235; Tue, 21 May 1996 14:34:14 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.5/8.6.9) with SMTP id OAA08170 for ; Tue, 21 May 1996 14:31:53 +0200 (MET DST) From: Wayne Hodgen Message-Id: <199605211231.OAA08170@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id OAA02439; Tue, 21 May 1996 14:31:04 +0200 Subject: Re: Kernel patches To: linux-mips@fnet.fr Date: Tue, 21 May 1996 14:31:03 +0200 (MET DST) In-Reply-To: <199605211217.OAA07441@informatik.uni-koblenz.de> from "Systemkennung Linux" at May 21, 96 02:17:34 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1046 Lines: 22 Hi, > > > > I must say that you've done a great job to get all this together. Now, > > > > can I bother you to put up a full 1.3.98 source tree snapshot...? It'll > > > > save me patching up a kernel all the way. > > > > > > Sorry, not until after the Linux Kongress. Fate has hit me hard twice > > > > Hmm, if someone gives me ftp access and a path to put it in, I can upload > > it (I have the full source tree that was used to generate the patches :) > > Small correction: you have the dirty .orig source tree that was used to > generate the patch thus generating a reject ;-) You mean you didn't create a new set of diffs using the source on my machine? With the few changes needed to get it working on the Olli? -- Wayne Hodgen | hodgen@informatik.uni-koblenz.de | #include From linux@mailhost.uni-koblenz.de Tue May 21 15:01:57 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id PAA19138 for ; Tue, 21 May 1996 15:01:57 +0200 (MET DST) Received: from relay2.EUnet.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA18874; Tue, 21 May 1996 15:01:39 +0200 (MET) Received: from informatik.uni-koblenz.de ([141.26.4.1]) by relay2.eunet.fr (5.65c8d/96.05.03) via EUnet-France id AA09551; Tue, 21 May 1996 14:52:53 +0200 (MET) Received: from thoma (thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.7.5/8.6.9) with SMTP id OAA08973 for ; Tue, 21 May 1996 14:46:54 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605211246.OAA08973@informatik.uni-koblenz.de> Received: by thoma (SMI-8.6/KO-2.0) id OAA11636; Tue, 21 May 1996 14:46:51 +0200 Subject: Re: Kernel patches To: linux-mips@fnet.fr Date: Tue, 21 May 1996 14:46:50 +0200 (MET DST) In-Reply-To: <199605210915.LAA03195@pallas.spacetec.no> from "Tor Arntsen" at May 21, 96 11:15:58 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1623 Lines: 41 Hi, > On May 20, 21:45, Systemkennung Linux wrote: > >> Have fun at the pre-2.0 conference to all those that are lucky enough > >> to be going. Say hi to Drew for me if you see him :-) > > > >I'll be there and I'd be happy to meet as many people from this list > >as possible. Who else will be there? > > It's clear to me that I'm currently suffering from not having time to read > news or anything.. where's that conference? (and when :-) Below is a list of speakers that will be at the congress. Last years congress was cool; this year's will be even better :-) For more information look on www.linux-kongress.de. As I said it will be cool, so just take (or steal ...) yourself the time and come :-) Ralf Alan Cox Alexandre Julliard Dirk H. Hohndel Drew Eckhardt Erik Troan Fritz Elfert Gerhard Friesland-Koepke Hans Lermen Harald Koenig Hardy Buchholz Lars Wirzenius Linus Torvalds Michael Hipp Olaf Kirch Raj Mathur Ralf Baechle Remy Card Roman Hodek Stephen Tweedie Theodore Ts'o Volker Lendecke Werner Almesberger From Stoned.Elipot@univ-evry.fr Tue May 21 14:55:29 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id OAA18661 for ; Tue, 21 May 1996 14:55:29 +0200 (MET DST) Received: from amiante.univ-evry.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA18701; Tue, 21 May 1996 14:55:11 +0200 (MET) Received: from bandsept.univ-evry.fr (bandsept.univ-evry.fr [192.93.214.19]) by amiante.univ-evry.fr with ESMTP (8.6.12/96.04.09/louis-blanc); id OAA02658; Tue, 21 May 1996 14:55:26 +0200 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 OAA13936; Tue, 21 May 1996 14:55:18 +0200 Message-Id: <199605211255.OAA13936@bandsept.univ-evry.fr> To: linux-mips@fnet.fr Subject: Re: Kernel patches In-Reply-To: Your message of "Tue, 21 May 1996 14:31:03 +0200." <199605211231.OAA08170@informatik.uni-koblenz.de> X-It: bandsept (wrong OS, but right CPU) X-Mailer: MH 6.8.3, MH-E 5.0.2 Date: Tue, 21 May 1996 14:55:08 +0200 From: Stoned Elipot Content-Length: 419 Lines: 12 Hi all, >>>>> "Wayne" == Wayne Hodgen writes: [SNIP] Wayne> You mean you didn't create a new set of diffs using the source Wayne> on my machine? With the few changes needed to get it working Wayne> on the Olli? Diffs 1.3.97 and 1.3.98 uploaded yesterday by Ralf produced this nigth a fine running kernel on my Olli with a wondeful fast black and white console. thanks Wayne! Stoned. From linux@mailhost.uni-koblenz.de Tue May 21 15:05:39 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id PAA19256 for ; Tue, 21 May 1996 15:05:38 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA18882; Tue, 21 May 1996 15:02:49 +0200 (MET) Received: from thoma (thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.7.5/8.6.9) with SMTP id PAA09636 for ; Tue, 21 May 1996 15:00:04 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605211300.PAA09636@informatik.uni-koblenz.de> Received: by thoma (SMI-8.6/KO-2.0) id PAA11811; Tue, 21 May 1996 15:00:02 +0200 Subject: Re: Kernel patches To: linux-mips@fnet.fr Date: Tue, 21 May 1996 15:00:01 +0200 (MET DST) In-Reply-To: <199605211231.OAA08170@informatik.uni-koblenz.de> from "Wayne Hodgen" at May 21, 96 02:31:03 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 294 Lines: 9 Hi, > You mean you didn't create a new set of diffs using the source on my machine? > With the few changes needed to get it working on the Olli? I did; the uploaded patch will however produce a reject on one file because the .orig tree on Ozzy wasn't as original as the name implied. Ralf From dom@algor.co.uk Tue May 21 15:44:25 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id PAA20058 for ; Tue, 21 May 1996 15:44:24 +0200 (MET DST) Received: from moorgate.algor.co.uk by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA19827; Tue, 21 May 1996 15:44:01 +0200 (MET) Received: from bank.algor.co.uk (bank.algor.co.uk [193.117.190.11]) by moorgate.algor.co.uk (8.7.4/8.7.3) with SMTP id OAA18097 for ; Tue, 21 May 1996 14:44:10 +0100 (BST) From: Dominic Sweetman Date: Tue, 21 May 96 14:44:10 +0100 Message-Id: <4066.9605211344@bank.algor.co.uk> To: linux-mips@fnet.fr Subject: Re: Cascade interrupt In-Reply-To: <199605211215.OAA07347@informatik.uni-koblenz.de> References: <199605210518.XAA26645@rover.village.org> <199605211215.OAA07347@informatik.uni-koblenz.de> Content-Length: 833 Lines: 20 > Hmm, I've never tried to get data-sheets via Alta-vista. But If you > want to try, the PIC's are from Intel type number 8259. It might be > difficult to get data sheets for these chips because nowadays the PICs > are integrated into either chipsets or even the CPU. Intel publish lots of books, just find your local Intel distributor. There are lots of datasheets on WWW and anonymous ftp servers. While an Alta Vista (I find Lycos is often better) search will sometimes turn up a data sheet in one go, there's an awful lot of ftp..com sites (eg man = opti... try it). File formats vary. Opti seem to like self-extracting executable DOS archives, and Adobe PDF is widespread (get xpdf... it's much more likely to run than Acrobat). It helps to have a DOS/Windows PC (spit!) on hand. Dominic Sweetman dom@algor.co.uk From imp@rover.village.org Tue May 21 15:54:00 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id PAA20237 for ; Tue, 21 May 1996 15:53:57 +0200 (MET DST) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA20063; Tue, 21 May 1996 15:53:25 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.7.5/8.6.6) with SMTP id HAA28270 for ; Tue, 21 May 1996 07:53:26 -0600 (MDT) Message-Id: <199605211353.HAA28270@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Cascade interrupt In-Reply-To: Your message of Tue, 21 May 1996 14:44:10 BST Date: Tue, 21 May 1996 07:53:25 -0600 From: Warner Losh Content-Length: 1186 Lines: 27 : Intel publish lots of books, just find your local Intel distributor. My day job gives me wonderful access to a huge library. Sadly, the volume that has the 82xx stuff in it that is interesting is on walkabout... : There are lots of datasheets on WWW and anonymous ftp servers. While : an Alta Vista (I find Lycos is often better) search will sometimes : turn up a data sheet in one go, there's an awful lot of ftp..com : sites (eg man = opti... try it). Yes. I noticed that. Intel has only a few of their docs online, especially for older chips. The newer chipsets seem to typically say "with two 8279A compatible chips built in" and then never describes tham further. I'll have to try Lycos, because I'm very happy with Alta Vista. : File formats vary. Opti seem to like self-extracting executable : DOS archives, and Adobe PDF is widespread (get xpdf... it's : much more likely to run than Acrobat). It helps to have a DOS/Windows : PC (spit!) on hand. There are a bunch of file unpackers for Unix that seem to get the job done most of the time.... Alas, I have a DOS/Windows box to run the accounting software for the Village and my consulting business. Warner From imp@rover.village.org Tue May 21 17:07:44 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id RAA21858 for ; Tue, 21 May 1996 17:07:43 +0200 (MET DST) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA21802; Tue, 21 May 1996 17:07:02 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.7.5/8.6.6) with SMTP id JAA29145 for ; Tue, 21 May 1996 09:06:51 -0600 (MDT) Message-Id: <199605211506.JAA29145@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Cascade interrupt In-Reply-To: Your message of Tue, 21 May 1996 14:15:15 +0200 Date: Tue, 21 May 1996 09:06:51 -0600 From: Warner Losh Content-Length: 1311 Lines: 27 : I'd assume the firmware initializes them correctly. Big trick with this : cascade mode is that the interrupt output of PIC 2 is wired to interrupt : input 2 of PIC 1 and then some cascade bit is set via software. I've : lost my data sheets when I left at Waldorf, so everything is just from : memory ... OK. I understand what you are saying now. When the IRQ 2 interrupt happens, it means that I have to manually go out and check interrupts 8-15 rather than having it just happen. : beq a0,2,poll_second # cascade? : li s1,1 # delay slot : ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ Insert these two lines Hmmm, similar lines in rpc44.S do indeed solve the problem. My Buslogic card seems to be properly initialized. Or at least the interrupts for it are happening now. It is having some problems on a bare naked scsi bus (well, one that has one device off attached externally), but I'm checking now with a populated scsi bus. Hmmm. I have a target 3 drive on the bus, properly terminated, and am seeing that it is aboriting commands and resetting the bus. I get a lot of bus reset for host 0 messages. In fact, all the commands seem to be coming back with all 0's. Hmmm, well, at least it is progress :-). Warner From bruce@pixar.com Tue May 21 18:28:01 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id SAA23727 for ; Tue, 21 May 1996 18:28:00 +0200 (MET DST) Received: from mongo.pixar.com by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA24069; Tue, 21 May 1996 18:27:40 +0200 (MET) Received: by mongo.pixar.com (Smail3.1.28.1 #15) id m0uLuHv-000524C; Tue, 21 May 96 09:27 PDT Message-Id: Date: Tue, 21 May 96 09:27 PDT From: bruce@pixar.com (Bruce Perens) To: linux-mips@fnet.fr, linux-mips@fnet.fr Subject: Re: Cascade interrupt Content-Length: 489 Lines: 12 The 8259 has _three_ cascade inputs (0, 1, and 2) and a slave-master input pin according to the "Microsoft Programmer's PC Sourcebook". Unfortunately, they don't give the register description :-( . Thus, I think it might be at least possible for a cascade interrupt to come up on another IRQ than 2 if someone wires it that way. Who is Intel's second-source for this chip? If you search for *8259* on other manufacturers WWW pages (TI, Motorola, etc.) you might find it. Thanks Bruce From linux@mailhost.uni-koblenz.de Tue May 21 21:08:23 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id VAA27191 for ; Tue, 21 May 1996 21:08:22 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA26954; Tue, 21 May 1996 21:07:56 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.5/8.6.9) with SMTP id VAA27706 for ; Tue, 21 May 1996 21:06:10 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605211906.VAA27706@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id VAA03328; Tue, 21 May 1996 21:05:21 +0200 Subject: Re: Cascade interrupt To: linux-mips@fnet.fr Date: Tue, 21 May 1996 21:05:21 +0200 (MET DST) In-Reply-To: from "Bruce Perens" at May 21, 96 09:27:00 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 294 Lines: 9 Hi, > Who is Intel's second-source for this chip? If you search for *8259* on > other manufacturers WWW pages (TI, Motorola, etc.) you might find it. Last information I heared about one year ago is that this chip is no longer manufactured but only integrated into some VLSI chipsets. Ralf From linux@mailhost.uni-koblenz.de Tue May 21 21:27:09 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id VAA27534 for ; Tue, 21 May 1996 21:27:08 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA27230; Tue, 21 May 1996 21:26:14 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.5/8.6.9) with SMTP id VAA28301 for ; Tue, 21 May 1996 21:25:15 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605211925.VAA28301@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id VAA03357; Tue, 21 May 1996 21:24:26 +0200 Subject: Re: Cascade interrupt To: linux-mips@fnet.fr Date: Tue, 21 May 1996 21:24:25 +0200 (MET DST) In-Reply-To: <199605211506.JAA29145@rover.village.org> from "Warner Losh" at May 21, 96 09:06:51 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1852 Lines: 38 Hi, > OK. I understand what you are saying now. When the IRQ 2 interrupt > happens, it means that I have to manually go out and check interrupts > 8-15 rather than having it just happen. > > : beq a0,2,poll_second # cascade? > : li s1,1 # delay slot > : ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ Insert these two lines > > Hmmm, similar lines in rpc44.S do indeed solve the problem. My > Buslogic card seems to be properly initialized. Or at least the > interrupts for it are happening now. It is having some problems on a > bare naked scsi bus (well, one that has one device off attached > externally), but I'm checking now with a populated scsi bus. > > Hmmm. I have a target 3 drive on the bus, properly terminated, and am > seeing that it is aboriting commands and resetting the bus. I get a > lot of bus reset for host 0 messages. In fact, all the commands seem > to be coming back with all 0's. Hmmm, well, at least it is progress > :-). I had those aborts with the DPT, too. This could either mean that the driver still doesn't get interrupts or you have problems with the caches. The DPT eg writes status information about a command into the memory. You now read something from this data which also (KSEG0!) is written into the cache while to command isn't complete yet. Later on this (or another) command is complete, you attempt to read the stale from the memory but instead get the data from the cache - the controller waits longer for the data ... Sometimes this effect is quite invisible; eg. the DPT needed 25 minutes for a certain disk benchmark. After adding the cacheflush only 13:24 left with the data flowing continuously, not in bursts. Ralf PS: Rule of thumb for MIPS programmers: If it doesn't work blame the caches ;-) From bruce@pixar.com Tue May 21 21:38:10 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id VAA27740 for ; Tue, 21 May 1996 21:38:09 +0200 (MET DST) Received: from mongo.pixar.com by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA27417; Tue, 21 May 1996 21:37:51 +0200 (MET) Received: by mongo.pixar.com (Smail3.1.28.1 #15) id m0uLxG1-000524C; Tue, 21 May 96 12:38 PDT Message-Id: Date: Tue, 21 May 96 12:38 PDT From: bruce@pixar.com (Bruce Perens) To: linux-mips@fnet.fr Subject: Re: Cascade interrupt Content-Length: 69 Lines: 3 The 8259 is still on Intel's current price list dated 4/96 . Bruce From digitech!digitech.co.nz!arnim@atlantis.actrix.gen.nz Tue May 21 23:26:44 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id XAA29928 for ; Tue, 21 May 1996 23:26:43 +0200 (MET DST) Received: from atlantis.actrix.gen.nz by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA29995; Tue, 21 May 1996 23:26:22 +0200 (MET) Received: (from uucp@localhost) by atlantis.actrix.gen.nz (8.6.11/8.6.9) id JAA15548 for linux-mips@fnet.fr; Wed, 22 May 1996 09:26:33 +1200 >Received: by digitech.co.nz (UUPC/extended 1.12r); Wed, 22 May 1996 09:01:05 +1200 Message-Id: <31a22101.digitech@digitech.co.nz> Date: Wed, 22 May 1996 09:01:01 +1200 From: "Arnim Littek" Organization: Digi-Tech Ltd, Wellington, New Zealand Reply-To: "Arnim Littek" To: linux-mips@fnet.fr Subject: Re: Cascade interrupt Content-Type: text Content-Length: 312 Lines: 9 On Tue, 21 May 1996 18:30:25 +0200 (MET DST), linux-mips@fnet.fr wrote: > Who is Intel's second-source for this chip? If you search for *8259* on Try NEC, Oki, Harris. Definitely not so popular as a standalone part anymore, with integration being the rule among the "avante garde" of the PC set... Arnim. -- From imp@rover.village.org Wed May 22 00:14:58 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id AAA01144 for ; Wed, 22 May 1996 00:14:56 +0200 (MET DST) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA01212; Wed, 22 May 1996 00:14:31 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.7.5/8.6.6) with SMTP id QAA01004 for ; Tue, 21 May 1996 16:14:43 -0600 (MDT) Message-Id: <199605212214.QAA01004@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Cascade interrupt In-Reply-To: Your message of Tue, 21 May 1996 21:24:25 +0200 Date: Tue, 21 May 1996 16:14:43 -0600 From: Warner Losh Content-Length: 404 Lines: 11 : PS: Rule of thumb for MIPS programmers: If it doesn't work blame the : caches ;-) Wonder if that is my floppy drive problem as well :-) :-) :-) Warner P.S. I have my ISA (well EISA) buss mapped in to an address range, but the TLB entry says that it is non-cachable. Would that still cause problems? Or is that data that is passed back a DMA thing and I need to be careful about my caches... From linux@mailhost.uni-koblenz.de Wed May 22 00:32:42 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id AAA01473 for ; Wed, 22 May 1996 00:32:41 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA01585; Wed, 22 May 1996 00:32:18 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.5/8.6.9) with SMTP id AAA04428 for ; Wed, 22 May 1996 00:32:04 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605212232.AAA04428@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id AAA03671; Wed, 22 May 1996 00:31:15 +0200 Subject: Re: Cascade interrupt To: linux-mips@fnet.fr Date: Wed, 22 May 1996 00:31:15 +0200 (MET DST) In-Reply-To: <199605212214.QAA01004@rover.village.org> from "Warner Losh" at May 21, 96 04:14:43 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1015 Lines: 26 Hi, > : PS: Rule of thumb for MIPS programmers: If it doesn't work blame the > : caches ;-) > > Wonder if that is my floppy drive problem as well :-) :-) :-) > > Warner > > P.S. I have my ISA (well EISA) buss mapped in to an address range, > but the TLB entry says that it is non-cachable. Would that still > cause problems? Or is that data that is passed back a DMA thing and I > need to be careful about my caches... The (E)ISA bus should be mapped uncachable; but as you say the floppy data is DMA thing. You need to enshure that you write all data to be written on the floppy back into the RAM. For reading flush the RAM address range to be read from the cache then read the data. It is important that you don't access the address range to be read from the floppy during the read or your cache might then once again contain old data. For the chipset controlled second level caches you don't need to care about flushing; the chipset should guarante consistence between l2 cache and RAM. Ralf From bruce@pixar.com Wed May 22 01:23:33 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id BAA02586 for ; Wed, 22 May 1996 01:23:32 +0200 (MET DST) Received: from mongo.pixar.com by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA02544; Wed, 22 May 1996 01:23:13 +0200 (MET) Received: by mongo.pixar.com (Smail3.1.28.1 #15) id m0uM0m1-000524C; Tue, 21 May 96 16:23 PDT Message-Id: Date: Tue, 21 May 96 16:23 PDT From: bruce@pixar.com (Bruce Perens) To: linux-mips@fnet.fr, linux-mips@fnet.fr Subject: Re: Cascade interrupt Content-Length: 210 Lines: 5 The key turned out to be searching for 82C59, not 8259. OKI has an 82C59A but no on-line doc. Harris will FAX the document to US phone numbers if you call (407) 724-7800 and ask for document # 2784 . Bruce From linux@mailhost.uni-koblenz.de Wed May 22 04:09:50 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id EAA07419 for ; Wed, 22 May 1996 04:09:49 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA04601; Wed, 22 May 1996 04:09:29 +0200 (MET) Received: from ozzy (ozzy.uni-koblenz.de [141.26.5.8]) by informatik.uni-koblenz.de (8.7.5/8.6.9) with SMTP id EAA22026 for ; Wed, 22 May 1996 04:09:15 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605220209.EAA22026@informatik.uni-koblenz.de> Received: by ozzy (SMI-8.6/KO-2.0) id EAA03900; Wed, 22 May 1996 04:08:26 +0200 Subject: Re: Cascade interrupt To: linux-mips@fnet.fr Date: Wed, 22 May 1996 04:08:26 +0200 (MET DST) In-Reply-To: from "Bruce Perens" at May 21, 96 04:23:00 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 366 Lines: 11 Hi, > The key turned out to be searching for 82C59, not 8259. > OKI has an 82C59A but no on-line doc. Harris will FAX the document to US > phone numbers if you call (407) 724-7800 and ask for document # 2784 . Well, in that case I'm definately on the wrong continent :-( Could someone with a faxmodem save these faxes as data files and send them to me? Ralf From munira@Crissy.Stanford.EDU Thu May 23 09:11:22 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id JAA17296 for ; Thu, 23 May 1996 09:11:20 +0200 (MET DST) Received: from crissy.Stanford.EDU by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA04274; Thu, 23 May 1996 09:10:56 +0200 (MET) Received: by Crissy.Stanford.EDU (8.6.8/25-LAGUNITA-eef) id AAA26965; Thu, 23 May 1996 00:11:53 -0700 From: Munira Rahemtulla Message-Id: <199605230711.AAA26965@Crissy.Stanford.EDU> Subject: Re: Kernel patches To: linux-mips@fnet.fr Date: Thu, 23 May 96 0:11:52 PDT In-Reply-To: <199605201840.MAA23176@rover.village.org>; from "Warner Losh" at May 20, 96 12:40 pm X-Mailer: ELM [version 2.3 PL11] Content-Length: 615 Lines: 16 Hi, > However, there is one minor problem with this idea. There are > currently (as of .26 anyway) parts of MILO that cannot be distributed > as source and are binary only (although some of the information in the > binary only code is also in the sources to OpenBSD). It may be tricky > to arrange for things of this nature to be in the tree. Could someone provide us with information about what happens in the parts of milo that cannot be distributed? We have been looking for the definitions of the bootinfo and screeninfo structures because we are trying to simulate what happens in milo. Thanks, Munira From Stoned.Elipot@univ-evry.fr Thu May 23 11:27:48 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id LAA19918 for ; Thu, 23 May 1996 11:27:46 +0200 (MET DST) Received: from amiante.univ-evry.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA07042; Thu, 23 May 1996 11:27:26 +0200 (MET) Received: from bandsept.univ-evry.fr (bandsept.univ-evry.fr [192.93.214.19]) by amiante.univ-evry.fr with ESMTP (8.6.12/96.04.09/louis-blanc); id LAA09653; Thu, 23 May 1996 11:27:43 +0200 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 LAA17773; Thu, 23 May 1996 11:27:37 +0200 Message-Id: <199605230927.LAA17773@bandsept.univ-evry.fr> To: linux-mips@fnet.fr Subject: Re: Kernel patches In-Reply-To: Your message of "Thu, 23 May 1996 00:11:52 PDT." <199605230711.AAA26965@Crissy.Stanford.EDU> X-It: bandsept (wrong OS, but right CPU) X-Mailer: MH 6.8.3, MH-E 5.0.2 Date: Thu, 23 May 1996 11:27:27 +0200 From: Stoned Elipot Content-Length: 1058 Lines: 23 Hi, >>>>> "Munira" == Munira Rahemtulla writes: [SNIP] Munira> Could someone provide us with information about what happens Munira> in the parts of milo that cannot be distributed? We have been Munira> looking for the definitions of the bootinfo and screeninfo Munira> structures because we are trying to simulate what happens in Munira> milo. bootinfo structure was defined in linux/include/asm-mips/bootinfo.h, but since Milo 0.26 is no longer used even defined in kernel's includes, we use what we call tgas instead, have a look in milo's bitags.c and identify.c, and in kernel's arch/mips/setup.c. Fortunately the tags creation process is much more commented in Milo 0.27 which I haven't released yet due to lack of time and lack of sleep :( Sorry for this. And well screeninfo struct is defined in linux/include/tty.h if I recall well, in Milo 0.26 the kernel's image screeninfo struct is still manipulated by Milo before the kernel is launched but this is remplaced by new tags in the coming Milo 0.27. Cheers, Stoned. From ralf@Julia.DE Thu May 23 13:20:08 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id NAA28032 for ; Thu, 23 May 1996 13:20:07 +0200 (MET DST) Received: from alles.intern.julia.de (loehnberg1.core.julia.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA09643; Thu, 23 May 1996 13:19:18 +0200 (MET) Received: from bryant.intern.julia.de (ralf@bryant.intern.julia.de [194.221.49.34]) by alles.intern.julia.de (8.7.5/8.7.3) with ESMTP id NAA05553 for ; Thu, 23 May 1996 13:17:26 +0200 From: Ralf Baechle Received: (from ralf@localhost) by bryant.intern.julia.de (8.7.5/8.7.3) id NAA08216 for linux-mips@fnet.fr; Thu, 23 May 1996 13:14:03 +0200 Message-Id: <199605231114.NAA08216@bryant.intern.julia.de> Subject: Re: Kernel patches To: linux-mips@fnet.fr Date: Thu, 23 May 1996 13:14:03 +0200 (MET DST) In-Reply-To: <199605230711.AAA26965@Crissy.Stanford.EDU> from "Munira Rahemtulla" at May 23, 96 00:11:52 am X-Mailer: ELM [version 2.4 PL24 PGP6] Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 846 Lines: 20 Hi, (so here I'm again reporting from the Linux Kongress in Berlin live and exclusive for the Linux/MIPS mailing list ...) > > However, there is one minor problem with this idea. There are > > currently (as of .26 anyway) parts of MILO that cannot be distributed > > as source and are binary only (although some of the information in the > > binary only code is also in the sources to OpenBSD). It may be tricky > > to arrange for things of this nature to be in the tree. > > Could someone provide us with information about what happens in the > parts of milo that cannot be distributed? We have been looking for > the definitions of the bootinfo and screeninfo structures because we > are trying to simulate what happens in milo. The definitions for most of the bootinformation passed to the kernel is defined in Ralf From paul@suite.sw.oz.au Thu May 23 13:20:16 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id NAA28053 for ; Thu, 23 May 1996 13:20:15 +0200 (MET DST) Received: from staff.cs.su.OZ.AU by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA09647; Thu, 23 May 1996 13:19:34 +0200 (MET) Received: from suite.sw.oz.au by swallow.sw.oz.au with SMTP id LAA25267; Thu, 23 May 1996 11:19:26 GMT (8.6.10/Unixware) (from paul@suite.sw.oz.au for ) Received: by suite.sw.oz.au id AA28979; Thu, 23 May 1996 21:19:24 +1000 (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: <199605231119.AA28979@suite.sw.oz.au> Subject: Re: Kernel patches To: linux-mips@fnet.fr Date: Thu, 23 May 1996 21:19:23 +1000 (EST) In-Reply-To: <199605230927.LAA17773@bandsept.univ-evry.fr> from "Stoned Elipot" at May 23, 96 11:27:27 am Organization: Softway Pty Ltd X-Face: U)Kb/c%d9`JI>Y>s'evmMaR`z*WE$<,["^2)I8o?jkx/d^=dMa$GD(s6[p|npE}r0?)qQ7H *3S2vbK66POj?=%pCn?Bp}D/3A{`-.v0=%L9emS+|Pc91$EP0-;i\ON*p;Mwp$tt3wO~g1Z W.mIz'n~$U&uIM9ix=\1=~""=W?t$3b)>`foC8+3g$m2l;^D$?i}WI(Uxc@THJ9OBbk3g6 X-Mailer: ELM [version 2.4 PL24 PGP3 *ALPHA*] Content-Type: text Content-Length: 1519 Lines: 35 Hi folks, > >>>>> "Munira" == Munira Rahemtulla writes: > [SNIP] > Munira> Could someone provide us with information about what happens > Munira> in the parts of milo that cannot be distributed? We have been > Munira> looking for the definitions of the bootinfo and screeninfo > Munira> structures because we are trying to simulate what happens in > Munira> milo. > > bootinfo structure was defined in linux/include/asm-mips/bootinfo.h, > but since Milo 0.26 is no longer used even defined in kernel's > includes, we use what we call tgas instead, have a look in milo's > bitags.c and identify.c, and in kernel's arch/mips/setup.c. > Fortunately the tags creation process is much more commented in Milo > 0.27 which I haven't released yet due to lack of time and lack of > sleep :( Sorry for this. I have some sample code for the DECstation that creates tags with config info (it even has some comments). You could have that if you want. Now if I could only get Ralf's nice shiny new kernel to compile for the R3000... :-) 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 Thu May 23 13:23:29 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id NAA28309 for ; Thu, 23 May 1996 13:23:28 +0200 (MET DST) Received: from staff.cs.su.OZ.AU by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA09713; Thu, 23 May 1996 13:23:03 +0200 (MET) Received: from suite.sw.oz.au by swallow.sw.oz.au with SMTP id LAA25390; Thu, 23 May 1996 11:23:05 GMT (8.6.10/Unixware) (from paul@suite.sw.oz.au for ) Received: by suite.sw.oz.au id AA29081; Thu, 23 May 1996 21:23:04 +1000 (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: <199605231123.AA29081@suite.sw.oz.au> Subject: Kernels being upside down (was Re: Olivetti console) To: linux-mips@fnet.fr Date: Thu, 23 May 1996 21:23:03 +1000 (EST) In-Reply-To: <199605201100.NAA26535@informatik.uni-koblenz.de> from "Systemkennung Linux" at May 20, 96 12:59:13 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: 1143 Lines: 26 Hi folks, > > YAY! The hacks to vga.c were horrible... and now I can use g364.c as > > a template for the DECStation console! I love it when you guys > > reduce the amount of work to be done.. :-) > > Ha! Don't be happy to early - last night I got 1.3.98 on the SNI machine > not just booting but also useable (undocumented R4600 "features"); so > I made the 1.3.98 snapshot which turns large parts of arch/mips/ upside > down. Have fun with your patches ... I am!! (Californian sarcasm indicator) Just when I'm begining to cope with a kernel, you turn it all upside down again. I think this is just an R4x00-lover's plot to make sure the DECstation kernel never boots! :-) 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 Thu May 23 13:25:06 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id NAA28481 for ; Thu, 23 May 1996 13:25:05 +0200 (MET DST) Received: from staff.cs.su.OZ.AU by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA09726; Thu, 23 May 1996 13:24:39 +0200 (MET) Received: from suite.sw.oz.au by swallow.sw.oz.au with SMTP id LAA25399; Thu, 23 May 1996 11:24:31 GMT (8.6.10/Unixware) (from paul@suite.sw.oz.au for ) Received: by suite.sw.oz.au id AA29106; Thu, 23 May 1996 21:24:29 +1000 (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: <199605231124.AA29106@suite.sw.oz.au> Subject: Re: Kernel patches To: linux-mips@fnet.fr Date: Thu, 23 May 1996 21:24:29 +1000 (EST) In-Reply-To: <199605231114.NAA08216@bryant.intern.julia.de> from "Ralf Baechle" at May 23, 96 01:14:03 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: 718 Lines: 19 Hi Ralf! > (so here I'm again reporting from the Linux Kongress in Berlin live > and exclusive for the Linux/MIPS mailing list ...) Have a beer with Linus for those of us that can't make it. How about suggesting the next Kongress be in Australia??? :-) Have fun! 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 imp@rover.village.org Fri May 24 02:14:25 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id CAA15702 for ; Fri, 24 May 1996 02:14:22 +0200 (MET DST) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA22637; Fri, 24 May 1996 02:13:21 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.7.5/8.6.6) with SMTP id SAA11808 for ; Thu, 23 May 1996 18:13:29 -0600 (MDT) Message-Id: <199605240013.SAA11808@rover.village.org> To: linux-mips@fnet.fr Subject: Re: Kernel patches In-Reply-To: Your message of Thu, 23 May 1996 00:11:52 PDT Date: Thu, 23 May 1996 18:13:28 -0600 From: Warner Losh Content-Length: 443 Lines: 10 : Could someone provide us with information about what happens in the : parts of milo that cannot be distributed? Basically the interface to the ARC BIOS that MILO calls is what in the files that can't be distributed. They were written with information that is copyright by Microsoft that has all kinds of scary DO NOT DISTRIBUTE warnings on them. It should have no effect on simulating the envirnment after MILO loads the kernel. Warner From nat@nataa.frmug.fr.net Tue May 28 10:27:21 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id KAA26032 for ; Tue, 28 May 1996 10:27:20 +0200 (MET DST) Received: from gatekeeper.waldorf-gmbh.de by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA28723; Tue, 28 May 1996 10:26:44 +0200 (MET) Received: from itesec.hsc.fr (root@itesec.hsc.fr [192.70.106.33]) by gatekeeper.waldorf-gmbh.de (8.6.12/8.6.9) with ESMTP id KAA01158 for ; Tue, 28 May 1996 10:25:16 +0200 Received: from frmug.fr.net (frmug.hsc.fr [192.70.106.65]) by itesec.hsc.fr (8.7.5/8.7.3/itesec-1.8) with ESMTP id KAA06101 for ; Tue, 28 May 1996 10:18:56 +0200 (MET DST) Received: (from uucp@localhost) by frmug.fr.net (8.6.8/8.6.9) with UUCP id KAA01203 for linux@waldorf-gmbh.de; Tue, 28 May 1996 10:18:55 +0200 Received: (from nat@localhost) by nataa.frmug.fr.net (8.6.12/8.6.12) id UAA01654; Mon, 27 May 1996 20:48:32 +0200 To: linux@waldorf-gmbh.de Subject: [comp.sys.mips,comp.os.linux.misc,comp.os.linux.development.system] Looking for Linux hacker for MIPS 4300 Linux port From: Nat Makarevitch Date: 27 May 1996 20:48:30 +0200 Message-Id: X-Mailer: September Gnus v0.89/XEmacs 19.13 Content-Length: 1053 Lines: 23 just in case you missed it ... sorry if it's useless ------- Start of forwarded message ------- From: eugen@research.nj.nec.com (Eugen Schenfeld) Newsgroups: comp.sys.mips,comp.os.linux.misc,comp.os.linux.development.system Subject: Looking for Linux hacker for MIPS 4300 Linux port Date: 3 May 1996 16:28:29 -0400 Organization: NEC Research Institute, 4 Independence Way, Princeton, NJ Message-ID: <4mdq9d$9a3@iris49.nj.nec.com> A system programmer is needed for a Linux/MIPS port to a Massivelly Parallel Processing new architecture. The intention is to create a MP MIPS version based on the MIPS 4300 CPUs, and other aspects of a distribited shared memory architecture. The work location is at the NEC Research Institute, in Princeton, NJ USA, for one year. A good Ph.D. student (hacker) that would like to combine this work as part of his/hers Ph.D. thesis is also encouraged to consider. For more info. please contact Dr. Eugen Schenfeld at eugen@research.nj.nec.com or (609)951-2742. ------- End of forwarded message ------- -- Nat Linux From linux@mailhost.uni-koblenz.de Wed May 29 03:14:01 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id DAA17571 for ; Wed, 29 May 1996 03:14:00 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA17685; Wed, 29 May 1996 03:13:28 +0200 (MET) Received: from thoma (thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.7.5/8.6.9) with SMTP id DAA10597; Wed, 29 May 1996 03:18:41 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605290118.DAA10597@informatik.uni-koblenz.de> Received: by thoma (SMI-8.6/KO-2.0) id DAA10495; Wed, 29 May 1996 03:13:03 +0200 Subject: Re: [comp.sys.mips,comp.os.linux.misc,comp.os.linux.development.system] Looking for Linux hacker for MIPS 4300 Linux port To: linux-mips@fnet.fr, nat@nataa.frmug.fr.net Date: Wed, 29 May 1996 03:13:03 +0200 (MET DST) In-Reply-To: from "Nat Makarevitch" at May 27, 96 08:48:30 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 748 Lines: 20 Hi, > just in case you missed it ... sorry if it's useless > > ------- Start of forwarded message ------- > From: eugen@research.nj.nec.com (Eugen Schenfeld) > Newsgroups: comp.sys.mips,comp.os.linux.misc,comp.os.linux.development.system > Subject: Looking for Linux hacker for MIPS 4300 Linux port > Date: 3 May 1996 16:28:29 -0400 > Organization: NEC Research Institute, 4 Independence Way, Princeton, NJ > Message-ID: <4mdq9d$9a3@iris49.nj.nec.com> > > A system programmer is needed for a Linux/MIPS port to a Massivelly Parallel > Processing new architecture. The intention is to create a MP MIPS version based And now guess who made the NEC guy post this :-) Ralf PS: The mailing list address has changed; it's now linux-mips@fnet.fr. From linux@mailhost.uni-koblenz.de Wed May 29 03:19:23 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id DAA17708 for ; Wed, 29 May 1996 03:19:22 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA17796; Wed, 29 May 1996 03:18:51 +0200 (MET) Received: from thoma (thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.7.5/8.6.9) with SMTP id DAA10746 for ; Wed, 29 May 1996 03:24:20 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605290124.DAA10746@informatik.uni-koblenz.de> Received: by thoma (SMI-8.6/KO-2.0) id DAA10509; Wed, 29 May 1996 03:18:41 +0200 Subject: SNI & rPC Floopy To: linux-mips@fnet.fr Date: Wed, 29 May 1996 03:18:41 +0200 (MET DST) Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 436 Lines: 12 Hi all, last night I tortured the floppy code for the SNI a bit and found it to be working perfectly. It's exactly the same code that is also being used for Warner's rPC44 with the exception that it fails on the rPC. So I've now profen this code to work on a machine equvalent to the model that I have in mind of the rPC. In other words: this must be a Deskstation specific bug. Sorry Warner, this is not good news for you, Ralf From pasec@can.nl Wed May 29 15:58:47 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id PAA03665 for ; Wed, 29 May 1996 15:58:41 +0200 (MET DST) Received: from cand.can.nl by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA02588; Wed, 29 May 1996 15:58:10 +0200 (MET) Received: by cand.can.nl with SMTP; Wed, 29 May 1996 15:58:39 +0200 (MET) Sender: pasec@can.nl Message-Id: <31AC57A1.41C6@can.nl> Date: Wed, 29 May 1996 15:56:49 +0200 From: Dima Pasechnik X-Mailer: Mozilla 2.01 (X11; I; IRIX 5.2 IP22) Mime-Version: 1.0 To: linux-mips@fnet.fr Cc: dima@can.nl Subject: linux on indy R4600 X-Url: http://lena.fnet.fr/linux-mips-howto.html Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Length: 560 Lines: 21 Dear Sir or Madam, Excuse my ignorance, I wonder if Linux-mips could run on an Indy R4600 (a Silicon Graphics workstation). The processor seem to be one of the supported by you. The problem to try it for us is that not a single workstation has a floppy-drive, so I cannot follow your HOWTO. Thanks very much for your attention, Dima Dmitrii V. Pasechnik RIACA Department of Mathematics and Informatics Eindhoven University of Technology PO Box 513 5600 MB Eindhoven The Netherlands email: pasec@can.nl dima@win.tue.nl www: http://www.can.nl/~pasec From imp@rover.village.org Wed May 29 16:07:43 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id QAA04048 for ; Wed, 29 May 1996 16:07:40 +0200 (MET DST) Received: from rover.village.org by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA02757; Wed, 29 May 1996 16:07:04 +0200 (MET) Received: from localhost (localhost [127.0.0.1]) by rover.village.org (8.7.5/8.6.6) with SMTP id IAA12999 for ; Wed, 29 May 1996 08:07:18 -0600 (MDT) Message-Id: <199605291407.IAA12999@rover.village.org> To: linux-mips@fnet.fr Subject: Re: SNI & rPC Floopy In-Reply-To: Your message of Wed, 29 May 1996 03:18:41 +0200 Date: Wed, 29 May 1996 08:07:18 -0600 From: Warner Losh Content-Length: 1224 Lines: 28 : last night I tortured the floppy code for the SNI a bit and found it : to be working perfectly. It's exactly the same code that is also being : used for Warner's rPC44 with the exception that it fails on the rPC. : So I've now profen this code to work on a machine equvalent to the model : that I have in mind of the rPC. In other words: this must be a : Deskstation specific bug. Bummer. : Sorry Warner, this is not good news for you, Yes. The more I hear you talk about cache problems, the more I start to think that the floppy problem may be a cache problem. I've rerun the seek tests and it could be seeking, or not. There is a lot of background noise in the room it is in. I'll have to shake lose about 4-8 hours in a row to look into it. Does it use exactly the same glue code as the rPC44, or just the same floppy driver? Maybe one of the DMA calls is boogered up on the rPC version. I'll have to take a look at that when I get a free minute. I've been swamped with house guests and spring cleaning (and a cool "new" used radial arm saw) for the past two weeks or so, so I haven't done much on the rPC. BTW, when will the SNI code be released, or is it in the massive patch I've not grabbed? Warner From linux@mailhost.uni-koblenz.de Wed May 29 17:11:22 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id RAA05351 for ; Wed, 29 May 1996 17:11:21 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA04312; Wed, 29 May 1996 17:10:48 +0200 (MET) Received: from thoma (thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.7.5/8.6.9) with SMTP id RAA25562 for ; Wed, 29 May 1996 17:17:08 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605291517.RAA25562@informatik.uni-koblenz.de> Received: by thoma (SMI-8.6/KO-2.0) id RAA14919; Wed, 29 May 1996 17:10:37 +0200 Subject: Re: SNI & rPC Floopy To: linux-mips@fnet.fr Date: Wed, 29 May 1996 17:10:37 +0200 (MET DST) In-Reply-To: <199605291407.IAA12999@rover.village.org> from "Warner Losh" at May 29, 96 08:07:18 am Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 2103 Lines: 52 Hi, > Yes. The more I hear you talk about cache problems, the more I start > to think that the floppy problem may be a cache problem. I've rerun > the seek tests and it could be seeking, or not. There is a lot of > background noise in the room it is in. I'll have to shake lose about > 4-8 hours in a row to look into it. You might open the floppy drive and look at the heads directly. With most of the floppies I know this is no big mechanical probblem. > Does it use exactly the same glue code as the rPC44, or just the same > floppy driver? Both should be identical down to the bit. Maybe one of the DMA calls is boogered up on the rPC > version. I wouldn't wonder if the Deskstation guys modified the DMA logik of the board in some difficile way ... I'll have to talk to them again also. I'll have to take a look at that when I get a free minute. > I've been swamped with house guests and spring cleaning (and a cool > "new" used radial arm saw) for the past two weeks or so, Doom ?-) so I haven't > done much on the rPC. > > BTW, when will the SNI code be released, or is it in the massive patch > I've not grabbed? The SNI stuff is in 1.3.98. I'll be doing a little hack session the next night and upgrade to 1.99.8 aka pre2.0.8. 1.99.7 is already running :-) The SNI looks enough like a "normal" Linux machine so that Drew attempted in Berlin to copy some binaries on it. If not a permission denied had stopped him a "wrong architecture" would have done :-) He's not the only guy who got fooled by the machine; like the other machines the SNI was abused as stupid telnet terminal for the most of the time ... I'll have to excuse once again for not doing all the things that I've promised on the mailinglist and some people in private. I recently had a car accident and 'till I've can replace it (Send me all your cars $-) public traffic/biking costs me some extra hours each day. Anyway, email get's outdated as it is being written - 1.99.8 is running on the SNI now ... Ralf From Stoned.Elipot@univ-evry.fr Wed May 29 17:33:27 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id RAA05759 for ; Wed, 29 May 1996 17:33:26 +0200 (MET DST) Received: from amiante.univ-evry.fr by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA04844; Wed, 29 May 1996 17:32:55 +0200 (MET) Received: from bandsept.univ-evry.fr (bandsept.univ-evry.fr [192.93.214.19]) by amiante.univ-evry.fr with ESMTP (8.6.12/96.04.09/louis-blanc); id RAA26865; Wed, 29 May 1996 17:33:23 +0200 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 RAA07905; Wed, 29 May 1996 17:33:17 +0200 Message-Id: <199605291533.RAA07905@bandsept.univ-evry.fr> To: linux-mips@fnet.fr Subject: Re: SNI & rPC Floopy In-Reply-To: Your message of "Wed, 29 May 1996 17:10:37 +0200." <199605291517.RAA25562@informatik.uni-koblenz.de> X-It: bandsept (wrong OS, but right CPU) X-Mailer: MH 6.8.3, MH-E 5.0.2 Date: Wed, 29 May 1996 17:33:11 +0200 From: Stoned Elipot Content-Length: 665 Lines: 15 Hi, >>>>> "Ralf" == Systemkennung Linux writes: [SNIP] Ralf> The SNI stuff is in 1.3.98. I'll be doing a little hack session Ralf> the next night and upgrade to 1.99.8 aka pre2.0.8. 1.99.7 is Ralf> already running :-) Ralf, I catch you here as I appears you don't have seen the last email I sent you (can't remember at which address I'd sent it)... I got a bunch of patch for you against 1.3.97, and Milo is about to be out but I'd like you to see this email I had sent (at waldorf I think) before I made it available, and please ............ wait till I send my kernel patches before realeasing a new kernel :) Cheers, Stoned. From linux@mailhost.uni-koblenz.de Wed May 29 20:24:29 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id UAA09322 for ; Wed, 29 May 1996 20:24:28 +0200 (MET DST) Received: from informatik.uni-koblenz.de (mailhost.uni-koblenz.de) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA08750; Wed, 29 May 1996 20:23:55 +0200 (MET) Received: from thoma (thoma.uni-koblenz.de [141.26.4.61]) by informatik.uni-koblenz.de (8.7.5/8.6.9) with SMTP id UAA02759 for ; Wed, 29 May 1996 20:30:30 +0200 (MET DST) From: Systemkennung Linux Message-Id: <199605291830.UAA02759@informatik.uni-koblenz.de> Received: by thoma (SMI-8.6/KO-2.0) id UAA15286; Wed, 29 May 1996 20:23:45 +0200 Subject: Re: SNI & rPC Floopy To: linux-mips@fnet.fr Date: Wed, 29 May 1996 20:23:44 +0200 (MET DST) In-Reply-To: <199605291533.RAA07905@bandsept.univ-evry.fr> from "Stoned Elipot" at May 29, 96 05:33:11 pm Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1022 Lines: 27 Hi, > >>>>> "Ralf" == Systemkennung Linux writes: > [SNIP] > Ralf> The SNI stuff is in 1.3.98. I'll be doing a little hack session > Ralf> the next night and upgrade to 1.99.8 aka pre2.0.8. 1.99.7 is > Ralf> already running :-) > > Ralf, I catch you here as I appears you don't have seen the last email > I sent you (can't remember at which address I'd sent it)... > I got a bunch of patch for you against 1.3.97, and Milo is about to be > out but I'd like you to see this email I had sent (at waldorf I think) > before I made it available, and please ............ wait till I send > my kernel patches before realeasing a new kernel :) Well, the problem with the account in Waldorf is that the line to there is usually worse than the Ebone, so sending me something there might result in long times 'till I answer it. Here there more addresses from my collection: ralf@uni-koblenz linux@uni-koblenz.de ralf@julia.de Of course I can't expect to get your patches :-) Ralf From engel@vespa.unix-ag.uni-siegen.de Fri May 31 16:40:00 1996 Received: from ns.fnet.fr (ns.fnet.fr [192.134.192.2]) by guadalquivir.fnet.fr (8.7.5/8.7.3) with SMTP id QAA04267 for ; Fri, 31 May 1996 16:39:48 +0200 (MET DST) Received: from si-nic.hrz.uni-siegen.de ([141.99.128.16]) by ns.fnet.fr (5.65c8d/AFUU-4.2.3) via EUnet-France id AA05156; Fri, 31 May 1996 16:39:14 +0200 (MET) Received: from vespa.unix-ag.uni-siegen.de by si-nic.hrz.uni-siegen.de with SMTP (5.67b/UniSiegen 1.1) id AA04949; Fri, 31 May 1996 16:39:12 +0200 Received: by vespa.unix-ag.uni-siegen.de (8.7.5/8.7.1) id QAA09837 for linux-mips@fnet.fr; Fri, 31 May 1996 16:36:48 +0200 (MET DST) From: Michael Engel Message-Id: <199605311436.QAA09837@vespa.unix-ag.uni-siegen.de> Subject: DS5000/120 To: linux-mips@fnet.fr Date: Fri, 31 May 1996 16:36:47 +0200 (MET DST) X-Mailer: ELM [version 2.4 PL23] Content-Type: text Content-Length: 626 Lines: 14 Hello to all you Linux/MIPS hackers, a friend of mine and I recently acquired two DecStations 5000/120 in working condition (for free :)). After stealing 16 MB out of a dead DS3100, both machines work fine with a terminal connected to serial port 2 as console. I know from the NetBSD sources that there exist several documents from DEC describing the 5000/1xx hardware, but I wasn't able to find those on the net (only the DS3100 and 5000/200 docs on gatekeeper). Does anyone know where we might get a version (either printed or Postscript/PDF) of these documents ? regards, Michael Engel (engel@unix-ag.uni-siegen.de)