From daemon Mon May 5 14:10:59 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id OAA03769 for vsta-xplod; Mon, 5 May 1997 14:10:59 GMT Received: from baruk2.zendo.com (baruk2.zendo.com [205.187.71.65]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id OAA03763 for ; Mon, 5 May 1997 14:10:57 GMT Received: (from root@localhost) by baruk2.zendo.com (8.8.5/8.6.12) id QAA11742; Mon, 5 May 1997 16:32:43 -0700 (PDT) Date: Mon, 5 May 1997 16:32:43 -0700 (PDT) To: vsta@zendo.com From: s2156593@cse.unsw.edu.au (Binh Thai) Message-Id: <970505225653.3325@cse.unsw.edu.au> Subject: Information on interrupt handling Hi everyone, I'm currently trying write a task which triggers an interrupt from an external event, and I need some information on how vsta handle interrupts in this kind of manner. Does anyone know where I can obtain such information? Thank you. From daemon Fri May 9 06:12:31 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id GAA08894 for vsta-xplod; Fri, 9 May 1997 06:12:31 GMT Received: (from root@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id GAA08887; Fri, 9 May 1997 06:12:28 GMT Date: Fri, 9 May 1997 06:12:28 GMT Message-Id: <199705090612.GAA08887@bodhi.zendo.com> From: "Bart Sekura (WAW-P)" To: "'vsta@zendo.com'" Subject: gunzippin' problems Hey, I'm having problems gunzipping vsta.tz file. I downloaded it couple of times, and every time I try to gzip -d it, it says gzip format violated. Please let me know if I'm the only one having problems meaning there's something wrong with my downloading. Thanks Bart Sekura bsekura@impaq.com.pl From daemon Fri May 9 07:31:19 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id HAA08989 for vsta-xplod; Fri, 9 May 1997 07:31:19 GMT Received: from puli.cisco.com (puli.cisco.com [171.69.1.174]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id HAA08983 for ; Fri, 9 May 1997 07:31:18 GMT Received: from localhost.cisco.com (localhost.cisco.com [127.0.0.1]) by puli.cisco.com (8.6.12/8.6.5) with SMTP id JAA25093; Fri, 9 May 1997 09:48:29 -0700 Message-Id: <199705091648.JAA25093@puli.cisco.com> To: "Bart Sekura (WAW-P)" cc: "'vsta@zendo.com'" Subject: Re: gunzippin' problems In-reply-to: Your message of "Fri, 09 May 1997 06:12:28 GMT." <199705090612.GAA08887@bodhi.zendo.com> Date: Fri, 09 May 1997 09:48:28 -0700 From: Andrew Valencia ["Bart Sekura (WAW-P)" writes:] >I'm having problems gunzipping vsta.tz file. I downloaded it couple >of times, and every time I try to gzip -d it, it says gzip format >violated. Please let me know if I'm the only one having problems >meaning there's something wrong with my downloading. I just pulled a copy from ftp.zendo.com, and gzip -tv tells me it's fine. FWIW, Andy From daemon Thu May 15 12:59:14 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id MAA19526 for vsta-xplod; Thu, 15 May 1997 12:59:14 GMT Received: from apollo.is.co.za (apollo.is.co.za [196.4.160.2]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id MAA19520 for ; Thu, 15 May 1997 12:59:07 GMT Received: (from datrix@localhost) by apollo.is.co.za (8.8.5/8.7.5/IShub#2) id AAA17137; Fri, 16 May 1997 00:20:17 +0200 (GMT) Date: Fri, 16 May 1997 00:20:17 +0200 (GMT) From: Datrix Solutions To: vsta@zendo.com Subject: Networking revisited. Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hi, I have tried tonight to get VSTa networking going :-( still no success. 1) When I start the ne server with ne 340 11 it report the MAC address, but with IRQ 3 (11 under Linux). It does this for all ne 340 XX. 2) I also get a hang on attach eth eth0 1500 after having mounted ne on /dev/eth. (after net has run and detached how do I get back in to do a telnet session...? I'd like to get it working before reading the source ;-) regards, Wayne From daemon Sun May 18 07:12:59 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id HAA25310 for vsta-xplod; Sun, 18 May 1997 07:12:59 GMT Received: from apollo.is.co.za (apollo.is.co.za [196.4.160.2]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id HAA25304 for ; Sun, 18 May 1997 07:12:52 GMT Received: (from datrix@localhost) by apollo.is.co.za (8.8.5/8.7.5/IShub#2) id SAA07683; Sun, 18 May 1997 18:34:04 +0200 (GMT) Date: Sun, 18 May 1997 18:34:04 +0200 (GMT) From: Datrix Solutions To: vsta@zendo.com Subject: Status of Mailing List Archives. Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hi, I am unable to connect to the VSTa Mailing list archives at: http://ftoomsh.socs.uts.EDU.AU/mlists/vsta Is this temporary? Is there another server available? cheers, Wayne From daemon Mon May 19 07:51:16 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id HAA26598 for vsta-xplod; Mon, 19 May 1997 07:51:16 GMT Received: from terra.igcom.net (jeske@terra.igcom.net [206.98.13.66]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id HAA26592 for ; Mon, 19 May 1997 07:51:07 GMT Received: (from jeske@localhost) by terra.igcom.net (8.8.5/8.8.5) id MAA07338; Mon, 19 May 1997 12:12:24 -0500 (CDT) Message-ID: Date: Mon, 19 May 1997 12:12:23 -0500 From: jeske@igcom.net (David Jeske) To: vsta@zendo.com Subject: Re: Status of Mailing List Archives. References: X-Mailer: Mutt 0.56e Mime-Version: 1.0 X-Uri: In-Reply-To: ; from Datrix Solutions on May 18, 1997 18:34:04 +0200 On May 18, datrix@is.co.za (Datrix Solutions) wrote: > Hi, > > I am unable to connect to the VSTa Mailing list archives at: > > http://ftoomsh.socs.uts.EDU.AU/mlists/vsta > > Is this temporary? Is there another server available? That dropped out quite a while ago as far as I know. If we need a new one setup, I can probably get one up. -- David Jeske (N9LCA) + jeske@igcom.net + http://www.igcom.net/~jeske/ From daemon Sat May 24 06:33:26 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id GAA05294 for vsta-xplod; Sat, 24 May 1997 06:33:26 GMT Received: from puli.cisco.com (puli.cisco.com [171.69.1.174]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id GAA05288 for ; Sat, 24 May 1997 06:33:21 GMT Received: from localhost.cisco.com (localhost.cisco.com [127.0.0.1]) by puli.cisco.com (8.6.12/8.6.5) with SMTP id IAA29447; Sat, 24 May 1997 08:49:22 -0700 Message-Id: <199705241549.IAA29447@puli.cisco.com> To: Datrix Solutions cc: vsta@zendo.com Subject: Re: Networking revisited. In-reply-to: Your message of "Fri, 16 May 1997 00:20:17 +0200." Date: Sat, 24 May 1997 08:49:22 -0700 From: Andrew Valencia [Datrix Solutions writes:] >1) When I start the ne server with ne 340 11 it report the >MAC address, but with IRQ 3 (11 under Linux). >It does this for all ne 340 XX. I think the syntax is 0x340,11 (comma, sigh). >2) I also get a hang on attach eth eth0 1500 after having >mounted ne on /dev/eth. This makes sense. The driver's probably waiting for interrupt I/O completions which never happen. >(after net has run and detached how do I get back in to >do a telnet session...? There's a hack telnet function under cmds/inetdb.c. Once you detach, telnet'ing back into the box gets you talking again to the networking stack. >I'd like to get it working before reading the source ;-) I'd have to say that the networking portion of VSTa is even less amenable to this than usual. :-) Andy From daemon Tue May 27 06:45:14 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id GAA09191 for vsta-xplod; Tue, 27 May 1997 06:45:14 GMT Received: from puli.cisco.com (puli.cisco.com [171.69.1.174]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id GAA09185 for ; Tue, 27 May 1997 06:45:12 GMT Received: from localhost.cisco.com (localhost.cisco.com [127.0.0.1]) by puli.cisco.com (8.6.12/8.6.5) with SMTP id JAA19400; Tue, 27 May 1997 09:05:58 -0700 Message-Id: <199705271605.JAA19400@puli.cisco.com> To: jeske@igcom.net (David Jeske) cc: vsta@zendo.com Subject: Re: Status of Mailing List Archives. In-reply-to: Your message of "Mon, 19 May 1997 12:12:23 CDT." Date: Tue, 27 May 1997 09:05:58 -0700 From: Andrew Valencia [jeske@igcom.net (David Jeske) writes:] >That dropped out quite a while ago as far as I know. If we need a new >one setup, I can probably get one up. That would be nice. I have archives of all the messages, and have been looking at the mail2html perl script. But if you had something nicer, it'd be welcomed. Andy From daemon Tue May 27 12:24:20 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id MAA09690 for vsta-xplod; Tue, 27 May 1997 12:24:20 GMT Received: from puli.cisco.com (puli.cisco.com [171.69.1.174]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id MAA09683 for ; Tue, 27 May 1997 12:24:17 GMT Received: from localhost.cisco.com (localhost.cisco.com [127.0.0.1]) by puli.cisco.com (8.6.12/8.6.5) with SMTP id OAA27897 for ; Tue, 27 May 1997 14:45:36 -0700 Message-Id: <199705272145.OAA27897@puli.cisco.com> To: vsta@zendo.com Subject: VSTa mail archives back on-line Date: Tue, 27 May 1997 14:45:36 -0700 From: Andrew Valencia Thanks to Rob Savoye of Cygnus for pointing me at a tool which makes this a breeze. The VSTa mail archives are now available in their HTML form from: http://www.zendo.com/vandys/vsta-mail I will tool up the automatic archive update from cron so the index will be updated on a nightly basis. On a completely unrelated note, I today put in an order for a dual processor Supermicro P6DOF motherboard with two 200 Mhz PPro 512k cache CPU's. Expect VSTa SMP to be up Real Soon Now. :-) Andy From daemon Fri May 30 19:34:54 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id TAA15119 for vsta-xplod; Fri, 30 May 1997 19:34:54 GMT Received: from puli.cisco.com (puli.cisco.com [171.69.1.174]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id TAA15111 for ; Fri, 30 May 1997 19:34:52 GMT Received: from localhost.cisco.com (localhost.cisco.com [127.0.0.1]) by puli.cisco.com (8.6.12/8.6.5) with SMTP id VAA29652; Fri, 30 May 1997 21:43:05 -0700 Message-Id: <199705310443.VAA29652@puli.cisco.com> To: eld@jewel.ucsd.edu (Eric Dorman) cc: vsta@zendo.com Subject: Re: Where is fsck_vfs, fsdb_vfs? In-reply-to: Your message of "Fri, 30 May 1997 16:25:09 PDT." <199705302325.QAA03558@jewel.ucsd.edu> Date: Fri, 30 May 1997 21:43:05 -0700 From: Andrew Valencia [eld@jewel.ucsd.edu (Eric Dorman) writes:] >I've been unable to find sources for fsck_vfs >and fsdb_vfs. Do these exist somewhere? Yes, see the cmd/ subdir of src/srv/vstafs. They're renamed from "fs*" to "fs*_vfs" during "make install". Andy From daemon Tue Jun 17 16:25:17 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id QAA20256 for vsta-xplod; Tue, 17 Jun 1997 16:25:17 GMT Received: from network.ucsd.edu (root@network.ucsd.edu [132.239.1.20]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id QAA20250 for ; Tue, 17 Jun 1997 16:25:06 GMT Received: from Tanya (tanya.ucsd.edu [132.239.141.90]) by network.ucsd.edu (8.8.5/8.6.9) with SMTP id SAA01587 for <@network.ucsd.edu:vsta@zendo.com>; Tue, 17 Jun 1997 18:47:54 -0700 (PDT) Received: by Tanya (950413.SGI.8.6.12/940406.SGI.AUTO) for vsta@zendo.com id SAA19897; Tue, 17 Jun 1997 18:55:36 -0700 Date: Tue, 17 Jun 1997 18:55:36 -0700 From: edorman@Tanya.ucsd.edu (Eric Dorman) Message-Id: <199706180155.SAA19897@Tanya> To: vsta@zendo.com Subject: test test From daemon Thu Jun 19 16:48:28 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id QAA23580 for vsta-xplod; Thu, 19 Jun 1997 16:48:28 GMT Received: from network.ucsd.edu (root@network.ucsd.edu [132.239.1.20]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id QAA23574 for ; Thu, 19 Jun 1997 16:48:25 GMT Received: from Tanya (tanya.ucsd.edu [132.239.141.90]) by network.ucsd.edu (8.8.5/8.6.9) with SMTP id TAA16579 for <@network.ucsd.edu:vsta@bodhi.zendo.com>; Thu, 19 Jun 1997 19:11:00 -0700 (PDT) Received: by Tanya (950413.SGI.8.6.12/940406.SGI.AUTO) for vsta@baruk.zendo.com id TAA29938; Thu, 19 Jun 1997 19:19:00 -0700 Date: Thu, 19 Jun 1997 19:19:00 -0700 From: edorman@Tanya.ucsd.edu (Eric Dorman) Message-Id: <199706200219.TAA29938@Tanya> To: vsta@bodhi.zendo.com Subject: VSTa homepage Hello; it appears that http://www.igcom.net/~jeske is dead.. is this an igcom.net problem, or are the pages being maintained somewhere else (or at all?) Thanks, Eric Dorman Department of Radiology University of California at San Diego edorman@tanya.ucsd.edu From daemon Thu Jun 19 17:08:21 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id RAA23604 for vsta-xplod; Thu, 19 Jun 1997 17:08:21 GMT Received: from puli.cisco.com (puli.cisco.com [171.69.1.174]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id RAA23598 for ; Thu, 19 Jun 1997 17:08:19 GMT Received: from localhost.cisco.com (localhost.cisco.com [127.0.0.1]) by puli.cisco.com (8.6.12/8.6.5) with SMTP id TAA13597; Thu, 19 Jun 1997 19:30:03 -0700 Message-Id: <199706200230.TAA13597@puli.cisco.com> To: edorman@Tanya.ucsd.edu (Eric Dorman) cc: vsta@zendo.com Subject: Re: VSTa homepage In-reply-to: Your message of "Thu, 19 Jun 1997 19:19:00 PDT." <199706200219.TAA29938@Tanya> Date: Thu, 19 Jun 1997 19:30:03 -0700 From: Andrew Valencia [edorman@Tanya.ucsd.edu (Eric Dorman) writes:] > it appears that http://www.igcom.net/~jeske is dead.. is this an igcom.net >problem, or are the pages being maintained somewhere else (or at all?) I've sent a note to David Jeske, who previously hosted them. No answer back (yet). I rather wish I had taken a snapshot of it all (hindsight). Andy From daemon Thu Jun 19 17:17:00 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id RAA23625 for vsta-xplod; Thu, 19 Jun 1997 17:17:00 GMT Received: from network.ucsd.edu (root@network.ucsd.edu [132.239.1.20]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id RAA23619 for ; Thu, 19 Jun 1997 17:16:57 GMT Received: from Tanya (tanya.ucsd.edu [132.239.141.90]) by network.ucsd.edu (8.8.5/8.6.9) with SMTP id TAA16673; Thu, 19 Jun 1997 19:38:29 -0700 (PDT) Received: by Tanya (950413.SGI.8.6.12/940406.SGI.AUTO) id TAA00030; Thu, 19 Jun 1997 19:46:12 -0700 Date: Thu, 19 Jun 1997 19:46:12 -0700 From: edorman@Tanya.ucsd.edu (Eric Dorman) Message-Id: <199706200246.TAA00030@Tanya> To: Andrew Valencia Subject: Re: VSTa homepage Cc: vsta@zendo.com Well, I've got most of it printed out; not that much really, so if it's vanished I've got hardcopy floating around. regards, eld From daemon Thu Jun 19 17:47:45 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id RAA23691 for vsta-xplod; Thu, 19 Jun 1997 17:47:45 GMT Received: from cypher.net (black@zen.pratt.edu [205.232.115.155]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id RAA23685 for ; Thu, 19 Jun 1997 17:47:42 GMT Received: (from black@localhost) by cypher.net (8.8.5/8.7.1) id XAA02010; Thu, 19 Jun 1997 23:11:26 -0400 Date: Thu, 19 Jun 1997 23:11:25 -0400 (EDT) From: Ben Black To: Eric Dorman cc: Andrew Valencia , vsta@zendo.com Subject: Re: VSTa homepage In-Reply-To: <199706200246.TAA00030@Tanya> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII i would be more than happy to home it at my office (multihomed, even!)... On Thu, 19 Jun 1997, Eric Dorman wrote: > Well, I've got most of it printed out; not that much really, so if > it's vanished I've got hardcopy floating around. > > regards, > > eld > From daemon Sat Jun 21 11:14:17 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id LAA28365 for vsta-xplod; Sat, 21 Jun 1997 11:14:17 GMT Received: from baruk2.zendo.com (baruk2.zendo.com [205.187.71.65]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id LAA28358 for ; Sat, 21 Jun 1997 11:14:13 GMT Received: from localhost.zendo.com (localhost.zendo.com [127.0.0.1]) by baruk2.zendo.com (8.8.5/8.6.12) with SMTP id NAA28579 for ; Sat, 21 Jun 1997 13:41:03 -0700 (PDT) Message-Id: <199706212041.NAA28579@baruk2.zendo.com> X-Authentication-Warning: baruk2.zendo.com: localhost.zendo.com [127.0.0.1] didn't use HELO protocol To: vsta@zendo.com Subject: VSTa WWW pages re-homed Date: Sat, 21 Jun 1997 13:41:02 -0700 From: Andy Valencia Thanks to Abdallah Chatila, who had archived the pages, the VSTa WWW pages are now available again at: http://www.zendo.com/vsta I've fixed up the mailing archive change, as well as some of the relative links. Please send me private E-mail if you find further rough edges. Andy Valencia From daemon Sat Jun 28 15:40:15 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id PAA10587 for vsta-xplod; Sat, 28 Jun 1997 15:40:15 GMT Received: from deathstar.ml.org (adrian@deathstar.ml.org [203.62.152.4]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id PAA10581 for ; Sat, 28 Jun 1997 15:40:09 GMT Received: from localhost (adrian@localhost) by deathstar.ml.org (8.8.5/8.7.3) with SMTP id IAA15052 for ; Sun, 29 Jun 1997 08:37:37 +0800 Date: Sun, 29 Jun 1997 08:37:37 +0800 (WST) From: Adrian Chadd To: vsta@zendo.com Subject: Swap space going under 1.5.2 .. Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hi. I have VSTa 1.5.2 going on my 386sx25 w/ 2mb RAM however I'd like to get the swapping going. I have created a second partition on the HDD, what else do I need to do to get it enabled? -- Adrian Chadd | "Unix doesn't stop you from doing | stupid things because that would | stop you from doing clever things" From daemon Sun Jun 29 06:44:28 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id GAA11520 for vsta-xplod; Sun, 29 Jun 1997 06:44:28 GMT Received: from puli.cisco.com (puli.cisco.com [171.69.1.174]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id GAA11514 for ; Sun, 29 Jun 1997 06:44:25 GMT Received: from localhost.cisco.com (localhost.cisco.com [127.0.0.1]) by puli.cisco.com (8.6.12/8.6.5) with SMTP id JAA13080; Sun, 29 Jun 1997 09:07:07 -0700 Message-Id: <199706291607.JAA13080@puli.cisco.com> To: Adrian Chadd cc: vsta@zendo.com Subject: Re: Swap space going under 1.5.2 .. In-reply-to: Your message of "Sun, 29 Jun 1997 08:37:37 +0800." Date: Sun, 29 Jun 1997 09:07:06 -0700 From: Andrew Valencia [Adrian Chadd writes:] >I have VSTa 1.5.2 going on my 386sx25 w/ 2mb RAM however I'd like to get >the swapping going. I have created a second partition on the HDD, what >else do I need to do to get it enabled? A couple problems to be aware of... first, rebuilding the C library requires an FPU. VSTa never implemented FPU emulation, and given the direction of the CPU industry, I doubt there will ever be sufficient demand to warrant implementing it. 2MB is very, very thin, if you intend to do much compilation on your system. If you're interested in finding and fixing swap bugs (both plain old crashes, as well as deadlocks), then you've find an ideal starting point! Regards, Andy Valencia From daemon Wed Jul 9 14:02:18 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id OAA01359 for vsta-xplod; Wed, 9 Jul 1997 14:02:18 GMT Received: from network.ucsd.edu (root@network.ucsd.edu [132.239.1.20]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id OAA01353 for ; Wed, 9 Jul 1997 14:02:09 GMT Received: from Tanya (tanya.ucsd.edu [132.239.141.90]) by network.ucsd.edu (8.8.5/8.6.9) with SMTP id QAA03080 for <@network.ucsd.edu:vsta@zendo.com>; Wed, 9 Jul 1997 16:25:52 -0700 (PDT) Received: by Tanya (950413.SGI.8.6.12/940406.SGI.AUTO) for vsta@zendo.com id QAA14368; Wed, 9 Jul 1997 16:36:40 -0700 Date: Wed, 9 Jul 1997 16:36:40 -0700 From: edorman@Tanya.ucsd.edu (Eric Dorman) Message-Id: <199707092336.QAA14368@Tanya> To: vsta@zendo.com Subject: bugs in vfs Is there another version of src/srv/vstafs/alloc.c other than that in the 1.5.2 release? I've found a number of bugs in it and want to make sure I'm using the 'latest' stuff. The bugs all concern construction of free extent lists longer than 1 sector. The code is supposed to automagically extend the freelist but it's broken in at least 3 places that'll prevent this from working (write_freelist() and move_forward()). Also cmd/fsck.c can't handle freelist extending past the first block. I wasn't quite sure how to get the write ordering of freelist blocks to be always correct on-disk; (which is totally broken in the code I have)... I guess one could either doubly-link the incore freelist (bleh) and write it out backwards from end to beginning, or somehow recurse forward down the linklist and write the blocks out as we unwind the recursion. Recursion is I think probably the best bet, as long as one knows it won't go too deep... Regards, Eric Dorman edorman@tanya.ucsd.edu From daemon Wed Jul 9 14:57:28 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id OAA01468 for vsta-xplod; Wed, 9 Jul 1997 14:57:28 GMT Received: from puli.cisco.com (puli.cisco.com [171.69.1.174]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id OAA01462 for ; Wed, 9 Jul 1997 14:57:26 GMT Received: from localhost.cisco.com (localhost.cisco.com [127.0.0.1]) by puli.cisco.com (8.6.12/8.6.5) with SMTP id RAA21073; Wed, 9 Jul 1997 17:13:59 -0700 Message-Id: <199707100013.RAA21073@puli.cisco.com> To: edorman@Tanya.ucsd.edu (Eric Dorman) cc: vsta@zendo.com Subject: Re: bugs in vfs In-reply-to: Your message of "Wed, 09 Jul 1997 16:36:40 PDT." <199707092336.QAA14368@Tanya> Date: Wed, 09 Jul 1997 17:13:59 -0700 From: Andrew Valencia [edorman@Tanya.ucsd.edu (Eric Dorman) writes:] >The bugs all concern construction of free extent lists longer >than 1 sector. The code is supposed to automagically extend the >freelist but it's broken in at least 3 places that'll prevent >this from working (write_freelist() and move_forward()). Also >cmd/fsck.c can't handle freelist extending past the first block. I fixed some bugs along these lines several months ago (very, *very* broken, sigh). I also fixed up fsck so it fixed such damage, and I think that was around the time I taught fsck to repair things as well as detect them. If you could take a look at: ftp://ftp.zendo.com/vsta/snap/*061297 and tell me whether the vstafs stuff in there is better, I'd be eager to know. Regards, Andy From daemon Fri Jul 11 07:30:21 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id HAA04469 for vsta-xplod; Fri, 11 Jul 1997 07:30:21 GMT Received: from network.ucsd.edu (root@network.ucsd.edu [132.239.1.195]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id HAA04463 for ; Fri, 11 Jul 1997 07:30:17 GMT Received: from Tanya (tanya.ucsd.edu [132.239.141.90]) by network.ucsd.edu (8.8.5/8.6.9) with SMTP id JAA05961 for <@network.ucsd.edu:vsta@zendo.com>; Fri, 11 Jul 1997 09:54:06 -0700 (PDT) Received: by Tanya (950413.SGI.8.6.12/940406.SGI.AUTO) for vsta@zendo.com id KAA19409; Fri, 11 Jul 1997 10:05:03 -0700 Date: Fri, 11 Jul 1997 10:05:03 -0700 From: edorman@Tanya.ucsd.edu (Eric Dorman) Message-Id: <199707111705.KAA19409@Tanya> To: vsta@zendo.com Subject: Re: bugs in vfs From: Andrew Valencia >[edorman@Tanya.ucsd.edu (Eric Dorman) writes:] >>[vfs bugs] > >I fixed some bugs along these lines several months ago (very, *very* broken, >sigh). I also fixed up fsck so it fixed such damage, and I think that was >around the time I taught fsck to repair things as well as detect them. If >you could take a look at: > ftp://ftp.zendo.com/vsta/snap/*061297 >and tell me whether the vstafs stuff in there is better, I'd be eager to >know. > Regards, > Andy Cool.. I'll try to give that a shot tonight/tomorrow; I've got ftp downloading 061297 as I write :) .. I had worked up a fix that would extend the freelist correctly, but it always seemed to copy extents into the last added freelist sector rather than balancing the contents correctly. Ugh. I think I was confused about the meaning of f_nfree for awhile =:) vsta is cool stuff; it's nice to have similar flexibility to Plan9 w/o the horrid distribution restrictions. Regards, Eric Dorman edorman@tanya.ucsd.edu From daemon Fri Jul 11 08:12:27 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id IAA04594 for vsta-xplod; Fri, 11 Jul 1997 08:12:27 GMT Received: from network.ucsd.edu (root@network.ucsd.edu [132.239.1.195]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id IAA04588 for ; Fri, 11 Jul 1997 08:12:23 GMT Received: from Tanya (tanya.ucsd.edu [132.239.141.90]) by network.ucsd.edu (8.8.5/8.6.9) with SMTP id KAA06061 for <@network.ucsd.edu:vsta@zendo.com>; Fri, 11 Jul 1997 10:36:09 -0700 (PDT) Received: by Tanya (950413.SGI.8.6.12/940406.SGI.AUTO) for vsta@zendo.com id KAA19483; Fri, 11 Jul 1997 10:47:18 -0700 Date: Fri, 11 Jul 1997 10:47:18 -0700 From: edorman@Tanya.ucsd.edu (Eric Dorman) Message-Id: <199707111747.KAA19483@Tanya> To: vsta@zendo.com Subject: Re: vfs bugs From: Andrew Valencia >[edorman@Tanya.ucsd.edu (Eric Dorman) writes:] >>[vfs bugs] > >I fixed some bugs along these lines several months ago (very, *very* broken, >sigh). I also fixed up fsck so it fixed such damage, and I think that was >around the time I taught fsck to repair things as well as detect them. If >you could take a look at: > ftp://ftp.zendo.com/vsta/snap/*061297 >and tell me whether the vstafs stuff in there is better, I'd be eager to >know. > Regards, > Andy Cool.. I'll try to give that a shot tonight/tomorrow; I've got ftp downloading 061297 as I write :) .. I had worked up a fix that would extend the freelist correctly, but it always seemed to copy extents into the last added freelist sector rather than balancing the contents correctly. Ugh. I think I was confused about the meaning of f_nfree for awhile =:) vsta is cool stuff; it's nice to have similar flexibility to Plan9 w/o the horrid distribution restrictions. Regards, Eric Dorman edorman@tanya.ucsd.edu From daemon Mon Jul 14 10:17:34 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id KAA12286 for vsta-xplod; Mon, 14 Jul 1997 10:17:34 GMT Received: from network.ucsd.edu (root@network.ucsd.edu [132.239.1.195]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id KAA12279 for ; Mon, 14 Jul 1997 10:16:13 GMT Received: from Tanya (tanya.ucsd.edu [132.239.141.90]) by network.ucsd.edu (8.8.5/8.6.9) with SMTP id MAA26124 for <@network.ucsd.edu:vsta@zendo.com>; Mon, 14 Jul 1997 12:40:18 -0700 (PDT) Received: by Tanya (950413.SGI.8.6.12/940406.SGI.AUTO) for vsta@zendo.com id MAA27583; Mon, 14 Jul 1997 12:51:54 -0700 Date: Mon, 14 Jul 1997 12:51:54 -0700 From: edorman@Tanya.ucsd.edu (Eric Dorman) Message-Id: <199707141951.MAA27583@Tanya> To: vsta@zendo.com Subject: 061297: bugs in vfs Hello, 061297 vfs seems to work alot better! I have been able to get through several iterations of compile/builds without losing any blocks, though I seems to have managed to stay within a single freelist sector, which is fine with me :) I'lll get to unpacking the other tars as well and fill the thing up. I did notice however that the response of the system as a whole seems to be more sluggish. Startup is somewhat slower and response in different vtys is, well, 'sticky'. It's kinda hard to explain.. I'll try to quantify it more this wk. I had a few comments about the src tree: srv/mach/fd, srv/mach/scsi, bin/init needs -lsrv in LDFLAGS. srv/mach/joystick stuffers from coming from a dosfs; libjoystick.c and libjoystick.h get truncated filenames so the makefile dies. ls -l output format isin't conducive to pipelining, and I personally find it hard to read. I was wondering what negative results might come from doing something like mount fs/root:vsta / all the time, to construct a union mount in / of the dos stuff and the vsta stuff. The intent is to establish a Plan9ish consistency, with /lib, /bin, /include etc. and writing the src tree to assume these are the locations of stuff it needs. This mount would eliminate a depth dependency that exists in the src tree (all the -I/-L../../../ stuff) and make 'mkall install' work regardless of where the src tree resided, and more easily support more architectures. I haven't run into any trouble with this mount even though it hides whatever bin/lib/include dos junk is beneath it, but I have very little DOSstuff :) Sincerely, Eric Dorman edorman@tanya.ucsd.edu From daemon Tue Jul 15 05:36:07 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id FAA13780 for vsta-xplod; Tue, 15 Jul 1997 05:36:07 GMT Received: from puli.cisco.com (puli.cisco.com [171.69.1.174]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id FAA13774 for ; Tue, 15 Jul 1997 05:35:55 GMT Received: from localhost.cisco.com (localhost.cisco.com [127.0.0.1]) by puli.cisco.com (8.6.12/8.6.5) with SMTP id HAA05069; Tue, 15 Jul 1997 07:59:32 -0700 Message-Id: <199707151459.HAA05069@puli.cisco.com> To: edorman@Tanya.ucsd.edu (Eric Dorman) cc: vsta@zendo.com Subject: Re: 061297: bugs in vfs In-reply-to: Your message of "Mon, 14 Jul 1997 12:51:54 PDT." <199707141951.MAA27583@Tanya> Date: Tue, 15 Jul 1997 07:59:32 -0700 From: Andrew Valencia [edorman@Tanya.ucsd.edu (Eric Dorman) writes:] >I did notice however that the response of the system as a whole >seems to be more sluggish. Startup is somewhat slower and response >in different vtys is, well, 'sticky'. It's kinda hard to explain.. >I'll try to quantify it more this wk. Check "ps" to see who's running. You might have a runaway process. I saw this sort of behavior once, when nothing showed up in ps. It turned out to be the power management mode running wild; I had to reinitialize its flash storage and reboot, after which things went back to normal. >I had a few comments about the src tree: > srv/mach/fd, srv/mach/scsi, bin/init needs -lsrv in LDFLAGS. Fixed in my latest private tree. I'll put up a new snapshot when I get back to the states. > srv/mach/joystick stuffers from coming from a dosfs; libjoystick.c >and libjoystick.h get truncated filenames so the makefile dies. Ah yes. > ls -l output format isin't conducive to pipelining, and I personally >find it hard to read. There was a poprt of GNU "ls". That'd help, I suppose. >I was wondering what negative results might come from doing something >like > mount fs/root:vsta / >all the time, to construct a union mount in / of the dos stuff and the >vsta stuff. The intent is to establish a Plan9ish consistency, >with /lib, /bin, /include etc. and writing the src tree to assume >these are the locations of stuff it needs. This mount would eliminate >a depth dependency that exists in the src tree (all the -I/-L../../../ >stuff) and make 'mkall install' work regardless of where the src tree >resided, and more easily support more architectures. No, actually all the relative stuff is so that you can move the tree anywhere. I don't want to change that. The only remaining wart is that /vsta/include should have a matching /vsta/src/include. > I haven't run into any trouble with this mount even though >it hides whatever bin/lib/include dos junk is beneath it, but I >have very little DOSstuff :) Me, either. I wrote Hearts for MGR so I could bear to de-install Windows 3.1. :-) Andy From daemon Tue Jul 15 08:05:54 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id IAA13988 for vsta-xplod; Tue, 15 Jul 1997 08:05:54 GMT Received: from network.ucsd.edu (root@network.ucsd.edu [132.239.1.195]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id IAA13982 for ; Tue, 15 Jul 1997 08:05:17 GMT Received: from Tanya (tanya.ucsd.edu [132.239.141.90]) by network.ucsd.edu (8.8.5/8.6.9) with SMTP id KAA29267 for <@network.ucsd.edu:vsta@zendo.com>; Tue, 15 Jul 1997 10:29:09 -0700 (PDT) Received: by Tanya (950413.SGI.8.6.12/940406.SGI.AUTO) for vsta@zendo.com id KAA00451; Tue, 15 Jul 1997 10:40:57 -0700 Date: Tue, 15 Jul 1997 10:40:57 -0700 From: edorman@Tanya.ucsd.edu (Eric Dorman) Message-Id: <199707151740.KAA00451@Tanya> To: vsta@zendo.com Subject: 061297: more stuff Andy writes: >[edorman@Tanya.ucsd.edu (Eric Dorman) writes:] >>I did notice however that the response of the system as a whole >>seems to be more sluggish. Startup is somewhat slower and response >>in different vtys is, well, 'sticky'. It's kinda hard to explain.. >>I'll try to quantify it more this wk. > >Check "ps" to see who's running. You might have a runaway process. That's what I thought the first time, but haven't found anything yet. I'm *fairly* sure it's not something 'subliminal' like (you mentioned) power mgmt; that sort of ick is turned off unless the BIOS is lying :) Maybe 061297 is tickling something in my box that 1.5.2 didn't.... >> ls -l output format isin't conducive to pipelining, and I personally >>find it hard to read. >There was a poprt of GNU "ls". That'd help, I suppose. True, but it wouldn't include coping with capabilities. >>I was wondering what negative results might come from doing something >>like >> mount fs/root:vsta / >>all the time, to construct a union mount in / of the dos stuff and the >>vsta stuff. The intent is to establish a Plan9ish consistency, >>with /lib, /bin, /include etc. and writing the src tree to assume >>these are the locations of stuff it needs. This mount would eliminate >>a depth dependency that exists in the src tree (all the -I/-L../../../ >>stuff) and make 'mkall install' work regardless of where the src tree >>resided, and more easily support more architectures. > >No, actually all the relative stuff is so that you can move the tree >anywhere. Uh, I can't see how you can move the tree anywhere if there is -I../../.. specifying where standard includes are (for example); it reads to me that you have to have /include exactly 3 directories up. So if your src tree is in, say, /usr/blah/src and your includes are in /vsta/include (like my test tree) it breaks compiles and installs... I've found it easier to just union mount stuff into / (like /include, /lib, /boot) and then just say -I/include, -L/lib, etc. rather than counting dir depths.. what am I missing? >The only remaining wart is that >/vsta/include should have a matching /vsta/src/include. Mmm, two separate include dirs, /vsta/include and /vsta/src/include with the same contents? >> I haven't run into any trouble with this mount even though >>it hides whatever bin/lib/include dos junk is beneath it, but I >>have very little DOSstuff :) > >Me, either. I wrote Hearts for MGR so I could bear to de-install Windows >3.1. :-) Hehehe. I think I have a 150M dos part and a 1.4G vfs part; just enough dos stuff to keep a full vsta tree on dosfs. I guess that shows where my priorities are :) Now if I could get that jaz working... > Andy Sincerely, Eric Dorman edorman@tanya.ucsd.edu From daemon Wed Jul 16 11:31:08 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id LAA16045 for vsta-xplod; Wed, 16 Jul 1997 11:31:08 GMT Received: from puli.cisco.com (puli.cisco.com [171.69.1.174]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id LAA16039 for ; Wed, 16 Jul 1997 11:31:05 GMT Received: from localhost.cisco.com (localhost.cisco.com [127.0.0.1]) by puli.cisco.com (8.6.12/8.6.5) with SMTP id NAA07159; Wed, 16 Jul 1997 13:36:09 -0700 Message-Id: <199707162036.NAA07159@puli.cisco.com> To: edorman@Tanya.ucsd.edu (Eric Dorman) cc: vsta@zendo.com Subject: Re: 061297: more stuff In-reply-to: Your message of "Tue, 15 Jul 1997 10:40:57 PDT." <199707151740.KAA00451@Tanya> Date: Wed, 16 Jul 1997 13:36:09 -0700 From: Andrew Valencia [edorman@Tanya.ucsd.edu (Eric Dorman) writes:] >>No, actually all the relative stuff is so that you can move the tree >>anywhere. >Uh, I can't see how you can move the tree anywhere if there is >-I../../.. specifying where standard includes are (for example); it >reads to me that you have to have /include exactly 3 directories up. Well, that's the point of breaking out src/include from include/. The intent is that the src/ sub-tree will build self-consistently, so you can change your source files (including .h files) without impacting your "real" system files. >Mmm, two separate include dirs, /vsta/include and /vsta/src/include >with the same contents? No; you'd "co -l ..." exclusively in the src/include one. The other is a read-only copy of the release distribution. > Now if I could get that jaz working... I got a Zip drive sort of working; it looks like a defective unit, since FreeBSD can't make it work right, either. The problem is in the SCSI server, which doesn't do the right thing for dismountable media. The workaround is to patch out the special case for dismountable and treat it like a fixed disk. Andy From daemon Wed Jul 30 05:38:57 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id FAA15939 for vsta-xplod; Wed, 30 Jul 1997 05:38:57 GMT Received: from baruk2.zendo.com (baruk2.zendo.com [205.187.71.65]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id FAA15933 for ; Wed, 30 Jul 1997 05:38:53 GMT Received: (from vandys@localhost) by baruk2.zendo.com (8.8.5/8.6.12) id IAA15456; Wed, 30 Jul 1997 08:22:13 -0700 (PDT) Date: Wed, 30 Jul 1997 08:22:13 -0700 (PDT) Message-Id: <199707301522.IAA15456@baruk2.zendo.com> From: mikk0022@maroon.tc.umn.edu To: vsta@bodhi.zendo.com Subject: SPAMCHECK Can't boot 1.5.2 I just attempted an install of 1.5.2, and have not been successful so far. Symptom: 'init' hangs with "cannot find root" My setup: As far as disks: I have two 1G disks, and an Adaptec 1542C. I am trying to use the DOS partition as root. It is the first partition on the first drive, which has SCSI address 0. I have no IDE drives, so I did not instruct the wd server to load. Basically, I modified boot.lst in exactly the way the CAM server documentation suggests. Any suggestions? -Chris From daemon Wed Jul 30 06:15:13 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id GAA16101 for vsta-xplod; Wed, 30 Jul 1997 06:15:13 GMT Received: (from vandys@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id GAA16094; Wed, 30 Jul 1997 06:15:12 GMT Date: Wed, 30 Jul 1997 06:15:12 GMT Message-Id: <199707300615.GAA16094@bodhi.zendo.com> From: mikk0022@maroon.tc.umn.edu To: vsta@bodhi.zendo.com Subject: Can't boot 1.5.2 I just attempted an install of 1.5.2, and have not been successful so far. Symptom: 'init' hangs with "cannot find root" My setup: As far as disks: I have two 1G disks, and an Adaptec 1542C. I am trying to use the DOS partition as root. It is the first partition on the first drive, which has SCSI address 0. I have no IDE drives, so I did not instruct the wd server to load. Basically, I modified boot.lst in exactly the way the CAM server documentation suggests. Any suggestions? -Chris From daemon Thu Aug 7 13:17:49 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id NAA02009 for vsta-xplod; Thu, 7 Aug 1997 13:17:49 GMT Received: from albeniz.orchestra.cse.unsw.EDU.AU (root@albeniz.orchestra.cse.unsw.EDU.AU [129.94.242.13]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id NAA02003 for ; Thu, 7 Aug 1997 13:17:45 GMT Received: From flute10 With LocalMail ; Fri, 8 Aug 97 08:42:48 +1000 From: s2156593@cse.unsw.edu.au (Binh Thai) To: vsta@zendo.com Date: Fri, 8 Aug 97 08:42:48 +1000 Message-Id: <970807224248.4900@cse.unsw.edu.au> Subject: Mgr Hi everyone, I would like to know is there anyone managed to get mgr working? My thesis supervisor and I are trying to get it to work - but without success. When I start mgr, all i get is a blank screen, and the whole system locks up. Thanks Binh Thai From daemon Sat Aug 9 07:58:50 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id HAA06807 for vsta-xplod; Sat, 9 Aug 1997 07:58:50 GMT Received: from baruk2.zendo.com (baruk2.zendo.com [205.187.71.65]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id HAA06801 for ; Sat, 9 Aug 1997 07:57:27 GMT Received: from localhost.zendo.com (localhost.zendo.com [127.0.0.1]) by baruk2.zendo.com (8.8.5/8.6.12) with SMTP id KAA20594; Sat, 9 Aug 1997 10:21:46 -0700 (PDT) Message-Id: <199708091721.KAA20594@baruk2.zendo.com> X-Authentication-Warning: baruk2.zendo.com: localhost.zendo.com [127.0.0.1] didn't use HELO protocol To: s2156593@cse.unsw.edu.au (Binh Thai) cc: vsta@zendo.com Subject: Re: Mgr In-reply-to: Your message of "Fri, 08 Aug 1997 08:42:48 +1000." <970807224248.4900@cse.unsw.edu.au> Date: Sat, 09 Aug 1997 10:21:46 -0700 From: Andy Valencia [s2156593@cse.unsw.edu.au (Binh Thai) writes:] > I would like to know is there anyone managed to get mgr working? My the >sis >supervisor and I are trying to get it to work - but without success. > When I start mgr, all i get is a blank screen, and the whole system loc >ks up. Make sure you have a mouse driver running before you start mgr. After that, make sure your video card is in 640x480 mode (mgr tries to switch it, but I've seen cases where the video card gets locked into its mode, especially if you boot VSTa from under Win95). There's a test program in vsta/src/srv/mach/mouse which'll tell you if your mouse driver is present and working. Andy From daemon Sun Aug 10 13:38:15 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id NAA08574 for vsta-xplod; Sun, 10 Aug 1997 13:38:15 GMT Received: from albeniz.orchestra.cse.unsw.EDU.AU (root@albeniz.orchestra.cse.unsw.EDU.AU [129.94.242.13]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id NAA08568 for ; Sun, 10 Aug 1997 13:38:12 GMT Received: From pipe10 With LocalMail ; Mon, 11 Aug 97 09:02:37 +1000 From: s2156593@cse.unsw.edu.au (Binh Thai) To: vsta@zendo.com Date: Mon, 11 Aug 97 09:02:36 +1000 Message-Id: <970810230236.14646@cse.unsw.edu.au> Subject: The Next Step in Mgr Hi everyone, Thanks for giving me some ideas on how to get mgr to work. My initial errors were not installing the mouse driver. I now have managed to get past the opening screen and into a screen with b&w pattern background (i suspect that is the worksapce). Here's where my next problem starts: my mouse is not responding the way it should. The cursor is going all over place, and Mgr eventaully locks up (I suspect it's waiting for some mouse response). I have read on the mailing list that there are some errors in the mouse driver, and the variable "y_off" needs an - sign. If that is cause of the problem, can anyone tell me which file that variable resides? Thanks. From daemon Mon Aug 11 05:24:19 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id FAA09601 for vsta-xplod; Mon, 11 Aug 1997 05:24:19 GMT Received: from puli.cisco.com (puli.cisco.com [171.69.1.174]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id FAA09595 for ; Mon, 11 Aug 1997 05:24:15 GMT Received: from localhost.cisco.com (localhost.cisco.com [127.0.0.1]) by puli.cisco.com (8.6.12/8.6.5) with SMTP id HAA02027; Mon, 11 Aug 1997 07:42:47 -0700 Message-Id: <199708111442.HAA02027@puli.cisco.com> To: s2156593@cse.unsw.edu.au (Binh Thai) cc: vsta@zendo.com Subject: Re: The Next Step in Mgr In-reply-to: Your message of "Mon, 11 Aug 1997 09:02:36 +1000." <970810230236.14646@cse.unsw.edu.au> Date: Mon, 11 Aug 1997 07:42:47 -0700 From: Andrew Valencia [s2156593@cse.unsw.edu.au (Binh Thai) writes:] > I now have managed to get past the opening screen and into a screen wi >th b&w >pattern background (i suspect that is the worksapce). Here's where my next pro >blem starts: >my mouse is not responding the way it should. The cursor is going all over pla >ce, and Mgr >eventaully locks up (I suspect it's waiting for some mouse response). Is it a serial mouse? You have to make sure the mouse driver is not compiled with KDB, otherwise the first ^Z it sees will drop you into the kernel debugger. Of course, you're in graphics mode, so you can't see the text screen debugger.... Try typing "reb" a couple times; if your machine restarts that's what happened. > I have read on the mailing list that there are some errors in the mouse > driver, >and the variable "y_off" needs an - sign. If that is cause of the problem, can > anyone >tell me which file that variable resides? I think all such things are fixed in current code. You should run the test program in src/srv/mach/mouse, and see if it shows X and Y values and buttons being tracked correctly. I've seen one mouse where it was basically a ps2aux model, but with the bits all hashed around. Mostly, the mouse models provided by Dave Hudson all those years ago have worked fine. Especially on the serial mouse models. Andy From daemon Wed Aug 13 01:47:17 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id BAA12770 for vsta-xplod; Wed, 13 Aug 1997 01:47:17 GMT Received: from albeniz.orchestra.cse.unsw.EDU.AU (root@albeniz.orchestra.cse.unsw.EDU.AU [129.94.242.13]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id BAA12764 for ; Wed, 13 Aug 1997 01:47:12 GMT Received: From mozart With LocalMail ; Wed, 13 Aug 97 21:12:28 +1000 From: s2156593@cse.unsw.edu.au (Binh Thai) To: vsta@zendo.com Date: Wed, 13 Aug 97 21:12:28 +1000 Message-Id: <970813111228.27194@cse.unsw.edu.au> Subject: Mouse driver Hi everyone, I'm sorry about this, but this is starting to bug me. I have followed Andy's instruction and run /src/srv/mach/mouse/test to test the mouse (serial one), and I noticed the co-ordinates given are not correct - this explains why the cursor is all over the place in mgr. Also every so often it drops into the kernel debugger - a ^Z has come from somewhere. Is there anyway I can fix this problem? Thanks. binh thai From daemon Wed Aug 13 06:05:27 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id GAA13212 for vsta-xplod; Wed, 13 Aug 1997 06:05:27 GMT Received: from puli.cisco.com (puli.cisco.com [171.69.1.174]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id GAA13206 for ; Wed, 13 Aug 1997 06:05:06 GMT Received: from localhost.cisco.com (localhost.cisco.com [127.0.0.1]) by puli.cisco.com (8.6.12/8.6.5) with SMTP id IAA03218; Wed, 13 Aug 1997 08:29:33 -0700 Message-Id: <199708131529.IAA03218@puli.cisco.com> To: s2156593@cse.unsw.edu.au (Binh Thai) cc: vsta@zendo.com Subject: Re: Mouse driver In-reply-to: Your message of "Wed, 13 Aug 1997 21:12:28 +1000." <970813111228.27194@cse.unsw.edu.au> Date: Wed, 13 Aug 1997 08:29:33 -0700 From: Andrew Valencia [s2156593@cse.unsw.edu.au (Binh Thai) writes:] > I have followed Andy's instruction and run /src/srv/mach/mouse/test to >test the mouse (serial one), and I noticed the co-ordinates given are not >correct - this explains why the cursor is all over the place in mgr. Also ever >y >so often it drops into the kernel debugger - a ^Z has come from somewhere. > Is there anyway I can fix this problem? You still haven't said what kind of mouse it is... serial? Bus? PS/2? If serial, did you recompile the rs232 server with KDB turned off? If not, perhaps you're crashing. What's the stack backtrace (kernel debugger command "bt") say? Andy From daemon Thu Aug 14 11:00:51 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id LAA15272 for vsta-xplod; Thu, 14 Aug 1997 11:00:51 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id LAA15266 for ; Thu, 14 Aug 1997 11:00:49 GMT Received: from localhost (vandys@localhost) by zipper.cisco.com (8.8.4-Cisco.1/8.6.5) with SMTP id NAA19260; Thu, 14 Aug 1997 13:25:58 -0700 (PDT) Message-Id: <199708142025.NAA19260@zipper.cisco.com> To: Larry Barchett cc: vsta@zendo.com Subject: Re: Web Pages In-reply-to: Your message of "Thu, 14 Aug 1997 10:18:51 CDT." <199708141518.KAA19016@pclsys82.dcrl.nd.edu> Date: Thu, 14 Aug 1997 13:25:58 -0700 From: Andrew Valencia [Larry Barchett writes:] >I missed something somewhere. The web pages seem to have moved. Can someone >please tell me where? http://www.zendo.com/vsta Enjoy, Andy From daemon Mon Aug 18 01:35:11 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id BAA22882 for vsta-xplod; Mon, 18 Aug 1997 01:35:11 GMT Received: from albeniz.orchestra.cse.unsw.EDU.AU (root@[129.94.242.13]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id BAA22876 for ; Mon, 18 Aug 1997 01:35:01 GMT Received: From ives With LocalMail ; Mon, 18 Aug 97 21:00:26 +1000 From: s2156593@cse.unsw.edu.au (Binh Thai) To: vsta@zendo.com Date: Mon, 18 Aug 97 21:00:26 +1000 Message-Id: <970818110026.11868@cse.unsw.edu.au> Subject: Mouse driver trouble (again) Hi everyone, I have tried very hard during the past week in an attempt to get the mouse driver to work properly, so far I have NO success whatsoever. The mouse i am using is a SERIAL one, i have commented out the KBD linesin the source code, so now it doesn't drop into the kernel debugger. However, the movement of the mouse cursor one the screen is inconsistantwith my actual mouse movement, and using /src/srv/mach/mouse/test, it basically confirms by findings - a down movement registered dx values etc. My question is this - can any tell me in the ibmrs232.c file, the function mouse(), uses a buffer replied from the rs232 driver to manipulate the dx and dy values. How does those values values came about? What's written doesn't seem to be 100% correct. If I know how those values are derived, I might have a chance to solve the problem. Thank you very much Binh thai / From daemon Mon Aug 18 15:26:43 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id PAA24078 for vsta-xplod; Mon, 18 Aug 1997 15:26:43 GMT Received: from staff.cs.su.OZ.AU (staff.cs.su.OZ.AU [129.78.8.1]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id PAA24072 for ; Mon, 18 Aug 1997 15:26:32 GMT Received: from suede.sw.oz.au by swallow.sw.oz.au with ESMTP id AAA03342; Tue, 19 Aug 1997 00:51:50 GMT (8.6.10/Unixware) (from jeremy@suede.sw.oz.au for ) Received: by suede.sw.oz.au id AAA05850; Tue, 19 Aug 1997 00:51:48 GMT (SMI-8.6/1.34) (from jeremy for ) From: "Jeremy Fitzhardinge" Message-Id: <9708191051.ZM5848@suede.sw.oz.au> Date: Tue, 19 Aug 1997 10:51:47 -0500 In-Reply-To: s2156593@cse.unsw.edu.au (Binh Thai) "Mouse driver trouble (again)" (Aug 18, 9:00pm) References: <970818110026.11868@cse.unsw.edu.au> X-Face: '6U=%Tv\k1l-:?\$C[D@G 7(vl~w8&y}!f\bh#wL#n,TGKh>T.c7eT5-y)Hl'i;A1z$9?*lD.k}yqshddFb l[EC}c=;uc%x'}uh3E91p&oE Subject: Mouse driver trouble (again) > Hi everyone, > I have tried very hard during the past week in an attempt to get the > mouse driver to work properly, so far I have NO success whatsoever. > The mouse i am using is a SERIAL one, i have commented out the KBD linesin the source code, so now it doesn't drop into the kernel debugger. > However, the movement of the mouse cursor one the screen is inconsistantwith my actual mouse movement, and using /src/srv/mach/mouse/test, it basically > confirms by findings - a down movement registered dx values etc. > My question is this - can any tell me in the ibmrs232.c file, the > function mouse(), uses a buffer replied from the rs232 driver to manipulate > the dx and dy values. How does those values values came about? What's written > doesn't seem to be 100% correct. If I know how those values are derived, I > might have a chance to solve the problem. Hi, I had some problems with mice a while ago. I found that my Logitech Series 9 mouse wouldn't work without having DTR and RTS cycled at it, and sending "*B" (I think). I modified my copy of the mouse server to do this; I'll see if I can dig it up for you. Otherwise, I had no problems with my very old Genius mouse; I guess it just depends on lots of detail... J From daemon Mon Aug 18 19:26:10 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id TAA24368 for vsta-xplod; Mon, 18 Aug 1997 19:26:10 GMT Received: from mantis.radlogic.com.au ([192.84.219.3]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id TAA24362 for ; Mon, 18 Aug 1997 19:26:01 GMT Date: Tue, 19 Aug 1997 14:11:09 +0930 From: Ivan Curtis Message-Id: <199708190441.OAA29260@cicada.radlogic.com.au> To: s2156593@cse.unsw.edu.au In-reply-to: Binh Thai's message of Mon, 18 Aug 97 21:00:26 +1000 <970818110026.11868@cse.unsw.edu.au> Subject: Mouse driver trouble (again) Binh, > I have tried very hard during the past week in an attempt to get the > mouse driver to work properly, so far I have NO success whatsoever. I have seen your past postings, and have refrained from replying with this because it seems so obvious. But you sound desperate, so just in case you haven't considered this: Is your mouse protocol what the driver expects? There are two main protocols: Microsoft (2 button mouse) and Mouse Systems (typically for a 3 button mouse). Some mice use a special handshake sequence to select the protocol, which the supplied DOS or windows driver knows about, but the vsta driver may not. If you think this may be your problem, my suggestion is to find a mouse with a 2/3 button select switch on the bottom and try it in both positions. (sadly, this sort of mouse is getting harder to find...) Ivan icurtis@radlogic.com.au ph: 61 8 8373-3990 Adelaide, South AUSTRALIA fx: 61 8 8373-3251 From daemon Sat Aug 23 15:01:25 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id PAA05138 for vsta-xplod; Sat, 23 Aug 1997 15:01:25 GMT Received: from baruk2.zendo.com (baruk2.zendo.com [205.187.71.65]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id PAA05132 for ; Sat, 23 Aug 1997 15:01:21 GMT Received: from localhost.zendo.com (localhost.zendo.com [127.0.0.1]) by baruk2.zendo.com (8.8.5/8.6.12) with SMTP id RAA25616 for ; Sat, 23 Aug 1997 17:33:06 -0700 (PDT) Message-Id: <199708240033.RAA25616@baruk2.zendo.com> X-Authentication-Warning: baruk2.zendo.com: localhost.zendo.com [127.0.0.1] didn't use HELO protocol To: vsta@zendo.com Subject: 64 MB memory config under VSTa Date: Sat, 23 Aug 1997 17:33:06 -0700 From: Andy Valencia Hello, I'm hoping somebody out there can cast a little light on this subject. I'm playing with > 16 MB memory configs, and ran into a PC limitation. I extract the memory size from a pair of RTC locations, which are combined as: (hi << 8) | low. This value is in K, and thus overflows at 64 megs. Does somebody know offhand how I decode memory size in this range? Thanks! Andy Valencia From daemon Sat Aug 23 18:39:59 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id SAA05340 for vsta-xplod; Sat, 23 Aug 1997 18:39:59 GMT Received: from uruk.org (root@ns.uruk.org [198.145.95.253]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id SAA05334; Sat, 23 Aug 1997 18:39:52 GMT Received: from uruk.org [127.0.0.1] (erich) by uruk.org with esmtp (Exim 0.53 #1) id E0x2Tww-0000Pb-00; Sat, 23 Aug 1997 21:06:50 -0700 To: Andy Valencia cc: vsta@zendo.com Subject: Re: 64 MB memory config under VSTa In-reply-to: Your message of "Sat, 23 Aug 1997 17:33:06 PDT." <199708240033.RAA25616@baruk2.zendo.com> Date: Sat, 23 Aug 1997 21:06:50 -0700 From: Erich Boleyn Message-Id: Andy Valencia wrote: > I'm hoping somebody out there can cast a little light on this subject. > I'm playing with > 16 MB memory configs, and ran into a PC limitation. I > extract the memory size from a pair of RTC locations, which are combined > as: (hi << 8) | low. This value is in K, and thus overflows at 64 megs. > Does somebody know offhand how I decode memory size in this range? Yes. My bootloader GRUB both recognizes and documents the newer BIOS interfaces used by PCs. See the web page at: http://www.uruk.org/~erich/grub/ ...and look under the "technical" heading. In general, there are 3 interfaces: -- Original PC interface (the one you're talking about) which can only represent up to about 64 MB, but on some machines actually only works up to about 15 or 16 MB (generally depending on if you have an ISA memory hole between 15-16 MB). -- Newer (EISA?) interface which allows for the ISA memory hole and can go up to 4 GB. -- Modern interface which actually maps out an arbitrary number of memory "regions" which have 64-bit start and length values. The third one is the way to go if possible. Again, see GRUB and the "multiboot" standard for details. GRUB first tries the modern "memory map" interface, then falls back on the "EISA" interface, then uses the original one if it needs to. Since many OSes expect the "640K + contiguous memory above 1MB" format, it also converts as much of what it finds into that format as possible to be passed to the OS in question (Multiboot-compliant OSes, and for Linux it passes it as a "mem=XXXk" option on the command-line, for example). -- Erich Stefan Boleyn \_ E-mail (preferred): Mad Genius wanna-be, CyberMuffin \__ (finger me for other stats) Web: http://www.uruk.org/~erich/ Motto: "I'll live forever or die trying" From daemon Mon Aug 25 22:39:21 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id WAA08586 for vsta-xplod; Mon, 25 Aug 1997 22:39:21 GMT Received: from relay.xlink.net (relay.xlink.net [193.141.40.5]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id WAA08580; Mon, 25 Aug 1997 22:39:10 GMT Received: from asysha.asys-h.de (asysha.asys-h.de [193.98.1.1]) by relay.xlink.net (8.8.6/8.8.6) with SMTP id KAA35375; Tue, 26 Aug 1997 10:05:17 +0200 Received: from localhost (ilse@localhost) by asysha.asys-h.de (8.6.12/ni-1.0) with SMTP id JAA22944; Tue, 26 Aug 1997 09:50:19 +0200 Date: Tue, 26 Aug 1997 09:50:18 +0200 (MET DST) From: Juergen Ilse To: Andy Valencia cc: vsta@zendo.com Subject: Re: 64 MB memory config under VSTa In-Reply-To: <199708240033.RAA25616@baruk2.zendo.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hello, On Sat, 23 Aug 1997, Andy Valencia wrote: > Hello, > I'm hoping somebody out there can cast a little light on this subject. > I'm playing with > 16 MB memory configs, and ran into a PC limitation. I > extract the memory size from a pair of RTC locations, which are combined > as: (hi << 8) | low. This value is in K, and thus overflows at 64 megs. > Does somebody know offhand how I decode memory size in this range? > I don't know a way to come around this limitation, but can't we use a command-line-switch for the start of vsta, which overrides the detected amount of memory? This is the way, that is used by linux (commandline parameter mem=xxx of the linux kernel). ciao, Juergen Ilse (ilse@asys-h.de) From daemon Tue Aug 26 05:22:29 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id FAA09119 for vsta-xplod; Tue, 26 Aug 1997 05:22:29 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id FAA09113 for ; Tue, 26 Aug 1997 05:22:27 GMT Received: from localhost (vandys@localhost) by zipper.cisco.com (8.8.4-Cisco.1/8.6.5) with SMTP id HAA14603; Tue, 26 Aug 1997 07:47:38 -0700 (PDT) Message-Id: <199708261447.HAA14603@zipper.cisco.com> To: Juergen Ilse cc: vsta@zendo.com Subject: Re: 64 MB memory config under VSTa In-reply-to: Your message of "Tue, 26 Aug 1997 09:50:18 +0200." Date: Tue, 26 Aug 1997 07:47:37 -0700 From: Andrew Valencia [Juergen Ilse writes:] >I don't know a way to come around this limitation, but can't we use a >command-line-switch for the start of vsta, which overrides the detected >amount of memory? This is the way, that is used by linux (commandline >parameter mem=xxx of the linux kernel). GRUB (GRand Unified Bootloader) seems to do a good job of figuring this out. It's also a very nice boot mechanism! Barring unforeseen issues, I expect the next release of VSTa to use GRUB (and Multiboot format) as its boot mechanism. This also dovetails with the fact that DOS is fading fast, and while VSTa's boot mechanism works from within Windows 95, it slows down the VSTa boot time a LOT. Native boot capability is really called for at this point. Andy From daemon Tue Sep 2 15:13:26 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id PAA21957 for vsta-xplod; Tue, 2 Sep 1997 15:13:26 GMT Received: from baruk2.zendo.com (baruk2.zendo.com [205.187.71.65]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id NAA21850; Tue, 2 Sep 1997 13:49:19 GMT Received: from localhost.zendo.com (localhost.zendo.com [127.0.0.1]) by baruk2.zendo.com (8.8.5/8.6.12) with SMTP id QAA00553; Tue, 2 Sep 1997 16:11:03 -0700 (PDT) Message-Id: <199709022311.QAA00553@baruk2.zendo.com> X-Authentication-Warning: baruk2.zendo.com: localhost.zendo.com [127.0.0.1] didn't use HELO protocol Date: Tue, 02 Sep 1997 16:11:02 -0700 From: Andy Valencia Subject: zendo.com outage over the weekend ------- Blind-Carbon-Copy To: vandys Subject: zendo.com outage over the weekend Date: Tue, 02 Sep 1997 16:11:02 -0700 From: Andy Valencia Due to a failure requiring physical access to the backbone router carrying zendo.com traffic, and due to the long weekend making said facility inaccessible, all zendo.com resources were unavailable from the Internet for the last couple of days. The situation has been corrected, and is unlikely to be repeated in the foreseeable future. Apologies for any inconvenience. Andy Valencia ------- End of Blind-Carbon-Copy From daemon Wed Sep 3 06:33:19 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id GAA22986 for vsta-xplod; Wed, 3 Sep 1997 06:33:19 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id GAA22980 for ; Wed, 3 Sep 1997 06:33:15 GMT Received: from localhost (vandys@localhost) by zipper.cisco.com (8.8.4-Cisco.1/8.6.5) with SMTP id IAA21018 for ; Wed, 3 Sep 1997 08:59:20 -0700 (PDT) Message-Id: <199709031559.IAA21018@zipper.cisco.com> To: vsta@zendo.com Subject: GCC 2.X tool chain up on VSTa Date: Wed, 03 Sep 1997 08:59:20 -0700 From: Andrew Valencia Well, here's a long trail of twisty passages. I started playing with a move to GRUB, and realized that I had to change how ld layed out the executable. This was hard to do with the 1.X ld, and the map files available in 2.X were the ticket. So I went and ported late-model ld, as, and gcc. As a part of this, I also had to port gawk, and fortunately I'd already done sed. I've successfully compiled and booted VSTa running under VSTa with the 2.X tool chain, so expect this switch in the next major release. I'm a little torn on what to play with next--I think I might take a little bit of time to port gdb, as debugging C programs with adb gets old quickly. OTOH, the sooner I move to GRUB, the sooner I can get native boots, 64 meg support, and all those other goodies. :-) Andy From daemon Mon Sep 8 20:25:40 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id UAA03993 for vsta-xplod; Mon, 8 Sep 1997 20:25:40 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id UAA03987 for ; Mon, 8 Sep 1997 20:25:38 GMT Received: from localhost (vandys@localhost) by zipper.cisco.com (8.8.4-Cisco.1/8.6.5) with SMTP id WAA02077 for ; Mon, 8 Sep 1997 22:51:59 -0700 (PDT) Message-Id: <199709090551.WAA02077@zipper.cisco.com> To: vsta@zendo.com Subject: VSTa catapults into the 70's! Date: Mon, 08 Sep 1997 22:51:59 -0700 From: Andrew Valencia I have gdb ported to VSTa and it seems to be working correctly. Breakpoints, single step, backtrace, step-over-function, and so forth are all operation--all at the source level! I like adb as much as the next guy, but things sure are a lot easier when you don't have to pore over individual instructions and hex dumps of structs. :-) I guess I need to push out a new version. There's a whole bunch of fixes, and now an upgrade to the tool chain itself. Of course, my dual processor Pentium Pro motherboard just croaked, so first I have to go see what it'll take to get it fixed. :-( Andy From daemon Tue Sep 9 05:47:56 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id FAA04683 for vsta-xplod; Tue, 9 Sep 1997 05:47:56 GMT Received: from baruk2.zendo.com (baruk2.zendo.com [205.187.71.65]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id FAA04677 for ; Tue, 9 Sep 1997 05:47:52 GMT Received: from localhost.zendo.com (localhost.zendo.com [127.0.0.1]) by baruk2.zendo.com (8.8.5/8.6.12) with SMTP id IAA26794; Tue, 9 Sep 1997 08:10:26 -0700 (PDT) Message-Id: <199709091510.IAA26794@baruk2.zendo.com> X-Authentication-Warning: baruk2.zendo.com: localhost.zendo.com [127.0.0.1] didn't use HELO protocol To: CChuang cc: vsta@zendo.com Subject: Re: how to make a cross complier targeted on vsta ? In-reply-to: Your message of "Tue, 09 Sep 1997 11:15:02 +0800." <3414BF36.C97E499C@iii.org.tw> Date: Tue, 09 Sep 1997 08:10:26 -0700 From: Andy Valencia [CChuang writes:] >Does anyone who has the successful story of building a cross complier >which cross from i386-winnt3.5(host) to i386-vsta(target) ? >I failed to pass the whole make process becuz too many error occurs. My >environment is Windows NT and I complied the GNU gcc source by Cygnus's >cdk package in the "bash" shell. I wonder if it will be a better idea to >build the cross complier in a freebsd environment. Could anyone give me >some suggestions ? thanks. Nope, never heard of cross compiling from NT. I *have* heard of FreeBSD, Linux, and Solaris cross-compilations. VSTa just uses a.out format, paged format, a.out header at 0x1000, text starting at 0x1020, and data starting at the first 4 meg boundary above the end of text. Andy From daemon Wed Sep 10 05:33:37 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id FAA06282 for vsta-xplod; Wed, 10 Sep 1997 05:33:37 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id FAA06276 for ; Wed, 10 Sep 1997 05:33:33 GMT Received: from localhost (vandys@localhost) by zipper.cisco.com (8.8.4-Cisco.1/8.6.5) with SMTP id HAA14879; Wed, 10 Sep 1997 07:59:57 -0700 (PDT) Message-Id: <199709101459.HAA14879@zipper.cisco.com> To: David Jeske cc: vsta@zendo.com Subject: Re: MP on P5 vs P6 In-reply-to: Your message of "Tue, 09 Sep 1997 23:31:06 PDT." <19970909233106.07310@home.chat.net> Date: Wed, 10 Sep 1997 07:59:57 -0700 From: Andrew Valencia [David Jeske writes:] >What's your feeling and experience on P5 SMP vs P6 SMP? It's very >inexpensive to get the required equipment to build a P5 SMP system ($190 >for a Tyan Tomcat 4 dual P5 board, $150 for a P5/166MMX CPU). While a P6 >is somewhat more costly ($290 for a SuperMicro dual P6 board, $275 for a >P6/180), I would expect it's "SMP" related performance to be significantly >higher, due both to the better processor bus protocol, and to the dual, >and private L2 caches. Yes, you've hit the nail on the head. Also there's the question of interrupt distribution, which was often assymetric in those low-end P5 "SMP" systems. >Of course these are all the bargan basement boards and CPUs, and I would >guess there would be significant gains to using 512k L2 instead of 256k L2 >P6 chips. Yes, but a coherent L2 cache at 256k is still usable. >Anyhow, my question goes something like, is the P6 SMP performance >enough to make it not reasonable to make a dual P5? Knowing that my goal >here is to put togeather a fairly inexpensive SMP system with the intent >of experimenting with SMP and responsiveness vs speed tradeoffs that SMP >provides. (i.e. not necessarily to make the fastest machine I can for the >dollar) My recommendation would be to go PPro for sure. If you want to economize, getting the 256k cache configuration is a better compromise than going to P5's. Shared L2 cache with write-through L1 is going to make it hard to scale to any extent at all. $0.02, Andy From daemon Wed Sep 10 07:16:09 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id HAA06428 for vsta-xplod; Wed, 10 Sep 1997 07:16:09 GMT Received: from network.ucsd.edu (network.ucsd.edu [132.239.1.195]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id HAA06422 for ; Wed, 10 Sep 1997 07:16:07 GMT Received: from Tanya (tanya.ucsd.edu [132.239.141.90]) by network.ucsd.edu (8.8.5/8.6.9) with SMTP id JAA27188 for <@network.ucsd.edu:vsta@zendo.com>; Wed, 10 Sep 1997 09:42:57 -0700 (PDT) Received: by Tanya (950413.SGI.8.6.12/940406.SGI.AUTO) for vsta@zendo.com id JAA14414; Wed, 10 Sep 1997 09:54:29 -0700 Date: Wed, 10 Sep 1997 09:54:29 -0700 From: edorman@Tanya.ucsd.edu (Eric Dorman) Message-Id: <199709101654.JAA14414@Tanya> To: vsta@zendo.com Subject: Re: MP on P5 vs P6 >[David Jeske writes:] > >>What's your feeling and experience on P5 SMP vs P6 SMP? It's very >>inexpensive to get the required equipment to build a P5 SMP system ($190 >>for a Tyan Tomcat 4 dual P5 board, $150 for a P5/166MMX CPU). While a P6 >>is somewhat more costly ($290 for a SuperMicro dual P6 board, $275 for a >>P6/180), I would expect it's "SMP" related performance to be significantly >>higher, due both to the better processor bus protocol, and to the dual, >>and private L2 caches. I've seen PPRO-150s around here for $215 each, and can almost always be throttled to -166. The trouble with the -180s is that you must run a memory bus speed of 60Mhz (180/3) and thus a PCI busspeed of 30Mhz. If OTOH you can run -166 or -200 (expensive) you can strap the memory bus speed to 66Mhz (166/5) and PCI speed of 33Mhz. From what the *BSD crew have been saying it seems the 66Mhz multiple processors (133, 166, 200 aka 198) tend to be a better bet if you're running stuff that bounces out of the cache (e.g X11 :) ) or if you want the most out of your I/O. I think the speed bump I saw going from -150 to -166 on my PPROs was about 10-15% but that's purely subjective. Plan9 certainly runs faster :) Unfortunately PPRO-150s are getting increasingly hard to find. I've had zero trouble with the Tyan S1668 ATX 2xPPRO board running everything from FreeBSD-3.0-SNAP (SMP) to WinNT4. The SuperMicro boards are supposedly pretty stable too (except yours, Andy? :< ) >>Of course these are all the bargan basement boards and CPUs, and I would >>guess there would be significant gains to using 512k L2 instead of 256k L2 >>P6 chips. >Yes, but a coherent L2 cache at 256k is still usable. I've found some vendors who didn't know how to tell externally between a PPRO200-256k and PPRO200-512k. Needless to say they'll never be seeing my business again... one has to be careful about these guys. I'm sure there are numbers on the chip somewhere, but I'm still not absolutely sure. AFAIK the 150s were only built with -256k L1s. [xx] >My recommendation would be to go PPro for sure. If you want to economize, >getting the 256k cache configuration is a better compromise than going to >P5's. Shared L2 cache with write-through L1 is going to make it hard to >scale to any extent at all. I think 'disappointing' was the word used when the FreeBSD crowed surveyed P5-SMP performance :) Their caches just aren't big enough I guess. > Andy Regards, Eric Dorman edorman@tanya.ucsd.edu From daemon Wed Sep 10 08:19:02 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id IAA06510 for vsta-xplod; Wed, 10 Sep 1997 08:19:02 GMT Received: from proxy3.ba.best.com (root@proxy3.ba.best.com [206.184.139.14]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id IAA06501 for ; Wed, 10 Sep 1997 08:18:50 GMT Received: from shellx.best.com (shellx.best.com [206.86.0.11]) by proxy3.ba.best.com (8.8.7/8.8.BEST) with ESMTP id KAA23538 for ; Wed, 10 Sep 1997 10:44:22 -0700 (PDT) Received: (from tm2@localhost) by shellx.best.com (8.8.6/8.8.3) id KAA20509 for vsta@zendo.com; Wed, 10 Sep 1997 10:44:20 -0700 (PDT) From: Toshi Morita Message-Id: <199709101744.KAA20509@shellx.best.com> Subject: Re: MP on P5 vs P6 To: vsta@zendo.com Date: Wed, 10 Sep 1997 10:44:20 -0700 (PDT) In-Reply-To: <199709101654.JAA14414@Tanya> from "Eric Dorman" at Sep 10, 97 09:54:29 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit > I've had zero trouble with the Tyan S1668 ATX 2xPPRO board running > everything from FreeBSD-3.0-SNAP (SMP) to WinNT4. The SuperMicro > boards are supposedly pretty stable too (except yours, Andy? :< ) I have a SuperMicro P6DOF and it works fine. The main thing to remember is the 450GX chipset has a few bugs, and there are a few options you shouldn't turn on. If you have a P6DNF, it should work fine, since it's a 440FX chipset. > >>Of course these are all the bargan basement boards and CPUs, and I would > >>guess there would be significant gains to using 512k L2 instead of 256k L2 > >>P6 chips. > >Yes, but a coherent L2 cache at 256k is still usable. > > I've found some vendors who didn't know how to tell externally > between a PPRO200-256k and PPRO200-512k. Needless to say they'll never > be seeing my business again... one has to be careful about these guys. > I'm sure there are numbers on the chip somewhere, but I'm still not > absolutely sure. AFAIK the 150s were only built with -256k L1s. Nope, there were some buit with 512k cache. Intel was offering a choice of 512k L2 cache only on PP150s initially, then some mags benched all the PPros and found that the PP150-512k was as fast as the PP180 in most applications, which was a bit embarassing, so Intel started offering 512k cache on the PP200 as well (if I remember correctly). > >My recommendation would be to go PPro for sure. If you want to economize, > >getting the 256k cache configuration is a better compromise than going to > >P5's. Shared L2 cache with write-through L1 is going to make it hard to > >scale to any extent at all. > > I think 'disappointing' was the word used when the FreeBSD crowed surveyed > P5-SMP performance :) Their caches just aren't big enough I guess. I had a dual P120 for a while. It was about 10-15% faster when runnning dual CPUS than when running a single CPU. The dual PP180 I have now is something like 70% faster when running with both CPUs. Toshi From daemon Wed Sep 10 09:20:34 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id JAA06625 for vsta-xplod; Wed, 10 Sep 1997 09:20:34 GMT Received: from network.ucsd.edu (network.ucsd.edu [132.239.1.195]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id JAA06619 for ; Wed, 10 Sep 1997 09:20:32 GMT Received: from Tanya (tanya.ucsd.edu [132.239.141.90]) by network.ucsd.edu (8.8.5/8.6.9) with SMTP id LAA28146 for <@network.ucsd.edu:vsta@zendo.com>; Wed, 10 Sep 1997 11:47:25 -0700 (PDT) Received: by Tanya (950413.SGI.8.6.12/940406.SGI.AUTO) for vsta@zendo.com id LAA14954; Wed, 10 Sep 1997 11:58:59 -0700 From: "Eric Dorman" Message-Id: <9709101158.ZM14952@Tanya.ucsd.edu> Date: Wed, 10 Sep 1997 11:58:58 -0700 In-Reply-To: Toshi Morita "Re: MP on P5 vs P6" (Sep 10, 10:44am) References: <199709101744.KAA20509@shellx.best.com> X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail) To: vsta@zendo.com Subject: Re: MP on P5 vs P6 Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii On Sep 10, 10:44am, Toshi Morita wrote: > Subject: Re: MP on P5 vs P6 > > I've found some vendors who didn't know how to tell externally > > between a PPRO200-256k and PPRO200-512k. Needless to say they'll never > > be seeing my business again... one has to be careful about these guys. > > I'm sure there are numbers on the chip somewhere, but I'm still not > > absolutely sure. AFAIK the 150s were only built with -256k L1s. > > Nope, there were some buit with 512k cache. Intel was offering a choice of > 512k L2 cache only on PP150s initially, then some mags benched all the > PPros and found that the PP150-512k was as fast as the PP180 in most > applications, which was a bit embarassing, so Intel started offering 512k > cache on the PP200 as well (if I remember correctly). Cool! It'd be interesting to find some of those :) All I've been able to find anywhere has been the -256k's; must've made a lot more of them than the -512ks. > Toshi Sincerely, Eric Dorman edorman@tanya.ucsd.edu From daemon Wed Sep 10 11:20:54 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id LAA06796 for vsta-xplod; Wed, 10 Sep 1997 11:20:54 GMT Received: from cs.utah.edu (cs.utah.edu [128.110.4.21]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id LAA06790 for ; Wed, 10 Sep 1997 11:20:51 GMT Received: from fast.cs.utah.edu by cs.utah.edu (8.8.4/utah-2.21-cs) id OAA19318; Wed, 10 Sep 1997 14:47:42 -0600 (MDT) Received: by fast.cs.utah.edu (8.6.10/utah-2.15-leaf) id OAA00665; Wed, 10 Sep 1997 14:47:42 -0600 Date: Wed, 10 Sep 1997 14:47:42 -0600 From: vanmaren@fast.cs.utah.edu (Kevin Van Maren) Message-Id: <199709102047.OAA00665@fast.cs.utah.edu> To: edorman@Tanya.ucsd.edu, vsta@zendo.com Subject: Re: MP on P5 vs P6 > Cool! It'd be interesting to find some of those :) All I've > been able to find anywhere has been the -256k's; must've made a lot > more of them than the -512ks. These are the Pentium Pro processors made by intel, in order of increasing (SMP) performance: Pro 150/256 Pro 180/256 Pro 200/256 Pro 166/512 Pro 200/512 Pro 200/1024 Note that the 166 is hard to come by (but still being sold), and costs a little more than the 180 (if you can find it), but less than the 200/256. (Last I checked). The 200/512 is significantly more than the 256k version, and the 1024 k is astronomical. Most places still only carry the 180/256 and the 200/256. From daemon Fri Sep 12 12:27:07 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id MAA10330 for vsta-xplod; Fri, 12 Sep 1997 12:27:07 GMT Received: from baruk2.zendo.com (baruk2.zendo.com [205.187.71.65]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id MAA10324 for ; Fri, 12 Sep 1997 12:27:04 GMT Received: from localhost.zendo.com (localhost.zendo.com [127.0.0.1]) by baruk2.zendo.com (8.8.5/8.6.12) with SMTP id OAA02164 for ; Fri, 12 Sep 1997 14:52:32 -0700 (PDT) Message-Id: <199709122152.OAA02164@baruk2.zendo.com> X-Authentication-Warning: baruk2.zendo.com: localhost.zendo.com [127.0.0.1] didn't use HELO protocol To: vsta@zendo.com Subject: Fwd: Re: MP on P5 vs P6 Date: Fri, 12 Sep 1997 14:52:32 -0700 From: Andy Valencia The spam filter got in the way of this... Andy ------- Forwarded Message Return-Path: daemon@zendo.com Received: from bodhi.zendo.com (bodhi.zendo.com [205.187.71.2]) by baruk2.zendo.com (8.8.5/8.6.12) with ESMTP id TAA00688 for ; Thu, 11 Sep 1997 19:31:20 -0700 (PDT) Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id RAA08951 for owner-vsta; Thu, 11 Sep 1997 17:06:10 GMT Received: from gateway.s3.com (gateway.s3.com [198.211.58.4]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id RAA08944 for ; Thu, 11 Sep 1997 17:06:07 GMT Received: from chat.net (home.chat.net [198.211.58.200]) by gateway.s3.com (8.6.12/8.6.12) with ESMTP id TAA02996 for ; Thu, 11 Sep 1997 19:12:47 -0700 Received: by chat.net (SMI-8.6/SMI-SVR4) id TAA21927; Thu, 11 Sep 1997 19:34:37 -0700 Message-ID: <19970911193436.28143@home.chat.net> Date: Thu, 11 Sep 1997 19:34:36 -0700 From: David Jeske To: vsta@zendo.com Subject: SPAMCHECK Re: MP on P5 vs P6 References: <199709101654.JAA14414@Tanya> <199709101744.KAA20509@shellx.best.com> Mime-Version: 1.0 XXContent-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.79e In-Reply-To: <199709101744.KAA20509@shellx.best.com>; from Toshi Morita on Wed, Sep 10, 1997 at 10:44:20AM -0700 On Wed, Sep 10, 1997 at 10:44:20AM -0700, Toshi Morita wrote: > > I think 'disappointing' was the word used when the FreeBSD crowed surveyed > > P5-SMP performance :) Their caches just aren't big enough I guess. > > I had a dual P120 for a while. It was about 10-15% faster when runnning > dual CPUS than when running a single CPU. The dual PP180 I have now is > something like 70% faster when running with both CPUs. Really? What were you running that was showing this little speedup? Some friends of mine did some "real world" test on Linux-SMP, like doing parallel makes of the kernel, and a dual P5 system fared much better than this. They are getting me the numbers. - -- David Jeske (N9LCA) + jeske@chat.net ------- End of Forwarded Message From daemon Mon Sep 15 12:01:08 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id MAA17116 for vsta-xplod; Mon, 15 Sep 1997 12:01:08 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id MAA17110 for ; Mon, 15 Sep 1997 12:01:04 GMT Received: from localhost (vandys@localhost) by zipper.cisco.com (8.8.4-Cisco.1/8.6.5) with SMTP id OAA08982 for ; Mon, 15 Sep 1997 14:27:43 -0700 (PDT) Message-Id: <199709152127.OAA08982@zipper.cisco.com> To: vsta@zendo.com Subject: Node transparent IPC for VSTa Date: Mon, 15 Sep 1997 14:27:43 -0700 From: Andrew Valencia In bringing up VSTa clustering, I've run into a kernel IPC limitation which is going to result in a minor but annoying change in VSTa 1.6 (due shortly). In the current code, each time a "struct msg" is copied out to user space, the m_op field has the M_READ field cleared. This is absolutely fine when the message has arrived at its ultimate destination, because m_op can be decoded, and the needed read/write implemented. However, in the case where the message has arrived at a proxy agent, the problem is that the agent can no longer tell if the request is a read with a given buffer size (said size is required to be in m_arg), or a simple message with no associated receive buffer required. The M_READ bit of the m_op field needs to be preserved so that a message--read or write--can traverse multiple hops without losing this context. Thus, msg_receive() will now get a "struct msg" with the M_READ bit set if the sender so specified. The new manifest constant bitmask MSG_MASK must be applied before looking at m_op's value to compare/switch on filesystem operation type (FS_READ, FS_WRITE, etc.). All servers in the distribution have been updated, but if any of you have personal stashes of server code, please keep this in mind when you start fooling with 1.6. Enjoy! Andy From daemon Fri Sep 19 07:00:18 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id HAA23807 for vsta-xplod; Fri, 19 Sep 1997 07:00:18 GMT Received: (from root@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id HAA23800; Fri, 19 Sep 1997 07:00:16 GMT Date: Fri, 19 Sep 1997 07:00:16 GMT Message-Id: <199709190700.HAA23800@bodhi.zendo.com> From: CChuang X-Mailer: Mozilla 4.02 [en] (Win95; I) MIME-Version: 1.0 To: vsta@zendo.com Subject: PE -> a.out translator ? I am searching for a utility which can translate the PE(.exe) format to a.out format. Does anyone know such a tool ? or, any detour but the same effect. thanks.. Sam From daemon Thu Sep 25 18:28:47 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id SAA06511 for vsta-xplod; Thu, 25 Sep 1997 18:28:47 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id SAA06505 for ; Thu, 25 Sep 1997 18:28:44 GMT Received: from localhost (vandys@localhost) by zipper.cisco.com (8.8.4-Cisco.1/8.6.5) with SMTP id UAA01499 for ; Thu, 25 Sep 1997 20:55:54 -0700 (PDT) Message-Id: <199709260355.UAA01499@zipper.cisco.com> To: vsta@zendo.com Subject: VSTa native boot Date: Thu, 25 Sep 1997 20:55:54 -0700 From: Andrew Valencia Well, after a bit of work, I have VSTa converted to using the Multiboot specification: http://www.uruk.org/~erich/grub/boot-proposal.html and have it booting without having to use DOS/Windows 95, instead using the Multiboot compliant boot loader "GRUB" (GRand Unified Bootloader): http://www.uruk.org/grub/ This is a very nice little bootloader which can boot FreeBSD, DOS, Windows 95, NT, Hurd, Linux, and now, VSTa. In converting to use it, I moved VSTa so the kernel and boot tasks are above the 1 meg mark. Thus, the old limitations of making the kernel and boot tasks all fit in 640k is now history, too. Finally (!), GRUB can handle compressed images, so there are definite possibilities for packing a whole bunch of VSTa tasks onto a bootable floppy. Along the way, I stumbled across a whole collection of ports which Dave Hudson did ages ago, and shamefacedly rolled them into the core VSTa distribution. This includes a bunch of text, file, and language utilities which will be nice to have on tap. Assuming everybody is up for grappling with a new bootloader scheme, I think VSTa 1.6 will be ready to roll shortly. Andy Valencia From daemon Thu Sep 25 20:46:19 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id UAA06678 for vsta-xplod; Thu, 25 Sep 1997 20:46:19 GMT Received: from gateway.s3.com (gateway.s3.com [198.211.58.4]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id UAA06672 for ; Thu, 25 Sep 1997 20:46:17 GMT Received: from chat.net (home.chat.net [198.211.58.200]) by gateway.s3.com (8.6.12/8.6.12) with ESMTP id WAA28897; Thu, 25 Sep 1997 22:53:13 -0700 Received: by chat.net (SMI-8.6/SMI-SVR4) id XAA11674; Thu, 25 Sep 1997 23:15:32 -0700 Message-ID: <19970925231532.08189@home.chat.net> Date: Thu, 25 Sep 1997 23:15:32 -0700 From: David Jeske To: Andrew Valencia Cc: vsta@zendo.com Subject: Re: VSTa native boot References: <199709260355.UAA01499@zipper.cisco.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.79e In-Reply-To: <199709260355.UAA01499@zipper.cisco.com>; from Andrew Valencia on Thu, Sep 25, 1997 at 08:55:54PM -0700 On Thu, Sep 25, 1997 at 08:55:54PM -0700, Andrew Valencia wrote: > Well, after a bit of work, I have VSTa converted to using the Multiboot > specification: > > http://www.uruk.org/~erich/grub/boot-proposal.html > > and have it booting without having to use DOS/Windows 95, instead using the > Multiboot compliant boot loader "GRUB" (GRand Unified Bootloader): Great! > This is a very nice little bootloader which can boot FreeBSD, DOS, Windows > 95, NT, Hurd, Linux, and now, VSTa. In converting to use it, I moved VSTa > so the kernel and boot tasks are above the 1 meg mark. Thus, the old > limitations of making the kernel and boot tasks all fit in 640k is now > history, too. Finally (!), GRUB can handle compressed images, so there are > definite possibilities for packing a whole bunch of VSTa tasks onto a > bootable floppy. Hmm... do you think there is a need for me to fix the VSTa boot image tools to work with this new kernel location expectation? If I fix it, then you will also be able to create a LILO compatible boot image which can be launched from LILO or GRUB as a single image. > Assuming everybody is up for grappling with a new bootloader scheme, I think > VSTa 1.6 will be ready to roll shortly. I already have GRUB installed anyhow. Now the only complication is that one of the nicest parts about grub is it's ability to directly read filesystems, however, it won't be able to read VSTa FS. So that means either a FAT based VSTa, or a FAT boot partition, or giving Grub specific location information for booting images. -- David Jeske (N9LCA) + jeske@chat.net From daemon Fri Sep 26 05:06:25 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id FAA07304 for vsta-xplod; Fri, 26 Sep 1997 05:06:25 GMT Received: from cs.utah.edu (cs.utah.edu [128.110.4.21]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id FAA07298 for ; Fri, 26 Sep 1997 05:06:23 GMT Received: from fast.cs.utah.edu by cs.utah.edu (8.8.4/utah-2.21-cs) id IAA26103; Fri, 26 Sep 1997 08:33:54 -0600 (MDT) Received: by fast.cs.utah.edu (8.6.10/utah-2.15-leaf) id IAA12129; Fri, 26 Sep 1997 08:33:53 -0600 Date: Fri, 26 Sep 1997 08:33:53 -0600 From: vanmaren@fast.cs.utah.edu (Kevin Van Maren) Message-Id: <199709261433.IAA12129@fast.cs.utah.edu> To: jeske@home.chat.net, vandys@cisco.com Subject: Re: VSTa native boot Cc: vsta@zendo.com > Hmm... do you think there is a need for me to fix the VSTa boot image > tools to work with this new kernel location expectation? If I fix it, then > you will also be able to create a LILO compatible boot image which can be > launched from LILO or GRUB as a single image. We use multiboot extensivly here in the Flux group. We have tools that we have developed that give us a lot of flexibility: we can take a multiboot kernel and an arbitrary number of additional files and link them with a wrapper which allows it to be booted by BSD bootblocks. We can do the same for Linux and Lilo, as well as boot from DOS. We also have a `netboot' kernel: this is a 2nd level bootloader that loads a kernel over NFS. The newest versions of these tools should be available soon. > I already have GRUB installed anyhow. Now the only complication is that > one of the nicest parts about grub is it's ability to directly read > filesystems, however, it won't be able to read VSTa FS. So that means > either a FAT based VSTa, or a FAT boot partition, or giving Grub specific > location information for booting images. I'm sure Erich isn't opposed to having Grub support VSTa FS. ;) Kevin From daemon Fri Sep 26 09:23:12 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id JAA07711 for vsta-xplod; Fri, 26 Sep 1997 09:23:12 GMT Received: from daemun.ilt.com (DAEMUN.ILT.COM [204.247.102.21]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id JAA07705 for ; Fri, 26 Sep 1997 09:23:09 GMT Received: from venus (venus [199.26.203.13]) by daemun.ilt.com (8.8.5/8.8.5) with SMTP id LAA02948 for ; Fri, 26 Sep 1997 11:50:50 -0700 (PDT) Received: by venus (4.1/ILT-1.1) id AA01051; Fri, 26 Sep 97 11:50:49 PDT Date: Fri, 26 Sep 97 11:50:49 PDT From: drgeorge@ilt.com (Dr_George_D_Detlefsen) Message-Id: <9709261850.AA01051@venus> To: vsta@zendo.com Subject: KA9Q executable Hello VSTA folks Does anyone have an executable file for the KA9Q port to VSTA ??? I have finally gotten a NE2000 card and the 'ne2000' driver seems to recognize it, but the KA9Q executable seems to be missing from the vsta_152 'tz' files. Thank you for any help Dr George D Detlefsen, InterLinear Technology, phone: 1-(510)-522-5077 1420 Harbor Bay Pkwy # 281 fax: 1-(510)-522-1228 Alameda CA 94502-6556 USA email: drgeorge@ilt.com `All progress comes thru the work of heretics` -- Dr George From daemon Fri Sep 26 09:26:52 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id JAA07724 for vsta-xplod; Fri, 26 Sep 1997 09:26:52 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id JAA07718 for ; Fri, 26 Sep 1997 09:26:50 GMT Received: from localhost (vandys@localhost) by zipper.cisco.com (8.8.4-Cisco.1/8.6.5) with SMTP id LAA17955; Fri, 26 Sep 1997 11:53:30 -0700 (PDT) Message-Id: <199709261853.LAA17955@zipper.cisco.com> To: drgeorge@ilt.com (Dr_George_D_Detlefsen) cc: vsta@zendo.com Subject: Re: KA9Q executable In-reply-to: Your message of "Fri, 26 Sep 1997 11:50:49 PDT." <9709261850.AA01051@venus> Date: Fri, 26 Sep 1997 11:53:29 -0700 From: Andrew Valencia [drgeorge@ilt.com (Dr_George_D_Detlefsen) writes:] >Hello VSTA folks > >Does anyone have an executable file for the KA9Q port to VSTA ??? I can cook one up for you tonight. >I have finally gotten a NE2000 card and the 'ne2000' driver seems >to recognize it, but the KA9Q executable seems to be missing from >the vsta_152 'tz' files. Yes, because the executable (named "net", I believe) was changing a little too often. Really, you probably want to cook your own from source, since you'll probably break it as soon as you apply a little pressure, and then you'll want to peruse the source to see why. FWIW, debugging KA9Q is a whole lot more pleasant with gdb, rather than just the "adb" from 1.5.2. Andy From daemon Mon Sep 29 10:32:19 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id KAA14669 for vsta-xplod; Mon, 29 Sep 1997 10:32:19 GMT Received: from ngcpost.ngc.com (ngcpost.ngc.com [161.69.1.1]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id KAA14663 for ; Mon, 29 Sep 1997 10:32:16 GMT Received: (from uucp@localhost) by ngcpost.ngc.com (8.8.5/8.8.5) id MAA18989 for ; Mon, 29 Sep 1997 12:59:32 -0700 (PDT) From: Michael_Paremski@NGC.COM Received: from vegas(161.69.2.3) by ngcpost via smap (V1.3) id sma018980; Mon Sep 29 12:59:01 1997 Received: from internet-mail.ngc.com by NGC.COM (4.1/SMI-4.1/Paul-NetGen-vegas-950920) id AA15563; Mon, 29 Sep 97 12:59:00 PDT Received: from ccMail by internet-mail.ngc.com (IMA Internet Exchange 2.11 Enterprise) id 0009150E; Mon, 29 Sep 1997 12:54:24 -0700 Mime-Version: 1.0 Date: Mon, 29 Sep 1997 12:53:39 -0700 Message-Id: <0009150E.1559@ngc.com> Subject: Novice questions. To: vsta@zendo.com Cc: Michael_Paremski@NGC.COM Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Description: cc:Mail note part Hi guys, I am novice in VSTa, but I have enough background in operating system kernel,and device driver programming. I am trying to build driver for 10/100Mbit DEC21140 chip based Ethernet card. Where I can get information about: 1. How small is smallest delay, which I can make by VSTa system calls. 2. How I can get chunk of memory in physical address space more than 4K. I need it to build descriptors ring and buffer pool. 3. How I can calculate physical address from virtual address. 4. Have anyone information how fast massage passing from kernel interrupt routine to driver. MikeP. michaelp@ngc.com From daemon Mon Sep 29 10:52:19 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id KAA14698 for vsta-xplod; Mon, 29 Sep 1997 10:52:19 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id KAA14692 for ; Mon, 29 Sep 1997 10:52:16 GMT Received: from localhost (vandys@localhost) by zipper.cisco.com (8.8.4-Cisco.1/8.6.5) with SMTP id NAA17988; Mon, 29 Sep 1997 13:19:37 -0700 (PDT) Message-Id: <199709292019.NAA17988@zipper.cisco.com> To: Michael_Paremski@NGC.COM cc: vsta@zendo.com Subject: Re: Novice questions. In-reply-to: Your message of "Mon, 29 Sep 1997 12:53:39 PDT." <0009150E.1559@ngc.com> Date: Mon, 29 Sep 1997 13:19:37 -0700 From: Andrew Valencia [Michael_Paremski@NGC.COM writes:] > 1. How small is smallest delay, which I can make by VSTa system calls. You mean the time to switch into the kernel from user mode? I think we had numbers ages ago, but you may as well just measure it with a current model CPU. > 2. How I can get chunk of memory in physical address space more than > 4K. I need it to build descriptors ring and buffer pool. There is no code in the distribution to do this... the initial answer is to see if your ring descriptors can accomodate scatter/gather of 4k pages. If not, I have written some contiguous allocation code which I'd be happy to supply. > 3. How I can calculate physical address from virtual address. In the kernel, vtop() does it. From user mode, see how the floppy driver sets up for a DMA transfer using page_wire()/page_release(). > 4. Have anyone information how fast massage passing from kernel > interrupt routine to driver. Again, you should probably get current numbers. It was in the single digit microsecond range, I think. Andy From daemon Mon Sep 29 22:41:16 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id WAA15489 for vsta-xplod; Mon, 29 Sep 1997 22:41:16 GMT Received: from a01-unix.gsyc.inf.uc3m.es (a01-unix.gsyc.inf.uc3m.es [163.117.137.180]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id WAA15480 for ; Mon, 29 Sep 1997 22:38:57 GMT Received: (from jcasilla@localhost) by a01-unix.gsyc.inf.uc3m.es (8.8.5/8.6.12) id JAA12573; Tue, 30 Sep 1997 09:02:35 +0100 (MET) Date: Tue, 30 Sep 1997 09:02:35 +0100 (MET) From: Juan Manuel Casillas Perez Message-Id: <199709300802.JAA12573@a01-unix.gsyc.inf.uc3m.es> Subject: Problems with 3com 3c509 net driver To: vsta@zendo.com Hello ! Well, I'm making a 3c 3c509 net driver (Just like "ne") and I have some problems with it. Can Anyone send me two clients (one to read, one to write) ? the server ? (The clients write/read ethernet pakages to the server). The clients must be very simplex. -Only the necesary - Thanks a lot, and sorry for my bad english ! P.D. Send me the clients as quick you can, because I have to expose my PFC this week. ___________________________________________________________________________ %=x%=x%=x%=x%=x%=x%=x%=x%=x%=x%=x%=x%=x%=x%=x%=x%=x%=x%=x%=x%=x%=x%=x%=x%=x ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ G R E E T I N G S ! ! ! F R O M ./ .aMMMb .dMMMb .dMMMb dMMMMMMMMb .aMMMb dMMMMb \. <_n_ dMP"dMP dMP" VP dMP" VP dMP dMP dMP dMP"dMP dMP dMP _n_> `H'\) dMMMMMP VMMMb VMMMb dMP dMP dMP dMMMMMP dMP dMP (/`H' /^> dMP dMP dP .dMP dP .dMP dMP dMP dMP dMP dMP dMP dMP <^\ ` ` dMP dMP VMMMP" VMMMP" dMP dMP dMP dMP dMP dMP dMP ' ' T H E A S S E M B L E R M A N http://a01-unix.gsyc.inf.ucm.es/~jcasilla/home2.html assman@ordago.gsyc.inf.uc3m.es | jcasilla@a01-unix.gsyc.inf.uc3m.es ___________________________________________________________________________ %=x%=x%=x%=x%=x%=x%=x%=x%=x%=x%=x%=x%=x%=x%=x%=x%=x%=x%=x%=x%=x%=x%=x%=x%=x ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ From daemon Tue Sep 30 07:05:07 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id HAA16169 for vsta-xplod; Tue, 30 Sep 1997 07:05:07 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id HAA16163 for ; Tue, 30 Sep 1997 07:05:05 GMT Received: from localhost (vandys@localhost) by zipper.cisco.com (8.8.4-Cisco.1/8.6.5) with SMTP id IAA11733; Tue, 30 Sep 1997 08:44:48 -0700 (PDT) Message-Id: <199709301544.IAA11733@zipper.cisco.com> To: Juan Manuel Casillas Perez cc: vsta@zendo.com Subject: Re: Problems with 3com 3c509 net driver In-reply-to: Your message of "Tue, 30 Sep 1997 09:02:35 BST." <199709300802.JAA12573@a01-unix.gsyc.inf.uc3m.es> Date: Tue, 30 Sep 1997 08:44:48 -0700 From: Andrew Valencia [Juan Manuel Casillas Perez writes:] > Can Anyone send me two clients (one to read, one to write) ? > the server ? (The clients write/read ethernet pakages to the server). > The clients must be very simplex. -Only the necesary - The "net" program from KA9Q actually makes a pretty nice little client for the ether driver. "ping" generates packets, and I recollect that the TRACE #ifdef of main.c gives a decent IP packet dump facility. I haven't played with the latter since I brought up the ether driver years ago, but I just riffled through the code and it looks like it ought to work. Andy From daemon Thu Oct 2 15:23:08 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id PAA19744 for vsta-xplod; Thu, 2 Oct 1997 15:23:08 GMT Received: from note.orchestra.cse.unsw.EDU.AU (root@[129.94.242.29]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id PAA19738 for ; Thu, 2 Oct 1997 15:22:58 GMT Received: From ives With LocalMail ; Fri, 3 Oct 97 10:50:48 +1000 From: s2156593@cse.unsw.edu.au (Binh Thai) To: vsta@zendo.com Date: Fri, 3 Oct 97 10:50:48 +1000 Message-Id: <971003005048.23954@cse.unsw.edu.au> Subject: What does RUN_TICKS really means? Hi all, I have been spending some time looking at the vsta scheduling, and I have a question. I wonder if someone can help me... In the file sched.c, there is a parameter called RUN_TICKS, which is set to Hz/4. I have found out from another file that Hz = 20. This makes RUN_TICKS = 5. What does that 5 means? Is it 5ms? 5 clock cycles? or 5Hz = 0.2s? Thank you. Binh Thai From daemon Thu Oct 2 17:18:44 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id RAA19915 for vsta-xplod; Thu, 2 Oct 1997 17:18:44 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id RAA19909 for ; Thu, 2 Oct 1997 17:18:41 GMT Received: from localhost (vandys@localhost) by zipper.cisco.com (8.8.4-Cisco.1/8.6.5) with SMTP id TAA15202; Thu, 2 Oct 1997 19:46:01 -0700 (PDT) Message-Id: <199710030246.TAA15202@zipper.cisco.com> To: s2156593@cse.unsw.edu.au (Binh Thai) cc: vsta@zendo.com Subject: Re: What does RUN_TICKS really means? In-reply-to: Your message of "Fri, 03 Oct 1997 10:50:48 +1000." <971003005048.23954@cse.unsw.edu.au> Date: Thu, 02 Oct 1997 19:46:01 -0700 From: Andrew Valencia [s2156593@cse.unsw.edu.au (Binh Thai) writes:] > In the file sched.c, there is a parameter called RUN_TICKS, which is >set to Hz/4. I have found out from another file that Hz = 20. This makes >RUN_TICKS = 5. What does that 5 means? Is it 5ms? 5 clock cycles? or >5Hz = 0.2s? HZ defines the rate at which the system clock ticks. Each clock tick results in hardclock() being called. hardclock(), among other things, decrements the current running thread's (if any) t_runticks counter, which starts at RUN_TICKS and counts down. When it reaches zero, the process is flagged to consider being preempted. RUN_TICKS, therefore, is in units of clock ticks. It is the granularity at which CPU bound processes will be timesliced among each other. Interactive tasks, on the other hand, generally do not run long enough to expire this count, and the "cheated" runtime queue lets such sporadic tasks receive a much more favorable scheduling treatment. You should also consider the role of the t_oink field, as it is important in correctly detecting processes which should receive this preferential treatment. Andy From daemon Sat Oct 4 17:37:33 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id RAA24972 for vsta-xplod; Sat, 4 Oct 1997 17:37:33 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id RAA24966 for ; Sat, 4 Oct 1997 17:37:30 GMT Received: from localhost (vandys@localhost) by zipper.cisco.com (8.8.4-Cisco.1/8.6.5) with SMTP id UAA09069 for ; Sat, 4 Oct 1997 20:04:26 -0700 (PDT) Message-Id: <199710050304.UAA09069@zipper.cisco.com> To: vsta@zendo.com Subject: PS/2 mouse driver for MGR on VSTa Date: Sat, 04 Oct 1997 20:04:26 -0700 From: Andrew Valencia Some of you have found that the "mouse -type ps2aux" would find your mouse device, but mouse motion would be totally erratic. I finally found a laptop that showed this behavior, and have added a "mouse -type ps2aux -bus" option, which decodes the incoming data stream correctly. The encoding looks quite a bit like a standard bus mouse (thus the -bus switch), but the device tests, buffers, and interrupts like a PS/2 type of mouse, thus I added it as a switch there. I've done some test installs of 1.6 here, and found a couple platform dependencies (somehow, depending on the BIOS, EFLAGS may or may not have the "nested task" flag, for instance), which I've cleaned up. I was hoping to do the final test and build on my PPro system, but since Aberdeen hasn't managed to return it, I may have to go ahead and do it off my Pentium system. I'll see what I can do tomorrow. Enjoy! Andy From daemon Wed Oct 8 19:19:09 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id TAA01798 for vsta-xplod; Wed, 8 Oct 1997 19:19:09 GMT Received: from baruk2.zendo.com (baruk2.zendo.com [205.187.71.65]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id TAA01792 for ; Wed, 8 Oct 1997 19:19:06 GMT Received: from localhost.zendo.com (localhost.zendo.com [127.0.0.1]) by baruk2.zendo.com (8.8.5/8.6.12) with SMTP id VAA04551 for ; Wed, 8 Oct 1997 21:42:35 -0700 (PDT) Message-Id: <199710090442.VAA04551@baruk2.zendo.com> X-Authentication-Warning: baruk2.zendo.com: localhost.zendo.com [127.0.0.1] didn't use HELO protocol To: vsta@zendo.com Subject: VSTa 1.6 available Date: Wed, 08 Oct 1997 21:42:34 -0700 From: Andy Valencia For those of you plagued with PC's which reliably boot and run their current operating system, may I introduce version 1.6? :-) The latest version of VSTa has been dropped in: ftp://ftp.zendo.com/vsta/vsta_16 for your booting pleasure. Note that this version now boots directly (rather than via DOS), although it's easy enough to set up a floppy as the boot loader, if you want to keep stakes low. Thanks to the folks in Utah for creating the Multiboot concept, and implementing the GRUB bootloader. In addition to this change in VSTa's boot, the largest changes are that a up-to-date gcc 2.X toolchain has replaced the 1.X version of 1.5.2. A matching gdb has been ported, and seems to work well. A number of previously ported commands by Dave Hudson finally make their appearance as well. The arrangement of files in vsta.tz and vsta_src.tz is still familiar, but the source to other ported commands has been somewhat rearranged. I also read through some of the doc files, and replaced some obsolete information. Enjoy! Andy Valencia From daemon Thu Oct 9 07:57:36 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id HAA02936 for vsta-xplod; Thu, 9 Oct 1997 07:57:36 GMT Received: from gateway.s3.com (gateway.s3.com [198.211.58.4]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id HAA02930 for ; Thu, 9 Oct 1997 07:57:32 GMT Received: from chat.net (home.chat.net [198.211.58.200]) by gateway.s3.com (8.6.12/8.6.12) with ESMTP id KAA07891; Thu, 9 Oct 1997 10:04:10 -0700 Received: by chat.net (SMI-8.6/SMI-SVR4) id KAA02041; Thu, 9 Oct 1997 10:26:55 -0700 Message-ID: <19971009102655.56833@home.chat.net> Date: Thu, 9 Oct 1997 10:26:55 -0700 From: David Jeske To: vsta@zendo.com Cc: fellow@iii.org.tw Subject: Documentation of a.out? Reply-To: fellow@iii.org.tw, vsta@zendo.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.79e I received this question from a VSTa inquirer and I don't have a satisfactory answer. From: Fellow Chiou To: jeske@uiuc.edu Subject: Documents of a.out! Hi! I know the execution file's format of VSTa is a.out, but I can't find any related documents or references of a.out. The only document I got was from the general UNIX by typing man a.out. Unfortunately it is too rough, could you tell me where I can find some detail informations or books of a.out. Thanks... Fellow Chiou TAIWAN From daemon Tue Oct 14 09:48:51 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id JAA14265 for vsta-xplod; Tue, 14 Oct 1997 09:48:51 GMT Received: from mail11.digital.com (mail11.digital.com [192.208.46.10]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id JAA14259 for ; Tue, 14 Oct 1997 09:48:18 GMT Received: from zeus.unx.dec.com (zeus.unx.dec.com [16.69.224.1]) by mail11.digital.com (8.7.5/UNX 1.5/1.0/WV) with SMTP id PAA28687 for ; Tue, 14 Oct 1997 15:02:41 -0400 (EDT) Received: from orion.unx.dec.com by zeus.unx.dec.com (5.65v4.0/1.1.10.5/28Jun96-0151PM) id AA08392; Tue, 14 Oct 1997 15:02:41 -0400 From: Pat Villani Received: by orion.unx.dec.com (5.65v3.2/1.1.10.5/09Apr97-0408PM) id AA07478; Tue, 14 Oct 1997 15:02:35 -0400 Date: Tue, 14 Oct 1997 15:02:35 -0400 Message-Id: <9710141902.AA07478@orion.unx.dec.com> To: vsta@zendo.com Subject: Building v1.6 Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Md5: iDZHdI5ExnW6ew1e1x6yXQ== I'm having some trouble with the latest release. Aside from a missing mkall, it looks like gcc always fails with a perm error. This error occurs in the lib directory. Permissions seem ok and relative directories are in place, i.e., ../../include exists. If I copy a small, stand alone file to /vsta/src/lib and compile it there, it compiles fine yet if I try to compile any VSTa file through the makefile or by hand I get the perm error. Can anyone help? Pat +--------------------------------+------------------------------------+ | Pat Villani | Email: patv@unx.dec.com | | UBPG/UESG | unxa::villani | | Digital Equipment Corporation | | | Manalapan, NJ | Amateur call: WB2GBF | +--------------------------------+------------------------------------+ | Disclaimer: | | All opinions, ideas and comments expressed are strictly mine and do | | not reflect on my employer. | | | | I've wrestled with reality for 35 years, doctor, and I'm happy to | | state I finally won out over it. -- Elwood P. Dowd | +---------------------------------------------------------------------+ From daemon Fri Oct 17 11:15:55 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id LAA20497 for vsta-xplod; Fri, 17 Oct 1997 11:15:55 GMT Received: from daemun.ilt.com (DAEMUN.ILT.COM [204.247.102.21]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id LAA20491 for ; Fri, 17 Oct 1997 11:15:46 GMT Received: from venus (venus [199.26.203.13]) by daemun.ilt.com (8.8.5/8.8.5) with SMTP id NAA03844 for ; Fri, 17 Oct 1997 13:44:27 -0700 (PDT) Received: by venus (4.1/ILT-1.1) id AA02361; Fri, 17 Oct 97 13:44:27 PDT Date: Fri, 17 Oct 97 13:44:27 PDT From: drgeorge@ilt.com (Dr_George_D_Detlefsen) Message-Id: <9710172044.AA02361@venus> To: vsta@zendo.com Vsta 1.6 experiences Hello Andy and Vsta Folks So far I have installed VSTA 1.6 on two computers using the FAT filesystem only, a 386sx-16 and a 486-25 laptop. I used a home grown TFTP server which I can boot from a floppy to transfer the *.tz files to the target machin from my UNIX hosts. The GRUB booting system seems to work well although it took some learning. I had problems with using large FAT file systems. The VSTA 'dos' does not seem to like FAT file systems larger than 127 mega bytes. It seems to require 2048 byte cluster sizes. With a larger FAT file system, the screen is cleared and 'init' cannot open '/vsta/etc/inittab'. There is an extra screen clear which wipes out the syslog trace. One has to have a 'fast' eye to discern that 'dos' apparently mounts the FAT filesystem, but it does not work. I have been unable to get the VFS filesystem to work using the directions in the FAQ. I made an 'Extended DOS Partition" with a D: drive. I can do the inital 'mount' and 'stat' and 'mkfs_vfs'. However the subsiquent 'vstafs' and 'mount' commands do not complain, but there is no file system. I can not 'cd' to it, or 'ls' it or copy a file to it with 'cp' Any hints. The sx386 has a NE2000 clone ethernet card. The 'ne' server finds it and does not report any errors. I edited '/vsta/etc/autoexec.net' to use local ip address, hostname, and domain. The 'net' command is run, but I cannot 'ping' the sx386 from a UNIX host or ping the UNIX host. Thanks for any enlightenment. Sincerely Dr George D Detlefsen, InterLinear Technology, phone: 1-(510)-522-5077 1420 Harbor Bay Pkwy # 281 fax: 1-(510)-522-1228 Alameda CA 94502-6556 USA email: drgeorge@ilt.com `All progress comes thru the work of heretics` -- Dr George From daemon Sun Oct 19 09:28:51 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id JAA26126 for vsta-xplod; Sun, 19 Oct 1997 09:28:51 GMT Received: from baruk2.zendo.com (baruk2.zendo.com [205.187.71.65]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id JAA26120 for ; Sun, 19 Oct 1997 09:28:47 GMT Received: from localhost.zendo.com (localhost.zendo.com [127.0.0.1]) by baruk2.zendo.com (8.8.5/8.6.12) with SMTP id LAA27647; Sun, 19 Oct 1997 11:55:25 -0700 (PDT) Message-Id: <199710191855.LAA27647@baruk2.zendo.com> X-Authentication-Warning: baruk2.zendo.com: localhost.zendo.com [127.0.0.1] didn't use HELO protocol To: Pat Villani cc: vsta@zendo.com Subject: Re: Building v1.6 In-reply-to: Your message of "Tue, 14 Oct 1997 15:02:35 EDT." <9710141902.AA07478@orion.unx.dec.com> Date: Sun, 19 Oct 1997 11:55:25 -0700 From: Andy Valencia [Pat Villani writes:] >I'm having some trouble with the latest release. Aside from a missing >mkall, it looks like gcc always fails with a perm error. This error occurs >in the lib directory. Hmmm... could you "stat /vsta/src/lib" and see what it says? >Permissions seem ok and relative directories are in place, i.e., >../../include exists. If I copy a small, stand alone file to /vsta/src/lib >and compile it there, it compiles fine yet if I try to compile any VSTa >file through the makefile or by hand I get the perm error. Can you use "gcc -v" and see more about exactly where it's dying? I've now reproduced a misbehavior where "tar" somehow ends up with a unwritable chmod mask when creating directories. I hadn't brought along the sources to tar on the laptop at hand, so I'll nail this now that I'm back from vacation. For now you might cd to /vsta and do a "rh -e '((mode & IFMT) == IFDIR) && !(mode & 0222)' to see what directories have no write access, then do a chmod 777 to make them writable. Andy From daemon Sun Oct 19 09:32:13 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id JAA26145 for vsta-xplod; Sun, 19 Oct 1997 09:32:13 GMT Received: from baruk2.zendo.com (baruk2.zendo.com [205.187.71.65]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id JAA26139 for ; Sun, 19 Oct 1997 09:32:10 GMT Received: from localhost.zendo.com (localhost.zendo.com [127.0.0.1]) by baruk2.zendo.com (8.8.5/8.6.12) with SMTP id MAA27668; Sun, 19 Oct 1997 12:00:02 -0700 (PDT) Message-Id: <199710191900.MAA27668@baruk2.zendo.com> X-Authentication-Warning: baruk2.zendo.com: localhost.zendo.com [127.0.0.1] didn't use HELO protocol To: drgeorge@ilt.com (Dr_George_D_Detlefsen) cc: vsta@zendo.com In-reply-to: Your message of "Fri, 17 Oct 1997 13:44:27 PDT." <9710172044.AA02361@venus> Date: Sun, 19 Oct 1997 12:00:02 -0700 From: Andy Valencia [drgeorge@ilt.com (Dr_George_D_Detlefsen) writes:] >I had problems with using large FAT file systems. The VSTA 'dos' does >not seem to like FAT file systems larger than 127 mega bytes. It seems >to require 2048 byte cluster sizes. With a larger FAT file system, >the screen is cleared and 'init' cannot open '/vsta/etc/inittab'. There >is an extra screen clear which wipes out the syslog trace. One has >to have a 'fast' eye to discern that 'dos' apparently mounts the >FAT filesystem, but it does not work. Can you check to see if it's FAT-32? I know for a fact there's code to be added to the DOS server before that format will work. I use DOS partitions much greater than 127 megs, but I bet a FAT-32 type of system will default to FAT-32 at around that size, and that will break the DOS server. >I have been unable to get the VFS filesystem to work using the directions >in the FAQ. I made an 'Extended DOS Partition" with a D: drive. I can >do the inital 'mount' and 'stat' and 'mkfs_vfs'. However the >subsiquent 'vstafs' and 'mount' commands do not complain, but >there is no file system. I can not 'cd' to it, or 'ls' it or copy a file >to it with 'cp' Check to see if something's really mounted there with "fstab". It should show a mount entry. Kill the vstafs, and now fstab should report a port number of -1 instead. This is an easy way to ensure the server you expect is mounted in your namespace where expected. >The sx386 has a NE2000 clone ethernet card. The 'ne' server finds it >and does not report any errors. I edited '/vsta/etc/autoexec.net' to >use local ip address, hostname, and domain. The 'net' command is >run, but I cannot 'ping' the sx386 from a UNIX host or ping the UNIX >host. Try pinging the host from VSTa, and use "netstat -i" from the UNIX side to see if the packets are arriving, and are being returned. I think "netstat -p icmp" should tell you if it thinks they are ICMP echoes it sees, otherwise "netstat -p ip" might show invalid checksum packets, or something along those lines. Andy From daemon Sun Oct 19 22:17:08 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id WAA27142 for vsta-xplod; Sun, 19 Oct 1997 22:17:08 GMT Received: from nenuphar.saclay.cea.fr (nenuphar.saclay.cea.fr [132.166.192.7]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id WAA27136 for ; Sun, 19 Oct 1997 22:16:46 GMT Received: from vega.serma.cea.fr (vega.serma.cea.fr [132.166.149.56]) by nenuphar.saclay.cea.fr (8.8.5/ CEANET-ROUTER-3.0) with ESMTP id JAA06291 for ; Mon, 20 Oct 1997 09:44:58 +0200 (MET DST) Received: (from basile@localhost) by vega.serma.cea.fr (8.8.6/8.8.5) id JAA15101; Mon, 20 Oct 1997 09:46:23 +0200 Date: Mon, 20 Oct 1997 09:46:23 +0200 Message-Id: <199710200746.JAA15101@vega.serma.cea.fr> From: Basile STARYNKEVITCH MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: vsta@zendo.com Subject: Q: mirrors of Vsta 1.6? X-Mailer: VM 6.31 under 20.2 XEmacs Lucid Hello all, The official Vsta ftp site is awefully slow (<0.1Kb/sec) from Paris suburb, France (although I do have a good Internet connection, even to several US sites). Is there an up to date Vsta ftp mirror site somewhere in Europa. (ftp.lip6.fr:/pub/vsta is no more updated; it was previously known as ftp.ibp.fr:/pub/vsta) Thanks N.B. Any opinions expressed here are solely mine, and not of my organization. N.B. Les opinions exprimees ici me sont personnelles et n engagent pas le CEA. ---------------------------------------------------------------------- Basile STARYNKEVITCH ---- Commissariat à l Energie Atomique DRN/DMT/SERMA * CEA/Saclay bat.470 * 91191 GIF/YVETTE CEDEX * France fax: (33) 01,69.08.85.68; phone: 01,69.08.40.66; home: 01,46.65.45.53 email: Basile . Starynkevitch @ cea . fr (but remove white space) I speak french, english, russian. Je parle français, anglais, russe. ---------------------------------------------------------------------- From daemon Mon Oct 20 08:37:13 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id IAA00170 for vsta-xplod; Mon, 20 Oct 1997 08:37:13 GMT Received: from mail12.digital.com (mail12.digital.com [192.208.46.20]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id IAA00164; Mon, 20 Oct 1997 08:36:51 GMT Received: from zeus.unx.dec.com (zeus.unx.dec.com [16.69.224.1]) by mail12.digital.com (8.7.5/UNX 1.5/1.0/WV) with SMTP id NAA08302; Mon, 20 Oct 1997 13:56:18 -0400 (EDT) Received: from orion.unx.dec.com by zeus.unx.dec.com (5.65v4.0/1.1.10.5/28Jun96-0151PM) id AA01420; Mon, 20 Oct 1997 13:56:17 -0400 From: Pat Villani Received: by orion.unx.dec.com (5.65v3.2/1.1.10.5/09Apr97-0408PM) id AA20810; Mon, 20 Oct 1997 13:56:03 -0400 Date: Mon, 20 Oct 1997 13:56:03 -0400 Message-Id: <9710201756.AA20810@orion.unx.dec.com> To: vandys@zendo.com Subject: Re: Building v1.6 Cc: vsta@zendo.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Md5: e+GknUccjqM5puNRgzUPDA== > Hmmm... could you "stat /vsta/src/lib" and see what it says? Logged in as root: vsta$ stat /vsta/src/lib perm=1/1 acc=5/0/2 size=4096 type=d owner=0 inode=1073769920 mtime=244912628 > Can you use "gcc -v" and see more about exactly where it's dying? It dies in cc1. > For now you might cd to /vsta and do a "rh -e '((mode & IFMT) == IFDIR) && > !(mode & 0222)' to see what directories have no write access, then do > a chmod 777 to make them writable. The list comes back empty. What does rh do? I've never encountered it on any unix system that I've worked with. I only seem to have /vsta/bin/rh and no man pages or source. Pat +--------------------------------+------------------------------------+ | Pat Villani | Email: patv@unx.dec.com | | UBPG/UESG | unxa::villani | | Digital Equipment Corporation | | | Manalapan, NJ | Amateur call: WB2GBF | +--------------------------------+------------------------------------+ | Disclaimer: | | All opinions, ideas and comments expressed are strictly mine and do | | not reflect on my employer. | | | | I've wrestled with reality for 35 years, doctor, and I'm happy to | | state I finally won out over it. -- Elwood P. Dowd | +---------------------------------------------------------------------+ From daemon Mon Oct 20 10:04:06 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id KAA00295 for vsta-xplod; Mon, 20 Oct 1997 10:04:06 GMT Received: from amri.cisco.com (amri.cisco.com [198.92.29.50]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id KAA00289 for ; Mon, 20 Oct 1997 10:04:04 GMT Received: from amri.cisco.com (localhost.cisco.com [127.0.0.1]) by amri.cisco.com (8.8.5/8.8.5) with ESMTP id NAA03079; Mon, 20 Oct 1997 13:06:20 -0700 (PDT) Message-Id: <199710202006.NAA03079@amri.cisco.com> To: Pat Villani cc: vsta@zendo.com Subject: Re: Building v1.6 In-reply-to: Your message of "Mon, 20 Oct 1997 13:56:03 EDT." <9710201756.AA20810@orion.unx.dec.com> Date: Mon, 20 Oct 1997 13:06:20 -0700 From: Andy Valencia [Pat Villani writes:] >> Can you use "gcc -v" and see more about exactly where it's dying? >It dies in cc1. What're the error messages? Does it look like the /tmp file? You should be able to reproduce this by typing in the compiler pass invocations manually, then play around with the input files and what-not. >> For now you might cd to /vsta and do a "rh -e '((mode & IFMT) == IFDIR) && >> !(mode & 0222)' to see what directories have no write access, then do >> a chmod 777 to make them writable. >The list comes back empty. What does rh do? I've never encountered it on >any unix system that I've worked with. I only seem to have /vsta/bin/rh and >no man pages or source. It's in the text.tz archive. It's basically "find", except with C syntax rather than find's clunky command line. Your directories look fine, so something else is going on. For instance, does your CPU have a floating point unit? Andy From daemon Tue Oct 21 07:11:37 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id HAA02195 for vsta-xplod; Tue, 21 Oct 1997 07:11:37 GMT Received: from mail11.digital.com (mail11.digital.com [192.208.46.10]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id HAA02189 for ; Tue, 21 Oct 1997 07:11:33 GMT Received: from zeus.unx.dec.com (zeus.unx.dec.com [16.69.224.1]) by mail11.digital.com (8.7.5/UNX 1.5/1.0/WV) with ESMTP id MAA11537; Tue, 21 Oct 1997 12:41:48 -0400 (EDT) Received: from orion.unx.dec.com by zeus.unx.dec.com (8.7.6/UNX 1.7/1.1.10.5/28Jun96-0151PM) id MAA0000016146; Tue, 21 Oct 1997 12:41:47 -0400 (EDT) From: Pat Villani Received: by orion.unx.dec.com (5.65v3.2/1.1.10.5/09Apr97-0408PM) id AA25107; Tue, 21 Oct 1997 12:41:46 -0400 Date: Tue, 21 Oct 1997 12:41:46 -0400 Message-Id: <9710211641.AA25107@orion.unx.dec.com> To: vandys@cisco.com Subject: Re: Building v1.6 Cc: vsta@zendo.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Md5: rxBQ5II0ESjwBu0e9wGdBQ== > For instance, does your CPU have a floating > point unit? Bingo! The laptop I'm using is an old NCR 3150 with _no_ co-processor. I've been using it for so long I forgot that it was just an old 486SX25. Since the kernel doesn't seem to require fp, why the limitation starting with 1.6. Is it the tools? Pat +--------------------------------+------------------------------------+ | Pat Villani | Email: patv@unx.dec.com | | UBPG/UESG | unxa::villani | | Digital Equipment Corporation | | | Manalapan, NJ | Amateur call: WB2GBF | +--------------------------------+------------------------------------+ | Disclaimer: | | All opinions, ideas and comments expressed are strictly mine and do | | not reflect on my employer. | | | | I've wrestled with reality for 35 years, doctor, and I'm happy to | | state I finally won out over it. -- Elwood P. Dowd | +---------------------------------------------------------------------+ From daemon Tue Oct 21 07:15:49 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id HAA02212 for vsta-xplod; Tue, 21 Oct 1997 07:15:49 GMT Received: from amri.cisco.com (amri.cisco.com [198.92.29.50]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id HAA02206 for ; Tue, 21 Oct 1997 07:15:46 GMT Received: from amri.cisco.com (localhost.cisco.com [127.0.0.1]) by amri.cisco.com (8.8.5/8.8.5) with ESMTP id KAA05791; Tue, 21 Oct 1997 10:18:06 -0700 (PDT) Message-Id: <199710211718.KAA05791@amri.cisco.com> To: Pat Villani cc: vsta@zendo.com Subject: Re: Building v1.6 In-reply-to: Your message of "Tue, 21 Oct 1997 12:41:46 EDT." <9710211641.AA25107@orion.unx.dec.com> Date: Tue, 21 Oct 1997 10:18:06 -0700 From: Andy Valencia [Pat Villani writes:] >> For instance, does your CPU have a floating >> point unit? >Bingo! The laptop I'm using is an old NCR 3150 with _no_ co-processor. >I've been using it for so long I forgot that it was just an old 486SX25. >Since the kernel doesn't seem to require fp, why the limitation starting >with 1.6. Is it the tools? GCC has this really lame register allocation algorithm which uses, of all things, floating point. djgpp of old, as well as the GCC 1.X port to VSTa, had a hack to work around it. With the conversion to 2.X GCC, I stopped hacking up the source. If somebody's game to hunt down this stuff and fix it again, I'd be happy to fold it back in. Andy From daemon Tue Oct 21 08:28:35 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id IAA02361 for vsta-xplod; Tue, 21 Oct 1997 08:28:35 GMT Received: from amri.cisco.com (amri.cisco.com [198.92.29.50]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id IAA02355 for ; Tue, 21 Oct 1997 08:28:33 GMT Received: from amri.cisco.com (localhost.cisco.com [127.0.0.1]) by amri.cisco.com (8.8.5/8.8.5) with ESMTP id LAA06066; Tue, 21 Oct 1997 11:30:53 -0700 (PDT) Message-Id: <199710211830.LAA06066@amri.cisco.com> To: Pat Villani cc: vsta@zendo.com Subject: Re: Building v1.6 In-reply-to: Your message of "Tue, 21 Oct 1997 14:01:38 EDT." <9710211801.AA25074@orion.unx.dec.com> Date: Tue, 21 Oct 1997 11:30:53 -0700 From: Andy Valencia [Pat Villani writes:] >> ... If somebody's game to hunt down this stuff and fix >> it again, I'd be happy to fold it back in. >It's worth taking a look at. BTW -- is VSTa still "buildable" via djgpp? At least one fellow is building parts of it via djgpp... the current djgpp defaults to ELF, but it's basically a GNU ld which can be switched back to a.out. I'm not aware of any other particular problems using cross tools--just make sure you pick the right text/data addresses. For kernel this is text @ 1 meg, data immediately following. For user programs, text is at 0x1020 (a.out header at 0x1000), with data at first 4 meg boundary above and of text. 2.x GNU ld with its load maps makes this easy to do. Andy From daemon Tue Oct 21 08:24:45 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id IAA02339 for vsta-xplod; Tue, 21 Oct 1997 08:24:45 GMT Received: from mail11.digital.com (mail11.digital.com [192.208.46.10]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id IAA02333 for ; Tue, 21 Oct 1997 08:24:43 GMT Received: from zeus.unx.dec.com (zeus.unx.dec.com [16.69.224.1]) by mail11.digital.com (8.7.5/UNX 1.5/1.0/WV) with ESMTP id OAA24927; Tue, 21 Oct 1997 14:01:39 -0400 (EDT) Received: from orion.unx.dec.com by zeus.unx.dec.com (8.7.6/UNX 1.7/1.1.10.5/28Jun96-0151PM) id OAA0000003293; Tue, 21 Oct 1997 14:01:38 -0400 (EDT) From: Pat Villani Received: by orion.unx.dec.com (5.65v3.2/1.1.10.5/09Apr97-0408PM) id AA25074; Tue, 21 Oct 1997 14:01:38 -0400 Date: Tue, 21 Oct 1997 14:01:38 -0400 Message-Id: <9710211801.AA25074@orion.unx.dec.com> To: vandys@cisco.com Subject: Re: Building v1.6 Cc: vsta@zendo.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Md5: 1bXcM1QN2DLsp/t1l3SOLQ== > ... If somebody's game to hunt down this stuff and fix > it again, I'd be happy to fold it back in. It's worth taking a look at. BTW -- is VSTa still "buildable" via djgpp? Pat +--------------------------------+------------------------------------+ | Pat Villani | Email: patv@unx.dec.com | | UBPG/UESG | unxa::villani | | Digital Equipment Corporation | | | Manalapan, NJ | Amateur call: WB2GBF | +--------------------------------+------------------------------------+ | Disclaimer: | | All opinions, ideas and comments expressed are strictly mine and do | | not reflect on my employer. | | | | I've wrestled with reality for 35 years, doctor, and I'm happy to | | state I finally won out over it. -- Elwood P. Dowd | +---------------------------------------------------------------------+ From daemon Wed Oct 29 06:39:40 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id GAA01937 for vsta-xplod; Wed, 29 Oct 1997 06:39:40 GMT Received: from amri.cisco.com (amri.cisco.com [198.92.29.50]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id GAA01931 for ; Wed, 29 Oct 1997 06:39:34 GMT Received: from amri.cisco.com (localhost.cisco.com [127.0.0.1]) by amri.cisco.com (8.8.5/8.8.5) with ESMTP id IAA27619; Wed, 29 Oct 1997 08:38:15 -0800 (PST) Message-Id: <199710291638.IAA27619@amri.cisco.com> To: Pavel Machek cc: vsta@zendo.com Subject: Re: 1.5.2 bugs In-reply-to: Your message of "Wed, 29 Oct 1997 15:50:46 +0100." <19971029155046.50954@atrey.karlin.mff.cuni.cz> Date: Wed, 29 Oct 1997 08:38:14 -0800 From: Andy Valencia [Pavel Machek writes:] >* shift tracking in cons2 is broken: Press and hold left ctrl, click >right ctrl, press any key - ctrl will be forgotten. Well, correcting >this would require tracking of which keys are done - which would make >it a lot more complex. True. As you noted, perhaps not really worth fixing. If you do fix it and submit patches, I'll roll it in. >* boot.exe does not honour 3rd argument correctly (simple typo) >* vsta will not boot from Windows 95 unless Safe mode command prompt >was selected. (I'm talking about plain dos - real mode) 1.6 moves away from the DOS boot loader, so these are basically fixed by supporting Multiboot standard and booting directly. >Is there bash for vsta? No, but note that canonical input automatically provides command line editing (using the getline() library), so ash is surprisingly usable. getline() has a hook for doing completion, and file name completion is the thing I miss most. >Is there ext2 fs support (even read only is ok) for vsta? Yes; search through www.zendo.com/vsta >Is there someone on this mailing list? :-) About 150 entries at last count. But we're a quiet bunch. :-) Andy From daemon Thu Oct 30 01:33:25 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id BAA03521 for vsta-xplod; Thu, 30 Oct 1997 01:33:25 GMT Received: from atrey.karlin.mff.cuni.cz (pavel@atrey.karlin.mff.cuni.cz [195.113.31.123]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id BAA03515 for ; Thu, 30 Oct 1997 01:33:22 GMT Received: (from pavel@localhost) by atrey.karlin.mff.cuni.cz (8.8.5/8.8.5) id MAA09961; Thu, 30 Oct 1997 12:32:15 +0100 Message-ID: <19971030123215.64450@atrey.karlin.mff.cuni.cz> Date: Thu, 30 Oct 1997 12:32:15 +0100 From: Pavel Machek To: vsta@zendo.com Subject: PID==0 problems Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.84 X-Warning: Not only using Windows can be dangerous to your mental health. Hi! At least under vsta 1.5.2, 0 is perfectly valid PID. There's trouble with this pid, however - with notify, it means 'send notification to itself'. As a result, process with pid==0 is unkillable. (In my case it was console driver). I removed this concrete problem by setting static variable 'last_pid' to 1, but in case PID's can wrap (can they?) this is not a solution. Pavel -- -- This is my little buggy signature... Pavel GCM d? s-: !g p?:+ au- a--@ w+ v- C++@ UL+++ L++ N++ E++ W--- M- Y- R+ From daemon Thu Oct 30 05:34:02 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id FAA03957 for vsta-xplod; Thu, 30 Oct 1997 05:34:02 GMT Received: from amri.cisco.com (amri.cisco.com [198.92.29.50]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id FAA03951 for ; Thu, 30 Oct 1997 05:33:51 GMT Received: from amri.cisco.com (localhost.cisco.com [127.0.0.1]) by amri.cisco.com (8.8.5/8.8.5) with ESMTP id HAA01182; Thu, 30 Oct 1997 07:32:25 -0800 (PST) Message-Id: <199710301532.HAA01182@amri.cisco.com> To: Pavel Machek cc: vsta@zendo.com Subject: Re: PID==0 problems In-reply-to: Your message of "Thu, 30 Oct 1997 12:32:15 +0100." <19971030123215.64450@atrey.karlin.mff.cuni.cz> Date: Thu, 30 Oct 1997 07:32:25 -0800 From: Andy Valencia [Pavel Machek writes:] >At least under vsta 1.5.2, 0 is perfectly valid PID. There's trouble >with this pid, however - with notify, it means 'send notification to >itself'. As a result, process with pid==0 is unkillable. (In my case >it was console driver). I removed this concrete problem by setting >static variable 'last_pid' to 1, but in case PID's can wrap (can >they?) this is not a solution. I guess in almost all cases, not killing your console is a feature. :-) Since 0 is a magic value, I believe you're right; we should start carving PID's at 1. I believe on wrap the scan starts above 0 (200, if memory serves), so that case is already covered. Note that the number of free PID's, which now starts at 2^32-1, must now be set to 2^32-2. Andy From daemon Thu Oct 30 15:16:31 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id PAA04610 for vsta-xplod; Thu, 30 Oct 1997 15:16:31 GMT Received: from baruk2.zendo.com (baruk2.zendo.com [205.187.71.65]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id PAA04604 for ; Thu, 30 Oct 1997 15:16:26 GMT Received: (from root@localhost) by baruk2.zendo.com (8.8.5/8.6.12) id RAA20759; Thu, 30 Oct 1997 17:20:07 -0800 (PST) Date: Thu, 30 Oct 1997 17:20:07 -0800 (PST) Message-Id: <199710310120.RAA20759@baruk2.zendo.com> To: vsta@zendo.com From: Mike Larson USG Subject: SPAMCHECK Re: VSTa 1.6 available Cc: larz@ultranet.com Hi, Andy wrote: > ... for your booting pleasure. Note that this version now boots directly > (rather than via DOS), although it's easy enough to set up a floppy as > the boot loader, if you want to keep stakes low. I'd like to setup vsta to boot off a floppy first, and perhaps later I will install grub on the appropriate hard drive. Unfortunately, I haven't figured out how make a grub boot floppy yet. I got as far as rawrite'ing grub.raw to a floppy and booting grub from the floppy. Is there a way to build the floppy so that it has menu.lst and all the other necessary boot files? Thanks, Mike Larson larz@ultranet.com From daemon Fri Oct 31 06:18:05 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id GAA05606 for vsta-xplod; Fri, 31 Oct 1997 06:18:05 GMT Received: from amri.cisco.com (amri.cisco.com [198.92.29.50]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id GAA05600 for ; Fri, 31 Oct 1997 06:18:01 GMT Received: from amri.cisco.com (localhost.cisco.com [127.0.0.1]) by amri.cisco.com (8.8.5/8.8.5) with ESMTP id IAA01418; Fri, 31 Oct 1997 08:16:25 -0800 (PST) Message-Id: <199710311616.IAA01418@amri.cisco.com> To: Mike Larson USG cc: vsta@zendo.com, larz@ultranet.com Subject: Re: VSTa 1.6 available In-reply-to: Your message of "Thu, 30 Oct 1997 17:20:07 PST." <199710310120.RAA20759@baruk2.zendo.com> Date: Fri, 31 Oct 1997 08:16:24 -0800 From: Andy Valencia [Mike Larson USG writes:] >I'd like to setup vsta to boot off a floppy first, and perhaps later I >will install grub on the appropriate hard drive. Unfortunately, I >haven't figured out how make a grub boot floppy yet. I got as far as >rawrite'ing grub.raw to a floppy and booting grub from the floppy. Is >there a way to build the floppy so that it has menu.lst and all the >other necessary boot files? I don't know about that, but I thought the instructions included a way to create a boot floppy which pulled the secondary loader as well as the menu.lst from the hard disk. You have to keep a couple files on your hard disk, but since grub supports most filesystems that doesn't seem like an undue hardship. And it spares you having to mess with your current hard disk boot mechanism. Andy From daemon Sat Nov 1 01:22:57 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id BAA06716 for vsta-xplod; Sat, 1 Nov 1997 01:22:57 GMT Received: from atrey.karlin.mff.cuni.cz (root@atrey.karlin.mff.cuni.cz [195.113.31.123]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id BAA06710 for ; Sat, 1 Nov 1997 01:22:41 GMT Received: from Elf.mj.gts.cz (root@slip8.ms.mff.cuni.cz [195.113.20.208]) by atrey.karlin.mff.cuni.cz (8.8.5/8.8.5) with ESMTP id MAA17146; Sat, 1 Nov 1997 12:23:05 +0100 Received: (from pavel@localhost) by Elf.mj.gts.cz (8.8.5/8.8.5) id UAA00448; Fri, 31 Oct 1997 20:32:41 +0100 Message-ID: <19971031203240.57296@Elf.mj.gts.cz> Date: Fri, 31 Oct 1997 20:32:40 +0100 From: Pavel Machek To: Andy Valencia Cc: vsta@zendo.com Subject: Re: PID==0 problems References: <19971030123215.64450@atrey.karlin.mff.cuni.cz> <199710301532.HAA01182@amri.cisco.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.81 In-Reply-To: <199710301532.HAA01182@amri.cisco.com>; from Andy Valencia on Thu, Oct 30, 1997 at 07:32:25AM -0800 X-Warning: Not only using Windows can be dangerous to your mental health. Hi! > I guess in almost all cases, not killing your console is a > feature. :-) Well - no if you are trying your own great new console and can not figure out why the old one does not die. > Since 0 is a magic value, I believe you're right; we should start carving > PID's at 1. I believe on wrap the scan starts above 0 (200, if memory > serves), so that case is already covered. Note that the number of free > PID's, which now starts at 2^32-1, must now be set to 2^32-2. Ok, here's the diff. It is rather big diff, as everything that bitten me at the first pass. So I a) Added '?' command to kernel debugger b) Added Meta key support to console (right thing as it allows you to move by whole words in editors and shell) c) Shortened console quite a bit by making Fxx code a bit wiser d) [not sure this is right thing!!!] made arrows send emacs-style escape sequences instead of vt100s proper arrow ones. I know I should fix applications, not the console driver, but this is sooooo much easier. e) Small modification to wd driver - 5 sec just is not enough. (This error did not actually bite me) Diff is against 1.5.2, but I *think* (from a look at 1.5.2 to 1.6 diff) that it should patch cleanly to 1.6. Note that you may to kill modifications to arrows as they are pretty emacs-centric. Pavel diff -ur vsta.old/src/os/dbg/dbgmain.c /c/vsta/src/os/dbg/dbgmain.c --- vsta.old/src/os/dbg/dbgmain.c Sun Oct 8 21:49:46 1995 +++ /c/vsta/src/os/dbg/dbgmain.c Mon Oct 27 20:05:34 1997 @@ -8,7 +8,7 @@ extern void dump_phys(), dump_virt(), dump_procs(), dump_pset(), dump_instr(), trace(), trapframe(), dump_vas(), dump_port(), dump_pview(), dump_thread(), dump_ref(), reboot(), memleaks(), - dump_sysmsg(), dump_core(); + dump_sysmsg(), dump_core(), help(); extern void dbg_inport(), dbg_outport(); static void quit(), calc(), set(), set_mem(); extern int get_num(); @@ -46,6 +46,7 @@ "trace", trace, "vas", dump_vas, "writemem", set_mem, + "?", help, 0, 0 }; @@ -137,6 +138,20 @@ x = get_num(str); printf("%s 0x%x %d\n", symloc(x), x, x); } +/* + * help() + * List all commands + */ +static void +help(void) +{ + int x; + + printf( "Available commands: " ); + for (x = 0; cmdtab[x].c_name; ++x) + printf( "%s ", cmdtab[x].c_name ); + printf( "\n" ); +} /* * do_cmd() @@ -161,7 +176,7 @@ } } if (matches == 0) { - printf("No such command\n"); + printf("No such command (type ? for help)\n"); return; } if (matches > 1) { diff -ur vsta.old/src/os/kern/proc.c /c/vsta/src/os/kern/proc.c --- vsta.old/src/os/kern/proc.c Mon Apr 8 18:31:30 1996 +++ /c/vsta/src/os/kern/proc.c Mon Oct 27 13:29:40 1997 @@ -26,7 +26,8 @@ extern lock_t runq_lock; -ulong npid_free = (ulong)-1; /* # PIDs free in pool */ +ulong npid_free = (ulong)-2; /* # PIDs free in pool */ -ulong pid_nextfree = 0L; /* Next free PID number */ +ulong pid_nextfree = 1L; /* Next free PID number */ + /* We got to use 1 or process #0 is unkillable */ struct proc *allprocs = 0; /* List of all procs */ sema_t pid_sema; /* Mutex for PID pool and proc lists */ diff -ur vsta.old/src/srv/mach/cons2/isr.c /c/vsta/src/srv/mach/cons2/isr.c --- vsta.old/src/srv/mach/cons2/isr.c Tue Jan 30 08:32:36 1996 +++ /c/vsta/src/srv/mach/cons2/isr.c Tue Oct 28 20:30:12 1997 @@ -12,6 +12,9 @@ capstoggle = 0, /* For toggling effect of CAPS */ numtoggle = 0, /* ...NUM lock */ isE0 = 0; /* Prefix for extended keys (FN1, etc.) */ + +#define C( x ) x & 0x1f /* Turn key into ctrl-one */ +#define M( x ) x | 0x80 /* Turn key into meta key, do not use for 0 */ /* Map scan codes to ASCII, one table for normal, one for shifted */ static char normal[] = { @@ -82,11 +85,19 @@ #endif /* + * Meta keys are badly needed - if Alt is pressed, + * it sends ESC before actual character. + */ + + if (alt || (ch&0x80)) + kbd_enqueue(s, 033); + + /* * Hand off straight data now. The keyboard always enters * data for the virtual screen currently being displayed * on the hardware screen. */ - kbd_enqueue(s, ch); + kbd_enqueue(s, ch&0x7f); } /* @@ -113,26 +124,29 @@ static int cursor_key(struct screen *s, uchar c) { - char *cp; + char buf[5]; + char *cp = buf; + cp[0] = ' '; + cp[1] = 0; switch (c) { case 72: /* up */ - cp = "\033OA"; + cp[0] = C('p'); break; case 80: /* down */ - cp = "\033OB"; + cp[0] = C('n'); break; case 77: /* right */ - cp = "\033OC"; + cp[0] = C('f'); break; case 75: /* left */ - cp = "\033OD"; + cp[0] = C('b'); break; case 73: /* pg up */ - cp = "\033[5~"; + cp = "\033v"; break; case 81: /* pg down */ - cp = "\033[6~"; + cp[0] = C('v'); break; case 82: /* insert */ cp = "\033[2~"; @@ -172,46 +186,11 @@ { char *p; - switch (c) { - case 59: /* F1 */ - p = "\033OP"; - break; - case 60: /* F2 */ - p = "\033OQ"; - break; - case 61: /* F3 */ - p = "\033OR"; - break; - case 62: /* F4 */ - p = "\033OS"; - break; - case 63: /* F5 */ - p = "\033OT"; - break; - case 64: /* F6 */ - p = "\033OU"; - break; - case 65: /* F7 */ - p = "\033OV"; - break; - case 66: /* F8 */ - p = "\033OW"; - break; - case 67: /* F9 */ - p = "\033OX"; - break; - case 68: /* F10 */ - case 87: /* F11 */ - case 88: /* F12 */ - p = 0; - break; - default: - return 0; - } - if (p) { - enqueue_string(s, p); - } - return(1); + if ((c<59) || (c>68)) return 0; + p = "\033Q?"; + p[2] = 'P'+c-59; + enqueue_string(s, p); + return 1; } /* diff -ur vsta.old/src/srv/mach/wd/wd.c /c/vsta/src/srv/mach/wd/wd.c --- vsta.old/src/srv/mach/wd/wd.c Thu Oct 12 07:09:46 1995 +++ /c/vsta/src/srv/mach/wd/wd.c Wed Oct 29 21:37:52 1997 @@ -376,10 +376,13 @@ * For laptops with power management, our inportb() here can * cause the disk to wake up; the controller can then flag * a busy disk until the spinup is complete. Give it 5 seconds. + * + * I feel that 5 seconds is not enough. (My drive takes about + * 6, my other drive is even slower). */ cyl = 0; while (inportb(base + WD_STATUS) & WDS_BUSY) { - if (++cyl > 50) { + if (++cyl > 150) { ASSERT(0, "wd_start: busy"); } __msleep(100); -- I'm really pavel@atrey.karlin.mff.cuni.cz. Pavel Look at http://atrey.karlin.mff.cuni.cz/~pavel/ ;-). From daemon Sat Nov 1 06:02:09 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id GAA09330 for vsta-xplod; Sat, 1 Nov 1997 06:02:09 GMT Received: from atrey.karlin.mff.cuni.cz (pavel@atrey.karlin.mff.cuni.cz [195.113.31.123]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id GAA09324 for ; Sat, 1 Nov 1997 06:02:05 GMT Received: (from pavel@localhost) by atrey.karlin.mff.cuni.cz (8.8.5/8.8.5) id RAA23535; Sat, 1 Nov 1997 17:02:31 +0100 Message-ID: <19971101170231.32381@atrey.karlin.mff.cuni.cz> Date: Sat, 1 Nov 1997 17:02:31 +0100 From: Pavel Machek To: vsta@zendo.com Subject: Cross-compiling for vsta Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.84 X-Warning: Not only using Windows can be dangerous to your mental health. Hi! I would like to ask if some cross-compiler compiling for vsta running under linux is available. I tried to port bash to vsta, but bash requires yacc or bison for its compilation and I just do not have those two ported. Seemed to me like cross-compilation could be pretty easy solution... Pavel -- This is my little buggy signature... Pavel GCM d? s-: !g p?:+ au- a--@ w+ v- C++@ UL+++ L++ N++ E++ W--- M- Y- R+ From daemon Sat Nov 1 06:14:38 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id GAA09358 for vsta-xplod; Sat, 1 Nov 1997 06:14:38 GMT Received: from amri.cisco.com (amri.cisco.com [198.92.29.50]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id GAA09352 for ; Sat, 1 Nov 1997 06:14:36 GMT Received: from amri.cisco.com (localhost.cisco.com [127.0.0.1]) by amri.cisco.com (8.8.5/8.8.5) with ESMTP id IAA08173; Sat, 1 Nov 1997 08:14:21 -0800 (PST) Message-Id: <199711011614.IAA08173@amri.cisco.com> To: Pavel Machek cc: vsta@zendo.com Subject: Re: Cross-compiling for vsta In-reply-to: Your message of "Sat, 01 Nov 1997 17:02:31 +0100." <19971101170231.32381@atrey.karlin.mff.cuni.cz> Date: Sat, 01 Nov 1997 08:14:21 -0800 From: Andy Valencia [Pavel Machek writes:] >I would like to ask if some cross-compiler compiling for vsta running >under linux is available. I tried to port bash to vsta, but bash >requires yacc or bison for its compilation and I just do not have >those two ported. They're in 1.6. :-) >Seemed to me like cross-compilation could be pretty >easy solution... Either way. Many people have successfully cross-compiled; you just need to build cross tools that generate a.out. Cygnus was kind enough to include some VSTa support in the core language tools. Andy From daemon Tue Nov 4 22:03:06 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id WAA15157 for vsta-xplod; Tue, 4 Nov 1997 22:03:06 GMT Received: from envy.ugcs.caltech.edu (root@envy.ugcs.caltech.edu [131.215.128.135]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id WAA15151 for ; Tue, 4 Nov 1997 22:03:03 GMT Received: from tammananny.tiger (ppp68.humboldt1.com [206.13.45.168]) by envy.ugcs.caltech.edu with ESMTP (8.8.5/UGCS:4.44) id AAA28741 for ; Wed, 5 Nov 1997 00:03:34 -0800 (PST) Recieved: from tammananny.tiger (tammananny.tigertammananny.tiger [127.0.0.1]) by tammananny.tiger (8.8.7/) with ESMTP id AAA00535 for ; Wed, 5 Nov 1997 00:02:28 -0800 Message-Id: <199711050802.AAA00535@tammananny.tiger> To: vsta@zendo.com Subject: new vsta user (windbag) Date: Wed, 05 Nov 1997 00:02:28 -0800 From: Quinn Dunkan Well, I'm just getting started with vsta. I have had it for a year or so, but only recently got an extra hard drive for it (my vsta boot process currently involves unplugging one hard drive and plugging another in since my machine only supports 2 ide drives). So: nice system! I'd like to help out with it, but unfortunately I'm not a very knowlegable programmer. I have never written anything like a device driver, but I'll learn someday! Perhaps vsta would be a good place to start. I've been reading source and skimming the list archive (number 11 seems not to be in the vsta directory, what's up?). So anyway, I went and upgraded to 1.6 and set up grub on my vsta hd. After changing cons2/isr.c to be dvorak and changing caps lock to control, the system was reasonably usable. I do have a few questions which may be silly, but I haven't seen them on the list, so I'll ask anyway. Linux, by default, has a high keyboard repeat rate which I rather like. Not that I know anything about assembly, but I found the following lines in src/linux/arch/i386/boot/setup.S: ! Set the keyboard repeat rate to the max mov ax,#0x0305 xor bx,bx ! clear bx int 0x16 so it seems to be a rather simple operation. Also, how easy would it be to switch console video modes? I'm used to a 80x50 console, would the cons server be the best place for something like this? Maybe have -80x50 and -80x25 command line switches or something? Unfortunately I know nothing of the inner workings of video cards, or I'd do it myself, but perhaps it would be a mere moment's hack for someone in the know. I'm sure relevant details can be borrowed from linux source, or the settextmode program. When under load, vsta seems a tad unresponsive. What I mean is, alt-f# vc switches give a noticable pause before switching, and character input is like typing over a 9600 baud modem line. Sometimes it seems as if characters are being dropped. Is this normal? Maybe since the console server is just another user-mode process, it is having to fight over the cpu? Also, vls -F takes quite a long time to display a directory. Is this normal overhead in interaction with the fs server? Is stat() much slower than on unix? Do processes have nice values and priority? My machine is a 486/66, btw. I was wondering how vsta did job control and ttys, etc. As much as I can tell, the answer is "it doesn't", but thought I heard a mention of setsid() somewhere back there on the mailing list. I mention this because a while back, Dan Bernstein wrote a suite of programs called `pty' that use pty sessions. He had a lot of interesting things to say on about how unix and posix really screws up with ttys and job control, and how such things could be improved greatly with a few simple changes. I wonder if anyone has thought of job control + vsta? Things like this were mentioned in the "Ideas Summary (2/3)" message of a while back. Perhaps the tty server is intended to do this, but I had a thought that it would be cool to have a "job control server". What would happen is that the stdio library (or whatever) would connect to this server, and processes would be connected to the other end. When you suspend a process, the server moves your shell's connection around, and maybe merges the stdout of the backgrounded process into your shell's. Say you start a long find or something, then get sick of waiting, so you suspend and bg. Then you don't want the garbage on your screen, so you tell the server to connect its stdout to a file. Or say you start up a bunch of processes, then log out. Later you can log back in and reconnect, or give them away to someone else ("here, finish up my game of nethack"). So the server acts like a sort of patchbay, allowing you to connect and disconnect processes at will. Sorry if this is exactly what the tty server you were talking about is supposed to do, it wasn't to clear to me. Anyway, then things like ^Z and ^C can be interpreted in the console driver, to switch tty connections around, and maybe freeze/melt processes. Then we can just tell everyone (like shells) that we don't have job control, and they'll never worry about it (or know). Here's another random little idea: how about a "config server"? The unix method is to have a million little text files spewed all over /etc, ~, /usr/lib, ..., which all have their own syntax and quirks. Under vsta, someone could write a "libconf" which provides a common interface for program config. It then registers with namer conf/progname, and makes files in there. So at any time, you can change the configuration of a daemon while it is running by doing, say, "echo 240 >/conf/ppp/mru", or view it with cat. Then the libconf has routines to save the config to a binary file in /etc/conf or something. Then programs don't have to worry about all this file parsing nonsense, they all have a consistent interface, etc. The only thing I would worry about is whether libconf could be made flexible enough to handle all purposes. Obviously, the simple case of config variables, like mtu in a net driver, can be easily represented. But what about conditionals, loops, etc.? rc files wouldn't change, of course, but maybe there could be a clear seperation between rcs and conf files So a shell, say, would have it's .shellrc which does the usual thing, and then a /conf/shell entry which replaces setopt (zsh), or set -o (bash) builtins. An even more random idea: I'm not sure what to call this, since I've never heard of it before, but one could have "standalone threads". What I mean is, executables which, instead of doing a fork()/exec(), do a tfork(). This would allow you to make commands like "cd", and more importantly, "mount" external rather than shell builtins. They might not actually be very useful, though, since they'd have to know all about the program they were running under. Standalone threads which had to do with common libraries (a mount command would just have to talk to stdio libc structures, right?) could actually be useful, though. What do people think? Well, as I said, I'm not a great programmer. I can write documentation, though (provided I understand what I'm documenting :) ). I figure I will first port "sps", my /proc-using linux ps replacement, and have a look at the proc server. Then maybe I'll write my "ls" replacement, "s", for vsta (it's one letter shorter, there's an improvement right there!). Along the way, I would be happy to write readmes and man pages about things I learn along the way. A "porting" document, which includes common things involved in porting posix programs to vsta, would be very useful and informative, not only to porters, but to users who want to know in what ways vsta differs from unix. Speaking of documentation, I've been having a little trouble with man. I figured out about adding things into the map file, but roff seems to be having trouble parsing non-vsta man pages. If I do "roff rc.1" I get pages and pages and pages of "unknown roff macro", or something similar. I think having a bare-bones nroff is a good idea, but it would be nice to be able to read or convert not-so-simple man pages. Speaking of mans, has anyone given any thought to the structure of documentation under vsta? As long as we're breaking from tradition, why not give some thought to the online documentation situation? I'm not a big fan of nroff (it's hard to read plaintext, it has developed lots of hackery over the years, etc.) myself. I'm not sure what could be used as a replacement, but my only requirement would be that it produces good console output (so tex + tty2dvi is out), and is easily greppable. I don't like having docs where the whole thing is organized around hypertext, so you have to _find_ all the documentation, but I think hyperlinks within a page could be very handy. I guess another possibility is TeXinfo, which presumably produces nice dvis as well as info files. I disagree with the gnu info reader, but given a decent reader and a linear document format, info might not be too bad. Of course, a solution that lets one use plain text pagers like less and more would be the best. So maybe nroff is doing pretty well after all. Plain text with no markup is fine, too. I have always thought that man should be more general: it should be able to look up arbitrary pages, and execute arbitrary commands on them. So for traditional man pages, it would run roff. For text files (and stray cats), it would run the pager with no preprocessing. For .tex it would run tex or latex and, err... madodvi the output. Ok, so I'm getting a bit ahead of things here :) I just think it would be nice to have a unified, well defined thing to do with documentation, that can be easily indexed and read by a user. The zsh people are using something called yodl which can compile to nroff, TeX, info, html, and maybe some more. Then of course there's sgml, but I haven't seen any console viewers for that. And then html, which I haven't seen any good console viewers for either (don't like lynx). Also, while watching all those roff errors go by, I found out about ctrl-C, or rather, that it seems to be lacking. I gathered that the only signal (event?) which can't be caught is "kill". Is there an equivalent of SIGINT? All I could find was some fakery in , and notify(2). Could cons2/isr.c be modified in the same place as the #define KDB to do a notify(0, 0, "interrupt");, or is it not so simple? How can I install an eventaction() sort of thing? Where can I find out more info about them? Has there been any sort of attempt at a filesystem structure standard? Since vsta's fs structure is all fakery anyway, it seems like agreeing on standard places for things would be pretty important. For example, all the makefiles in the src/ dirs want to link with libs in ../../lib. Unfortunately, given /vsta/src/bin/whatever, that doesn't work to well. I thought "oh, I'm supposed to put srcs in the root", but then makefiles in srv/ were tacking on too many ..s. I like the plan9 style /bin union dir, with sub dirs, so you could type disk/fsck_vfs or something. BTW again, why do all these programs have to manually link with crt0.o and -I/vsta/include? Couldn't gcc's specs file (which is undocumented as far as I can tell, I can't figure it out) do this automatically? From what I can tell, vstafs is nothing to rely on right now. There's also this bfs thing (boot file system?). It looks like it's not really intended for use on the root partition. So is everyone just using old 8.3, no owner/permissions FAT? Everyone seemed to agree that ext2 would be easy to port, but it seems that the only port is currently alpha dont-use-this-if-you-value-your-data. It would be nice to have a well tested, high performance filesystem for vsta. I'd imagine extensions to support vsta style file permissions wouldn't be hard, I think the hurd adds some extensions of it's own as well. Perhaps people would be interested in (if they don't already know of) the reiserfs, a high performance file-system being developed. There is info at http://www.idiom.com/~beverly/reiserfs.html. I'm not sure how well it would work with Andy's "reset at any time" policy, though. I've had some crashes when attempting to move directories. I'll do something like "mv src .." and that vc will freeze. All the other vcs are fine, until I type something and press enter, at which point they freeze too. Speaking of freezing vcs, does vsta (cons?) have type-ahead? Looking at a struct screen s.s_buf[], it apppears it does, but you can't see it. I'm used to typing one command while the first is completing, and I find it disconcerting if the characters don't immediately appear. Would it be a simple modification to have the cons server put all keystrokes on the screen immediately, or is there no way to do it unless the app at the other end does a read? Thank you (Andy?) very much for the vim port, I'm glad to see my favorite editor in vsta. Too bad it's an ancient version. Maybe the person who ported it could tell me what exactly was involved in doing it, so I could maybe try and port a new version (5.0something as of now, but I'm ignoring 5.0 in favor of 4.x)? Also, someone mentioned dynamic loading under vsta. Am I safe assuming no one has written dlopen()? The new version of zsh uses dynamically loaded modules, this would be a good way to encapsulate vsta-specifics (mount builtin, spiffy protections, etc.) into the mainstream distribution. I would like zsh mostly for the command-line editing, programmable completion, and decent globbing. How difficult is a shell to port? Something as complex as zsh might be very difficult. As a system shell, I think es would be very cool. It's small, reasonably fast, and very extensible and powerful. Maybe I'll try to port that, as rc has already been done it may not be too hard. It appears getline's idea of configuration is "modify the libc source". Unfortunately my libc won't compile (details later). Maybe I'll get the getline source to see if there's an easy way to add vi-style command line editing. I'm perfectly happy with rc, I just want my customizable command line editing and completion! Speaking of getline, I think it's mucking some things up: if I do "cat >foo" in the interests of making a file, I get "getline(): not interactive" or something and my prompt back. Speaking of porting, it would be nice if porters made a little document that describes changes they had to make to the program source. That way, when someone wants to port a new version, or a similar command, all that work isn't duplicated. Maybe it could even be sent to the program mainainer to put in the official release. I wonder if anyone is interested in a tar filesystem. I'm not sure exactly how it would work... you'd need some sort of auto-mounting. Maybe you could modify your 'cd' to say "if it's magic number indicates a tar or gz file, then mount it with the fs/tgz server, and cd to it". Also, how about an ftp filesystem, ala plan9? It seems it wouldn't be too difficult to write, in the same manner as a fs/tgz server: if you cd to, say, /n/ftp/ftp.zendo.com, it would try to connect to zendo. Like the tgz fs, it should do heavy caching to keep things nice. After no access for a while, it disconnects, and removes the /n/ftp/directory. Another advantage of this is if one user has been recently using /n/ftp.zendo.com, another one can use it too, without reconnecting or losing the cache. I don't think ftp allows multiple files to be transmitted at the same time, though, so the server may have to open another connection. Speaking of magic numbers, I've never really been very happy with them. Now that vsta is getting its own filesystem, perhaps that hack could be eliminated? I'm not exactly sure how to do it. Apple's type/creator fields seemed to work pretty well, although I have heard complaints about them being prone to trouble somehow (besides the trouble of having to register them with a central authority). But magic numbers rely totally on convention, too. I envision a filesystem with a type attribute, so all your gzs and gifs can lose their DOS-like filename extensions. It would still be easy to see what they are, since ls would of course display the file type if you wanted, and later on, what icon they get (hopefully we'll get that far :) ). But you could also not display them if you wanted (and maybe have ls color-code them or something), and save directory clutter. It might be hard to get away from those silly three letter extenions, since they are already dug in so firmly. A simple 32 bit field (or maybe 8 printable chars) would actually be pretty similar to .blah extensions, but more well-defined (all text files would have 'nl-ascii' or something, and .conf files would still end in .conf), more powerful (file type globbing, etc.), and more unobtrusive (what if you get sick of seeing and typing that extension, and you already know what file type it is, or don't care?). I've always thought programs that rely completely on three letter extensions were silly (DOS crashes if you rename a text file with .exe and try to run it). Of course, a type field is the same sort of thing, but much harder to mess up. And I'm sure something better than a plain type field could be imagined. Dare I mention mime? I remember a debate a while back among the plan9ers about whether arrow keys and like should be escape sequences or unicode 'chars' (runes?). cons2 seems to do ^[O{A,B,C,D}, which is vt100 standard, I guess. I've never really liked the concept of escape sequences, but single characters isn't actually that different. Some sort of seperate data path? Perhaps there could be a a vt100 cons server that supports all the old unix cruftiness, and then a special vsta cons that's all cool and spiffy and experimental? A problem I can think of is how would one be using the vsta shell under the vsta cons server, and then immediately switch around as soon as one runs a unix port? Now that we no longer have gazillions of different terminals, we could avoid the unix term{cap,info} and have a few seperate cons servers: vt100 for compatibility, telnet to non-vsta, etc., since it seems all the world's a vt100 nowadays, and vstacons, which has neat vsta-specific features. Another idea I had for a project I could possibly handle would be a syslog server. I've always thought the traditional syslog style was rather ad hoc, especially the facility part. A more general, flexible syslog would be a good thing, and maybe even a secure one (so if a message says it comes from a certain place, you know it did). I thought that facilities could somehow register themselves so the syslog knew about them, but then we already have the namer, which does something like that. My ideas are not clear at this point. Perhaps the syslog() call could create a /namer/thisserver/syslog file which is written to, and then read from by the syslogd? Or maybe the syslog server could create its own filesystem or something. Would it be beneficial to have permissions for a file specific to the process that created it? For example, so a process can store info that is readable, but not writable by even the user who started the process. This would suggest a "fake filesystem", since then those files would disappear when the process died, and not litter undeletable files around. Who handles #! in vsta? When using rc, I get an exec format error, whereas sh handles it ok. There is no mention of #! anywhere in sh.c, nor could I find anything in kern/exec.c (or anywhere else). I wonder what's happening with MADO? Is it still happening, slowly? Last I heard it sounded like significant work was done, but proceding at a leisurely pace. I have some ideas about UI sorts of things, but the mado person (Gavin?) would probably not be interested if he was not at the point to be implementing that sort of thing. Should I post them to vsta-win, just to see what people think? Oh, one more thing. What I am ultimately interested is a midi sequencer. I have had plans to write one of my own, but unfortunately the existing midi drivers under linux are limited in accuracy to 1/100 sec. I was wondering what kind of accuracy could be implemented in vsta. I know it's not a real time system, but given a need to tag incoming data with timestamps, how accurate could vsta be? Could we give the server a nice high priority so it would have lots of interrupts to service the midi port? In the vsta readme Andy said he wanted "enough RT support to do com port drivers", which implies yes. BTW, how do you nice things under vsta? The docs mentioned that, given it's leaf/node scheduling algorithm, you could gaurantee a certain process group would never use more than, say, 50% cpu time. How would you do this? Something I've always thought unix should have is a major page fault limit: limit the number of pages of swap a process can pull in every second. That would alleviate the problem of one misbehaved process thrashing the disk and cheating everyone else, while the cpu is sitting around at 3% (I see a lot of this, since I have 6MB RAM). I also had a crash: I compiled the libc, and it got to the mkshlib part. It printed two lines (Library libc.a @ 0x60001000) and froze. After a bit, I eventually got: Assertion Failed line 323 file ../kern/vm_page.c set_core: bad index and was dumped into the kernel debugger. I don't know many commands in the kdb (quit and proc :) ), so I reset eventually. Every time I try to compile the libc, it crashes at that same line, although in different ways (sometimes it goes into ignore-the-user mode). I've got vim 4.2 working except one thing: vsta doesn't seem to have select() (or poll(), I guess that's a systemV thing), and vim wants to use that for, well, waiting on fds. I looked at the vim 1.24 port, and was more confused than enlightened (in the routine supposed to wait for fd input, uses select in 4.2): RealWaitForChar(int ticks) { if (ticks != -1) return 0; return 1; } Notice it doesn't actually do any waiting? Anyway, the only non working thing about my vim is multi character bindings, which I figure is since it isn't waiting properly... but they work on vim 1.24!! How are you supposed to wait on multiple fds in vsta? Seems rename() hasn't made it into the headers yet. No big deal, just a reminder. So, anyway, if anyone has a simple project for me to do, I would be interested in helping out. I could write documentation: as you can see, I am capable of being very verbose :) Oh, btw: in case anyone is interested, there is a libgc with a drop-in garbage-collecting replacement for malloc() freely available, developed by the Xerox PARC people. It shouldn't be too hard to port to vsta, since it also compiles on the mac, os/2, amiga, and windows (as well as unix, of course). I don't know how it stacks up in efficiency, strength, etc. with other gcs, but it's there and it's free and portable. It's designed to be thread-safe. If people are interested, I could try to port that too. btw again, vsta's cpp doesn't seem to be the same as the gnu one. It chokes on certain #if lines, which makes porting rather more difficult. btw again again, "root" is a more appropriate name than ever in vsta, since it is the root of the permissions tree. Does anyone know where that name on unix came from? From daemon Tue Nov 4 22:15:19 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id WAA15176 for vsta-xplod; Tue, 4 Nov 1997 22:15:19 GMT Received: from nenuphar.saclay.cea.fr (nenuphar.saclay.cea.fr [132.166.192.7]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id WAA15170 for ; Tue, 4 Nov 1997 22:15:15 GMT Received: from vega.serma.cea.fr (vega.serma.cea.fr [132.166.149.56]) by nenuphar.saclay.cea.fr (8.8.5/ CEANET-ROUTER-3.0) with ESMTP id JAA02972 for ; Wed, 5 Nov 1997 09:15:53 +0100 (MET) Received: (from basile@localhost) by vega.serma.cea.fr (8.8.6/8.8.5) id JAA23864; Wed, 5 Nov 1997 09:17:30 +0100 Date: Wed, 5 Nov 1997 09:17:30 +0100 Message-Id: <199711050817.JAA23864@vega.serma.cea.fr> From: Basile STARYNKEVITCH MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: vsta@zendo.com Subject: Q: GRUB & LILO? X-Mailer: VM 6.31 under 20.2 XEmacs Lucid Hello all, I would like to boot VSTa from a hard disk partition (/dev/hda2 on linux, ie 2nd partition is a DOS/FAT filesystem of about 150mb containing /vsta) on my home PC. The main (everyday) OS is Linux (binutils 8.1.0.15, ELF libc5.4.39, kernel 2.0.31, gcc-2.7.2.3) booted with Lilo (Linux root is /dev/hda1 which contains the LILO loader) I do succeed in booting the GRUB (grub.raw) from a floppy. I would like to boot GRUB from LILO (and then boot VSTa). However, I failed to install grub sucessfully on the harddisk (actually, I'm not sure i fully understood all of GRUB features; I tried several GRUB install= commands without success) I added a other=/dos/vsta/grub/stage1 entry into /etc/lilo.conf. Since I'm using the debug version of GRUB, I did found that GRUB stage1 run ok and stage2 is loaded. (machine is a Cyrix150 -soon will be a AmdK6200- on an Asus motherboard with PCI) Any clue? N.B. Any opinions expressed here are solely mine, and not of my organization. N.B. Les opinions exprimees ici me sont personnelles et n engagent pas le CEA. ---------------------------------------------------------------------- Basile STARYNKEVITCH ---- Commissariat à l Energie Atomique DRN/DMT/SERMA * CEA/Saclay bat.470 * 91191 GIF/YVETTE CEDEX * France fax: (33) 01,69.08.85.68; phone: 01,69.08.40.66; home: 01,46.65.45.53 email: Basile . Starynkevitch @ cea . fr (but remove white space) I speak french, english, russian. Je parle français, anglais, russe. ---------------------------------------------------------------------- From daemon Wed Nov 5 07:31:52 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id HAA15892 for vsta-xplod; Wed, 5 Nov 1997 07:31:52 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id HAA15886 for ; Wed, 5 Nov 1997 07:31:49 GMT Received: from localhost (vandys@localhost) by zipper.cisco.com (8.8.4-Cisco.1/8.6.5) with SMTP id JAA09861; Wed, 5 Nov 1997 09:31:42 -0800 (PST) Message-Id: <199711051731.JAA09861@zipper.cisco.com> To: Basile STARYNKEVITCH cc: vsta@zendo.com Subject: Re: Q: GRUB & LILO? In-reply-to: Your message of "Wed, 05 Nov 1997 09:17:30 +0100." <199711050817.JAA23864@vega.serma.cea.fr> Date: Wed, 05 Nov 1997 09:31:42 -0800 From: Andrew Valencia [Basile STARYNKEVITCH writes:] >I would like to boot GRUB from LILO (and then boot VSTa). However, I >failed to install grub sucessfully on the harddisk (actually, I'm not >sure i fully understood all of GRUB features; I tried several GRUB >install= commands without success) I thought GRUB could boot Linux in *place* of LILO. I don't recollect any instructions on how to use LILO to load GRUB. But, then, my other primary OS is FreeBSD, so I'm not very LILO literate. >I added a other=/dos/vsta/grub/stage1 entry into /etc/lilo.conf. Since >I'm using the debug version of GRUB, I did found that GRUB stage1 run >ok and stage2 is loaded. So LILO knew how to bring in the primary stage1 loader, and it successfully pulled in stage2? Then you have the menu prompt, and could now enter root= command and so forth? It seems if so, all that's left is to get it to honor a menu file, yes? Andy From daemon Wed Nov 5 15:33:01 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id PAA16455 for vsta-xplod; Wed, 5 Nov 1997 15:33:01 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id PAA16449 for ; Wed, 5 Nov 1997 15:32:59 GMT Received: from localhost (vandys@localhost) by zipper.cisco.com (8.8.4-Cisco.1/8.6.5) with SMTP id RAA17956; Wed, 5 Nov 1997 17:33:11 -0800 (PST) Message-Id: <199711060133.RAA17956@zipper.cisco.com> To: OooYoUooO@aol.com cc: vsta@zendo.com Subject: Re: GCC locks up on my 386. In-reply-to: Your message of "Wed, 05 Nov 1997 20:27:02 EST." <971105202701_445105629@mrin41.mail.aol.com> Date: Wed, 05 Nov 1997 17:33:11 -0800 From: Andrew Valencia [OooYoUooO@aol.com writes:] >I am using VSTa 1.5.2 on my 386sx/16 with 2mb of RAM, and GCC locks up as >soon as I press enter to run it! And when I start MGR, I get about two >frames of the animation, then it just sits there and responds to nothing. > Any ideas? Do ^Z (control-Z) to get the kernel debugger. Do "dv freemem", which "dump virtual address" of the "freemem" variable. If it displays 0, you've run out of memory. That'd be my first guess, due to your configuration. 4 megs was the baseline for pre-1.6, and with gcc 2.x in 1.6, I think 8 may have to be the baseline. Beware that in 1.6, you need an FPU to use GCC due to some lame techniques used in code generation. Andy From daemon Wed Nov 5 15:27:26 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id PAA16442 for vsta-xplod; Wed, 5 Nov 1997 15:27:26 GMT Received: from mrin41.mail.aol.com (mrin41.mx.aol.com [198.81.19.151]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id PAA16436 for ; Wed, 5 Nov 1997 15:27:23 GMT From: OooYoUooO@aol.com Received: (from root@localhost) by mrin41.mail.aol.com (8.8.5/8.7.3/AOL-2.0.0) id UAA14639; Wed, 5 Nov 1997 20:27:02 -0500 (EST) Date: Wed, 5 Nov 1997 20:27:02 -0500 (EST) Message-ID: <971105202701_445105629@mrin41.mail.aol.com> To: vandys@cisco.com cc: vsta@zendo.com Subject: GCC locks up on my 386. I am using VSTa 1.5.2 on my 386sx/16 with 2mb of RAM, and GCC locks up as soon as I press enter to run it! And when I start MGR, I get about two frames of the animation, then it just sits there and responds to nothing. Any ideas? From daemon Wed Nov 5 16:39:36 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id QAA16524 for vsta-xplod; Wed, 5 Nov 1997 16:39:36 GMT Received: from elektra.ultra.net (elektra.ultra.net [199.232.56.13]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id QAA16518 for ; Wed, 5 Nov 1997 16:39:34 GMT From: larz@ultranet.com Received: from [146.115.41.136] (d8.dial-5.ltn.ma.ultra.net [146.115.41.136]) by elektra.ultra.net (8.8.5/ult.n14191) with SMTP id VAA02530; Wed, 5 Nov 1997 21:40:03 -0500 (EST) Date: Wed, 5 Nov 1997 21:40:03 -0500 (EST) Message-Id: <199711060240.VAA02530@elektra.ultra.net> Subject: Re: VSTa 1.6 available To: , Mime-Version: 1.0 Content-Type: text/plain; charset="US-ASCII" Hi, [Andy Valencia writes regarding a grub boot floppy] >I don't know about that, but I thought the instructions included a way to >create a boot floppy which pulled the secondary loader as well as the >menu.lst from the hard disk. You have to keep a couple files on your hard >disk, but since grub supports most filesystems that doesn't seem like an >undue hardship. And it spares you having to mess with your current hard >disk boot mechanism. Ok, grub boot from floppy works just fine. I pretty much used the second example in the grub install page. Next - on to getting 1.6 up and running. There were a few problems (see below). All-in-all, though, VSTa performed very well. Its certainly a lot more stable than it used to be. Here some things that needed fixing and possible fixes: 1) the cam server was not handling message op's w/ the high bit set. I see that other servers mask m_op w/ MSG_MASK. Unfortunately, the cam server defines some messages (that it sends to itself) that use the low 16-bits. So for now, m_op gets AND'ed w/ 0xffff. 2) the ka9q net program would freeze as soon as it executed an "attach ether eth0 1500" command. It turns out that the function keep_things_going() will always return true (ie, didstuff) after the attach is done. This caused one of the threads to continously loop w/ ka9q_lock held. I modified keep_things_going() to always return 0. This probably isn't the right way to fix this problem (I don't know this code very well at all), but it works around the freeze. 3) it doesn't seem like the /vsta/etc/hosts.net file is being read by net, so I added a call to xresolve() from resolve() (domain.c). Mike Larson From daemon Wed Nov 5 18:34:36 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id SAA16652 for vsta-xplod; Wed, 5 Nov 1997 18:34:36 GMT Received: from mail-out1.apple.com (mail-out1.apple.com [17.254.0.52]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id SAA16646 for ; Wed, 5 Nov 1997 18:34:34 GMT Received: from scv1.apple.com (A17-128-100-139.apple.com [17.128.100.139]) by mail-out1.apple.com (8.8.5/8.8.5) with ESMTP id UAA16384 for ; Wed, 5 Nov 1997 20:02:55 -0800 Received: from [17.201.21.202] (bandsu.apple.com [17.201.21.202]) by scv1.apple.com (8.8.5/8.8.5) with ESMTP id UAA32014 for ; Wed, 5 Nov 1997 20:02:48 -0800 X-Sender: web12188@thematech.com Message-Id: In-Reply-To: <199711051731.JAA09861@zipper.cisco.com> References: Your message of "Wed, 05 Nov 1997 09:17:30 +0100." <199711050817.JAA23864@vega.serma.cea.fr> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Wed, 5 Nov 1997 20:02:49 -0800 To: vsta@zendo.com From: Sumit Bandyopadhyay Subject: Anybody use VSTA on k6 and/or PA-2007 by FIC I am considering buying the PA-2007 motherboard by FIC, and put a K6/200 on it. VSTA will be one of the OSes I would like to run on this system. Anyone have experience related to VSTA with this combination, or just the motherboard or K6? Regards, sumit@thematech.com From daemon Thu Nov 6 08:41:51 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id IAA17714 for vsta-xplod; Thu, 6 Nov 1997 08:41:51 GMT Received: from gateway.s3.com (gateway.s3.com [198.211.58.4]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id IAA17707 for ; Thu, 6 Nov 1997 08:41:43 GMT Received: from chat.net (home.chat.net [198.211.58.200]) by gateway.s3.com (8.6.12/8.6.12) with ESMTP id KAA11721 for ; Thu, 6 Nov 1997 10:20:34 -0800 Received: by chat.net (SMI-8.6/SMI-SVR4) id KAA02493; Thu, 6 Nov 1997 10:44:20 -0800 Message-ID: <19971106104419.23929@home.chat.net> Date: Thu, 6 Nov 1997 10:44:19 -0800 From: David Jeske To: vsta@zendo.com Subject: Re: Q: GRUB & LILO? References: <199711050817.JAA23864@vega.serma.cea.fr> <199711051731.JAA09861@zipper.cisco.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.79e In-Reply-To: <199711051731.JAA09861@zipper.cisco.com>; from Andrew Valencia on Wed, Nov 05, 1997 at 09:31:42AM -0800 On Wed, Nov 05, 1997 at 09:31:42AM -0800, Andrew Valencia wrote: > >I would like to boot GRUB from LILO (and then boot VSTa). However, I > >failed to install grub sucessfully on the harddisk (actually, I'm not > >sure i fully understood all of GRUB features; I tried several GRUB > >install= commands without success) > > I thought GRUB could boot Linux in *place* of LILO. I don't recollect any > instructions on how to use LILO to load GRUB. But, then, my other primary > OS is FreeBSD, so I'm not very LILO literate. You can boot GRUB from LILO. Standard "disk" or "partition" booting, involves loading the first 512 bytes from the disk. That's how GRUB gets loaded initially, that's how LILO gets loaded, that's how either of them "chain-load" other standard operating systems like DOS or OS/2. If you install LILO on your master boot record (i.e. use the hard drive device, not a partition device) and you install GRUB onto a partition's boot record, then you can setup LILO to chain-load that partition, and you'll hit GRUB. It sounds like you may have already accomplished this. -- David Jeske (N9LCA) + jeske@chat.net From daemon Thu Nov 6 10:48:39 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id KAA17883 for vsta-xplod; Thu, 6 Nov 1997 10:48:39 GMT Received: from atrey.karlin.mff.cuni.cz (root@atrey.karlin.mff.cuni.cz [195.113.31.123]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id KAA17877 for ; Thu, 6 Nov 1997 10:48:36 GMT Received: from Elf.mj.gts.cz (root@slip8.ms.mff.cuni.cz [195.113.20.208]) by atrey.karlin.mff.cuni.cz (8.8.5/8.8.5) with ESMTP id VAA02204; Thu, 6 Nov 1997 21:49:08 +0100 Received: (from pavel@localhost) by Elf.mj.gts.cz (8.8.5/8.8.5) id LAA10840; Thu, 6 Nov 1997 11:24:25 +0100 Message-ID: <19971106112424.24361@Elf.mj.gts.cz> Date: Thu, 6 Nov 1997 11:24:24 +0100 From: Pavel Machek To: Andrew Valencia Cc: Basile STARYNKEVITCH , vsta@zendo.com Subject: Re: Q: GRUB & LILO? References: <199711050817.JAA23864@vega.serma.cea.fr> <199711051731.JAA09861@zipper.cisco.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.81 In-Reply-To: <199711051731.JAA09861@zipper.cisco.com>; from Andrew Valencia on Wed, Nov 05, 1997 at 09:31:42AM -0800 X-Warning: Not only using Windows can be dangerous to your mental health. Hi! > >I would like to boot GRUB from LILO (and then boot VSTa). However, I > >failed to install grub sucessfully on the harddisk (actually, I'm not > >sure i fully understood all of GRUB features; I tried several GRUB > >install= commands without success) > > I thought GRUB could boot Linux in *place* of LILO. I don't > recollect any Yes, you could. But, GRUB has HORRIBLE user interface. Lilo is much better in this: It automagically boots last system you booted (unless you hold shift when booting, in such case command line appears), its interface is single line - so it will not wipe BIOS's messages. I see that GRUB is more generic, and probably better than lilo, but lilo's user interface suits much more to me. (No menus, just command line, it does not appear unless I want it to, ...). I patched grub a bit (to make it look better), but much more should be done. I would need things like safing last booted system, macros for command line etc. Also, MultiBOOT protocol should be redone a bit ;-). ... and maybe creating completely new MultiBOOT compliant loader would be less work ... Pavel -- I'm really pavel@atrey.karlin.mff.cuni.cz. Pavel Look at http://atrey.karlin.mff.cuni.cz/~pavel/ ;-). From daemon Fri Nov 7 01:41:11 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id BAA18736 for vsta-xplod; Fri, 7 Nov 1997 01:41:11 GMT Received: from atrey.karlin.mff.cuni.cz (pavel@atrey.karlin.mff.cuni.cz [195.113.31.123]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id BAA18730 for ; Fri, 7 Nov 1997 01:41:01 GMT Received: (from pavel@localhost) by atrey.karlin.mff.cuni.cz (8.8.5/8.8.5) id MAA23424; Fri, 7 Nov 1997 12:39:34 +0100 Message-ID: <19971107123934.35011@atrey.karlin.mff.cuni.cz> Date: Fri, 7 Nov 1997 12:39:34 +0100 From: Pavel Machek To: Gregory Gerard Cc: vsta@zendo.com Subject: Re: Q: GRUB & LILO? References: <199711070647.AAA21808@larissa.cs.trinity.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.84 In-Reply-To: <199711070647.AAA21808@larissa.cs.trinity.edu>; from Gregory Gerard on Thu, Nov 06, 1997 at 10:42:15PM -0800 X-Warning: Not only using Windows can be dangerous to your mental health. > >I see that GRUB is more generic, and probably better than lilo, but > >lilo's user interface suits much more to me. (No menus, just command > >line, it does not appear unless I want it to, ...). I patched grub a > >bit (to make it look better), but much more should be done. I would > >need things like safing last booted system, macros for command line > >etc. > > > >Also, MultiBOOT protocol should be redone a bit ;-). > > > >... and maybe creating completely new MultiBOOT compliant loader > >would be less work ... > > > > Pavel > > What would you propose in place of MultiBoot? Well. I have not seen anything else, but I saw way how to improve multiboot. 1) It passes some config info, but it does not tell you where config info is, so you can not reserve area for config without walking all those config tables 2) Also, config tables are complicated, and you can see history on them. It would be much easier ... to create link list of detected hardware. (Instead of having entry in structures for floppies and for ... and for ... and for ...). Current specification is limited in 32-bits somewhere, and structures are already pretty big. > BTW, have you been able to compile grub under linux? After minor hackery, I was. Makefiles needed to be changed. May me personally and I'll send you my GRUB tree. Pavel -- -- This is my little buggy signature... Pavel GCM d? s-: !g p?:+ au- a--@ w+ v- C++@ UL+++ L++ N++ E++ W--- M- Y- R+ From daemon Fri Nov 7 03:29:33 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id DAA18980 for vsta-xplod; Fri, 7 Nov 1997 03:29:33 GMT Received: from cs.utah.edu (cs.utah.edu [128.110.4.21]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id DAA18974 for ; Fri, 7 Nov 1997 03:29:21 GMT Received: from fast.cs.utah.edu by cs.utah.edu (8.8.4/utah-2.21-cs) id GAA21281; Fri, 7 Nov 1997 06:29:47 -0700 (MST) Received: by fast.cs.utah.edu (8.6.10/utah-2.15-leaf) id GAA01098; Fri, 7 Nov 1997 06:29:26 -0700 Date: Fri, 7 Nov 1997 06:29:26 -0700 From: vanmaren@fast.cs.utah.edu (Kevin Van Maren) Message-Id: <199711071329.GAA01098@fast.cs.utah.edu> To: pavel@Elf.mj.gts.cz, vandys@cisco.com Subject: Re: Q: GRUB & LILO? Cc: Basile.Starynkevitch@cea.fr, vsta@zendo.com > Yes, you could. But, GRUB has HORRIBLE user interface. Lilo is much > better in this: It automagically boots last system you booted (unless > you hold shift when booting, in such case command line appears), its > interface is single line - so it will not wipe BIOS's messages. LILO has a lot of other limitations, though -- like you have to re-run LILO to every time you change a kernel. At least with Grub (or BSD bootloaders) you can boot *any* kernel on your disk. > Also, MultiBOOT protocol should be redone a bit ;-). > > ... and maybe creating completely new MultiBOOT compliant loader > would be less work ... We link Multiboot kernels with a BSD boot adapter, which allows us to load mutiboot kernels from BSD bootblocks. This is pretty simple. What part of the multiboot specification are you not happy with? Kevin From daemon Thu Nov 13 23:55:31 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id XAA02353 for vsta-xplod; Thu, 13 Nov 1997 23:55:31 GMT Received: from atrey.karlin.mff.cuni.cz (atrey.karlin.mff.cuni.cz [195.113.31.123]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id XAA02347 for ; Thu, 13 Nov 1997 23:55:19 GMT Received: from Elf.mj.gts.cz (root@slip8.ms.mff.cuni.cz [195.113.20.208]) by atrey.karlin.mff.cuni.cz (8.8.5/8.8.5) with ESMTP id KAA21385 for ; Fri, 14 Nov 1997 10:55:38 +0100 Received: (from pavel@localhost) by Elf.mj.gts.cz (8.8.5/8.8.5) id WAA00106; Thu, 13 Nov 1997 22:45:18 +0100 Message-ID: <19971113224517.06638@Elf.mj.gts.cz> Date: Thu, 13 Nov 1997 22:45:17 +0100 From: Pavel Machek To: vsta@zendo.com Subject: REPOST: Patch for things I did not like on VSTa Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.81 X-Warning: Not only using Windows can be dangerous to your mental health. Hi! I posted rather big patch to this list, but got no reply. Does it mean that it was silently applied, or silently discarded, or silently lost by mail? I believe everything in here is Right Think (tm), with possible exception of turning arrows from ^[[.. into ^f, ^b, ^p, ^n. As termcaps are not too working, yet, it seemed to me like least-force solution - ignore hunk @@ -113,26 +124,29 @@ from isr.c if you want. Pavel Anyway, here it is: diff -ur vsta.old/src/os/dbg/dbgmain.c /c/vsta/src/os/dbg/dbgmain.c --- vsta.old/src/os/dbg/dbgmain.c Sun Oct 8 21:49:46 1995 +++ /c/vsta/src/os/dbg/dbgmain.c Mon Oct 27 20:05:34 1997 @@ -8,7 +8,7 @@ extern void dump_phys(), dump_virt(), dump_procs(), dump_pset(), dump_instr(), trace(), trapframe(), dump_vas(), dump_port(), dump_pview(), dump_thread(), dump_ref(), reboot(), memleaks(), - dump_sysmsg(), dump_core(); + dump_sysmsg(), dump_core(), help(); extern void dbg_inport(), dbg_outport(); static void quit(), calc(), set(), set_mem(); extern int get_num(); @@ -46,6 +46,7 @@ "trace", trace, "vas", dump_vas, "writemem", set_mem, + "?", help, 0, 0 }; @@ -137,6 +138,20 @@ x = get_num(str); printf("%s 0x%x %d\n", symloc(x), x, x); } +/* + * help() + * List all commands + */ +static void +help(void) +{ + int x; + + printf( "Available commands: " ); + for (x = 0; cmdtab[x].c_name; ++x) + printf( "%s ", cmdtab[x].c_name ); + printf( "\n" ); +} /* * do_cmd() @@ -161,7 +176,7 @@ } } if (matches == 0) { - printf("No such command\n"); + printf("No such command (type ? for help)\n"); return; } if (matches > 1) { diff -ur vsta.old/src/os/kern/proc.c /c/vsta/src/os/kern/proc.c --- vsta.old/src/os/kern/proc.c Mon Apr 8 18:31:30 1996 +++ /c/vsta/src/os/kern/proc.c Mon Oct 27 13:29:40 1997 @@ -26,7 +26,8 @@ extern lock_t runq_lock; -ulong npid_free = (ulong)-1; /* # PIDs free in pool */ +ulong npid_free = (ulong)-2; /* # PIDs free in pool */ -ulong pid_nextfree = 0L; /* Next free PID number */ +ulong pid_nextfree = 1L; /* Next free PID number */ + /* We got to use 1 or process #0 is unkillable */ struct proc *allprocs = 0; /* List of all procs */ sema_t pid_sema; /* Mutex for PID pool and proc lists */ diff -ur vsta.old/src/srv/mach/cons2/isr.c /c/vsta/src/srv/mach/cons2/isr.c --- vsta.old/src/srv/mach/cons2/isr.c Tue Jan 30 08:32:36 1996 +++ /c/vsta/src/srv/mach/cons2/isr.c Tue Oct 28 20:30:12 1997 @@ -12,6 +12,9 @@ capstoggle = 0, /* For toggling effect of CAPS */ numtoggle = 0, /* ...NUM lock */ isE0 = 0; /* Prefix for extended keys (FN1, etc.) */ + +#define C( x ) x & 0x1f /* Turn key into ctrl-one */ +#define M( x ) x | 0x80 /* Turn key into meta key, do not use for 0 */ /* Map scan codes to ASCII, one table for normal, one for shifted */ static char normal[] = { @@ -82,11 +85,19 @@ #endif /* + * Meta keys are badly needed - if Alt is pressed, + * it sends ESC before actual character. + */ + + if (alt || (ch&0x80)) + kbd_enqueue(s, 033); + + /* * Hand off straight data now. The keyboard always enters * data for the virtual screen currently being displayed * on the hardware screen. */ - kbd_enqueue(s, ch); + kbd_enqueue(s, ch&0x7f); } /* @@ -113,26 +124,29 @@ static int cursor_key(struct screen *s, uchar c) { - char *cp; + char buf[5]; + char *cp = buf; + cp[0] = ' '; + cp[1] = 0; switch (c) { case 72: /* up */ - cp = "\033OA"; + cp[0] = C('p'); break; case 80: /* down */ - cp = "\033OB"; + cp[0] = C('n'); break; case 77: /* right */ - cp = "\033OC"; + cp[0] = C('f'); break; case 75: /* left */ - cp = "\033OD"; + cp[0] = C('b'); break; case 73: /* pg up */ - cp = "\033[5~"; + cp = "\033v"; break; case 81: /* pg down */ - cp = "\033[6~"; + cp[0] = C('v'); break; case 82: /* insert */ cp = "\033[2~"; @@ -172,46 +186,11 @@ { char *p; - switch (c) { - case 59: /* F1 */ - p = "\033OP"; - break; - case 60: /* F2 */ - p = "\033OQ"; - break; - case 61: /* F3 */ - p = "\033OR"; - break; - case 62: /* F4 */ - p = "\033OS"; - break; - case 63: /* F5 */ - p = "\033OT"; - break; - case 64: /* F6 */ - p = "\033OU"; - break; - case 65: /* F7 */ - p = "\033OV"; - break; - case 66: /* F8 */ - p = "\033OW"; - break; - case 67: /* F9 */ - p = "\033OX"; - break; - case 68: /* F10 */ - case 87: /* F11 */ - case 88: /* F12 */ - p = 0; - break; - default: - return 0; - } - if (p) { - enqueue_string(s, p); - } - return(1); + if ((c<59) || (c>68)) return 0; + p = "\033Q?"; + p[2] = 'P'+c-59; + enqueue_string(s, p); + return 1; } /* diff -ur vsta.old/src/srv/mach/wd/wd.c /c/vsta/src/srv/mach/wd/wd.c --- vsta.old/src/srv/mach/wd/wd.c Thu Oct 12 07:09:46 1995 +++ /c/vsta/src/srv/mach/wd/wd.c Wed Oct 29 21:37:52 1997 @@ -376,10 +376,13 @@ * For laptops with power management, our inportb() here can * cause the disk to wake up; the controller can then flag * a busy disk until the spinup is complete. Give it 5 seconds. + * + * I feel that 5 seconds is not enough. (My drive takes about + * 6, my other drive is even slower). */ cyl = 0; while (inportb(base + WD_STATUS) & WDS_BUSY) { - if (++cyl > 50) { + if (++cyl > 150) { ASSERT(0, "wd_start: busy"); } __msleep(100); -- I'm really pavel@atrey.karlin.mff.cuni.cz. Pavel Look at http://atrey.karlin.mff.cuni.cz/~pavel/ ;-). From daemon Fri Nov 14 07:06:59 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id HAA02947 for vsta-xplod; Fri, 14 Nov 1997 07:06:59 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id HAA02941 for ; Fri, 14 Nov 1997 07:06:55 GMT Received: from localhost (vandys@localhost) by zipper.cisco.com (8.8.4-Cisco.1/8.6.5) with SMTP id JAA29362; Fri, 14 Nov 1997 09:06:58 -0800 (PST) Message-Id: <199711141706.JAA29362@zipper.cisco.com> To: Pavel Machek cc: vsta@zendo.com Subject: Re: REPOST: Patch for things I did not like on VSTa In-reply-to: Your message of "Thu, 13 Nov 1997 22:45:17 +0100." <19971113224517.06638@Elf.mj.gts.cz> Date: Fri, 14 Nov 1997 09:06:58 -0800 From: Andrew Valencia [Pavel Machek writes:] >I posted rather big patch to this list, but got no reply. Does it mean >that it was silently applied, or silently discarded, or silently lost >by mail? Our mail services tend to be pretty reliable, and yes, your patch has been accepted. I think all of it looked pretty reasonable except--as you said--the cursor key hack. So that "help" for kdb, and more time for slow IDE disks, and avoiding PID 0. I have the patches in my "vsta_todo" folder, which gets emptied as I head towards a next source drop. Quinn Dunkan's long and interesting message, on the other hand, is still in my inbox, as I intend to answer it when I have a bunch of minutes all in a row. :-) Regards, Andy From daemon Mon Nov 17 10:10:20 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id KAA09926 for vsta-xplod; Mon, 17 Nov 1997 10:10:20 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id KAA09920 for ; Mon, 17 Nov 1997 10:10:17 GMT Received: from localhost (vandys@localhost) by zipper.cisco.com (8.8.4-Cisco.1/8.6.5) with SMTP id MAA15495; Mon, 17 Nov 1997 12:11:02 -0800 (PST) Message-Id: <199711172011.MAA15495@zipper.cisco.com> To: Quinn Dunkan cc: vsta@zendo.com Subject: Re: new vsta user (windbag) In-reply-to: Your message of "Wed, 05 Nov 1997 00:02:28 PST." <199711050802.AAA00535@tammananny.tiger> Date: Mon, 17 Nov 1997 12:11:00 -0800 From: Andrew Valencia [Quinn Dunkan writes:] > Linux, by default, has a >high keyboard repeat rate which I rather like. Not that I know anything about >assembly, but I found the following lines in src/linux/arch/i386/boot/setup.S: >! Set the keyboard repeat rate to the max > mov ax,#0x0305 > xor bx,bx ! clear bx > int 0x16 I'm not sure I *like* the hyper repeat rate.... >so it seems to be a rather simple operation. Also, how easy would it be to >switch console video modes? I'm used to a 80x50 console, would the cons serve >r >be the best place for something like this? Maybe have -80x50 and -80x25 >command line switches or something? Unfortunately I know nothing of the inner >workings of video cards, or I'd do it myself, but perhaps it would be a mere >moment's hack for someone in the know. I'm sure relevant details can be >borrowed from linux source, or the settextmode program. Actually, it'd probably be better to spend that time on putting a decent video mode into MGR. Right now it tops out at 640x480, which is getting long in the tooth with a 8 MB video card playing into a 20" monitor. Then you could get multi-fonts into all the windows, which would be more useful yet. >When under load, vsta seems a tad unresponsive. What I mean is, alt-f# vc >switches give a noticable pause before switching, and character input is like >typing over a 9600 baud modem line. Sometimes it seems as if characters are >being dropped. Is this normal? Maybe since the console server is just anothe >r >user-mode process, it is having to fight over the cpu? Also, vls -F takes >quite a long time to display a directory. Yes, the scheduler treatment of cons2 is unkind. The problem is that the process looks like a CPU-bound one, with reasonably large sleeps in between. This would *really* be solved by some sort of inheritance, but that's something which loads a system with more overhead. A simple fix would be to give cons2 a real-time priority. >Is this normal overhead in >interaction with the fs server? Is stat() much slower than on unix? Do >processes have nice values and priority? My machine is a 486/66, btw. I've noticed this, too, but [gv]ls are new additions, ported years ago by Dave Hudson, and I didn't spend any time analyzing where they spend their time. >I was wondering how vsta did job control and ttys, etc. As much as I can tell >, >the answer is "it doesn't", but thought I heard a mention of setsid() somewher >e >back there on the mailing list. I mention this because a while back, Dan >Bernstein wrote a suite of programs called `pty' that use pty sessions. He ha >d >a lot of interesting things to say on about how unix and posix really screws u >p >with ttys and job control, and how such things could be improved greatly with >a >few simple changes. I wonder if anyone has thought of job control + vsta? Yes, and we ended up decicing that the signal-based job control model actually is better than you'd think up front. But first the basics of keyboards generating interrupts. Note that MGR already does this (well, it uses a pull-down item, but the *mechanism* is all there). And processes do have signal groups. I actually have some code I'm cooking to keep track of TTY mode and setting of VINTR, and generate signals to the right processes at the right time. >Here's another random little idea: how about a "config server"? The unix >method is to have a million little text files spewed all over /etc, ~, >/usr/lib, ..., which all have their own syntax and quirks. Actually, it was proposed as simply a config library. I'm not sure an additional server adds all that much over editing of config files. Of course, if you want to hack it out and see if it's nice in reality, I can fold it in. >An even more random idea: I'm not sure what to call this, since I've never >heard of it before, but one could have "standalone threads". What I mean is, >executables which, instead of doing a fork()/exec(), do a tfork(). This would >allow you to make commands like "cd", and more importantly, "mount" external >rather than shell builtins. They might not actually be very useful, though, >since they'd have to know all about the program they were running under. >Standalone threads which had to do with common libraries (a mount command woul >d >just have to talk to stdio libc structures, right?) could actually be useful, >though. What do people think? Since it's going to intimately re-tread the shared data structures, right off the bat you need to figure out how you protect critical sections. > I figure I will first port >"sps", my /proc-using linux ps replacement, and have a look at the proc server >. A nice ps(1) replacement would be fine. In fact, it would just become the VSTa ps, I think. Did you write the original sps, from the days of V7? >Then maybe I'll write my "ls" replacement, "s", for vsta (it's one letter >shorter, there's an improvement right there!). Along the way, I would be >happy to write readmes and man pages about things I learn along the way. A >"porting" document, which includes common things involved in porting posix >programs to vsta, would be very useful and informative, not only to porters, >but to users who want to know in what ways vsta differs from unix. That would be cool. Like most programmers, I think documentation is a fine thing, but I always end up doing the coding part most of the time. >Speaking of documentation, I've been having a little trouble with man. I >figured out about adding things into the map file, but roff seems to be having >trouble parsing non-vsta man pages. If I do "roff rc.1" I get pages and pages >and pages of "unknown roff macro", or something similar. I think having a >bare-bones nroff is a good idea, but it would be nice to be able to read or >convert not-so-simple man pages. On a 386, it was a real feature that "roff" (which is a clone of the original; NROFF was a spin-off of it, thus "new roff") was much, much faster than GNU [nt]roff. With even the smallest machines sporting Pentium II's and 16 megs of RAM, perhaps the time has come to take on some more overhead on this front. I'm open to suggestions. >Speaking of mans, has anyone given any >thought to the structure of documentation under vsta? As long as we're >breaking from tradition, why not give some thought to the online documentation >situation? I'm not a big fan of nroff (it's hard to read plaintext, it has >developed lots of hackery over the years, etc.) myself. I'm not sure what >could be used as a replacement, but my only requirement would be that it >produces good console output (so tex + tty2dvi is out), and is easily >greppable. I don't like having docs where the whole thing is organized around >hypertext, so you have to _find_ all the documentation, but I think hyperlinks >within a page could be very handy. I guess another possibility is TeXinfo, >which presumably produces nice dvis as well as info files. I disagree with th >e >gnu info reader, but given a decent reader and a linear document format, info >might not be too bad. Of course, a solution that lets one use plain text >pagers like less and more would be the best. So maybe nroff is doing pretty >well after all. Plain text with no markup is fine, too. I have always though >t >that man should be more general: it should be able to look up arbitrary pages, >and execute arbitrary commands on them. So for traditional man pages, it woul >d >run roff. For text files (and stray cats), it would run the pager with no >preprocessing. For .tex it would run tex or latex and, err... madodvi the >output. Ok, so I'm getting a bit ahead of things here :) I just think it >would be nice to have a unified, well defined thing to do with documentation, >that can be easily indexed and read by a user. The zsh people are using >something called yodl which can compile to nroff, TeX, info, html, and maybe >some more. Then of course there's sgml, but I haven't seen any console >viewers for that. And then html, which I haven't seen any good console >viewers for either (don't like lynx). I like lynx pretty well, but it *is* a little, um, rustic. These days I can't imagine building a hypertext structure using anything but HTML. Perhaps HTML generated by some other tool and source language, but HTML nonetheless. >... For example, all the makefiles i >n >the src/ dirs want to link with libs in ../../lib. Unfortunately, given >/vsta/src/bin/whatever, that doesn't work to well. I thought "oh, I'm suppose >d >to put srcs in the root", Hmmm, I thought I fixed all of that. Relative paths should point at the right thing, unless I made a mistake. Note that especially in the src/bin area, include/ shifted "up" a level relative to them several years ago. > BTW again, why do all these programs have to >manually link with crt0.o and -I/vsta/include? Couldn't gcc's specs file >(which is undocumented as far as I can tell, I can't figure it out) do this >automatically? Mostly, history. In the case of libraries, it's intended that you build against the libraries freshly cooked from source. Since the advent of shared libraries, this doesn't make all that much sense any more. >From what I can tell, vstafs is nothing to rely on right now. >There's also this bfs thing (boot file system?). It looks like it's not reall >y >intended for use on the root partition. So is everyone just using old 8.3, no >owner/permissions FAT? Everyone seemed to agree that ext2 would be easy to >port, but it seems that the only port is currently alpha >dont-use-this-if-you-value-your-data. It would be nice to have a well tested, >high performance filesystem for vsta. I'd imagine extensions to support vsta >style file permissions wouldn't be hard, I think the hurd adds some extensions >of it's own as well. Perhaps people would be interested in (if they don't >already know of) the reiserfs, a high performance file-system being developed. >There is info at http://www.idiom.com/~beverly/reiserfs.html. I'm not sure >how well it would work with Andy's "reset at any time" policy, though. :->. Yeah, somebody asked me if I designed my car so I crash into it every night. No, but if I crashed my car as often as I crash my computer, I'd have to see about it! >I've had some crashes when attempting to move directories. I'll do >something like "mv src .." and that vc will freeze. All the other vcs are >fine, until I type something and press enter, at which point they freeze too. On CONS:0 (Alt-F1) you should see a server croaking. Probably DOS. I'm not aware of any outstanding problems in this area, but many of these are dependent on the geometry of your filesystem. >Speaking of freezing vcs, does vsta (cons?) have type-ahead? Looking at a >struct screen s.s_buf[], it apppears it does, but you can't see it. I'm used >to typing one command while the first is completing, and I find it >disconcerting if the characters don't immediately appear. Would it be a >simple modification to have the cons server put all keystrokes on the screen >immediately, or is there no way to do it unless the app at the other end >does a read? All cooked mode processing (including editing, echoing, and so forth) is done by getline() within libc. The console server doesn't know about TTY modes, editing, or anything. If you really wanted to do flashy hacks, you could have a typeahead "popup" show up as you typed ahead.... It'd have to be behind a command switch, since I know it would drive *me* crazy. >Thank you (Andy?) very much for the vim port, I'm glad to see my favorite >editor in vsta. Too bad it's an ancient version. Maybe the person who ported >it could tell me what exactly was involved in doing it, so I could maybe try >and port a new version (5.0something as of now, but I'm ignoring 5.0 in favor >of 4.x)? You're welcome. It's my favorite editor, too. I don't recollect anything earth-shaking, and a port should be much easier now, since a lot more of VSTa exists than back then. The one real bug I had to hunt was a place where vim was accessing a buffer after free()'ing it. It made regular expressions work incorrectly, I think. I can't believe they haven't found this problem and fixed it by now. > Speaking of getline, I think it's mucking some things >up: if I do "cat >foo" in the interests of making a file, I get "getline(): no >t >interactive" or something and my prompt back. I had a problem with window resizing in the same area. getline() should probably be taught how to try stdin or stderr if stdout appears redirected. >Speaking of porting, it would be nice if porters made a little document that >describes changes they had to make to the program source. That way, when >someone wants to port a new version, or a similar command, all that work isn't >duplicated. Maybe it could even be sent to the program mainainer to put >in the official release. Guilty as charged. All I can tell you is that many of those commands ported as I wrote missing pieces of libc, found kernel bugs and/or bugs in the parts of libc I had previously written, and tried to guess whether it would be easier to hack the ported source, or hack the system. >Who handles #! in vsta? When using rc, I get an exec format error, whereas >sh handles it ok. There is no mention of #! anywhere in sh.c, nor could >I find anything in kern/exec.c (or anywhere else). It's in the shell. See the exec.c area. >I wonder what's happening with MADO? Is it still happening, slowly? Last I >heard it sounded like significant work was done, but proceding at a leisurely >pace. I'm afraid it's dead. I would cheerfully be proved wrong. >Oh, one more thing. What I am ultimately interested is a midi sequencer. I >have had plans to write one of my own, but unfortunately the existing midi >drivers under linux are limited in accuracy to 1/100 sec. I was wondering >what kind of accuracy could be implemented in vsta. I know it's not a real >time system, but given a need to tag incoming data with timestamps, how >accurate could vsta be? Could we give the server a nice high priority so it >would have lots of interrupts to service the midi port? In the vsta readme >Andy said he wanted "enough RT support to do com port drivers", which implies >yes. Probably you could. I'd guess that in addition to low latency, you need a high resolution clock. You could build a kernel with HZ set to 1000, and accept a little more overhead, for instance. Or on Pentium and later, there are CPU resources which help with fine resolution timing. >BTW, how do you nice things under vsta? The docs mentioned that, given >it's leaf/node scheduling algorithm, you could gaurantee a certain process >group would never use more than, say, 50% cpu time. How would you do this? Heh. Write a UI, and then exercise the kernel mechanism. During bringup I hacked it to see if it would work (it did), but it never became pressing enough to code up. It'd probably be by way of the scheduler control syscall. >Something I've always thought unix should have is a major page fault limit: >limit the number of pages of swap a process can pull in every second. That >would alleviate the problem of one misbehaved process thrashing the disk >and cheating everyone else, while the cpu is sitting around at 3% (I see >a lot of this, since I have 6MB RAM). There was a paper which described an "inactivated" state which controlled the damage a single thrashing process could cause. It was present in HP/UX, for instance. It mostly helped for those then-popular test programs which malloc()'ed a maximum size VM area, then modified it randomly. >I also had a crash: I compiled the libc, and it got to the mkshlib part. It >printed two lines (Library libc.a @ 0x60001000) and froze. After a bit, I >eventually got: >Assertion Failed line 323 file ../kern/vm_page.c >set_core: bad index >and was dumped into the kernel debugger. I don't know many commands in >the kdb (quit and proc :) ), so I reset eventually. Every time I try to >compile the libc, it crashes at that same line, although in different ways >(sometimes it goes into ignore-the-user mode). Wow! That's a pretty serious crash, and not one I recognize. Can you get a "bt" output, and send it to me? >I've got vim 4.2 working except one thing: vsta doesn't seem to have select() >(or poll(), I guess that's a systemV thing), and vim wants to use that for, >well, waiting on fds. Implementing select() requires an async messaging connection to the servers of your target FD's. So until a datagram service is put in the kernel, the best you can do is poll the target FD's yourself. Now, an immediate question for this port is why is there a need to block on an arbitrary set of file descriptors? In my port of MGR I converted to using threads, and ended up with a much more efficient implementation. >btw again, vsta's cpp doesn't seem to be the same as the gnu one. It chokes o >n >certain #if lines, which makes porting rather more difficult. No, it's the standard 2.x one which matches the rest of the tools. Use -v (or was it -V?) to see who's invoking what. Regards, Andy From daemon Tue Nov 18 02:12:40 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id CAA10865 for vsta-xplod; Tue, 18 Nov 1997 02:12:40 GMT Received: from bach.videotron.net (bach.videotron.net [205.151.222.10]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id CAA10859 for ; Tue, 18 Nov 1997 02:12:37 GMT From: dufrp@oricom.ca Received: from oricom.ca (ppp222.oricom.ca [206.108.60.222]) by bach.videotron.net (8.8.5/8.8.2) with SMTP id HAA20978 for vsta@zendo.com; Tue, 18 Nov 1997 07:13:24 -0500 (EST) Date: Tue, 18 Nov 1997 07:13:24 -0500 (EST) Message-Id: <199711181213.HAA20978@bach.videotron.net> Subject: Is my computer too small for VSTa? >>I also had a crash: I compiled the libc, and it got to the mkshlib part. It >>printed two lines (Library libc.a @ 0x60001000) and froze. After a bit, I >>eventually got: >>Assertion Failed line 323 file ../kern/vm_page.c >>set_core: bad index >>and was dumped into the kernel debugger. I don't know many commands in >>the kdb (quit and proc :) ), so I reset eventually. Every time I try to >>compile the libc, it crashes at that same line, although in different ways >>(sometimes it goes into ignore-the-user mode). >Wow! That's a pretty serious crash, and not one I recognize. Can you get a >"bt" output, and send it to me? I had the same problem with version 1.6. whatever I was compiling. I was thinking that it was either because my computer was too small (2Mb extended memory+256k which is remapped from the first Mb by the motherboard -386sx16Mhz WITHOUT ANY 387-). Because of this I have returned to 1.5.2. and now I remember why I don't use it. Because gcc systematically hangs the system after 5 to 15 seconds. In the first 5 to 10 seconds HD light flash normally, but after that nothings happen (I have never waited more than about 10 minutes). When the system hangs I can switch to others consoles, I can enter a login name but nothings happens when I hit return. I can always go to the debugger screen however. Maybe there is a port of a language that would work in the version 1.6 on my small machine? ---Paul Dufresne From daemon Tue Nov 18 06:08:05 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id GAA11167 for vsta-xplod; Tue, 18 Nov 1997 06:08:05 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id GAA11161 for ; Tue, 18 Nov 1997 06:08:01 GMT Received: from localhost (vandys@localhost) by zipper.cisco.com (8.8.4-Cisco.1/8.6.5) with SMTP id IAA10724; Tue, 18 Nov 1997 08:08:01 -0800 (PST) Message-Id: <199711181608.IAA10724@zipper.cisco.com> To: Adrian Chadd cc: dufrp@oricom.ca, vsta@zendo.com Subject: Re: Is my computer too small for VSTa? In-reply-to: Your message of "Tue, 18 Nov 1997 21:07:02 +0800." <199711181307.VAA01463@mail.creative.net.au> Date: Tue, 18 Nov 1997 08:08:01 -0800 From: Andrew Valencia [Adrian Chadd writes:] >>motherboard -386sx16Mhz WITHOUT ANY 387-). >Funny, I had the same problem. >GCC uses FPU to do some optimisations. So is 1.6 crashing when it tries to use the FPU? I would like to fix that. Sadly, I no longer own a 386 system (or, for that matter, a system with a CPU which doesn't have an FPU). >I think I might just download vsta 1.6 and try porting a math emulator. >Unless someone has already done it. :-) Nope, nobody's ever even started on such a project. There's some OK emulation stuff out on the net, so the only interest part is how to connect it in. The most trivial would be to make it a library and use gcc's option to turn FPU operations into function calls. Of course, then you need a special version of each program. It really shouldn't go in the kernel (this *is* a microkernel!), so the next possibility would be to have the C startup register an FPU emulation trampoline. Then when the kernel saw an FPU fault on a system without FPU, and saw that the program had registered code for this, it would turn it into a event handler, much like the usual even delivery mechanism. I'll do the kernel part, if you'd prefer. >Andy - have you actually put it anywhere that you need an FPU to do stuff ? Yes, see the latest release documents for 1.6. It's in both the per-release "features" (doc/features.txt) as well as the "getting started" document (doc/roadmap.txt). >I'm looking at revamping vsta's IP stack and network card support to make >some mini self-contained WAN bridges/routers, and using 386's wout FPUs :) VSTa itself does not require an FPU, and never will. It's the way gcc is written that requires the FPU. If it *really* bothers somebody, you might go look at how it's used, and then look at how it was changed in the old 1.x VSTa port (I stole the technique from an old port of djgpp), and see if you couldn't patch it out again. That'd be fine with me. Andy From daemon Fri Nov 21 22:37:14 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id WAA16626 for vsta-xplod; Fri, 21 Nov 1997 22:37:14 GMT Received: from bach.videotron.net (bach.videotron.net [205.151.222.10]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id WAA16620 for ; Fri, 21 Nov 1997 22:37:10 GMT From: dufrp@oricom.ca Received: from oricom.ca (ppp246.oricom.ca [206.108.60.246]) by bach.videotron.net (8.8.5/8.8.2) with SMTP id DAA08505 for vsta@zendo.com; Sat, 22 Nov 1997 03:38:08 -0500 (EST) Date: Sat, 22 Nov 1997 03:38:08 -0500 (EST) Message-Id: <199711220838.DAA08505@bach.videotron.net> Subject: Some questions About the bug I had (assertion failed line 323 of vm_page.c) like Andy told me it seems to happens on the last page available to the system, at least I can say that when it happens and I type 'dv freemem' in the debugger it says 0 (no more memory). What I dont likes is that often when I quit the debugger (always if I quit the debugger after having typed 'dv freemem') and that I try to go to an other console (Alt-F2) my system crashed saying that boot process 0 died. I didn't knew I had a boot process but it make me sad to learn that it died. :) An other thing I don't understand is why I miss so easily memory (I have 2Mb physical extended). But why virtual memory is not used on my dos partition to compensate for my little physical memory? About 'di'ssassemble command in the debugger... /doc/roadmap.txt explains: >... You can look at the current >user process' address space by OR'ing in 0x80000000 to the user address. But how *exactly* can I get this user address for a specific process? ---Paul Dufresne From daemon Sat Nov 22 21:23:15 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id VAA20122 for vsta-xplod; Sat, 22 Nov 1997 21:23:15 GMT Received: from bach.videotron.net (bach.videotron.net [205.151.222.10]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id VAA20116 for ; Sat, 22 Nov 1997 21:23:11 GMT From: dufrp@oricom.ca Received: from oricom.ca (ppp227.oricom.ca [206.108.60.227]) by bach.videotron.net (8.8.5/8.8.2) with SMTP id CAA22187 for vsta@zendo.com; Sun, 23 Nov 1997 02:24:14 -0500 (EST) Date: Sun, 23 Nov 1997 02:24:14 -0500 (EST) Message-Id: <199711230724.CAA22187@bach.videotron.net> Subject: How can I setup a swap partition? >>An other thing I don't understand is why I miss so easily memory (I have >>2Mb physical extended). But why virtual memory is not used on my dos >>partition to compensate for my little physical memory? >No swapping occurs by default. You need to configure the swap daemon, and >explicitly point him at a swap partition. Nice, I think I really need to do this. Could someone give me the steps needed to make this? ---Paul Dufres From daemon Sun Nov 23 12:16:04 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id MAA21157 for vsta-xplod; Sun, 23 Nov 1997 12:16:04 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id MAA21151 for ; Sun, 23 Nov 1997 12:16:01 GMT Received: from localhost (vandys@localhost) by zipper.cisco.com (8.8.4-Cisco.1/8.6.5) with SMTP id OAA01711; Sun, 23 Nov 1997 14:17:07 -0800 (PST) Message-Id: <199711232217.OAA01711@zipper.cisco.com> To: dufrp@oricom.ca Cc: vsta@zendo.com Subject: Re: How can I setup a swap partition? In-reply-to: Your message of "Sun, 23 Nov 1997 02:24:14 EST." <199711230724.CAA22187@bach.videotron.net> Date: Sun, 23 Nov 1997 14:17:07 -0800 From: Andrew Valencia [dufrp@oricom.ca writes:] >>No swapping occurs by default. You need to configure the swap daemon, and >>explicitly point him at a swap partition. >Nice, I think I really need to do this. >Could someone give me the steps needed to make this? First, you have to have a disk partition available. Let's say it has a partition type of "Linux swap", 0x82, so it'd be known as wd0_swap0 under the first IDE drive. In your menu.lst file, you have to add loading of the /vsta/boot/swap. No command line options. With this, swapping is possible, though inactive. Once booted, you need to run /vsta/bin/swapd. The argument is either "-n", which only steals clean text pages (i.e., it flushes things out of the page cache), or a swap partition (in "disk/wd:wd0_swap0" format), or no arguments (just starts the machinery, without adding a swap partition). From looking at the code, command line parsing may have gone quietly broken for the case of passing in the swap partition name. I suggest that life will be more than interesting enough with just clean text page stealing. Consider building a kernel with the code under a WATCHMEM ifdef in os/kern/vm_steal.c enabled. This'll give you a feel for whether the page stealer is running and/or finding memory to steal. The page stealing functions work OK, but the last time we hacked on this code there were still clear signs that you could deadlock in the VM system. There were scenarios where stealing pages was going to take some memory (to support the needed exchange of microkernel messages) when none was available. If your memory use doesn't plunge the freemem count down to 0 before more pages can be stolen, you'll probably be OK. VSTa and even MGR are small enough that you don't need VM to do C development on even an 8 meg system. Trying to run gcc 2.X (and maybe gdb, and MGR) will stress things in brand new ways, so don't be too surprised if you're breaking new ground (and debugging the fallout). Andy From daemon Fri Nov 28 12:48:28 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id MAA28570 for vsta-xplod; Fri, 28 Nov 1997 12:48:28 GMT Received: from baruk2.zendo.com (baruk2.zendo.com [205.187.71.65]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id MAA28564 for ; Fri, 28 Nov 1997 12:48:24 GMT Received: from localhost.zendo.com (localhost.zendo.com [127.0.0.1]) by baruk2.zendo.com (8.8.5/8.6.12) with SMTP id PAA17909 for ; Fri, 28 Nov 1997 15:04:20 -0800 (PST) Message-Id: <199711282304.PAA17909@baruk2.zendo.com> X-Authentication-Warning: baruk2.zendo.com: localhost.zendo.com [127.0.0.1] didn't use HELO protocol To: vsta@zendo.com Subject: VSTa happenings Date: Fri, 28 Nov 1997 15:04:20 -0800 From: Andy Valencia Hello, I got to spend some hours on VSTa this weekend, and fixed a number of mouldering bugs. First, KA9Q was locking up if you tried to turn on a physical interface (looping packets to itself works fine, for all the good that'd do you...). Fixed; it was a regression caused by fixing some edge conditions with when it chooses to run its scheduling loop. I turned on its packet tracing, and saw that it was ARP'ing for very nearly every packet out of the box. Hunted this back to a bug with how it ran its timer for ARP entries. Fixed. With these, it's running reasonably well. FreeBSD's complaining on the PORT command it generates (FTP), so I have a BPF tracing kernel ready to boot on the box so I can see what's going wrong on the BSD side. A long standing (as in, from the original port) bug in the Bourne shell port was finally exterminated. There was an edge condition in the local shell handling of variables which it had received out of the /env server. Fixed. This would show up as command in a shell script which would fail with something like ": not found" even though they were valid commands in your PATH. If you run with the standard path, you probably wouldn't see this. And the emulation of environ[] (the UNIX thing which kind of assumes that the environment is static, sigh) had a bug where a junk character could end up at the end of some array entries. Missing null termination, oops. Fixed. Raw booting is working OK. I had occasion to make myself a boot floppy which ran a menu.lst and secondary loader off of the hard disk. Works fine, and the GRUB instructions make it easy to do. gdb is working very well! I just kick myself for not doing the port a LOT sooner. Hope some of you break it out and give it a spin for your own code. Regards, Andy From daemon Sun Nov 30 13:12:04 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id NAA03506 for vsta-xplod; Sun, 30 Nov 1997 13:12:04 GMT Received: from atrey.karlin.mff.cuni.cz (root@atrey.karlin.mff.cuni.cz [195.113.31.123]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id NAA03500 for ; Sun, 30 Nov 1997 13:11:59 GMT Received: from Elf.mj.gts.cz (root@slip8.ms.mff.cuni.cz [195.113.20.208]) by atrey.karlin.mff.cuni.cz (8.8.5/8.8.5) with ESMTP id AAA08357 for ; Mon, 1 Dec 1997 00:13:43 +0100 Received: (from pavel@localhost) by Elf.mj.gts.cz (8.8.5/8.8.5) id WAA00244; Sun, 30 Nov 1997 22:19:53 +0100 Message-ID: <19971130221953.51362@Elf.mj.gts.cz> Date: Sun, 30 Nov 1997 22:19:53 +0100 From: Pavel Machek To: vsta@zendo.com Subject: Select behaviour Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.81 X-Warning: Not only using Windows can be dangerous to your mental health. Hi! When talking with my friend (mj) about mixie implementation, we got to select() behaviour. Select behaviour is pretty strange, as it can be turned into attempt to read/write 0 bytes, which is kind of braindamage we do not want in mixie. How is this solved in VSTa? I looked for select in lib/ and found nothing. (Maybe I'm just plain blind.) Pavel -- I'm really pavel@atrey.karlin.mff.cuni.cz. Pavel Look at http://atrey.karlin.mff.cuni.cz/~pavel/ ;-). From daemon Mon Dec 1 06:18:48 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id GAA04716 for vsta-xplod; Mon, 1 Dec 1997 06:18:48 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id GAA04710 for ; Mon, 1 Dec 1997 06:18:46 GMT Received: from localhost (vandys@localhost) by zipper.cisco.com (8.8.4-Cisco.1/8.6.5) with SMTP id IAA21431; Mon, 1 Dec 1997 08:18:51 -0800 (PST) Message-Id: <199712011618.IAA21431@zipper.cisco.com> To: Pavel Machek cc: vsta@zendo.com Subject: Re: Select behaviour In-reply-to: Your message of "Sun, 30 Nov 1997 22:19:53 +0100." <19971130221953.51362@Elf.mj.gts.cz> Date: Mon, 01 Dec 1997 08:18:51 -0800 From: Andrew Valencia [Pavel Machek writes:] > When talking with my friend (mj) about mixie implementation, we got >to select() behaviour. > Select behaviour is pretty strange, as it can be turned into attempt >to read/write 0 bytes, which is kind of braindamage we do not want in >mixie. How is this solved in VSTa? I looked for select in lib/ and >found nothing. (Maybe I'm just plain blind.) I never coded select(), primarily because it's hard to make it scale, especially in a preemptive and SMP environment. For cases like MGR and KA9Q ports, I added multiple threads to them, with each thread handling a particular kind of resource. I then used a global mutex to protect the main program, which continued to be single-threaded. We're getting to the point where we could take a shot at select() again, if somebody wants to take it on.... Speaking of KA9Q, over the weekend I debugged why FTP's to another host on the local LAN were so slow... the NE driver was working fine, but it would overrun the ethernet thread in KA9Q. I added some buffering to NE, and now transfers zip right along. I'll push out a 1.6.1 pretty soon. Andy From daemon Wed Dec 17 15:17:22 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id PAA20556 for vsta-xplod; Wed, 17 Dec 1997 15:17:22 GMT Received: from note.orchestra.cse.unsw.EDU.AU (root@note.orchestra.cse.unsw.EDU.AU [129.94.242.29]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id PAA20550 for ; Wed, 17 Dec 1997 15:17:19 GMT Received: From ives With LocalMail ; Thu, 18 Dec 97 12:20:02 +1100 From: s2156593@cse.unsw.edu.au (Binh Thai) To: vsta@zendo.com Date: Thu, 18 Dec 97 12:19:44 +1100 Message-Id: <971218011944.7816@cse.unsw.edu.au> Subject: screen capture in MGR Hi everyone, Is there anyone tried doing a screen capture in MGR? If so, can you kindly tell me how to do it? Thank you. Binh Thai. From daemon Thu Dec 18 11:19:44 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id LAA22173 for vsta-xplod; Thu, 18 Dec 1997 11:19:44 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id LAA22167 for ; Thu, 18 Dec 1997 11:19:42 GMT Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id NAA18268; Thu, 18 Dec 1997 13:21:57 -0800 (PST) Message-Id: <199712182121.NAA18268@vandys-pc.cisco.com> To: s2156593@cse.unsw.edu.au (Binh Thai) cc: vsta@zendo.com Subject: Re: screen capture in MGR In-reply-to: Your message of "Thu, 18 Dec 1997 12:19:44 +1100." <971218011944.7816@cse.unsw.edu.au> Date: Thu, 18 Dec 1997 13:21:57 -0800 From: Andy Valencia [s2156593@cse.unsw.edu.au (Binh Thai) writes:] > Is there anyone tried doing a screen capture in MGR? If so, can you >kindly tell me how to do it? I've never tried it. I assume you'd use m_bitget(), and save off the image in that raw format, or maybe you could marry it with the PBM tools if you wanted a compressed format. I don't see any applications for it in the MGR distribution. Andy From daemon Sun Dec 21 18:21:57 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id SAA29588 for vsta-xplod; Sun, 21 Dec 1997 18:21:57 GMT Received: from baruk2.zendo.com (baruk2.zendo.com [205.187.71.65]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id SAA29582 for ; Sun, 21 Dec 1997 18:21:54 GMT Received: from localhost.zendo.com (localhost.zendo.com [127.0.0.1]) by baruk2.zendo.com (8.8.5/8.6.12) with SMTP id UAA19355 for ; Sun, 21 Dec 1997 20:28:23 -0800 (PST) Message-Id: <199712220428.UAA19355@baruk2.zendo.com> X-Authentication-Warning: baruk2.zendo.com: localhost.zendo.com [127.0.0.1] didn't use HELO protocol To: vsta@zendo.com Subject: Happenings with VSTa Date: Sun, 21 Dec 1997 20:28:23 -0800 From: Andy Valencia Hello, This is being typed over a telnet link over KA9Q from an MGR window. KA9Q with an ether interface is back in order, and some enhancements to the NE2000 driver really spiffed up the performance. I also finally got around to converting all the TTY stuff to permit keyboard interrupt generation, so (finally... I know it's been a long wait) you can now type control-C (or whatever you set your INTR and QUIT characters to) and have your application interrupt. Internally it's implemented with the POSIX termios emulation detecting underlying changes and sending FS_WSTAT messages to the TTY server. The server records these, and thus knows if the application is in a TTY mode which calls for signals. If a signal should be sent, it is sent to the process group specified by another FS_WSTAT (bin/login does this for you as you log in, for instance). For protection, the TTY server also records your capabilities at the time you set a process group, and assumes only those capabilities at the time it sends the signal. Thus, you can't point the TTY server at, say, the disk server, and then hit ^C and watch it croak. I'll bundle up a 1.6.1 RSN. I think I've laid to rest all the gotchas from 1.6. Andy From daemon Mon Dec 22 18:44:37 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id SAA01700 for vsta-xplod; Mon, 22 Dec 1997 18:44:37 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id SAA01694 for ; Mon, 22 Dec 1997 18:44:35 GMT Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id UAA01783 for ; Mon, 22 Dec 1997 20:47:07 -0800 (PST) Message-Id: <199712230447.UAA01783@vandys-pc.cisco.com> To: vsta@zendo.com Subject: VSTa 1.6.1 available Date: Mon, 22 Dec 1997 20:47:07 -0800 From: Andy Valencia I've just dropped 1.6.1 onto ftp://ftp.zendo.com/vsta/current. 1.6.1 has a number of networking fixes, the new TTY interrupt code, and fixes some other misfeatures of 1.6. See vsta/doc/features.txt for more detail. Enjoy! Andy Valencia From daemon Tue Dec 23 12:40:56 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id MAA03182 for vsta-xplod; Tue, 23 Dec 1997 12:40:56 GMT Received: from baruk2.zendo.com (baruk2.zendo.com [205.187.71.65]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id MAA03176 for ; Tue, 23 Dec 1997 12:40:50 GMT From: pavel@atrey.karlin.mff.cuni.cz Received: (from root@localhost) by baruk2.zendo.com (8.8.5/8.6.12) id OAA22235; Tue, 23 Dec 1997 14:48:04 -0800 (PST) Date: Tue, 23 Dec 1997 14:48:04 -0800 (PST) Message-Id: <199712232248.OAA22235@baruk2.zendo.com> To: vsta@zendo.com Subject: Grub patch (slightly offtopic) Hi! When I heard that new vsta is moving tovards grub, I decided to make grub actually usable. (I really did not like its old interface). So I made following patch which might be quite interesting. Sorry for being slightly offtopic, and merry chrismas, Pavel (Hmm, is there better place for this mail?) -- -- I'm really pavel@atrey.karlin.mff.cuni.cz. Pavel Look at http://atrey.karlin.mff.cuni.cz/~pavel ;-). Hi! I patched grub to make it compile (for me) and to make it user interface better. (Much better in my eyes, now grub is usable for me.) Timeout=0 now has good sense, and only key in buffer or holding of shift will stop booting sequence. You can jump directly to entry in menu by selecting first character of its title. Space is bind to editing of second line of current config entry - this is for lilo-like editing of command line. Share and enjoy! Pavel PS: I know some of my changes are dirty hack which should be implemented other way. If you are willing to accept changes (and tell me how to code that better in one case), I'll clean it up. diff -ur 0.4-grub/configure grub-mine/configure --- 0.4-grub/configure Mon Oct 7 06:45:42 1996 +++ grub-mine/configure Tue Aug 5 02:10:32 1997 @@ -1,4 +1,5 @@ #!/bin/sh +PATH=$PATH:. # # My mini-configure, used only for: @@ -374,7 +370,7 @@ for link_addr in $addrs_to_check ; do echo "Linking test_asm.o with ${LD} at $link_addr" >> ../config.log - ${LD} -Ttext $link_addr -o test_asm test_asm.o >> ../config.log 2>&1 + ${LD} -N -Ttext $link_addr -o test_asm test_asm.o >> ../config.log 2>&1 check_exit_status ${LD} echo "Running test_offset on test_asm" >> ../config.log @@ -439,6 +435,5 @@ # Delete test directory # -rm -rf .conf_test_dir diff -ur 0.4-grub/docs/commands.txt grub-mine/docs/commands.txt --- 0.4-grub/docs/commands.txt Mon Oct 7 09:32:27 1996 +++ grub-mine/docs/commands.txt Tue Aug 5 03:28:07 1997 @@ -20,7 +20,7 @@ -- Options are separated by spaces. - -- All numbers can be either decimal or hexidecimal. A hexidecimal number + -- Any number can be either decimal or hexadecimal. A hexadecimal number must be preceeded by "0x", and is case insensitive. -- Extra options/text at the end of the line is ignored unless otherwise @@ -208,5 +208,4 @@ (so "sector" is offset from the start of the disk). Filesystem test mode is turned off by any uses of the "install=" or - "testload=" commands. - + "testload=" commands. \ No newline at end of file diff -ur 0.4-grub/shared_src/asm.S grub-mine/shared_src/asm.S --- 0.4-grub/shared_src/asm.S Sun Sep 29 02:35:28 1996 +++ grub-mine/shared_src/asm.S Sun Dec 21 23:49:47 1997 @@ -120,9 +120,10 @@ call EXT_C(prot_to_real) /* - * This next part is sort of evil. It takes advantage of the - * byte ordering on the x86 to work in either 16-bit or 32-bit - * mode, so think about it before changing it. + * ? This next part is sort of evil. It takes advantage of the + * ? byte ordering on the x86 to work in either 16-bit or 32-bit + * ? mode, so think about it before changing it. + * XXX comment seems outdated to me, pavel@atrey.karlin.mff.cuni.cz */ ENTRY(hard_stop) @@ -639,7 +640,7 @@ push %eax push %ebx - movl 0x10(%esp), %bl + movb 0x10(%esp), %bl /* if not '\n', just print the character */ cmpb $0xa, %bl @@ -1373,6 +1374,11 @@ jmp pending notpending: + movb $0x2, %ah + int $0x16 + testb $0x43, %al + mov $0x0, %ebx + jnz pending movl $0xFFFFFFFF, %ebx pending: diff -ur 0.4-grub/shared_src/char_io.c grub-mine/shared_src/char_io.c --- 0.4-grub/shared_src/char_io.c Thu Sep 19 11:03:58 1996 +++ grub-mine/shared_src/char_io.c Tue Aug 5 13:22:05 1997 @@ -130,9 +130,7 @@ void init_page(void) { - cls(); - - printf("\n GRUB version %s (%dK lower / %dK upper memory)\n\n", + printf("Grub V%s (mem %dK/%dK)", version_string, mbi.mem_lower, mbi.mem_upper); } @@ -180,7 +178,7 @@ void cl_init() { /* distinguish us from other lines and error messages! */ - putchar('\n'); + /* putchar('\n'); - better not, it looks horibly */ /* print full line and set position here */ ystart = (getxy() & 0xFF); Only in grub-mine/shared_src: char_io.c~ diff -ur 0.4-grub/shared_src/cmdline.c grub-mine/shared_src/cmdline.c --- 0.4-grub/shared_src/cmdline.c Thu Oct 3 17:26:24 1996 +++ grub-mine/shared_src/cmdline.c Sun Dec 21 21:14:47 1997 @@ -77,9 +77,7 @@ void init_cmdline(void) { - printf(" [ Minimal BASH-like line editing is supported. For the first word, TAB - lists possible command completions. Anywhere else TAB lists the possible - completions of a device/filename. ESC at any time exits. ]\n"); + printf(" (tab help, esc exit)\n" ); } @@ -162,6 +160,7 @@ if (!script) { + cls(); init_page(); init_cmdline(); } @@ -203,7 +202,7 @@ print_error(); } - if (run_cmdline && get_cmdline("command> ", commands, cur_heap, 2048)) + if (run_cmdline && get_cmdline("Grub: ", commands, cur_heap, 2048)) return 1; if (strcmp("boot", cur_heap) == 0 || (script && !*cur_heap)) Only in grub-mine/shared_src: cmdline.c~ diff -ur 0.4-grub/stage1/Makefile grub-mine/stage1/Makefile --- 0.4-grub/stage1/Makefile Wed Sep 18 05:13:00 1996 +++ grub-mine/stage1/Makefile Tue Aug 5 01:54:41 1997 @@ -13,7 +13,6 @@ $(LD) -N -Ttext 7C00 -o stage1.exec stage1.o dd if=stage1.exec of=stage1.noheader bs=$(HEADER_7C00) skip=1 dd if=stage1.noheader of=../bin/stage1 bs=512 count=1 - rm stage1.exec stage1.noheader clean: rm -f stage1.o Only in grub-mine/stage1: Makefile~ diff -ur 0.4-grub/stage1/stage1.S grub-mine/stage1/stage1.S --- 0.4-grub/stage1/stage1.S Wed Sep 18 08:51:49 1996 +++ grub-mine/stage1/stage1.S Tue Aug 5 03:07:49 1997 @@ -44,8 +44,8 @@ #define CHAR_a 'a' /* messages displayed at all times */ -#define CHAR_S 'S' -#define CHAR_J 'J' +#define CHAR_S '.' +#define CHAR_J '.' /* error messages */ Only in grub-mine/stage1: stage1.S~ Only in grub-mine/stage1: stage1.exec Only in grub-mine/stage1: stage1.noheader Only in grub-mine/stage1: stage1.o diff -ur 0.4-grub/stage2/Makefile grub-mine/stage2/Makefile --- 0.4-grub/stage2/Makefile Wed Sep 18 05:17:11 1996 +++ grub-mine/stage2/Makefile Tue Aug 5 02:35:50 1997 @@ -61,9 +61,8 @@ # "asm.o" absolutely has to come first in the link line! ../bin/stage2: $(SHARED_OBJS) $(OBJS) Makefile ../Makefile $(LD) -N -Ttext 8000 -o stage2.exec $(SHARED_OBJS) $(OBJS) - $(STRIP) stage2.exec - dd if=stage2.exec of=../bin/stage2 bs=$(HEADER_8000) skip=1 - rm stage2.exec +# $(STRIP) stage2.exec + dd if=stage2.exec of=../bin/stage2 bs=128 skip=1 clean: rm -f $(SHARED_OBJS) $(OBJS) diff -ur 0.4-grub/stage2/main.c grub-mine/stage2/main.c --- 0.4-grub/stage2/main.c Thu Sep 19 02:01:02 1996 +++ grub-mine/stage2/main.c Sun Dec 21 23:40:21 1997 @@ -136,7 +136,7 @@ void run_menu(char *menu_entries, char *config_entries, int num_entries, - char *heap, int entryno) + char *heap, int entryno, int xfer) { int c, time1, time2 = -1, first_entry = 0; char *cur_entry; @@ -151,22 +151,19 @@ first_entry++; entryno--; } + cls(); init_page(); print_border(3, 12); - printf("\n - Use the \x18 and \x19 keys for selecting which entry is highlighted.\n"); + printf("\n"); if (config_entries) - printf(" Press enter to boot the selected OS, \'e\' to edit the - commands before booting, or \'c\' for a command-line."); - else - printf( -" Press \'b\' to boot, enter to edit the selected command in the - boot sequence, \'c\' for a command-line, \'o\' to open a new line - after (\'O\' for before) the selected line, \'d\' to remove the - selected line, or escape to go back to the main menu."); + printf( "\n \x18, \x19 move, enter boot, end edit, tab command mode, space edit & launch." ); + else { + printf( "\n \x18, \x19 move, Boot, enter edit, tab command mode, esc back.\n" ); + printf( " INSert line, Insert line after, DELete line, esc back." ); + } print_entries(3, 12, first_entry, menu_entries); @@ -181,25 +178,10 @@ /* initilize to NULL just in case... */ cur_entry = NULL; - if (timeout >= 0 && (time1 = getrtsecs()) != time2 && time1 != 0xFF) + if (xfer || checkkey() != -1) { - if (timeout <= 0) - { - timeout = -1; - break; - } - - /* else not booting yet! */ - time2 = time1; - gotoxy(3, 22); - printf("The highlighted entry will be booted automatically in %d seconds. ", timeout); - gotoxy(74, 4+entryno); - timeout--; - } - - if (checkkey() != -1) - { - c = getkey(); + if (!(c = xfer)) + c = getkey(); if (timeout >= 0) { @@ -242,28 +224,26 @@ } } - c = ASCII_CHAR(c); - if (config_entries) { - if ((c == '\n') || (c == '\r')) + if ((ASCII_CHAR(c) == '\n') || (ASCII_CHAR(c) == '\r')) break; } else { - if ((c == 'd') || (c == 'o') || (c == 'O')) + if ((c == KEY_DELETE) || (c == KEY_INSERT) || (ASCII_CHAR(c) == 'i')) { set_line(4+entryno, 0x7); /* insert after is almost exactly like insert before */ - if (c == 'o') + if (ASCII_CHAR(c) == 'i') { entryno++; - c = 'O'; + c = KEY_INSERT; } cur_entry = get_entry(menu_entries, first_entry+entryno, 0); - if (c == 'O') + if (c == KEY_INSERT) { bcopy(cur_entry, cur_entry+2, ((int)heap) - ((int)cur_entry)); @@ -295,14 +275,14 @@ } cur_entry = menu_entries; - if (c == 27) + if (ASCII_CHAR(c) == 27) return; - if (c == 'b') + if (ASCII_CHAR(c) == 'b') break; } - if ((config_entries && (c == 'e')) - || (!config_entries && ((c == '\n') || (c == '\r')))) + if ((config_entries && ((c == KEY_END) || (ASCII_CHAR(c) == ' '))) + || (!config_entries && ((ASCII_CHAR(c) == '\n') || (ASCII_CHAR(c) == '\r')))) { int num_entries = 0, i = 0; char *new_heap; @@ -332,7 +312,11 @@ *(new_heap++) = 0; if (config_entries) - run_menu(heap, NULL, num_entries, new_heap, 0); + { + cls(); + run_menu(heap, NULL, num_entries, new_heap, + (ASCII_CHAR(c)==' '?1:0), (ASCII_CHAR(c)==' '?'\n':0)); + } else { cls(); @@ -366,16 +350,53 @@ bcopy(new_heap, cur_entry, j); heap += (j - i); + if (xfer) + { xfer = 0; break; } } + xfer = 0; } goto restart; } - if (c == 'c') + if (ASCII_CHAR(c) == 9) { + cls(); enter_cmdline(NULL, heap); goto restart; } + c = ASCII_CHAR(c) | 0x20; + if (config_entries && (((c > 'a') && (c < 'z')) || ((c > '0') && (c < '9')))) { + /* Unknown key, try comparing it to first char of title of each */ + int i; + char *cur_entry; + for(i=0; i= 0 && (time1 = getrtsecs()) != time2 && time1 != 0xFF) + { + if (timeout <= 0) + { + timeout = -1; + break; + } + + /* else not booting yet! */ + time2 = time1; + gotoxy(3, 22); + printf("The highlighted entry will be booted automatically in %d seconds. ", timeout); + gotoxy(74, 4+entryno); + timeout--; } } @@ -467,6 +488,7 @@ char *config_entries = (char *)(mbi.mmap_addr + mbi.mmap_length); char *menu_entries = (char *)(BUFFERADDR + (32 * 1024)); + printf( ". " ); /* * Here load the configuration file. */ @@ -554,8 +576,8 @@ * Run menu interface (this shouldn't return!). */ + cls(); run_menu(menu_entries, config_entries, num_entries, - menu_entries+menu_len, default_entry); + menu_entries+menu_len, default_entry, 0); + while(1); } - - From daemon Mon Dec 29 14:17:46 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id OAA15631 for vsta-xplod; Mon, 29 Dec 1997 14:17:46 GMT Received: from note.orchestra.cse.unsw.EDU.AU (root@note.orchestra.cse.unsw.EDU.AU [129.94.242.29]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id OAA15625 for ; Mon, 29 Dec 1997 14:17:43 GMT Received: From dvorak With LocalMail ; Tue, 30 Dec 97 11:18:14 +1100 From: s2156593@cse.unsw.edu.au (Binh Thai) To: vsta@zendo.com Date: Mon, 29 Dec 97 23:46:03 +1100 Message-Id: <971229124603.5795@cse.unsw.edu.au> To: vsta@zendo.com Subject: Found the real problem Hi everyone, I have finally worked out where are problem is regarding to the mouse driver :-) The mouse driver itself is fine. However, the problem occurs at the rs232 server. The problem I described in my previous posting was due to the fact that the rs232 was reading garbage values, such as a long sequence of hex 0 and hex 10, even when the mouse is not moving! this is why the cursor is not moving consistantly in MGR. My question is now this... Does anyone know why the rs232 server is not reading the mouse data correctly? Is there anyway to fix it? Thank you so much Binh Thai From daemon Mon Dec 29 14:17:01 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id OAA15623 for vsta-xplod; Mon, 29 Dec 1997 14:17:01 GMT Received: from note.orchestra.cse.unsw.EDU.AU (root@note.orchestra.cse.unsw.EDU.AU [129.94.242.29]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id OAA15617 for ; Mon, 29 Dec 1997 14:16:54 GMT Received: From dvorak With LocalMail ; Tue, 30 Dec 97 11:17:30 +1100 From: s2156593@cse.unsw.edu.au (Binh Thai) To: vsta@zendo.com Date: Sun, 28 Dec 97 21:33:14 +1100 Message-Id: <971228103314.5771@cse.unsw.edu.au> Subject: serial mouse driver Hi everyone, I have spent the last few days working on the mouse driver, and I continuously getting problems to work properly with MGR. The type of mouse I have is a 2 button microsoft model. BTW, I am still using VSTa 1.5.2. I have used the test program, and every so often a crazy reading is resulted. eg. it tells me i have pressed the left mouse button when I haven't. Anyway, The major problem is when I use MGR, the mouse movement is not consistant. By that I mean when I move the mouse to the left, the cursor on the screen sometimes moves the left correctly, sometimes it doesn't even move, and sometimes it thinks I've pressed the mouse button. Can anyone help me solve this problem? or has anyone got a version of mouse driver that works properly with a 2 button microsoft mouse with MGR? Any suggestions and / or solutions are greatly appreciated. Binh Thai From daemon Mon Dec 29 15:40:43 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id PAA15766 for vsta-xplod; Mon, 29 Dec 1997 15:40:43 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id PAA15760 for ; Mon, 29 Dec 1997 15:40:35 GMT Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id RAA15693; Mon, 29 Dec 1997 17:43:23 -0800 (PST) Message-Id: <199712300143.RAA15693@vandys-pc.cisco.com> To: s2156593@cse.unsw.edu.au (Binh Thai) cc: vsta@zendo.com In-reply-to: Your message of "Mon, 29 Dec 1997 23:46:03 +1100." <971229124603.5795@cse.unsw.edu.au> Date: Mon, 29 Dec 1997 17:43:23 -0800 From: Andy Valencia [s2156593@cse.unsw.edu.au (Binh Thai) writes:] > I have finally worked out where are problem is regarding to the mouse >driver :-) Note that I did a bunch of work to make the driver play better with some flavors of mouse; that would've shown up in 1.6. > The mouse driver itself is fine. However, the problem occurs at the >rs232 server. The problem I described in my previous posting was due to the >fact that the rs232 was reading garbage values, such as a long sequence of >hex 0 and hex 10, even when the mouse is not moving! this is why the cursor >is not moving consistantly in MGR. > My question is now this... Does anyone know why the rs232 server is not >reading the mouse data correctly? Is there anyway to fix it? You have to go back and see why your UART is telling the driver that there's receive data. Either it's because there *is* data streaming in on the port (if you have an RS-232 breakout box you can check this), or because the UART status is confused. For the latter, there needs to be both an interrupt (to create the M_ISR message to start servicing of the port), then the "receive ready" status needs to be set, and only then will the data port be read and the data queued for application consumption. One possiblity is that your mouse has "jitter" and generates an endless stream of single-bit delta changes. Especially if the UART isn't buffered, you can then have port overruns, and you start reading garbage from the port. You can "stat //tty/tty01" (or tty02 for the 2nd COM port) and check the error counters on the device. A couple overruns are not actually all that big a deal; the serial protocol is pretty forgiving. Andy From daemon Mon Dec 29 23:36:24 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id XAA16596 for vsta-xplod; Mon, 29 Dec 1997 23:36:24 GMT Received: from mail.creative.net.au (mail.creative.net.au [203.56.168.4]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id XAA16590 for ; Mon, 29 Dec 1997 23:36:19 GMT Received: from mail (localhost.creative.net.au [127.0.0.1]) by mail.creative.net.au (8.8.5/8.7) with ESMTP id RAA25749 for ; Tue, 30 Dec 1997 17:39:36 +0800 (WST) Message-Id: <199712300939.RAA25749@mail.creative.net.au> To: vsta@zendo.com Subject: 'dos' process dying while booting vsta 1.6.1 ? Date: Tue, 30 Dec 1997 17:39:33 +0800 From: Adrian Chadd Hi. I'm installing vsta 1.6.1 on a laptop I just bought. Trouble is, whenever I boot it , the thing starts booting, and then spits: Boot process 7 dies [Kernel debugger @ spl 0] > So I do a proc, and find: .. 7 dos 8 ONPROC kstack c0e000 uregs c0efc0 wchan 0 / Its juyst a DX4-100 laptop with 8mb RAM and a 500ish mbyte HDD. Its running FAT16. Does anyone have any ideas why its barfing? (I'm new to the GRUB scene since erm, 4 hours ago :) Thanks, Adrian -- Adrian Chadd | "I used to be thin, handsome and smart. | Then I discovered UNIX." | From daemon Tue Dec 30 06:11:50 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id GAA17151 for vsta-xplod; Tue, 30 Dec 1997 06:11:50 GMT Received: from Tandem.com (suntan.tandem.com [192.216.221.8]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id GAA17145 for ; Tue, 30 Dec 1997 06:11:32 GMT Received: from tenchi (tenchi.loc251.tandem.com [155.186.86.137]) by Tandem.com (8.8.8/2.0.1) with SMTP id IAA13253 for ; Tue, 30 Dec 1997 08:14:51 -0800 (PST) Message-Id: <3.0.3.32.19971230081354.009d95f0@mail.gerf.org> Organization: Gerf.org X-Sender: docwhat@mail.gerf.org X-Mailer: QUALCOMM Windows Eudora Pro Version 3.0.3 (32) Date: Tue, 30 Dec 1997 08:13:54 -0800 To: vsta@zendo.com From: "docwhat@gerf.org ;The Doctor What" Subject: Re: Mouse Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Repeating what Andy said: ->One possiblity is that your mouse has "jitter" and generates an endless ->stream of single-bit delta changes. Especially if the UART isn't buffered, ->you can then have port overruns, and you start reading garbage from the ->port. You can "stat //tty/tty01" (or tty02 for the 2nd COM port) and check ->the error counters on the device. A couple overruns are not actually all ->that big a deal; the serial protocol is pretty forgiving. Also, from my own experience: It's possible that the "jitter" or other mouse problems can be caused by a mouse that's improperly shielded against light. Try putting the mouse under a box or a towel and see if the "jitter" garbage stops. Example 1: Generic Brand ($18) Circa 1994 I had a mouse that failed to work vertically (i.e. it would only move horizontally) in the afternoons. Cause: Sunlight coming in at just the right angle to hit the optical sensor inside the mouse. Cure: Moved monitor in front of mouse. ;) Example 2: Generic Brand ($10) Circa 1991 or 1992 I had a mouse that would jitter at night. Sometimes vertically, sometimes horizontally, rarely both. Cause: Florescent lighting and the angle of the mouse to this lighting. Cure: Buy a new mouse. (I thought of putting a box which I'd put my hand in, but it wasn't worth the greif.) :( Just thought I'd put my two cents in. Ciao! -=Doc -- "If only you'd listened to me, I could have saved you from all that yukkiness." --Kryten The Doctor What: Un-Humble http://www.gerf.org/~docwhat/ docwhat@gerf.org (finger docwhat@gerf.org for PGP key) PGPkey Fingerprint: EA 4C 8C FC 5C F0 14 78 9C 02 B9 A1 83 54 7C 8D From daemon Tue Dec 30 07:04:44 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id HAA17241 for vsta-xplod; Tue, 30 Dec 1997 07:04:44 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id HAA17235 for ; Tue, 30 Dec 1997 07:04:41 GMT Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id JAA16741; Tue, 30 Dec 1997 09:05:22 -0800 (PST) Message-Id: <199712301705.JAA16741@vandys-pc.cisco.com> To: Adrian Chadd cc: vsta@zendo.com Subject: Re: 'dos' process dying while booting vsta 1.6.1 ? In-reply-to: Your message of "Tue, 30 Dec 1997 17:39:33 +0800." <199712300939.RAA25749@mail.creative.net.au> Date: Tue, 30 Dec 1997 09:05:21 -0800 From: Andy Valencia [Adrian Chadd writes:] >I'm installing vsta 1.6.1 on a laptop I just bought. >Trouble is, whenever I boot it , the thing starts booting, and then spits: >Boot process 7 dies >[Kernel debugger @ spl 0] Do a "q" and see what the syslog output says (because of queueing it may show up on the screen after the actual process has exited). Typical reasons are that you're running a compressed drive, VSTa's gotten confused on your disk geometry, or it's FAT-32 (though in this case you're using FAT-16), or it's partitioned in some non-obvious way. If you want to get a lot more detail, see the "Problem with booting VSTa" and "Booting into testsh does work" message threads under: http://www.vsta.org/vsta/mail/1 (BTW, yes, we now have our own vsta.org domain. The real workhorse machine will continue to be bodhi.zendo.com for the time being, but will probably cut to a VSTa server some time this year.) Andy From daemon Tue Dec 30 15:58:06 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id PAA18091 for vsta-xplod; Tue, 30 Dec 1997 15:58:06 GMT Received: from melia.qut.edu.au (melia.qut.edu.au [131.181.127.2]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id PAA18085 for ; Tue, 30 Dec 1997 15:57:51 GMT Received: from pigeon.qut.edu.au (meissner@pigeon.qut.edu.au) by melia.qut.edu.au (PMDF V5.1-7 #20429) with ESMTP id <01IRTN4N57NK00VRZN@melia.qut.edu.au> for vsta@zendo.com; Wed, 31 Dec 1997 12:03:37 +1000 Received: from localhost (meissner@localhost) by pigeon.qut.edu.au (8.8.5/8.8.5) with SMTP id MAA18867; Wed, 31 Dec 1997 12:00:53 +1000 (EST) Date: Wed, 31 Dec 1997 12:00:52 +1000 (EST) From: Wayne Meissner Subject: Re: 'dos' process dying while booting vsta 1.6.1 ? In-reply-to: <199712300939.RAA25749@mail.creative.net.au> To: Adrian Chadd Cc: vsta@zendo.com Message-id: MIME-version: 1.0 Content-type: TEXT/PLAIN; charset=US-ASCII On Tue, 30 Dec 1997, Adrian Chadd wrote: > > Hi. > > I'm installing vsta 1.6.1 on a laptop I just bought. > > Trouble is, whenever I boot it , the thing starts booting, and then spits: > Boot process 7 dies > [Kernel debugger @ spl 0] I get exactly the same result. If I hit 'q' to leave the debugger, I then see the WD server fire up and display the HD info(which it gets correct), and init complaining that it can't find the root filesystem (gee, I wonder why .... :-). Mine is on a VFAT (FAT16) partition. I've been meaning to try it on a raw FAT partition, on one of my spare drives, or one of my machines at work, but have been too busy. > So I do a proc, and find: > .. > 7 dos > 8 ONPROC kstack c0e000 uregs c0efc0 wchan 0 / > > Its juyst a DX4-100 laptop with 8mb RAM and a 500ish mbyte HDD. Its running > FAT16. Does anyone have any ideas why its barfing? (I'm new to the GRUB > scene since erm, 4 hours ago :) If you get that far, then its pretty likely that grub is working, so that can be discounted. --wayne From daemon Tue Dec 30 16:10:14 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id QAA18131 for vsta-xplod; Tue, 30 Dec 1997 16:10:14 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id QAA18125 for ; Tue, 30 Dec 1997 16:10:12 GMT Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id SAA17859; Tue, 30 Dec 1997 18:12:22 -0800 (PST) Message-Id: <199712310212.SAA17859@vandys-pc.cisco.com> To: Wayne Meissner cc: Adrian Chadd , vsta@zendo.com Subject: Re: 'dos' process dying while booting vsta 1.6.1 ? In-reply-to: Your message of "Wed, 31 Dec 1997 12:00:52 +1000." Date: Tue, 30 Dec 1997 18:12:22 -0800 From: Andy Valencia [Wayne Meissner writes:] >I get exactly the same result. If I hit 'q' to leave the debugger, I then >see the WD server fire up and display the HD info(which it gets >correct), and init complaining that it can't find the root filesystem >(gee, I wonder why .... :-). (Light goes on over my head.) Oops. The arguments to the DOS server changed, and although I fixed my menu.lst, I managed to not copy this back into the boot/menu.lst sample one. You should have an entry like: /vsta/boot/dos -n fs/root -d //disk/wd:wd0_dos0 -B 512 instead of the old one which is in there. I'm going to go copy the sample entry into the distribution, and also convert the usage() message to use syslog. I bet you'll find things work a bit better with the DOS server not picking up its toys and going home because it doesn't like its command line! Note that the vstafs server was also changed to use this format. Andy From daemon Tue Dec 30 19:01:45 1997 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id TAA18335 for vsta-xplod; Tue, 30 Dec 1997 19:01:45 GMT Received: from mail.creative.net.au (mail.creative.net.au [203.56.168.4]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id TAA18329 for ; Tue, 30 Dec 1997 19:01:41 GMT Received: from mail.creative.net.au (localhost.creative.net.au [127.0.0.1]) by mail.creative.net.au (8.8.5/8.7) with ESMTP id NAA28063 for ; Wed, 31 Dec 1997 13:04:58 +0800 (WST) Message-Id: <199712310504.NAA28063@mail.creative.net.au> To: vsta@zendo.com Subject: Another roadmap.txt change. Date: Wed, 31 Dec 1997 13:04:57 +0800 From: Adrian Chadd While you're at it :) roadmap.txt references some file to rawrite/dd to the floppy, making it boot GRUB. But the file isn't there, instead, there is grub.raw. (which on a hunch I rawrite'd, and it worked :) Thought you might want to know. Adrian -- Adrian Chadd | "I used to be thin, handsome and smart. | Then I discovered UNIX." | From daemon Sun Jan 4 12:34:27 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id MAA29743 for vsta-xplod; Sun, 4 Jan 1998 12:34:27 GMT Received: from baruk2.zendo.com (baruk2.zendo.com [205.187.71.65]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id MAA29737 for ; Sun, 4 Jan 1998 12:34:23 GMT Received: (from vandys@localhost) by baruk2.zendo.com (8.8.5/8.6.12) id OAA14868; Sun, 4 Jan 1998 14:46:40 -0800 (PST) X-Authentication-Warning: baruk2.zendo.com: vandys set sender to pavel@elf.ucw.cz using -f Received: from bodhi.zendo.com (bodhi.zendo.com [205.187.71.2]) by baruk2.zendo.com (8.8.5/8.6.12) with ESMTP id EAA14045 for ; Sun, 4 Jan 1998 04:56:34 -0800 (PST) Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id CAA28955 for owner-vsta; Sun, 4 Jan 1998 02:44:24 GMT Received: from atrey.karlin.mff.cuni.cz (root@atrey.karlin.mff.cuni.cz [195.113.31.123]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id CAA28948 for ; Sun, 4 Jan 1998 02:44:16 GMT Received: from Elf.mj.gts.cz (root@slip7.ms.mff.cuni.cz [195.113.20.207]) by atrey.karlin.mff.cuni.cz (8.8.5/8.8.5) with ESMTP id NAA24836 for ; Sun, 4 Jan 1998 13:47:50 +0100 Received: (from pavel@localhost) by Elf.mj.gts.cz (8.8.5/8.8.5) id NAA00094; Sun, 4 Jan 1998 13:46:40 +0100 Message-ID: <19980104134639.26787@Elf.mj.gts.cz> Date: Sun, 4 Jan 1998 13:46:39 +0100 From: Pavel Machek To: vsta@zendo.com Subject: 1.6.1 impressions Mime-Version: 1.0 XXContent-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.81 X-Warning: Not only using Windows can be dangerous to your mental health. Hi! First: Thanx for 1.6.1. It's really nice. I even tried TCP/IP package, and it quite works to my big surprise. And now, usual set of bugreports, questions, etc: * Seems like process #1 (usually cons) disappears from ps listing after some time of heavy activity. (Try rebuilding whole sources.) It is also no longer in /proc filesystem. * telnetd does not work too well for me. 'plain' version justed printed login banner and existed immediately, with comment 'IO server: all clients done' in syslog. After I patched (nrefs<1) to (nrefs<0), I can log in and work, but it will not disconnect at the end. * I have made 'blinkedlights' extension to uKernel: it now displays current PID of process and CPU busy state on debugging LEDs attached to printer port. If anyone has similar hardware, PLEASE let me know * Question: What are reasons behind having processes organized in tree? I have thought about putting part of scheduler outside of microkernel (making uKernel capable only of strictly-prioritive scheduler, and external daemon, which would play with process's priorities based on how much they run). Is there way to see tree structure of processes in ps? * ps tries to show some kind of times with each process. This times look like min:sec, but they are not. They are running much faster. What is that? * here is patch for srv/mach/cons2/isr.c - it makes 'Meta' key work, so Alt+letters have good meanings. (Console is unusable for me without this). It also simplifies handling of F1, F2, ... keys. * printink 'feel free to hit reset button' on attempt to reboot is bad think (tm), because user has to go and push reset button himself. On the end, I append simple program to reset machine. I think that this should be put in place of reboot script Keep hacking, Pavel --- isrpmk.c Sun Jan 4 00:24:34 1998 GMT +++ isr.c Sun Jan 4 13:26:54 1998 GMT @@ -13,6 +13,9 @@ capstoggle = 0, /* For toggling effect of CAPS */ numtoggle = 0, /* ...NUM lock */ isE0 = 0; /* Prefix for extended keys (FN1, etc.) */ + +#define C( x ) x & 0x1f /* Turn key into ctrl-one */ +#define M( x ) x | 0x80 /* Turn key into meta key, do not use for 0 */ /* Map scan codes to ASCII, one table for normal, one for shifted */ static char normal[] = { @@ -83,11 +86,19 @@ #endif /* + * Meta keys are badly needed - if Alt is pressed, + * it sends ESC before actual character. + */ + + if (alt || (ch&0x80)) + kbd_enqueue(s, 033); + + /* * Hand off straight data now. The keyboard always enters * data for the virtual screen currently being displayed * on the hardware screen. */ - kbd_enqueue(s, ch); + kbd_enqueue(s, ch&0x7f); } /* @@ -171,48 +185,15 @@ static int function_key(struct screen *s, uchar c) { - char *p; + char p[4]; - switch (c) { - case 59: /* F1 */ - p = "\033OP"; - break; - case 60: /* F2 */ - p = "\033OQ"; - break; - case 61: /* F3 */ - p = "\033OR"; - break; - case 62: /* F4 */ - p = "\033OS"; - break; - case 63: /* F5 */ - p = "\033OT"; - break; - case 64: /* F6 */ - p = "\033OU"; - break; - case 65: /* F7 */ - p = "\033OV"; - break; - case 66: /* F8 */ - p = "\033OW"; - break; - case 67: /* F9 */ - p = "\033OX"; - break; - case 68: /* F10 */ - case 87: /* F11 */ - case 88: /* F12 */ - p = 0; - break; - default: - return 0; - } - if (p) { - enqueue_string(s, p); - } - return(1); + if ((c<59) || (c>68)) return 0; + p[0] = '\033'; + p[1] = 'Q'; + p[2] = 'P'+c-59; + p[3] = 0; + enqueue_string(s, p); + return 1; } /* Reboot program: /* * reboot.c * Reboot your PC * * Copyright (C) 1998 by Pavel Machek, GPL */ #include #include #include #include #include int main( void ) { printf( "Reseting" ); fflush( stdout ); if (enable_io( 0x64, 0x64 )) { printf( "Can not enable io.\n" ); return 1; } printf( "." ); fflush( stdout ); while( inportb( 0x64 ) & 0x02 ); printf( "." ); fflush( stdout ); outportb( 0x64, 0xfe ); } -- I'm really pavel@atrey.karlin.mff.cuni.cz. Pavel Look at http://atrey.karlin.mff.cuni.cz/~pavel/ ;-). From daemon Tue Jan 6 11:35:24 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id LAA04291 for vsta-xplod; Tue, 6 Jan 1998 11:35:24 GMT Received: from atrey.karlin.mff.cuni.cz (pavel@atrey.karlin.mff.cuni.cz [195.113.31.123]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id LAA04285 for ; Tue, 6 Jan 1998 11:35:14 GMT Received: (from pavel@localhost) by atrey.karlin.mff.cuni.cz (8.8.5/8.8.5) id WAA05660; Tue, 6 Jan 1998 22:38:56 +0100 Message-ID: <19980106223856.52174@atrey.karlin.mff.cuni.cz> Date: Tue, 6 Jan 1998 22:38:56 +0100 From: Pavel Machek To: vsta@zendo.com Subject: GNU ports - where Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.84 X-Warning: Not only using Windows can be dangerous to your mental health. Hi! There used to be GNU ports at ftp://ftp.crystald.com/pub/vsta/ports. Where are they now? Pavel -- Do *NOT* buy software, GNU software is better and free! Pavel GCM d? s-: !g p?:+ au- a--@ w+ v- C++@ UL+++ L++ N++ E++ W--- M- Y- R+ From daemon Tue Jan 6 16:39:48 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id QAA04716 for vsta-xplod; Tue, 6 Jan 1998 16:39:48 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id QAA04710 for ; Tue, 6 Jan 1998 16:39:45 GMT Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id SAA03420; Tue, 6 Jan 1998 18:42:58 -0800 (PST) Message-Id: <199801070242.SAA03420@vandys-pc.cisco.com> To: Pavel Machek cc: vsta@zendo.com Subject: Re: GNU ports - where In-reply-to: Your message of "Tue, 06 Jan 1998 22:38:56 +0100." <19980106223856.52174@atrey.karlin.mff.cuni.cz> Date: Tue, 06 Jan 1998 18:42:58 -0800 From: Andy Valencia [Pavel Machek writes:] >There used to be GNU ports at >ftp://ftp.crystald.com/pub/vsta/ports. Where are they now? They're folded into the main distribution. You'll find all of them (unless I goofed) in the 1.6 (and later) distribution filesets. Andy From daemon Sat Jan 10 10:52:10 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id KAA13440 for vsta-xplod; Sat, 10 Jan 1998 10:52:10 GMT Received: from atrey.karlin.mff.cuni.cz (root@atrey.karlin.mff.cuni.cz [195.113.31.123]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id KAA13434 for ; Sat, 10 Jan 1998 10:52:06 GMT Received: from Elf.mj.gts.cz (root@slip8.ms.mff.cuni.cz [195.113.20.208]) by atrey.karlin.mff.cuni.cz (8.8.5/8.8.5) with ESMTP id VAA30513 for ; Sat, 10 Jan 1998 21:55:52 +0100 Received: (from pavel@localhost) by Elf.mj.gts.cz (8.8.5/8.8.5) id UAA00307; Sat, 10 Jan 1998 20:14:03 +0100 Message-ID: <19980110201359.55562@Elf.mj.gts.cz> Date: Sat, 10 Jan 1998 20:13:59 +0100 From: Pavel Machek To: vsta@zendo.com Subject: UMC U5S CPU (486SX clone) Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.81 X-Warning: Not only using Windows can be dangerous to your mental health. Hi! I just replaced my amd486 to UMC U5S cpu, and VSTA -say- not works really well. When program exits, it in sometimes (70% on first program to exit, much less than, really strange) raises following 'assertion failed': ASSERT_DEBUG(a->a_flags & ATL_CACHE, "pset_free: non-cache ref"); The cpu is rather strange (althrough it seems to work well with linux). I would like to ask where is TLB invalidation being done? I did not found that place. Pavel One more question: This CPU does not have FPU. Is it problem? I know there's no emulation, but is missing FPU likely causing assertion failed when logging in? (This is linux /proc/cpuinfo:) processor : 0 cpu family : 4 model : U5S vendor_id : UMC UMC UMC stepping : 3 fdiv_bug : no hlt_bug : no sep_bug : no f00f_bug : no fpu : no fpu_exception : no cpuid level : 1 wp : yes flags : bogomips : 26.42 -- I'm really pavel@atrey.karlin.mff.cuni.cz. Pavel Look at http://atrey.karlin.mff.cuni.cz/~pavel/ ;-). From daemon Mon Jan 12 05:57:02 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id FAA16528 for vsta-xplod; Mon, 12 Jan 1998 05:57:02 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id FAA16522 for ; Mon, 12 Jan 1998 05:56:59 GMT Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id IAA17477; Mon, 12 Jan 1998 08:00:24 -0800 (PST) Message-Id: <199801121600.IAA17477@vandys-pc.cisco.com> To: Pavel Machek cc: vsta@zendo.com Subject: Re: UMC U5S CPU (486SX clone) In-reply-to: Your message of "Sat, 10 Jan 1998 20:13:59 +0100." <19980110201359.55562@Elf.mj.gts.cz> Date: Mon, 12 Jan 1998 08:00:23 -0800 From: Andy Valencia [Pavel Machek writes:] >I just replaced my amd486 to UMC U5S cpu, and VSTA -say- not works >really well. When program exits, it in sometimes (70% on first program >to exit, much less than, really strange) raises following 'assertion >failed': > ASSERT_DEBUG(a->a_flags & ATL_CACHE, > "pset_free: non-cache ref"); Interesting. It would appear that you have an extra translation which could not be found by walking the active list of translations. I recollect that the most common way to cause this is if your chip ever causes page faults even though the PTE is present and valid--because then VSTa (which trusts the hardware) will insert the translation again, and this adds another ATL entry. On cleanup the pview/pset slot being marked active will cause VSTa to clean up the underlying structures--but only once. Then, during this final walk, it is noticed that we managed to miss a translation. If you're up for hacking, you can either try and detect this with some extra logic in vas_fault(). You might also want to look at this extra ATL and make sure it looks at least basically sane--matches the pview being deleted, has a reasonable linked list structure, and so forth. >The cpu is rather strange (althrough it seems to work well with >linux). I would like to ask where is TLB invalidation being done? I >did not found that place. See hat_deletetrans(). >One more question: This CPU does not have FPU. Is it problem? I know >there's no emulation, but is missing FPU likely causing assertion >failed when logging in? Nope, to the best of my knowledge, it'll only bite you with gcc. Andy From daemon Wed Jan 14 06:52:45 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id GAA20850 for vsta-xplod; Wed, 14 Jan 1998 06:52:45 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id GAA20844 for ; Wed, 14 Jan 1998 06:52:42 GMT Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id IAA22030; Wed, 14 Jan 1998 08:56:12 -0800 (PST) Message-Id: <199801141656.IAA22030@vandys-pc.cisco.com> To: docwhat@gerf.org cc: vsta@zendo.com Subject: Re: VSTa slow... In-reply-to: Your message of "Wed, 14 Jan 1998 08:44:57 PST." <3.0.3.32.19980114084457.009d8c20@mail.gerf.org> Date: Wed, 14 Jan 1998 08:56:12 -0800 From: Andy Valencia ["docwhat@gerf.org ;The Doctor What" writes:] > Just got GRUB and co. working on my new system (well, it's a >year old, but I've never had VSTa running on it). > It's a 64mb PentiumMMX 200Mhz. and VSTa is extremely slow to >respond. It seems to me that it is one of two things that make it slow: > 1) Disk access is very slow > 2) Process creation is very slow > I suppose I could just hit ^Z and look at traces to see what's >up, but that seems so...so...inelegant! > What would I have to do to figure out what's wrong, since I >wanted to start looking at the security model in VSTa again. Well, if it really seems like something's wrong, you should at least look at ps and see if something seems to be running like crazy. Generally, either you'll find it there, or else you've tickled a bug in your chip's power management. VSTa flies on my 200 Mhz Pentium Pro, for instance. If you can't see the CPU in a "ps", then ^Z probably won't help much, since all ^Z does is drop into the kernel debugger from the console driver (where the ^Z is detected). You'd have to hack some sort of clock tick driven profiling to find it otherwise. >Did we >have a paper/description describing how we wanted security to be in >VSTa? I mean besides the file in the distribution. > I'll have to go get a copy of the POSIX security commands and >try lining things up in a "coherent" set of English rules. Off to the >library I'll go. ;) What I have, you have. :-) "Security" is a big area, it'd probably be better to figure out what problem you're trying to solve, and go from there. You can hash it out offline with me in private E-mail for the initial cycles. Regards, Andy From daemon Wed Jan 14 06:42:59 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id GAA20827 for vsta-xplod; Wed, 14 Jan 1998 06:42:59 GMT Received: from Tandem.com (suntan.tandem.com [192.216.221.8]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id GAA20821 for ; Wed, 14 Jan 1998 06:42:54 GMT Received: from tenchi (tenchi.loc251.tandem.com [155.186.86.137]) by Tandem.com (8.8.8/2.0.1) with SMTP id IAA01052; Wed, 14 Jan 1998 08:46:58 -0800 (PST) Message-Id: <3.0.3.32.19980114084457.009d8c20@mail.gerf.org> Organization: Gerf.org X-Sender: docwhat@mail.gerf.org X-Mailer: QUALCOMM Windows Eudora Pro Version 3.0.3 (32) Date: Wed, 14 Jan 1998 08:44:57 -0800 To: Andy Valencia From: "docwhat@gerf.org ;The Doctor What" Subject: VSTa slow... Cc: vsta@zendo.com In-Reply-To: <199801121600.IAA17477@vandys-pc.cisco.com> References: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Howdy all.... Just got GRUB and co. working on my new system (well, it's a year old, but I've never had VSTa running on it). It's a 64mb PentiumMMX 200Mhz. and VSTa is extremely slow to respond. It seems to me that it is one of two things that make it slow: 1) Disk access is very slow 2) Process creation is very slow I suppose I could just hit ^Z and look at traces to see what's up, but that seems so...so...inelegant! What would I have to do to figure out what's wrong, since I wanted to start looking at the security model in VSTa again. Did we have a paper/description describing how we wanted security to be in VSTa? I mean besides the file in the distribution. I'll have to go get a copy of the POSIX security commands and try lining things up in a "coherent" set of English rules. Off to the library I'll go. ;) -=Doc -- "When you have eliminated the impossible, whatever remains, however improbable, must be the truth." -- Sherlock Holmes, "The Sign of Four" The Doctor What: A Holtje Production http://www.gerf.org/~docwhat/ docwhat@gerf.org (finger docwhat@gerf.org for PGP key) PGPkey Fingerprint: EA 4C 8C FC 5C F0 14 78 9C 02 B9 A1 83 54 7C 8D From daemon Thu Jan 15 08:40:50 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id IAA23136 for vsta-xplod; Thu, 15 Jan 1998 08:40:50 GMT Received: from malasada.lava.net (root@malasada.lava.net [199.222.42.2]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id IAA23130 for ; Thu, 15 Jan 1998 08:40:35 GMT Received: from localhost (1282 bytes) by malasada.lava.net via sendmail with P:stdio/R:inet_hosts/T:smtp (sender: ) (ident using unix) id for ; Thu, 15 Jan 1998 08:44:26 -1000 (HST) (Smail-3.2.0.96 1997-Jun-2 #3 built 1997-Aug-20) Message-Id: From: newsham@lava.net (Tim Newsham) Subject: Re: VSTa slow... To: vandys@cisco.com (Andy Valencia) Date: Thu, 15 Jan 1998 08:44:26 -1000 (HST) Cc: docwhat@gerf.org, vsta@zendo.com In-Reply-To: <199801141656.IAA22030@vandys-pc.cisco.com> from "Andy Valencia" at Jan 14, 98 08:56:12 am X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit > >Did we > >have a paper/description describing how we wanted security to be in > >VSTa? I mean besides the file in the distribution. > > I'll have to go get a copy of the POSIX security commands and > >try lining things up in a "coherent" set of English rules. Off to the > >library I'll go. ;) > > What I have, you have. :-) "Security" is a big area, it'd probably be > better to figure out what problem you're trying to solve, and go from there. > You can hash it out offline with me in private E-mail for the initial > cycles. I would be very interested in seeing any security discussions, either by being CC'ed, having a seperate mailing list, or including the discussions on the main list. > Andy Tim N. From daemon Mon Feb 9 06:01:37 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id GAA22383 for vsta-xplod; Mon, 9 Feb 1998 06:01:37 GMT Received: from spock.lpr.e-technik.tu-muenchen.de (root@spock.lpr.e-technik.tu-muenchen.de [129.187.151.1]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id GAA22377 for ; Mon, 9 Feb 1998 06:01:33 GMT Received: from nelion.lpr.e-technik.tu-muenchen.de (root@nelion.lpr.e-technik.tu-muenchen.de [129.187.151.44]) by spock.lpr.e-technik.tu-muenchen.de (8.8.7/8.8.7) with ESMTP id RAA21211 for ; Mon, 9 Feb 1998 17:06:50 +0100 (MET) Received: from nelion (erdost@localhost [127.0.0.1]) by nelion.lpr.e-technik.tu-muenchen.de (8.6.12/8.6.6) with SMTP id RAA18877 for ; Mon, 9 Feb 1998 17:06:49 +0100 Sender: erdost@lpr.e-technik.tu-muenchen.de Message-ID: <34DF2998.446B@lpr.e-technik.tu-muenchen.de> Date: Mon, 09 Feb 1998 17:06:48 +0100 From: Thomas Erdoes X-Mailer: Mozilla 3.0 (X11; I; OSF1 V4.0 alpha) MIME-Version: 1.0 To: vsta@zendo.com Subject: cons2 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hello, does someone work with non-US keybord under VSTa? I combined cons2 with cons3 and the result was that now I am able to use the AltGr-Key combination. If someone is interested, I could post the diffs here. I have a question, too. I tried to compile gcc under VSTa but right at the begin I got some errors. I don't know right now what errors occured because I tried to compile the compiler a few weeks ago and I didn't record the errors. The question is: Need I start the configure script or is gcc already configured? Thomas From daemon Mon Feb 9 06:22:05 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id GAA22414 for vsta-xplod; Mon, 9 Feb 1998 06:22:05 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id GAA22408 for ; Mon, 9 Feb 1998 06:22:03 GMT Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id IAA08695; Mon, 9 Feb 1998 08:25:06 -0800 (PST) Message-Id: <199802091625.IAA08695@vandys-pc.cisco.com> To: Thomas Erdoes cc: vsta@zendo.com Subject: Re: cons2 In-reply-to: Your message of "Mon, 09 Feb 1998 17:06:48 +0100." <34DF2998.446B@lpr.e-technik.tu-muenchen.de> Date: Mon, 09 Feb 1998 08:25:06 -0800 From: Andy Valencia [Thomas Erdoes writes:] >I have a question, too. I tried to compile gcc under VSTa but right at >the begin I got some errors. I don't know right now what errors occured >because I tried to compile the compiler a few weeks ago and I didn't >record the errors. The question is: Need I start the configure >script or is gcc already configured? gcc should already be configured, and was and is built under VSTa. It should continue to build, delta any accidental damage I've done to it. Andy From daemon Mon Feb 9 13:38:20 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id NAA23197 for vsta-xplod; Mon, 9 Feb 1998 13:38:20 GMT Received: from valhalla.cs.wright.edu (valhalla.cs.wright.edu [130.108.2.10]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id NAA23191 for ; Mon, 9 Feb 1998 13:38:17 GMT Received: from gamma.cs.wright.edu (gamma.cs.wright.edu [130.108.2.22]) by valhalla.cs.wright.edu (8.8.7/8.8.5) with ESMTP id SAA20451 for ; Mon, 9 Feb 1998 18:43:39 -0500 (EST) Received: from localhost (mjack@localhost) by gamma.cs.wright.edu (8.8.7/8.8.6) with SMTP id SAA01191 for ; Mon, 9 Feb 1998 18:43:39 -0500 (EST) X-Authentication-Warning: gamma.cs.wright.edu: mjack owned process doing -bs Date: Mon, 9 Feb 1998 18:43:39 -0500 (EST) From: Marcel Jackowski To: vsta@zendo.com Subject: fd server Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII I've been trying to mount a dos volume from a floppy disk but so far I have not been able to make it work. I am wondering if I am doing something silly here: I do 'fd &' then the device server loads and it correctly prints the floppy drive information. Then I try to make 'dos' talk to the device by issuing a 'dos -n fs/fd -d //disk/fd:fd0_1440 &' The drive led turns on, but 'dos' never replies with a 'file system established' message. It seems that something is wrong with the fd server. Another interesting thing is that if you issue a 'cat //disk/fd:fd0_1440' nothing is printed. I wonder if other people have successfully made it. Marcel mjack@cs.wright.edu From daemon Fri Feb 13 06:36:16 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id GAA03189 for vsta-xplod; Fri, 13 Feb 1998 06:36:16 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id GAA03183 for ; Fri, 13 Feb 1998 06:35:58 GMT Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id IAA18866; Fri, 13 Feb 1998 08:40:07 -0800 (PST) Message-Id: <199802131640.IAA18866@vandys-pc.cisco.com> To: Pavel Machek cc: vsta@zendo.com Subject: Re: 1.6.1 impressions In-reply-to: Your message of "Sun, 04 Jan 1998 13:46:39 +0100." <19980104134639.26787@Elf.mj.gts.cz> Date: Fri, 13 Feb 1998 08:40:07 -0800 From: Andy Valencia [Pavel Machek writes:] >First: Thanx for 1.6.1. It's really nice. I even tried TCP/IP package, and >it quite works to my big surprise. (grin) >* Seems like process #1 (usually cons) disappears from ps listing after some >time of heavy activity. (Try rebuilding whole sources.) It is also no longer i >n >/proc filesystem. I've tried this again and again, even under MGR, and don't see it happen. You'll need to pick apart your proc chain from the kernel debugger, or give me more clues as to what causes it. >* telnetd does not work too well for me. 'plain' version justed printed login >banner and existed immediately, with comment 'IO server: all clients done' in >syslog. After I patched (nrefs<1) to (nrefs<0), I can log in and work, but it >will not disconnect at the end. I can't reproduce this, either. For me telnet works fine for multiple logins, and "tcp stat" shows correct timeout and cleanup. Also, no funny reference counts under "vsta stat". I remade from scratch to make sure I ws using current code. >* Question: What are reasons behind having processes organized in tree? I have >thought about putting part of scheduler outside of microkernel (making uKernel >capable only of strictly-prioritive scheduler, and external daemon, which >would play with process's priorities based on how much they run). Is there way >to see tree structure of processes in ps? The tree gives you a "fair share" distribution of CPU. It was an idea I wanted to code up, so I put it in VSTa way back when I was first writing it. I exercised the tree distribution a bit early on (with direct hacking to make the trees I wanted) and it worked OK. But I never fleshed out a regular syscall API to control/monitor it. BTW, the "cheated" queue came as a result of the latency a fair share allocation imposes on interactive processes. Basically, trees and fair share are good when you're dealing with CPU-heavy processes, but if a process takes only small slices at a time (typing to a shell or editor, for instance) then it suffers unduly if it always has to go back into the tree. >* ps tries to show some kind of times with each process. This times look like >min:sec, but they are not. They are running much faster. What is that? Looks like at a minimum a units problem in pulling the values from pstat. I played with this a little bit, and still am not satisfied I understand everything that's going wrong here. >* here is patch for srv/mach/cons2/isr.c - it makes 'Meta' key work, so >Alt+letters have good meanings. (Console is unusable for me without this). >It also simplifies handling of F1, F2, ... keys. Ok, I think I'll fold this one in. >* printink 'feel free to hit reset button' on attempt to reboot is >bad think (tm), because user has to go and push reset button >himself. On the end, I append simple program to reset machine. I think >that this should be put in place of reboot script We probably also want a call into the kernel to do the current shutdown trick. I've heard of PC's where the keyboard wouldn't work but CR3/shutdown would, and also vice versa. If we do both, we're probably pretty safe. I'll play with this, too, unless you want to. >Keep hacking, Always. Andy From daemon Fri Feb 13 06:35:56 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id GAA03182 for vsta-xplod; Fri, 13 Feb 1998 06:35:56 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id GAA03176 for ; Fri, 13 Feb 1998 06:35:44 GMT Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id IAA18874; Fri, 13 Feb 1998 08:40:10 -0800 (PST) Message-Id: <199802131640.IAA18874@vandys-pc.cisco.com> To: Marcel Jackowski cc: vsta@zendo.com Subject: Re: fd server In-reply-to: Your message of "Mon, 09 Feb 1998 18:43:39 EST." Date: Fri, 13 Feb 1998 08:40:10 -0800 From: Andy Valencia [Marcel Jackowski writes:] >I've been trying to mount a dos volume from a floppy disk but so far >I have not been able to make it work. I am wondering if I am doing >something silly here: Nope, it was me being silly. As of 1.6, M_TIME is no longer a special case, and thus can not be sent by non-kernel code. Unfortunately, I missed the FD server. Please pick up the patch at: ftp://ftp.vsta.org/vsta/pickup/pat_fd.1 and tell me if it works for you. BTW, this was a 5-minute debug job with gdb. I just set a breakpoint at the ISR message handler, and saw that it never got entered following floppy spinup. So I set a breakpoint in the time handling, and saw that my time message was sent (with error) but not received. Then I checked os/kern/msg.c and saw the cap on m_type field values, and realized that M_TIME was now being disabled. So I renumbered it, and things started to work. Sorry for breaking it, though, and sorry for the delay in getting around to hooking up a floppy drive and managing to try it locally. Regards, Andy Valencia From daemon Wed Feb 25 00:26:46 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id AAA04532 for vsta-xplod; Wed, 25 Feb 1998 00:26:46 GMT Received: from gateway.s3.com (gateway.s3.com [198.211.58.4]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id AAA04526 for ; Wed, 25 Feb 1998 00:26:44 GMT Received: from chat.net (genesis.jeske.meer.net [204.94.139.85]) by gateway.s3.com (8.6.12/8.6.12) with ESMTP id CAA21767 for ; Wed, 25 Feb 1998 02:07:28 -0800 Received: by chat.net (SMI-8.6/SMI-SVR4) id CAA29099; Wed, 25 Feb 1998 02:34:49 -0800 Message-ID: <19980225023449.22612@home.chat.net> Date: Wed, 25 Feb 1998 02:34:49 -0800 From: David Jeske To: *** VSTa *** Subject: VSTa bootable floppy troubles... Mail-Followup-To: *** VSTa *** Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.90.2 Hey all, I dropped 1.6.1 on my machine this evening, and proceeded to make a bootable VSTa floppy. By this I mean, I formatted a floppy as DOS FAT, copied what I could surmise were the required files onto it, installed grub to load the stage 2 loader off the floppy, and let 'er rip. Before doing any of this I: 1) installed the fd_pat.1 patch to the fd server to get it to work 2) changed the floppy to build as a static server (ugh it's big now) by making the libraries "-lc_s -lsrv" instead of "-lusr -lc" and using "crt0srv.o" instead of "crt0.o". Now, the floppy server works when launched from VSTa, and it works when used as a boot server from the floppy. I tried to do a basic boot, and after GRUB loads all the servers, the screen clears (good sign) and I see the two messages from the fd server. I think I saw the filesystem established message also (more on this), the drive churns in nice patterns like it's running software, and then after a few seconds the system locks hard. No drop into the kernel debugger, just everything stops. So I changed to using testsh instead of init. I successfully mounted "fs/root" to "/" and perused around it, so the fd server and dos filesystem on the fd server at least _seem_ to be working somewhat. Any ideas about what might be locking it hard? I have all the standard /vsta/etc files on the floppy, and most of the important stuff from /vsta/bin (i.e. login, cp, echo, setime, runrc, ls, pwd, and a few more) I'm going to look at this again tommorow. If anyone wants to take a look, grab the image of the floppy from: http://www.chat.net/~jeske/VSTa/vstaflop.gz just ungzip it and then either use rawrite.exe or "cat" or "dd" to get it onto a floppy. -- David Jeske (N9LCA) + http://www.chat.net/~jeske/ + jeske@chat.net From daemon Wed Feb 25 04:16:34 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id EAA04909 for vsta-xplod; Wed, 25 Feb 1998 04:16:34 GMT Received: from spock.lpr.e-technik.tu-muenchen.de (root@spock.lpr.e-technik.tu-muenchen.de [129.187.151.1]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id EAA04903 for ; Wed, 25 Feb 1998 04:16:28 GMT Received: from nelion.lpr.e-technik.tu-muenchen.de (root@nelion.lpr.e-technik.tu-muenchen.de [129.187.151.44]) by spock.lpr.e-technik.tu-muenchen.de (8.8.7/8.8.7) with ESMTP id PAA12574 for ; Wed, 25 Feb 1998 15:21:01 +0100 (MET) Received: from nelion (erdost@localhost [127.0.0.1]) by nelion.lpr.e-technik.tu-muenchen.de (8.6.12/8.6.6) with SMTP id PAA01974 for ; Wed, 25 Feb 1998 15:21:00 +0100 Sender: erdost@lpr.e-technik.tu-muenchen.de Message-ID: <34F428CB.3F54@lpr.e-technik.tu-muenchen.de> Date: Wed, 25 Feb 1998 15:20:59 +0100 From: Thomas Erdoes X-Mailer: Mozilla 3.0 (X11; I; OSF1 V4.0 alpha) MIME-Version: 1.0 To: vsta@zendo.com Subject: Re:VSTa bootable floppy troubles... Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hello, did you get a message like: init: can't find root, sleeping ? Tom From daemon Wed Feb 25 09:20:23 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id JAA05513 for vsta-xplod; Wed, 25 Feb 1998 09:20:23 GMT Received: from gateway.s3.com (gateway.s3.com [198.211.58.4]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id JAA05507 for ; Wed, 25 Feb 1998 09:20:14 GMT Received: from chat.net (genesis.jeske.meer.net [204.94.139.85]) by gateway.s3.com (8.6.12/8.6.12) with ESMTP id LAA24101 for ; Wed, 25 Feb 1998 11:01:23 -0800 Received: by chat.net (SMI-8.6/SMI-SVR4) id LAA29611; Wed, 25 Feb 1998 11:28:48 -0800 Message-ID: <19980225112847.58799@home.chat.net> Date: Wed, 25 Feb 1998 11:28:47 -0800 From: David Jeske To: vsta@zendo.com Subject: Re: VSTa bootable floppy troubles... (fixed) Mail-Followup-To: vsta@zendo.com References: <34F428CB.3F54@lpr.e-technik.tu-muenchen.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.90.2 In-Reply-To: <34F428CB.3F54@lpr.e-technik.tu-muenchen.de>; from Thomas Erdoes on Wed, Feb 25, 1998 at 03:20:59PM +0100 On Wed, Feb 25, 1998 at 03:20:59PM +0100, Thomas Erdoes wrote: > did you get a message like: > > init: can't find root, sleeping ? > > Tom Nope, but I figured out why it didn't boot. I made the /vsta/lib directory but forgot to populate it with the two shared libs. So init was launching programs but those programs couldn't find ld.shl. I've fixed the image and it's up at the same location: http://www.chat.net/~jeske/VSTa/vstaflop.gz -- David Jeske (N9LCA) + http://www.chat.net/~jeske/ + jeske@chat.net From daemon Fri Feb 27 07:32:02 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id HAA01286 for vsta-xplod; Fri, 27 Feb 1998 07:32:02 GMT Received: from valhalla.cs.wright.edu (valhalla.cs.wright.edu [130.108.2.10]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id HAA01280 for ; Fri, 27 Feb 1998 07:31:57 GMT Received: from obr.cs.wright.edu.mayberry (obr.cs.wright.edu [130.108.16.44]) by valhalla.cs.wright.edu (8.8.7/8.8.5) with SMTP id MAA22468 for ; Fri, 27 Feb 1998 12:59:07 -0500 (EST) Received: from obr.cs.wright.edu by obr.cs.wright.edu.mayberry (SMI-8.6/SMI-SVR4) id MAA03537; Fri, 27 Feb 1998 12:59:09 -0500 Message-Id: <199802271759.MAA03537@obr.cs.wright.edu.mayberry> Date: Fri, 27 Feb 1998 12:59:09 -0500 (EST) From: Marcel Jackowski Reply-To: Marcel Jackowski Subject: can't kill fd server To: vsta@zendo.com MIME-Version: 1.0 Content-Type: TEXT/plain; charset=us-ascii Content-MD5: 6Kgn4OPpDoudQnZJ0dGvUA== X-Mailer: dtmail 1.2.0 CDE Version 1.2 SunOS 5.6 sun4u sparc Hi all, It seems like that the fd server once is running cannot be killed. It looks the problem is related somehow to this second thread which sends timer messages to the server, but I am still unable to nail it down. I have written a device server for an old Mitsumi 1x CDROM, which mimics most of the functionality of our fd server. As a result, it suffers from the same problem - the server works perfectly, but if I say "kill server-pid" from the prompt, it just hangs. Note that, a 'ps' will list 2 threads for the server. The first gets killed but never the last one. Any suggestions ? Marcel. mjack@cs.wright.edu From daemon Fri Feb 27 09:03:58 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id JAA01475 for vsta-xplod; Fri, 27 Feb 1998 09:03:58 GMT Received: from gateway.s3.com (gateway.s3.com [198.211.58.4]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id JAA01469 for ; Fri, 27 Feb 1998 09:03:56 GMT Received: from chat.net (genesis.jeske.meer.net [204.94.139.85]) by gateway.s3.com (8.6.12/8.6.12) with ESMTP id LAA13106 for ; Fri, 27 Feb 1998 11:06:01 -0800 Received: by chat.net (SMI-8.6/SMI-SVR4) id LAA04865; Fri, 27 Feb 1998 11:33:29 -0800 Message-ID: <19980227113329.35318@home.chat.net> Date: Fri, 27 Feb 1998 11:33:29 -0800 From: David Jeske To: vsta@zendo.com Subject: Re: can't kill fd server Mail-Followup-To: vsta@zendo.com References: <199802271759.MAA03537@obr.cs.wright.edu.mayberry> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.90.2 In-Reply-To: <199802271759.MAA03537@obr.cs.wright.edu.mayberry>; from Marcel Jackowski on Fri, Feb 27, 1998 at 12:59:09PM -0500 I've seen this problem in the past with "unkillable servers" I'm not sure if it was fd or not. However, I just recently noticed the same thing once with fd. Not only can you not kill fd, but you can quickly create a "chain of no-death" as trying to kill the "kill fd" process will also hang. On Fri, Feb 27, 1998 at 12:59:09PM -0500, Marcel Jackowski wrote: > Hi all, > > It seems like that the fd server once is running cannot be killed. > It looks the problem is related somehow to this second thread which > sends timer messages to the server, but I am still unable to nail it down. > I have written a device server for an old Mitsumi 1x CDROM, which mimics > most of the functionality of our fd server. As a result, it suffers from > the same problem - the server works perfectly, but if I say "kill server-pid" > from the prompt, it just hangs. Note that, a 'ps' will list 2 threads for > the server. The first gets killed but never the last one. > > Any suggestions ? > > Marcel. > mjack@cs.wright.edu > > -- David Jeske (N9LCA) + http://www.chat.net/~jeske/ + jeske@chat.net From daemon Wed Mar 11 11:24:51 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id LAA05048 for vsta-xplod; Wed, 11 Mar 1998 11:24:51 GMT Received: from note.orchestra.cse.unsw.EDU.AU (root@note.orchestra.cse.unsw.EDU.AU [129.94.242.29]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id LAA05042 for ; Wed, 11 Mar 1998 11:24:47 GMT Received: From flute09 With LocalMail ; Thu, 12 Mar 98 08:54:15 +1100 From: Binh Thai To: vsta@zendo.com Date: Thu, 12 Mar 1998 08:54:14 +1100 (EST) X-Sender: s2156593@flute09.orchestra.cse.unsw.EDU.AU Subject: MGR Programming Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hi everyone, I'm trying to write an user interface using MGR, but so far without much success. This is what I'm trying to do... * When the user starts the program, a new window should pop up. This new window will display a plot of some activities. This is my 1st step. Unfornately I'm having trouble creating a new window. I used the macro in term.h m_newwin() A new window is created, but I can't do anything to that window (not even printing hello world). I tried using the examples given in the directory /vsta/mgr/src/clients/tests/half.c, but the macro m_halfwin() crashes mgr. I am just wondering has anyone out there tried to do this sort of stuff? If so, can you tell me how to do it properly? Thank you. Binh Thai From daemon Thu Mar 12 08:22:11 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id IAA07054 for vsta-xplod; Thu, 12 Mar 1998 08:22:11 GMT Received: from gateway.s3.com (gateway.s3.com [198.211.58.4]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id IAA07048 for ; Thu, 12 Mar 1998 08:22:01 GMT Received: from chat.net (genesis.jeske.meer.net [204.94.139.85]) by gateway.s3.com (8.6.12/8.6.12) with ESMTP id KAA26407 for ; Thu, 12 Mar 1998 10:26:16 -0800 Received: by chat.net (SMI-8.6/SMI-SVR4) id KAA00054; Thu, 12 Mar 1998 10:54:03 -0800 Message-ID: <19980312105402.49585@home.chat.net> Date: Thu, 12 Mar 1998 10:54:02 -0800 From: David Jeske To: vsta@zendo.com Subject: Re: MGR Programming Mail-Followup-To: vsta@zendo.com References: <199803121703.JAA29325@vandys-pc.cisco.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.90.2 In-Reply-To: <199803121703.JAA29325@vandys-pc.cisco.com>; from Andy Valencia on Thu, Mar 12, 1998 at 09:03:39AM -0800 On Thu, Mar 12, 1998 at 09:03:39AM -0800, Andy Valencia wrote: > The real pain with debugging this level of MGR is that your screen and > keyboard are held by MGR, so there's not a good place to run the debugger. > Back when I was hacking on this, I was on the road with my 2-pound laptop, > and was actually running adb on the terminal emulator of my Psion 3 > connected to the COM1 port! oh my... -- David Jeske (N9LCA) + http://www.chat.net/~jeske/ + jeske@chat.net From daemon Thu Mar 12 06:34:43 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id GAA06884 for vsta-xplod; Thu, 12 Mar 1998 06:34:43 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id GAA06878 for ; Thu, 12 Mar 1998 06:34:39 GMT Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id JAA29325; Thu, 12 Mar 1998 09:03:39 -0800 (PST) Message-Id: <199803121703.JAA29325@vandys-pc.cisco.com> To: Binh Thai cc: vsta@zendo.com Subject: Re: MGR Programming In-reply-to: Your message of "Thu, 12 Mar 1998 08:54:14 +1100." Date: Thu, 12 Mar 1998 09:03:39 -0800 From: Andy Valencia [Binh Thai writes:] > * When the user starts the program, a new window should pop up. >This new window will display a plot of some activities. So it will still be a application window? > This is my 1st step. Unfornately I'm having trouble creating a new >window. I used the macro in term.h > m_newwin() > A new window is created, but I can't do anything to that window >(not even printing hello world). I tried using the examples given in the >directory /vsta/mgr/src/clients/tests/half.c, but the macro m_halfwin() >crashes mgr. Groan, yes. As I recollect, halfwin's are a hack to marry a window with a pseudo-tty. I never worked on this part, aside from making it compile during the initial port. A better bet would be to look at the pop-up windows which icon/zoom.c uses. > I am just wondering has anyone out there tried to do this sort of >stuff? If so, can you tell me how to do it properly? I wrestled with some thorny edge cases in the multiple window handling, and got it working acceptably about a year ago. Bit rot could certainly have won back this fix, or perhaps a new variation has come up. The real pain with debugging this level of MGR is that your screen and keyboard are held by MGR, so there's not a good place to run the debugger. Back when I was hacking on this, I was on the road with my 2-pound laptop, and was actually running adb on the terminal emulator of my Psion 3 connected to the COM1 port! Andy From daemon Tue Mar 17 12:41:21 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id MAA20273 for vsta-xplod; Tue, 17 Mar 1998 12:41:21 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id MAA20267 for ; Tue, 17 Mar 1998 12:41:17 GMT Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id PAA13101; Tue, 17 Mar 1998 15:09:52 -0800 (PST) Message-Id: <199803172309.PAA13101@vandys-pc.cisco.com> To: pramod@rishi.serc.iisc.ernet.in (Pramod B. S.) cc: vsta@zendo.com Subject: Re: Strange output with threads In-reply-to: Your message of "Wed, 18 Mar 1998 00:01:30 +0530." <9803171831.AA29036@rishi.serc.iisc.ernet.in> Date: Tue, 17 Mar 1998 15:09:52 -0800 From: Andy Valencia [pramod@rishi.serc.iisc.ernet.in (Pramod B. S.) writes:] > I am running VSTa 1.6 on a i486. The following code produces the > output listed below. Please observe that the Output shown does > not list all the four printf messages and prints the same > message("Main thread exiting") twice. > Any ideas on what could be the possible mistake? The default VSTa C library has a standard I/O which is not thread safe. It's a tradeoff of performance of the 99% case (single threaded POSIX app) versus multi-threaded applications. Andy Valencia From daemon Fri Apr 17 13:54:38 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id NAA08516 for vsta-xplod; Fri, 17 Apr 1998 13:54:38 GMT Received: from baruk2.zendo.com (baruk2.zendo.com [205.187.71.65]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id NAA08510 for ; Fri, 17 Apr 1998 13:54:34 GMT Received: (from root@localhost) by baruk2.zendo.com (8.8.7/8.8.7) id RAA09617; Fri, 17 Apr 1998 17:59:24 -0700 (PDT) (envelope-from bmccoy@bbnplanet.com) Message-ID: <3537C970.CF933333@bbnplanet.com> Date: Fri, 17 Apr 1998 14:28:16 -0700 From: Brett McCoy Organization: GTE Internetworking X-Mailer: Mozilla 4.04 [en] (WinNT; U) MIME-Version: 1.0 To: vsta@zendo.com Subject: bugs (features) in vstafs? Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit After several years of being too busy to do anything but work, I've finally gotten a reprieve and decided to take a look at vsta again. 1.6.1 is very nice and other than some stupid end of file mangling by winzip installed without any major problems. Booting with grub is also very cool. While playing around with the vstafs I discovered what I consider a bug. If I try to 'tar xvf' any of the source archives into a vstafs filesystem, I get a ton of error messages about not being able to set mtime and atime. I dug through the vstafs code and found in stat.c where it will allow mtime to be set, but no other stat. Since there isn't an atime stat for vstafs files I added some code to just ignore the request instead of returning an error (a NOP in effect). That fixed the tar problem and I was able to unpack everything into the vstafs filesystem. I'm wondering if this behavior is what was intended? Second, I discovered, after accidently removing chmod permissions from a directory, that I couldn't change it's acc list, even though I owned the file. I dug through the stat.c code some more and discovered where it was specifically allowing the owner of a file to do things even if the permissions didn't allow it. However, that didn't seem to make any difference, and after some debugging discovered that it was never making it to the stat write routine. I then dug some more and found that the open routine was also checking permissions and was the one blocking my request. I added the same extra check to open and then I was able to update the acc list. I made the changes on my machine at home, so I don't have the source to diff right now. I'll send out the diff's tonight or tomorrow. Finally, I thought I would mention that I'm running vsta on a Syquest SPARQ 1G removable drive (it's the only IDE drive in my system). It works fine. I have grub installed in the bootblock and the system boots from it just fine. It's not the worlds fastest drive (significantly slower than my barracuda), but it's livable. I'm thinking about stopping by Fry's to pick up one of their dirt cheap 6G IDE drives to make things go a little faster, but it's cool that it works and is a good way to have a dedicated vsta drive if you can't spare a drive in your system. ++Brett; From daemon Sat May 2 03:08:50 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id DAA16058 for vsta-xplod; Sat, 2 May 1998 03:08:50 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id DAA16052 for ; Sat, 2 May 1998 03:08:47 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by zipper.cisco.com (8.8.5-Cisco.2-SunOS.5.5.1.sun4/8.6.5) with ESMTP id GAA28827; Sat, 2 May 1998 06:40:26 -0700 (PDT) Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id GAA27221; Sat, 2 May 1998 06:40:25 -0700 (PDT) Message-Id: <199805021340.GAA27221@vandys-pc.cisco.com> To: "roman puls" cc: vsta@zendo.com Subject: Re: please help: booting vasta fails with "boot process x dies" In-reply-to: Your message of "Sat, 02 May 1998 15:26:19 +0200." <01bd75cd$e41d1aa0$2ae79582@puls.prz.tu-berlin.de> Date: Sat, 02 May 1998 06:40:25 -0700 From: Andy Valencia ["roman puls" writes:] >Actually found vsta on the web and and think it is a good thing. But I still >have a big problem: >I tried to install vsta version 161 on my machine. After unpacking the files >to the according paths I tried to boot from a bootdisk with grub. Parameters >in grub: > >root= (hd0,0) >kernel= /vsta/boot/vsta >module= /vsta/boot/cons >module= /vsta/boot/namer >module= /vsta/boot/wd d0:readp >module= /vsta/boot/dos -p disk/wd:wd0_dos0 fs/root Change this to "-d //disk/wd:wd0_dos0 -n fs/root". >after looking into the sources, I assume that the WD driver is having >problems with my harddisk. But I have no Idea why the boot processes die. The usage complaint message gets hidden behind the drop into the kernel debugger. Try with the new arguments and see if it fixes things. Andy From daemon Sat May 2 07:12:32 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id HAA18578 for vsta-xplod; Sat, 2 May 1998 07:12:32 GMT Received: from apollo.prz.tu-berlin.de (apollo.prz.tu-berlin.de [130.149.226.78]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id HAA18572 for ; Sat, 2 May 1998 07:12:28 GMT Received: from puls.prz.tu-berlin.de (puls.prz.tu-berlin.de [130.149.231.42]) by apollo.prz.tu-berlin.de (8.8.7/8.8.7) with SMTP id TAA15420; Sat, 2 May 1998 19:40:05 +0200 (METDST) From: "roman puls" To: "Andy Valencia" Cc: Subject: Re: please help: booting vasta fails with "boot process x dies" Date: Sat, 2 May 1998 19:43:11 +0200 Message-ID: <01bd75f1$c6a1c2e0$2ae79582@puls.prz.tu-berlin.de> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 4.71.1712.3 X-MimeOLE: Produced By Microsoft MimeOLE V4.71.1712.3 Hi Andy, [and all the other folks,] >>module= /vsta/boot/dos -p disk/wd:wd0_dos0 fs/root > >Change this to "-d //disk/wd:wd0_dos0 -n fs/root". > >>after looking into the sources, I assume that the WD driver is having >>problems with my harddisk. But I have no Idea why the boot processes die. > >The usage complaint message gets hidden behind the drop into the kernel >debugger. Try with the new arguments and see if it fixes things. thank you for the fast response! Everything is working fine now! vsta starts up and I can login. After reading the mail-archives there still are questions: 1) it seems as every access to the harddisk is _very_ slow. It took my machine 3-4 seconds from "less readme" until the first line is displayed. The command "ls" takes a lot of time to react, too. Is the FAT-filesystem as slow or is this behaviour machine (/hd) -dependent? (actually i use a pentium 200 with a fast hd) 2) does there is an actual todo-list? could only find elder ones in the mailing-archive. Im interested in doing some network stuff ;) 3) Just a remark: the installation tutorial on the website is probably outdated (http://www.zendo.com/vsta/Tutorial/Install.html). have a nice day roman From daemon Sat May 2 07:15:53 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id HAA18593 for vsta-xplod; Sat, 2 May 1998 07:15:53 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id HAA18587 for ; Sat, 2 May 1998 07:15:50 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by zipper.cisco.com (8.8.5-Cisco.2-SunOS.5.5.1.sun4/8.6.5) with ESMTP id KAA29813; Sat, 2 May 1998 10:47:30 -0700 (PDT) Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id KAA27612; Sat, 2 May 1998 10:47:30 -0700 (PDT) Message-Id: <199805021747.KAA27612@vandys-pc.cisco.com> To: "roman puls" cc: vsta@zendo.com Subject: Re: please help: booting vasta fails with "boot process x dies" In-reply-to: Your message of "Sat, 02 May 1998 19:43:11 +0200." <01bd75f1$c6a1c2e0$2ae79582@puls.prz.tu-berlin.de> Date: Sat, 02 May 1998 10:47:29 -0700 From: Andy Valencia ["roman puls" writes:] >1) it seems as every access to the harddisk is _very_ slow. It took my >machine 3-4 seconds from >"less readme" >until the first line is displayed. The command "ls" takes a lot of time to >react, too. >Is the FAT-filesystem as slow or is this behaviour machine (/hd) -dependent? >(actually i use a pentium 200 with a fast hd) No, it isn't that slow! :-) I'd guess something funny in your disk completions... it's a very straightforward IDE driver (except, of course, that it runs in user mode). >2) does there is an actual todo-list? could only find elder ones in the >mailing-archive. Im interested in doing some network stuff ;) It depends on what kind of networking you want to do. Basic IP stack is operational (though I found a memory leak I need to hunt down), as is basic clustering (i.e., transparent forwarding of VSTa kernel messages). >3) Just a remark: the installation tutorial on the website is probably >outdated (http://www.zendo.com/vsta/Tutorial/Install.html). I'll have a look again. I updated many of the files in the actual distribution in 1.6.1. Andy From daemon Sun May 3 06:09:56 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id GAA20412 for vsta-xplod; Sun, 3 May 1998 06:09:56 GMT Received: from baruk2.zendo.com (baruk2.zendo.com [205.187.71.65]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id GAA20406 for ; Sun, 3 May 1998 06:09:53 GMT Received: (from root@localhost) by baruk2.zendo.com (8.8.7/8.8.7) id KAA11971; Sun, 3 May 1998 10:21:19 -0700 (PDT) (envelope-from shawn@holwegner.com) Message-ID: <19980502191652.39098@shawn.holwegner.com> Date: Sat, 2 May 1998 19:16:52 -0700 From: Shawn Holwegner To: vsta@zendo.com Subject: Re: please help: booting vasta fails with "boot process x dies" References: <01bd75cd$e41d1aa0$2ae79582@puls.prz.tu-berlin.de> <199805021340.GAA27221@vandys-pc.cisco.com> Mime-Version: 1.0 XXContent-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.85e In-Reply-To: <199805021340.GAA27221@vandys-pc.cisco.com>; from Andy Valencia on Sat, May 02, 1998 at 06:40:25AM -0700 On Sat, May 02, 1998 at 06:40:25AM -0700, Andy Valencia wrote: > ["roman puls" writes:] > >module= /vsta/boot/dos -p disk/wd:wd0_dos0 fs/root > > Change this to "-d //disk/wd:wd0_dos0 -n fs/root". (grin) how this is familiar so many months ago. Turns out that my 386SX craptop drive died two minutes after the reinstall so I gave up on it, now i have a small 386 system i am going to install vSTA on as soon as I get a video card for it... hopefully the phantom of death will leave my old hardware alone for awhile. Shawn -- Why warm thyself beside the fire when you can be the flame? From daemon Sun May 3 09:46:46 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id JAA20694 for vsta-xplod; Sun, 3 May 1998 09:46:46 GMT Received: from email.mcmail.com (email.mcmail.com [195.44.0.12]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id JAA20688 for ; Sun, 3 May 1998 09:46:43 GMT Received: from localhostmcmail.com ([195.44.20.144]) by email.mcmail.com (9.9.9/8.8.8) with ESMTP id VAA04790 for ; Sun, 3 May 1998 21:18:41 +0100 (BST) Received: (from root@localhost) by localhostmcmail.com (8.8.7/8.8.7) id UAA00106; Sun, 3 May 1998 20:11:44 GMT Message-ID: <19980503201144.13796@mcmail.com> Date: Sun, 3 May 1998 20:11:44 +0000 From: David Welch To: vsta@zendo.com Subject: MGR doesn't work Reply-To: welch@mcmail.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.79e From daemon Mon May 4 10:05:31 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id KAA23421 for vsta-xplod; Mon, 4 May 1998 10:05:31 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id KAA23415 for ; Mon, 4 May 1998 10:05:28 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by zipper.cisco.com (8.8.5-Cisco.2-SunOS.5.5.1.sun4/8.6.5) with ESMTP id NAA12569; Mon, 4 May 1998 13:37:14 -0700 (PDT) Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id NAA01215; Mon, 4 May 1998 13:37:14 -0700 (PDT) Message-Id: <199805042037.NAA01215@vandys-pc.cisco.com> To: welch@mcmail.com cc: vsta@zendo.com Subject: Re: MGR doesn't work In-reply-to: Your message of "Sun, 03 May 1998 20:11:44 -0000." <19980503201144.13796@mcmail.com> Date: Mon, 04 May 1998 13:37:14 -0700 From: Andy Valencia Bummer. The #1 reason is that you didn't run an operable mouse driver first. There's a test program in src/srv/mach/mouse (called "test.c") which you can run after you think you have a working mouse driver running, and it'll track X/Y and button changes. After that's working right (note for RS-232 mouse you also have to run the rs232 driver first), MGR usually flies. If it doesn't, see what kind of video mode your card's in, and play with locking it to the mode MGR's expecting to use (640x480). Andy From daemon Thu May 7 23:43:57 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id XAA03469 for vsta-xplod; Thu, 7 May 1998 23:43:57 GMT Received: from mail.ptt.ru ([195.34.0.100]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id XAA03463 for ; Thu, 7 May 1998 23:43:53 GMT Received: (qmail 1363 invoked from network); 8 May 1998 10:26:15 -0000 Received: from unknown (HELO default) (195.34.27.107) by dialup.ptt.ru with SMTP; 8 May 1998 10:26:15 -0000 From: "Yuriy" To: Subject: VFS Date: Fri, 8 May 1998 14:28:13 +0400 Message-ID: <01bd7a6c$0175dcc0$6b1b22c3@default> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0018_01BD7A8D.88877CC0" X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 4.71.1712.3 X-MimeOLE: Produced By Microsoft MimeOLE V4.71.1712.3 This is a multi-part message in MIME format. ------=_NextPart_000_0018_01BD7A8D.88877CC0 Content-Type: text/plain; charset="koi8-r" Content-Transfer-Encoding: quoted-printable Where i can get VSTa VFS (or system that replase it in VSTa) information = (is it part of kernel ore server ? what massages are standart for all = object and what not ? and so so on...) ? Yuriy ------=_NextPart_000_0018_01BD7A8D.88877CC0 Content-Type: text/html; charset="koi8-r" Content-Transfer-Encoding: quoted-printable
Where i can get VSTa VFS (or = system that=20 replase it in VSTa) information (is it part of kernel ore server ? what = massages=20 are standart for all object and what not ? and so so on...) = ?
 
Yuriy
------=_NextPart_000_0018_01BD7A8D.88877CC0-- From daemon Fri May 8 03:42:12 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id DAA03986 for vsta-xplod; Fri, 8 May 1998 03:42:12 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id DAA03980 for ; Fri, 8 May 1998 03:42:10 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by zipper.cisco.com (8.8.5-Cisco.2-SunOS.5.5.1.sun4/8.6.5) with ESMTP id HAA04886; Fri, 8 May 1998 07:24:05 -0700 (PDT) Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id HAA12828; Fri, 8 May 1998 07:24:04 -0700 (PDT) Message-Id: <199805081424.HAA12828@vandys-pc.cisco.com> To: "Yuriy" cc: vsta@zendo.com Subject: Re: VFS In-reply-to: Your message of "Fri, 08 May 1998 14:28:13 +0400." <01bd7a6c$0175dcc0$6b1b22c3@default> Date: Fri, 08 May 1998 07:24:04 -0700 From: Andy Valencia ["Yuriy" writes:] >Where i can get VSTa VFS (or system that replase it in VSTa) information = >(is it part of kernel ore server ? There are no filesystems and no drivers in the kernel. vstafs, like all other filesystems, runs as its own user-mode process. >what massages are standart for all = >object and what not ? and so so on...) ? defines the message types which are understood outside the kernel but are well-known among filesystems. defines the messages types which are known by the kernel. The former outnumber the latter by a large margin. The best way to understand filesystems is to look at their source... /vsta/src/srv/{vstafs, bfs, cdfs, tmpfs, dos}. By convention, the main processing loop is in main.c, handling of open/close is in open.c, and I/O is in rw.c. This is not true as a rule, but it's a good starting point. Enjoy, Andy Valencia From daemon Mon May 11 18:19:06 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id SAA12966 for vsta-xplod; Mon, 11 May 1998 18:19:06 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id SAA12960 for ; Mon, 11 May 1998 18:19:04 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by zipper.cisco.com (8.8.5-Cisco.2-SunOS.5.5.1.sun4/8.6.5) with ESMTP id WAA25111; Mon, 11 May 1998 22:01:10 -0700 (PDT) Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id WAA29336; Mon, 11 May 1998 22:01:09 -0700 (PDT) Message-Id: <199805120501.WAA29336@vandys-pc.cisco.com> To: magpie@techline.com cc: vsta@zendo.com Subject: Re: Installing vsta-1.6.1 on other than 1st Dos partition? In-reply-to: Your message of "Mon, 11 May 1998 21:25:50 PDT." <3.0.2.32.19980511212550.00909200@mail.techline.com> Date: Mon, 11 May 1998 22:01:09 -0700 From: Andy Valencia [magpie@techline.com writes:] > Dos f: (VSTa; 240Mb available) >What arguments should I pass to /vsta/boot/dos on the command line (from >Grub) to establish the root filesystem? Grub accepts these lines: >... > module= /vsta/boot/dos -d //disk/wd:wd0_dos0 > -n fs/root (all on one line) This is the line you'll have to play with. >The system actually establishes a filesystem (on the wrong partition?), but >then can't find inittab and 'proc 9 dies'. I think the 'wd0_dos0' is wrong >for the dos f: partition, but what should it be? I tried reading the source >files (I am not a software person, alas, but an ex-EE) and have tried some >guesses: wd0_dos3, wd0_dos6, wd0_p3, wd0_p6 ...but I'm shooting in the >dark. Time to ask those who know :) Now, these are all partitions on the FIRST IDE disk, right? Because we'll have to start playing with "wd1_*" stuff if not. Replace the last "/vsta/boot/init" line with "/vsta/boot/testsh", and now you should drop into a little shell environment (if your DOS server bombs and drops you into the kernel debugger, just "q" and you'll be back to the running system, sans DOS filesystem, but that's OK). Now do: % mount disk/wd /wd % cd /wd % ls (You should see the WD partitions) You can then do ^Z (control-Z) to get back to the kernel debugger, and then "reboot". Hopefully this'll give you an idea of what DOS partitions can be found. Note that VSTa does not understand compressed drives, nor FAT-32 (actually, FAT-32 wouldn't be hard to add, but I just can't find time these days). Also, if your drive's using a pseudo-geometry (i.e, how Windows numbers the tracks does not match the natural geometry reported by the controller--this is common for disks above 1 gigabyte in size), it's easy for the filesystem to be incomprehensible to VSTa. The most common sympton of this is for the partition to be there, but the DOS server takes one look at it and exits in abject terror. Andy Valencia From daemon Mon May 11 17:51:19 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id RAA12924 for vsta-xplod; Mon, 11 May 1998 17:51:19 GMT Received: from mx2.techline.com (mx2.techline.com [204.201.36.17]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id RAA12918 for ; Mon, 11 May 1998 17:51:16 GMT From: magpie@techline.com Received: from lucy (ip451.pm9net.ips.techline.com [204.201.38.166]) by mx2.techline.com (8.8.5/8.8.5) with SMTP id VAA27774 for ; Mon, 11 May 1998 21:35:13 -0700 Message-Id: <3.0.2.32.19980511212550.00909200@mail.techline.com> X-Sender: magpie@mail.techline.com X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.2 (32) Date: Mon, 11 May 1998 21:25:50 -0700 To: vsta@zendo.com Subject: Installing vsta-1.6.1 on other than 1st Dos partition? Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Hello! In the last 2 days I have downloaded & installed vsta-1.6.1, but the disk space I have available for VSTa is not the first Dos partition. Rather, it is the 3rd logical partition inside a Dos extended partition, and there is of course a primary partition. Like this: Primary - Dos c: (dos & win3.1; full up) extended - (contains 3 logical parts) Dos d: (NT 4.0) Dos e: (archive) Dos f: (VSTa; 240Mb available) What arguments should I pass to /vsta/boot/dos on the command line (from Grub) to establish the root filesystem? Grub accepts these lines: root= (hd0,6) (filesystem is FAT) kernel= /vsta/boot/vsta (kernel loads...) module= /vsta/boot/cons module= /vsta/boot/namer module= /vsta/boot/wd d0:readp module= /vsta/boot/dos -d //disk/wd:wd0_dos0 -n fs/root (all on one line) module= /vsta/boot/init boot The system actually establishes a filesystem (on the wrong partition?), but then can't find inittab and 'proc 9 dies'. I think the 'wd0_dos0' is wrong for the dos f: partition, but what should it be? I tried reading the source files (I am not a software person, alas, but an ex-EE) and have tried some guesses: wd0_dos3, wd0_dos6, wd0_p3, wd0_p6 ...but I'm shooting in the dark. Time to ask those who know :) Thankyou for taking the time to read this long-winded post and for any help you can give. Scott W. Lucas, from the time-warp of Gray's Harbor, Washington, USA From daemon Tue May 12 07:02:53 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id HAA14115 for vsta-xplod; Tue, 12 May 1998 07:02:53 GMT Received: from mx1.techline.com (mx1.techline.com [204.201.36.16]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id HAA14109 for ; Tue, 12 May 1998 07:02:50 GMT From: magpie@techline.com Received: from lucy (ip464.pm9net.ips.techline.com [204.201.38.179]) by mx1.techline.com (8.8.5/8.8.5) with SMTP id KAA08406 for ; Tue, 12 May 1998 10:47:08 -0700 Message-Id: <3.0.2.32.19980512104040.0090cdf0@mail.techline.com> X-Sender: magpie@mail.techline.com (Unverified) X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.2 (32) Date: Tue, 12 May 1998 10:40:40 -0700 To: vsta@zendo.com Subject: Re: Installing vsta-1.6.1 on other than 1st Dos partition? In-Reply-To: <199805120501.WAA29336@vandys-pc.cisco.com> References: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Thankyou for your very quick answer, Andy! I tried out your suggestions immediately last night, with mixed results (detailed below). At 10:01 PM 5/11/98 -0700, you wrote: >[magpie@techline.com writes:] >>... >> module= /vsta/boot/dos -d //disk/wd:wd0_dos0 >> -n fs/root (all on one line) > >This is the line you'll have to play with. > >Now, these are all partitions on the FIRST IDE disk, right? Because >we'll have to start playing with "wd1_*" stuff if not. Yes, all of the Dos partitions (FAT 16, noncompressed) are on wd0, and I'm am trying to install into "Dos f:". There *is* a second harddrive on the system, but it is exclusively ext2fs for RedHat Linux 4.2, and (hopefully) doesn't contribute to this problem. Does wd need to scan it (wd1) also, for sake of it's sanity or completeness? >Replace the last "/vsta/boot/init" line with "/vsta/boot/testsh", >and now you should drop into a little shell environment (if your DOS >server bombs and drops you into the kernel debugger, just "q" and >you'll be back to the running system, sans DOS filesystem, but >that's OK). Now do: >% mount disk/wd /wd Mount returned 0 >% cd /wd New dir: /wd >% ls wd0 wd0_dos0 wd0_dos1 wd0_dos2 wd0_dos3 >(You should see the WD partitions) Yes, as I have typed-in above. I gather that wd0 represents the overall disk and _dos0,1,2,3 are Dos c: d: e: f: respectively. >You can then do ^Z (control-Z) to get back to the kernel debugger, >and then "reboot". > >Hopefully this'll give you an idea of what DOS partitions can be >found. Note that VSTa does not understand compressed drives, nor >FAT-32 (actually, FAT-32 wouldn't be hard to add, but I just can't >find time these days). I don't doubt it! Your return address says 'cisco.com' and VSTa is something you do for fun?? You must have the metabolism of a hummingbird :) >Also, if your drive's using a pseudo-geometry (i.e, how Windows >numbers the tracks does not match the natural geometry reported by >the controller--this is common for disks above 1 gigabyte in size), >it's easy for the filesystem to be incomprehensible to VSTa. The >most common sympton of this is for the partition to be there, but >the DOS server takes one look at it and exits in abject terror. I changed the /vsta/boot/dos command line to: module= /vsta/boot/dos -d //disk/wd:wd0_dos3 -n fs/root and put the line to load /vsta/boot/init back in. Upon rebooting, these messages scroll themselves off the screen: Init: can't find root, sleeping syslog: wd (pid 5) unit 0 1222.5M 16 heads 2484 cyl 63 sectors syslog: dos (pid 7) info: filesystem established syslog: error: init: /vsta/etc/inittab: can't open inittab Boot process 9 dies ...and it enters the debugger. Quitting the debugger sends it into never-never land. The reported disk geometry correctly corresponds to the physical geometry of the Western Digital AC31200 drive. The Dos server doesn't *appear* to be exiting, but the system can neither find root nor /vsta/etc/inittab. Any idea what I might try next? I'm not trying to do something impossible, am I? I have no idea what implementation rules I may be transgressing. Lastly, are there any functional descriptions of the server modules that I might download? I don't presently read source code well enough to work it all out ('though I'll keep trying...). Thanks again for taking the time to help! Scott W. Lucas, from the time-warp of Gray's Harbor, Washington, USA From daemon Tue May 12 10:14:18 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id KAA14343 for vsta-xplod; Tue, 12 May 1998 10:14:18 GMT Received: from ganymede.or.intel.com (ganymede.or.intel.com [134.134.248.3]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id KAA14337 for ; Tue, 12 May 1998 10:14:16 GMT Received: from ichips-ra.pdx.intel.com (ichips-ra.pdx.intel.com [137.102.192.31]) by ganymede.or.intel.com (8.8.6/8.8.5) with ESMTP id OAA22870 for ; Tue, 12 May 1998 14:12:17 -0700 (PDT) Received: from pdxcs314 (pdxcs314.pdx.intel.com [137.102.207.13]) by ichips-ra.pdx.intel.com (8.8.8/8.8.7) with ESMTP id NAA19753 for ; Tue, 12 May 1998 13:56:53 -0700 (PDT) From: Gordon Neal Received: by pdxcs314 (8.8.8/WW2.1) id NAA54772; Tue, 12 May 1998 13:56:53 -0700 Message-Id: <199805122056.NAA54772@pdxcs314> Subject: booting from floppy To: vsta@zendo.com Date: Tue, 12 May 1998 13:56:53 -0700 (PDT) X-Mailer: ELM [version 2.4ME+ PL37 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Has anyone ever tried booting vsta from floppy? Is this possible? Thanks Gordon Neal From daemon Tue May 12 10:42:38 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id KAA14391 for vsta-xplod; Tue, 12 May 1998 10:42:38 GMT Received: from home.chat.net (genesis.jeske.meer.net [204.94.139.85]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id KAA14385 for ; Tue, 12 May 1998 10:42:35 GMT Received: (from jeske@localhost) by home.chat.net (8.8.8/8.8.8) id OAA05976 for vsta@zendo.com; Tue, 12 May 1998 14:28:21 -0700 (PDT) Message-ID: <19980512142821.B5822@home.chat.net> Date: Tue, 12 May 1998 14:28:21 -0700 From: David Jeske To: vsta@zendo.com Subject: Re: booting from floppy Mail-Followup-To: vsta@zendo.com References: <199805122056.NAA54772@pdxcs314> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.92.1 In-Reply-To: <199805122056.NAA54772@pdxcs314>; from Gordon Neal on Tue, May 12, 1998 at 01:56:53PM -0700 On Tue, May 12, 1998 at 01:56:53PM -0700, Gordon Neal wrote: > Has anyone ever tried booting vsta from floppy? Is this possible? Yes, and yes... I made a boot floppy which has a FAT filesystem, and uses GRUB to launch the VSTa kernel and boot servers. It works quite well... I just took a look at my webpage and it looks like I was too lazy or forgetful to put the disk image online. So I'll do it tonight when I get home. -- David Jeske (N9LCA) + http://www.chat.net/~jeske/ + jeske@chat.net From daemon Tue May 12 13:32:17 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id NAA14651 for vsta-xplod; Tue, 12 May 1998 13:32:17 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id NAA14645 for ; Tue, 12 May 1998 13:32:15 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by zipper.cisco.com (8.8.5-Cisco.2-SunOS.5.5.1.sun4/8.6.5) with ESMTP id RAA07282; Tue, 12 May 1998 17:14:23 -0700 (PDT) Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id RAA01568; Tue, 12 May 1998 17:14:23 -0700 (PDT) Message-Id: <199805130014.RAA01568@vandys-pc.cisco.com> To: magpie@techline.com cc: vsta@zendo.com Subject: Re: Installing vsta-1.6.1 on other than 1st Dos partition? In-reply-to: Your message of "Tue, 12 May 1998 10:40:40 PDT." <3.0.2.32.19980512104040.0090cdf0@mail.techline.com> Date: Tue, 12 May 1998 17:14:23 -0700 From: Andy Valencia [magpie@techline.com writes:] >>% ls > wd0 > wd0_dos0 > wd0_dos1 > wd0_dos2 > wd0_dos3 Ah, good, all the DOS partitions. >I changed the /vsta/boot/dos command line to: > module= /vsta/boot/dos -d //disk/wd:wd0_dos3 -n fs/root >and put the line to load /vsta/boot/init back in. Upon rebooting, these >messages scroll themselves off the screen: > Init: can't find root, sleeping > syslog: wd (pid 5) unit 0 1222.5M 16 heads 2484 cyl 63 sectors > syslog: dos (pid 7) info: filesystem established > syslog: error: init: /vsta/etc/inittab: can't open inittab Ok, now remove init again, and put back testsh. Once you're up and running do: % mount fs/root /x % cd /x % ls See if you recognize any of the files. :-) You should see a "vsta", and you should be able to cd into it, and then see "boot", "etc", and so forth. You're right; since your DOS filesystem server process didn't die, it probably thinks it see *something*. From this hopefully you can find out where the disconnect is, and make "init" happy by giving him his inittab. >I'm not trying to do something impossible, am I? No, nothing stands out so far. >I have no idea what >implementation rules I may be transgressing. Lastly, are there any >functional descriptions of the server modules that I might download? I >don't presently read source code well enough to work it all out ('though >I'll keep trying...). Thanks again for taking the time to help! Hmmm, just read the comments at the front of the "main.c" file for a given server. That one, or perhaps any .h file you see in the dir, will tell you about the given module. Regards, Andy From daemon Wed May 13 19:40:32 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id TAA17584 for vsta-xplod; Wed, 13 May 1998 19:40:32 GMT Received: from home.chat.net (genesis.jeske.meer.net [204.94.139.85]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id TAA17578 for ; Wed, 13 May 1998 19:40:29 GMT Received: (from jeske@localhost) by home.chat.net (8.8.8/8.8.8) id XAA11649 for vsta@zendo.com; Wed, 13 May 1998 23:26:19 -0700 (PDT) Message-ID: <19980513232619.B11625@home.chat.net> Date: Wed, 13 May 1998 23:26:19 -0700 From: David Jeske To: *** VSTa *** Subject: VSTa Bootable Floppy Mail-Followup-To: *** VSTa *** Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.92.1 Okay, I put the VSTa bootable floppy online. If you have any trouble with it let me know.. http://www.chat.net/~jeske/VSTa/Software/ *** What it is: It's a DOS FAT formatted floppy. GRUB stage1 is installed on the boot sector and stage 2 is installed on the FAT filesystem. GRUB is setup to boot the VSTa kernel and the required servers to get the root filesystem up on a floppy... The only thing I had to do special to get this to work was to compile the fd server with static libraries. -- David Jeske (N9LCA) + http://www.chat.net/~jeske/ + jeske@chat.net From daemon Fri May 15 13:27:40 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id NAA03777 for vsta-xplod; Fri, 15 May 1998 13:27:40 GMT Received: from shell9.ba.best.com (root@shell9.ba.best.com [206.184.139.140]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id NAA03771 for ; Fri, 15 May 1998 13:27:38 GMT Received: (from hjh@localhost) by shell9.ba.best.com (8.8.8/8.8.BEST) id RAA09817; Fri, 15 May 1998 17:11:29 -0700 (PDT) From: "J. Han" Message-Id: <199805160011.RAA09817@shell9.ba.best.com> Subject: Globalized VSTa In-Reply-To: <199805130014.RAA01568@vandys-pc.cisco.com> from Andy Valencia at "May 12, 98 05:14:23 pm" To: vandys@cisco.com (Andy Valencia) Date: Fri, 15 May 1998 17:11:29 -0700 (PDT) Cc: vsta@zendo.com X-Mailer: ELM [version 2.4ME+ PL38 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Hello, I just made an indexed HTML version of VSTa source tree (sans RCS) using GLOBAL. You can access it from my homepage: http://www.best.com/~hjh Any comments welcome. yours, J Han hjh@best.com From daemon Sun May 17 01:08:45 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id BAA08833 for vsta-xplod; Sun, 17 May 1998 01:08:45 GMT Received: from mail.ptt.ru ([195.34.0.100]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id BAA08827 for ; Sun, 17 May 1998 01:08:37 GMT Received: (qmail 26966 invoked from network); 17 May 1998 11:52:30 -0000 Received: from unknown (HELO dialup.ptt.ru) (195.34.27.99) by dialup.ptt.ru with SMTP; 17 May 1998 11:52:30 -0000 Sender: root Message-ID: <355ED023.8850F79C@dialup.ptt.ru> Date: Sun, 17 May 1998 15:55:15 +0400 From: Yoda X-Mailer: Mozilla 4.04 [en] (X11; I; FreeBSD 2.2.5-RELEASE i386) MIME-Version: 1.0 To: vsta@zendo.com Subject: Spin Lock & p_sema func ? Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hi Andy ! Can you describe me how works spin-locks when somebody take the semaphor ? I don't see any "hold chek" in p_lock declaration (except debug assert). And why you don't put there (on semaphor fields) ordinary sleep-locks ? P.S. I learn OS disign (and english too) so please be tolerant even if i ask something obvious Thank you ! Yoriy Sydorin From daemon Mon May 18 08:46:20 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id IAA11867 for vsta-xplod; Mon, 18 May 1998 08:46:20 GMT Received: from isp1.alhsys.es (root@isp1.alhsys.es [195.53.58.2]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id IAA11861 for ; Mon, 18 May 1998 08:46:17 GMT Received: from mail.alhsys.es (ppp241.alhsys.es [195.53.58.241]) by isp1.alhsys.es (8.8.5/8.7.3) with ESMTP id UAA09467 for ; Mon, 18 May 1998 20:36:17 +0200 Message-ID: <35608C66.AB494DCD@mail.alhsys.es> Date: Mon, 18 May 1998 21:30:46 +0200 From: Alberto Molina X-Mailer: Mozilla 4.03 [es] (Win95; I) MIME-Version: 1.0 To: vsta@zendo.com Subject: Re: Globalized VSTa References: <199805160011.RAA09817@shell9.ba.best.com> Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Sooooo..... Is there any port of RCS to VSTa? I have just begun experimenting with it and I was thinking on a port of RCS. Can anyone help me? Thanks, Alberto P.S.: If RCS exists for VSTa, is there any other important GNU tool that needs to be ported? (Maybe we can vote the suggestions in the future) J. Han escribió: > > Hello, > > I just made an indexed HTML version of VSTa source tree (sans RCS) > using GLOBAL. You can access it from my homepage: > > http://www.best.com/~hjh > > Any comments welcome. > > yours, > J Han hjh@best.com From daemon Mon May 18 08:55:15 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id IAA11884 for vsta-xplod; Mon, 18 May 1998 08:55:15 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id IAA11878 for ; Mon, 18 May 1998 08:55:13 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by zipper.cisco.com (8.8.5-Cisco.2-SunOS.5.5.1.sun4/8.6.5) with ESMTP id MAA15553; Mon, 18 May 1998 12:39:02 -0700 (PDT) Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id MAA21928; Mon, 18 May 1998 12:39:01 -0700 (PDT) Message-Id: <199805181939.MAA21928@vandys-pc.cisco.com> To: Alberto Molina cc: vsta@zendo.com Subject: Re: Globalized VSTa In-reply-to: Your message of "Mon, 18 May 1998 21:30:46 +0200." <35608C66.AB494DCD@mail.alhsys.es> Date: Mon, 18 May 1998 12:39:01 -0700 From: Andy Valencia [Alberto Molina writes:] >Is there any port of RCS to VSTa? I have just begun experimenting with >it and I was thinking on a port of RCS. Can anyone help me? Yes, there's a port, and I use it to maintain all the VSTa source. >P.S.: If RCS exists for VSTa, is there any other important GNU tool that >needs to be ported? (Maybe we can vote the suggestions in the future) GNU emacs without unexec would probably be a reasonably interesting task. Most of the base GNU tools of the textutil/fileutil/gcc/gas/gdb ilk are already ported. bash might be nice. Andy From daemon Mon May 18 09:01:23 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id JAA11931 for vsta-xplod; Mon, 18 May 1998 09:01:23 GMT Received: from bbnplanet.com (cambridge6-mail-router1.netops.gte.com [4.0.63.75]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id JAA11925 for ; Mon, 18 May 1998 09:01:20 GMT Received: from bbnplanet.com (cam-sysadmin-dynamic23.bbnplanet.com [4.2.7.23]) by bbnplanet.com (8.8.7/8.8.7) with ESMTP id PAA05837; Mon, 18 May 1998 15:45:38 -0400 (EDT) Message-ID: <35609011.5B8EB656@bbnplanet.com> Date: Mon, 18 May 1998 12:46:25 -0700 From: Brett McCoy X-Mailer: Mozilla 4.05 [en] (WinNT; I) MIME-Version: 1.0 To: vsta@zendo.com Subject: long filenames in FAT filesystem Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Can anyone point me at a resource that explains how Win95 and NT do long filenames in FAT filesystems? I can't *stand* 8.3 filenames and I really want something better. However, I can't use vstafs as my main filesystem right now because it crashes periodically, specifically when compiling everything. I'd like to work on debugging the vstafs filesystem code, but I need a stable filesystem to work on in the meantime, and the FAT filesystem code seems stable (if a bit slow). So, I'd like to look at adding long filenames to it. If it's done in a really, really grody way, then maybe I'll work on making vstafs more stable from the getgo (but it would be nice to use the long filenames under FAT). ++Brett; From daemon Mon May 18 09:43:39 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id JAA12022 for vsta-xplod; Mon, 18 May 1998 09:43:39 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id JAA12016 for ; Mon, 18 May 1998 09:43:36 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by zipper.cisco.com (8.8.5-Cisco.2-SunOS.5.5.1.sun4/8.6.5) with ESMTP id NAA05275; Mon, 18 May 1998 13:27:26 -0700 (PDT) Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id NAA22083; Mon, 18 May 1998 13:27:25 -0700 (PDT) Message-Id: <199805182027.NAA22083@vandys-pc.cisco.com> To: Brett McCoy cc: vsta@zendo.com Subject: Re: long filenames in FAT filesystem In-reply-to: Your message of "Mon, 18 May 1998 12:46:25 PDT." <35609011.5B8EB656@bbnplanet.com> Date: Mon, 18 May 1998 13:27:25 -0700 From: Andy Valencia [Brett McCoy writes:] >Can anyone point me at a resource that explains how Win95 and NT do long >filenames in FAT filesystems? I found a reference which described much of it (enough to avoid it breaking the server, which was aborting after seeing embedded nulls in filenames), but not enough to implement it. Since then I've seen versions of mtools which appear to know how to do it, but I've never gone back to hack up the code. Andy From daemon Mon May 18 10:23:50 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id KAA12123 for vsta-xplod; Mon, 18 May 1998 10:23:50 GMT Received: from home.chat.net (genesis.jeske.meer.net [204.94.139.85]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id KAA12117 for ; Mon, 18 May 1998 10:23:47 GMT Received: (from jeske@localhost) by home.chat.net (8.8.8/8.8.8) id OAA21363 for vsta@zendo.com; Mon, 18 May 1998 14:11:15 -0700 (PDT) Message-ID: <19980518141115.B21304@home.chat.net> Date: Mon, 18 May 1998 14:11:15 -0700 From: David Jeske To: vsta@zendo.com Subject: Re: long filenames in FAT filesystem Mail-Followup-To: vsta@zendo.com References: <35609011.5B8EB656@bbnplanet.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.92.1 In-Reply-To: <35609011.5B8EB656@bbnplanet.com>; from Brett McCoy on Mon, May 18, 1998 at 12:46:25PM -0700 On Mon, May 18, 1998 at 12:46:25PM -0700, Brett McCoy wrote: > Can anyone point me at a resource that explains how Win95 and NT do long > filenames in FAT filesystems? > > I can't *stand* 8.3 filenames and I really want something better. > However, I can't use vstafs as my main filesystem right now because it > crashes periodically, specifically when compiling everything. I'd like > to work on debugging the vstafs filesystem code, but I need a stable > filesystem to work on in the meantime, and the FAT filesystem code seems > stable (if a bit slow). > > So, I'd like to look at adding long filenames to it. If it's done in a > really, really grody way, then maybe I'll work on making vstafs more > stable from the getgo (but it would be nice to use the long filenames > under FAT). It's pretty UGLY (VFAT) They use "extra" directory entries to store long filename information. Basically: 1) every file has a short filename 2) long filename entries can take up multiple directory entries. 3) MS found that most programs "leave alone" files which have all attribute bits set (or cleared, I forget), so the long filename entries have the attribute bits set this way. I seem to remember something specifically having to do with volume/archive bits. 4) the long filename for a file has a checksum of the short filename, so that if a non-long filename aware program changes the short filename only, it breaks the link with the long filename. 5) Other information can also be stored in these "modified directory entries", I don't remember what exactly I think you would be better off implementing something like UMSDOS, where you just store extended information in a "hidden" file. VSTa's handling of VFAT, if I remember correctly, is particularly not good, because VSTa reads a complete directory entry, and if it's changed, it writes out the whole entry again. Thus, just "skipping over" VFAT filenames means that when VSTa writes out the directory again, it writes right over what would be still valid VFAT long filenames. However, I could be wrong about this, it was a long time ago that I looked at it. -- David Jeske (N9LCA) + http://www.chat.net/~jeske/ + jeske@chat.net From daemon Mon May 18 10:56:38 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id KAA12174 for vsta-xplod; Mon, 18 May 1998 10:56:38 GMT Received: from shell9.ba.best.com (root@shell9.ba.best.com [206.184.139.140]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id KAA12168 for ; Mon, 18 May 1998 10:56:35 GMT Received: (from hjh@localhost) by shell9.ba.best.com (8.8.8/8.8.BEST) id OAA16677; Mon, 18 May 1998 14:39:47 -0700 (PDT) From: "J. Han" Message-Id: <199805182139.OAA16677@shell9.ba.best.com> Subject: Re: Globalized VSTa In-Reply-To: <35608C66.AB494DCD@mail.alhsys.es> from Alberto Molina at "May 18, 98 09:30:46 pm" To: amolina@mail.alhsys.es (Alberto Molina) Date: Mon, 18 May 1998 14:39:46 -0700 (PDT) Cc: vsta@zendo.com X-Mailer: ELM [version 2.4ME+ PL38 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Globalized VSTa contains source of VSTa only. Source of userland tools e.g. rcs are not included. Also I didn't include the files in rcs directories in Globalized VSTa: they mostly duplicated good deal of the source. best, J Han hjh@best.com From daemon Mon May 18 21:55:58 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id VAA13274 for vsta-xplod; Mon, 18 May 1998 21:55:58 GMT Received: from atrey.karlin.mff.cuni.cz (atrey.karlin.mff.cuni.cz [195.113.31.123]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id VAA13268 for ; Mon, 18 May 1998 21:55:55 GMT Received: (from pavel@localhost) by atrey.karlin.mff.cuni.cz (8.8.8/8.8.8) id KAA08745; Tue, 19 May 1998 10:38:43 +0200 Message-ID: <19980519103843.60390@atrey.karlin.mff.cuni.cz> Date: Tue, 19 May 1998 10:38:43 +0200 From: Pavel Machek To: David Jeske Cc: vsta@zendo.com Subject: Re: long filenames in FAT filesystem References: <35609011.5B8EB656@bbnplanet.com> <19980518141115.B21304@home.chat.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.84 In-Reply-To: <19980518141115.B21304@home.chat.net>; from David Jeske on Mon, May 18, 1998 at 02:11:15PM -0700 X-Warning: Not only using Windows can be dangerous to your mental health. HI! > > Can anyone point me at a resource that explains how Win95 and NT do long > > filenames in FAT filesystems? > > > > I can't *stand* 8.3 filenames and I really want something better. > > However, I can't use vstafs as my main filesystem right now because it > > crashes periodically, specifically when compiling everything. I'd like > > to work on debugging the vstafs filesystem code, but I need a stable > > filesystem to work on in the meantime, and the FAT filesystem code seems > > stable (if a bit slow). > > > > So, I'd like to look at adding long filenames to it. If it's done in a > > really, really grody way, then maybe I'll work on making vstafs more > > stable from the getgo (but it would be nice to use the long filenames > > under FAT). > > It's pretty UGLY (VFAT) They use "extra" directory entries to store long > filename information. Basically: [snip] If you want better docs, look at linux-2.1.102.tar.gz#utar/linux/fs/vfat/ or so. Pavel -- The best software in life is free (not shareware)! Pavel GCM d? s-: !g p?:+ au- a--@ w+ v- C++@ UL+++ L++ N++ E++ W--- M- Y- R+ From daemon Wed May 20 06:01:56 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id GAA16877 for vsta-xplod; Wed, 20 May 1998 06:01:56 GMT Received: from mx2.techline.com (mx2.techline.com [204.201.36.17]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id GAA16871 for ; Wed, 20 May 1998 06:01:54 GMT From: magpie@techline.com Received: from lucy (ip439.pm9net.ips.techline.com [204.201.38.154]) by mx2.techline.com (8.8.5/8.8.5) with SMTP id JAA02412 for ; Wed, 20 May 1998 09:47:47 -0700 Message-Id: <3.0.2.32.19980520094042.00910100@mail.techline.com> X-Sender: magpie@mail.techline.com X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.2 (32) Date: Wed, 20 May 1998 09:40:42 -0700 To: vsta@zendo.com Subject: Unpacking source files, & cet. Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" A few, simple, newbie questions: in what directory should I unpack gcc, make & other source distributions such that they are in the right place relative to everything else? Are there established conventions for VSTa's directory tree (as in unix)? Lastly, what needs to be done to set up gcc? A mail-archive message from 9feb98 says: "...gcc should already be configured, and was and is built under VSTa. It should continue to build, delta any accidental damage I've done to it." ..and I see a file 'gcc' (52k) in /vsta/bin that suggests that the compiler may just run as installed, provided it is installed in the right place. Is this so, or does it need to be 'built'? I am (clearly) not a programmer, but I can turn cranks... There are many other files in /vsta/bin (make, gmake, gzip, yacc, etc.) that seem to relate to the other software distributions available from the VSTa website. How do I make use of these files? Someone has already invested a lot of work here; I would like to take advantage of that, with gratitude. And muchas thanks for bearing with these "entry level" questions. Scott W. Lucas, from the time-warp of Gray's Harbor, Washington, USA From daemon Wed May 20 18:55:28 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id SAA17913 for vsta-xplod; Wed, 20 May 1998 18:55:28 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id SAA17907 for ; Wed, 20 May 1998 18:55:26 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by zipper.cisco.com (8.8.5-Cisco.2-SunOS.5.5.1.sun4/8.6.5) with ESMTP id WAA00155; Wed, 20 May 1998 22:39:22 -0700 (PDT) Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id WAA28134; Wed, 20 May 1998 22:39:21 -0700 (PDT) Message-Id: <199805210539.WAA28134@vandys-pc.cisco.com> To: magpie@techline.com cc: vsta@zendo.com Subject: Re: Unpacking source files, & cet. In-reply-to: Your message of "Wed, 20 May 1998 09:40:42 PDT." <3.0.2.32.19980520094042.00910100@mail.techline.com> Date: Wed, 20 May 1998 22:39:21 -0700 From: Andy Valencia [magpie@techline.com writes:] >A few, simple, newbie questions: in what directory should I unpack gcc, >make & other source distributions such that they are in the right place >relative to everything else? I tried to make everything so it would unpack from /vsta. Generally you can just do "cd /vsta ; tar -zxvf /.tz". >Are there established conventions for VSTa's directory tree (as in unix)? Yes, most of which are hopefully apparent in the current tree. >Lastly, what needs to be done to set up gcc? A mail-archive message from >9feb98 says: >"...gcc should already be configured, and was and is built under VSTa. It >should continue to build, delta any accidental damage I've done to it." >..and I see a file 'gcc' (52k) in /vsta/bin that suggests that the compiler >may just run as installed, provided it is installed in the right place. Is >this so, or does it need to be 'built'? I am (clearly) not a programmer, >but I can turn cranks... When you install VSTa you get a fully usable gcc binary. I'd suggest avoiding rebuilding all of that unless you want to do compiler work. >There are many other files in /vsta/bin (make, gmake, gzip, yacc, etc.) >that seem to relate to the other software distributions available from the >VSTa website. How do I make use of these files? Someone has already >invested a lot of work here; I would like to take advantage of that, with >gratitude. Your default shell should have /vsta/bin in its path. Just enter the command the same as ever. >And muchas thanks for bearing with these "entry level" questions. It's a pretty unexceptional UN*X-ish shell environment. You should refer to the documentation in /vsta/readme and /vsta/doc/roadmap.txt and /vsta/doc/faq. If you follow those instructions, you should end up with an installation which'll boot, let you log in (as shipped, /vsta/etc/passwd has an account for "vandys" with no password) and edit, compile, ls, and so forth. It's a plain old Bourne shell, except that it has emacs-style command line editing (actually, it doesn't, but when you're in canonical input for a TTY, you get it for free). Andy From daemon Thu May 21 06:07:21 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id GAA18908 for vsta-xplod; Thu, 21 May 1998 06:07:21 GMT Received: from home.chat.net (genesis.jeske.meer.net [204.94.139.85]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id GAA18902 for ; Thu, 21 May 1998 06:07:19 GMT Received: (from jeske@localhost) by home.chat.net (8.8.8/8.8.8) id JAA27320 for vsta@zendo.com; Thu, 21 May 1998 09:54:55 -0700 (PDT) Message-ID: <19980521095454.A27308@home.chat.net> Date: Thu, 21 May 1998 09:54:54 -0700 From: David Jeske To: vsta@zendo.com Subject: [meneal@ibm.net: include files for vsta?] Mail-Followup-To: vsta@zendo.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.92.1 ----- Forwarded message from Marsha Neal ----- Reply-To: "Marsha Neal" From: "Marsha Neal" To: Subject: include files for vsta? Date: Fri, 22 May 1998 00:52:46 -0400 I noticed that the .c files refer to includes such as: #include and three is no sys directory but pset.h exists in the mach directory. What is with this? Also the makefile references includes as ../../../include and this directory is not part of the distribution? Thanks Gordon ----- End forwarded message ----- -- David Jeske (N9LCA) + http://www.chat.net/~jeske/ + jeske@chat.net From daemon Thu May 21 06:13:04 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id GAA18932 for vsta-xplod; Thu, 21 May 1998 06:13:04 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id GAA18926 for ; Thu, 21 May 1998 06:13:02 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by zipper.cisco.com (8.8.5-Cisco.2-SunOS.5.5.1.sun4/8.6.5) with ESMTP id JAA09368; Thu, 21 May 1998 09:56:59 -0700 (PDT) Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id JAA29648; Thu, 21 May 1998 09:56:59 -0700 (PDT) Message-Id: <199805211656.JAA29648@vandys-pc.cisco.com> To: David Jeske cc: vsta@zendo.com Subject: Re: [meneal@ibm.net: include files for vsta?] In-reply-to: Your message of "Thu, 21 May 1998 09:54:54 PDT." <19980521095454.A27308@home.chat.net> Date: Thu, 21 May 1998 09:56:59 -0700 From: Andy Valencia [David Jeske writes:] >I noticed that the .c files refer to includes such as: >#include >and three is no sys directory but pset.h exists in the mach directory. >What is with this? Also the makefile references includes as ../../../include >and this directory is not part of the distribution? Seems like the files were not unpacked properly? From the distribution files on ftp.zendo.com: bodhi$ tar -ztvf vsta.tz | grep "pset\.h" -r-xr-xr-x 0/0 7834 Nov 9 15:32 1997 include/sys/rcs/pset.h -r-xr-xr-x 0/0 4287 Sep 22 11:48 1997 include/sys/pset.h bodhi$ So as long as you unpacked the files per the instructions (in /vsta), pset.h *is* in sys/, not mach/. The CWD relative references of source, as long as they were unpacked in the /vsta root, should walk up and over to the /vsta/include/ hierarchy. I can not, of course, speak to whether a specific relative reference is correct unless a specific source directory is identified. Andy Valencia From daemon Thu May 21 06:44:32 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id GAA18967 for vsta-xplod; Thu, 21 May 1998 06:44:32 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id GAA18961 for ; Thu, 21 May 1998 06:44:30 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by zipper.cisco.com (8.8.5-Cisco.2-SunOS.5.5.1.sun4/8.6.5) with ESMTP id KAA13969; Thu, 21 May 1998 10:28:28 -0700 (PDT) Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id KAA29845; Thu, 21 May 1998 10:28:27 -0700 (PDT) Message-Id: <199805211728.KAA29845@vandys-pc.cisco.com> To: David Jeske cc: vsta@zendo.com Subject: Re: can't kill fd server In-reply-to: Your message of "Fri, 27 Feb 1998 11:33:29 PST." <19980227113329.35318@home.chat.net> Date: Thu, 21 May 1998 10:28:27 -0700 From: Andy Valencia [David Jeske writes:] >I've seen this problem in the past with "unkillable servers" I'm not sure >if it was fd or not. However, I just recently noticed the same thing once >with fd. Not only can you not kill fd, but you can quickly create a "chain >of no-death" as trying to kill the "kill fd" process will also hang. Yes, I've looked into it, and it poses an interesting issue. The problem occurs with multi-threaded servers where threads talk to each other with kernel IPC. What happens is that a client thread is blocked waiting for an answer from the server, and the server thread dies without unblocking this client. If the server were in a different process, the exit of the last thread causes all ports to be torn down, which would unblock the thread. However, in this case, the waiting thread is in the same process, so the server port is not scrubbed, and a deadlock results. I'm looking at a change so that server ports are related to the thread which created them. Thus, it only requires the death of the server thread to cause a port to be scrubbed, which breaks this deadlock. This should work for most server which come to mind, but I want to look at some "pool of thread" servers which exist. The KA9Q port comes to mind, for instance. Andy From daemon Thu May 21 07:33:44 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id HAA19038 for vsta-xplod; Thu, 21 May 1998 07:33:44 GMT Received: from home.chat.net (genesis.jeske.meer.net [204.94.139.85]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id HAA19032 for ; Thu, 21 May 1998 07:33:42 GMT Received: (from jeske@localhost) by home.chat.net (8.8.8/8.8.8) id LAA27595 for vsta@zendo.com; Thu, 21 May 1998 11:21:19 -0700 (PDT) Message-ID: <19980521112118.A27528@home.chat.net> Date: Thu, 21 May 1998 11:21:18 -0700 From: David Jeske To: vsta@zendo.com Subject: Re: can't kill fd server Mail-Followup-To: vsta@zendo.com References: <19980227113329.35318@home.chat.net> <199805211728.KAA29845@vandys-pc.cisco.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.92.1 In-Reply-To: <199805211728.KAA29845@vandys-pc.cisco.com>; from Andy Valencia on Thu, May 21, 1998 at 10:28:27AM -0700 On Thu, May 21, 1998 at 10:28:27AM -0700, Andy Valencia wrote: > [David Jeske writes:] > > >I've seen this problem in the past with "unkillable servers" I'm not sure > >if it was fd or not. However, I just recently noticed the same thing once > >with fd. Not only can you not kill fd, but you can quickly create a "chain > >of no-death" as trying to kill the "kill fd" process will also hang. > > Yes, I've looked into it, and it poses an interesting issue. The problem > occurs with multi-threaded servers where threads talk to each other with > kernel IPC. What happens is that a client thread is blocked waiting for an > answer from the server, and the server thread dies without unblocking this > client. If the server were in a different process, the exit of the last Do you mean "if the server were in another process and _it_ was being closed down" or "if the server were in another process and the client was being closed down"? Or to rephrase, does the problem occur because the server side of the IPC deadlocks, or because the client side of the IPC deadlocks? > thread causes all ports to be torn down, which would unblock the thread. > However, in this case, the waiting thread is in the same process, so the > server port is not scrubbed, and a deadlock results. > I'm looking at a change so that server ports are related to the thread which > created them. Thus, it only requires the death of the server thread to > cause a port to be scrubbed, which breaks this deadlock. This should work > for most server which come to mind, but I want to look at some "pool of > thread" servers which exist. The KA9Q port comes to mind, for instance. I don't understand why there is a relationship between killing a thread and unblocking it. Why does it need to be unblocked at all? -- David Jeske (N9LCA) + http://www.chat.net/~jeske/ + jeske@chat.net From daemon Thu May 21 08:23:49 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id IAA19106 for vsta-xplod; Thu, 21 May 1998 08:23:49 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id IAA19100 for ; Thu, 21 May 1998 08:23:47 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by zipper.cisco.com (8.8.5-Cisco.2-SunOS.5.5.1.sun4/8.6.5) with ESMTP id MAA25222; Thu, 21 May 1998 12:07:44 -0700 (PDT) Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id MAA00505; Thu, 21 May 1998 12:07:44 -0700 (PDT) Message-Id: <199805211907.MAA00505@vandys-pc.cisco.com> To: newsham@lava.net (Tim Newsham) cc: vsta@zendo.com Subject: Re: can't kill fd server In-reply-to: Your message of "Thu, 21 May 1998 08:54:25 -1000." Date: Thu, 21 May 1998 12:07:43 -0700 From: Andy Valencia [newsham@lava.net (Tim Newsham) writes:] >Hmm.. would it still be possible to hand off ports between threads >after this change was made? That's precisely the issue. No, it wouldn't, and yes, I think it would cause trouble with current applications. As to David's questions: >Do you mean "if the server were in another process and _it_ was being >closed down" or "if the server were in another process and the client was >being closed down"? Server in another process, and server closing down (being killed, i.e., an asynchronous event to server execution). >Or to rephrase, does the problem occur because the server side of the >IPC deadlocks, or because the client side of the IPC deadlocks? It's a process which is a server, but also has another thread within the process which is a client of this same server. >I don't understand why there is a relationship between killing a thread >and unblocking it. Why does it need to be unblocked at all? Because a thread in the blocked state represents a number of data structures which must be moved through further state transitions. For instance, it has taken the portref, and must come back of it for any other thread to be able to use that connection. There's also the portref's state, which also goes back to the mapping of client buffers into the server, and how that reference is cleaned up. Also, for the last thread in a process, it is within this thread context that a number of potentially slow actions are taken--for instance, closing all open connections. This all comes down to synchronous IPC. Synchronous IPC means synchronous with respect to, basically, a thread. If the thread suddenly becomes asynchronous, most of the design optimizations you gain from synchronous IPC go away. At least as I envision the fallout of changing VSTa to, for instance, permit the thread to be silently axed. Andy From daemon Thu May 21 08:10:10 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id IAA19089 for vsta-xplod; Thu, 21 May 1998 08:10:10 GMT Received: from malasada.lava.net (root@malasada.lava.net [199.222.42.2]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id IAA19083 for ; Thu, 21 May 1998 08:10:08 GMT Received: from localhost (1060 bytes) by malasada.lava.net via sendmail with P:stdio/R:inet_hosts/T:smtp (sender: ) (ident using unix) id for ; Thu, 21 May 1998 08:54:26 -1000 (HST) (Smail-3.2.0.101 1997-Dec-17 #2 built 1998-Mar-12) Message-Id: From: newsham@lava.net (Tim Newsham) Subject: Re: can't kill fd server To: vandys@cisco.com (Andy Valencia) Date: Thu, 21 May 1998 08:54:25 -1000 (HST) Cc: jeske@home.chat.net, vsta@zendo.com In-Reply-To: <199805211728.KAA29845@vandys-pc.cisco.com> from "Andy Valencia" at May 21, 98 10:28:27 am X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit > I'm looking at a change so that server ports are related to the thread which > created them. Thus, it only requires the death of the server thread to > cause a port to be scrubbed, which breaks this deadlock. This should work > for most server which come to mind, but I want to look at some "pool of > thread" servers which exist. The KA9Q port comes to mind, for instance. Hmm.. would it still be possible to hand off ports between threads after this change was made? > Andy Tim N. From daemon Thu May 21 12:39:13 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id MAA19443 for vsta-xplod; Thu, 21 May 1998 12:39:13 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id MAA19437 for ; Thu, 21 May 1998 12:39:10 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by zipper.cisco.com (8.8.5-Cisco.2-SunOS.5.5.1.sun4/8.6.5) with ESMTP id QAA10320; Thu, 21 May 1998 16:23:08 -0700 (PDT) Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id QAA02060; Thu, 21 May 1998 16:23:08 -0700 (PDT) Message-Id: <199805212323.QAA02060@vandys-pc.cisco.com> To: Yoda cc: vsta@zendo.com Subject: Re: Spin Lock & p_sema func ? In-reply-to: Your message of "Sun, 17 May 1998 15:55:15 +0400." <355ED023.8850F79C@dialup.ptt.ru> Date: Thu, 21 May 1998 16:23:08 -0700 From: Andy Valencia [Yoda writes:] >Can you describe me how works spin-locks when somebody take the semaphor >? Semaphores and spinlocks occupy different levels in the hierarchy of locking within VSTa. Spinlocks are at the lowest level; it is a fatal error to sleep on a semaphore while holding a lock. Since spinlocks do not relinquish the CPU, they have only two uses: - On uniprocessor systems, they disable interrupts - On multiprocessor systems, they coordinate low-level access to a shared structure. >I don't see any "hold check" in p_lock declaration (except debug >assert). On a multiprocessor, when you see the lock held, you spin until some other CPU who holds the lock releases it. On a uniprocessor, if you see the lock held, since there isn't another CPU, you can assume you have a locking bug. >And why you don't put there (on semaphor fields) ordinary sleep-locks ? On fields within a sema_t? A semaphore is the basic primitive for sleeping. So the mutex to manipulate this primitive has to be non-sleeping--that is, a spinlock. Andy From daemon Mon May 25 05:54:09 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id FAA00173 for vsta-xplod; Mon, 25 May 1998 05:54:09 GMT Received: from mail.ptt.ru ([195.34.0.100]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id FAA00167 for ; Mon, 25 May 1998 05:54:02 GMT Received: (qmail 20315 invoked from network); 25 May 1998 16:40:28 -0000 Received: from unknown (HELO dialup.ptt.ru) (195.34.27.112) by dialup.ptt.ru with SMTP; 25 May 1998 16:40:28 -0000 Sender: root Message-ID: <35699FA6.3A9CD934@dialup.ptt.ru> Date: Mon, 25 May 1998 20:43:18 +0400 From: Yoda X-Mailer: Mozilla 4.04 [en] (X11; I; FreeBSD 2.2.5-RELEASE i386) MIME-Version: 1.0 To: vsta@zendo.com Subject: Terminal Help ! Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hi everybody ! Question number 1: I found interesting thing: there is no TERMCAP or TERMPATH variable are set intialy... Code from termcap.c, according to the source comments, should search termcap database in /etc/termcap while it in /vsta/lib/tarmcap but all works fine !? I can't understand this ! Did i miss something ? Question number 2: I can't move coursor in Emacs (but i can type text normally) - it just writes some strange symbols on the screen and nothing more ! I tryed set TERM to 'nansi', 'ansi', 'vt100'... tryed install cons2 server - all useless ! P.S. I guess it can be interesting for novice... When i untarred VSTa by WinZIP in Windoze i found huge amount of bugs... but when i did this normly (under FreeBSD) all bugs gone ... WinZIP adds \cr \nl to the end of string when it wark i guess it was main cause of bugs... Yuriy Sidorin From daemon Mon May 25 07:25:18 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id HAA00262 for vsta-xplod; Mon, 25 May 1998 07:25:18 GMT Received: from home.chat.net (genesis.jeske.meer.net [204.94.139.85]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id HAA00256 for ; Mon, 25 May 1998 07:25:16 GMT Received: (from jeske@localhost) by home.chat.net (8.8.8/8.8.8) id LAA20856 for vsta@zendo.com; Mon, 25 May 1998 11:15:00 -0700 (PDT) Message-ID: <19980525111459.D20068@home.chat.net> Date: Mon, 25 May 1998 11:14:59 -0700 From: David Jeske To: vsta@zendo.com Subject: Re: Terminal Help ! Mail-Followup-To: vsta@zendo.com References: <35699FA6.3A9CD934@dialup.ptt.ru> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.92.1 In-Reply-To: <35699FA6.3A9CD934@dialup.ptt.ru>; from Yoda on Mon, May 25, 1998 at 08:43:18PM +0400 On Mon, May 25, 1998 at 08:43:18PM +0400, Yoda wrote: > Question number 2: > I can't move coursor in Emacs (but i can type text normally) - it just > writes some strange symbols on the screen and nothing more ! > I tryed set TERM to 'nansi', 'ansi', 'vt100'... tryed install cons2 > server - all useless ! Try the "real" emacs movement characters: control-N - next line (down) control-P - previous line (up) control-F - forward (right) control-B - backwards (left) I think that's right anyhow. -- David Jeske (N9LCA) + http://www.chat.net/~jeske/ + jeske@chat.net From daemon Mon May 25 18:40:35 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id SAA01228 for vsta-xplod; Mon, 25 May 1998 18:40:35 GMT Received: from mx1.techline.com (mx1.techline.com [204.201.36.16]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id SAA01222 for ; Mon, 25 May 1998 18:40:33 GMT From: magpie@techline.com Received: from lucy (ip377.pm10net.ips.techline.com [204.201.38.92]) by mx1.techline.com (8.8.5/8.8.5) with SMTP id WAA24565 for ; Mon, 25 May 1998 22:29:03 -0700 Message-Id: <3.0.2.32.19980525222446.00911680@mail.techline.com> X-Sender: magpie@mail.techline.com X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.2 (32) Date: Mon, 25 May 1998 22:24:46 -0700 To: vsta@zendo.com Subject: re: Gcc can't execute cpp Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Thanks, Yoda! I am a novice too! When I read about your experience of many bugs related to the use of WinZip, I blew off my whole installation & started over using Linux to untar VSTa. Wonderful !! Emacs, vi, less and gcc all work flawlessly now with no other changes! Oh, I did have to change one line in the fd makefile to: fd: $(OBJS) $(LD) $(LDFLAGS) -o fd ../../../../lib/crt0.o $(OBJS) $(LIBS) (cd up one more level so it would find crt0.o) There were also other troubles running rc and in /env, all gone now in one stroke. Whatever corruption WinZip is causing, it is to be avoided. Thankyou, Yuriy! Yoda wrote: >P.S. I guess it can be interesting for novice...When i >untarred VSTa by WinZIP in Windoze i found huge amount of >bugs...but when i did this normly (under FreeBSD) all bugs >gone ... WinZIPadds \cr \nl to the end of string when it >wark i guess it wasmain cause of bugs... > >Yuriy Sidorin > Scott W. Lucas, from the time-warp of Gray's Harbor, Washington, USA From daemon Mon May 25 17:52:43 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id RAA01177 for vsta-xplod; Mon, 25 May 1998 17:52:43 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id RAA01171 for ; Mon, 25 May 1998 17:52:42 GMT Received: from localhost (vandys@localhost) by zipper.cisco.com (8.8.5-Cisco.2-SunOS.5.5.1.sun4/8.6.5) with SMTP id VAA12896; Mon, 25 May 1998 21:38:06 -0700 (PDT) Message-Id: <199805260438.VAA12896@zipper.cisco.com> To: Yoda cc: vsta@zendo.com Subject: Re: Terminal Help ! In-reply-to: Your message of "Mon, 25 May 1998 20:43:18 +0400." <35699FA6.3A9CD934@dialup.ptt.ru> Date: Mon, 25 May 1998 21:38:06 -0700 From: Andrew Valencia [Yoda writes:] >I found interesting thing: there is no TERMCAP or TERMPATH variable are >set intialy... >Code from termcap.c, according to the source comments, should search >termcap database in /etc/termcap while it in /vsta/lib/tarmcap but all >works fine !? See and _PATH_TERMCAP. >When i untarred VSTa by WinZIP in Windoze i found huge amount of bugs... >but when i did this normly (under FreeBSD) all bugs gone ... WinZIP >adds \cr \nl to the end of string when it wark i guess it was main cause >of bugs... Right, \r\n line termination breaks the C preprocessor. I hacked the previous port of GNU C to handle this, but that was when a DOS cross compilation environment was fresher in my memory. For the current port of GNU C, you need to strip the \r's off to build under GCC (this is the same requirement any standard UNIX would impose). Andy From daemon Mon May 25 19:39:10 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id TAA01292 for vsta-xplod; Mon, 25 May 1998 19:39:10 GMT Received: from mail.ptt.ru ([195.34.0.100]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id TAA01286 for ; Mon, 25 May 1998 19:39:07 GMT Received: (qmail 14127 invoked from network); 26 May 1998 06:25:35 -0000 Received: from dialup-30019.dialup.ptt.ru (HELO default) (195.34.30.19) by dialup.ptt.ru with SMTP; 26 May 1998 06:25:35 -0000 From: "Yoda" To: Subject: =?iso-8859-1?B?zvLi5fI6IFRlcm1pbmFsIEhlbHAgIQ==?= Date: Tue, 26 May 1998 10:28:30 +0400 Message-ID: <01bd886f$80174d00$131e22c3@default> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 8bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 4.71.1712.3 X-MimeOLE: Produced By Microsoft MimeOLE V4.71.1712.3 -----Èñõîäíîå ñîîáùåíèå----- Îò: David Jeske Êîìó: vsta@zendo.com Äàòà: 26 ìàÿ 1998 ã. 7:04 Òåìà: Re: Terminal Help ! >On Mon, May 25, 1998 at 08:43:18PM +0400, Yoda wrote: >> Question number 2: >> I can't move coursor in Emacs (but i can type text normally) - it just >> writes some strange symbols on the screen and nothing more ! >> I tryed set TERM to 'nansi', 'ansi', 'vt100'... tryed install cons2 >> server - all useless ! > >Try the "real" emacs movement characters: > >control-N - next line (down) >control-P - previous line (up) >control-F - forward (right) >control-B - backwards (left) > >I think that's right anyhow. Thank you ! Have Microemacs color mode (i mean "c" highlite mode) ? And how turn it on ? > >-- >David Jeske (N9LCA) + http://www.chat.net/~jeske/ + jeske@chat.net > From daemon Tue May 26 07:21:23 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id HAA02516 for vsta-xplod; Tue, 26 May 1998 07:21:23 GMT Received: from bbnplanet.com (cambridge6-mail-router1.netops.gte.com [4.0.63.75]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id HAA02510 for ; Tue, 26 May 1998 07:21:19 GMT Received: from bbnplanet.com (meson.barrnet.net [204.160.72.220]) by bbnplanet.com (8.8.7/8.8.7) with ESMTP id OAA03870; Tue, 26 May 1998 14:07:48 -0400 (EDT) Message-ID: <356B050C.25A8407B@bbnplanet.com> Date: Tue, 26 May 1998 11:08:12 -0700 From: Brett McCoy Organization: GTE Internetworking X-Mailer: Mozilla 4.04 [en] (WinNT; U) MIME-Version: 1.0 To: magpie@techline.com CC: vsta@zendo.com Subject: Re: Gcc can't execute cpp References: <3.0.2.32.19980525222446.00911680@mail.techline.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit If you use the tar.exe that is in the vsta distribution directory on the ftp server it seems to do the right thing. ++Brett; magpie@techline.com wrote: > > Thanks, Yoda! > I am a novice too! When I read about your experience of many bugs related > to the use of WinZip, I blew off my whole installation & started over using > Linux to untar VSTa. Wonderful !! Emacs, vi, less and gcc all work > flawlessly now with no other changes! Oh, I did have to change one line in > the fd makefile to: > fd: $(OBJS) > $(LD) $(LDFLAGS) -o fd ../../../../lib/crt0.o $(OBJS) $(LIBS) > > (cd up one more level so it would find crt0.o) > There were also other troubles running rc and in /env, all gone now in one > stroke. Whatever corruption WinZip is causing, it is to be avoided. > Thankyou, Yuriy! From daemon Sat May 30 02:23:09 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id CAA11546 for vsta-xplod; Sat, 30 May 1998 02:23:09 GMT Received: from mail.ptt.ru ([195.34.0.100]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id CAA11540 for ; Sat, 30 May 1998 02:23:04 GMT Received: (qmail 26537 invoked from network); 30 May 1998 13:09:23 -0000 Received: from unknown (HELO dialup.ptt.ru) (195.34.27.82) by dialup.ptt.ru with SMTP; 30 May 1998 13:09:23 -0000 Sender: root Message-ID: <35700592.719807B8@dialup.ptt.ru> Date: Sat, 30 May 1998 17:11:47 +0400 From: Yoda X-Mailer: Mozilla 4.04 [en] (X11; I; FreeBSD 2.2.5-RELEASE i386) MIME-Version: 1.0 To: vsta@zendo.com Subject: Memory mapping... Help ! Content-Type: multipart/mixed; boundary="------------3AE568BB1AA5E86E80A10618" This is a multi-part message in MIME format. --------------3AE568BB1AA5E86E80A10618 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hi everybody ! I trying to make some sort of PCI server... I applied patch to mmap.c and cons2 (from David's page) to make possible to flags=MAP_PHYS |MAP_FIXED and write little test code thet search for BIOS32 signature in memory... It leads to very strange error: (under GDB, when trying to call last printf()) "Hmmm, unexpected stop reason 0x4 Program received signal SIGTRAP Trace/breakpoint trap 0x6000295a in ??()" But all before that printf() work fine ... When it occured first time i change patch (offset now is "where-to-map" instead of "what-to-map"), rebuild libs, servers and ... all useless... Help me please (if it possible) ! My code: #include #include #include #include #define BIOS32_SIGNATURE (('_'<<0) + ('3'<<8) + ('2'<<16) + ('_'<<24)) void main() { unsigned long *base, *check; unsigned short a; size_t len = 0x100000-0xe0000; int num = len/sizeof(*base); base = mmap((void *)0xe0000, len, PROT_READ, MAP_PHYS|MAP_FIXED, 0, 0xe0000L); if (base == (void *)0 || base == (void *)-1) { perror("mmap failed"); } else { printf("Maping Ok! at %p\n", (void *)base); } for (check = base; check < base+num; ++check) { if (*check == BIOS32_SIGNATURE) printf("BIOS32 signature found at %p\n", check); } munmap((void *)base, len); asm("mov %%cs, %w0": "=r" (a)); printf("Current segmet is: %sx\n", a); } I attached last version of mmap.c... Yuriy. --------------3AE568BB1AA5E86E80A10618 Content-Type: text/plain; charset=us-ascii; name="mmap.c" Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="mmap.c" /* * mmap.c * Primary user interface to the VM system */ #include #include #include #include #include #include #include #include #include #include #include #include "../mach/mutex.h" #include "pset.h" /* Flag that we're not allowing a map hash to be created */ #define NO_MAP_HASH ((struct hash *)1) /* * mmap() * Map something into the address space * * Many combinations of options are not allowed. */ void * mmap(caddr_t addr, ulong len, int prot, int flags, port_t port, ulong offset) { struct proc *p = curthread->t_proc; struct vas *vas = &p->p_vas; struct pview *pv; struct pset *ps; void *vaddr; /* * Anonymous memory */ if (flags & MAP_ANON) { /* * Keep it simple, eh? Read-only ZFOD??? */ if ((flags & (MAP_FILE|MAP_FIXED|MAP_PHYS)) || !(prot & PROT_WRITE)) { err(EINVAL); return(0); } /* * Get view/set */ ps = alloc_pset_zfod(btorp(len)); pv = alloc_pview(ps); /* * Flag view as from mmap(). We use this to keep * munmap() honest; it only works on views which * were created by the user. */ pv->p_prot |= PROT_MMAP; /* * If shared, turn on bit */ if (flags & MAP_SHARED) { ps->p_flags |= PF_SHARED; } /* * Try to attach */ pv->p_vaddr = addr; vaddr = attach_pview(vas, pv); if (vaddr == 0) { free_pview(pv); } return(vaddr); } /* * Physical mapping */ if (flags & MAP_PHYS) { /* * Sanity */ if (flags & MAP_FILE) { err(EINVAL); return(0); } /* * Protection */ if (!issys()) { return(0); } /* * Get a physical pset, create a view, map it in. */ ps = physmem_pset(btop(addr), btorp(len)); pv = alloc_pview(ps); pv->p_prot = PROT_MMAP; if (flags & MAP_FIXED) pv->p_vaddr = (void *)offset; vaddr = attach_pview(vas, pv); if (vaddr == 0) { free_pview(pv); } return(vaddr); } /* * Create read-only or copy-on-write view of a file */ if (flags & MAP_FILE) { struct portref *pr; /* * If writing, make sure they're asking for COW; we * don't support writing back to a server via mmap(). * It's nasty to do. */ if ((prot & PROT_WRITE) && !(flags & MAP_PRIVATE)) { err(EINVAL); return(0); } /* * Get the portref. add_map() creates new portrefs * as necessary, so just use the open file. */ pr = find_portref(p, port); if (pr == 0) { return(0); } v_lock(&pr->p_lock, SPL0_SAME); /* * Try to generate a view. Return address or error. */ addr = add_map(vas, pr, addr, btorp(len + (offset - ptob(btop(offset)))), btop(offset), !(prot & PROT_WRITE)); v_sema(&pr->p_sema); if (addr) { return(addr); } /* VVV fall down to common error case VVV */ } err(EINVAL); return(0); } /* * munmap() * Unmap a region given an address within */ int munmap(caddr_t vaddr, ulong len) { struct vas *vas = &curthread->t_proc->p_vas; struct pview *pv; struct pset *ps; /* * Look for the view which matches */ pv = find_pview(vas, vaddr); if (!pv) { return(err(EINVAL)); } ps = pv->p_set; /* * Make sure it's an mmap()'ed view. Turn off bit so once we * release the lock and call detach_view(), we can't race with * another thread coming into munmap(). */ if ((pv->p_prot & PROT_MMAP) == 0) { v_lock(&ps->p_lock, SPL0_SAME); return(err(EBUSY)); } pv->p_prot &= ~(PROT_MMAP); v_lock(&ps->p_lock, SPL0_SAME); /* * Blow it away */ remove_pview(vas, vaddr); return(0); } /* * get_map_pset() * Return pset view of named file * * Servers must support the FS_FID function in order to allow * mapping of files. */ static struct pset * get_map_pset(struct portref *pr) { struct pset *ps; long args[3]; struct port *port; /* * Hold mutex so we're the only one searching/updating the * cache of mappings. This also keeps the port from shutting * on us. */ p_lock_void(&pr->p_lock, SPL0); port = pr->p_port; if (port) { if ((port->p_flags & P_CLOSING) || (port->p_maps == NO_MAP_HASH)) { v_lock(&pr->p_lock, SPL0_SAME); port = 0; } else { p_sema_v_lock(&port->p_mapsema, PRIHI, &pr->p_lock); } } /* * Try to get file ID. This also gets us the file's * size. */ if (!port) { return(0); } if (kernmsg_send(pr, FS_FID, args)) { v_sema(&port->p_mapsema); return(0); } /* * If there's a hash already, search it for our ID. Otherwise * create a new hash and flag that we didn't find the pset. */ if (port->p_maps) { ps = hash_lookup(port->p_maps, args[0]); } else { /* * New hash table. Is 32 a good number? Who knows? */ port->p_maps = hash_alloc(32); ps = 0; } /* * If the file's changed size, invalidate the old executable * cache so we can set up the new one. */ if (ps && (ps->p_len != args[1])) { (void)hash_delete(port->p_maps, args[0]); deref_pset(ps); ps = 0; } /* * If no pset, create one and insert it. The mapping in * the hash counts as a reference. */ if (ps == 0) { struct portref *newpr; newpr = dup_port(pr); if (newpr == 0) { v_sema(&port->p_mapsema); return(0); } ps = alloc_pset_fod(newpr, args[1]); (void)hash_insert(port->p_maps, args[0], ps); ref_pset(ps); } /* * Leave an extra ref on the pset so it won't go away * while we add views to it. Our caller will release * this ref once he's done his own refs. */ ref_pset(ps); /* * Release sema and return pset */ v_sema(&port->p_mapsema); return(ps); } /* * add_map() * Add a mmap view of the given file * * Returns 0 on failure, attach address on success */ void * add_map(struct vas *vas, struct portref *pr, caddr_t vaddr, ulong len, ulong off, int readonly) { struct pset *ps; struct pview *pv; int err; /* * Get underlying pset; might be cached */ ps = get_map_pset(pr); if (ps == 0) { return(0); } /* * Create a pview of the given size */ if (readonly) { /* * Read-only is a simple view into the pset */ pv = alloc_pview(ps); pv->p_prot = PROT_RO; if (pv->p_len > off) { pv->p_off = off; pv->p_len -= off; } } else { struct pset *ps2; /* * Read/write means the view is copy-on-write; * insert a COW pset on top of the straight * view. */ ps2 = alloc_pset_cow(ps, off, len); pv = alloc_pview(ps2); pv->p_prot = 0; pv->p_off = 0; } if (len < pv->p_len) pv->p_len = len; pv->p_vaddr = vaddr; /* * Try to attach. Throw away if it's rejected. Most * likely this is a bogus filemap, with an address * which the HAT rejected. */ err = (attach_pview(vas, pv) == 0); if (err) { free_pview(pv); } /* * Drop "placeholder" ref to pset (see get_map_pset()) */ deref_pset(ps); return(err ? 0 : pv->p_vaddr); } /* * do_derefport() * Release a cached access to a port */ static int do_derefport(long keydummy, struct pset *ps, void *dummy) { deref_pset(ps); return(0); } /* * mmap_cleanup() * Clean up mapped file cache, if any */ void mmap_cleanup(struct port *port) { /* * Take map mutex */ p_sema(&port->p_mapsema, PRIHI); /* * No cache--nothing to do */ if (port->p_maps == 0) { v_sema(&port->p_mapsema); return; } /* * Remove reference from pset for each entry * in our cache. */ hash_foreach(port->p_maps, do_derefport, 0); /* * Free the hash storage, flag port as shutting down */ hash_dealloc(port->p_maps); port->p_maps = NO_MAP_HASH; v_sema(&port->p_mapsema); } /* * unhash() * Unhash any reference to the indicated hash # */ int unhash(port_t arg_port, long arg_fid) { struct port *port; struct proc *p = curthread->t_proc; /* * Verify range */ if (arg_port < PROCOPENS) { return(err(EBADF)); } arg_port -= PROCOPENS; if (arg_port >= PROCPORTS) { return(err(EBADF)); } /* * Interlock, get access to port, atomically switch * to mapsema. */ if (p_sema(&p->p_sema, PRICATCH)) { return(err(EINTR)); } port = p->p_ports[arg_port]; if (port == 0) { v_sema(&p->p_sema); return(err(EBADF)); } p_lock_void(&port->p_lock, SPLHI); v_sema(&p->p_sema); p_sema_v_lock(&port->p_mapsema, PRIHI, &port->p_lock); /* * If there's a map, delete entry from it */ if (port->p_maps && (port->p_maps != NO_MAP_HASH)) { struct pset *ps; ps = hash_lookup(port->p_maps, arg_fid); if (ps) { (void)hash_delete(port->p_maps, arg_fid); deref_pset(ps); } } /* * Release mutex, return success */ v_sema(&port->p_mapsema); return(0); } --------------3AE568BB1AA5E86E80A10618-- From daemon Sat May 30 04:26:12 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id EAA13960 for vsta-xplod; Sat, 30 May 1998 04:26:12 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id EAA13954 for ; Sat, 30 May 1998 04:26:10 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by zipper.cisco.com (8.8.5-Cisco.2-SunOS.5.5.1.sun4/8.6.5) with ESMTP id IAA15084; Sat, 30 May 1998 08:12:29 -0700 (PDT) Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id IAA15310; Sat, 30 May 1998 08:12:29 -0700 (PDT) Message-Id: <199805301512.IAA15310@vandys-pc.cisco.com> To: Yoda cc: vsta@zendo.com Subject: Re: Memory mapping... Help ! In-reply-to: Your message of "Sat, 30 May 1998 17:11:47 +0400." <35700592.719807B8@dialup.ptt.ru> Date: Sat, 30 May 1998 08:12:29 -0700 From: Andy Valencia [Yoda writes:] >I trying to make some sort of PCI server... A worthy goal! >It leads to very strange error: (under GDB, when trying to call last >printf()) "Hmmm, unexpected stop reason 0x4 Program received signal >SIGTRAP Trace/breakpoint trap 0x6000295a in ??()" But all before that >printf() work fine ... I think all that's wrong here is the debugger's comment about why the program stopped. > unsigned short a; >... > printf("Current segmet is: %sx\n", a); You've asked the printf code to use a short number (promoted to int, no doubt) as a pointer to chars to display in the "%s" format. Andy From daemon Fri Jun 19 06:12:45 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id GAA03139 for vsta-xplod; Fri, 19 Jun 1998 06:12:45 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id GAA03133 for ; Fri, 19 Jun 1998 06:12:43 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by zipper.cisco.com (8.8.5-Cisco.2-SunOS.5.5.1.sun4/8.6.5) with ESMTP id KAA05564; Fri, 19 Jun 1998 10:00:00 -0700 (PDT) Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id JAA09312; Fri, 19 Jun 1998 09:59:59 -0700 (PDT) Message-Id: <199806191659.JAA09312@vandys-pc.cisco.com> To: Ondrej Martinek cc: vsta@zendo.com Subject: Re: console problem In-reply-to: Your message of "Thu, 18 Jun 1998 13:15:54 +0200." Date: Fri, 19 Jun 1998 09:59:59 -0700 From: Andy Valencia [Ondrej Martinek writes:] >hi all, > >i'm intertested in OS architectures and i like some nice features of >Vsta. > >i have installed Vsta on poor 486sx 25mhz 4mb with dos on the same >partition. unfortunately, keypad control arrows don't work with both >"cons2" and "cons3". please don't laught! ;) it's very >unconvenient for me and blocks me to work more with Vsta. I believe cons2 generates ANSI standard output from the keypad. Do you mean the arrows don't work from the shell? You'd need to teach the getline() routine how to interpret arrow keys. I'm not sure it'd be worth it. >Also pressing >backspace ,when the input line is empty, displays its ascii char. >what's the problem? can you help me? Again, you mean while typing to the shell? Note that "del" isn't bound to the "backspace" function; you need to use "backspace". >moreover, executing "mgr" frooze the system imediately. Two things: did you start a mouse driver? And also you might well run out of memory. I'd guess 4 megs can boot VSTa and run MGR, but don't be surprised if a gcc within a window runs you out of RAM. Andy From daemon Sun Jun 21 02:24:39 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id CAA09612 for vsta-xplod; Sun, 21 Jun 1998 02:24:39 GMT Received: from atrey.karlin.mff.cuni.cz (atrey.karlin.mff.cuni.cz [195.113.31.123]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id CAA09606 for ; Sun, 21 Jun 1998 02:24:31 GMT Received: from Elf.mj.gts.cz (root@slip9.ms.mff.cuni.cz [195.113.20.209]) by atrey.karlin.mff.cuni.cz (8.8.8/8.8.8) with ESMTP id PAA06510; Sun, 21 Jun 1998 15:12:08 +0200 Received: (from pavel@localhost) by Elf.mj.gts.cz (8.8.5/8.8.5) id MAA00435; Sat, 20 Jun 1998 12:11:04 +0200 Message-ID: <19980620120943.05279@Elf.ucw.cz> Date: Sat, 20 Jun 1998 12:09:43 +0200 From: Pavel Machek To: Andy Valencia Cc: Ondrej Martinek , vsta@zendo.com Subject: Re: console problem References: <199806191659.JAA09312@vandys-pc.cisco.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.81 In-Reply-To: <199806191659.JAA09312@vandys-pc.cisco.com>; from Andy Valencia on Fri, Jun 19, 1998 at 09:59:59AM -0700 X-Warning: Not only using Windows can be dangerous to your mental health. Hi! > [Ondrej Martinek writes:] Hmm, greetings to Mala Strana. > >i'm intertested in OS architectures and i like some nice features of > >Vsta. > > > >i have installed Vsta on poor 486sx 25mhz 4mb with dos on the same > >partition. unfortunately, keypad control arrows don't work with both > >"cons2" and "cons3". please don't laught! ;) it's very > >unconvenient for me and blocks me to work more with Vsta. I solved this by making consoles send ^P, ^N, ^F, ^B instead of standart ANSI bindings. Very convient. Pavel -- I'm really pavel@atrey.karlin.mff.cuni.cz. Pavel Look at http://atrey.karlin.mff.cuni.cz/~pavel/ ;-). From daemon Sun Jun 21 22:06:59 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id WAA11370 for vsta-xplod; Sun, 21 Jun 1998 22:06:59 GMT Received: from atlas.cs.york.ac.uk (atlas.cs.york.ac.uk [144.32.176.2]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id WAA11364 for ; Sun, 21 Jun 1998 22:06:52 GMT From: "Robin J Bradley (3Y0)" Message-Id: <9806220947.ZM18931@ugrad.cs.york.ac.uk> Date: Mon, 22 Jun 1998 09:47:45 +0100 X-Mailer: Z-Mail (3.2.1 6apr95 MediaMail) To: vsta@zendo.com Subject: Re: console problem Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii >i have installed Vsta on poor 486sx 25mhz 4mb with dos on the same >partition. unfortunately, keypad control arrows don't work with both >"cons2" and "cons3". please don't laught! ;) it's very >unconvenient for me and blocks me to work more with Vsta. I belive hte main problem is with the curses library which last time I looked did not support handling ansi escape sequances. It looks like the old BSD curses library not the newer ncurses style which will convert escape codes into something meaningfull. Perhaps we should update the curses library ? Robin From daemon Mon Jun 22 04:16:18 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id EAA12078 for vsta-xplod; Mon, 22 Jun 1998 04:16:18 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id EAA12072 for ; Mon, 22 Jun 1998 04:16:16 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by zipper.cisco.com (8.8.5-Cisco.2-SunOS.5.5.1.sun4/8.6.5) with ESMTP id IAA28704; Mon, 22 Jun 1998 08:03:42 -0700 (PDT) Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id IAA17539; Mon, 22 Jun 1998 08:03:42 -0700 (PDT) Message-Id: <199806221503.IAA17539@vandys-pc.cisco.com> To: "Robin J Bradley (3Y0)" cc: vsta@zendo.com Subject: Re: console problem In-reply-to: Your message of "Mon, 22 Jun 1998 09:47:45 BST." <9806220947.ZM18931@ugrad.cs.york.ac.uk> Date: Mon, 22 Jun 1998 08:03:42 -0700 From: Andy Valencia ["Robin J Bradley (3Y0)" writes:] >I belive hte main problem is with the curses library which last time I looked >did not support handling ansi escape sequances. It looks like the old BSD >curses library not the newer ncurses style which will convert escape codes >into something meaningfull. Yes, that'd be whatever was easy to grab ~6 years ago. >Perhaps we should update the curses library ? Sounds good to me! I guess I could go try and port the FreeBSD version. Andy From daemon Thu Jun 25 11:33:37 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id LAA20802 for vsta-xplod; Thu, 25 Jun 1998 11:33:37 GMT Received: from mumble.frotz.net (qmailr@mumble.frotz.net [198.88.183.235]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id LAA20796 for ; Thu, 25 Jun 1998 11:33:34 GMT Received: (qmail 27744 invoked by uid 500); 25 Jun 1998 22:24:55 -0000 Message-ID: <19980625172455.58631@frotz.net> Date: Thu, 25 Jun 1998 17:24:55 -0500 From: Brian J Swetland To: vsta@zendo.com Subject: SCSI / PCI Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.88.14 Organization: Frotz Communications, Ltd. X-Mailer-Holy-War: Get Mutt, it bites! Is PCI support in the works at all? I was about to take a crack at porting my BeOS BusLogic BT-9x8 driver (BeOS also uses CAM, making it a pretty simple task), and realized that I need to be able to detect and potentially configure the PCI card. Brian -- Brian J. Swetland - swetland@frotz.net | "If a tree falls in the forest, and http://www.frotz.net/swetland/ | we've already sold the tree, does it | have Quality?" -- The Boss, Dilbert From daemon Thu Jun 25 11:46:55 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id LAA20871 for vsta-xplod; Thu, 25 Jun 1998 11:46:55 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id LAA20865 for ; Thu, 25 Jun 1998 11:46:52 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by zipper.cisco.com (8.8.5-Cisco.2-SunOS.5.5.1.sun4/8.6.5) with ESMTP id PAA03413; Thu, 25 Jun 1998 15:34:28 -0700 (PDT) Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id PAA28953; Thu, 25 Jun 1998 15:34:27 -0700 (PDT) Message-Id: <199806252234.PAA28953@vandys-pc.cisco.com> To: Brian J Swetland cc: vsta@zendo.com Subject: Re: SCSI / PCI In-reply-to: Your message of "Thu, 25 Jun 1998 17:24:55 CDT." <19980625172455.58631@frotz.net> Date: Thu, 25 Jun 1998 15:34:27 -0700 From: Andy Valencia [Brian J Swetland writes:] >Is PCI support in the works at all? I was about to take a crack at >porting my BeOS BusLogic BT-9x8 driver (BeOS also uses CAM, making it >a pretty simple task), and realized that I need to be able to detect >and potentially configure the PCI card. There was some PCI work quite a while back, which got far enough to map in the card and run some of its code. I recall that there was a real problem that the only rules for what BIOS-type code was allowed to do was basically "if it works for Windoze...". The most interesting work was probably: http://www.zendo.com/vsta/mail/10/0009.html Although PCI pops up here and there throughout the mail archives. Andy From daemon Fri Jun 26 08:12:53 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id IAA22777 for vsta-xplod; Fri, 26 Jun 1998 08:12:53 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id IAA22771 for ; Fri, 26 Jun 1998 08:12:46 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by zipper.cisco.com (8.8.5-Cisco.2-SunOS.5.5.1.sun4/8.6.5) with ESMTP id MAA18373; Fri, 26 Jun 1998 12:00:20 -0700 (PDT) Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id MAA00558; Fri, 26 Jun 1998 12:00:20 -0700 (PDT) Message-Id: <199806261900.MAA00558@vandys-pc.cisco.com> To: Jeremy Fitzhardinge cc: vsta@zendo.com, Brian J Swetland Subject: Re: SCSI / PCI In-reply-to: Your message of "Thu, 25 Jun 1998 23:33:21 PDT." Date: Fri, 26 Jun 1998 12:00:20 -0700 From: Andy Valencia [Jeremy Fitzhardinge writes:] >> There was some PCI work quite a while back, which got far enough to map in >> the card and run some of its code. I recall that there was a real problem >> that the only rules for what BIOS-type code was allowed to do was basically >> "if it works for Windoze...". >Yeah, I was forced to stop working on that approach because it broke completel >y >when I changed motherboards/bioses. The BIOS32 mechanism which is supposed to >allow a 32-bit kernel make use of BIOS code is flawed, because there's no way >of telling what memory it wants access to, so you can't do the appropriate >mmaps. Is there any possibility we could install a SEGV handler and manipulate the environment on the fly? >There's also the secondary problem that you can't identity map all >parts of the physical address space. I also needed to change the mmap() >syscall itself to allow the use of MAP_FIXED and MAP_PHYS together, but I >haven't looked at recent kernels to see if this is still the case. Well, changing the kernel to allow this (especially for a privileged process) would be the least of the problem. >I suspect that using direct IO to the PCI chipset is more likely to work in th >e >VSTa environment. If there's a modest set of chip types to support, I'm sure this is workable. It seems like the PC video arena is an example where there's just barely a contained enough set for this approach to work. Andy From daemon Fri Jun 26 17:18:45 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id RAA23839 for vsta-xplod; Fri, 26 Jun 1998 17:18:45 GMT Received: (from root@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id RAA23832; Fri, 26 Jun 1998 17:18:42 GMT Message-ID: X-Mailer: XFMail 1.3 [p0] on Linux X-Priority: 3 (Normal) Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 8bit MIME-Version: 1.0 In-Reply-To: <199806252234.PAA28953@vandys-pc.cisco.com> Date: Thu, 25 Jun 1998 23:33:21 -0700 (PDT) From: Jeremy Fitzhardinge To: vsta@zendo.com Subject: Re: SCSI / PCI On 25-Jun-98 Andy Valencia wrote: > [Brian J Swetland writes:] > >>Is PCI support in the works at all? I was about to take a crack at >>porting my BeOS BusLogic BT-9x8 driver (BeOS also uses CAM, making it >>a pretty simple task), and realized that I need to be able to detect >>and potentially configure the PCI card. > > There was some PCI work quite a while back, which got far enough to map in > the card and run some of its code. I recall that there was a real problem > that the only rules for what BIOS-type code was allowed to do was basically > "if it works for Windoze...". The most interesting work was probably: > > http://www.zendo.com/vsta/mail/10/0009.html > > Although PCI pops up here and there throughout the mail archives. Yeah, I was forced to stop working on that approach because it broke completely when I changed motherboards/bioses. The BIOS32 mechanism which is supposed to allow a 32-bit kernel make use of BIOS code is flawed, because there's no way of telling what memory it wants access to, so you can't do the appropriate mmaps. There's also the secondary problem that you can't identity map all parts of the physical address space. I also needed to change the mmap() syscall itself to allow the use of MAP_FIXED and MAP_PHYS together, but I haven't looked at recent kernels to see if this is still the case. I suspect that using direct IO to the PCI chipset is more likely to work in the VSTa environment. J From daemon Fri Jun 26 18:21:31 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id SAA23961 for vsta-xplod; Fri, 26 Jun 1998 18:21:31 GMT Received: from home.chat.net (genesis.jeske.meer.net [204.94.139.85]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id SAA23955 for ; Fri, 26 Jun 1998 18:21:29 GMT Received: (from jeske@localhost) by home.chat.net (8.8.8/8.8.8) id WAA27148 for vsta@zendo.com; Fri, 26 Jun 1998 22:12:48 -0700 (PDT) Message-ID: <19980626221247.E26766@home.chat.net> Date: Fri, 26 Jun 1998 22:12:47 -0700 From: David Jeske To: vsta@zendo.com Subject: Re: SCSI / PCI Mail-Followup-To: vsta@zendo.com References: <199806252234.PAA28953@vandys-pc.cisco.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.92.1 In-Reply-To: ; from Jeremy Fitzhardinge on Thu, Jun 25, 1998 at 11:33:21PM -0700 On Thu, Jun 25, 1998 at 11:33:21PM -0700, Jeremy Fitzhardinge wrote: > Yeah, I was forced to stop working on that approach because it broke completely > when I changed motherboards/bioses. The BIOS32 mechanism which is supposed to > allow a 32-bit kernel make use of BIOS code is flawed, because there's no way > of telling what memory it wants access to, so you can't do the appropriate > mmaps. There's also the secondary problem that you can't identity map all > parts of the physical address space. I also needed to change the mmap() > syscall itself to allow the use of MAP_FIXED and MAP_PHYS together, but I > haven't looked at recent kernels to see if this is still the case. > > I suspect that using direct IO to the PCI chipset is more likely to work in the > VSTa environment. What current systems actually use BIOS32 to do this anyhow? I am under the impression that this is usually done by accessing the chipset directly always. -- David Jeske (N9LCA) + http://www.chat.net/~jeske/ + jeske@chat.net From daemon Fri Jun 26 19:05:46 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id TAA24008 for vsta-xplod; Fri, 26 Jun 1998 19:05:46 GMT Received: from mail1.sirius.com (mail1.sirius.com [205.134.253.131]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id TAA24002 for ; Fri, 26 Jun 1998 19:05:44 GMT Received: from ixodes.ssa.au (ppp-asfm04--207.sirius.net [205.134.240.207]) by mail1.sirius.com (8.8.7/Sirius-8.8.7-97.08.12) with ESMTP id WAA29663; Fri, 26 Jun 1998 22:53:59 -0700 (PDT) Message-ID: X-Mailer: XFMail 1.3 [p0] on Linux X-Priority: 3 (Normal) Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 8bit MIME-Version: 1.0 In-Reply-To: <199806261900.MAA00558@vandys-pc.cisco.com> Date: Fri, 26 Jun 1998 22:47:06 -0700 (PDT) From: Jeremy Fitzhardinge To: Andy Valencia Subject: Re: SCSI / PCI Cc: Brian J Swetland , vsta@zendo.com On 26-Jun-98 Andy Valencia wrote: > Is there any possibility we could install a SEGV handler and manipulate the > environment on the fly? Potentially. Seems very cumbersome though... >>I suspect that using direct IO to the PCI chipset is more likely to work in >>th >>e >>VSTa environment. > > If there's a modest set of chip types to support, I'm sure this is workable. > It seems like the PC video arena is an example where there's just barely a > contained enough set for this approach to work. While in principle cards can supply their own drivers via BIOS32, I don't think this is used very much. The only exception I know of is putting a CAM layer into SCSI controllers, but I think most people avoid it and go straight to the hardware. BIOS32's main purpose is to simply provide the OS with the PCI bus configuration. Linux used to use it exclusively, but it now seems to go straight to the hardware for the more common Intel and other PCI controllers. I think this allows you to tweak performance settings (burst modes, auto-retry, timing, etc). Since there don't seem to be very many incompatible PCI controllers around, this is probably a reasonable approach. J From daemon Sat Jun 27 03:04:12 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id DAA24700 for vsta-xplod; Sat, 27 Jun 1998 03:04:12 GMT Received: from wrath.cs.utah.edu (wrath.cs.utah.edu [155.99.198.100]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id DAA24694 for ; Sat, 27 Jun 1998 03:04:08 GMT Received: from fast.cs.utah.edu (fast.cs.utah.edu [155.99.212.1]) by wrath.cs.utah.edu (8.8.8/8.8.8) with SMTP id HAA03115; Sat, 27 Jun 1998 07:52:19 -0600 (MDT) Received: by fast.cs.utah.edu (8.6.10/utah-2.15-leaf) id HAA05961; Sat, 27 Jun 1998 07:52:18 -0600 Date: Sat, 27 Jun 1998 07:52:18 -0600 From: vanmaren@fast.cs.utah.edu (Kevin Van Maren) Message-Id: <199806271352.HAA05961@fast.cs.utah.edu> To: jeske@home.chat.net, vsta@zendo.com Subject: Re: SCSI / PCI > On Thu, Jun 25, 1998 at 11:33:21PM -0700, Jeremy Fitzhardinge wrote: > > Yeah, I was forced to stop working on that approach because it broke completely > > when I changed motherboards/bioses. The BIOS32 mechanism which is supposed to > > allow a 32-bit kernel make use of BIOS code is flawed, because there's no way > > of telling what memory it wants access to, so you can't do the appropriate > > mmaps. There's also the secondary problem that you can't identity map all > > parts of the physical address space. I also needed to change the mmap() > > syscall itself to allow the use of MAP_FIXED and MAP_PHYS together, but I > > haven't looked at recent kernels to see if this is still the case. > > I suspect that using direct IO to the PCI chipset is more likely to work in the > > VSTa environment. > > What current systems actually use BIOS32 to do this anyhow? I am under the > impression that this is usually done by accessing the chipset directly > always. Linux does :-) FreeBSD and NetBSD program the chipset directly. I have used the BIOS32 interface in usermode before. I can check the BIOS32 specification later, but I'm fairly certain that you only have to map the 128k BIOS region, and provide a writable stack. There are two `standard' chipset interfaces, and of course, non of this is portable to other PCI platforms (Alpha, Sparc, etc), as this section of the PCI spec is the "x86-specific" part. Also, the BIOS does not have to be mapped at virt--phys; you can map the 128k region anywhere. [But I won't claim that there are no broken BIOSes out there.] Kevin From daemon Sat Jun 27 04:56:57 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id EAA27009 for vsta-xplod; Sat, 27 Jun 1998 04:56:57 GMT Received: from wrath.cs.utah.edu (wrath.cs.utah.edu [155.99.198.100]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id EAA27003 for ; Sat, 27 Jun 1998 04:56:46 GMT Received: from fast.cs.utah.edu (fast.cs.utah.edu [155.99.212.1]) by wrath.cs.utah.edu (8.8.8/8.8.8) with SMTP id JAA03831; Sat, 27 Jun 1998 09:44:53 -0600 (MDT) Received: by fast.cs.utah.edu (8.6.10/utah-2.15-leaf) id JAA06768; Sat, 27 Jun 1998 09:44:52 -0600 Date: Sat, 27 Jun 1998 09:44:52 -0600 From: vanmaren@fast.cs.utah.edu (Kevin Van Maren) Message-Id: <199806271544.JAA06768@fast.cs.utah.edu> To: jeremy@zip.com.au, vsta@zendo.com Subject: Re: SCSI / PCI I was mostly right, but the details are slightly different from what I said. From the PCI BIOS Specification, version 2.1: "The PCI BIOS routines (for both 16-bit and 32-bit callers) must be invoked with appropriate privlidge so that interrupts can be enabled/ disabled and the routines can access IO space. Implementors of the PCI BIOS must assume that CS is execute-only and DS is read-only" "The 32-bit PCI BIOS functions must be accessed using CALL FAR. The CS and DS register descriptors must be setup to encompass the physical addresses specified by the Base and Length parameters returned by the BIOS32 Service Directory. The CS and DS descriptors must have the same base. The calling environment must allow access to IO space and provide at least 1K of stack space." I didn't have any problems using BIOS32 in user-mode, at least on the machines I tried. It was intentionally made easy to use. Kevin T. Van Maren ps: You get the BIOS32 Service Directory by searching for it in the range 0xE0000 to 0xFFFFF (physical). "$PCI" is the PCI BIOS interface. From daemon Sat Jun 27 06:20:17 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id GAA27177 for vsta-xplod; Sat, 27 Jun 1998 06:20:17 GMT Received: from home.chat.net (genesis.jeske.meer.net [204.94.139.85]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id GAA27171 for ; Sat, 27 Jun 1998 06:20:14 GMT Received: (from jeske@localhost) by home.chat.net (8.8.8/8.8.8) id KAA27818 for vsta@zendo.com; Sat, 27 Jun 1998 10:11:36 -0700 (PDT) Message-ID: <19980627101135.F26766@home.chat.net> Date: Sat, 27 Jun 1998 10:11:35 -0700 From: David Jeske To: vsta@zendo.com Subject: Re: SCSI / PCI Mail-Followup-To: vsta@zendo.com References: <199806271544.JAA06768@fast.cs.utah.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.92.1 In-Reply-To: <199806271544.JAA06768@fast.cs.utah.edu>; from Kevin Van Maren on Sat, Jun 27, 1998 at 09:44:52AM -0600 On Sat, Jun 27, 1998 at 09:44:52AM -0600, Kevin Van Maren wrote: > "The 32-bit PCI BIOS functions must be accessed using CALL FAR. The > CS and DS register descriptors must be setup to encompass the physical > addresses specified by the Base and Length parameters returned by the > BIOS32 Service Directory. The CS and DS descriptors must have the same > base. The calling environment must allow access to IO space and provide > at least 1K of stack space." The biggest problem with this is exactly as Jeremy said. There is not a convinent way to open 'all of IO space' and I'm not sure you would even want to do this. I think the best solution here would be a signal handler which would allow the program to handle unallowed accesses to IO or MEMORY space. This would not only allow this to work (i.e. the signal handler would open up the IO register when the BIOS32 code tried to access it, and then it would let it continue to run) but this would also allow in program page fault handlers and such later. OS/2 actually uses in program page fault handlers to handle all your memory fault needs. They use an in program fault handler to handle growing the stack. They used a system called 'guard pages' where you would setup a guard page, and if someone hit a guard page, then your handler was called. -- David Jeske (N9LCA) + http://www.chat.net/~jeske/ + jeske@chat.net From daemon Sat Jun 27 15:49:50 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id PAA27922 for vsta-xplod; Sat, 27 Jun 1998 15:49:50 GMT Received: from wrath.cs.utah.edu (wrath.cs.utah.edu [155.99.198.100]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id PAA27916 for ; Sat, 27 Jun 1998 15:49:46 GMT Received: from fast.cs.utah.edu (fast.cs.utah.edu [155.99.212.1]) by wrath.cs.utah.edu (8.8.8/8.8.8) with SMTP id UAA08091; Sat, 27 Jun 1998 20:37:58 -0600 (MDT) Received: by fast.cs.utah.edu (8.6.10/utah-2.15-leaf) id UAA12537; Sat, 27 Jun 1998 20:37:57 -0600 Date: Sat, 27 Jun 1998 20:37:57 -0600 From: vanmaren@fast.cs.utah.edu (Kevin Van Maren) Message-Id: <199806280237.UAA12537@fast.cs.utah.edu> To: jeske@home.chat.net, vsta@zendo.com Subject: Re: SCSI / PCI > The biggest problem with this is exactly as Jeremy said. There is not a > convinent way to open 'all of IO space' and I'm not sure you would even > want to do this. I think the best solution here would be a signal handler > which would allow the program to handle unallowed accesses to IO or MEMORY > space. This would not only allow this to work (i.e. the signal handler > would open up the IO register when the BIOS32 code tried to access it, and > then it would let it continue to run) but this would also allow in program > page fault handlers and such later. I ran with IOPL=3 for a while. When I went to using per-process IO bitmaps, I changed the PCI configuration space accesses to use an IPC (essentially to a PCI config space arbiter, which could enforce access controls). The IPC was not a performance problem, as PCI configuration registers are only read during the driver initialization phase. Allowing access to the low-level registers or BIOS32 is problematic as both are (by their nature) non-reentrant, so only one process can use them at a time (technically that is only true for one of the config mechanisms). There is also the security issue: if a process can "disable" arbitrary PCI devices, you might as well just give it full IO permissions. Having a trusted process handle all accesses to PCI configuration space isn't so bad (IMHO). BTW, the PCI BIOS32 doesn't need all of IO space, but it does need IOPL3 to disable interrupts, so the issue is moot. Anyway, this has gone on long enough. If you want code to access PCI config space directly, look at the FreeBSD code; if you want to try BIOS32, I was able to adapt the Linux 2.0.x code by mapping 128k (and not worrying about the BASE/BOUNDS) of BIOS and adjusting the search to use the virtual instead of physical addresses. (I now use the FreeBSD code to avoid dealing with it.) Kevin From daemon Sat Jun 27 13:21:55 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id NAA27651 for vsta-xplod; Sat, 27 Jun 1998 13:21:55 GMT Received: from mail1.sirius.com (mail1.sirius.com [205.134.253.131]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id NAA27645 for ; Sat, 27 Jun 1998 13:21:52 GMT Received: from ixodes.ssa.au (ppp-asfm06--076.sirius.net [205.134.241.76]) by mail1.sirius.com (8.8.7/Sirius-8.8.7-97.08.12) with ESMTP id RAA01182; Sat, 27 Jun 1998 17:10:07 -0700 (PDT) Message-ID: X-Mailer: XFMail 1.3 [p0] on Linux X-Priority: 3 (Normal) Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 8bit MIME-Version: 1.0 In-Reply-To: <199806271544.JAA06768@fast.cs.utah.edu> Date: Sat, 27 Jun 1998 17:03:06 -0700 (PDT) From: Jeremy Fitzhardinge To: (Kevin Van Maren) Subject: Re: SCSI / PCI Cc: vsta@zendo.com On 27-Jun-98 Kevin Van Maren wrote: > I didn't have any problems using BIOS32 in user-mode, at least on the > machines I tried. It was intentionally made easy to use. > > Kevin T. Van Maren > > ps: You get the BIOS32 Service Directory by searching for it in > the range 0xE0000 to 0xFFFFF (physical). "$PCI" is the PCI BIOS interface. Its been a while, but I didn't remember it saying that you need to map at least X k - if its 128k, then that's fine. And I have to admit I just assumed that you'd need to identity map things - I didn't try it otherwise. Perhaps it is less moribund than I thought. J From daemon Mon Jun 29 04:27:15 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id EAA01251 for vsta-xplod; Mon, 29 Jun 1998 04:27:15 GMT Received: from ss1000.ms.mff.cuni.cz (root@ss1000.ms.mff.cuni.cz [195.113.19.221]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id EAA01245 for ; Mon, 29 Jun 1998 04:27:08 GMT Received: from ulab-7.ms.mff.cuni.cz (omar4275@ulab-7.ms.mff.cuni.cz [195.113.16.37]) by ss1000.ms.mff.cuni.cz (8.8.8/8.8.8) with ESMTP id RAA06856; Mon, 29 Jun 1998 17:15:11 +0200 (MET DST) Received: from localhost (omar4275@localhost) by ulab-7.ms.mff.cuni.cz (8.8.8/8.8.8) with SMTP id RAA15823; Mon, 29 Jun 1998 17:15:10 +0200 (MET DST) X-Authentication-Warning: ulab-7.ms.mff.cuni.cz: omar4275 owned process doing -bs Date: Mon, 29 Jun 1998 17:15:06 +0200 (MET DST) From: Ondrej Martinek X-Sender: omar4275@ulab-7.ms.mff.cuni.cz To: Andy Valencia cc: vsta@zendo.com Subject: Re: console problem In-Reply-To: <199806191659.JAA09312@vandys-pc.cisco.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Fri, 19 Jun 1998, Andy Valencia wrote: Hello, > [Ondrej Martinek writes:] > > >hi all, > > > >i'm intertested in OS architectures and i like some nice features of > >Vsta. > > > >i have installed Vsta on poor 486sx 25mhz 4mb with dos on the same > >partition. unfortunately, keypad control arrows don't work with both > >"cons2" and "cons3". please don't laught! ;) it's very > >unconvenient for me and blocks me to work more with Vsta. > > I believe cons2 generates ANSI standard output from the keypad. Do you mean > the arrows don't work from the shell? You'd need to teach the getline() > routine how to interpret arrow keys. I'm not sure it'd be worth it. > > >Also pressing > >backspace ,when the input line is empty, displays its ascii char. > >what's the problem? can you help me? > > Again, you mean while typing to the shell? Note that "del" isn't bound to > the "backspace" function; you need to use "backspace". I have already fixed it and patched cons3. The actual problem is getline() which doesn't use TERMCAP capabilities. I changed cons3/isr.c:cursor_key() to enqueue different sequences (emacs like ones) eg.: left arrow - ^B, PgDn - ^V, Del - ^D, Home - ^A, etc... Also added '\a' (BELL) handling in cons3. The result works great but cons3 isn't real VT100 now which can raise difficulities in future. I guess the best way would be in implementing better getline(). > > >moreover, executing "mgr" frooze the system imediately. > > Two things: did you start a mouse driver? And also you might well run out > of memory. I'd guess 4 megs can boot VSTa and run MGR, but don't be > surprised if a gcc within a window runs you out of RAM. I succeded to start MGR but it crashed the system after several seconds. (btw, the mouse moved very slowly) > > Andy > Moreover, I have other questions: 1. I want to learn cons3 to send signal (aka SIGKILL) on ^C. Is there any way how a server can find out the PID of it's client? Is it in msg.m_sender field of struct msg? 2. Vsta sometimes stops responding me. I find out (using kernel debbuger) that one thread is in page_out(). Is it necessary to have swap installed? Can I do it the following way for instance? mkfs_bfs /bfs 100 bfs /bfs fs/bfs swap swapd fs/bfs 3. During porting some shell scripts I find strange behaviour of "sh": -The shell started from another shell by commnad "sh" never inherits the environment but running a script is ok. eg: $export x=1; sh -c "echo x=\$x" x= $echo "echo x=\$x" > t x= $t x=1 -The script with #!/vsta/bin/sh in the first line is marked as "-sh" in the list of processes. -Strange reaction to commands with nonzero retcode. eg: suppose this script: test 1 = 0 echo done. when executed it prints: perm done. 4. I have root privileges but still can create file and change its access flags so that I can't access anymore. Is it alright? (btw, on DOS filesystem) 5. How can I link dynamic and static libraries? What is default? Thanks for answering my last questions. Can I send the next ones to vsta@cisco.com too? Is anyone of you, Vsta gurus, using IRC (internet relay chat) to talk interactivelly? Regards --ondra From daemon Mon Jun 29 06:03:53 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id GAA01417 for vsta-xplod; Mon, 29 Jun 1998 06:03:53 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id GAA01411 for ; Mon, 29 Jun 1998 06:03:51 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by zipper.cisco.com (8.8.5-Cisco.2-SunOS.5.5.1.sun4/8.6.5) with ESMTP id JAA17135; Mon, 29 Jun 1998 09:51:37 -0700 (PDT) Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id JAA08089; Mon, 29 Jun 1998 09:51:36 -0700 (PDT) Message-Id: <199806291651.JAA08089@vandys-pc.cisco.com> To: Ondrej Martinek cc: vsta@zendo.com Subject: Re: console problem In-reply-to: Your message of "Mon, 29 Jun 1998 17:15:06 +0200." Date: Mon, 29 Jun 1998 09:51:36 -0700 From: Andy Valencia [Ondrej Martinek writes:] > I succeded to start MGR but it crashed the system after several seconds. > (btw, the mouse moved very slowly) "crashed" in what sense? Kernel panic? Hang? > 1. I want to learn cons3 to send signal (aka SIGKILL) on ^C. Is there any > way how a server can find out the PID of it's client? > Is it in msg.m_sender field of struct msg? No, see the s_pgrp field in cons2 for how to do it. > 2. Vsta sometimes stops responding me. I find out (using kernel debbuger) > that one thread is in page_out(). Do "dv freemem" and if it's zero, yes, you're out of memory. >Is it necessary to have swap > installed? Can I do it the following way for instance? > mkfs_bfs /bfs 100 > bfs /bfs fs/bfs > swap > swapd fs/bfs Swap uses raw partitions, not filesystems. So pick a free partition, and then swap onto it. Note that this code is rarely used (because VSTa generally fits all in RAM anyway), so you should only go down this path if you feel ready to do kernel debugging. > 3. During porting some shell scripts I find strange behaviour of "sh": > -The shell started from another shell by commnad "sh" never inherits > the environment but running a script is ok. > eg: > $export x=1; sh -c "echo x=\$x" I don't believe "export var=" is standard. "x=1 ; export x" may be more reliable. > eg: suppose this script: > test 1 = 0 > echo done. > when executed it prints: > perm > done. Right, termination code versus errno. VSTa unifies some of this (including signals) which can have interesting interactions. > 4. I have root privileges but still can create file and change its > access flags so that I can't access anymore. Is it alright? > (btw, on DOS filesystem) DOS does not implement VSTa privs. It merely keeps track of R/W or R/O, and also you have to have sys.sys to write the filesystem (by default). > 5. How can I link dynamic and static libraries? What is default? Static is the default. > Thanks for answering my last questions. Can I send the next ones to > vsta@cisco.com too? Is anyone of you, Vsta gurus, using IRC (internet > relay chat) to talk interactivelly? Personally, I do not have enough time to participate in RT. Mail lets me schedule a little time here and there. Andy From daemon Wed Sep 2 20:33:04 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id UAA16322 for vsta-xplod; Wed, 2 Sep 1998 20:33:04 GMT Received: from NII.itb.ac.id (nii.itb.ac.id [167.205.48.134]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id UAA16316 for ; Wed, 2 Sep 1998 20:32:58 GMT Received: from localhost (helmi@localhost) by NII.itb.ac.id (8.8.8/8.8.7) with SMTP id OAA04690 for ; Thu, 3 Sep 1998 14:33:17 +0700 (JAVT) (envelope-from helmi@NII.itb.ac.id) Date: Thu, 3 Sep 1998 14:33:16 +0700 (JAVT) From: Helmi Zain To: vsta@zendo.com Subject: problem startup Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII I have same problem when I try boot vsta from floppy with GRUB. The error messege is: init can't find root, sleeping ! init can't find root, sleeping ! and than i can't do anything, i loss control my machine :-) Anyone can help me ? what should i do ? Regards, Helmi Zain From daemon Thu Sep 3 04:54:10 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id EAA17504 for vsta-xplod; Thu, 3 Sep 1998 04:54:10 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id EAA17498 for ; Thu, 3 Sep 1998 04:54:07 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by zipper.cisco.com (8.8.5-Cisco.2-SunOS.5.5.1.sun4/8.6.5) with ESMTP id IAA02621; Thu, 3 Sep 1998 08:54:13 -0700 (PDT) Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id IAA12361; Thu, 3 Sep 1998 08:54:13 -0700 (PDT) Message-Id: <199809031554.IAA12361@vandys-pc.cisco.com> To: Helmi Zain cc: vsta@zendo.com Subject: Re: problem startup In-reply-to: Your message of "Thu, 03 Sep 1998 14:33:16 +0700." Date: Thu, 03 Sep 1998 08:54:13 -0700 From: Andy Valencia [Helmi Zain writes:] >I have same problem when I try boot vsta from floppy with GRUB. >The error messege is: >init can't find root, sleeping ! >init can't find root, sleeping ! Since you didn't drop into the debugger, none of the processes exited. Init is trying to mount the DOS root filesystem, and this filesystem is trying to open the FD disk device. >and than i can't do anything, i loss control my machine :-) Can you ^Z (control key, and Z) to get into the kernel debugger? >Anyone can help me ? what should i do ? Assuming you can get into the kernel debugger, do a "pr" to see which processes exist. Then work your way back. Find the FD task, and do "pr " for it. You'll see each thread, and do "bt " for each, to see what they're blocked on. Then do the same thing for the DOS filesystem task, and for grins, init. Hopefully you'll find one of them blocked on something unexpected. Normally you'd expect to see the FD thread blocked on a msg_receive(), and maybe one on timed_sleep(). The DOS one should show threads in msg_receive() (if it's idle, waiting for work) or down in msg_send() (if it has a request outstanding to the disk). For msg_send(), the first argument is the port # it's talking to. You can then do "pr " and look at the "ports: " and count over for this port #, then do "ref " (which looks up a port reference, i.e., an open port) for this port. The "state" is of interest, and also "port". You can then do "port " for the latter, and it'll tell you about the port, including the PID which serves that port. You can then do some sanity cross checking. We can work this offline if you get some more data and can't find the problem. Andy From daemon Thu Sep 3 14:15:51 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id OAA18654 for vsta-xplod; Thu, 3 Sep 1998 14:15:51 GMT Received: from dfw-ix7.ix.netcom.com (dfw-ix7.ix.netcom.com [206.214.98.7]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id OAA18648 for ; Thu, 3 Sep 1998 14:15:47 GMT Received: (from smap@localhost) by dfw-ix7.ix.netcom.com (8.8.4/8.8.4) id TAA10089; Thu, 3 Sep 1998 19:55:49 -0500 (CDT) Received: from nbw-nj10-49.ix.netcom.com(207.94.119.113) by dfw-ix7.ix.netcom.com via smap (V1.3) id rma010056; Thu Sep 3 19:55:38 1998 Sender: jb@ix.netcom.com Message-ID: <35EF161D.8ED24A32@ix.netcom.com> Date: Thu, 03 Sep 1998 18:20:13 -0400 From: Jean-Baptiste Reply-To: jbnivoit@ix.netcom.com X-Mailer: Mozilla 4.04 [en] (X11; I; FreeBSD 2.2.6-RELEASE i386) MIME-Version: 1.0 To: Helmi Zain CC: vsta@zendo.com Subject: Re: problem startup References: <199809031554.IAA12361@vandys-pc.cisco.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit > >I have same problem when I try boot vsta from floppy with GRUB. > >The error messege is: > >init can't find root, sleeping ! > >init can't find root, sleeping ! > > Since you didn't drop into the debugger, none of the processes exited. Init > is trying to mount the DOS root filesystem, and this filesystem is trying to > open the FD disk device. > > >and than i can't do anything, i loss control my machine :-) > > Can you ^Z (control key, and Z) to get into the kernel debugger? > > >Anyone can help me ? what should i do ? The command line you use to start the DOS server is probably wrong. I think i remember it being mistyped in the roadmap, or something like this... HTH, jb From daemon Fri Sep 4 03:53:15 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id DAA20400 for vsta-xplod; Fri, 4 Sep 1998 03:53:15 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id DAA20394 for ; Fri, 4 Sep 1998 03:53:11 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by zipper.cisco.com (8.8.5-Cisco.2-SunOS.5.5.1.sun4/8.6.5) with ESMTP id HAA29435; Fri, 4 Sep 1998 07:53:21 -0700 (PDT) Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id HAA14116; Fri, 4 Sep 1998 07:53:20 -0700 (PDT) Message-Id: <199809041453.HAA14116@vandys-pc.cisco.com> To: "Michael Henry" cc: vsta@zendo.com Subject: Re: Can't boot In-reply-to: Your message of "Fri, 04 Sep 1998 21:30:41 +1000." <199809040030.AAA19945@bodhi.zendo.com> Date: Fri, 04 Sep 1998 07:53:20 -0700 From: Andy Valencia ["Michael Henry" writes:] > Boot process 7 dies > [Kernel debugger @ spl 0] > > >When I get this message, the HDD light is permanently on. At a guess, your IDE disk takes a long time to reset, and the driver's giving up on you. > Boot process 5 dies > [Kernel debugger @ spl 0] > > >The HDD light is off when I get this message. You probably inherited one of my mistakes, which is the arguments to the filesystem process. Your "DOS" command line in boot/grub/menu.lst should look like: module= /vsta/boot/dos -d disk/wd:wd0_dos0 -n fs/root >I don't know how to use the debugger, and if I did, I wouldn't know >what to look for, so I'd appreciate it if someone could give me some >pointers as to what's going on here and how I can fix it. If that doesn't help, a "pr" will at least give you a list of the processes with their name, which'll tell you who died by name. "pr " tells you about the process, including the threads within it. "bt " tells you the stack backtrace of a given thread. You can tell me more in direct E-mail if the dos arguments change doesn't help. Andy From daemon Fri Sep 4 06:09:26 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id GAA20925 for vsta-xplod; Fri, 4 Sep 1998 06:09:26 GMT Received: from home.chat.net (genesis.jeske.meer.net [204.94.139.85]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id GAA20919 for ; Fri, 4 Sep 1998 06:09:23 GMT Received: (qmail 23644 invoked by uid 120); 4 Sep 1998 17:13:09 -0000 Message-ID: <19980904101309.W10199@home.chat.net> Date: Fri, 4 Sep 1998 10:13:09 -0700 From: David Jeske To: vsta@zendo.com Subject: Re: Can't boot Mail-Followup-To: vsta@zendo.com References: <199809040030.AAA19945@bodhi.zendo.com> <199809041453.HAA14116@vandys-pc.cisco.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.92.1 In-Reply-To: <199809041453.HAA14116@vandys-pc.cisco.com>; from Andy Valencia on Fri, Sep 04, 1998 at 07:53:20AM -0700 On Fri, Sep 04, 1998 at 07:53:20AM -0700, Andy Valencia wrote: > ["Michael Henry" writes:] > > > Boot process 7 dies > > [Kernel debugger @ spl 0] > > > > >When I get this message, the HDD light is permanently on. > > At a guess, your IDE disk takes a long time to reset, and the driver's > giving up on you. Michael, Perhaps you should try pulling down: http://www.chat.net/~jeske/VSTa/Software/vstaflop.zip http://sunsite.unc.edu/pub/Linux/system/install/rawwrite/rawrite2.exe 1) Just unzip the first one, what you'll get out is a disk image. Use rawrite2 to put the disk image onto a floppy. 2) boot the floppy It should bring up vsta running completely off the floppy. I believe I included the wd server on that disk, so you should then be able to start wd against your hard drive manually and see what it does. -- David Jeske (N9LCA) + http://www.chat.net/~jeske/ + jeske@chat.net From daemon Tue Sep 15 03:19:30 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id DAA20304 for vsta-xplod; Tue, 15 Sep 1998 03:19:30 GMT Received: from NII.itb.ac.id (nii.itb.ac.id [167.205.48.134]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id DAA20298 for ; Tue, 15 Sep 1998 03:19:21 GMT Received: from localhost (helmi@localhost) by NII.itb.ac.id (8.8.8/8.8.7) with SMTP id VAA09686 for ; Tue, 15 Sep 1998 21:21:18 +0700 (JAVT) (envelope-from helmi@NII.itb.ac.id) Date: Tue, 15 Sep 1998 21:21:18 +0700 (JAVT) From: Helmi Zain To: vsta@zendo.com Subject: Window system In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hi.. is there any windows system with graphic capabilities for vsta ? i want the small one and i can use internet broswer on it helmi z From daemon Tue Sep 15 03:26:58 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id DAA20328 for vsta-xplod; Tue, 15 Sep 1998 03:26:58 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id DAA20322 for ; Tue, 15 Sep 1998 03:26:54 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by zipper.cisco.com (8.8.5-Cisco.2-SunOS.5.5.1.sun4/8.6.5) with ESMTP id HAA05981; Tue, 15 Sep 1998 07:27:34 -0700 (PDT) Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id HAA09190; Tue, 15 Sep 1998 07:27:34 -0700 (PDT) Message-Id: <199809151427.HAA09190@vandys-pc.cisco.com> To: Helmi Zain cc: vsta@zendo.com Subject: Re: Window system In-reply-to: Your message of "Tue, 15 Sep 1998 21:21:18 +0700." Date: Tue, 15 Sep 1998 07:27:34 -0700 From: Andy Valencia [Helmi Zain writes:] >is there any windows system with graphic capabilities for vsta ? I ported MGR quite a while ago. >i want >the small one and i can use internet broswer on it I'm afraid MGR was written long before WWW existed, and I haven't heard of such a beast. If you really needed one, probably the path of least resistance is to port X11 to VSTa. A base-level port of an unaccelerated SVGA video server probably wouldn't be *that* bad. Andy Valencia From daemon Wed Sep 16 05:26:01 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id FAA22782 for vsta-xplod; Wed, 16 Sep 1998 05:26:01 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id FAA22776 for ; Wed, 16 Sep 1998 05:25:59 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by zipper.cisco.com (8.8.5-Cisco.2-SunOS.5.5.1.sun4/8.6.5) with ESMTP id JAA27823; Wed, 16 Sep 1998 09:26:12 -0700 (PDT) Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id JAA11439; Wed, 16 Sep 1998 09:26:11 -0700 (PDT) Message-Id: <199809161626.JAA11439@vandys-pc.cisco.com> To: goconne@csc.UVic.CA (Gordon W. O'Connell) cc: vsta@zendo.com, goconne@gulf.csc.UVic.CA Subject: Re: gcc compiler In-reply-to: Your message of "Wed, 16 Sep 1998 09:01:59 PDT." <199809161601.JAA09756@valdes.csc.UVic.CA> Date: Wed, 16 Sep 1998 09:26:10 -0700 From: Andy Valencia [goconne@csc.UVic.CA (Gordon W. O'Connell) writes:] > I'm using VSTa 1.6.1 on a 386 with 8M of memory. I've been trying > to build the VSTa kernel in /vsta/src/os/make; but shortly after > the make begins, gcc fails with an error code 1. Is gcc trying > to tell me I don't have enough memory? No, as of 1.6.1, gcc requires an FPU, and I'm assuming that your 386 is not endowed with a 387. :-( > If I am out of memory; would it help to load the swap daemon > (/vsta/boot/swap) at boot time? Where exactly would the module > line for /vsta/boot/swap have to go? There also seems to be > an executable in /vsta/bin/swapd. What does this do? Should this > be placed in the inittab? If so, where should it go? > Would having the swap service running allow gcc to run in an > 8M system? gcc would hang (well, block for more memory) if that was the problem. You can use the older gcc port, which had some hacks to avoid using floating point. I can't think of anything which would break offhand. Andy From daemon Wed Sep 16 08:08:30 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id IAA23061 for vsta-xplod; Wed, 16 Sep 1998 08:08:30 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id IAA23055 for ; Wed, 16 Sep 1998 08:08:27 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by zipper.cisco.com (8.8.5-Cisco.2-SunOS.5.5.1.sun4/8.6.5) with ESMTP id MAA00419; Wed, 16 Sep 1998 12:09:08 -0700 (PDT) Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id MAA12344; Wed, 16 Sep 1998 12:09:07 -0700 (PDT) Message-Id: <199809161909.MAA12344@vandys-pc.cisco.com> To: Helmi Zain cc: vsta@zendo.com Subject: Re: MGR In-reply-to: Your message of "Thu, 17 Sep 1998 02:05:29 +0700." Date: Wed, 16 Sep 1998 12:09:07 -0700 From: Andy Valencia [Helmi Zain writes:] >i try to make MGR >i have some error when i try to ./configure It's already pre-configured. You should just "cd" down into the source dir in which you want to compile. Andy From daemon Wed Sep 16 08:03:39 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id IAA23046 for vsta-xplod; Wed, 16 Sep 1998 08:03:39 GMT Received: from NII.itb.ac.id (nii.itb.ac.id [167.205.48.134]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id IAA23040 for ; Wed, 16 Sep 1998 08:03:34 GMT Received: from localhost (helmi@localhost) by NII.itb.ac.id (8.8.8/8.8.7) with SMTP id CAA15517; Thu, 17 Sep 1998 02:05:29 +0700 (JAVT) (envelope-from helmi@NII.itb.ac.id) Date: Thu, 17 Sep 1998 02:05:29 +0700 (JAVT) From: Helmi Zain To: Andy Valencia cc: vsta@zendo.com Subject: MGR In-Reply-To: <199809151427.HAA09190@vandys-pc.cisco.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII hi i try to make MGR i have some error when i try to ./configure m4: not found Eeeek, you don't have m4! you will have to install it first what is m4 ? helmi z From daemon Wed Sep 16 08:24:16 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id IAA23104 for vsta-xplod; Wed, 16 Sep 1998 08:24:16 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id IAA23098 for ; Wed, 16 Sep 1998 08:24:13 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by zipper.cisco.com (8.8.5-Cisco.2-SunOS.5.5.1.sun4/8.6.5) with ESMTP id MAA02043; Wed, 16 Sep 1998 12:24:27 -0700 (PDT) Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id MAA12412; Wed, 16 Sep 1998 12:24:26 -0700 (PDT) Message-Id: <199809161924.MAA12412@vandys-pc.cisco.com> To: goconne@csc.UVic.CA (Gordon W. O'Connell) cc: vsta@zendo.com, goconne@gulf.csc.UVic.CA Subject: Re: Problems with gcc In-reply-to: Your message of "Wed, 16 Sep 1998 12:20:43 PDT." <199809161920.MAA13501@valdes.csc.UVic.CA> Date: Wed, 16 Sep 1998 12:24:26 -0700 From: Andy Valencia [goconne@csc.UVic.CA (Gordon W. O'Connell) writes:] > You were correct. I am trying to compile using a 386SX and no > FPU. You mentioned that it is okay to use the compiler which > comes with 1.5.2. I'm assuming that the executables I need > to replace are: as, cc1, gcc, ld and cc. Is this a correct > assumption? Or can I get away with just replaceing the > gcc and cc1 components? Have I missed some components? I would strongly recommend replacing all of them at once. > Can the kernel build still use vsta.lnk to bind? I believe the old "ld" didn't use this kind of file. The correct parameters were compiled into it. Andy From daemon Fri Sep 18 02:55:15 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id CAA27710 for vsta-xplod; Fri, 18 Sep 1998 02:55:15 GMT Received: from staff.cs.usyd.edu.au ([129.78.8.1]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id CAA27704 for ; Fri, 18 Sep 1998 02:55:10 GMT Message-Id: <199809180255.CAA27704@bodhi.zendo.com> Subject: Boot probs... To: vsta@zendo.com Date: Fri, 18 Sep 1998 23:56:18 +1000 (EST) From: "Michael Henry" Content-Type: text I am finally able to start my experimenting with VSTa again! (The last couple of weeks have been busy) The story so far... I was unable to boot, and was getting messages about "boot process 7" dying. Vandys' suggestion was to use: module=/vsta/boot/dos -d disk/wd0_dos0 -n fs/root instead of: module=/vsta/boot/dos -p disk/wd0_dos0 fs/root I did this and now I'm getting: init: can't find root, sleeping (This happens about three times) syslog: dos (pid 7) error: unable to open 'disk/wd:wd0_dos0' Does anyone have any suggestions? From daemon Fri Sep 18 07:41:32 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id HAA28291 for vsta-xplod; Fri, 18 Sep 1998 07:41:32 GMT Received: from sims-ha.videotron.net (faure.videotron.net [205.151.222.100]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id HAA28285 for ; Fri, 18 Sep 1998 07:41:29 GMT From: dufrp@oricom.ca Received: from 207.96.204.178 (ts4-178.oricom.ca) by sims-ha.videotron.net (Sun Internet Mail Server sims.3.5.1998.07.14.10.43) with SMTP id <0EZH00LM2T9Z8G@sims-ha.videotron.net> for vsta@zendo.com; Fri, 18 Sep 1998 14:42:04 -0400 (EDT) Date: Fri, 18 Sep 1998 14:42:04 -0400 (EDT) Date-warning: Date header was inserted by sims-ha.videotron.net Subject: boot process 7 dying, I have no solution To: vsta@zendo.com Cc: dufrp@oricom.ca Message-id: <0EZH00LM6TA08G@sims-ha.videotron.net> MIME-version: 1.0 X-Mailer: MailSender for Oberon (ejz) Content-type: TEXT/PLAIN; CHARSET=US-ASCII The only thing I'd like to add is that I have blocked on this too, about two months ago whan I wanted to try a small OS on my new 486DX33 motherboard with 8Mb ram. I was not feeling like posting here, so I did move to an other OS (FreeBSD, than later to Oberon when I saw that FreeBSD was taking all my boths drives 85Mb+43Mb). Now I still have plenty of room because Oberon so small even with graphics and PPP. Now I still like to install VSTA on a free 40Mb partition I have, that is, if a solution to the Boot process 7 dying problem get resolved. ---Paul Dufresne From daemon Fri Sep 18 08:12:40 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id IAA28330 for vsta-xplod; Fri, 18 Sep 1998 08:12:40 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id IAA28324 for ; Fri, 18 Sep 1998 08:12:37 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by zipper.cisco.com (8.8.5-Cisco.2-SunOS.5.5.1.sun4/8.6.5) with ESMTP id MAA07655; Fri, 18 Sep 1998 12:13:28 -0700 (PDT) Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id MAA16924; Fri, 18 Sep 1998 12:13:27 -0700 (PDT) Message-Id: <199809181913.MAA16924@vandys-pc.cisco.com> To: "Michael Henry" cc: vsta@zendo.com Subject: Re: Boot probs... In-reply-to: Your message of "Fri, 18 Sep 1998 23:56:18 +1000." <199809180255.CAA27704@bodhi.zendo.com> Date: Fri, 18 Sep 1998 12:13:27 -0700 From: Andy Valencia ["Michael Henry" writes:] >I was unable to boot, and was getting messages about "boot process 7" >dying. > module=/vsta/boot/dos -d disk/wd0_dos0 -n fs/root >instead of: > module=/vsta/boot/dos -p disk/wd0_dos0 fs/root >I did this and now I'm getting: Ok, good. So now your disk server may be starting. Are you sure you said "disk/wd_dos0"? Your error message below looks like you configured disk/wd:wd0_dos0, which would be right for the first DOS partition on the first IDE disk. > init: can't find root, sleeping >(This happens about three times) > syslog: dos (pid 7) error: unable to open 'disk/wd:wd0_dos0' Do ^Z to enter the kernel debugger, and "pr" to see the proceses. Is there still a "wd" process? Do "pr " on it, and then do "bt " on the thread within it. Let me know what the backtrace looks like (just routine names will be fine) and we can see where Mr. Disk has gone. Thanks, Andy From daemon Sat Sep 19 20:06:52 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id UAA03360 for vsta-xplod; Sat, 19 Sep 1998 20:06:52 GMT Received: from staff.cs.usyd.edu.au (staff.cs.usyd.edu.au [129.78.8.1]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id UAA03354 for ; Sat, 19 Sep 1998 20:06:47 GMT Message-Id: <199809192006.UAA03354@bodhi.zendo.com> Subject: Re: Boot probs... To: vsta@zendo.com Date: Sun, 20 Sep 1998 17:08:03 +1000 (EST) From: "Michael Henry" In-Reply-To: <199809181913.MAA16924@vandys-pc.cisco.com> from "Andy Valencia" at Sep 18, 98 12:13:27 pm Content-Type: text > Ok, good. So now your disk server may be starting. Are you sure you said > "disk/wd_dos0"? No, I said "disk/wd:wd0_dos0" > Your error message below looks like you configured > disk/wd:wd0_dos0, which would be right for the first DOS partition on the > first IDE disk. That's the one. > > Do ^Z to enter the kernel debugger, and "pr" to see the proceses. Is there > still a "wd" process? Yes. > Do "pr " on it, and then do "bt " on the > thread within it. The output of "pr 5" looks like: 5 wd vas 181ea10 threads 181fa00 runq 181dda0 sys/usr 0/0 . . . 6 SLEEP kstack c0b000 uregs cobfc0 wchan 1819950 / . . . I assume the "6 SLEEP" entry is the thread". (?) > Let me know what the backtrace looks like (just routine > names will be fine) and we can see where Mr. Disk has gone. "bt 6" produces: _switch() called from _p_sema_v_lock + 18a _p_sema_v_lock() called from _msg_receive + 44 _msg_receive() called from _syscall +dl _syscall() called from _Tsyscall + 17 Does this give you any clues? From daemon Sun Sep 20 03:13:26 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id DAA03982 for vsta-xplod; Sun, 20 Sep 1998 03:13:26 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id DAA03976 for ; Sun, 20 Sep 1998 03:13:23 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by zipper.cisco.com (8.8.5-Cisco.2-SunOS.5.5.1.sun4/8.6.5) with ESMTP id HAA07176; Sun, 20 Sep 1998 07:14:19 -0700 (PDT) Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id HAA22279; Sun, 20 Sep 1998 07:14:18 -0700 (PDT) Message-Id: <199809201414.HAA22279@vandys-pc.cisco.com> To: "Michael Henry" cc: vsta@zendo.com Subject: Re: Boot probs... In-reply-to: Your message of "Sun, 20 Sep 1998 17:08:03 +1000." <199809192006.UAA03354@bodhi.zendo.com> Date: Sun, 20 Sep 1998 07:14:18 -0700 From: Andy Valencia ["Michael Henry" writes:] >I assume the "6 SLEEP" entry is the thread". (?) Yes. >"bt 6" produces: > _switch() called from _p_sema_v_lock + 18a > _p_sema_v_lock() called from _msg_receive + 44 > _msg_receive() called from _syscall +dl > _syscall() called from _Tsyscall + 17 >Does this give you any clues? This looks like a normal disk driver server, blocked waiting to receive more requests. Very healthy, very normal. Now I guess do the same thing for the "dos" process; you should see two threads there. Also, make sure your "dos" invocation line has him advertise his filesystem as "fs/root". Andy From daemon Tue Sep 22 15:10:14 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id PAA10490 for vsta-xplod; Tue, 22 Sep 1998 15:10:14 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id PAA10484 for ; Tue, 22 Sep 1998 15:10:12 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by zipper.cisco.com (8.8.5-Cisco.2-SunOS.5.5.1.sun4/8.6.5) with ESMTP id TAA03966; Tue, 22 Sep 1998 19:11:15 -0700 (PDT) Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id TAA27584; Tue, 22 Sep 1998 19:11:15 -0700 (PDT) Message-Id: <199809230211.TAA27584@vandys-pc.cisco.com> To: rayr@rramos.ucsc.edu cc: vsta@zendo.com Subject: Re: networking questions In-reply-to: Your message of "Tue, 22 Sep 1998 18:11:35 MDT." <199809230111.SAA09298@rramos.ucsc.edu> Date: Tue, 22 Sep 1998 19:11:15 -0700 From: Andy Valencia [rayr@rramos.ucsc.edu writes:] >I have a 486/100 running VSTa 1.6.1. I've gotten ka9q to work >and telnetted to my Linux box successfully using a NE2K card. >What do I need to create a standalone telnet program without >going through ka9q? You need to mosey on down to the "cmds" subdirectory and write such a thing! ;-> Since I never got around to emulating sockets (tsk, tsk, but emulating select() requires some fairly fundamental decisions), you need to code it using wstat()/rstat(). There's some sample code down there which knows how to initiate connections. You'll also want to surf over to www.ietf.org, and find the RFC for the telnet protocol. You can implement a very small subset of what exists, and the telnet server in KA9Q will give you some good hints of how to handle the options. >Also, how would I start getting a ppp server >running on vsta? (I'm assuming you mean async PPP....) I'd start by going over the SLIP code. This'll show you how to connect to an async source and frame it into IP. Then I'd grab one of the free PPP implementations for Linux or FreeBSD, and start adapting. PPP is an even more ornate protocol than telnet, but again, there's a lot you can skip. Regards, Andy From daemon Tue Sep 22 14:26:08 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id OAA10432 for vsta-xplod; Tue, 22 Sep 1998 14:26:08 GMT Received: from rramos.ucsc.edu (root@rramos.UCSC.EDU [128.114.163.51]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id OAA10426 for ; Tue, 22 Sep 1998 14:26:06 GMT From: rayr@rramos.ucsc.edu Received: from rramos.ucsc.edu (rayr@localhost [127.0.0.1]) by rramos.ucsc.edu (8.8.7/8.8.7) with ESMTP id SAA09298 for ; Tue, 22 Sep 1998 18:11:35 -0700 Message-Id: <199809230111.SAA09298@rramos.ucsc.edu> To: vsta@zendo.com Subject: networking questions Date: Tue, 22 Sep 1998 18:11:35 -0600 Hi all, I have a 486/100 running VSTa 1.6.1. I've gotten ka9q to work and telnetted to my Linux box successfully using a NE2K card. What do I need to create a standalone telnet program without going through ka9q? Also, how would I start getting a ppp server running on vsta? I'm looking for a guide, I'm a newbie to network programming... -ray From daemon Thu Sep 24 16:39:51 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id QAA16362 for vsta-xplod; Thu, 24 Sep 1998 16:39:51 GMT Received: from staff.cs.usyd.edu.au (staff.cs.usyd.edu.au [129.78.8.1]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id QAA16355 for ; Thu, 24 Sep 1998 16:39:47 GMT Message-Id: <199809241639.QAA16355@bodhi.zendo.com> Subject: Re: Boot probs... To: vsta@zendo.com Date: Thu, 24 Sep 1998 20:44:05 +1000 (EST) From: "Michael Henry" In-Reply-To: <199809201414.HAA22279@vandys-pc.cisco.com> from "Andy Valencia" at Sep 20, 98 07:14:18 am Content-Type: text > >"bt 6" produces: > > _switch() called from _p_sema_v_lock + 18a > > _p_sema_v_lock() called from _msg_receive + 44 > > _msg_receive() called from _syscall +dl > > _syscall() called from _Tsyscall + 17 > >Does this give you any clues? > > This looks like a normal disk driver server, blocked waiting to receive more > requests. Very healthy, very normal. > > Now I guess do the same thing for the "dos" process; you should see two > threads there. There is only one thread. It is called "onproc". bt produces: _swtch(0x1, 0x1a, 0x1185a8, 0x1) called from (b) > Also, make sure your "dos" invocation line has him advertise > his filesystem as "fs/root". module=/vsta/boot/dos -d disk/wd:wd0_dos0 -n fs/root is what I use. > > Andy > From daemon Mon Sep 28 04:30:43 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id EAA27113 for vsta-xplod; Mon, 28 Sep 1998 04:30:43 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id EAA27107 for ; Mon, 28 Sep 1998 04:30:38 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by zipper.cisco.com (8.8.5-Cisco.2-SunOS.5.5.1.sun4/8.6.5) with ESMTP id IAA25709; Mon, 28 Sep 1998 08:31:57 -0700 (PDT) Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id IAA12116; Mon, 28 Sep 1998 08:31:57 -0700 (PDT) Message-Id: <199809281531.IAA12116@vandys-pc.cisco.com> To: "Michael Henry" cc: vsta@zendo.com Subject: Re: Boot probs... In-reply-to: Your message of "Thu, 24 Sep 1998 20:44:05 +1000." <199809241639.QAA16355@bodhi.zendo.com> Date: Mon, 28 Sep 1998 08:31:56 -0700 From: Andy Valencia ["Michael Henry" writes:] >> Now I guess do the same thing for the "dos" process; you should see two >> threads there. >There is only one thread. It is called "onproc". >bt produces: > _swtch(0x1, 0x1a, 0x1185a8, 0x1) called from (b) Hmmm, usually there should be two threads, one doing async I/O to the disk, and the other doing the filesystem. It looks like one of your threads died while holding a mutex, and now the other thread is spinning, waiting for this mutex. Common causes include disks with pseudo-geometries, disks with certain kinds of compressed filesystems, and disks with really odd FDISK entries. Usually something like this causes the FS to try and do I/O to a location off the end of the disk, which results in a failed I/O. Andy From daemon Mon Sep 28 10:28:14 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id KAA27995 for vsta-xplod; Mon, 28 Sep 1998 10:28:14 GMT Received: from atlas.cs.york.ac.uk (Atlas.cs.york.ac.uk [144.32.176.2]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id KAA27989 for ; Mon, 28 Sep 1998 10:28:11 GMT From: "Robin J Bradley (4Y0)" Message-Id: <9809282224.ZM14052@ugrad.cs.york.ac.uk> Date: Mon, 28 Sep 1998 22:24:24 +0100 In-Reply-To: Andy Valencia "Re: Boot probs..." (Sep 20, 7:14am) References: <199809201414.HAA22279@vandys-pc.cisco.com> X-Mailer: Z-Mail (3.2.1 6apr95 MediaMail) To: vsta@zendo.com Subject: MADO and Graphics Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Does MADO still exist as a project and is there a working version of the bitblt server it was going to use. I was looking at doing some graphics under VSTa and a bltbit server would be ideal for high level language graphics. Robin -- Robin J. Bradley (rjb108@cs.york.ac.uk) ____________________________________________________________________________ If we define things by there opposits, then reality is defined by our dreams From daemon Mon Sep 28 12:09:19 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id MAA28184 for vsta-xplod; Mon, 28 Sep 1998 12:09:19 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id MAA28178 for ; Mon, 28 Sep 1998 12:09:17 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by zipper.cisco.com (8.8.5-Cisco.2-SunOS.5.5.1.sun4/8.6.5) with ESMTP id QAA15013; Mon, 28 Sep 1998 16:10:37 -0700 (PDT) Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id QAA13827; Mon, 28 Sep 1998 16:10:37 -0700 (PDT) Message-Id: <199809282310.QAA13827@vandys-pc.cisco.com> To: "Robin J Bradley (4Y0)" cc: vsta@zendo.com Subject: Re: MADO and Graphics In-reply-to: Your message of "Mon, 28 Sep 1998 22:24:24 BST." <9809282224.ZM14052@ugrad.cs.york.ac.uk> Date: Mon, 28 Sep 1998 16:10:36 -0700 From: Andy Valencia ["Robin J Bradley (4Y0)" writes:] >Does MADO still exist as a project and is there a working version of the bitbl >t >server it was going to use. I was looking at doing some graphics under VSTa >and a bltbit server would be ideal for high level language graphics. Sadly, MADO has gathered dust for many years now. It will probably stay quiescent unless some brave new soul picks it up and runs with it. I ported MGR to get some base-level windowing functionality, but that's the only alternative I'm aware of. Andy Valencia From daemon Tue Sep 29 19:13:11 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id TAA02023 for vsta-xplod; Tue, 29 Sep 1998 19:13:11 GMT Received: from staff.cs.usyd.edu.au (staff.cs.usyd.edu.au [129.78.8.1]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id TAA02017 for ; Tue, 29 Sep 1998 19:12:57 GMT Message-Id: <199809291912.TAA02017@bodhi.zendo.com> Subject: Re: Boot probs... To: vsta@zendo.com Date: Wed, 30 Sep 1998 16:14:38 +1000 (EST) From: "Michael Henry" In-Reply-To: <199809281531.IAA12116@vandys-pc.cisco.com> from "Andy Valencia" at Sep 28, 98 08:31:56 am Content-Type: text I found the solution to the problem! I downloaded Jeske's floppy image and used it to boot vsta from the floppy, and then from the hard drive, without any problems. So I looked at his boot sequence, and he had (from memory) module=/vsta/boot/dos -n fs/root -d //dos/fd:fd0_dos0 I'd been using "-d dos/fd:fd0_dos0", as per the recommendations given in this mailing list. (Since the documentation was wrong). Note to the documentors: Could you please make sure the documentation is upgraded to reflect the proper boot procedure. Thanks, Michael > > ["Michael Henry" writes:] > > >> Now I guess do the same thing for the "dos" process; you should see two > >> threads there. > >There is only one thread. It is called "onproc". > >bt produces: > > _swtch(0x1, 0x1a, 0x1185a8, 0x1) called from (b) > > Hmmm, usually there should be two threads, one doing async I/O to the disk, > and the other doing the filesystem. It looks like one of your threads died > while holding a mutex, and now the other thread is spinning, waiting for > this mutex. > > Common causes include disks with pseudo-geometries, disks with certain kinds > of compressed filesystems, and disks with really odd FDISK entries. Usually > something like this causes the FS to try and do I/O to a location off the > end of the disk, which results in a failed I/O. > > Andy > From daemon Wed Sep 30 04:22:31 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id EAA03024 for vsta-xplod; Wed, 30 Sep 1998 04:22:31 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id EAA03018 for ; Wed, 30 Sep 1998 04:22:28 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by zipper.cisco.com (8.8.5-Cisco.2-SunOS.5.5.1.sun4/8.6.5) with ESMTP id IAA09517; Wed, 30 Sep 1998 08:23:53 -0700 (PDT) Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id IAA17420; Wed, 30 Sep 1998 08:23:53 -0700 (PDT) Message-Id: <199809301523.IAA17420@vandys-pc.cisco.com> To: "Michael Henry" cc: vsta@zendo.com Subject: Re: Boot probs... In-reply-to: Your message of "Wed, 30 Sep 1998 16:14:38 +1000." <199809291912.TAA02017@bodhi.zendo.com> Date: Wed, 30 Sep 1998 08:23:52 -0700 From: Andy Valencia ["Michael Henry" writes:] >Could you please make sure the documentation is upgraded to reflect the >proper boot procedure. Sure. Can you list which doc files mislead you? Sorry for the inconvenience. Regards, Andy Valencia From daemon Fri Oct 2 01:38:34 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id BAA07672 for vsta-xplod; Fri, 2 Oct 1998 01:38:34 GMT Received: from atlas.cs.york.ac.uk (Atlas.cs.york.ac.uk [144.32.176.2]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id BAA07666 for ; Fri, 2 Oct 1998 01:38:31 GMT From: "Robin J Bradley (4Y0)" Message-Id: <9810021331.ZM20403@ugrad.cs.york.ac.uk> Date: Fri, 2 Oct 1998 13:31:05 +0100 In-Reply-To: Andy Valencia "Re: MADO and Graphics" (Sep 28, 4:10pm) References: <199809282310.QAA13827@vandys-pc.cisco.com> X-Mailer: Z-Mail (3.2.1 6apr95 MediaMail) To: Andy Valencia Subject: Re: MADO and Graphics Cc: vsta@zendo.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Since I'm feeling motivated at the moment (workload hasnt kicked in yet) ios the source for mado and bltbit such as it was arround somewhere where I could get my hands on it ? Robin From daemon Fri Oct 2 09:16:30 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id JAA08862 for vsta-xplod; Fri, 2 Oct 1998 09:16:30 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id JAA08856 for ; Fri, 2 Oct 1998 09:16:27 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by zipper.cisco.com (8.8.5-Cisco.2-SunOS.5.5.1.sun4/8.6.5) with ESMTP id NAA14854; Fri, 2 Oct 1998 13:17:55 -0700 (PDT) Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id NAA23945; Fri, 2 Oct 1998 13:17:54 -0700 (PDT) Message-Id: <199810022017.NAA23945@vandys-pc.cisco.com> To: "Robin J Bradley (4Y0)" cc: vsta@zendo.com Subject: Re: MADO and Graphics In-reply-to: Your message of "Fri, 02 Oct 1998 13:31:05 BST." <9810021331.ZM20403@ugrad.cs.york.ac.uk> Date: Fri, 02 Oct 1998 13:17:54 -0700 From: Andy Valencia ["Robin J Bradley (4Y0)" writes:] >Since I'm feeling motivated at the moment (workload hasnt kicked in yet) is >the source for mado and bltbit such as it was arround somewhere where I could >get my hands on it ? Nobody ever submitted anything to me. I believe MADO never went beyond the concept stage, but somebody out there might have bitblt server code of some sort or other. I've included an old MADO doc which Gavin wrote quite a while ago. Andy From daemon Fri Oct 2 11:51:37 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id LAA09104 for vsta-xplod; Fri, 2 Oct 1998 11:51:37 GMT Received: from home.chat.net (genesis.jeske.meer.net [204.94.139.85]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id LAA09098 for ; Fri, 2 Oct 1998 11:51:34 GMT Received: (qmail 22223 invoked by uid 120); 2 Oct 1998 22:56:44 -0000 Message-ID: <19981002155643.L17785@home.chat.net> Date: Fri, 2 Oct 1998 15:56:43 -0700 From: David Jeske To: vsta@zendo.com Subject: Re: MADO and Graphics Mail-Followup-To: vsta@zendo.com References: <9810021331.ZM20403@ugrad.cs.york.ac.uk> <199810022017.NAA23945@vandys-pc.cisco.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.92.1 In-Reply-To: <199810022017.NAA23945@vandys-pc.cisco.com>; from Andy Valencia on Fri, Oct 02, 1998 at 01:17:54PM -0700 On Fri, Oct 02, 1998 at 01:17:54PM -0700, Andy Valencia wrote: > ["Robin J Bradley (4Y0)" writes:] > > >Since I'm feeling motivated at the moment (workload hasnt kicked in yet) is > >the source for mado and bltbit such as it was arround somewhere where I could > >get my hands on it ? > > Nobody ever submitted anything to me. I believe MADO never went beyond the > concept stage, but somebody out there might have bitblt server code of some > sort or other. I've included an old MADO doc which Gavin wrote quite a > while ago. I remember Dave Hudson saying something about having a prototype bitblit almost working for VGA. However, I wouldn't be surprised if I'm wrong, or if he can't find it anymore. Anyone up for porting X? -- David Jeske (N9LCA) + http://www.chat.net/~jeske/ + jeske@chat.net From daemon Sat Oct 3 04:19:43 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id EAA12774 for vsta-xplod; Sat, 3 Oct 1998 04:19:43 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id EAA12768 for ; Sat, 3 Oct 1998 04:19:41 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by zipper.cisco.com (8.8.5-Cisco.2-SunOS.5.5.1.sun4/8.6.5) with ESMTP id IAA08444; Sat, 3 Oct 1998 08:21:15 -0700 (PDT) Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id IAA28051; Sat, 3 Oct 1998 08:21:14 -0700 (PDT) Message-Id: <199810031521.IAA28051@vandys-pc.cisco.com> To: Brett McCoy cc: vsta@zendo.com Subject: Re: MADO and Graphics In-reply-to: Your message of "Sat, 03 Oct 1998 01:35:00 EDT." Date: Sat, 03 Oct 1998 08:21:14 -0700 From: Andy Valencia [Brett McCoy writes:] >You might want to take a look at GGI. It's way more flexible and useful >than straight X, and there is an X server already ported to talk to GGI >instead of native frame buffers. This is the GGI from: http://synergy.caltech.edu/~ggi/ yes? Looks pretty good, although the X port could still probably benefit from a sockets emulation.... Andy From daemon Sat Oct 3 08:40:17 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id IAA13122 for vsta-xplod; Sat, 3 Oct 1998 08:40:17 GMT Received: from home.chat.net (genesis.jeske.meer.net [204.94.139.85]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id IAA13116 for ; Sat, 3 Oct 1998 08:40:14 GMT Received: (qmail 25248 invoked by uid 120); 3 Oct 1998 19:45:26 -0000 Message-ID: <19981003124526.K22462@home.chat.net> Date: Sat, 3 Oct 1998 12:45:26 -0700 From: David Jeske To: vsta@zendo.com Subject: Re: MADO and Graphics Mail-Followup-To: vsta@zendo.com References: <19981002155643.L17785@home.chat.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.92.1 In-Reply-To: ; from Brett McCoy on Sat, Oct 03, 1998 at 01:35:00AM -0400 On Sat, Oct 03, 1998 at 01:35:00AM -0400, Brett McCoy wrote: > On Fri, 2 Oct 1998, David Jeske wrote: > > >I remember Dave Hudson saying something about having a prototype > >bitblit almost working for VGA. > > > >However, I wouldn't be surprised if I'm wrong, or if he can't find it > >anymore. > > > >Anyone up for porting X? > > You might want to take a look at GGI. It's way more flexible and useful > than straight X, and there is an X server already ported to talk to GGI > instead of native frame buffers. GGI is definetly the way to go. Last time I paid attention to their stuff, they had finished their 'proof of concept' and had mostly started over from scratch to do the real implementation. On VSTa, there would be a user level GGI server which would load ggi drivers and handle the console issues. Programs would communicate with GGI via VSTa IPC instead of kernel calls. There are a few cases where the GGI server would want to setup a shared memory region to let the client talk directly to the hardware (most often the framebuffer). Andy: if a server setup a shared memory segment on 'physically mapped pages', and then a client connects to it, is it currently possible for the server to change the physical pages that shared memory segment maps to after the fact? I believe this would be required for GGI to work, because the server needs to be able to switch the client mapping from the 'real framebuffer' to a 'backing store' without the client knowing about it. -- David Jeske (N9LCA) + http://www.chat.net/~jeske/ + jeske@chat.net From daemon Sat Oct 3 09:47:46 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id JAA13187 for vsta-xplod; Sat, 3 Oct 1998 09:47:46 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id JAA13181 for ; Sat, 3 Oct 1998 09:47:44 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by zipper.cisco.com (8.8.5-Cisco.2-SunOS.5.5.1.sun4/8.6.5) with ESMTP id NAA20148; Sat, 3 Oct 1998 13:49:19 -0700 (PDT) Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id NAA28371; Sat, 3 Oct 1998 13:49:19 -0700 (PDT) Message-Id: <199810032049.NAA28371@vandys-pc.cisco.com> To: David Jeske cc: vsta@zendo.com Subject: Re: MADO and Graphics In-reply-to: Your message of "Sat, 03 Oct 1998 12:45:26 PDT." <19981003124526.K22462@home.chat.net> Date: Sat, 03 Oct 1998 13:49:18 -0700 From: Andy Valencia [David Jeske writes:] >Andy: if a server setup a shared memory segment on 'physically mapped >pages', and then a client connects to it, is it currently possible for >the server to change the physical pages that shared memory segment >maps to after the fact? Hmmm, not really, since there's not (currently) any good way to name a shared object distinct from naming what it maps. I think it'd require a new page set (struct pset, psop_* stuff) object type. In fact, I think it would have to be a lot like a ZFOD set, with an operation to switch from the backing store memory set to a physical map set. >I believe this would be required for GGI to work, because the server >needs to be able to switch the client mapping from the 'real >framebuffer' to a 'backing store' without the client knowing about it. Delta some interesting race conditions. Switching the mapping of the page set isn't enough, because you also have to hammer the client's PTE's and TLB (if they're running when you do the switch). I think some of the current page stealing code would have to be generalized to support this. I'll do the VM work if somebody actually wants to try a GGI port! Andy From daemon Mon Oct 5 17:57:57 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id RAA20207 for vsta-xplod; Mon, 5 Oct 1998 17:57:57 GMT Received: from rramos.ucsc.edu (root@rramos.UCSC.EDU [128.114.163.51]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id RAA20201 for ; Mon, 5 Oct 1998 17:57:55 GMT From: rayr@rramos.ucsc.edu Received: from rramos.ucsc.edu (rayr@localhost [127.0.0.1]) by rramos.ucsc.edu (8.8.7/8.8.7) with ESMTP id VAA15792 for ; Mon, 5 Oct 1998 21:42:02 -0700 Message-Id: <199810060442.VAA15792@rramos.ucsc.edu> To: vsta@zendo.com Subject: compile problems Date: Mon, 05 Oct 1998 21:42:02 -0600 I'm running this on a 486DX33 with 8 MB RAM, with just the base vsta.tz installed (version 1.6.1) gcc -v hello.c shows this: gcc version 2.7.2.2 cpp -lang-c -v -undef -D__GNUC__=2 -D__GNUC_MINOR__=7 -Dunix -Di386 -DVSTA -D__ unix__ -D__i386__ -D__VSTA__ -D__unix -D__i386 -D__VSTA -Asystem(unix) -Asystem( vsta) -Acpu(i386) -Amachine(i386) hello.c /tmp/cc000094.i cc1 /tmp/cc000094.i -quiet -dumpbase hello.c -version -o /tmp/cc000094.s GNU C version 2.7.2.2 (80386, BSD syntax) compiled by GNU C version 2.7.2.2. include /vsta/lib/lib/include /usr/include End of search list. cc1 /tmp/cc000094.i -quiet -dumpbase hello.c -version -o /tmp/cc000094.s GNU C version 2.7.2.2 (80386, BSD syntax) compile by GNU C version 2.7.2.2. perm I don't know what cc1 does, but it doesn't happen - I get a 0 byte .s file when I run it by hand. Can someone give me insight as to what is happening? btw- I searched the archives and there was something about unmounting the tmpfs so I just knocked it out of etc/fstab. doing a cc -c hello.c doesn't help either. -ray From daemon Tue Oct 6 09:33:16 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id JAA21942 for vsta-xplod; Tue, 6 Oct 1998 09:33:16 GMT Received: from zipper.cisco.com (zipper.cisco.com [171.69.63.31]) by bodhi.zendo.com (8.8.5/8.6.12) with ESMTP id JAA21936 for ; Tue, 6 Oct 1998 09:33:13 GMT Received: from vandys-pc.cisco.com (vandys-pc.cisco.com [171.69.51.26]) by zipper.cisco.com (8.8.5-Cisco.2-SunOS.5.5.1.sun4/8.6.5) with ESMTP id NAA22372; Tue, 6 Oct 1998 13:34:51 -0700 (PDT) Received: from vandys-pc.cisco.com (localhost.cisco.com [127.0.0.1]) by vandys-pc.cisco.com (8.8.5/8.8.5) with ESMTP id NAA03367; Tue, 6 Oct 1998 13:34:51 -0700 (PDT) Message-Id: <199810062034.NAA03367@vandys-pc.cisco.com> To: rayr@rramos.ucsc.edu cc: vsta@zendo.com Subject: Re: compile problems In-reply-to: Your message of "Mon, 05 Oct 1998 21:42:02 MDT." <199810060442.VAA15792@rramos.ucsc.edu> Date: Tue, 06 Oct 1998 13:34:50 -0700 From: Andy Valencia [rayr@rramos.ucsc.edu writes:] >I don't know what cc1 does, but it doesn't happen - I get a 0 byte >.s file when I run it by hand. Does the input ".i" look OK? Can you run cc1 under gdb and see if it exits normally? Also, can you feed some simple (no CPP constructs) C source into cc1 and see if, in general, it can turn C into assembly? Andy From daemon Tue Oct 6 21:20:18 1998 Received: (from daemon@localhost) by bodhi.zendo.com (8.8.5/8.6.12) id VAA24287 for vsta-xplod; Tue, 6 Oct 1998 21:20:18 GMT Received: from home.chat.net (genesis.jeske.meer.net [204.94.139.85]) by bodhi.zendo.com (8.8.5/8.6.12) with SMTP id VAA24281 for ; Tue, 6 Oct 1998 21:20:16 GMT Received: (qmail 10660 invoked by uid 120); 7 Oct 1998 08:25:39 -0000 Message-ID: <19981007012538.D10557@home.chat.net> Date: Wed, 7 Oct 1998 01:25:38 -0700 From: David Jeske To: vsta@zendo.com Subject: Re: MADO and Graphics Mail-Followup-To: vsta@zendo.com References: <19981003124526.K22462@home.chat.net> <199810032049.NAA28371@vandys-pc.cisco.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.92.1 In-Reply-To: <199810032049.NAA28371@vandys-pc.cisco.com>; from Andy Valencia on Sat, Oct 03, 1998 at 01:49:18PM -0700 On Sat, Oct 03, 1998 at 01:49:18PM -0700, Andy Valencia wrote: > [David Jeske writes:] > > >Andy: if a server setup a shared memory segment on 'physically mapped > >pages', and then a client connects to it, is it currently possible for > >the server to change the physical pages that shared memory segment > >maps to after the fact? > > Hmmm, not really, since there's not (currently) any good way to name a > shared object distinct from naming what it maps. I think it'd require a new > page set (struct pset, psop_* stuff) object type. In fact, I think it would > have to be a lot like a ZFOD set, with an operation to switch from the > backing store memory set to a physical map set. > > >I believe this would be required for GGI to work, because the server > >needs to be able to switch the client mapping from the 'real > >framebuffer' to a 'backing store' without the client knowing about it. > > Delta some interesting race conditions. Switching the mapping of the page > set isn't enough, because you also have to hammer the client's PTE's and TLB > (if they're running when you do the switch). I think some of the current > page stealing code would have to be generalized to support this. > > I'll do the VM work if somebody actually wants to try a GGI port! I'll have to take another look at the current GGI stuff.. last I looked, their stuff wasn't too ingrained into the 'Linux kernel view of the world', but that was when they were doing proof of concept. Things have certainly changed since then. -- David Jeske (N9LCA) + http://www.chat.net/~jeske/ + jeske@chat.net