From owner-amigappc-dev Mon Mar 9 17:43:17 1998 Received: from haigate.haitec.de (haigate.haitec.de [195.27.6.201]) by amiga.com.pl (8.7.5/8.7.3) with ESMTP id RAA16775 for ; Mon, 9 Mar 1998 17:36:53 +0100 (MET) From: hwhippenstiel@haitec.de Received: from hai1isdn.ueb.haitec.de (hai1isdn.ueb.haitec.de [194.55.3.201]) by haigate.haitec.de (8.8.7/8.8.7) with ESMTP id RAA20866 for ; Mon, 9 Mar 1998 17:38:00 +0100 Received: from hailg-ts03.haitec.de (hailg-ts03 [194.55.3.109]) by hai1isdn.ueb.haitec.de (8.8.7/8.8.7) with SMTP id RAA14763 for ; Mon, 9 Mar 1998 17:30:59 +0100 Received: by hailg-ts03.haitec.de(Lotus SMTP MTA SMTP v4.6 (462.2 9-3-1997)) id 412565C2.005AEC90 ; Mon, 9 Mar 1998 17:33:08 +0100 X-Lotus-FromDomain: HAITEC To: "AmigaPPC developers' mailing list" Message-ID: Date: Mon, 9 Mar 1998 17:36:29 +0100 Subject: Unsubscribe Mime-Version: 1.0 Content-type: text/plain; charset=us-ascii Unsubscribe From owner-amigappc-dev Wed Mar 11 21:13:09 1998 Received: from server.fiber.net.pl (server.fiber.net.pl [195.117.208.101]) by amiga.com.pl (8.7.5/8.7.3) with ESMTP id UAA27101 for ; Wed, 11 Mar 1998 20:48:07 +0100 (MET) Received: from server.fiber.net.pl (pppa247.szczecin.tpnet.pl [194.204.141.247]) by server.fiber.net.pl (Netscape Mail Server v2.02) with SMTP id AAB20779 for ; Wed, 11 Mar 1998 20:46:41 +0100 From: Marcin Orlowski To: "[WFMH] AmigaPPC-Dev ML" Date: Wed, 11 Mar 1998 10:40:53 +0100 Message-ID: X-Mailer: YAM 2.0.0 beta10 [020] - Amiga Mailer by Marcel Beck - http://www.yam.ch Organization: W.F.M.H. (http://wfmh.org.pl) Subject: Aministrator's note - please read this! MIME-Version: 1.0 Content-Type: text/plain Hi, Please note, that when you receive _removal_ confirmation from the listserv some day, and it wasn't your will to be removed, check who has send the removal request (his email is written in the confirmation body). If there's me (or anyone of WFMH staff, or the listowner himself) it means your email address/server started to produce too many "bounce" errors or "quota exceeded" reports or any other simmilar troubles while receiving ML messages. As almost every mail server sends information about the problems it encounters back to the sender, it causes dozens of mails and unnecesary increases the bandwitch. So if you wasn't kicked nor told to go away, just fix the cause of mail problems and simply resubscribe. You _don't_ need to ask nor contact listowner first. Cheers, -- My software never has bugs. It just develops random features. Marcin http://wfmh.org.pl/~carlos/ mailto:carlos@wfmh.org.pl From owner-amigappc-dev Thu Mar 12 02:13:10 1998 Received: from shell.wco.com (root@shell.wco.com [199.4.94.16]) by amiga.com.pl (8.7.5/8.7.3) with ESMTP id CAA29650 for ; Thu, 12 Mar 1998 02:11:25 +0100 (MET) Received: from paix-alg-gw1-11.ncal.verio.com (fw@paix-alg-gw1-11.ncal.verio.com [207.21.130.202]) by shell.wco.com (8.8.5/8.8.5/WCO-18jul97) with SMTP id RAA11581 for ; Wed, 11 Mar 1998 17:08:53 -0800 (PST) Date: Wed, 11 Mar 1998 17:08:41 -0800 (PST) From: Fred Wright X-Sender: fw@paix-alg-gw1-11.ncal.verio.com Reply-To: Fred Wright To: amigappc-dev@amiga.com.pl Subject: Test Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Test message - please ignore. From owner-amigappc-dev Thu Mar 12 04:13:12 1998 Received: from mail-r1.shlink.de (mail-r1.shlink.de [194.64.6.19]) by amiga.com.pl (8.7.5/8.7.3) with SMTP id DAA29755 for ; Thu, 12 Mar 1998 03:43:24 +0100 (MET) Received: from mail.shlink.de (194.64.6.3) by mail-r1.shlink.de with smtp (Smail3.2.0.95 #1) id m0yCxvR-000141C; Thu, 12 Mar 1998 03:40:53 +0100 (MET) To: amigappc-dev@amiga.com.pl Message-ID: <1BmG6MD5131aUz1@ark.news.wwbnet.de> From: ARK@News.wwbnet.de (Andreas R. Kleinert) Path: apg.wwbnet.de!News.wwbnet.de Organization: PerSuaSiVe SoftWorX Subject: Re: SteveK? Date: Wed, 11 Mar 1998 01:00:00 +0100 X-Mailer: MicroDot 1.17beta3 [REGISTERED 005131] References: <19980308.83BD300.8F41@spode.mindspring.net> Received: by News.wwbnet.de ; Wed, 11 Mar 1998 12:07:18 +0100 Received: by apg.wwbnet.de ; Wed, 11 Mar 1998 19:16:49 +0100 X-Gateway: ZCONNECT apg.wwbnet.de [UNIX/Connect v0.76b1-t1] X-Comment-To: Steve Krueger X-ZC-TELEFON: 0271-22869 (PHONE/AM/FAX) - NO BBS X-ZC-POST: Sandstrasse 1, D-57072 Siegen, Germany. You wrote at 08.03.1998/11:57:17: > I have an alternative page setup now. > www.thule.no/~stevek the old page seems to work again, too now :-) (it's just, that it still carries beta 17 ;-) -- Andreas_Kleinert@t-online.de | ARK@amigaworld.com | Fido 2:2457/350.18 ARK@superview.ftn.neckar-alb.de | ARK@News.wwbnet.de | - *SuperView* - PerSuaSiVe SoftWorX - http://home.t-online.de/home/Andreas_Kleinert/ Image Engineer Registration Site Europe --- Software Engineering -- MicroDot 1.17beta3 registered From owner-amigappc-dev Fri Mar 13 09:13:13 1998 Received: from mh2.cts.com (root@mh2.cts.com [205.163.24.68]) by amiga.com.pl (8.7.5/8.7.3) with ESMTP id IAA04970 for ; Fri, 13 Mar 1998 08:47:47 +0100 (MET) Received: from king.cts.com (root@king.cts.com [198.68.168.21]) by mh2.cts.com (8.8.7/8.8.5) with ESMTP id XAA28784; Thu, 12 Mar 1998 23:45:24 -0800 (PST) Received: from crash.cts.com (root@crash.cts.com [192.188.72.17]) by king.cts.com (8.8.7/8.8.7) with SMTP id XAA13597; Thu, 12 Mar 1998 23:45:20 -0800 (PST) Received: from outside.cts.com by crash.cts.com with smtp (Smail3.1.29.1 #5) id m0yDP9a-0001pJC; Thu, 12 Mar 98 23:45 PST Message-Id: X-SMTP-Client: VoodooSMTP_AmiTCP 1.5 by Finale Development, Inc. From: "David A. Newman" Reply-To: dan@cts.com MIME-Version: 1.0 X-Mailer: Serious Voodoo 1.388 (Registered to David A Newman ) In-Reply-To: To: amigappc-dev@amiga.com.pl Cc: PowerUP@lists.best.com Subject: Re: ppc-amigaos-ld problems Date: Thu, 12 Mar 1998 23:41:58 (-0800) Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Hello All, > I'm not having problems developing for PPC other than the linker crashes for me very regularly (slowing things down). > Unfortunately GCC and SAS/C use the same linker, so the problem is still around. So does anyone know why ppc-amigaos-ld crashes > and invalidates my hard drive so often? This is part of old message of my own, and this condition is still screwing things up for me. The machine just freeze 50% of the time when running ppc-amiagos-ld, there is no info from CyberGuard, and snoopdos shows little. Note: why does Snoopdos3.0 prevent SCPPC from compiling? Snoopdos 1.5 reports ppc-amiagos-ld Open '*' when it crashes. Anyone have any ideas. cya -- David A Newman dan@cts.com (Home) Applied Magic Inc dan@amagic.com (Work) Producer Tools http://www.free.cts.com/crash/d/dan/ Graffiti Wall http://www.sorceror.com/thewall.html From owner-amigappc-dev Fri Mar 13 18:13:26 1998 Received: from mb05.swip.net (mb05.swip.net [193.12.122.209]) by amiga.com.pl (8.7.5/8.7.3) with ESMTP id SAA06646 for ; Fri, 13 Mar 1998 18:08:23 +0100 (MET) Received: from hehe.com (dialup154-4-11.swipnet.se [130.244.154.203]) by mb05.swip.net (8.8.8/8.8.8) with SMTP id SAA14611 for ; Fri, 13 Mar 1998 18:06:03 +0100 (MET) From: Linus McCabe Reply-To: sparkle@hehe.com To: "AmigaPPC developers' mailing list" Date: Fri, 13 Mar 1998 18:00:02 +0200 Message-ID: In-Reply-To: X-Mailer: YAM 2.0.0 Preview [020] - Amiga Mailer by Marcel Beck - http://www.yam.ch Organization: schmuck universe Subject: Re[2]: ppc-amigaos-ld problems MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable Den 13-Mar-98, skrev David A. Newman: > = > Note: why does Snoopdos3.0 prevent SCPPC from compiling? I cannot use SnoopDos any more.. At every OpenLibrary, Snoopdos prevents = the library to open, 'freezing' the task in a 'wait for openlibrary to return= ' state. Is anyone using Snoopdos without problems w/=A0their PuP system, or is it= something else that's troubeling snoopdos for me?? > Snoopdos 1.5 reports ppc-amiagos-ld Open '*' when it crashes. > = > Anyone have any ideas. Sorry, no. > cya > -- > David A Newman dan@cts.com (Home) /Linus -- = Linus McCabe * Sparkle@hehe.com = * www.come.to/Sparkle * Sparkle, #Snoopy, DalNet From owner-amigappc-dev Sun Mar 15 15:24:48 1998 Received: from kdn0.attnet.or.jp (kdn0.attnet.or.jp [165.76.124.5]) by amiga.com.pl (8.7.5/8.7.3) with ESMTP id OAA01155 for ; Sun, 15 Mar 1998 14:59:20 +0100 (MET) Received: from 30.gate1.kadena.attnet.or.jp (mclaughd@localhost [165.76.124.109]) by kdn0.attnet.or.jp (8.8.8+Spin/3.6Wbeta7-CONS(11/18/97)) id WAA27069; Sun, 15 Mar 1998 22:58:43 +0900 (JST) From: "Douglas McLaughlin" Organization: STR Programming Services Date: 15 Mar 98 23:03:43 +0900 Subject: Re: ppc-amigaos-ld problems Message-Id: <350C5E4F.MD-0.214.mclaughd@kdn0.attnet.or.jp> In-Reply-To: To: "AmigaPPC developers' mailing list" Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit X-Mailer: MicroDot-II/AmigaOS 0.214 On Fri, 13 Mar 1998 18:00:02 +0200, Linus McCabe had this to say about Re[2]: ppc-amigaos-ld problems... > > Note: why does Snoopdos3.0 prevent SCPPC from compiling? > > I cannot use SnoopDos any more.. At every OpenLibrary, Snoopdos prevents the > library to open, 'freezing' the task in a 'wait for openlibrary to return' > state. > > Is anyone using Snoopdos without problems w/ their PuP system, or is it > something else that's troubeling snoopdos for me?? > > > Snoopdos 1.5 reports ppc-amiagos-ld Open '*' when it crashes. > > > > Anyone have any ideas. > > Sorry, no. Same here. We've discussed this before, but nothing has come of it yet. There is a connection with the PPC software and SnoopDOS's tracking of OpenLibrary(). I don't think anyone know what it is yet. Doug From owner-amigappc-dev Sun Mar 15 16:54:55 1998 Received: from nevald.k-net.dk (nevald.k-net.dtu.dk [130.225.71.226]) by amiga.com.pl (8.7.5/8.7.3) with SMTP id QAA01451 for ; Sun, 15 Mar 1998 16:37:25 +0100 (MET) Received: (qmail 8015 invoked from network); 15 Mar 1998 15:36:42 -0000 Received: from carlsberg.kampsax.dtu.dk (qmailr@192.38.212.2) by nevald.k-net.dtu.dk with SMTP; 15 Mar 1998 15:36:42 -0000 Received: (qmail 450 invoked from network); 15 Mar 1998 15:36:40 -0000 Received: from k4315.kampsax.dtu.dk (rask@192.38.215.253) by carlsberg.kampsax.dtu.dk with SMTP; 15 Mar 1998 15:36:40 -0000 Received: from k4315.kampsax.dtu.dk (rask@192.38.215.253) by k4315.kampsax.dtu.dk with SMTP; 15 Mar 1998 15:33:22 -0000 Date: 15 Mar 98 16:30:20 +0100 From: "Rask Ingemann Lambertsen" Subject: Asynchronous I/O (Was: AmigaAMP M68K/PPC - Strange Feature!) To: "A mailing list for discussion of phase5 PowerUp boards" In-Reply-To: <199803121756.JAA29276@lists1.best.com> Message-ID: <1186.378T1009T9904081@kampsax.k-net.dk> MIME-Version: 1.0 Content-type: text/plain; charset=iso-8859-1 Content-transfer-encoding: 8bit Organization: Me? Organised? Dream on... X-Files: The truth is out there. X-INUFS: This message was sent from an INetUtils Free System. Cc: amigappc-dev@amiga.com.pl X-Mailer: THOR 2.5a (Amiga;TCP/IP) Den 12-Mar-98 18:14:25 skrev Thomas Wenzel fĝlgende om "Re: AmigaAMP M68K/PPC - Strange Feature!": > Well, at first I thought it must be quite stupid to use an asynchronous > shared 68k library for a PPC task. I think you're right. What we really need is a PowerUp version of asyncio.library (not necessarily a shared one at first). Not only would it pretty much solve the problems with interrupted audio, it would also reduce the rate of PPCCallOS() calls nicely as a side effect. What do you think, should we give it a shot (as opposed to waiting for someone else to do it)? Perhaps other people would like to join the show too? Regards, /ŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻTŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻ\ | Rask Ingemann Lambertsen | E-mail: mailto:rask@kampsax.k-net.dk | | Registered Phase5 developer | WWW: http://www.gbar.dtu.dk/~c948374/ | | A4000, 775 kkeys/s (RC5-64) | "ThrustMe" on XPilot, ARCnet and IRC | | LOAD "emacs",8,1 | From owner-amigappc-dev Mon Mar 16 01:54:48 1998 Received: from shell.wco.com (root@shell.wco.com [199.4.94.16]) by amiga.com.pl (8.7.5/8.7.3) with ESMTP id BAA02808 for ; Mon, 16 Mar 1998 01:52:58 +0100 (MET) Received: from paix-alg-gw1-19.ncal.verio.com (fw@paix-alg-gw1-19.ncal.verio.com [207.21.130.210]) by shell.wco.com (8.8.5/8.8.5/WCO-18jul97) with SMTP id QAA24674 for ; Sun, 15 Mar 1998 16:52:03 -0800 (PST) Date: Sun, 15 Mar 1998 16:51:50 -0800 (PST) From: Fred Wright X-Sender: fw@paix-alg-gw1-19.ncal.verio.com Reply-To: Fred Wright To: "AmigaPPC developers' mailing list" Subject: Re: ppc-amigaos-ld problems In-Reply-To: <350C5E4F.MD-0.214.mclaughd@kdn0.attnet.or.jp> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On 15 Mar 1998, Douglas McLaughlin wrote: > Same here. We've discussed this before, but nothing has come of it > yet. There is a connection with the PPC software and SnoopDOS's > tracking of OpenLibrary(). I don't think anyone know what it is yet. Aarrgh. That seems like a serious problem, considering how valuable SnoopDOS is. Does it perhaps relate to the LoadSeg patch? Fred Wright From owner-amigappc-dev Mon Mar 16 04:54:46 1998 Received: from kdn0.attnet.or.jp (kdn0.attnet.or.jp [165.76.124.5]) by amiga.com.pl (8.7.5/8.7.3) with ESMTP id EAA04728 for ; Mon, 16 Mar 1998 04:47:13 +0100 (MET) Received: from 44.gate3.kadena.attnet.or.jp (mclaughd@localhost [165.76.124.219]) by kdn0.attnet.or.jp (8.8.8+Spin/3.6Wbeta7-CONS(11/18/97)) id MAA05007; Mon, 16 Mar 1998 12:46:20 +0900 (JST) From: "Douglas McLaughlin" Organization: STR Programming Services Date: 16 Mar 98 12:51:21 +0900 Subject: Re: ppc-amigaos-ld problems Message-Id: <350D2049.MD-0.214.mclaughd@kdn0.attnet.or.jp> In-Reply-To: To: "AmigaPPC developers' mailing list" Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: MicroDot-II/AmigaOS 0.214 On Sun, 15 Mar 1998 16:51:50 -0800 (PST), Fred Wright had this to say about Re: ppc-amigaos-ld problems... > > Same here. We've discussed this before, but nothing has come of it > > yet. There is a connection with the PPC software and SnoopDOS's > > tracking of OpenLibrary(). I don't think anyone know what it is yet. > > Aarrgh. That seems like a serious problem, considering how valuable > SnoopDOS is. Does it perhaps relate to the LoadSeg patch? No. I don't have the LoadSeg patch installed. Anyway, I noticed the problem back when I first got my PowerUp card. Waaaay back in September (or thereabouts.) Doug From owner-amigappc-dev Mon Mar 16 06:54:46 1998 Received: from rock101.genie.net (genie.com [199.164.140.11]) by amiga.com.pl (8.7.5/8.7.3) with ESMTP id GAA04827 for ; Mon, 16 Mar 1998 06:43:10 +0100 (MET) Received: from rock103.genie.net (rock103.is.ge.com [204.90.131.103]) by rock101.genie.net (8.8.8/8.8.8) with ESMTP id FAA23789 for ; Mon, 16 Mar 1998 05:42:28 GMT Received: (from uucp@localhost) by rock103.genie.net (8.8.8/8.8.8) id FAA29931 for amigappc-dev@amiga.com.pl; Mon, 16 Mar 1998 05:42:27 GMT From: k.barkley@genie.geis.com Message-Id: <199803160542.FAA29931@rock103.genie.net> Received: by (genie.)rock103.genie.net ( 2rem/1.44 ) ; Mon, 16 Mar 98 05:42:27 GMT ( from inet# ; Mon, 16 Mar 98 05:41:17 GMT ) Date: Mon, 16 Mar 98 05:14:00 GMT To: amigappc-dev@amiga.com.pl X-genie-QK-From: K.BARKLEY X-genie-QK-Id: 8443443 X-genie-Gateway-Id: 800837 Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii" Subject: Using libmoto.a with SAS/C PPC Like the subject says. since ELF is the standard will libmoto.a work like it does with gnu? Or is it un-neccesary since Steve already incorporated it into the SAS/C Math libraries? 8^) My software uses lots of ffp math calls and can use the speed up. k.barkley@genie.com (Yes Genie, my main email address got me kicked off the list.) barkleyk@macom.com From owner-amigappc-dev Mon Mar 16 09:24:54 1998 Received: from shell.wco.com (root@shell.wco.com [199.4.94.16]) by amiga.com.pl (8.7.5/8.7.3) with ESMTP id JAA05047 for ; Mon, 16 Mar 1998 09:08:42 +0100 (MET) Received: from paix-alg-gw1-3.ncal.verio.com (fw@paix-alg-gw1-3.ncal.verio.com [207.21.130.194]) by shell.wco.com (8.8.5/8.8.5/WCO-18jul97) with SMTP id AAA28643 for ; Mon, 16 Mar 1998 00:07:56 -0800 (PST) Date: Mon, 16 Mar 1998 00:07:46 -0800 (PST) From: Fred Wright X-Sender: fw@paix-alg-gw1-3.ncal.verio.com Reply-To: Fred Wright To: "AmigaPPC developers' mailing list" Subject: Re: ppc-amigaos-ld problems In-Reply-To: <350D2049.MD-0.214.mclaughd@kdn0.attnet.or.jp> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On 16 Mar 1998, Douglas McLaughlin wrote: > > No. I don't have the LoadSeg patch installed. Anyway, I noticed the > problem back when I first got my PowerUp card. Waaaay back in > September (or thereabouts.) OK, that was just a guess, since it seemed like the most obvious part of OpenLibrary() to interact with the PPC stuff. Another possibility is the funny signal interaction mentioned in the description of the [NO]PATCHRAMLIB options in SnoopDOS. Does it work any better with NOPATCHRAMLIB? I don't have my PowerUP board yet, so I can't try it myself. Fred Wright From owner-amigappc-dev Mon Mar 16 14:54:59 1998 Received: from spode (user-38ld8rc.dialup.mindspring.com [209.86.163.108]) by amiga.com.pl (8.7.5/8.7.3) with SMTP id OAA06066 for ; Mon, 16 Mar 1998 14:54:44 +0100 (MET) Received: by spode.mindspring.net (Amiga SMTPpost 1.04 December 9, 1994) id AA01; Mon, 16 Mar 98 08:52:09 Received: by spode.mindspring.net (Amiga SMTPpost 1.04 December 9, 1994) id AA01; Mon, 16 Mar 98 08:44:18 Subject: Re: Using libmoto.a with SAS/C PPC In-Reply-To: <199803160542.FAA29931@rock103.genie.net> from k.barkley@genie.geis.com at Mon, 16 Mar 98 05:14:00 GMT From: sekruege@mindspring.com (Steve Krueger) X-Mailer: GMail 0.55 (11.5.95) Date: Mon, 16 Mar 98 08:44:18 Message-ID: <19980316.83E1170.7E60@spode.mindspring.net> To: amigappc-dev@amiga.com.pl : Like the subject says. since ELF is the standard will libmoto.a work : like it does with gnu? Or is it un-neccesary since Steve already : incorporated it into the SAS/C Math libraries? 8^) : : My software uses lots of ffp math calls and can use the speed up. You can (and probably should :) ) use libmoto.a with SAS/C code. Motorola's routines are faster than the SAS versions. sk From owner-amigappc-dev Tue Mar 17 12:24:51 1998 Received: from kdn0.attnet.or.jp (kdn0.attnet.or.jp [165.76.124.5]) by amiga.com.pl (8.7.5/8.7.3) with ESMTP id LAA10785 for ; Tue, 17 Mar 1998 11:59:55 +0100 (MET) Received: from 28.gate1.kadena.attnet.or.jp (mclaughd@localhost [165.76.124.107]) by kdn0.attnet.or.jp (8.8.8+Spin/3.6Wbeta7-CONS(11/18/97)) id TAA24095; Tue, 17 Mar 1998 19:58:59 +0900 (JST) From: "Douglas McLaughlin" Organization: STR Programming Services Date: 16 Mar 98 19:24:14 +0900 Subject: Re: ppc-amigaos-ld problems Message-Id: <350D7C5E.MD-0.214.mclaughd@kdn0.attnet.or.jp> In-Reply-To: To: "AmigaPPC developers' mailing list" Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: MicroDot-II/AmigaOS 0.214 On Mon, 16 Mar 1998 00:07:46 -0800 (PST), Fred Wright had this to say about Re: ppc-amigaos-ld problems... > > No. I don't have the LoadSeg patch installed. Anyway, I noticed the > > problem back when I first got my PowerUp card. Waaaay back in > > September (or thereabouts.) > > OK, that was just a guess, since it seemed like the most obvious part of > OpenLibrary() to interact with the PPC stuff. Another possibility is the > funny signal interaction mentioned in the description of the > [NO]PATCHRAMLIB options in SnoopDOS. Does it work any better with > NOPATCHRAMLIB? I don't have my PowerUP board yet, so I can't try it > myself. I will give it a try and get back to you... Doug From owner-amigappc-dev Thu Mar 19 14:49:58 1998 Received: from mail-r1.shlink.de (mail-r1.shlink.de [194.64.6.19]) by amiga.com.pl (8.7.5/8.7.3) with SMTP id OAA07079 for ; Thu, 19 Mar 1998 14:34:16 +0100 (MET) Received: from mail.shlink.de (194.64.6.3) by mail-r1.shlink.de with smtp (Smail3.2.0.95 #1) id m0yFfRQ-00016dC; Thu, 19 Mar 1998 14:33:04 +0100 (MET) To: amigappc-dev@amiga.com.pl Message-ID: <1EM1GMD5131aUz1@ark.news.wwbnet.de> From: ARK@News.wwbnet.de (Andreas R. Kleinert) Path: apg.wwbnet.de!News.wwbnet.de Organization: PerSuaSiVe SoftWorX Subject: Re: ppc-amigaos-ld problems Date: Thu, 19 Mar 1998 01:00:00 +0100 X-Mailer: MicroDot 1.17beta3 [REGISTERED 005131] References: Received: by News.wwbnet.de ; Thu, 19 Mar 1998 07:56:45 +0100 Received: by apg.wwbnet.de ; Thu, 19 Mar 1998 09:57:07 +0100 X-Gateway: ZCONNECT apg.wwbnet.de [UNIX/Connect v0.76b1-t1] X-Comment-To: Fred Wright X-ZC-TELEFON: 0271-22869 (PHONE/AM/FAX) - NO BBS X-ZC-POST: Sandstrasse 1, D-57072 Siegen, Germany. You wrote at 16.03.1998/-6:51:50: > Aarrgh. That seems like a serious problem, considering how valuable > SnoopDOS is. Does it perhaps relate to the LoadSeg patch? No. -- Andreas_Kleinert@t-online.de | ARK@amigaworld.com | Fido 2:2457/350.18 ARK@superview.ftn.neckar-alb.de | ARK@News.wwbnet.de | - *SuperView* - PerSuaSiVe SoftWorX - http://home.t-online.de/home/Andreas_Kleinert/ Image Engineer Registration Site Europe --- Software Engineering -- MicroDot 1.17beta3 registered From owner-amigappc-dev Thu Mar 19 15:19:58 1998 Received: from mail-r1.shlink.de (mail-r1.shlink.de [194.64.6.19]) by amiga.com.pl (8.7.5/8.7.3) with SMTP id OAA07138 for ; Thu, 19 Mar 1998 14:53:02 +0100 (MET) Received: from mail.shlink.de (194.64.6.3) by mail-r1.shlink.de with smtp (Smail3.2.0.95 #1) id m0yFfjT-00016TC; Thu, 19 Mar 1998 14:51:43 +0100 (MET) To: amigappc-dev@amiga.com.pl Message-ID: <1EM4UMD5131asz2@ark.news.wwbnet.de> From: ARK@News.wwbnet.de (Andreas R. Kleinert) Path: apg.wwbnet.de!News.wwbnet.de Organization: PerSuaSiVe SoftWorX Subject: Re: Asynchronous I/O (Was: AmigaAMP M68K/PPC - Strange Feature!) Date: Thu, 19 Mar 1998 01:00:00 +0100 X-Mailer: MicroDot 1.17beta3 [REGISTERED 005131] References: <1186.378T1009T9904081@kampsax.k-net.dk> Received: by News.wwbnet.de ; Thu, 19 Mar 1998 07:56:45 +0100 Received: by apg.wwbnet.de ; Thu, 19 Mar 1998 09:57:07 +0100 X-Gateway: ZCONNECT apg.wwbnet.de [UNIX/Connect v0.76b1-t1] X-Comment-To: Rask Ingemann Lambertsen X-ZC-TELEFON: 0271-22869 (PHONE/AM/FAX) - NO BBS X-ZC-POST: Sandstrasse 1, D-57072 Siegen, Germany. You wrote at 15.03.1998/18:30:20: > > Well, at first I thought it must be quite stupid to use an asynchronous > > shared 68k library for a PPC task. > > I think you're right. What we really need is a PowerUp version of > asyncio.library (not necessarily a shared one at first). Not only would it > pretty much solve the problems with interrupted audio, it would also reduce > the rate of PPCCallOS() calls nicely as a side effect. What do you think, > should we give it a shot (as opposed to waiting for someone else to do it)? > Perhaps other people would like to join the show too? I've a half-ported version here. The only problem is, that SeekAsync() does not seem to work as it should. Since there's a lot of packet-sending and cache-flushing involved, it probably would make *more* sense to instead generate an PPC interface for calling the 68k asyncio.library (so just the includes). It still should be faster than dos.library. -- Andreas_Kleinert@t-online.de | ARK@amigaworld.com | Fido 2:2457/350.18 ARK@superview.ftn.neckar-alb.de | ARK@News.wwbnet.de | - *SuperView* - PerSuaSiVe SoftWorX - http://home.t-online.de/home/Andreas_Kleinert/ Image Engineer Registration Site Europe --- Software Engineering -- MicroDot 1.17beta3 registered From owner-amigappc-dev Thu Mar 19 18:19:58 1998 Received: from mopo.cc.lut.fi (tsuikki@mopo.cc.lut.fi [157.24.10.3]) by amiga.com.pl (8.7.5/8.7.3) with ESMTP id RAA07802 for ; Thu, 19 Mar 1998 17:55:49 +0100 (MET) Received: (from tsuikki@localhost) by mopo.cc.lut.fi (8.8.8/8.7.6/1.17.sn) id SAA24730; Thu, 19 Mar 1998 18:55:15 +0200 (EET) (for amigappc-dev@amiga.com.pl) From: Teemu Suikki Message-Id: <199803191655.SAA24730@mopo.cc.lut.fi> Subject: PPC devices In-Reply-To: <1EM4UMD5131asz2@ark.news.wwbnet.de> from "Andreas R. Kleinert" at "Mar 19, 98 01:00:00 am" To: amigappc-dev@amiga.com.pl Date: Thu, 19 Mar 1998 18:55:15 +0200 (EET) X-Mailer: ELM [version 2.4ME+ PL27 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit I have been thinking how devices should be implemented in PowerUp.. This article is posted to csa.programmers and PowerUP developer list. First, some background: On 68k, devices have a library interface with functions for OpenDevice, CloseDevice, BeginIO and AbortIO. But, most devices have a process that handles the iorequest, and BeginIO simply does PutMsg() to queue the request.. AbortIO removes requests from this list. In my opinion the API would be more clear and simple if OS would take care of BeginIO/AbortIO, and devices simply would have a port where requests are sent. So, here's the suggestion part. :) PPC device could be a simple ELF executable, which is executed when the device is opened. The device would then create a public named port, and start waiting for requests. When an application wants to open a device, it does the following: - Check if port "devicename/unit" exists. - If not, execute file devs:devicename, and the port becomes available. - Send a startup message to the device. This is similar to iorequest passed to OpenDevice; device can fill in the fields and increment its open count. Then, the application simply sends messages to the port, with PPCSendMessage(). When it wants to close the device, it sends a cleanup message, which tells device that it is no longer used and it can decrement the open count. The above actions could be implemented as PPCOpenDevice and PPCCloseDevice calls in PowerUp kernel, but interestingly enough, all above is possible with the _current_ ppc.library with no special support. But of course it would be better to have these as kernel functions. For AbortIO functionality, we would need a PPCAbortMessage function.. This should be quite simple to add to the kernel, just remove the message from the queue and reply it. If the message is already replied or is not on a port, just return without doing anything.. About devices that have library functions in them.. Anyone ever wanted to use timer.device functions like ReadEClock? Remember all the trouble creating iorequest and msgport which really aren't going to be used anywhere, you just want to call that darn function. Using the above style devices, a standard ELF library can simply create a msgport on its startup code, and suddenly it has a device API too! An application can use either device or library API without knowing that the other API even exists. I'm not sure which type of iorequest structure should be used.. 68k iorequest could be used just for clarity. This would also have the added benefit that you could create a very simple 68k library stub with BeginIO etc, that just feeds requests to the PPC port with PPCSendMessage.. So the PPC device could be used in 68k programs too. Tell me what you think.. And please note that this is just something I thought when I couldn't get sleep last night. :) -- int W,Y;main(){char*D="X`>@X@@~~`#~X~~##`##> "//"tsuikki@lut.fi IRC:Zuikkis" "# `>?X~#88` XC# ` #X## ` ~X~~ `@@",*S="\n""Petsoff Limited Partnership" ;for(Y=0;*++D;Y++)for(W=256;W/=2;)putchar((*D^64)&W?S[Y&7]&95:32);puts(S+8);} From owner-amigappc-dev Sat Mar 21 19:19:37 1998 Received: from mx1.polbox.com (mx1.polbox.com [195.116.5.16]) by amiga.com.pl (8.7.5/8.7.3) with ESMTP id TAA01152; Sat, 21 Mar 1998 19:07:12 +0100 (MET) Received: from polbox.com (neon@ppp133.radom.tpnet.pl [195.116.243.133]) by mx1.polbox.com (8.8.8/rev-A0) with SMTP id TAA02790; Sat, 21 Mar 1998 19:11:03 +0100 From: Kamil =?iso-8859-1?Q?Drzewi=EFski?= Reply-To: Kamil =?iso-8859-1?Q?Drzewi=EFski?= To: AmigaPL , CGX , PPC_PL , R3D , Marek Pampuch , Forum3D Date: Sat, 21 Mar 1998 19:02:11 +0100 Message-ID: X-Mailer: YAM 1.3.5 [040] - Amiga Mailer by Marcel Beck Subject: Lightwave for Amiga is going on!!! (fwd) MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable *** Wiadomo=F4=EA przeniesiona, oryginalnie napisana przez Michael Burkha= rdt dn. 21-Mar-98 *** Hi list-members, a good news from Petro Tyschtschenko: Lightwave is going to be developed again for Amiga! This is the result of a meeting of Petro and Newtek. They told him that Dweight, who is against Amiga is fired. Now Don Pence is our man. He will continue Lightwave developing and updating. Now we should ask Don what are the further plans for the Amiga. It is a very big chance for the Amiga. And it could be now possible to get PPC-Lightwave because AMIGA International Inc. has decided for 680x0 + PPC dual processor boards from Phase 5 for further Amigas. It should be our and the journalist's job to ask Don for all news we can get. I have mailed him and I hope he will answer fast. With a Lightwave which is the same on Amiga and PC there is the chance to get out of a forgotten hole. Therefore I think we should all work for and believe to Petro Tyschtschenko. He is a great man who want's to get back the market Commodore lost. And Lightwave could help him. I hope my english isn't as bad as I think myself. Please ask Don at Newtek: don_pence@newtek.com Here is the original message from Petro (German): >Michael, Ich habe gute Neuigkeiten. Bei NewTek gab es personelle >Ver=E4nderung. Dweight, der AMIGA geblockt hat ist gefeuert worden und die ganze >Firma atmet auf. >Don Pence, von NewTek war in St. Louis und ich hatte sehr interessante >Gespr=E4che mit Ihm. NewTek wird wieder f=FCr AMIGA entwickeln und updaten.... > >N=E4heres k=F6nnen Sie gerne von Don selbst erfahren....don_pence@newtek.com > >Bis bald Petro Tyschtschenko Hopefull greetings, -- Michael Burkhardt IOM Film- und Videoproduktion Hof + Mittweida Tel.: 09281-65325 ... 03727-601261 Fax: 09281-64705 ... 03727-601267 Hotline: 0171-3296371 ... email: iom@hof.de -- = ----------------------------------------------- Kamil Drzewinski, kgfx@polbox.com / Team Amiga = GSM: +48 602773956 ----------------------------------------------- ----- [ REKLAMA / ADVERTISEMENT ] ---------------------------------------- Lubisz film?! Wiec na co czekasz? Przylacz sie do FILMWEBU! Pierwszy Polski Serwis Filmowy w Internecie http://www.filmweb.pl -------------------------------------------------------------------------- From owner-amigappc-dev Sat Mar 21 21:20:00 1998 Received: from dns1.supermedia.pl (dns1.supermedia.pl [195.116.124.17]) by amiga.com.pl (8.7.5/8.7.3) with SMTP id UAA01524 for ; Sat, 21 Mar 1998 20:53:46 +0100 (MET) Received: from ppp-cst94.warszawa.tpnet.pl (ppp-cst94.warszawa.tpnet.pl [194.204.131.94]) by dns1.supermedia.pl (NTMail 3.03.0014/4c.aamn) with ESMTP id ga460674 for ; Sat, 21 Mar 1998 20:49:59 +0100 From: Peter Drapich Reply-To: Peter Drapich To: Lista dyskusyjna AmigaPL , Lista dyskusyjna WFMH -AmigaPL , Lista dyskusyjna Amiga PowerPC-Dev Date: Sat, 21 Mar 1998 08:31:39 +0100 Message-ID: X-Mailer: YAM 1.3.4 [020] - Amiga Mailer by Marcel Beck Organization: Union Systems Subject: Lightwave for Amiga is going on!!! (fwd) MIME-Version: 1.0 Content-Type: text/plain *** Forwarded message, originally written by Michael Burkhardt on 21-Mar-98 *** Hi list-members, a good news from Petro Tyschtschenko: Lightwave is going to be developed again for Amiga! This is the result of a meeting of Petro and Newtek. They told him that Dweight, who is against Amiga is fired. Now Don Pence is our man. He will continue Lightwave developing and updating. Now we should ask Don what are the further plans for the Amiga. It is a very big chance for the Amiga. And it could be now possible to get PPC-Lightwave because AMIGA International Inc. has decided for 680x0 + PPC dual processor boards from Phase 5 for further Amigas. It should be our and the journalist's job to ask Don for all news we can get. I have mailed him and I hope he will answer fast. With a Lightwave which is the same on Amiga and PC there is the chance to get out of a forgotten hole. Therefore I think we should all work for and believe to Petro Tyschtschenko. He is a great man who want's to get back the market Commodore lost. And Lightwave could help him. I hope my english isn't as bad as I think myself. Please ask Don at Newtek: don_pence@newtek.com *** End of forwarded message *** The future is brighter... Docent -- When the gods want to punish you, they answer your prayers. - O. Wilde From owner-amigappc-dev Sun Mar 22 18:19:57 1998 Received: from nevald.k-net.dk (nevald.k-net.dtu.dk [130.225.71.226]) by amiga.com.pl (8.7.5/8.7.3) with SMTP id SAA05732 for ; Sun, 22 Mar 1998 18:17:02 +0100 (MET) Received: (qmail 9407 invoked from network); 22 Mar 1998 17:15:49 -0000 Received: from carlsberg.kampsax.dtu.dk (qmailr@192.38.212.2) by nevald.k-net.dtu.dk with SMTP; 22 Mar 1998 17:15:49 -0000 Received: (qmail 19950 invoked from network); 22 Mar 1998 17:15:49 -0000 Received: from k4315.kampsax.dtu.dk (rask@192.38.215.253) by carlsberg.kampsax.dtu.dk with SMTP; 22 Mar 1998 17:15:49 -0000 Received: from k4315.kampsax.dtu.dk (rask@192.38.215.253) by k4315.kampsax.dtu.dk with SMTP; 22 Mar 1998 17:12:26 -0000 Date: 22 Mar 98 15:22:00 +0100 From: "Rask Ingemann Lambertsen" Subject: Re: Asynchronous I/O (Was: AmigaAMP M68K/PPC - StrangeFeature!) To: "Amiga PowerPC developer mailing list" In-Reply-To: <1EM4UMD5131asz2@ark.news.wwbnet.de> Message-ID: <2504.385T45T9223269@kampsax.k-net.dk> MIME-Version: 1.0 Content-type: text/plain; charset=iso-8859-1 Content-transfer-encoding: 8bit Organization: Me? Organised? Dream on... X-Files: The truth is out there. X-INUFS: This message was sent from an INetUtils Free System. X-Mailer: THOR 2.5a (Amiga;TCP/IP) Den 19-Mar-98 01:00:00 skrev Andreas R. Kleinert fĝlgende om "Re: Asynchronous I/O (Was: AmigaAMP M68K/PPC - StrangeFeature!)": > You wrote at 15.03.1998/18:30:20: >> > Well, at first I thought it must be quite stupid to use an asynchronous >> > shared 68k library for a PPC task. >> >> I think you're right. What we really need is a PowerUp version of >> asyncio.library (not necessarily a shared one at first). Not only would it >> pretty much solve the problems with interrupted audio, it would also reduce >> the rate of PPCCallOS() calls nicely as a side effect. What do you think, >> should we give it a shot (as opposed to waiting for someone else to do it)? >> Perhaps other people would like to join the show too? > I've a half-ported version here. How about making it available for download somewhere? > The only problem is, that SeekAsync() does not seem to work as it should. I can live with that because I don't need it. > Since there's a lot of packet-sending and cache-flushing involved, it > probably would make *more* sense to instead generate an PPC interface > for calling the 68k asyncio.library (so just the includes). I'm not so sure. I have an application (mpg123, actually) which reads it's input file in 4 and 164 byte chunks. Reading at 12-14 kB/s this way means lots of calls to the '060, and while I'm sure PPCRead() only flushes/invalidates only as much as necessary, I think this app would be much better off getting 32-64 kB (or even more, user specifiable) at a time, even if this would mean a full cache flush/invalidate each time the buffer is filled. > It still should be faster than dos.library. Uhm, the main purpose of asyncronous I/O is to improve latency, not so much speed. Buffering file I/O on the PPC side would help speed too, though, and reduce the load on the '060. It would also make the PPC app less vulnerable to '060 load spikes and starvation of the PPC:MsgHandler process (sortof related to the latency problem). Regards, /ŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻTŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻ\ | Rask Ingemann Lambertsen | E-mail: mailto:rask@kampsax.k-net.dk | | Registered Phase5 developer | WWW: http://www.gbar.dtu.dk/~c948374/ | | A4000, 775 kkeys/s (RC5-64) | "ThrustMe" on XPilot, ARCnet and IRC | | There are no absolutes...that's the absolute truth! | From owner-amigappc-dev Sun Mar 22 18:20:04 1998 Received: from nevald.k-net.dk (nevald.k-net.dtu.dk [130.225.71.226]) by amiga.com.pl (8.7.5/8.7.3) with SMTP id SAA05731 for ; Sun, 22 Mar 1998 18:16:58 +0100 (MET) Received: (qmail 9404 invoked from network); 22 Mar 1998 17:15:47 -0000 Received: from carlsberg.kampsax.dtu.dk (qmailr@192.38.212.2) by nevald.k-net.dtu.dk with SMTP; 22 Mar 1998 17:15:47 -0000 Received: (qmail 19943 invoked from network); 22 Mar 1998 17:15:47 -0000 Received: from k4315.kampsax.dtu.dk (rask@192.38.215.253) by carlsberg.kampsax.dtu.dk with SMTP; 22 Mar 1998 17:15:47 -0000 Received: from k4315.kampsax.dtu.dk (rask@192.38.215.253) by k4315.kampsax.dtu.dk with SMTP; 22 Mar 1998 17:12:23 -0000 Date: 22 Mar 98 15:06:07 +0100 From: "Rask Ingemann Lambertsen" Subject: Re: Cache optimising inline files To: "Amiga PowerPC developer mailing list" In-Reply-To: <1026.359T1397T9973829@kampsax.k-net.dk> Message-ID: <2119.385T384T9063092@kampsax.k-net.dk> MIME-Version: 1.0 Content-type: text/plain; charset=iso-8859-1 Content-transfer-encoding: 8bit Organization: Me? Organised? Dream on... X-Files: The truth is out there. X-INUFS: This message was sent from an INetUtils Free System. X-Mailer: THOR 2.5a (Amiga;TCP/IP) Den 24-Feb-98 16:37:27 skrev Rask Ingemann Lambertsen fĝlgende om "Re: Cache optimising inline files": > Den 01-Feb-98 12:12:55 skrev Rask Ingemann Lambertsen fĝlgende om "Cache > optimising inline files": >> Hi. >> Has anybody had success optimising GCC's inline headers to avoid >> unnecessary cache flushing for OS calls that don't need it? So far, I've >> tried disabling cache flushes for dos.library/FPutC(),FGetC() and >> locale.library/IsXXX(), but then I start seeing crashes. > With ppc.library v45.13, FPutC() and FGetC() work now, but I'm still not > having much luck with IsXXX() from locale.library. And then, testing a port of mpg123 using libRILc, I found a situation where FPutC() calls would always crash because Caos.a6 would be zero (when viewed from the '060). After a talk with Ralph Schmidt on IRC, we came to the conclusion that the problem was that the Caos structure passed to PPCCallOS() was not being flushed/invalidated properly. It turns out that this structure _must_ be 32 byte aligned when passed to PPCCallOS() since ppc.library currently does not check for this. With a 32 byte aligned struct Caos, dos.library/FPutC() and locale.library/IsXXX() calls with IF_CACHEFLUSHNO set for the PPC and the m68k now work properly. One way of ensuring 32 byte alignment of the Caos structure is: struct Caos *MyCaos; MyCaos = (struct Caos *) ((unsigned long int) ((char *) alloca (sizeof (struct Caos) + 31) + 31) & ~31); You can download a replacement file from . Regards, /ŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻTŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻ\ | Rask Ingemann Lambertsen | E-mail: mailto:rask@kampsax.k-net.dk | | Registered Phase5 developer | WWW: http://www.gbar.dtu.dk/~c948374/ | | A4000, 775 kkeys/s (RC5-64) | "ThrustMe" on XPilot, ARCnet and IRC | | "Compatible": Gracefully accepts erroneous data from any source. | From owner-amigappc-dev Mon Mar 23 09:23:34 1998 Received: from localhost (carlos@localhost) by amiga.com.pl (8.7.5/8.7.3) with SMTP id JAA09072 for ; Mon, 23 Mar 1998 09:23:34 +0100 (MET) Date: Mon, 23 Mar 1998 09:23:33 +0100 (MET) From: Marcin Orlowski To: amigappc-dev@amiga.com.pl Subject: Rejected article from list 'amigappc-dev' (fwd) Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Transfer-Encoding: QUOTED-PRINTABLE Please avoid sending from postmaster, root and simmilar acounts as they are banned from posting by default here. -- Daddy, what FORMATTING DRIVE C: means? Marcin mailto:carlos@amiga.com.pl http://www.wfmh.org.pl ---------- Forwarded message ---------- Subject: Rejected article from list 'amigappc-dev' Reason: Address 'postmaster@rask.nospam.kampsax.k-net.dk' is banned from po= sting. Hi. Version 0.20 of libRILc is now available for download: . LibRILc is a standard C library for GCC, written with the purpose of providing PowerUp boards with a runtime environment which is good enough to port many existing m68k programs relatively easily. Thus, it also includes some commonly used functions from the Unix environment in addition to the ones defined by ANSI C. Support for a m68k version was easy to include and useful for debugging purposes. It also has one exceptional feature; AFAIK it is the only Amiga C library other than ixemul.library which is capable of letting the user brea= k out of "for (;;) ;" type loops with Ctrl-C. The library requires OS 2.1 or later. In addition, the PowerUp version need= s ppc.library v45.20 or later. The PPC startup code is compatible with the LoadSeg() patch from phase5. Major news since version 0.15: Fixed some bad bugs in *alloc(), read(), __free_file(), atexit(), fseek() a= nd fgets(). The library should be much more stable now. Fixed (__float_)sprintf() return value. Cache flush/invalidate optimisations. May crash if not used with ppc.librar= y v45.13 or later due to a PPCCallOS() bug. See ppc.library release notes for v45.8. Tracked down and fixed a (cache) misalignment problem with the Caos structu= re passed to PPCCallOS(). Put back in all the cache optimised OS call macros. Added signal(), raise(), abort(), Ctrl-C/E/F check. Added strtod(), atof(), localeconv(), mktemp(). Added atexit(), getenv(), putenv(), system(), clock(), dummy access(). Completed M68k version (except for WBStartup support). Autoconfiscated libRILc. The library is now distributed under the terms of the GNU LGPL, with the following exception: As a special exception, if you link this library with files compiled with a GNU compiler to produce an executable, this does not cause the resulting executable to be covered by the GNU General Public License. This exception does not however invalidate any other reasons why the executable file might be covered by the GNU General Public License. Have fun and don't forget to report any bugs you find. Regards, /=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF= =AF=AF=AF=AF=AF=AFT=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF= =AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF\ | Rask Ingemann Lambertsen | postmaster@rask.nospam.kampsax.k-net.dk | | Registered Phase5 developer | WWW: http://www.gbar.dtu.dk/~c948374/ | | A4000, 775 kkeys/s (RC5-64) | "ThrustMe" on XPilot, ARCnet and IRC | | Windows NT is the OS of the future and always will be... | From owner-amigappc-dev Mon Mar 23 14:39:41 1998 Received: from mail-r1.shlink.de (mail-r1.shlink.de [194.64.6.19]) by amiga.com.pl (8.7.5/8.7.3) with SMTP id OAA00471 for ; Mon, 23 Mar 1998 14:33:19 +0100 (MET) Received: from mail.shlink.de (194.64.6.3) by mail-r1.shlink.de with smtp (Smail3.2.0.95 #1) id m0yH7K2-00013rC; Mon, 23 Mar 1998 14:31:26 +0100 (MET) To: amigappc-dev@amiga.com.pl Message-ID: <1Fh8lMD5131asz2@ark.news.wwbnet.de> From: ARK@News.wwbnet.de (Andreas R. Kleinert) Path: apg.wwbnet.de!News.wwbnet.de Organization: PerSuaSiVe SoftWorX Subject: Re: Asynchronous I/O (Was: AmigaAMP M68K/PPC - StrangeFeature!) Date: Mon, 23 Mar 1998 01:00:00 +0100 X-Mailer: MicroDot 1.17beta3 [REGISTERED 005131] References: <2504.385T45T9223269@kampsax.k-net.dk> Received: by News.wwbnet.de ; Mon, 23 Mar 1998 08:35:08 +0100 Received: by apg.wwbnet.de ; Mon, 23 Mar 1998 09:58:14 +0100 X-Gateway: ZCONNECT apg.wwbnet.de [UNIX/Connect v0.76b1-t1] X-Comment-To: Rask Ingemann Lambertsen X-ZC-TELEFON: 0271-22869 (PHONE/AM/FAX) - NO BBS X-ZC-POST: Sandstrasse 1, D-57072 Siegen, Germany. You wrote at 22.03.1998/17:22:00: [ Asyncio ] > > I've a half-ported version here. > How about making it available for download somewhere? I've deleted it again some days ago - but I can tell you, what I had done to get it working. Wasn't too difficult. > > The only problem is, that SeekAsync() does not seem to work as it should. > I can live with that because I don't need it. Yes. But there was no rational reason, why just Seek should not have worked - so there perhaps had been some more bugs in less-frequented functions, too. Only tested Open/Close/Read/Write/Seek. > > Since there's a lot of packet-sending and cache-flushing involved, it > > probably would make *more* sense to instead generate an PPC interface > > for calling the 68k asyncio.library (so just the includes). > > I'm not so sure. I have an application (mpg123, actually) which reads it's > input file in 4 and 164 byte chunks. Reading at 12-14 kB/s this way means > lots of calls to the '060, and while I'm sure PPCRead() only > flushes/invalidates only as much as necessary, I think this app would be much > better off getting 32-64 kB (or even more, user specifiable) at a time, even > if this would mean a full cache flush/invalidate each time the buffer is > filled. No, it's more than just that to flush: 1. The DOS packet structure to be send to the 68k (twice). 2. The internally used data buffer (twice). 3. Sometimes the buffer passed to Read/Write (twice). Additionally, there do appear some cases where e.g. file size is checked via Examine (flushing FileInfoBlock) and so on. The problem is, that cache first has to be flushed on the PPC side manually, then you have to exchange the data with the 68k and let it work on it, then again flush data on the 68k side. This is quite ineffective, since 68k cache flushing does require to call an exec-lib function. > > It still should be faster than dos.library. > Uhm, the main purpose of asyncronous I/O is to improve latency, not so much > speed. Buffering file I/O on the PPC side would help speed too, though, and > reduce the load on the '060. It would also make the PPC app less vulnerable > to '060 load spikes and starvation of the PPC:MsgHandler process (sortof > related to the latency problem). It may be more efficient to let the 68k do all the packet sending and data copying (which may be DMA into a buffer anyway), so that cache flushing only would be necessary for the buffer that actually is passed to Read/Write. The packet has to be processed by the 68k, anyway. Another advantage is, that the 68k library already may have returned control back to the PPC side (where the PPC can start processing the data) while on the 68k side the asynchroneous read-ahead already is in progress. When doing that from PPC side, some more context-switches and flushes would be involved (to even start the async operation). -- Andreas_Kleinert@t-online.de | ARK@amigaworld.com | Fido 2:2457/350.18 ARK@superview.ftn.neckar-alb.de | ARK@News.wwbnet.de | - *SuperView* - PerSuaSiVe SoftWorX - http://home.t-online.de/home/Andreas_Kleinert/ Image Engineer Registration Site Europe --- Software Engineering -- MicroDot 1.17beta3 registered From owner-amigappc-dev Tue Mar 24 01:09:42 1998 Received: from develop.livewire.com.au (denise.livewire.com.au [203.16.26.4]) by amiga.com.pl (8.7.5/8.7.3) with ESMTP id BAA02230 for ; Tue, 24 Mar 1998 01:00:16 +0100 (MET) Received: from gpsoft.com.au (203.16.26.95) by develop.livewire.com.au (WorldMail 1.3.122) for amigappc-dev@amiga.com.pl; 24 Mar 1998 10:01:11 +1000 From: Andrew Dunbar Reply-To: andrew@gpsoft.com.au To: "AmigaPPC developers' mailing list" Date: Tue, 24 Mar 1998 09:56:26 +1000 Message-ID: In-Reply-To: <1Fh8lMD5131asz2@ark.news.wwbnet.de> X-Mailer: YAM 2.0.0 Preview [020] - Amiga Mailer by Marcel Beck - http://www.yam.ch Organization: GP Software Subject: Re: Cache flushing woes. Was: Asynchronous I/O (Was: AmigaAMP M68K/PPC - StrangeFeature!) MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable Andreas R. Kleinert reckons: (well he/she/it did on 23-Mar-98) >> > Since there's a lot of packet-sending and cache-flushing involved, = it >> > probably would make *more* sense to instead generate an PPC interfa= ce >> > for calling the 68k asyncio.library (so just the includes). >> if this would mean a full cache flush/invalidate each time the buffer = is >> filled. > = > No, it's more than just that to flush: > = > 1. The DOS packet structure to be send to the 68k (twice). > 2. The internally used data buffer (twice). > 3. Sometimes the buffer passed to Read/Write (twice). > = > Additionally, there do appear some cases where e.g. file size is > checked via Examine (flushing FileInfoBlock) and so on. Etc etc etc. It seems to me that what is needed is a 'shared memory' concept. Both CPUs need to be able to mark some area of memory as not cacheable. Sure this would involve the MMU on the 68k side which can clash with other MMU software. But if a concerted effort is made, perhaps some kind of MMU.library can be devised and the existing MMU-banging software can be rewritten to support it. It may even be possible for such a library to work around well known MMU software such as Enforcer and VMem. Come to think of it, isn't chip memory already non-cacheable? Sure it ain't fast but it's way faster than all this cache flushing. Just my 2=A2 -- = Andrew Dunbar GP Software andrew@gpsoft.com.au - Skating rox - Gravel rash sux - From owner-amigappc-dev Wed Mar 25 03:39:44 1998 Received: from mail-r1.shlink.de (mail-r1.shlink.de [194.64.6.19]) by amiga.com.pl (8.7.5/8.7.3) with SMTP id DAA09181 for ; Wed, 25 Mar 1998 03:36:11 +0100 (MET) Received: from mail.shlink.de (194.64.6.3) by mail-r1.shlink.de with smtp (Smail3.2.0.95 #1) id m0yHg13-000141C; Wed, 25 Mar 1998 03:34:09 +0100 (MET) To: amigappc-dev@amiga.com.pl Message-ID: <1F3Z5MD5131aUz1@ark.news.wwbnet.de> From: ARK@News.wwbnet.de (Andreas R. Kleinert) Path: apg.wwbnet.de!News.wwbnet.de Organization: PerSuaSiVe SoftWorX Subject: Re: Cache flushing woes. Was: Asynchronous I/O (Was: AmigaAMP M68K/PPC - StrangeFeature!) Date: Tue, 24 Mar 1998 01:00:00 +0100 X-Mailer: MicroDot 1.17beta3 [REGISTERED 005131] References: Received: by News.wwbnet.de ; Tue, 24 Mar 1998 09:07:47 +0100 Received: by apg.wwbnet.de ; Tue, 24 Mar 1998 19:18:29 +0100 X-Gateway: ZCONNECT apg.wwbnet.de [UNIX/Connect v0.76b1-t1] X-Comment-To: Andrew Dunbar X-ZC-TELEFON: 0271-22869 (PHONE/AM/FAX) - NO BBS X-ZC-POST: Sandstrasse 1, D-57072 Siegen, Germany. MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit You wrote at 24.03.1998/11:56:26: > Both CPUs need to be able to mark some area of memory as not > cacheable. That's already possible, question is, whether this actually is wanted, here ;-) I mean, *if* the memory is cacheable one can profit from the cache while actually working on it, but has to flush it before and after. If it isn't cacheable at all, there's no need to flush, but no related speed gain either... Non-cacheable memory may be ideal for *passing* data, but not necessarily for working on it. So that would limit it's use for the exchange of the read/written data with the 68k side; one would not want to use it for the buffers supplied to AsyncRead/Write() by the application. Anyway, you're right. One could try this, but one would have to compare the two possible solutions, which one's slower. > Come to think of it, isn't chip memory already non-cacheable? Sure it > ain't fast but it's way faster than all this cache flushing. I wouldn't say that. There's no need to flush the *whole* cache, just do something like 'write back xxx bytes starting from address yyy' Still think, it would be better to call the 68k async.library from the PPC side - it's too much OS-ish anyway, so why trying to simulate all that low-level stuff on the PPC side in an appropriate way ? Why *build* a DOS packet on the PPC side when it has to be *processed* by the 68k anyway ? IMHO the 'asynchroneous' aspect here is more important than the 'native' one ;-) -- Andreas_Kleinert@t-online.de | ARK@amigaworld.com | Fido 2:2457/350.18 ARK@superview.ftn.neckar-alb.de | ARK@News.wwbnet.de | - *SuperView* - PerSuaSiVe SoftWorX - http://home.t-online.de/home/Andreas_Kleinert/ Image Engineer Registration Site Europe --- Software Engineering Using ISO 8859-1? Do you see İ as copyright, ÷ as division and ½ as 1/2? -- MicroDot 1.17beta3 registered From owner-amigappc-dev Wed Mar 25 11:39:49 1998 Received: from mail.nordwest.net (root@mail.nordwest.net [134.106.252.131]) by amiga.com.pl (8.7.5/8.7.3) with ESMTP id LAA10128 for ; Wed, 25 Mar 1998 11:34:14 +0100 (MET) Received: from nwn.de (c0009701@ip006-nordhorn.nordwest.net [134.106.237.198]) by mail.nordwest.net (8.8.8/8.8.8) with SMTP id MAA31332 for ; Wed, 25 Mar 1998 12:30:26 +0100 From: Andreas Etzrodt Reply-To: the_shark@gmx.de To: "AmigaPPC developers' mailing list" Date: Wed, 25 Mar 1998 11:34:26 +0100 Message-ID: In-Reply-To: <1F3Z5MD5131aUz1@ark.news.wwbnet.de> X-Mailer: YAM 2.0.0 Preview [020] - Amiga Mailer by Marcel Beck - http://www.yam.ch Subject: FrodoPPC? MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: binary Hi! Would it be too difficult to port Frodo 4.1 to PPC? Frodo68k is too slow. And as it is coded in C(++?) it should be possible ?! Is anyone working on it? Bye, Andy -- Andreas Etzrodt (T H E S H A R K) MEMBER OF APC&TCP - GERMANYS B(IGG)EST AMIGA-CLUB PGP-Public-Key available Anyone can make mistakes, but only an idiot persists in his error. -- Cicero From owner-amigappc-dev Thu Mar 26 02:39:47 1998 Received: from csbvannt2.csbsysbc.com (csbvannt2.csbsysbc.com [209.121.121.66]) by amiga.com.pl (8.7.5/8.7.3) with ESMTP id CAA15079 for ; Thu, 26 Mar 1998 02:26:25 +0100 (MET) Received: by csbvannt2.csbsysbc.com with Internet Mail Service (5.5.1960.3) id ; Wed, 25 Mar 1998 17:26:29 -0800 Message-ID: <015F48BEB57CD111A2C200805F50314104537D@csbvannt2.csbsysbc.com> From: Greg Condon To: "'amigappc-dev@amiga.com.pl'" Subject: subscribe Date: Wed, 25 Mar 1998 17:26:28 -0800 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.1960.3) Content-Type: text/plain From owner-amigappc-dev Thu Mar 26 16:40:07 1998 Received: from leporello.cs.unibo.it (leporello.cs.unibo.it [130.136.1.110]) by amiga.com.pl (8.7.5/8.7.3) with SMTP id QAA16875 for ; Thu, 26 Mar 1998 16:36:01 +0100 (MET) Received: from pinkerton.cs.unibo.it by leporello.cs.unibo.it (5.67b/96.09.13) id AA13384; Thu, 26 Mar 1998 16:32:30 +0100 Received: by pinkerton.cs.unibo.it (SMI-8.6/SMI-SVR4) id QAA29938; Thu, 26 Mar 1998 16:32:27 +0100 From: nardinoc@CS.UniBO.IT (Ilario Nardinocchi) Message-Id: <199803261532.QAA29938@pinkerton.cs.unibo.it> Subject: Ingenuous question... To: amigappc-dev@amiga.com.pl Date: Thu, 26 Mar 1998 16:32:25 +0100 (MET) X-Mailer: ELM [version 2.4 PL25] Content-Type: text I haven't followed the mailing list for a while, but... I've just found out that the last version of scppc (beta19) manages only the first eight arguments of a function and sets to zero the others. Guess what? I've a lot of functions with 11/12 arguments... Did anybody notice this problem? Bye, Illo de' Illis. From owner-amigappc-dev Fri Mar 27 00:09:56 1998 Received: from sulaco.novagate.net (root@sulaco.novagate.com [205.138.138.1]) by amiga.com.pl (8.7.5/8.7.3) with ESMTP id XAA18338 for ; Thu, 26 Mar 1998 23:41:50 +0100 (MET) Received: from bjkrueger (SAS1-K56-517.GrandHaven.novagate.net [206.26.15.17]) by sulaco.novagate.net (8.8.5/8.8.5) with SMTP id RAA01880 for ; Thu, 26 Mar 1998 17:39:53 -0500 (EST) Message-ID: <351B03CD.80F@mindspring.com> Date: Thu, 26 Mar 1998 17:41:33 -0800 From: Steve Krueger X-Mailer: Mozilla 2.02E-intu (Win95; U; 16bit) MIME-Version: 1.0 To: "AmigaPPC developers' mailing list" Subject: Re: Ingenuous question... References: <199803261532.QAA29938@pinkerton.cs.unibo.it> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Ilario Nardinocchi wrote: > > I haven't followed the mailing list for a while, but... > I've just found out that the last version of scppc (beta19) > manages only the first eight arguments of a function and sets > to zero the others. Guess what? I've a lot of functions with 11/12 > arguments... > Did anybody notice this problem? The first eight are passed in registers, and the rest should go on the stack. I thought I had tested that, but since most functions don't use that many parameters, it's possible that I have screwed something up. I'll look into it. sk From owner-amigappc-dev Fri Mar 27 10:10:00 1998 Received: from leporello.cs.unibo.it (leporello.cs.unibo.it [130.136.1.110]) by amiga.com.pl (8.7.5/8.7.3) with SMTP id KAA21317 for ; Fri, 27 Mar 1998 10:04:13 +0100 (MET) Received: from pinkerton.cs.unibo.it by leporello.cs.unibo.it (5.67b/96.09.13) id AA15972; Fri, 27 Mar 1998 10:00:36 +0100 Received: by pinkerton.cs.unibo.it (SMI-8.6/SMI-SVR4) id KAA00888; Fri, 27 Mar 1998 10:00:35 +0100 From: nardinoc@CS.UniBO.IT (Ilario Nardinocchi) Message-Id: <199803270900.KAA00888@pinkerton.cs.unibo.it> Subject: Re: Ingenuous question... To: amigappc-dev@amiga.com.pl Date: Fri, 27 Mar 1998 10:00:33 +0100 (MET) In-Reply-To: <351B03CD.80F@mindspring.com> from "Steve Krueger" at Mar 26, 98 05:41:33 pm X-Mailer: ELM [version 2.4 PL25] Content-Type: text > > Ilario Nardinocchi wrote: > > > > I haven't followed the mailing list for a while, but... > > I've just found out that the last version of scppc (beta19) > > manages only the first eight arguments of a function and sets > > to zero the others. Guess what? I've a lot of functions with 11/12 > > arguments... > > Did anybody notice this problem? > > The first eight are passed in registers, and the rest should go > on the stack. I thought I had tested that, but since most functions > don't use that many parameters, it's possible that I have screwed > something up. In fact, eight parameters should be enough for every application... ;) Bye, Illo de' Illis. From owner-amigappc-dev Fri Mar 27 12:10:09 1998 Received: from era.radiator.dk ([195.97.136.33]) by amiga.com.pl (8.7.5/8.7.3) with ESMTP id LAA21722 for ; Fri, 27 Mar 1998 11:54:07 +0100 (MET) Received: from dotweb.dk (test@[194.239.114.251]) by era.radiator.dk (8.8.7/8.8.7) with ESMTP id LAA20112 for ; Fri, 27 Mar 1998 11:48:53 +0100 Message-ID: <351B847C.D3E4D44E@dotweb.dk> Date: Fri, 27 Mar 1998 11:50:36 +0100 From: Joachim Larsen Reply-To: Joachim@dotweb.dk Organization: Dotweb X-Mailer: Mozilla 4.03 [en] (WinNT; I) MIME-Version: 1.0 To: amigappc-dev@amiga.com.pl Subject: Pre/box rewrite ? Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit I was thinking about Phase 5's announcement of their Pre/box project re: http://www.phase5.de/news/preboxe.html And the impact on the porting efforts . The Quad cpu design would make it necessary to rewrite alot of code to make it support ppc multiprocessing no ? or would the solution simply be a 68k emulator ? . Any thoughts ?. From owner-amigappc-dev Sat Mar 28 15:09:55 1998 Received: from spode (user-38ld870.dialup.mindspring.com [209.86.160.224]) by amiga.com.pl (8.7.5/8.7.3) with SMTP id PAA26868 for ; Sat, 28 Mar 1998 15:06:25 +0100 (MET) Received: by spode.mindspring.net (Amiga SMTPpost 1.04 December 9, 1994) id AA01; Sat, 28 Mar 98 09:03:26 Subject: Re: Ingenuous question... In-Reply-To: <199803270900.KAA00888@pinkerton.cs.unibo.it> from nardinoc@cs.unibo.it (Ilario Nardinocchi) at Fri, 27 Mar 1998 10:00:33 +0100 (MET) From: sekruege@mindspring.com (Steve Krueger) X-Mailer: GMail 0.55 (11.5.95) To: amigappc-dev@amiga.com.pl Date: Sat, 28 Mar 98 09:03:26 Message-ID: <19980328.85DD9A8.846B@spode.mindspring.net> : > : > Ilario Nardinocchi wrote: : > > : > > I haven't followed the mailing list for a while, but... : > > I've just found out that the last version of scppc (beta19) : > > manages only the first eight arguments of a function and sets : > > to zero the others. Guess what? I've a lot of functions with 11/12 : > > arguments... : > > Did anybody notice this problem? : > : > The first eight are passed in registers, and the rest should go : > on the stack. I thought I had tested that, but since most functions : > don't use that many parameters, it's possible that I have screwed : > something up. : : In fact, eight parameters should be enough for every application... ;) : Well, I tried the simple case of passing 10 integer paramenters, and it works ok. Below is what I tried. You can see that I tested both normal arguments in the call to foo(), and varargs to printf(). Can you send me a test case of what fails? foo(int a, int b, int c, int d, int e, int f, int g, int h, int i, int j) { printf("%d %d %d %d %d %d %d %d %d %d\n", a,b,c,d,e,f,g,h,i,j); } main() { foo(1,2,3,4,5,6,7,8,9,10); } From owner-amigappc-dev Mon Mar 30 02:19:33 1998 Received: from shell.wco.com (root@shell.wco.com [199.4.94.16]) by amiga.com.pl (8.8.8/8.7.3) with ESMTP id CAA02795 for ; Mon, 30 Mar 1998 02:07:14 +0200 (CEST) Received: from paix-alg-gw1-23.ncal.verio.com (fw@paix-alg-gw1-23.ncal.verio.com [207.21.130.214]) by shell.wco.com (8.8.5/8.8.5/WCO-18jul97) with SMTP id QAA17608 for ; Sun, 29 Mar 1998 16:06:28 -0800 (PST) Date: Sun, 29 Mar 1998 16:06:18 -0800 (PST) From: Fred Wright X-Sender: fw@paix-alg-gw1-23.ncal.verio.com Reply-To: Fred Wright To: "AmigaPPC developers' mailing list" Subject: Re: ppc-amigaos-ld problems In-Reply-To: <350D7C5E.MD-0.214.mclaughd@kdn0.attnet.or.jp> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On 16 Mar 1998, Douglas McLaughlin wrote: > On Mon, 16 Mar 1998 00:07:46 -0800 (PST), Fred Wright had this to say > about Re: ppc-amigaos-ld problems... > > > > OK, that was just a guess, since it seemed like the most obvious part of > > OpenLibrary() to interact with the PPC stuff. Another possibility is the > > funny signal interaction mentioned in the description of the > > [NO]PATCHRAMLIB options in SnoopDOS. Does it work any better with > > NOPATCHRAMLIB? I don't have my PowerUP board yet, so I can't try it > > myself. > > I will give it a try and get back to you... I just got my PowerUP board installed yesterday, and determined empirically that SnoopDOS is usable with PATCHRAMLIB=NO in the tooltypes. This means that you lose the ability to track file accesses from ramlib, but it's better than nothing. The objects being looked for by ramlib can be seen by tracking OpenLibrary(), OpenDevice(), etc., but you don't get to see the actual full path for the files. Fred Wright From owner-amigappc-dev Mon Mar 30 07:19:23 1998 Received: from kdn0.attnet.or.jp (kdn0.attnet.or.jp [165.76.124.5]) by amiga.com.pl (8.8.8/8.7.3) with ESMTP id GAA03457 for ; Mon, 30 Mar 1998 06:53:03 +0200 (CEST) Received: from 43.gate5.kadena.attnet.or.jp (mclaughd@localhost [165.76.125.59]) by kdn0.attnet.or.jp (8.8.8+Spin/3.6Wbeta7-CONS(11/18/97)) id NAA03279; Mon, 30 Mar 1998 13:52:06 +0900 (JST) From: "Douglas McLaughlin" Organization: STR Programming Services Date: 30 Mar 98 13:56:17 +0900 Subject: SnoopDOS vs. CyberstormPPC Message-Id: <351FA481.MD-0.214.mclaughd@kdn0.attnet.or.jp> In-Reply-To: To: "AmigaPPC developers' mailing list" Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: MicroDot-II/AmigaOS 0.214 On Sun, 29 Mar 1998 16:06:18 -0800 (PST), Fred Wright had this to say about Re: ppc-amigaos-ld problems... > > I will give it a try and get back to you... > > I just got my PowerUP board installed yesterday, and determined > empirically that SnoopDOS is usable with PATCHRAMLIB=NO in the tooltypes. > This means that you lose the ability to track file accesses from ramlib, > but it's better than nothing. The objects being looked for by ramlib can > be seen by tracking OpenLibrary(), OpenDevice(), etc., but you don't get > to see the actual full path for the files. Oops, forgot all about that. I'll add the tooltype in right now... Thanks for the info and enjoy your processer!!! Doug From owner-amigappc-dev Mon Mar 30 14:03:41 1998 Received: from server.fiber.net.pl (server.fiber.net.pl [195.117.208.101]) by amiga.com.pl (8.8.8/8.7.3) with ESMTP id NAA00823 for ; Mon, 30 Mar 1998 13:45:19 +0200 (CEST) Received: from nottingham.ac.uk (pat.ccc.nottingham.ac.uk [128.243.40.194]) by server.fiber.net.pl (Netscape Mail Server v2.02) with SMTP id AAA23970 for ; Mon, 30 Mar 1998 12:45:13 +0100 Received: from granby.ccc.nottingham.ac.uk (unix.ccc.nottingham.ac.uk) [128.243.40.43] by nottingham.ac.uk with smtp (Exim 1.82 #1) id 0yJcuU-0003rE-00; Mon, 30 Mar 1998 12:39:26 +0100 Received: from (PPCJMZ.chemistry) [128.243.121.236] by unix.ccc.nottingham.ac.uk with smtp (Exim 1.82 #1) id 0yJcuT-0006Vx-00; Mon, 30 Mar 1998 12:39:26 +0100 Message-Id: <3.0.1.32.19980330124148.00b7ad30@unix.ccc.nottingham.ac.uk> X-Sender: pczmhx@unix.ccc.nottingham.ac.uk X-Mailer: Windows Eudora Light Version 3.0.1 (32) Date: Mon, 30 Mar 1998 12:41:48 +0100 To: "AmigaPPC developers' mailing list" From: Mark Howson Subject: New A1200 board problems Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Hi Sorry if this question has been answered a thousand times before, but I'm new-ish here :) 1) Does anyone have any idea as to how you update the ppc.library on the new A1200 cards? 'version' reports 45.16, despite the fact I have 45.20beta in LIBS: Is this the infamous ROM? Is it updatable? Surely, it *must* be, right? And, while I'm here, does anyone have problems with libRILc.a not working properly? I'm using 'ppctools.lha' and upon trying a simple program (ie. a for loop with a 'puts()' inside, it works the first couple of times, then complains about 'PPC task already running' (or something), then causes gurus etc. Does it work for other people, and if so what ppc.library are you running? Thanks for any help. I'm stuck :( Mark From owner-amigappc-dev Tue Mar 31 00:03:30 1998 Received: from server.fiber.net.pl (server.fiber.net.pl [195.117.208.101]) by amiga.com.pl (8.8.8/8.7.3) with ESMTP id XAA02495 for ; Mon, 30 Mar 1998 23:52:45 +0200 (CEST) Received: from su1236.Mathematik.Uni-Marburg.DE ([137.248.122.100]) by server.fiber.net.pl (Netscape Mail Server v2.02) with ESMTP id AAA26027 for ; Mon, 30 Mar 1998 22:52:36 +0100 Received: from pollux.mathematik.uni-marburg.de (Jimmy@sppp068.Stud-PPP.Uni-Marburg.DE [137.248.9.200]) by su1236.Mathematik.Uni-Marburg.DE (8.8.8/8.8.5) with SMTP id XAA01864 for ; Mon, 30 Mar 1998 23:46:31 +0200 (MET DST) From: Jimmy Schmidt To: "AmigaPPC developers' mailing list" Date: Mon, 30 Mar 1998 22:46:30 CET-1 Message-ID: In-Reply-To: <351FA481.MD-0.214.mclaughd@kdn0.attnet.or.jp> X-Mailer: YAM 1.3.5 [040] - Amiga Mailer by Marcel Beck Subject: LinuxPPC MIME-Version: 1.0 Content-Type: text/plain Have you seen that? There is a Linux-PPC version for Amiga available at ... ftp://sunsite.auc.dk:21/pub/os/linux/680x0/testing/apus ... the only problem is, that there is no goot doc 4 installation yet. I've tried lots of things, but didn't get it work right now. Has somebody any experience with that. Could you help me ? Thanks, Jimmy ;-) From owner-amigappc-dev Tue Mar 31 00:23:36 1998 Received: from server.fiber.net.pl (server.fiber.net.pl [195.117.208.101]) by amiga.com.pl (8.8.8/8.7.3) with ESMTP id AAA02593 for ; Tue, 31 Mar 1998 00:06:52 +0200 (CEST) Received: from shell.wco.com ([199.4.94.16]) by server.fiber.net.pl (Netscape Mail Server v2.02) with ESMTP id AAA26143 for ; Mon, 30 Mar 1998 23:06:43 +0100 Received: from paix-alg-gw1-9.ncal.verio.com (fw@paix-alg-gw1-9.ncal.verio.com [207.21.130.200]) by shell.wco.com (8.8.5/8.8.5/WCO-18jul97) with SMTP id OAA25627 for ; Mon, 30 Mar 1998 14:00:57 -0800 (PST) Date: Mon, 30 Mar 1998 14:00:44 -0800 (PST) From: Fred Wright X-Sender: fw@paix-alg-gw1-9.ncal.verio.com Reply-To: Fred Wright To: "AmigaPPC developers' mailing list" Subject: Re: New A1200 board problems In-Reply-To: <3.0.1.32.19980330124148.00b7ad30@unix.ccc.nottingham.ac.uk> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Mon, 30 Mar 1998, Mark Howson wrote: > 1) Does anyone have any idea as to how you update the ppc.library on the > new A1200 cards? 'version' reports 45.16, despite the fact I have 45.20beta > in LIBS: The ppc.library is just an ordinary library in LIBS:. If your LIBS: assign ever changes, then be aware of which directory you place ppc.library in, and what the LIBS: assign is at the time of the first access to it. For example, if you use the PPC version of cgxsystem.library, then ppc.library will be opened during the monitor configuration scan, before User-Startup is executed. > Is this the infamous ROM? Is it updatable? Surely, it *must* be, right? No. There is something called the PowerUP kernel, which I believe is in ROM on the board. This should not need changing as frequently as ppc.library. I believe it's flash-upgradable when needed, but I have yet to see any documentation on how this is done. In the CSPPC at least, there doesn't appear to be any write-enable jumper for the flash ROM as there was on the MkII. Although this is more convenient, it creates a potential for insisdious viruses. Fred Wright From owner-amigappc-dev Tue Mar 31 11:43:37 1998 Received: from nevald.k-net.dk (nevald.k-net.dtu.dk [130.225.71.226]) by amiga.com.pl (8.8.8/8.7.3) with SMTP id LAA05438 for ; Tue, 31 Mar 1998 11:32:28 +0200 (CEST) Received: (qmail 2867 invoked from network); 31 Mar 1998 09:31:23 -0000 Received: from carlsberg.kampsax.dtu.dk (192.38.212.2) by nevald.k-net.dtu.dk with SMTP; 31 Mar 1998 09:31:23 -0000 Received: (qmail 11755 invoked from network); 31 Mar 1998 09:31:23 -0000 Received: from k4315.kampsax.dtu.dk (rask@192.38.215.253) by carlsberg.kampsax.dtu.dk with SMTP; 31 Mar 1998 09:31:23 -0000 Received: from k4315.kampsax.dtu.dk (rask@192.38.215.253) by k4315.kampsax.dtu.dk with SMTP; 31 Mar 1998 09:31:06 -0000 Date: 31 Mar 98 11:30:26 +0100 From: "Rask Ingemann Lambertsen" Subject: Re: New A1200 board problems To: "Amiga PowerPC developer mailing list" In-Reply-To: <3.0.1.32.19980330124148.00b7ad30@unix.ccc.nottingham.ac.uk> Message-ID: <1090.394T1337T6904025@kampsax.k-net.dk> MIME-Version: 1.0 Content-type: text/plain; charset=iso-8859-1 Content-transfer-encoding: 8bit Organization: Me? Organised? Dream on... X-Files: The truth is out there. X-INUFS: This message was sent from an INetUtils Free System. X-Mailer: THOR 2.5a (Amiga;TCP/IP) Den 30-Mar-98 12:41:48 skrev Mark Howson fĝlgende om "New A1200 board problems": > 1) Does anyone have any idea as to how you update the ppc.library on the > new A1200 cards? 'version' reports 45.16, despite the fact I have 45.20beta ^^^^^ > in LIBS: > Is this the infamous ROM? Is it updatable? Surely, it *must* be, right? I really hope so. > And, while I'm here, does anyone have problems with libRILc.a not working > properly? I think I saw a notice somewhere in the docs that ppc.library v45.20 is required for libRILc. Regards, /ŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻTŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻ\ | Rask Ingemann Lambertsen | E-mail: mailto:rask@kampsax.k-net.dk | | Registered Phase5 developer | WWW: http://www.gbar.dtu.dk/~c948374/ | | A4000, 775 kkeys/s (RC5-64) | "ThrustMe" on XPilot, ARCnet and IRC | | Whoever said you can't multitask a Mac didn't own an Amiga | From owner-amigappc-dev Tue Mar 31 14:23:49 1998 Received: from leporello.cs.unibo.it (leporello.cs.unibo.it [130.136.1.110]) by amiga.com.pl (8.8.8/8.7.3) with SMTP id OAA06017 for ; Tue, 31 Mar 1998 14:14:35 +0200 (CEST) Received: from pinkerton.cs.unibo.it by leporello.cs.unibo.it (5.67b/96.09.13) id AA05489; Tue, 31 Mar 1998 14:10:13 +0200 Received: by pinkerton.cs.unibo.it (SMI-8.6/SMI-SVR4) id OAA05966; Tue, 31 Mar 1998 14:10:13 +0200 From: nardinoc@CS.UniBO.IT (Ilario Nardinocchi) Message-Id: <199803311210.OAA05966@pinkerton.cs.unibo.it> Subject: Re: Ingenuous question... To: amigappc-dev@amiga.com.pl Date: Tue, 31 Mar 1998 14:10:12 +0200 (MET DST) In-Reply-To: <19980328.85DD9A8.846B@spode.mindspring.net> from "Steve Krueger" at Mar 28, 98 09:03:26 am X-Mailer: ELM [version 2.4 PL25] Content-Type: text > : > The first eight are passed in registers, and the rest should go > : > on the stack. I thought I had tested that, but since most functions > : > don't use that many parameters, it's possible that I have screwed > : > something up. > : > : In fact, eight parameters should be enough for every application... ;) > : > > Well, I tried the simple case of passing 10 integer paramenters, and > it works ok. Below is what I tried. You can see that I tested both normal > arguments in the call to foo(), and varargs to printf(). Can you send me > a test case of what fails? > > > foo(int a, int b, int c, int d, int e, int f, int g, int h, int i, int j) > { > printf("%d %d %d %d %d %d %d %d %d %d\n", > a,b,c,d,e,f,g,h,i,j); > } > > > main() > { > foo(1,2,3,4,5,6,7,8,9,10); > } > Well, my case is this: typedef unsigned short US; typedef unsigned long UL; US function(US* a, US* b, US* c, US* d, US* e, US* f, US* g, US* h, US* i, UL* j, US* k); The arguments from a to h have the correct value, while i, j and k are all NULLs. I didn't have the time to test any variation to this code so here it is (for the ones who are questioning about this queer function definition: I have to interface my code with old COBOL code on a MVS system... *sigh*). Bye, Illo de' Illis. From owner-amigappc-dev Tue Mar 31 18:23:40 1998 Received: from plukwa.pdi.net (robert@plukwa.pdi.net [195.116.162.7]) by amiga.com.pl (8.8.8/8.7.3) with SMTP id SAA06584 for ; Tue, 31 Mar 1998 18:09:42 +0200 (CEST) Received: (qmail 2028760224 invoked by uid 531); 31 Mar 1998 18:11:24 -0000 Message-ID: <19980331181122.64984@plukwa.pdi.net> Date: Tue, 31 Mar 1998 18:11:22 +0000 From: Robert Ramiega To: "AmigaPPC developers' mailing list" Subject: Re: New A1200 board problems References: <3.0.1.32.19980330124148.00b7ad30@unix.ccc.nottingham.ac.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.88 In-Reply-To: ; from Fred Wright on Mon, Mar 30, 1998 at 02:00:44PM -0800 Organization: PDi Ltd, Lodz, PL X-Personal: PlukwaZone Owner X-PGP-fingerprint: BB 1F 6D 5F CF AF 5A 67 E3 0C EA 68 E4 7F 74 B1 On Mon, Mar 30, 1998 at 02:00:44PM -0800, Fred Wright wrote: > > On Mon, 30 Mar 1998, Mark Howson wrote: > > > 1) Does anyone have any idea as to how you update the ppc.library on the > > new A1200 cards? 'version' reports 45.16, despite the fact I have 45.20beta > > in LIBS: > > The ppc.library is just an ordinary library in LIBS:. If your LIBS: Not in this case. In all Blizzards PPC ppc.library and 68040.library is burnt in to the flashROM (ppc.library is indeed version 45.16 and 68040.library in my case has version 43.4). So it doesn't matter what You put in libs: > > > Is this the infamous ROM? Is it updatable? Surely, it *must* be, right? According to P5 tech-support upgrades will be available as soon as flashburner program will be ready (NO, DO NOT USE scsi_xxx from scsi dir on P5 ftp site it is meant for Cyberstorm's only) -- Robert Ramiega | robert@pdi.net IRC: _Jedi_ | Don't underestimate Plukwa Zone Owner | http://plukwa.pdi.net/ | the power of Source