From amigappc-dev-owner@amiga.com.pl Thu Sep 3 04:00:25 1998 Received: from mail-r2.shlink.de (mail-r2.shlink.de [194.64.6.19]) by amiga.com.pl (8.8.8/8.7.3) with SMTP id DAA11241 for ; Thu, 3 Sep 1998 03:42:42 +0200 (CEST) Received: from mail-r1.shlink.de (194.64.6.3) by mail-r2.shlink.de with esmtp id m0zEOIy-0000XFC; Thu, 3 Sep 1998 03:35:20 +0200 (MEST) To: amigappc-dev@amiga.com.pl Message-ID: <17VSuMD5131asz2@ark.news.wwbnet.de> From: ARK@News.wwbnet.de (Andreas R. Kleinert) Subject: registration service Date: Wed, 02 Sep 1998 02:00:00 +0200 Received: by News.wwbnet.de ; Wed, 02 Sep 1998 14:23:35 +0100 Received: by apg.wwbnet.de ; Wed, 02 Sep 1998 19:13:51 +0100 Reply-To: AmigaPPC developers' mailing list Sender: amigappc-dev-owner@amiga.com.pl Precedence: list Status: O X-Status: Since this is a developer list: Have you already checked http://www.swregnet.com ? IMHO all Amiga shareware developers should join there. -- Andreas_Kleinert@t-online.de or via @gmx.de | ARK@News.wwbnet.de Or try good, old: FidoNet 2:2457/350.18 --- *SuperView* development PerSuaSiVe SoftWorX - Software Engineering Amiga/Wintel and more Visit http://wdo.de/ark/ for information, free downloads, registration Any sufficiently advanced technologie is indistinguishable from magic. -- MicroDot 1.17beta3 registered From amigappc-dev-owner@amiga.com.pl Sat Sep 5 18:01:42 1998 Received: from samantha.lysator.liu.se (root@samantha.lysator.liu.se [130.236.254.202]) by amiga.com.pl (8.8.8/8.7.3) with ESMTP id RAA25860 for ; Sat, 5 Sep 1998 17:50:17 +0200 (CEST) Received: from sara.lysator.liu.se (johane@sara.lysator.liu.se [130.236.254.204]) by samantha.lysator.liu.se (8.8.7/8.8.7) with ESMTP id RAA19948 for ; Sat, 5 Sep 1998 17:41:10 +0200 (MET DST) Received: from localhost (johane@localhost) by sara.lysator.liu.se (8.8.8/8.8.7) with SMTP id RAA05002 for ; Sat, 5 Sep 1998 17:41:03 +0200 (MET DST) Date: Sat, 5 Sep 1998 17:41:02 +0200 (MET DST) From: Johan Eriksson To: amigappc-dev@amiga.com.pl Subject: What's wrong here? Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: AmigaPPC developers' mailing list Sender: amigappc-dev-owner@amiga.com.pl Precedence: list Status: O X-Status: The silly program below causes an exception on exit. I have no idea what's up. #include #include #include #include int main(void) { struct ExecBase *SysBase = *(struct ExecBase **)4; struct DosLibrary *DOSBase; struct Library *GfxBase, *IntuitionBase; struct Window *win; int i, k; if ((DOSBase = (struct DosLibrary *)OpenLibrary("dos.library", 37L)) == NULL) return RETURN_FAIL; if ((GfxBase = OpenLibrary("graphics.library", 37L))) { if ((IntuitionBase = OpenLibrary("intuition.library", 37L))) { win = OpenWindowTags(NULL, WA_Left, 100, WA_Top, 100, WA_Width, 100, WA_Height,150, TAG_END); for (k = 10; k < 70; k++) { PPCprintf("%ld\n", k); for (i = 10; i < 70; i++) WritePixel(win->RPort, i, k); } CloseWindow(win); CloseLibrary(IntuitionBase); } CloseLibrary(GfxBase); } CloseLibrary((struct Library *)DOSBase); return 0; } /Johan Eriksson, johane@lysator.liu.se From amigappc-dev-owner@amiga.com.pl Mon Sep 7 23:41:02 1998 Received: from dedal.man.szczecin.pl (dedal.man.szczecin.pl [194.92.39.62]) by amiga.com.pl (8.8.8/8.7.3) with ESMTP id XAA08795 for ; Mon, 7 Sep 1998 23:35:45 +0200 (CEST) Received: (from adam@localhost) by dedal.man.szczecin.pl (8.8.8/8.8.8) id VAA11641 for amigappc-dev@amiga.com.pl; Mon, 7 Sep 1998 21:28:04 GMT From: Adam Ciarcinski Message-Id: <199809072128.VAA11641@dedal.man.szczecin.pl> Subject: Re: What's wrong here? To: amigappc-dev@amiga.com.pl Date: Mon, 7 Sep 1998 23:28:04 +0200 (MDT) In-Reply-To: from "Johan Eriksson" at Sep 05, 1998 05:41:02 PM MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Reply-To: AmigaPPC developers' mailing list Sender: amigappc-dev-owner@amiga.com.pl Precedence: list Status: O X-Status: > The silly program below causes an exception on exit. I have no idea > what's up. > win = OpenWindowTags(NULL, WA_Left, 100, > WA_Top, 100, > WA_Width, 100, > WA_Height,150, > TAG_END); first of all, try not to use vararg tag list, use pointer instead. regards - Adam From amigappc-dev-owner@amiga.com.pl Tue Sep 8 21:21:11 1998 Received: from front1.grolier.fr (front1.grolier.fr [194.158.96.51]) by amiga.com.pl (8.8.8/8.7.3) with ESMTP id VAA15388 for ; Tue, 8 Sep 1998 21:10:46 +0200 (CEST) Received: from club-internet.fr (lyon9-119.club-internet.fr [195.36.158.119]) by front1.grolier.fr (8.9.1/MGC-980407-Frontal-No_Relay) with SMTP id VAA16824 for ; Tue, 8 Sep 1998 21:01:23 +0200 (MET DST) From: Emmanuel Lesueur To: "AmigaPPC developers' mailing list" Date: Tue, 08 Sep 1998 17:47:40 +0100 Message-ID: In-Reply-To: Subject: Re: What's wrong here? MIME-Version: 1.0 Content-Type: text/plain Reply-To: AmigaPPC developers' mailing list Sender: amigappc-dev-owner@amiga.com.pl Precedence: list Hello Johan On 05-Sep-98, you wrote: > The silly program below causes an exception on exit. I have no idea > what's up. I compiled it with egcs 1.0.1 using: gcc -b ppc-amigaos -V egcs-2.90.23 -c -O tst.c ppc-amigaos-ld -r tst.o -o tst and it works fine on my 604e/233, with ppc.library 46.19. Besides, the source seems OK. How do you compile/link it ? -- Emmanuel Lesueur - lesueur@club-internet.fr - lesueur@desargues.univ-lyon1.fr From amigappc-dev-owner@amiga.com.pl Wed Sep 9 07:41:00 1998 Received: from vesuri.Helsinki.FI (jxsaarin@vesuri.helsinki.fi [128.214.205.10]) by amiga.com.pl (8.8.8/8.7.3) with ESMTP id HAA18885 for ; Wed, 9 Sep 1998 07:26:56 +0200 (CEST) Received: from localhost (jxsaarin@localhost) by vesuri.Helsinki.FI (8.8.8/8.8.0) with ESMTP id IAA03775 for ; Wed, 9 Sep 1998 08:17:42 +0300 (EET DST) Date: Wed, 9 Sep 1998 08:17:42 +0300 (EET DST) From: Jyrki O Saarinen To: "AmigaPPC developers' mailing list" Subject: Re: What's wrong here? In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: AmigaPPC developers' mailing list Sender: amigappc-dev-owner@amiga.com.pl Precedence: list > I compiled it with egcs 1.0.1 using: > > gcc -b ppc-amigaos -V egcs-2.90.23 -c -O tst.c > ppc-amigaos-ld -r tst.o -o tst > > and it works fine on my 604e/233, with ppc.library 46.19. > Besides, the source seems OK. > > How do you compile/link it ? Does anyone know how to have ppc-amigaos support for the newest egcs, 1.1b? From amigappc-dev-owner@amiga.com.pl Fri Sep 11 12:41:42 1998 Received: from matej.cerge.cuni.cz (matej.cerge.cuni.cz [195.113.12.40]) by amiga.com.pl (8.8.8/8.7.3) with SMTP id MAA03464 for ; Fri, 11 Sep 1998 12:34:36 +0200 (CEST) Received: from email.cerge.cuni.cz (hpserver.cerge.cuni.cz) by matej.cerge.cuni.cz (5.67a8/Ultrix4.2a) id AA08058; Fri, 11 Sep 1998 12:23:49 +0200 Received: by email.cerge.cuni.cz with NT SMTP Gateway ver 31 id <35F8FA6D@email.cerge.cuni.cz>; Fri, 11 Sep 98 12:24:45 C From: Cizek Pavel To: "amigappc-dev@amiga.com.pl" Subject: Just a note on context switching Date: Fri, 11 Sep 98 13:35:00 C Message-Id: <35F8FA6D@email.cerge.cuni.cz> Reply-To: AmigaPPC developers' mailing list Sender: amigappc-dev-owner@amiga.com.pl Precedence: list This is just a short note on the importance of internal procesor cache in the case of dual processor boards from Phase5. (I have BlizzPPC 603/200 + 040/25.) There was a discussion about the PowerUP vs. WarpUP and the speed of the context switching. Recently, I have installed VBCC for WarpOS (beta). A small program for the testing of the speed of context switches is included in the package (it just opens a window and then draw every pixel in the window using the pair of SetAPen and WritePixel calls about 10000 times). So, I compile it both for WarpOS (using VBCC) and powerpc.library (using SAS/C). The version for WarpOS shows cca 1000 context switches pe second, the version for powerpc.library does only 750 context switches per second. Now comes the tricky part. As I suspect both these systems to clear not only the whole data cache, but also the instruction cache (this is sure fo SAS/C + powerpc.lib combination at least), I changed this behavior. Because nearly for any application it is a pure nonsense to access Intuition's bitmap directly by PowerPC, I modified the calls of SetAPen and WritePixel so that only the part of the data cache corresponding to the memory area of the apropriate RastPort structure is flushed before / after these calls. (Actually, I was able to do it only in the case of SAS/C and powerpc.library by modification of the inline functions for system calls, as I have only a compiled version of amiga.lib for WarpOS.) The result of this change was more than 3000 (!) context switches per second under powerpc.library. I know that such an approach is not possible for every system function (e.g., GetMsg etc.), but in special cases when the maximum speed is required, a careful handling of cache flushes may help a lot. Pavel Cizek P.S.: I am curious about results of such an experiment on PPC604 boards. Thank. From amigappc-dev-owner@amiga.com.pl Fri Sep 11 21:41:39 1998 Received: from jehova.owl.de (jehova.owl.de [194.121.202.132]) by amiga.com.pl (8.8.8/8.7.3) with ESMTP id VAA05239 for ; Fri, 11 Sep 1998 21:28:16 +0200 (CEST) Received: from fiction.pb.owl.de (root@fiction.pb.owl.de [193.174.12.5]) by jehova.owl.de (8.8.8/8.8.8) with SMTP id VAA13291 for ; Fri, 11 Sep 1998 21:18:22 +0200 (MEST) Received: from phoenix.owl.de by fiction.pb.owl.de with bsmtp id m0zHZ4b-00000fC; Fri, 11 Sep 98 21:41 MET DST Received: by phoenix.owl.de (UMSRFC 0.11); Fri, 11 Sep 1998 20:11:08 +0200 To: "amigappc-dev" In-Reply-To: <35F8FA6D@email.cerge.cuni.cz> From: "Frank Wille" Date: Fri, 11 Sep 1998 19:18:10 +0200 Subject: Re: Just a note on context switching Message-ID: <42120665@phoenix.owl.de> Reply-To: AmigaPPC developers' mailing list Sender: amigappc-dev-owner@amiga.com.pl Precedence: list Cizek Pavel wrote: > This is just a short note on the importance of internal procesor cache > in the case of dual processor boards from Phase5. (I have BlizzPPC > 603/200 + 040/25.) > > There was a discussion about the PowerUP vs. WarpUP and the speed of the > context switching. Recently, I have installed VBCC for WarpOS (beta). A > small program for the testing of the speed of context switches is > included in the package (it just opens a window and then draw every > pixel in the window using the pair of SetAPen and WritePixel calls about > 10000 times). So, I compile it both for WarpOS (using VBCC) and > powerpc.library (using SAS/C). For a fair comparison, you should use vbcc for PowerUp from Aminet. SAS/C may be slower or faster in some cases. > The version for WarpOS shows cca 1000 > context switches pe second, the version for powerpc.library does only > 750 context switches per second. What a huge difference! In most cases I measured a small advantage for WarpOS. But today, I made the test again and measured 1606 for PowerUp and only 1560 for WarpOS (PPC604e/200). The latest power- pc.library seems to be slower (I was using powerpc.library 14.6 and ppc.library 45.20). > Now comes the tricky part. As I suspect both these systems to clear not > only the whole data cache, but also the instruction cache (this is sure > fo SAS/C + powerpc.lib combination at least), I changed this behavior. WarpOS does all required cache management operations automati- cally. > Because nearly for any application it is a pure nonsense to access > Intuition's bitmap directly by PowerPC, I modified the calls of SetAPen > and WritePixel so that only the part of the data cache corresponding to > the memory area of the apropriate RastPort structure is flushed before / > after these calls. (Actually, I was able to do it only in the case of > SAS/C and powerpc.library by modification of the inline functions for > system calls, as I have only a compiled version of amiga.lib for > WarpOS.) The source of the linker libraries and startup code is available at the same URL where you got the vbccWOS beta. But it wouldn't help as WarpOS' Run68k() clears the caches automagically. > The result of this change was more than 3000 (!) context > switches per second under powerpc.library. I know that such an approach Not bad. :) > is not possible for every system function (e.g., GetMsg etc.), but in > special cases when the maximum speed is required, a careful handling of > cache flushes may help a lot. Good idea. Yes, I should modify some functions of vbcc's ami- ga.lib (at least the PowerUp ones) and suggest Sam Jordan to implement manual cache clearing for WarpOS/Run68k(). -- _ Frank Wille (frank@phoenix.owl.de) _ // http://fiction.pb.owl.de/~frank/ \X/ Phx @ #AmigaGer From amigappc-dev-owner@amiga.com.pl Fri Sep 11 23:41:24 1998 Received: from avignon.pacwan.net (avignon.pacwan.net [194.250.111.129] (may be forged)) by amiga.com.pl (8.8.8/8.7.3) with ESMTP id XAA05883 for ; Fri, 11 Sep 1998 23:29:03 +0200 (CEST) Received: from pacwan.net (194.250.111.52) by avignon.pacwan.net with SMTP (Eudora Internet Mail Server 1.2); Fri, 11 Sep 1998 23:19:14 +0200 From: Vincent Morenas To: amigappc-dev@amiga.com.pl Date: Fri, 11 Sep 1998 23:15:51 +0100 Message-ID: Subject: EGCS ???? MIME-Version: 1.0 Content-Type: text/plain Reply-To: AmigaPPC developers' mailing list Sender: amigappc-dev-owner@amiga.com.pl Precedence: list Hello, I wan't to recompile by SAS 68K project under PPC with egcs who seem to have beter support than SAS PPC ;-( but i don't know where and what i need ?? Do i have to install the ADE GEEK GADGETS and after EGCS or only EGCS ?? Where to download EGCS PPC for Power-up ?? Do the dev ressource kit from PHASE5 work with EGCS and how to install it ?? (I think what it works but...) That all for the momment.... Kind regards Vincent MORENAS. FRANCE. From amigappc-dev-owner@amiga.com.pl Sat Sep 12 02:01:30 1998 Received: from samantha.lysator.liu.se (root@samantha.lysator.liu.se [130.236.254.202]) by amiga.com.pl (8.8.8/8.7.3) with ESMTP id BAA06257 for ; Sat, 12 Sep 1998 01:41:05 +0200 (CEST) Received: from sara.lysator.liu.se (johane@sara.lysator.liu.se [130.236.254.204]) by samantha.lysator.liu.se (8.8.7/8.8.7) with ESMTP id BAA05839 for ; Sat, 12 Sep 1998 01:31:03 +0200 (MET DST) Received: from localhost (johane@localhost) by sara.lysator.liu.se (8.8.8/8.8.7) with SMTP id BAA11449 for ; Sat, 12 Sep 1998 01:30:56 +0200 (MET DST) Date: Sat, 12 Sep 1998 01:30:56 +0200 (MET DST) From: Johan Eriksson To: amigappc-dev@amiga.com.pl Subject: Clarification Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: AmigaPPC developers' mailing list Sender: amigappc-dev-owner@amiga.com.pl Precedence: list Once and for all, to avoid confusion: ppc.library is in PowerUP powerpc.library is in WarpUP /Johan Eriksson, johane@lysator.liu.se From amigappc-dev-owner@amiga.com.pl Sat Sep 12 16:21:29 1998 Received: from avignon.pacwan.net (avignon.pacwan.net [194.250.111.129] (may be forged)) by amiga.com.pl (8.8.8/8.7.3) with ESMTP id QAA09989 for ; Sat, 12 Sep 1998 16:16:39 +0200 (CEST) Received: from pacwan.net (194.250.111.43) by avignon.pacwan.net with SMTP (Eudora Internet Mail Server 1.2); Sat, 12 Sep 1998 16:06:48 +0200 From: Vincent Morenas To: amigappc-dev@amiga.com.pl Date: Fri, 11 Sep 1998 23:15:51 +0100 Message-ID: Subject: EGCS ???? MIME-Version: 1.0 Content-Type: text/plain Reply-To: AmigaPPC developers' mailing list Sender: amigappc-dev-owner@amiga.com.pl Precedence: list Hello, I wan't to recompile by SAS 68K project under PPC with egcs who seem to have beter support than SAS PPC ;-( but i don't know where and what i need ?? Do i have to install the ADE GEEK GADGETS and after EGCS or only EGCS ?? Where to download EGCS PPC for Power-up ?? Do the dev ressource kit from PHASE5 work with EGCS and how to install it ?? (I think what it works but...) That all for the momment.... Kind regards Vincent MORENAS. FRANCE. From amigappc-dev-owner@amiga.com.pl Sun Sep 13 00:43:54 1998 Received: from mailhub2.ncal.verio.com (mailhub2.ncal.verio.com [204.247.247.54]) by amiga.com.pl (8.8.8/8.7.3) with ESMTP id AAA11472 for ; Sun, 13 Sep 1998 00:36:55 +0200 (CEST) Received: from paix-alg-gw1-49.ncal.verio.com (fw@paix-alg-gw1-49.ncal.verio.com [207.21.130.240]) by mailhub2.ncal.verio.com (8.8.7/8.8.7) with SMTP id PAA18624 for ; Sat, 12 Sep 1998 15:27:02 -0700 (PDT) Date: Sat, 12 Sep 1998 15:27:17 -0700 (PDT) From: Fred Wright To: "AmigaPPC developers' mailing list" Subject: Re: Just a note on context switching In-Reply-To: <35F8FA6D@email.cerge.cuni.cz> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: AmigaPPC developers' mailing list Sender: amigappc-dev-owner@amiga.com.pl Precedence: list On Fri, 11 Sep 1998, Cizek Pavel wrote: > Now comes the tricky part. As I suspect both these systems to clear > not only the whole data cache, but also the instruction cache (this is > sure fo SAS/C + powerpc.lib combination at least), I changed this That's weird. The circumstances requiring instruction-cache clearing are much more limited and fairly well defined. Doing it overzealously seems like a mistake. > behavior. Because nearly for any application it is a pure nonsense to > access Intuition's bitmap directly by PowerPC, I modified the calls of > SetAPen and WritePixel so that only the part of the data cache > corresponding to the memory area of the apropriate RastPort structure > is flushed before / after these calls. (Actually, I was able to do it > only in the case of SAS/C and powerpc.library by modification of the > inline functions for system calls, as I have only a compiled version > of amiga.lib for WarpOS.) The result of this change was more than 3000 > (!) context switches per second under powerpc.library. I know that > such an approach is not possible for every system function (e.g., > GetMsg etc.), but in special cases when the maximum speed is required, > a careful handling of cache flushes may help a lot. While there might be certain circumstances where selective cache flushing is desirable, it's important to remember that any cache flushing beyond a single line is iterative. Even apparently limited operations such as "flush one page" have to scan the *entire* cache since the cache isn't organized by physical pages. Clearing multiple pages requires multiple passes. On DEC mainframes the term used is "cache sweep", which I think is a better term since it helps remind the programmer that these are time-consuming operations. Also note that at least on the 68K (I'm not sure about the PPC) *any* cache flush requires the overhead of switching in and out of supervisor mode when done by a normal program. The best way to limit the impact of cache sweeping is to design the interprocessor communication to minimize the need for synchronous cache sweeps, and to track cache sweeps via globally accessible cache sweep serial numbers so that the cache coherence of a given object can be determined even when the actual sweeps are part of an unrelated activity. Data structures used by multiple processors should be kept in uncached pages to avoid the need for any cache concerns (though interprocessor interlocking is still an issue). Note in particular that using a ring buffer rather than a linked list for a queue can even avoid the need for interprocessor interlocking, with the downside being the fixed maximum queue length. Kludges like PPCCallOS are OK for quick-n'dirty PPC ports, but they don't represent an efficient way to use the PPC in a 68K-based system. Fred Wright From amigappc-dev-owner@amiga.com.pl Tue Sep 15 20:02:00 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 TAA29522 for ; Tue, 15 Sep 1998 19:51:00 +0200 (CEST) Received: from rostock.netsurf.de ([194.64.189.20]) by server.fiber.net.pl (Netscape Messaging Server 3.52) with ESMTP id AAA3ED8 for ; Tue, 15 Sep 1998 19:44:03 +0200 From: Frank Pagels To: amigappc-dev@amiga.com.pl Date: Tue, 15 Sep 1998 19:33:02 +0200 Message-ID: Subject: Timer MIME-Version: 1.0 Content-Type: text/plain Reply-To: AmigaPPC developers' mailing list Sender: amigappc-dev-owner@amiga.com.pl Precedence: list Hello Is there a possibility to to wait for less than 0.02 s with ppclib ? So i understand i can wait with the 50hz timer. But not microsecounds. Is there a way to wait e.g. for 0.0045 s ? Frank -- Frank (Copper) Pagels / Defect Softworks E-Mail: fp0001@rostock.netsurf.de Copper@Hysteria.dssd.sub.org IRC: Copper@#amigager and #rostock WWW: http://copper.home.pages.de/ <-- ArtPRO support ICQ: 16682026 From amigappc-dev-owner@amiga.com.pl Tue Sep 15 23:22:16 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 XAA00619 for ; Tue, 15 Sep 1998 23:17:34 +0200 (CEST) Received: from mailhub2.ncal.verio.com ([204.247.247.54]) by server.fiber.net.pl (Netscape Messaging Server 3.52) with ESMTP id AAA43CA for ; Tue, 15 Sep 1998 23:10:33 +0200 Received: from paix-alg-gw1-18.ncal.verio.com (fw@paix-alg-gw1-18.ncal.verio.com [207.21.130.209]) by mailhub2.ncal.verio.com (8.8.7/8.8.7) with SMTP id OAA19691 for ; Tue, 15 Sep 1998 14:02:09 -0700 (PDT) Date: Tue, 15 Sep 1998 14:02:22 -0700 (PDT) From: Fred Wright To: "AmigaPPC developers' mailing list" Subject: Re: Timer In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: AmigaPPC developers' mailing list Sender: amigappc-dev-owner@amiga.com.pl Precedence: list On Tue, 15 Sep 1998, Frank Pagels wrote: > Is there a possibility to to wait for less than 0.02 s with ppclib ? > So i understand i can wait with the 50hz timer. But not microsecounds. > Is there a way to wait e.g. for 0.0045 s ? There's a very nice feature for this in the PPC CPU, but I haven't looked enough to see whether the software supports it. Hardwarily, it's better than the 68K side, where you have to go through those blecherous 8520s. Fred Wright From amigappc-dev-owner@amiga.com.pl Wed Sep 16 21:42:41 1998 Received: from jehova.owl.de (jehova.owl.de [194.121.202.132]) by amiga.com.pl (8.8.8/8.7.3) with ESMTP id VAA07553 for ; Wed, 16 Sep 1998 21:30:19 +0200 (CEST) Received: from fiction.pb.owl.de (root@fiction.pb.owl.de [193.174.12.5]) by jehova.owl.de (8.8.8/8.8.8) with SMTP id VAA02319 for ; Wed, 16 Sep 1998 21:19:53 +0200 (MEST) Received: from phoenix.owl.de by fiction.pb.owl.de with bsmtp id m0zJNWF-00000gC; Wed, 16 Sep 98 21:45 MET DST Received: by phoenix.owl.de (UMSRFC 0.11); Wed, 16 Sep 1998 20:18:49 +0200 To: "AmigaPPC developers' mailing list" In-Reply-To: From: "Frank Wille" Date: Wed, 16 Sep 1998 19:31:24 +0200 Subject: Re: Timer Message-ID: <42120682@phoenix.owl.de> Reply-To: AmigaPPC developers' mailing list Sender: amigappc-dev-owner@amiga.com.pl Precedence: list Fred Wright wrote: > On Tue, 15 Sep 1998, Frank Pagels wrote: > > > Is there a possibility to to wait for less than 0.02 s with ppclib ? > > So i understand i can wait with the 50hz timer. But not microsecounds. > > Is there a way to wait e.g. for 0.0045 s ? > > There's a very nice feature for this in the PPC CPU, but I haven't TBR, the time base facility. Two 32 bit registers with a very high precision. IIRC using the PPC bus clock as input. > looked enough to see whether the software supports it. Hardwarily, it's There is support in WarpOS for it: GetSysTimePPC() and WaitTime(). > better than the 68K side, where you have to go through those blecherous > 8520s. Most definitely. -- _ Frank Wille (frank@phoenix.owl.de) _ // http://fiction.pb.owl.de/~frank/ \X/ Phx @ #AmigaGer From amigappc-dev-owner@amiga.com.pl Thu Sep 17 11:25:47 1998 Received: from rostock.netsurf.de (root@rostock.netsurf.de [194.64.189.20]) by amiga.com.pl (8.8.8/8.7.3) with ESMTP id LAA03837 for ; Thu, 17 Sep 1998 11:24:59 +0200 (CEST) From: Frank Pagels To: "AmigaPPC developers' mailing list" Date: Thu, 17 Sep 1998 11:00:23 +0200 Message-ID: In-Reply-To: <42120682@phoenix.owl.de> Subject: Re: Timer MIME-Version: 1.0 Content-Type: text/plain Reply-To: AmigaPPC developers' mailing list Sender: amigappc-dev-owner@amiga.com.pl Precedence: list Am 16-Sep-98 schrieb Frank Wille: >>> Is there a possibility to to wait for less than 0.02 s with ppclib ? >>> So i understand i can wait with the 50hz timer. But not microsecounds. >>> Is there a way to wait e.g. for 0.0045 s ? >> >> There's a very nice feature for this in the PPC CPU, but I haven't > > TBR, the time base facility. Two 32 bit registers with a very > high precision. IIRC using the PPC bus clock as input. Yep, i solve this problem with the time base registers. It works now fine :) Frank -- Frank (Copper) Pagels / Defect Softworks E-Mail: fp0001@rostock.netsurf.de Copper@Hysteria.dssd.sub.org IRC: Copper@#amigager and #rostock WWW: http://copper.home.pages.de/ <-- ArtPRO support ICQ: 16682026 From amigappc-dev-owner@amiga.com.pl Fri Sep 18 00:06:05 1998 Received: from oak (joesera@oak.liii.com [198.207.193.1]) by amiga.com.pl (8.8.8/8.7.3) with ESMTP id AAA06043 for ; Fri, 18 Sep 1998 00:01:20 +0200 (CEST) Received: from localhost by oak with ESMTP (8.8.6/8.8.5) id VAA18994; Thu, 17 Sep 1998 21:50:18 GMT Date: Thu, 17 Sep 1998 17:50:16 -0400 (EDT) From: Joe Sera To: "AmigaPPC developers' mailing list" Subject: Re: CyberGL PPC In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: AmigaPPC developers' mailing list Sender: amigappc-dev-owner@amiga.com.pl Precedence: list Hi all! Has anybody tried using the CyberGL PPC link lib yet? I am converting some of my tools over to the PPC version, but the PPC version of the link lib has no open or close GL window as the 68k shared library version does. Since no header files came with the link lib, I was wondering how can I create and delete a GL context using the PPC link library? TIA, Joe Sera From amigappc-dev-owner@amiga.com.pl Fri Sep 18 04:25:40 1998 Received: from mailhub2.ncal.verio.com (mailhub2.ncal.verio.com [204.247.247.54]) by amiga.com.pl (8.8.8/8.7.3) with ESMTP id EAA08458 for ; Fri, 18 Sep 1998 04:23:51 +0200 (CEST) Received: from paix-alg-gw1-3.ncal.verio.com (fw@paix-alg-gw1-3.ncal.verio.com [207.21.130.194]) by mailhub2.ncal.verio.com (8.8.7/8.8.7) with SMTP id TAA21896 for ; Thu, 17 Sep 1998 19:12:28 -0700 (PDT) Date: Thu, 17 Sep 1998 19:12:41 -0700 (PDT) From: Fred Wright To: "AmigaPPC developers' mailing list" Subject: Re: Timer In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: AmigaPPC developers' mailing list Sender: amigappc-dev-owner@amiga.com.pl Precedence: list On Thu, 17 Sep 1998, Frank Pagels wrote: > Am 16-Sep-98 schrieb Frank Wille: > > > TBR, the time base facility. Two 32 bit registers with a very > > high precision. IIRC using the PPC bus clock as input. > > Yep, i solve this problem with the time base registers. It works > now fine :) Yes, for timestamping or short loop-style delays, there shouldn't be a problem using the registers directly, although you have to keep track of the system-dependent rate, which can vary as a function of the PPC model as well as the clock speed. Using the interval timer isn't very practical without system support, though, since that has to be coordinated among the various uses. Fred Wright From amigappc-dev-owner@amiga.com.pl Fri Sep 18 14:46:26 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 OAA09913 for ; Fri, 18 Sep 1998 14:42:30 +0200 (CEST) Received: from rostock.netsurf.de ([194.64.189.20]) by server.fiber.net.pl (Netscape Messaging Server 3.52) with ESMTP id AAA2124 for ; Fri, 18 Sep 1998 14:35:11 +0200 From: Frank Pagels To: "AmigaPPC developers' mailing list" Date: Fri, 18 Sep 1998 14:24:11 +0200 Message-ID: In-Reply-To: Subject: Re: Timer MIME-Version: 1.0 Content-Type: text/plain Reply-To: AmigaPPC developers' mailing list Sender: amigappc-dev-owner@amiga.com.pl Precedence: list Am 18-Sep-98 schrieb Fred Wright: > Yes, for timestamping or short loop-style delays, there shouldn't be a > problem using the registers directly, although you have to keep track of > the system-dependent rate, which can vary as a function of the PPC model > as well as the clock speed. Yes, i check the ticks per second with the ppc lib to calibrate the timer. I hope it works. I need a tester with lower than 233 mhz to verify this :) It's a mpeg video player. > Fred Wright Frank -- Frank (Copper) Pagels / Defect Softworks E-Mail: fp0001@rostock.netsurf.de Copper@Hysteria.dssd.sub.org IRC: Copper@#amigager and #rostock WWW: http://copper.home.pages.de/ <-- ArtPRO support ICQ: 16682026 From amigappc-dev-owner@amiga.com.pl Fri Sep 18 23:45:51 1998 Received: from aix-mail.pacwan.net (aix-mail.pacwan.net [194.206.203.152]) by amiga.com.pl (8.8.8/8.7.3) with ESMTP id XAA12103 for ; Fri, 18 Sep 1998 23:41:11 +0200 (CEST) Received: from pacwan.net ([194.250.111.20]) by aix-mail.pacwan.net (Netscape Mail Server v2.02) with SMTP id AAB178 for ; Fri, 18 Sep 1998 23:49:51 +0200 From: Vincent Morenas To: amigappc-dev@amiga.com.pl Date: Fri, 18 Sep 1998 23:28:35 +0100 Message-ID: Subject: EGCS Help !!! (fwd) MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable Reply-To: AmigaPPC developers' mailing list Sender: amigappc-dev-owner@amiga.com.pl Precedence: list Hello, *** Message suivi, =E9crit =E0 l'origine par Vincent Morenas le 18-Sep-98= *** Hello, As a candide under Power-Up developpement, i need some information to por= t my C++ "Amiga Classic" project under EGCS : = 1 - What do i need to make and link Power-Up Code ? ADE ?? Where to get E= GCS Power-Up (1.1b ?) ?? 2 - Where to put the Phase5 Kit in EGCS ?? Di i need to "translated" the AmigaOS include ?? 3 - How to make a self running ELF whitout 68k Launcher and is .ELF ?? = Kind regards Thanks in advance... Vincent MORENAS. - FRANCE - (OK, i'm sorry but i still not read the Phase5 doc and i like to have lot= of informations before starting some job ;-) ) *** End of forwarded message *** Kind regards From amigappc-dev-owner@amiga.com.pl Thu Oct 1 08:47:13 1998 Received: from ns.sbs.de (ns.sbs.de [194.112.84.17]) by amiga.com.pl (8.8.8/8.7.3) with ESMTP id IAA07088 for ; Thu, 1 Oct 1998 08:34:38 +0200 (CEST) Received: from mail.fth.sbs.de (mail.fth.sbs.de [192.129.41.35]) by ns.sbs.de (8.8.8/8.8.6) with ESMTP id IAA21333 for ; Thu, 1 Oct 1998 08:42:11 +0200 (MDT) Received: from h62d6b0.hgw.scn.de (mail.hgw.scn.de [218.80.2.111]) by mail.fth.sbs.de (8.8.8/8.8.6) with SMTP id IAA20505 for ; Thu, 1 Oct 1998 08:27:30 +0200 (MET DST) Message-Id: <199810010627.IAA20505@mail.fth.sbs.de> Received: from grfw0673.hgw.scn.de by h62d6b0.hgw.scn.de with SMTP (1.38.193.4/16.2) id AA03302; Thu, 1 Oct 1998 07:35:09 +0100 From: "Alexander Fritsch" To: amigappc-dev@amiga.com.pl Date: Thu, 1 Oct 1998 08:38:20 +0000 Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT Subject: elf-progs, runelf still needed? In-Reply-To: Reply-To: AmigaPPC developers' mailing list Sender: amigappc-dev-owner@amiga.com.pl Precedence: list Hello all, I am sorry for this stupid question, but without your help I will propably loose my last hairs. I compiled 2 programs with SAS/C. I made a 68k and a ppc version. Both programs work nicely in the 68k version and as I remember the used to work even in the ppc-version. After a longer break I continiued my work and got a huge problem: I cant start these programs any longer. If I run a shell an type test_ppc I get the message file is not executable. Ok, I maight have forgotten to install the elfloader-Patch, but is is installed! I have ppc.library 46.15 and the v44 680x0.libraries. InstallPPC is in the startup-sequence. I also made an icon for the programs. When I doubleclick the icon my programs start fine(!!!), but they dont work from the shell. (???) Wen I use runelf from the Steve Kruegers SASPPC-betas the programs run from shell but I think due to InstallPPC this program is no longer needed or am I wrong? (And I think it worked before) (Of course I removed all other strange patches while testing.) Other ppc-programs like the lha from Phase5 or the RC5 clients work with no problems. Hope anybody can help me. Best Regards Alexander Fritsch