Re: [Dri-devel] Radeon PCI API
--- Chris Ison [EMAIL PROTECTED] wrote: I know you can't give details under NDA, but do the specs for the r200 show any API for PCI, or has the PCI work on the driver been done by trial and error. I'll have to check the docs when I get home. The original PCI GART work does done for HP for radeon support on Alpha. At the time it was just for regular radeons. Michel did some re-arranging fairly recently to unify the gart code between PCI and AGP. Also I am wondering if there is any word on specs for r250 and r300. If not, I would like to know what the best supported card is in DRI with the least software implemented GL. I assume you are talking about r350. the r250 is just a variation of the r200 core and docs are available for that. no 3D databooks are available yet for r300 to my understanding. I don't know if that will change in the future or not. I'd say the best supported cards are r200 based radeons (8500 to 9200 cards). Alex __ Do you Yahoo!? Free Pop-Up Blocker - Get it now http://companion.yahoo.com/ --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Dri-devel] Radeon PCI API
On Fri, 2003-11-28 at 07:37, Alex Deucher wrote: I'd say the best supported cards are r200 based radeons (8500 to 9200 cards). And which one is the fastest? -- Torgeir Veimo [EMAIL PROTECTED] --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Dri-devel] Radeon PCI API
Torgeir Veimo wrote: On Fri, 2003-11-28 at 07:37, Alex Deucher wrote: I'd say the best supported cards are r200 based radeons (8500 to 9200 cards). And which one is the fastest? Arguably the Radeon 9100. Just make sure you don't get some crappy card with a 64-bit memory interface; it'll perform like crap no matter how good the drivers are. David Bronaugh --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Dri-devel] Radeon PCI API
On Fri, 2003-11-28 at 08:49, David Bronaugh wrote: Torgeir Veimo wrote: On Fri, 2003-11-28 at 07:37, Alex Deucher wrote: I'd say the best supported cards are r200 based radeons (8500 to 9200 cards). And which one is the fastest? Arguably the Radeon 9100. Just make sure you don't get some crappy card with a 64-bit memory interface; it'll perform like crap no matter how good the drivers are. Well, I got a cheap 9100. It doesn't perform as good as my previous geforce Ti4200, and they managed to cripple it by having an analogue only DVI connector. But hey, it used open source drivers! The 9100 only exist in versions with 128bit memory interface, or is it dependent on manufacturer? -- Torgeir Veimo [EMAIL PROTECTED] --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Dri-devel] r200 Xserver hang: strange
hello unfortunately your mail got lost somehow. In the meantime I searched bugs.xfree86.org and found bugs #787 and #788 which I think are related to this problem. I added some info to bug #788. Later I noticed that his problem was on *BSD, not on Linux. Maybe you could have a look at it. http://bugs.xfree86.org/show_bug.cgi?id=788 Am 2003.11.22 23:29:47 +0100 schrieb(en) Michel Dänzer: [...] You're saying there are UT2003 binaries for PPC? ;) hmmm ;) maybe not. sorry. [...] how to trigger the bug: start glxgears, start quake3 in windowed mode, You're saying... ;) ;) I found another program using scissors that triggers (maybe) the same bug: lesson24 from nehe.gamedev.net http://nehe.gamedev.net/data/lessons/lesson.asp?lesson=24 (glut and sdl source) The difference is: when triggering the bug with q3a both gl-apps exit() with r200WaitIrq: drmRadeonIrqWait: -16 and the Xserver loops, no cursor-movement... when triggering the bug with lesson24, one program loops and gets #1 in top (99,x% CPU), and you have at least a moving cursor. After killing -KILL the programs, the Xserver becomes #1 - now you have the same result you get with quake3. go to the model choosing menu, fiddle around a bit, try different models, push the back/forw. buttons, choose another model... and somewhen it should happen: glxgears and quake3 exit with r200WaitIrq: drmRadeonIrqWait: -16r200WaitIrq: drmRadeonIrqWait: -16 Another symptom of a lockup: the awaited interrupts never arrive. I'm curious, does this only happen if the windows overlap, or regardless? I dont know. Im going to try it out. [...] I guess the chip lockup turns into a full system lockup when trying to (re-)initialise the chip or something. Couldnt it be possible to somehow regenerate/reset the card? Dunno, suspect not. maybe Hui Yu has some hints? At least the ati drivers for another os got a reinitialise the chip gracefully or something similar on the features/changes list sometime ago. What needs to be defined in host.def to get better debug-info in the static build? (XFree86 CVS HEAD) What kind of information are you thinking of? I thought of linenumbers and variable names. Both are there when building from DRI CVS, but none of them is there when building XFREE86 CVS HEAD. best regards, Andreas --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Dri-devel] Radeon PCI API
On Fri, 2003-11-28 at 22:36, Chris Ison wrote: I'd say the best supported cards are r200 based radeons (8500 to 9200 cards). well thats just great, I was hopeing to see something different other than r200 as I have had nothing but trouble and poor performance with them with DRI. I have yet to determine why but since AGP ppl have really good performance even in 1x, its looking like somethings happening with the PCI side, or mesa is using software rendering for PCI for some reason where it doesn't on the AGP. You can check the Mesa and 3D driver debugging output for fallbacks yourself, but both behave exactly the same regardless of the bus type. As you may have gathered from reading this list, it's basically a mystery. PS: Does anyone have an idea why Chris's posts seem to get through to the list directly, but also appear in the administrative queue? -- Earthling Michel Dnzer | Debian (powerpc), X and DRI developer Software libre enthusiast| http://svcs.affero.net/rm.php?r=daenzer --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
[Dri-devel] Wiki notification emails
Today another funny guy decided to tamper with the Wiki (this time by applying rot13 to a full page). Thankfully this doesn't happen often (I recall only 3 times so far). Still, since I'm the only one monitoring I'm afraid I could let attacks like pass though. I'm even more concerned about minor edits which could change drastically the meaning of sentences, leading users to do something unapproriate. (It may be a remote possibility that sombody goes through that kind of work, but still perfectly possible). Therefore I'd like to have more eyeballs by directing the nightly diffs emails to a mailing list. I was thinking of dri-patches, since it's just one mail per day, usually 1-2Kb. Below is yesterday's email as an example. Is it OK to send this to dri-patches? José Fonseca - Forwarded message from Jose Fonseca [EMAIL PROTECTED] - Date: Fri, 28 Nov 2003 04:02:13 -0800 From: Jose Fonseca [EMAIL PROTECTED] Subject: [Dri-wiki] Wiki Update To: [EMAIL PROTECTED] diff -urN /tmp/wiki.BjnDI7/wiki/text/WikiWikiWeb /home/groups/d/dr/dri/wiki/text/WikiWikiWeb --- /tmp/wiki.BjnDI7/wiki/text/WikiWikiWeb2003-11-01 15:58:10.0 -0800 +++ /home/groups/d/dr/dri/wiki/text/WikiWikiWeb 2003-11-27 09:38:14.0 -0800 @@ -1,9 +1,9 @@ -The [wiki:Wiki:FrontPage first ever wiki site] was founded in 1994 as an automated supplement to the Wiki:PortlandPatternRepository. The site was immediately popular within the pattern community, largely due to the newness of the internet and a good slate of Wiki:InvitedAuthors. The site was, and remains, dedicated to Wiki:PeopleProjectsAndPatterns. +Gur [jvxv:Jvxv:SebagCntr svefg rire jvxv fvgr] jnf sbhaqrq va 1994 nf na nhgbzngrq fhccyrzrag gb gur Jvxv:CbegynaqCnggreaErcbfvgbel. Gur fvgr jnf vzzrqvngryl cbchyne jvguva gur cnggrea pbzzhavgl, ynetryl qhr gb gur arjarff bs gur vagrearg naq n tbbq fyngr bs Jvxv:VaivgrqNhgubef. Gur fvgr jnf, naq erznvaf, qrqvpngrq gb Jvxv:CrbcyrCebwrpgfNaqCnggreaf. -Wiki:WardCunnigham created the site and the WikiWikiWeb machinery that operates it. He chose wiki-wiki as an alliterative substitute for quick and thereby avoided naming this stuff quick-web. An early page, Wiki:WikiWikiHyperCard, traces wiki ideas back to a Wiki:HyperCard stack he wrote in the late 80's. +Jvxv:JneqPhaavtunz perngrq gur fvgr naq gur JvxvJvxvJro znpuvarel gung bcrengrf vg. Ur pubfr jvxv-jvxv nf na nyyvgrengvir fhofgvghgr sbe dhvpx naq gurerol nibvqrq anzvat guvf fghss dhvpx-jro. Na rneyl cntr, Jvxv:JvxvJvxvUlcrePneq, genprf jvxv vqrnf onpx gb n Jvxv:UlcrePneq fgnpx ur jebgr va gur yngr 80'f. -See also one of these links: - * http://www.c2.com/cgi/wiki or wiki:Wiki/FrontPage - * get some answers on the Wiki:WikiWikiWebFaq - * get to know more about the Wiki:WikiHistory - * [http://news.mpr.org/programs/futuretense/daily_rafiles/20011220.ram Ward Cunningham Radio Interview] +Frr nyfb bar bs gurfr yvaxf: + * uggc://jjj.p2.pbz/ptv/jvxv be jvxv:Jvxv/SebagCntr + * trg fbzr nafjref ba gur Jvxv:JvxvJvxvJroSnd + * trg gb xabj zber nobhg gur Jvxv:JvxvUvfgbel + * [uggc://arjf.zce.bet/cebtenzf/shghergrafr/qnvyl_ensvyrf/20011220.enz Jneq Phaavatunz Enqvb Vagreivrj] - End forwarded message - --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
[Dri-devel] Wiki notification emails
Today another funny guy decided to tamper with the Wiki (this time by applying rot13 to a full page). Thankfully this doesn't happen often (I recall only 3 times so far). Still, since I'm the only one monitoring I'm afraid I could let attacks like pass though. I'm even more concerned about minor edits which could change drastically the meaning of sentences, leading users to do something unapproriate. (It may be a remote possibility that sombody goes through that kind of work, but still perfectly possible). Therefore I'd like to have more eyeballs by directing the nightly diffs emails to a mailing list. I was thinking of dri-patches, since it's just one mail per day, usually 1-2Kb. Below is yesterday's email as an example. Is it OK to send this to dri-patches? José Fonseca - Forwarded message from Jose Fonseca [EMAIL PROTECTED] - Date: Fri, 28 Nov 2003 04:02:13 -0800 From: Jose Fonseca [EMAIL PROTECTED] Subject: [Dri-wiki] Wiki Update To: [EMAIL PROTECTED] diff -urN /tmp/wiki.BjnDI7/wiki/text/WikiWikiWeb /home/groups/d/dr/dri/wiki/text/WikiWikiWeb --- /tmp/wiki.BjnDI7/wiki/text/WikiWikiWeb2003-11-01 15:58:10.0 -0800 +++ /home/groups/d/dr/dri/wiki/text/WikiWikiWeb 2003-11-27 09:38:14.0 -0800 @@ -1,9 +1,9 @@ -The [wiki:Wiki:FrontPage first ever wiki site] was founded in 1994 as an automated supplement to the Wiki:PortlandPatternRepository. The site was immediately popular within the pattern community, largely due to the newness of the internet and a good slate of Wiki:InvitedAuthors. The site was, and remains, dedicated to Wiki:PeopleProjectsAndPatterns. +Gur [jvxv:Jvxv:SebagCntr svefg rire jvxv fvgr] jnf sbhaqrq va 1994 nf na nhgbzngrq fhccyrzrag gb gur Jvxv:CbegynaqCnggreaErcbfvgbel. Gur fvgr jnf vzzrqvngryl cbchyne jvguva gur cnggrea pbzzhavgl, ynetryl qhr gb gur arjarff bs gur vagrearg naq n tbbq fyngr bs Jvxv:VaivgrqNhgubef. Gur fvgr jnf, naq erznvaf, qrqvpngrq gb Jvxv:CrbcyrCebwrpgfNaqCnggreaf. -Wiki:WardCunnigham created the site and the WikiWikiWeb machinery that operates it. He chose wiki-wiki as an alliterative substitute for quick and thereby avoided naming this stuff quick-web. An early page, Wiki:WikiWikiHyperCard, traces wiki ideas back to a Wiki:HyperCard stack he wrote in the late 80's. +Jvxv:JneqPhaavtunz perngrq gur fvgr naq gur JvxvJvxvJro znpuvarel gung bcrengrf vg. Ur pubfr jvxv-jvxv nf na nyyvgrengvir fhofgvghgr sbe dhvpx naq gurerol nibvqrq anzvat guvf fghss dhvpx-jro. Na rneyl cntr, Jvxv:JvxvJvxvUlcrePneq, genprf jvxv vqrnf onpx gb n Jvxv:UlcrePneq fgnpx ur jebgr va gur yngr 80'f. -See also one of these links: - * http://www.c2.com/cgi/wiki or wiki:Wiki/FrontPage - * get some answers on the Wiki:WikiWikiWebFaq - * get to know more about the Wiki:WikiHistory - * [http://news.mpr.org/programs/futuretense/daily_rafiles/20011220.ram Ward Cunningham Radio Interview] +Frr nyfb bar bs gurfr yvaxf: + * uggc://jjj.p2.pbz/ptv/jvxv be jvxv:Jvxv/SebagCntr + * trg fbzr nafjref ba gur Jvxv:JvxvJvxvJroSnd + * trg gb xabj zber nobhg gur Jvxv:JvxvUvfgbel + * [uggc://arjf.zce.bet/cebtenzf/shghergrafr/qnvyl_ensvyrf/20011220.enz Jneq Phaavatunz Enqvb Vagreivrj] - End forwarded message - --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Dri-devel] r200 Xserver hang: strange
On Fri, 2003-11-28 at 13:00, Andreas Stenglein wrote:
I found another program using scissors that triggers
(maybe) the same bug: lesson24 from nehe.gamedev.net
http://nehe.gamedev.net/data/lessons/lesson.asp?lesson=24
(glut and sdl source)
The difference is: when triggering the bug with q3a
both gl-apps exit() with r200WaitIrq: drmRadeonIrqWait: -16
and the Xserver loops, no cursor-movement...
when triggering the bug with lesson24, one program loops
and gets #1 in top (99,x% CPU), and you have at least
a moving cursor. After killing -KILL the programs, the Xserver
becomes #1 - now you have the same result you get with quake3.
Sounds like different symptoms of a chip lockup.
This patch doesn't happen to make a difference, does it?
What needs to be defined in host.def to get better debug-info in the
static build? (XFree86 CVS HEAD)
What kind of information are you thinking of?
I thought of linenumbers and variable names.
Both are there when building from DRI CVS, but none of them
is there when building XFREE86 CVS HEAD.
I guess XFree86 CVS doesn't build with -g by default. You can always
rebuild any parts you want debugging info for with makeg.
--
Earthling Michel Dnzer | Debian (powerpc), X and DRI developer
Software libre enthusiast| http://svcs.affero.net/rm.php?r=daenzer
Index: programs/Xserver/hw/xfree86/os-support/shared/drm/kernel/radeon_state.c
===
RCS file: /home/x-cvs/xc/programs/Xserver/hw/xfree86/os-support/shared/drm/kernel/radeon_state.c,v
retrieving revision 1.8
diff -p -u -r1.8 radeon_state.c
--- programs/Xserver/hw/xfree86/os-support/shared/drm/kernel/radeon_state.c 28 Sep 2003 20:16:09 - 1.8
+++ programs/Xserver/hw/xfree86/os-support/shared/drm/kernel/radeon_state.c 28 Nov 2003 15:11:17 -
@@ -47,7 +47,11 @@ static __inline__ void radeon_emit_clip_
DRM_DEBUG(box: x1=%d y1=%d x2=%d y2=%d\n,
box-x1, box-y1, box-x2, box-y2 );
- BEGIN_RING( 4 );
+ /* FIXME send a WAIT_UNTIL_3D_IDLE before writing to the scissor
+ * registers to avoid lockups
+ */
+ BEGIN_RING( 6 );
+ RADEON_WAIT_UNTIL_3D_IDLE();
OUT_RING( CP_PACKET0( RADEON_RE_TOP_LEFT, 0 ) );
OUT_RING( (box-y1 16) | box-x1 );
OUT_RING( CP_PACKET0( RADEON_RE_WIDTH_HEIGHT, 0 ) );
@@ -1941,23 +1945,6 @@ static int radeon_emit_packet3_cliprect(
if ( i cmdbuf-nbox ) {
if (DRM_COPY_FROM_USER_UNCHECKED( box, boxes[i], sizeof(box) ))
return DRM_ERR(EFAULT);
- /* FIXME The second and subsequent times round
- * this loop, send a WAIT_UNTIL_3D_IDLE before
- * calling emit_clip_rect(). This fixes a
- * lockup on fast machines when sending
- * several cliprects with a cmdbuf, as when
- * waving a 2D window over a 3D
- * window. Something in the commands from user
- * space seems to hang the card when they're
- * sent several times in a row. That would be
- * the correct place to fix it but this works
- * around it until I can figure that out - Tim
- * Smith */
- if ( i ) {
-BEGIN_RING( 2 );
-RADEON_WAIT_UNTIL_3D_IDLE();
-ADVANCE_RING();
- }
radeon_emit_clip_rect( dev_priv, box );
}
Re: [Dri-devel] Wiki notification emails
José Fonseca wrote: Today another funny guy decided to tamper with the Wiki (this time by applying rot13 to a full page). Thankfully this doesn't happen often (I recall only 3 times so far). Still, since I'm the only one monitoring I'm afraid I could let attacks like pass though. I'm even more concerned about minor edits which could change drastically the meaning of sentences, leading users to do something unapproriate. (It may be a remote possibility that sombody goes through that kind of work, but still perfectly possible). Therefore I'd like to have more eyeballs by directing the nightly diffs emails to a mailing list. I was thinking of dri-patches, since it's just one mail per day, usually 1-2Kb. Below is yesterday's email as an example. Is it OK to send this to dri-patches? I would think so... Keith --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
[Dri-devel] What's up with dri.freedesktop.org? Connection timed out
for several hours, now. ri-trunk/xc cvs update cvs [update aborted]: connect to dri.freedesktop.org(131.252.208.82):2401 failed: Connection timed out dri-trunk/xc ping dri.freedesktop.org PING dri.freedesktop.org (131.252.208.82) 56(84) bytes of data. --- dri.freedesktop.org ping statistics --- 75 packets transmitted, 0 received, 100% packet loss, time 74049ms Thanks, Dieter --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Dri-devel] What's up with dri.freedesktop.org? Connection timed out
Dieter Nützel wrote: for several hours, now. ri-trunk/xc cvs update cvs [update aborted]: connect to dri.freedesktop.org(131.252.208.82):2401 failed: Connection timed out dri-trunk/xc ping dri.freedesktop.org PING dri.freedesktop.org (131.252.208.82) 56(84) bytes of data. Seems down from here too. Being thanksgiving, I don't know when it will come back up... I'm also not sure exactly who's responsible for the box, hence the shotgun cc list... Keith --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
[Dri-devel] Ah Keith, while your up...What about the r200 speed decrease?
Which patch should I test? VTK demos are slower. Haven't ipers et. al (MesaDemos, Q3A etc.) at the moment 'cause my whole system was over the Jordan. Greetings, Dieter --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
[Dri-devel] [r200] What about the TCL color bugs (VTK)?
./PipelineParallelism ./ParallelIso ./ParallelIsoTest etc. Do not show _any_ colors. Only LIBGL_ALWAYS_INDIRECT works. Some weeks ago, R200_NO_TCL worked. Any hints? -Dieter --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Dri-devel] [r200] What about the TCL color bugs (VTK)?
Dieter Nützel wrote: ./PipelineParallelism ./ParallelIso ./ParallelIsoTest etc. Do not show _any_ colors. Only LIBGL_ALWAYS_INDIRECT works. Some weeks ago, R200_NO_TCL worked. Any hints? In general, these sorts of regressions are much easier to fix if they can be tracked down to a specific change which broke things... Keith --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Dri-devel] r200 Xserver hang: strange
Am 2003.11.28 16:15:55 +0100 schrieb(en) Michel Dänzer: On Fri, 2003-11-28 at 13:00, Andreas Stenglein wrote: I found another program using scissors that triggers (maybe) the same bug: lesson24 from nehe.gamedev.net http://nehe.gamedev.net/data/lessons/lesson.asp?lesson=24 (glut and sdl source) The difference is: when triggering the bug with q3a both gl-apps exit() with r200WaitIrq: drmRadeonIrqWait: -16 and the Xserver loops, no cursor-movement... when triggering the bug with lesson24, one program loops and gets #1 in top (99,x% CPU), and you have at least a moving cursor. After killing -KILL the programs, the Xserver becomes #1 - now you have the same result you get with quake3. Sounds like different symptoms of a chip lockup. This patch doesn't happen to make a difference, does it? unfortunately not. The lockup still occurs. [...] It looks like its easier to trigger the lockup if the windows overlap. But overlap isnt necessary. Andreas --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Dri-devel] What's up with dri.freedesktop.org? Connection timed out
On Fri, 2003-11-28 at 18:44, Keith Whitwell wrote: Dieter Ntzel wrote: for several hours, now. ri-trunk/xc cvs update cvs [update aborted]: connect to dri.freedesktop.org(131.252.208.82):2401 failed: Connection timed out dri-trunk/xc ping dri.freedesktop.org PING dri.freedesktop.org (131.252.208.82) 56(84) bytes of data. Seems down from here too. Being thanksgiving, I don't know when it will come back up... I'm also not sure exactly who's responsible for the box, hence the shotgun cc list... Keith From: Keith Packard [EMAIL PROTECTED] To: [EMAIL PROTECTED], [EMAIL PROTECTED] Cc: Keith Packard [EMAIL PROTECTED] Subject: Freedesktop server will be unavailable 0600-0800 PST Nov 27 Date: Fri, 28 Nov 2003 00:33:02 -0800 PSU (our fine hosting provider) is replacing the giant UPS serving the machine room where the freedesktop.org server is mounted. This will cause the machine to be off-line for a few hours tomorrow morning, and then again for a few hours on Saturday evening. I recommend a few hours visit to the big room with the wet gray ceiling during this period; we shouldn't have another opportunity like this for several years if all goes well :-) -keith -- Ronny V. Vindenes [EMAIL PROTECTED] --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
[Dri-devel] Secondary adapter init
When I bring X up dual headed (two cards, not one card two heads) on my system the secondary adapter needs to be initialized. Where does this happen in the code? = Jon Smirl [EMAIL PROTECTED] __ Do you Yahoo!? Free Pop-Up Blocker - Get it now http://companion.yahoo.com/ --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Dri-devel] [r200] What about the TCL color bugs (VTK)?
On Fri, 28 Nov 2003 18:58:46 +0100 Dieter Nützel [EMAIL PROTECTED] wrote: ./PipelineParallelism ./ParallelIso ./ParallelIsoTest etc. Do not show _any_ colors. Only LIBGL_ALWAYS_INDIRECT works. Some weeks ago, R200_NO_TCL worked. With recent drivers either use the configuration file or tcl_mode=0 to disable TCL. The neat thing about the configuration file is that you can disable TCL for specific applications and leave it enabled for the rest. Check out http://dri.sourceforge.net/cgi-bin/moin.cgi/ConfigurationInfrastructure for details. :) Any hints? -Dieter Best, Felix __\|/_____ ___ - Felix ___\_e -_/___/ __\___/ __\_ You can do anything, Kühling (_\Ä// /_/ /) just not everything [EMAIL PROTECTED] \___/ \___/ Uat the same time. --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Dri-devel] AGP 8x support
On Fri, 2003-11-28 at 00:35, Dmitri Katchalov wrote:
Quoting Michel Dnzer [EMAIL PROTECTED]:
On Thu, 2003-11-27 at 15:47, Dmitri Katchalov wrote:
I've made this quick and dirty patch for Radeon driver only.
I understand that it is not perfect as it has to be copied
into every other driver. A better solution is needed IMHO.
I've posted about http://penguinppc.org/~daenzer/DRI/radeon-agp8x.diff
before, which is a similar patch by Hui Yu.
mode = ~RADEON_AGP_MODE_MASK;
-switch (info-agpMode) {
-case 4: mode |= RADEON_AGP_4X_MODE;
-case 2: mode |= RADEON_AGP_2X_MODE;
-case 1: default: mode |= RADEON_AGP_1X_MODE;
+if ((mode RADEON_AGPv3_MODE)
+ (INREG(RADEON_AGP_STATUS) RADEON_AGPv3_MODE)) {
+ switch (info-agpMode) {
+ case 8: mode |= RADEON_AGPv3_8X_MODE;
+ case 4: default: mode |= RADEON_AGPv3_4X_MODE;
+ }
+ /*TODO: need to take care of other bits valid for v3 mode
+ * currently these bits are not used in all tested cards.
+ */
+} else {
+ switch (info-agpMode) {
+ case 4: mode |= RADEON_AGP_4X_MODE;
+ case 2: mode |= RADEON_AGP_2X_MODE;
+ case 1: default: mode |= RADEON_AGP_1X_MODE;
+ }
}
Sorry, don't we need a break (no pun intended) in those switch statements?
I think the idea is that the kernel will fall back to a lower mode like
this if the requested mode isn't possible.
I understand Keith Whitwell
has integrated yet another version on a CVS branch.
I'll have a look. Which CVS branch are you referring to?
Looking at the CVS log of radeon_dri.c, i865-agp-0-1-branch it seems.
--
Earthling Michel Dnzer | Debian (powerpc), X and DRI developer
Software libre enthusiast| http://svcs.affero.net/rm.php?r=daenzer
---
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive? Does it
help you create better code? SHARE THE LOVE, and help us help
YOU! Click Here: http://sourceforge.net/donate/
___
Dri-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/dri-devel
[Dri-devel] [Bug 271] euphoria (really-slick screensavers) repeatably locks up system
Please do not reply to this email: if you want to comment on the bug, go to the URL shown below and enter your comments there. http://bugs.xfree86.org/show_bug.cgi?id=271 [EMAIL PROTECTED] changed: What|Removed |Added CC||[EMAIL PROTECTED] --- Additional Comments From [EMAIL PROTECTED] 2003-11-28 21:13 --- Hi, I dunno if this is related but I get these (EE) R128(0): Idle timed out, resetting engine... (EE) R128(0): R128WaitForIdle: CCE reset -1007 (EE) R128(0): R128WaitForIdle: CCE start -1007 Log entries when I enable DRI in my dual screen configuration with a r128 fury pro (AGP) and a Matrox Millenium G450DH (PCI). This occurs during X startup (which hangs then and heavily writes these Log entries). I didn't find much on this issue yet. Any help would be nice! thanks, ron -- Configure bugmail: http://bugs.xfree86.org/userprefs.cgi?tab=email --- You are receiving this mail because: --- You are the assignee for the bug, or are watching the assignee. --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Dri-devel] MergedFB question.
the 2048 limitation of the scissor registers can be worked around. check the dri-devel archives for messages with 2048 in the subject. basically what we need to do is look at the cliprects in the client driver and everytime they are a multiple of 2048 divide into a zone. then when you are rendering iterate over those zones. I've looked into it, but I haven't had the time dig into the DRI driver enough to fully understand how the current code works and then how best to implement this on top of it. Someday... Anyone can submit a patch if they get to it first :) Also you can run an desktop of 2560x1024 with the option no2048limit (I should just remove that option and just start the xserver with DRI anyway...) and 3D will work fine up to 2048. if you extend beyond 2048, then window just goes blank. Also you can specify metamodes and switch the resolution of both crtcs on the fly. also if you specify clone modes in your metamodes, you can run your desktop at 2560x1024 and then when you start a game it will select one of the clone modes using the vidmode extension and then your game will run fine in 1024x768 or 1280x1024 on both heads. really the only time you will have a problem is if you try and run a 3D window that is larger than 2048; anything smaller works fine. Alex --- Mike Mestnik [EMAIL PROTECTED] wrote: Not at this time, however the work around I have been using is setting up more than one layout in my XF86Config. Then I write scripts for common apps/tasks that call the right xSession and layout. I had a dream that apps would be free from persecution of color depth and of screen/root number/size. That one day layouts could be change on the fly and apps could roam freely from server to server. I'm prepared for X11R7 or X12 if thats what it takes. --- Adam K Kirchhoff [EMAIL PROTECTED] wrote: I'm curious if MergedFB can be modified so that you can run with a resolution of 2560x1024 normally, and then drop it down to 2048x768 if you want to do anything 3D related? Adam __ Do you Yahoo!? Free Pop-Up Blocker - Get it now http://companion.yahoo.com/ --- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
