[Dri-devel] [Bug 44] Radeon Driver hard locks during VT switches/resolution changes 8500DV DRI only
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=44 --- Additional Comments From [EMAIL PROTECTED] 2004-02-04 16:29 --- Hey again, I just wanted to mention, in case someone reccomends it, that I also recently installed XFree86 4.4.0 RC2 (4.3.99.902), and the problem persists. -- 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. --- The SF.Net email is sponsored by EclipseCon 2004 Premiere Conference on Open Tools Development and Integration See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. http://www.eclipsecon.org/osdn -- ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
[Dri-devel] [Bug 44] Radeon Driver hard locks during VT switches/resolution changes 8500DV DRI only
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=44 [EMAIL PROTECTED] changed: What|Removed |Added Status|RESOLVED|REOPENED Resolution|FIXED | --- Additional Comments From [EMAIL PROTECTED] 2004-02-02 04:01 --- Sorry to reopen this guys... many eons after the fact. I'm running the latest DRI snapshot with the CVS XFree86 binary from AlanH's website. Here's my XFree86 -version information: XFree86 Version 4.3.99.16 Release Date: 20 November 2003 X Protocol Version 11, Revision 0, Release 6.6 Build Operating System: Linux 2.4.20-gentoo-r8 i686 [ELF] Current Operating System: Linux pixelmonkey 2.6.1 #2 SMP Tue Jan 27 19:01:39 EST 2004 i686 Build Date: 03 December 2003 Changelog Date: 02 December 2003 Before reporting problems, check http://www.XFree86.Org/ to make sure that you have the latest version. Module Loader present Since I last commented on this bug, I have switched hardware (except for this damn ATI card) to an Intel chipset, P4, etc. I have also upgraded to debian sid and also got my X packages from the X Strike Force's experimental branch. My DRI snapshot is from February 2nd. I downloaded the AlanH stuff today as well. Basically, my mode switching problems still exist. I think it's possible this has fixed the VT Switching problems. But mode switching still causes complete hard lock on my machine, with no way to recover. As before, when my machine locks up, monitor goes blank and I can ssh in, but cannot kill XFree86 for the life of me. To refresh your memory, the locks are happening when I run a 3D-enabled program that tries to go fullscreen with a resolution lower than my desktop. i.e. if I try to run tuxracer at 1024x768 when my desktop is running 1280x1024. I can, as a workaround, make all 3D programs run at my desktop resolution, and crashes NEVER occur. But with some things that means horrible performance. Note that the lock-ups during mode change are erratic. Sometimes they occur, sometimes they don't. But it's definitely true that usually they do occur. It's lucky when it doesn't occur. This is the same as before. So, basically it looks like the CVS modifications and latest work on DRI has fixed the VT switching problem (I think, I'm kinda afraid to test it extensively), but the mode switching problems are still there. Thanks for your time, Andrew -- 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. --- The SF.Net email is sponsored by EclipseCon 2004 Premiere Conference on Open Tools Development and Integration See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. http://www.eclipsecon.org/osdn -- ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
[Dri-devel] [Bug 44] Radeon Driver hard locks during VT switches/resolution changes 8500DV DRI only
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=44 [EMAIL PROTECTED] changed: What|Removed |Added Status|ASSIGNED|RESOLVED Resolution||FIXED --- Additional Comments From [EMAIL PROTECTED] 2003-06-11 21:44 --- reopen if still a problem with current XFree86 CVS -- 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
[Dri-devel] [Bug 44] Radeon Driver hard locks during VT switches/resolution changes 8500DV DRI only
http://bugs.xfree86.org/show_bug.cgi?id=44 --- Additional Comments From [EMAIL PROTECTED] 2003-28-07 02:35 --- Michel, I just noticed this patch on your site: http://penguinppc.org/~daenzer/DRI/radeon-reinit.diff To me, that looks like a patch that might solve this problem. Is that right? -Andrew -- 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 sponsored by: Free pre-built ASP.NET sites including Data Reports, E-commerce, Portals, and Forums are available now. Download today and enter to win an XBOX or Visual Studio .NET. http://aspnet.click-url.com/go/psa0013ave/direct;at.aspnet_072303_01/01 ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
[Dri-devel] [Bug 44] Radeon Driver hard locks during VT switches/resolution changes 8500DV DRI only
http://bugs.xfree86.org/show_bug.cgi?id=44 --- Additional Comments From [EMAIL PROTECTED] 2003-28-07 06:57 --- It won't help for the mode changes, but it's possible that it 'helps' for VT switching back to the server by avoiding the problem. -- 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 sponsored by: Free pre-built ASP.NET sites including Data Reports, E-commerce, Portals, and Forums are available now. Download today and enter to win an XBOX or Visual Studio .NET. http://aspnet.click-url.com/go/psa0013ave/direct;at.aspnet_072303_01/01 ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
[Dri-devel] [Bug 44] Radeon Driver hard locks during VT switches/resolution changes 8500DV DRI only
[This e-mail has been automatically generated.] 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//cgi-bin/bugzilla/show_bug.cgi?id=44 --- Additional Comments From [EMAIL PROTECTED] 2003-07-12 10:15 --- I've read through the bug descriptions, and I'm fairly certain that I have the same (sort) of problem, with slightly different propagation. I have two machines, both with Radeon AIW (Radeon QD). The machine with the AMD 762 / 768 chipset has not had any problem. The machine with the VIA KT266A / VT8233CE chipset, on the other hand, is having serious problems, similar to what has been posted. I use KDM to login to the machines (this happens with XDM, also). When the person logs out and logs in as another user on the VIA machine, the system hangs before it gets back to KDM. When trying to SSH into the box, nothing is there (not available on the network). This situation happens with XFree86 4.3 , the ~alanh drm driver and any kernel I use (currently 2.4.21). If I use the mis-matched kernel drm radeon.o module, the problem goes away, but of course dri is not available. This seems like the old soft-reset problem that the radeon had with the AMD (I think Irongate) chipsets about 2 years ago. But it follows that it is a VIA chipset problem that follows the 'mode change' scenario. --- You are receiving this mail because: --- You are the assignee for the bug, or are watching the assignee. --- This SF.Net email sponsored by: Parasoft Error proof Web apps, automate testing more. Download eval WebKing and get a free book. www.parasoft.com/bulletproofapps1 ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
[Dri-devel] [Bug 44] Radeon Driver hard locks during VT switches/resolution changes 8500DV DRI only
[This e-mail has been automatically generated.] 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//cgi-bin/bugzilla/show_bug.cgi?id=44 --- Additional Comments From [EMAIL PROTECTED] 2003-07-07 13:37 --- Thanks for the comments Roy. I agree with you: I think this is only occuring with certain combinations of mobo/chipsets and video cards. And, to tell you the truth, I wouldn't be surprised if it was Via's fault ;-). I hate that company more and more as I continue to painstakingly use their products. I did some scouring of the web and I found out that Mike Harris @ Redhat was actually working on this problem back when it was popping up for RH users while X was at 4.2. He wrote a patch for radeon_drv that claimed to FIX the problem, and that patch was integrated into the Redhat XFree86 4.2 rawhide. Unfortunately, the patch is no longer available on his site. I did see a bunch of other X patches there... ftp://people.redhat.com/mharris/ I personally have no clue how Redhat development works, but aren't patches like this sent back to the XFree team to be integrated into the source tree/CVS there? Or is this patch currently only available to Redhat users? In the errata that includes the patch, Redhat writes: Fix a problem which caused certain combinations of Radeon and Rage 128 hardware and particular motherboards to hang, due to bus mastering getting disabled when VT switching. Check out the bugzilla bug report here: http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=62171 And the errata here: http://rhn.redhat.com/errata/RHSA-2003-066.html Maybe this sheds some more light on the issue? -Andrew --- You are receiving this mail because: --- You are the assignee for the bug, or are watching the assignee. --- This SF.Net email sponsored by: Free pre-built ASP.NET sites including Data Reports, E-commerce, Portals, and Forums are available now. Download today and enter to win an XBOX or Visual Studio .NET. http://aspnet.click-url.com/go/psa0016ave/direct;at.asp_061203_01/01 ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
[Dri-devel] [Bug 44] Radeon Driver hard locks during VT switches/resolution changes 8500DV DRI only
[This e-mail has been automatically generated.] 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//cgi-bin/bugzilla/show_bug.cgi?id=44 --- Additional Comments From [EMAIL PROTECTED] 2003-07-07 17:45 --- The patch is no longer needed because the fix has been integrated. Still, I've been asking in this very bug about the status of bus mastering when it happens - have you checked it? The fact that it only happens when VT switching back to the server or when switching modes points towards starting the CP as causing the problems. Unfortunately, I don't have any idea what could go wrong there (and it seems to work perfectly on the vast majority of systems). I'm adding Hui Yu to the CC list, hoping he can think of something. --- You are receiving this mail because: --- You are the assignee for the bug, or are watching the assignee. --- This SF.Net email sponsored by: Free pre-built ASP.NET sites including Data Reports, E-commerce, Portals, and Forums are available now. Download today and enter to win an XBOX or Visual Studio .NET. http://aspnet.click-url.com/go/psa0016ave/direct;at.asp_061203_01/01 ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
[Dri-devel] [Bug 44] Radeon Driver hard locks during VT switches/resolution changes 8500DV DRI only
[This e-mail has been automatically generated.] 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//cgi-bin/bugzilla/show_bug.cgi?id=44 --- Additional Comments From [EMAIL PROTECTED] 2003-07-05 15:21 --- Michel, Sorry if I sounded rude before; I think it was the frustration of this problem talking ;) Thanks for taking the time for this relatively minor problem. I think we're getting closer to diagnosing it though. I got some new [testing] information that may prove helpful: a) I tried out Alex's suggestions. AGPMode and FastWrites do not seem to affect the behavior or my system. With the mode at 4 or with the mode at 1, and with FW enabled/disabled, I still get the random black screens of death I was getting before, and I still get the occasional good runs of DRI programs from before. b) PageFlip, OTOH, does affect the behavior or my system. With it enabled, I notice a 400+ frame improvement in glxgears (from 1200-1400 to 1600-2000). With this mode enabled, however, I simply HAVE NOT been able to get into a 3D accelerated program AT ALL. TuxRacer gives me black screens of death immediately, no fail, every time. But 2D work seems to be unaffected. c) I read through the man page for the radeon driver and noticed the ForcePCIMode option. I tried enabling this. Under this mode performance dropped considerably (but Direct Rendering was still enabled). glxgears produced 200 frames per sec. But TuxRacer started EVERY TIME. I tried about 15 consecutive times with no crash. Of course, TuxRacer was running alower than when agpgart was being used, but it definitely was using DRI. The most consecutive starts I could get for TuxRacer with no PageFlips (and no FastWrites, and AGPMode of 1) was four. The 5th time crashed. And the next time I rebooted, the 3rd time was a crash. So there definitely is a difference in behavior here. I am not sure what this all suggests. To me, it seems that if PCIMode works seemingly flawlessly, then there is a problem of the interaction between agpgart and the radeon driver. But I don't know for sure. What do you guys think? -Andrew --- You are receiving this mail because: --- You are the assignee for the bug, or are watching the assignee. --- This SF.Net email sponsored by: Free pre-built ASP.NET sites including Data Reports, E-commerce, Portals, and Forums are available now. Download today and enter to win an XBOX or Visual Studio .NET. http://aspnet.click-url.com/go/psa0016ave/direct;at.asp_061203_01/01 ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
[Dri-devel] [Bug 44] Radeon Driver hard locks during VT switches/resolution changes 8500DV DRI only
[This e-mail has been automatically generated.] 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//cgi-bin/bugzilla/show_bug.cgi?id=44 --- Additional Comments From [EMAIL PROTECTED] 2003-07-04 04:30 --- Who said it was a configuration problem? I certainly didn't. strace output usually isn't very useful to track down this kind of problem. The client stderr output with R200_DEBUG=all (caution, redirect stderr to a file) or the kernel output when loading the DRM with drm_opts=debug is better. And no, this isn't the same problem as on suspend/resume - it's just incomplete initialization of the hardware on resume there. --- You are receiving this mail because: --- You are the assignee for the bug, or are watching the assignee. --- This SF.Net email sponsored by: Free pre-built ASP.NET sites including Data Reports, E-commerce, Portals, and Forums are available now. Download today and enter to win an XBOX or Visual Studio .NET. http://aspnet.click-url.com/go/psa0016ave/direct;at.asp_061203_01/01 ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
[Dri-devel] [Bug 44] Radeon Driver hard locks during VT switches/resolution changes 8500DV DRI only
[This e-mail has been automatically generated.] 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//cgi-bin/bugzilla/show_bug.cgi?id=44 --- Additional Comments From [EMAIL PROTECTED] 2003-07-04 11:59 --- have you tried with a lower apgmode, or turning on/off pageflip or fastwrites? Alex --- You are receiving this mail because: --- You are the assignee for the bug, or are watching the assignee. --- This SF.Net email sponsored by: Free pre-built ASP.NET sites including Data Reports, E-commerce, Portals, and Forums are available now. Download today and enter to win an XBOX or Visual Studio .NET. http://aspnet.click-url.com/go/psa0016ave/direct;at.asp_061203_01/01 ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
[Dri-devel] [Bug 44] Radeon Driver hard locks during VT switches/resolution changes 8500DV DRI only
[This e-mail has been automatically generated.] 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//cgi-bin/bugzilla/show_bug.cgi?id=44 --- Additional Comments From [EMAIL PROTECTED] 2003-07-03 19:28 --- Okay, I have some more info for you guys. I did a stack trace of tuxracer when it WORKED for me, and then I did a stack trace of tuxracer when it DIDN'T WORK for me. You'll see marked differences, most notably that this line: read(3, 0xb630, 32) = -1 EAGAIN (Resource temporarily unavailable) select(4, [3], NULL, NULL, NULL Appears right before the strace output finishes, while the line: read(3, 0xb630, 32) = -1 EAGAIN (Resource temporarily unavailable) Appears 3x in the first stack trace of when tuxracer works. After that point in the trace is information of the game working as it should. But it seems suspect to me that even in the working trace I get a message that say Resource temporarily unavailable and that that message is the one right after which my tuxracer sometimes dies. Perhaps that explains (in part) the eraticness of this all? What is the resource that is unavailable, and how does it go unavailable indefinitely (causing my system to lock up). Perhaps if Resource temporarily unavailable message is debuged out of the drm/dri code, then my problem will go away as well? Here are the strace files: o tuxracer1.txt - pertains to the WORKING tuxracer run --http://www.pixelmonkey.org/pub/tuxracer1.txt o tuxracer2.txt - pertains to the CRASHED tuxracer run --http://www.pixelmonkey.org/pub/tuxracer2.txt Enjoy -Andrew --- You are receiving this mail because: --- You are the assignee for the bug, or are watching the assignee. --- This SF.Net email sponsored by: Free pre-built ASP.NET sites including Data Reports, E-commerce, Portals, and Forums are available now. Download today and enter to win an XBOX or Visual Studio .NET. http://aspnet.click-url.com/go/psa0016ave/direct;at.asp_061203_01/01 ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
[Dri-devel] [Bug 44] Radeon Driver hard locks during VT switches/resolution changes 8500DV DRI only
[This e-mail has been automatically generated.] 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//cgi-bin/bugzilla/show_bug.cgi?id=44 --- Additional Comments From [EMAIL PROTECTED] 2003-07-03 20:28 --- Michel, I don't discount the possibility that this is a red herring. My look over the strace file is uninformed. I only had a guess based on the location of those statements in the file, etc. I don't think just because my analysis of the issue is wrong that you should discount the raw materials, i.e. the two strace files. Surely there must be some information to extract there (perhaps for the informed reader). To reply to the second part--trust me, I _have_ searched through dri-devel's archive, and there are MANY, MANY posts in which people say Listen, Radeon + DRI = disaster... computer lockups, etc. but most all of these posts end with no conclusion, or someone saying, It's all ATI's fault, I just bought an nVidia card and everything is peachy. I'm convinced this is a X-related or DRI/DRM-related bug. A bug, not just a configuration problem. How could it be a configuration problem if it is only the actual VT switch that causes my crash (never does my 3D application crash mid-flight)? I don't know. What do you think? If it is configuration-related, what could it possibly be? --- You are receiving this mail because: --- You are the assignee for the bug, or are watching the assignee. --- This SF.Net email sponsored by: Free pre-built ASP.NET sites including Data Reports, E-commerce, Portals, and Forums are available now. Download today and enter to win an XBOX or Visual Studio .NET. http://aspnet.click-url.com/go/psa0016ave/direct;at.asp_061203_01/01 ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
[Dri-devel] [Bug 44] Radeon Driver hard locks during VT switches/resolution changes 8500DV DRI only
[This e-mail has been automatically generated.] 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//cgi-bin/bugzilla/show_bug.cgi?id=44 --- Additional Comments From [EMAIL PROTECTED] 2003-07-03 20:34 --- Sorry to post again. I just want to point out that the DRI web page has a reference to this very issue in their FAQ which they label Computer hangs up when using DRI + RADEON. http://dri.sourceforge.net/faq/faq_display.phtml?id=49 They also have a link to a solution for the DRI suspend/resume problem. http://cpbotha.net/dri_resume.html Is it possible these two issues are related in any way? -Andrew --- You are receiving this mail because: --- You are the assignee for the bug, or are watching the assignee. --- This SF.Net email sponsored by: Free pre-built ASP.NET sites including Data Reports, E-commerce, Portals, and Forums are available now. Download today and enter to win an XBOX or Visual Studio .NET. http://aspnet.click-url.com/go/psa0016ave/direct;at.asp_061203_01/01 ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
[Dri-devel] [Bug 44] Radeon Driver hard locks during VT switches/resolution changes 8500DV DRI only
[This e-mail has been automatically generated.] 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//cgi-bin/bugzilla/show_bug.cgi?id=44 --- Additional Comments From [EMAIL PROTECTED] 2003-04-05 03:56 --- When entering bug reports, please hit enter after 60 characters or so, so the bug report is readable. Currently it scrolls off my screen to the right 4 times. Just trying to provide a helpful suggestion to improve the chance your problem will be investigated by a developer. --- 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: ValueWeb: Dedicated Hosting for just $79/mo with 500 GB of bandwidth! No other company gives more support or power for your dedicated server http://click.atdmt.com/AFF/go/sdnxxaff00300020aff/direct/01/ ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
[Dri-devel] [Bug 44] Radeon Driver hard locks during VT switches/resolution changes 8500DV DRI only
[This e-mail has been automatically generated.] http://bugs.xfree86.org//cgi-bin/bugzilla/show_bug.cgi?id=44 [EMAIL PROTECTED] changed: What|Removed |Added AssignedTo|[EMAIL PROTECTED] |dri- ||[EMAIL PROTECTED] --- 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: ValueWeb: Dedicated Hosting for just $79/mo with 500 GB of bandwidth! No other company gives more support or power for your dedicated server http://click.atdmt.com/AFF/go/sdnxxaff00300020aff/direct/01/ ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
