Re: [Bug 21542] Radeon: No 50Hz (PAL) modes available after upgrading to 2.6.36
> I finally did the bisect and got this as a result after a lot of rebooting: > > " > 139315796778a6d5f67c644e2ff470ddc69efb7b is the first bad commit > commit 139315796778a6d5f67c644e2ff470ddc69efb7b > Author: Adam Jackson > Date: Tue Aug 3 14:38:19 2010 -0400 <...> > Maybe this should be changed to Video(Other) as this probably applies to all > graphic cards? > > Do you need the bisect log? > thanx. some time ago i've also noticed i can't set up 50hz modes on various hw. while previously i could, so i can confirm it affects other hw. 50hz modes are esp. usefull on analog LCD screens (slows down pixel clock, so improves sharpness) , slow hw (bit less stress on VRAM), and when using programs like vice , fuse, or other sw. emulating PAL computer with exact refresh rate (i do not know why, but they work much slower and stutter - probably because vblank signal is firing too often, trigering update which could be deferred a bit more). -- -- The modern datacenter depends on network connectivity to access resources and provide services. The best practices for maximizing a physical server's connectivity to a physical network are well understood - see how these rules translate into the virtual world? http://p.sf.net/sfu/oracle-sfdevnlfb -- ___ Dri-devel mailing list Dri-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: mesa 7.7.1 and 7.8.1 fails to compile under uclibc
On Thu, 27 May 2010, Brian Paul wrote: >> On Thu, 27 May 2010, Brian Paul wrote: >> > > Piotr Gluszenia Slawinski wrote: >> > > > > > > hello. problem as in topic. >> > > > > > > http://83.18.299.190/uclibc/ contain files which might add >> > > > > > > some light >> > > > > > > to subject :) >> > > > > > > >> > > > > > > it seems that problem is that _GNU_SOURCE is passed on (i >> > > > > > > know it from >> > > > > > > irc chat with someone bit more skilled) >> > > > > > > >> > > > > > > previous mesa versions compiled fine with uclibc >> > > > > >> > > > > http://pastebin.com/g20gCNxe >> > > > > >> > > > > following modification seems to fix the problem >> > > >> > > Did you forget an attachment? >> > > >> > > I don't see a difference in the code at the given URL. >> > > >> > > -Brian >> >> basically 'fix' (or rather quick hack) goes down to line : >> >> #if defined(_GNU_SOURCE) && !defined(__UCLIBC_MAJOR__) >> >> it should be ~801 line of original main/imports.c . >> >> while discussing it on #uclibc, conclusion was drawn that this is >> not very ellegant solution - program should check if locale are available, >> they can be both existant in uclibc, or missing in glibc (or elibc, libc5 >> , etc. implementation) - it is not considered 'given' part of libc. >> >> current 'fix' "works for me" though and allows at least uclibc users enjoy >> mesa :) > > Can you send a regular patch? I still don't understand what needs to be > changed. esna media-libs # diff mesa-7.8.1/work/Mesa-7.8.1/src/mesa/main/imports.c mesa-7.8.1-uclibc/work/Mesa-7.8.1/src/mesa/main/imports.c 802c802 < #ifdef _GNU_SOURCE --- > #if defined(_GNU_SOURCE) && !defined(__UCLIBC_MAJOR__) -- -- -- ___ Dri-devel mailing list Dri-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: mesa 7.7.1 and 7.8.1 fails to compile under uclibc
-- On Thu, 27 May 2010, Brian Paul wrote: > Piotr Gluszenia Slawinski wrote: >> > hello. problem as in topic. >> > http://83.18.299.190/uclibc/ contain files which might add some light >> > to subject :) >> > >> > it seems that problem is that _GNU_SOURCE is passed on (i know it from >> > irc chat with someone bit more skilled) >> > >> > previous mesa versions compiled fine with uclibc >> >> http://pastebin.com/g20gCNxe >> >> following modification seems to fix the problem > > Did you forget an attachment? > > I don't see a difference in the code at the given URL. > > -Brian basically 'fix' (or rather quick hack) goes down to line : #if defined(_GNU_SOURCE) && !defined(__UCLIBC_MAJOR__) it should be ~801 line of original main/imports.c . while discussing it on #uclibc, conclusion was drawn that this is not very ellegant solution - program should check if locale are available, they can be both existant in uclibc, or missing in glibc (or elibc, libc5 , etc. implementation) - it is not considered 'given' part of libc. current 'fix' "works for me" though and allows at least uclibc users enjoy mesa :) -- -- -- ___ Dri-devel mailing list Dri-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: mesa 7.7.1 and 7.8.1 fails to compile under uclibc
> hello. problem as in topic. > http://83.18.299.190/uclibc/ contain files which might add some light > to subject :) > > it seems that problem is that _GNU_SOURCE is passed on (i know it from > irc chat with someone bit more skilled) > > previous mesa versions compiled fine with uclibc http://pastebin.com/g20gCNxe following modification seems to fix the problem -- -- -- ___ Dri-devel mailing list Dri-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel
mesa 7.7.1 and 7.8.1 fails to compile under uclibc
hello. problem as in topic. http://83.18.299.190/uclibc/ contain files which might add some light to subject :) it seems that problem is that _GNU_SOURCE is passed on (i know it from irc chat with someone bit more skilled) previous mesa versions compiled fine with uclibc -- -- -- ___ Dri-devel mailing list Dri-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: mga drm compiled into kernel crashes hibernation
On Sun, 28 Mar 2010, Piotr Gluszenia Slawinski wrote: > On Sat, 27 Mar 2010, Piotr Gluszenia Slawinski wrote: > >> On Sat, 27 Mar 2010, Piotr Gluszenia Slawinski wrote: >> >>> kernel 2.6.32-tuxonice-r5 >>> >>> with matrox drm component compiled into kernel - hibernation >>> halts at 'doing atomic copy&restore' >>> >>> compiled as module - works fine. >> >> actually does not work fine. trying to run any glx app after >> un-hibernating results in locking of X >> and constant printout of >> "[drm:drm_lock_take] *ERROR* 1 holds heavyweight lock" >> to system log. >> >> killing X with signal 9 is possible then, but X do not restart : >> old , unchanged data pops-up back after restart, and one cannot move >> mouse nor press keys. > > also if one disables DRI (naively hoping this will somewhat help) > machine fails to un-hibernate no matter what. > > *sigh*. > > p.s. i've did some more dumpster diving and i've got nvidia riva tnt2. > suprisingly it works fine (except 3d, as it's 'legacy' drivers do not > compile anymore, so using 'nv' driver) and another side note - after trying various kernels and configs, it seems to be somehow related to SMP being enabled, or page allowance setting of tux on ice. disabling SMP, increasing page allowance setting, and enabling ACPI subsystem debug seems to allow hibernating of riva TNT quite reliably. i've quit debugging further, as after searching for some cheap spares i've got old agp nvidia quadro which works pretty seamless. -- Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev -- ___ Dri-devel mailing list Dri-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: mga drm compiled into kernel crashes hibernation
On Sat, 27 Mar 2010, Piotr Gluszenia Slawinski wrote: > On Sat, 27 Mar 2010, Piotr Gluszenia Slawinski wrote: > >> kernel 2.6.32-tuxonice-r5 >> >> with matrox drm component compiled into kernel - hibernation >> halts at 'doing atomic copy&restore' >> >> compiled as module - works fine. > > actually does not work fine. trying to run any glx app after > un-hibernating results in locking of X > and constant printout of > "[drm:drm_lock_take] *ERROR* 1 holds heavyweight lock" > to system log. > > killing X with signal 9 is possible then, but X do not restart : > old , unchanged data pops-up back after restart, and one cannot move > mouse nor press keys. also if one disables DRI (naively hoping this will somewhat help) machine fails to un-hibernate no matter what. *sigh*. p.s. i've did some more dumpster diving and i've got nvidia riva tnt2. suprisingly it works fine (except 3d, as it's 'legacy' drivers do not compile anymore, so using 'nv' driver) p.s.2 - can't those non-functional cards being masked in portage? as i smell that they will not be fixed anywhere in future... it is a bit confusing to user when he gets stunning choice of totally bitrot drivers... also for users of those old devices - perhaps some 'good, old, and verified to work' Xorg/X11 builds overlay could be handy? i.e. the matrox card works fine (did not checked hibernation though, but at least 3d works) with knoppix 6.2 (so not so long ago) savage works with Xorg 1.4 (also with 3d...) and stuff like riva tnt2 also had it's working Xorg+drivers pair somewhere in history AFAIR. this would make also bisecting to point where drivers got broken esier, which would allow at least drawing some sketch TODO for brave ones willing to fix the drivers in future... greetings -- -- Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev -- ___ Dri-devel mailing list Dri-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: mga drm compiled into kernel crashes hibernation
On Sat, 27 Mar 2010, Piotr Gluszenia Slawinski wrote: > kernel 2.6.32-tuxonice-r5 > > with matrox drm component compiled into kernel - hibernation > halts at 'doing atomic copy&restore' > > compiled as module - works fine. actually does not work fine. trying to run any glx app after un-hibernating results in locking of X and constant printout of "[drm:drm_lock_take] *ERROR* 1 holds heavyweight lock" to system log. killing X with signal 9 is possible then, but X do not restart : old , unchanged data pops-up back after restart, and one cannot move mouse nor press keys. -- -- Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev -- ___ Dri-devel mailing list Dri-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: mga drm compiled into kernel crashes hibernation
> also worth noting is that compiling via agpgart support into kernel > statically causes problem during un-hibernation (compiled as module > works fine) to clarify, whole agpgart must be compiled as module, not only driver, to make un-hibernation functional -- Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev -- ___ Dri-devel mailing list Dri-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel
mga drm compiled into kernel crashes hibernation
kernel 2.6.32-tuxonice-r5 with matrox drm component compiled into kernel - hibernation halts at 'doing atomic copy&restore' compiled as module - works fine. also worth noting is that compiling via agpgart support into kernel statically causes problem during un-hibernation (compiled as module works fine) -- -- Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev -- ___ Dri-devel mailing list Dri-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel
bugs in 3d display of matrox g450 identical to those of savage
Hello. Unsatisfied with current savage support , I've done more 'dumpster diving' and found much better computer, with AGP support. I've put matrox g450 dual head card inside, and it booted knoppix just fine, with (partial, no composite, but well) 3d acceleration, and glxgears happily working. BUT to my utter disappointment, after installing disk with gentoo, GLX acceleration did not work. everything seems to be detected, much like with savage chip on previous computer, but glxgears window remain blank, google earth starts up and displays earth and textures, but displays no labels, nor any text really (large white rectangles get displayed instead) i've upgraded to whatever is bleeding edge in gentoo (libdrm, mesa, xorg) so i just notify the issue . i will try to downgrade/bisect today, but as i've tried same thing without luck with savage it sounds i've hit same bug and there will be little luck in downgrades... -- -- Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev -- ___ Dri-devel mailing list Dri-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel
savage4 + dri fails to un-hibernate
with dri enabled, after un-hibernating machine mouse cursor gets split into individual lines , and can be moved only in corner of the screen. keyboard becomes unresponsive, one cannot switch to vt. with dri disabled, un-hibernating is fine. -- -- Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev -- ___ Dri-devel mailing list Dri-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: savage4 crashes
On Tue, 9 Mar 2010, Piotr Gluszenia Slawinski wrote: > > On Tue, 9 Mar 2010, James Simmons wrote: > >> >>> i've tried to run this old X server with new (2.6.32) kernel >>> >>> here is log from startup : >>> >>> http://83.18.229.190/Xorg/Xorg.0.log.old_X_new_kernel >>> >>> what is stunning - DRI works , and i get stunning 360fps >>> with gears turning _smoothly_ in glxgears. >>> >>> downsides - switching to VT doesn't work. >>> one can switch, keyboard works, but there is mess displayed >>> (white screen or remains of past console). >>> one can still type and execute commands (but nothing is displayed) >> >> Are you using the savage fbdev driver or vga console? >> > > > i was using savage fbdev. > i've removed it, recompiled kernel, > and now Xorg starts with DRI enabled > on one display! :) (other has still videoram > detected improperly to just 2048kb, instead of 8192) > kde starts just fine. > > also vt switches work now. > i am using plain vga console now, i've disabled > framebuffer support completely. > > > there is remaining problem with glx - glxgears show > even more stunning 600fps , but there is just blank void > staring at me from glxgears window... > > i recall from 'old times' there was some bug in mesa, > so i will try to downgrade, i've tried 7.7 and 7.5 so far. i've downgraded libdrm and mesa, rebuilt X and savage driver, and glxgears is still just black void. i can't downgrade below mesa 7.5.2 (no gentoo ebuild) also Xv is still broken. -- -- Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev -- ___ Dri-devel mailing list Dri-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: savage4 crashes
On Tue, 9 Mar 2010, James Simmons wrote: > >> i've tried to run this old X server with new (2.6.32) kernel >> >> here is log from startup : >> >> http://83.18.229.190/Xorg/Xorg.0.log.old_X_new_kernel >> >> what is stunning - DRI works , and i get stunning 360fps >> with gears turning _smoothly_ in glxgears. >> >> downsides - switching to VT doesn't work. >> one can switch, keyboard works, but there is mess displayed >> (white screen or remains of past console). >> one can still type and execute commands (but nothing is displayed) > > Are you using the savage fbdev driver or vga console? > i was using savage fbdev. i've removed it, recompiled kernel, and now Xorg starts with DRI enabled on one display! :) (other has still videoram detected improperly to just 2048kb, instead of 8192) kde starts just fine. also vt switches work now. i am using plain vga console now, i've disabled framebuffer support completely. there is remaining problem with glx - glxgears show even more stunning 600fps , but there is just blank void staring at me from glxgears window... i recall from 'old times' there was some bug in mesa, so i will try to downgrade, i've tried 7.7 and 7.5 so far. -- Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev -- ___ Dri-devel mailing list Dri-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: savage4 crashes
-- On Mon, 8 Mar 2010, Alex Deucher wrote: > On Mon, Mar 8, 2010 at 12:42 PM, Piotr Gluszenia Slawinski > wrote: >> i have dual head sa(l)vage(d) MX card. >> >> it works only in vesa mode, or with dri disabled. >> >> i am trying latest Xorg and mesa avail from gentoo ebuilds. >> >> also Xv doesn't work (tested with plain xawtv) - data is displayed >> not in proper window, but as many rectangles with empty holes, >> scattered over almost half of the screen (they move when one moves window) >> >> when one enables DRI - Xorg starts, xdm allows login, but it crashes >> in the middle of kde startup. >> >> i've tried changing almost all options i was able to find, >> and no help. >> >> EXA is not working either (Xorg starts, but there are no window >> decorations) >> >> >> peculiar thing is that not so long ago (perhaps year...) >> i've tried to use the card on exactly same computer, >> and it worked fine, including 3d. >> >> now only thing which differs is that >> i use fresh gentoo install. >> hopefully i've kept old gentoo on backups somewhere, so i will try to >> restore it - but bit later (today i've wasted WHOLE DAY hoping to make it >> work...) > > It's likely the driver has bitrotten since no one is actively working > on it anymore. i've tried to run this old X server with new (2.6.32) kernel here is log from startup : http://83.18.229.190/Xorg/Xorg.0.log.old_X_new_kernel what is stunning - DRI works , and i get stunning 360fps with gears turning _smoothly_ in glxgears. downsides - switching to VT doesn't work. one can switch, keyboard works, but there is mess displayed (white screen or remains of past console). one can still type and execute commands (but nothing is displayed) so . there was certainly something broken in kernel's drm which made savage driver to be sluggish (on old kernel one could enjoy at just ~140fps). now there is new bug in kernel which makes VT switching broken (it is fine on old kernel) i can also start kde out of chroot'ed 'new' gentoo (the one which has new Xorg build) and it starts just fine, and works pretty smoothly (with plasma desktop it's quite visible , even moving things around desktop seems smoother, not to mention switching workspaces...) so it sounds rest of problems are due to 'bitrot' of savage drivers... is there still anyone able to fix it ? :/ what i could do to help it happen? -- Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev -- ___ Dri-devel mailing list Dri-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: savage4 crashes
On Mon, 8 Mar 2010, Alex Deucher wrote: > On Mon, Mar 8, 2010 at 12:42 PM, Piotr Gluszenia Slawinski > wrote: >> i have dual head sa(l)vage(d) MX card. >> >> it works only in vesa mode, or with dri disabled. >> >> i am trying latest Xorg and mesa avail from gentoo ebuilds. >> >> also Xv doesn't work (tested with plain xawtv) - data is displayed >> not in proper window, but as many rectangles with empty holes, >> scattered over almost half of the screen (they move when one moves window) >> >> when one enables DRI - Xorg starts, xdm allows login, but it crashes >> in the middle of kde startup. >> >> i've tried changing almost all options i was able to find, >> and no help. >> >> EXA is not working either (Xorg starts, but there are no window >> decorations) >> >> >> peculiar thing is that not so long ago (perhaps year...) >> i've tried to use the card on exactly same computer, >> and it worked fine, including 3d. >> >> now only thing which differs is that >> i use fresh gentoo install. >> hopefully i've kept old gentoo on backups somewhere, so i will try to >> restore it - but bit later (today i've wasted WHOLE DAY hoping to make it >> work...) > > It's likely the driver has bitrotten since no one is actively working > on it anymore. > > Alex > i've found disk with backup :) first thing - switching to console and back to X work fine. (it is broken in recent versions) then - it's lightning fast compared to recent version (scrolling windows, moving stuff around, etc) stuff displyaing and scrolling in console does not slow down whole computer (as it happens with recent drivers/Xorg) as result whole machine works significantly FASTER. glxgears display something. it is not updating smoothly, but there is ~100fps rating, and _something_ gets displayed. also, both 'heads' get proper ram amount detected, which is 8192k. in recent driver, just one head (first) gets proper amount detected, other gets just 2048 - no idea why (i was even afraid card got somehow damaged...) i'm on my way to test xv, but i somehow suspect it will also work fine. here log of old driver can be seen : http://83.18.229.190/Xorg/Xorg.0.log.old_one -- -- Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev -- ___ Dri-devel mailing list Dri-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel
savage4 crashes
i have dual head sa(l)vage(d) MX card. it works only in vesa mode, or with dri disabled. i am trying latest Xorg and mesa avail from gentoo ebuilds. also Xv doesn't work (tested with plain xawtv) - data is displayed not in proper window, but as many rectangles with empty holes, scattered over almost half of the screen (they move when one moves window) when one enables DRI - Xorg starts, xdm allows login, but it crashes in the middle of kde startup. i've tried changing almost all options i was able to find, and no help. EXA is not working either (Xorg starts, but there are no window decorations) peculiar thing is that not so long ago (perhaps year...) i've tried to use the card on exactly same computer, and it worked fine, including 3d. now only thing which differs is that i use fresh gentoo install. hopefully i've kept old gentoo on backups somewhere, so i will try to restore it - but bit later (today i've wasted WHOLE DAY hoping to make it work...) -- -- Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev -- ___ Dri-devel mailing list Dri-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel
