Created attachment 25534 [details]
xorg conf file

I just tested mesa_7_5_branch here on my 965 and both pbo and texCube run correctly for me.

This might be a DRI1 vs. DRI2 thing (I'm still on DRI1) or i915 only.  Though I don't know of any differences in the 915 code between mesa_7_5_branch and master.

Yes, this is using UXA/DRI2.
Ian, do you have any clue what's the difference between master and mesa_7_5_branch here?

This bug would only show up with DRI2.  It should be reproducible on either i915 or i965 in that configuration.  In any case, it should be fixed by:

commit 64f36ff9fbe7e12c79cd72ceb68ed5967979445f
Author: Ian Romanick <ian.d.romanick@intel.com>
Date:   Tue Apr 28 10:08:57 2009 -0700

    Test either GL_FRONT_LEFT or GL_FRONT for front-buffer rendering
    
    For non-stereo visuals, which is all we support, we treat
    GL_FRONT_LEFT as GL_FRONT.  However, they are technically different,
    and they have different enum values.  Test for either one to determine
    if we're in front-buffer rendering mode.
    
    This fix was suggested by Pierre Willenbrock.
    
    Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
    (cherry picked from commit 2085cf24628be7cd297ab0f9ef5ce02bd5a006e2)

this issues happens again.I have tested Ian's fix and it works with that commit, and I will keep bisecting which commit cause this regression.

after bisecting ,I think the current failure is different from this bug. and when testing with below commit(after Ian's fix),this issue had gone.so let us close this bug. 

commit 0fc5fa85bf858ba2ad88995f65cc48b2dab1298d
Author: Eric Anholt <eric@anholt.net>
Date:   Tue May 12 10:03:08 2009 -0700

    i915: Fix driver after HW glGenerateMipmap commit.

verified.

verified.

Created attachment 31846 [details]
Xorg.conf file

Created attachment 31847 [details]
Xorg Log

(In reply to comment #0)
> Created an attachment (id=25533) [details]
> xorg.0.log
> 
> System Environment:
> --------------------------
> Host:           x-aspire1
> Arch:           i386
> Platform:       acer aspire1    
> OSD:            Fedora release 10 (Cambridge)
> Kernel_version:         2.6.30-rc4-KMS
> Libdrm:         (master)68103b2758029b3c1fbfcf995baa758bfd2676de
> Mesa:           (mesa_7_5_branch)dd4c142e90a0cba5b445990bb522ce9199d7f565
> Xserver:                (master)1b1b20d6e3e696e4437a9ef56128cde70a485f66
> Xf86_video_intel:              
> (master)a8a771a853478e5f45f71d0eff3c4d55bf24d0ad
> Kernel:       (for-linus)091438dd5668396328a3419abcbc6591159eb8d1
> 
> Bug detailed description:
> -------------------------
> glean cases pbo and texCube get segfault in intel_region_buffer
> (intel=0x8136e30, region=0x0, flag=2)at intel_regions.c:440 recently because we
> switch mesa to 7_5 branch. these two cases worked on mesa 7_4 branch and
> master.
> the backtrace of pbo from gdb is as below:
> 
> #0  intel_region_buffer (intel=0x8136e30, region=0x0, flag=2)
>     at intel_regions.c:440
> #1  0xb7af36d2 in intelClearWithBlit (ctx=0x8136e30, mask=1)
>     at intel_blit.c:493
> #2  0xb7aea75f in intelClear (ctx=0x8136e30, mask=1) at intel_clear.c:365
> #3  0xb7b40bae in _mesa_Clear (mask=0) at main/clear.c:182
> #4  0x080817a2 in GLEAN::PBOTest::testBitmap (this=0x8105400) at tpbo.cpp:455
> #5  0x0807fe05 in GLEAN::PBOTest::testFunctionality (this=0x8105400,
>     r=@0x816ac88) at tpbo.cpp:1083
> #6  0x0807fe71 in GLEAN::PBOTest::runSubTests (this=0x8105400, r=@0x816ac88)
>     at tpbo.cpp:1219
> #7  0x08082996 in GLEAN::PBOTest::runOne (this=0x8105400, r=@0x816ac88,
>     w=@0xbffff354) at tpbo.cpp:1236
> #8  0x0805ac36 in GLEAN::BaseTest<GLEAN::MultiTestResult>::run (
>     this=0x8105400, environment=@0xbffff3e0) at tbase.h:313
> #9  0x0805466b in main (argc=6, argv=0xbffff524) at main.cpp:129
> 
> Reproduce steps:
> ----------------
> 1. xinit&
> 2. run glean cases
> 

(In reply to comment #9)
> verified.
> 

I think I have the same issue. I got it while running an OpenCascade based application.

System Environment:
--------------------------
Arch:           i386    
OSD:            Ubuntu Karmic Koala (9.10)
Kernel_version:         2.6.31-16-generic


Here is the backtrace
                                                         
#0  0xb17fe7c0 in intel_region_buffer () from /usr/lib/dri/i965_dri.so
#1  0xb17ef545 in intelClearWithBlit () from /usr/lib/dri/i965_dri.so 
#2  0xb17f2f77 in ?? () from /usr/lib/dri/i965_dri.so                 
#3  0xb1866b9c in _mesa_Clear () from /usr/lib/dri/i965_dri.so        
#4  0x00c8f343 in TelInitWS () from /opt/OpenCASCADE6.3.0/lib/libTKOpenGl.so.0
#5  0x00c8cd40 in call_subr_open_ws () from /opt/OpenCASCADE6.3.0/lib/libTKOpenGl.so.0
#6  0x00c9c7be in call_togl_view () from /opt/OpenCASCADE6.3.0/lib/libTKOpenGl.so.0   
#7  0x00c527cd in OpenGl_GraphicDriver::View(CALL_DEF_VIEW&) ()                       
   from /opt/OpenCASCADE6.3.0/lib/libTKOpenGl.so.0                                    
#8  0x054b54ce in Visual3d_View::SetWindow(Handle_Aspect_Window const&) ()            
   from /opt/OpenCASCADE6.3.0/lib/libTKV3d.so.0                                       
#9  0x0549d06c in V3d_View::SetWindow(Handle_Aspect_Window const&) ()                 
   from /opt/OpenCASCADE6.3.0/lib/libTKV3d.so.0                                       
#10 0x05d9662d in Display3d::Init(int) ()                                             
   from /usr/local/lib/python2.6/dist-packages/OCC/_Visualization.so                  
#11 0x05d95641 in _wrap_Display3d_Init ()

(In reply to comment #9)
> verified.
> 

I think I got the same issue. 

System Environment:
--------------------------
Arch:           i386    
OSD:            Ubuntu Karmic Koala (9.10)
Kernel_version: 2.6.31-16-generic

Here is the backtrace                                                     
#0  0xb17fe7c0 in intel_region_buffer () from /usr/lib/dri/i965_dri.so
#1  0xb17ef545 in intelClearWithBlit () from /usr/lib/dri/i965_dri.so 
#2  0xb17f2f77 in ?? () from /usr/lib/dri/i965_dri.so                 
#3  0xb1866b9c in _mesa_Clear () from /usr/lib/dri/i965_dri.so        
#4  0x00c8f343 in TelInitWS () from /opt/OpenCASCADE6.3.0/lib/libTKOpenGl.so.0

I hope something can be done.

Sorry for the repetition in my previous comment. Something went wrong when committing the first time.

Additional info
OpenGL version string: 2.1 Mesa 7.6

However, I must mention that, even though the backtrace I get seems to reveal identical problem as the initial one, I couldn't reproduce it with the proposetd steps :
Indeed 
>> glean -r results-dir -t +pbo 
doesn't crash at all.

Note, that I got glean from piglit project, so maye the sources differ from the original glean.
Loic

Please open your own bug for your own issue.  You've now confused your potentially interesting bug report with an already solved bug report.  And in your report, we'll want the information requested in http://intellinuxgraphics.org/how_to_report_bug.html so we can figure out how to reproduce and fix your bug.