Skip to contentSkip to conversation

xf86-video-intel 2.11.901

ANNOUNCE Carl Worth 11 participants 59 messages Jun 15, 2010 – Jul 1, 2010

Subsystems (2)
lore
Share & tools
lore

Carl Worth

author
#0 lore raw

This is the first release candidate in preparation for the upcoming 2.12.0 release. We will appreciate any feedback we can get from testing of this snapshot to improve the 2.12.0 release.

-Carl

Signature · 268 lines
[email protected]

Where to obtain xf86-video-intel 2.11.901
-----------------------------------------
git tag: 2.11.901

http://xorg.freedesktop.org/archive/individual/driver/xf86-video-intel-2.11.901.tar.bz2
MD5: 03b989f0a910c3a5966cbcf7c059a8ef  xf86-video-intel-2.11.901.tar.bz2
SHA1: 51f657f8fc0e2a9d9ebd7355044c25c45c3714e6  xf86-video-intel-2.11.901.tar.bz2

http://xorg.freedesktop.org/archive/individual/driver/xf86-video-intel-2.11.901.tar.gz
MD5: a9da0890207cb4af23f76a2334f27b09  xf86-video-intel-2.11.901.tar.gz
SHA1: 05f77c8a170fc203efb0cf7f34784e44e3d81e58  xf86-video-intel-2.11.901.tar.gz

New requirements compared to 2.10
---------------------------------
 * Libdrm >= 2.4.19

Major performance improvements
------------------------------
There have been many performance improvements targeting the
915G/945G/Pineview graphics chipsets (thank to Chris Wilson!). Areas
improved include:

 * Major improvements to glyph rendering

 * Avoid software fallback when using extended desktops, (for example,
   video playback)

 * Better performance when using large windows, (for example with
   firefox scrolling)

 * Avoiding migration ping-pong with gigantic drawing

The following speedups (measured with cairo-perf-trace/cairo-traces)
capture some of the improvements. These were measured on a Pineview
system with a very recent X server, (from git just before 1.8.1). This
compares how this 2.12 release candidate improves performance compared
to 2.11.0:

	Trace			Speedup
	-----			-------
	ocitysmap		5.41x speedup
	firefox-talos-gfx       2.41x speedup
	gnome-terminal-vim      2.15x speedup
	poppler                 1.86x speedup
	firefox-planet-gnome	1.43x speedup
	midori-zoomed		1.39x speedup
	swfdec-giant-steps      1.36x speedup
	xfce4-terminal-a1	1.17x speedup
	evolution		1.09x speedup
	swfdec-youtube		1.06x speedup

Other improvements
------------------
Daniel Vetter improved Xvmc and overlay, adding GEM support.

Bug fixes
---------
There have been many conformance improvements targeting the
915G/945G/Pineview graphics chipsets. The rendercheck test suite
passes once again and there has been a significant reduction in the
number of errors detected by the cairo test suite (notably center
sampling and EXTEND_NONE behaviors are fixed).

Other notable bug fixes include:

 * Fix for bug 28446 Garbled fonts with Mathematica
   https://bugs.freedesktop.org/show_bug.cgi?id=28446

 * Fixes for page-flipping and other DRI2-handling bugs, (Thanks Jesse
   Barnes, and Kristian Høgsberg)

 * Avoid corruption when using extended desktops, (for example, video
   playback)

Log of all changes from 2.11.0 to 2.11.901
------------------------------------------
Carl Worth (3):
      Fix to require libdrm 2.4.19 or newer.
      NEWS: Add notes for 2.11.901
      Update version to 2.11.901

Chris Wilson (124):
      uxa: Add fallback warnings for PutImage.
      Review i830_pad_drawable_width()
      uxa: Try using put_image when copying from a memory buffer.
      Revert "uxa: Add fallback warnings for PutImage."
      i830: Use pixman_blt directly for performing the in-memory copy
      Revert "Revert "uxa: Try using put_image when copying from a memory buffer.""
      i830: Allocate bo's for glyphs larger than 32x32.
      uxa: Reuse in-flight bo
      i830: tidy in flight bo reuse.
      i830: if pixman_blt() fails fallback to fbCopyArea()
      i830: prepare the uxa pixmap for fbCopyArea.
      uxa: Extract sub-region from in-memory buffers.
      uxa: Transform composites with a simple translation into a blit
      uxa: Rearrange checking and preparing of composite textures.
      i915: Use 1x1R pixmap for solid drawables
      uxa,i915: Handle SourcePict through uxa_composite()
      i830: Update stride when swapping bo for PutImage
      i830: Minor cleanup
      uxa: Protect against valid SourcePict in uxa_acquire_mask()
      i915: Fix pixmap based masks.
      uxa: Recheck texture after acquiring pattern.
      i965: Add texformats mapping for additional pixman formats
      i830: Do not cache in-flight non-reusable buffers.
      i830: Use set_pixmap_bo() instead of open-coding.
      i830: Never create a bo for depth=1 pixmaps.
      drmmode: Reorder i830_set_pixmap_bo() so that the correct stride is used.
      xvmc: Build fix with -pedantic
      debug: Add names for operators
      uxa: solid rects
      uxa: Use accelerated PutImage for uploading pixman images.
      i830: Prevent allocation of bo larger than half the aperture
      uxa: Fallback when compositing to a !offscreen destination
      uxa: Avoid ping-pong with !offscreen destination and traps
      uxa: Avoid glyph ping-pong with !offscreen destination
      dri2: Handle reference counting across page flipping
      i915: Add a2r10g10b10 format and friends
      i915: Force output alpha to 1. if dst has no alpha channel.
      uxa: Fallback early if compositing with alphaMaps
      uxa: Replace source for CLEAR with a transparent solid
      uxa: Only use solid_fill for SRC.
      uxa: Convert 1x1R back to solid_fill
      dri: Hold reference to buffers across swap
      i830: Teardown batch entries on reset.
      drm: don't overwrite the old intel->front_buffer
      drm: Remove unused old_front parameter from drmmode_do_pageflip.
      drm: Remove pin(); unpin() sync
      uxa: enable solid rects for backends that require pixmaps
      i915: Don't force alpha=1 for RGB drawables in the shader.
      Split the prepare blitter functions into check + prepare.
      uxa: Parse BGRA pixel formats.
      i915: Remove a couple of unsupported 16bpp no-alpha tex formats
      uxa: Disable compatible src xrgb and dst argb
      i915: Load texture into directly into OC when possible.
      uxa: Call check_solid before running the solid blitter.
      uxa: Only use 1x1R as a solid with an opaque format or SRC
      uxa: Expand the range of compatible formats to cover all bpp.
      uxa: Fix order of conditionals to only run fill_region for SRC or opaque
      i830: Remove incorrectly mapped tex formats.
      i830: Remove xrgb conversion to argb, no longer required.
      uxa: Check the w-scaling component is 1 for an translation matrix
      uxa: Avoid using blits when with PictFilterConvolution
      uxa: Apply clip for solid rectangles.
      uxa: Force an alpha channel when rendering source fallbacks
      uxa: Patterns are acquired at 0,0
      uxa: Tidy uxa_solid_rects()
      i830: Encode surface bpp into format
      uxa: Replace solid planemask [0xffffffff] with FB_ALLONES
      dri: Check error code from GetScratchGC()
      i830: Remove vestigal debugging ALWAYS_FLUSH and ALWAYS_SYNC
      uxa: Spans! OMG!
      i915: amalgamate composite into a single primitive list
      i915: Emit composite primitive with specialised functions.
      Kill paranoid assertions on every write into the batchbuffer.
      Use pwrite to upload the batch buffer
      i915: Move vertices into a vertex buffer object.
      i915: Add special case primitive emitters for glyphs.
      uxa: translate the region in line for composites
      i915: compute normalized texcoords using a scale factor.
      uxa: Composite glyphs directly onto dst when possible.
      uxa: Use temporary dest when target is too large for compositor
      i915: Emit CA over using OutReverse + Add passes
      uxa: Apply the drawable offset to the solid rects
      i830: Skip an empty fill.
      uxa: Add PICT format mapping for depth 4 pixmaps.
      uxa: Use Composite rather than solid blitter for PolyRect
      uxa: Force the alpha value to 0xffff when treating Over as Src
      uxa: Perform manual damage for CompositeRects
      i965: Remove ATOMIC_BATCH.
      Add vertex bo to the list of buffers to be torn down.
      Decouple non-reusuable pixmaps from batch lists on unref.
      uxa: Use a glyph private rather than a hash table.
      uxa: Make the glyph caches' fixed size explicit.
      uxa: Skip the redundant miComputeCompositeRects() when adding to the mask
      i915: Don't re-emit vertex size unless it has changed.
      dri: Use size from backing pixmap when creating buffers.
      Revert "dri: Use size from backing pixmap when creating buffers."
      dri: Only flip if the front and back pixmaps match.
      dri: Compilation fix.
      i915: Support textured video on an extended desktop.
      uxa: Mega-Glyphs!
      Disable acceleration if we detect a hardware error.
      uxa: Sign reversal on early break from spans passing the YXband
      uxa: Spans, try again to get the early break correct.
      i915; Avoid the implicit flush on changing BUF_INFO
      i915: Centre sampling.
      i830: Remove unused coord-adjust.
      dri: Protect against NULL dereference following GPU hang.
      uxa: Force fallback for copies.
      Adapt glyphs for changes in devPrivates API
      Use the direct dixGevPrivate() API when available
      i915: Replace structure passing with macros for shader generation.
      i915: Remove screen size limit from video setup.
      i915: Fix typo from previous commit.
      uxa: Setup acceleration functions prior to the damage layer
      implicit-flush
      xp:trapezoids
      uxa: Handle all-clipped out case with destination glyphs.
      Revert "xp:trapezoids"
      Fallback implementation for trapezoids for hung GPUs.
      i830: Only emit the disabling GPU error message once.
      i830: Limit disabling acceleration following EIO to !i965
      uxa: Always clip glyphs to destination.
      Compile fix for alternate list.h from xserver-1.9
      i965: Sanity check ComponentAlpha status in prepare_composite

Daniel Vetter (35):
      XvMC: everyone's using execbuffer!
      i915 XvMC: kill more subpict support remnants
      XvMC: merge *_hwmc.c into i830_hwmc.c
      XvMC: unify PutImage handling
      XvMC: unify subpicture functions
      i965 XvMC: don't create any surface state in the ddx
      XvMC: unify surface functions
      XvMC: clean up i830_hwmc.c
      XvMC: unify destroy context function
      XvMC: move i915 specific stuff to common context
      XvMC: move i965 specific stuff to common context
      XvMC: init driver in one function call
      XvMC: directly set driver type in context creation
      XvMC: dynamically allocate adaptor
      XvMC: kill intel_xvmc_command
      libXvMC: derive driver surfaces from struct intel_xvmc_surface
      libXvMC: move drm_intel_bo to common intel_xvmc_surface
      libXvMC: unify PutSurface
      libXvMC i915: rip out unneeded surface fields
      libXvMC i915: s/i915XvMCSurface/struct intel_xvmc_surface/
      libXvMC: drop get_surface_status driver callback
      libXvMC: unify DestroySurface
      libXvMC: derive driver context from struct intel_xvmc_context
      libXvMC i915: kill unused context private fields
      libXvMC: unify CreateSurface
      libXvMC: rip out debug stuff
      i965 Xv: fix chroma pitch
      Xv: fixup overlay stride confusion
      XvMC: fix memory overflow
      i830 render: use tiling bits where possible
      i915 render: use tiling bits where possible
      render: tell the kernel explicitly when fences are needed
      i830 render: check aperture space requirements
      i810: adjust the pitch for DRI rendering
      Fix "make dist"

Eric Anholt (8):
      Revert "uxa: Try using put_image when copying from a memory buffer."
      Remove remaining REG_DUMPER build stuff.
      Add a workaround for Ironlake errata regarding blits and other engines.
      Add a workaround for Ironlake errata relating to disabling the clipper.
      uxa: Fix prepare_solid being called without check_solid first.
      Use libc instead of deprecated libc wrappers for malloc/calloc/free.
      Allocate a correctly sized framebuffer when tiling by using libdrm's support.
      Allocate rotate shadow buffers using the usual framebuffer allocator.

Gaetan Nadon (1):
      COPYING: replace stub file with actual Copyright notices.

Jesse Barnes (1):
      DRI2: fix new buffer exchange check

Keith Packard (2):
      Must call ValidateGC in i830_uxa_put_image for scratch GC
      Adapt to DevPrivate API changes

Kristian Høgsberg (1):
      dri: Clean up DRI2 API #ifdefs a bit

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

author
#2 lore raw
Quoted 7 lines · Julien Cristau
Quoted 4 lines · Carl Worth

New requirements compared to 2.10
---------------------------------
* Libdrm >= 2.4.19

The requirement is actually 2.4.21, afaict.

Can you give more details of what's required in 2.4.21 so that I can make a commit message that explains it?

I made the bump to require 2.4.19 when I noticed a failure to compile against 2.4.16 with an unresolved drm_intel_bo_emit_reloc_fence function. So the commit message I wrote was:

Fix to require libdrm 2.4.19 or newer.

This is necessary for the drm_intel_bo_emit_reloc_fence function which did not exist in 2.4.16 which is all we were requiring earlier.

I'd like something similar to accompany a commit requiring 2.4.21.

-Carl

Signature · 1 line
[email protected]

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

#3 lore raw
Quoted 16 lines · Carl Worth
Quoted 7 lines · Julien Cristau
Quoted 4 lines · Carl Worth

New requirements compared to 2.10
---------------------------------
* Libdrm >= 2.4.19

The requirement is actually 2.4.21, afaict.

Can you give more details of what's required in 2.4.21 so that I can make a commit message that explains it?

I made the bump to require 2.4.19 when I noticed a failure to compile against 2.4.16 with an unresolved drm_intel_bo_emit_reloc_fence function. So the commit message I wrote was:

Quoted 7 lines

Fix to require libdrm 2.4.19 or newer.

This is necessary for the drm_intel_bo_emit_reloc_fence function which did not exist in 2.4.16 which is all we were requiring earlier.

I'd like something similar to accompany a commit requiring 2.4.21.

Chris fixed this already,
http://cgit.freedesktop.org/xorg/driver/xf86-video-intel/commit/?id=254bcc40fca6ed9aa0b71f2115c654667f16aec6

Cheers,
Julien

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

#5 lore raw
Quoted 11 lines · Sergio Monteiro Basto
Quoted 7 lines · Julien Cristau
Quoted 4 lines · Carl Worth

New requirements compared to 2.10
---------------------------------
* Libdrm >= 2.4.19

The requirement is actually 2.4.21, afaict.

and libdrm >= 2.4.21 requires what kernel ?

You need Linux with KMS for xf86-video-intel. I'm not sure what you're getting at though.

Cheers,
Julien

#6 lore raw
Quoted 9 lines · Julien Cristau
Quoted 7 lines · Sergio Monteiro Basto
Quoted 6 lines · Julien Cristau
Quoted 4 lines · Carl Worth

New requirements compared to 2.10
---------------------------------
* Libdrm >= 2.4.19

Quoted 6 lines
Quoted 4 lines
Quoted 1 line

The requirement is actually 2.4.21, afaict.

and libdrm >= 2.4.21 requires what kernel ?

You need Linux with KMS for xf86-video-intel. I'm not sure what you're getting at though.

I am asking for kernel version, if we need a libdrm >= 2.4.21, what kernel version works with libdrm 2.4.21 ?

regards ,

Signature · 1 line
Sérgio M. B.

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

#7 lore raw
In reply to Carl Worth
Quoted 3 lines · On Tuesday June 15 2010 02:39:54 Carl Worth wrote:

This is the first release candidate in preparation for the upcoming 2.12.0 release. We will appreciate any feedback we can get from testing of this snapshot to improve the 2.12.0 release.

It makes my system really unstable. Sometimes I can't even login :S

The system just freezes :(

I'm running archlinux, Mesa 7.8.1, libdrm 2.4.21, Xorg-server 1.8.1.901, kernel 2.6.34.

I must say that the system did run faster while it was usable the times I tried.

Damnshock

Signature · 4 lines
A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?
#8 lore raw
Quoted 5 lines · Marc Deop i Argemí
Quoted 3 lines · On Tuesday June 15 2010 02:39:54 Carl Worth wrote:

This is the first release candidate in preparation for the upcoming 2.12.0 release. We will appreciate any feedback we can get from testing of this snapshot to improve the 2.12.0 release.

Quoted 7 lines

It makes my system really unstable. Sometimes I can't even login :S

The system just freezes :(

I'm running archlinux, Mesa 7.8.1, libdrm 2.4.21, Xorg-server 1.8.1.901, kernel 2.6.34.

On what hw?

Cheers,
Julien

#9 lore raw
Quoted 1 line · On Tuesday June 15 2010 16:18:07 Julien Cristau wrote:

On what hw?

My apologies:

Intel 945GMA, Centrino Duo T2400 1,83Ghz, 2'5Gb RAM, external monitor Samsung SyncMaster 2233SN

By the way (and not related) from kernel 2.6.34 the default resolution is not correct on my CRT getting the internal LCD 1280x800

Regards

Signature · 4 lines
A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?
author
#10 lore raw
In reply to Marc Deop i Argemí
Quoted 4 lines · Marc Deop i Argemí

The system just freezes :(

I'm running archlinux, Mesa 7.8.1, libdrm 2.4.21, Xorg-server 1.8.1.901, kernel 2.6.34.

If the other components remain the same and you change only the driver to an older version does it then work reliably? If so could you please bisect the commits in the driver repository to find the offending one?

Thanks,

-Carl

Signature · 1 line
[email protected]

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

#11 lore raw
Quoted 4 lines · On Tuesday June 15 2010 18:47:49 Carl Worth wrote:

If the other components remain the same and you change only the driver to an older version does it then work reliably? If so could you please bisect the commits in the driver repository to find the offending one?

That looks like a lot of work, ain't it? :S

Anyway, I could give it a try but I must admit I don't know where to start to look for the commits... can somebody give a hand on that?

Quoted 1 line

Thanks,

Naaah, we are here to help ;)

Regards

Signature · 4 lines
A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?
author
#12 lore raw
Quoted 1 line · Marc Deop i Argemí

That looks like a lot of work, ain't it? :S

It shouldn't be, once you get the hang of compiling and running against a locally-compiled driver. There are about 176 commits between 2.11.0 and 2.11.901 so that should only require about 8 compile/test cycles to determine what the buggy commit is.

Quoted 2 lines

Anyway, I could give it a try but I must admit I don't know where to start to look for the commits... can somebody give a hand on that?

Sure. Here's the recipe. First, checkout the driver from git with:

git clone git://anongit.freedesktop.org/xorg/driver/xf86-video-intel

After that, compile it with something like:

	cd xf86-video-intel
	./autogen.sh --prefix=/some/directory
        make
        make install

Then, you'll want to edit your xorg.conf file (likely /etc/X11/xorg.conf) to let it know about the new directory to which you installed this driver. So either add to the beginning of the Files section or make a new section named "Files" as follows:

Section "Files"
ModulePath "/some/directory/lib/xorg/modules"
ModulePath "/usr/lib/xorg/modules"
EndSection

[Obviously you might use something other than "/some/directory" in both of the above commands. Use whatever directory you'd like there.]

At that point, you should be able to restart X and run with your custom-compiled driver. And when you're done with all this, you'll be able to undo the changes to return to using your system's driver.

Next, to do the bisect, the first thing to do is to ensure that things work with the 2.11.0 release. That's as follows:

	git checkout 2.11.0
	./autogen.sh --prefix=/some/directory
        make
        make install

And then run the server. If it all works, then you're ready to bisect. Get back to master first with:

git checkout master

And then actually start bisecting with:

git bisect start
git bisect bad master
git bisect good 2.11.0

At that point git will checkout an intermediate commit for you. So then it's time to compile and test, (same commands as before: autogen.sh, make, make install). If it works, run "git bisect good". If it doesn't, run "git bisect bad". Then git will checkout another intermediate commit and you continue the process.

In the end, if all goes well, git should report "first bad commit is..." and that's the information we'll want to know.

Quoted 1 line

Naaah, we are here to help ;)

Fantastic. Please let me know if any of the above is not clear.

-Carl

Signature · 1 line
[email protected]

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

#13 lore raw
Quoted 1 line · On Tuesday June 15 2010 23:23:42 Carl Worth wrote:

On Tue, 15 Jun 2010 19:12:28 +0200, Marc Deop i Argemí <[email protected]>

Quoted 63 lines · wrote:
Quoted 1 line

That looks like a lot of work, ain't it? :S

It shouldn't be, once you get the hang of compiling and running against a locally-compiled driver. There are about 176 commits between 2.11.0 and 2.11.901 so that should only require about 8 compile/test cycles to determine what the buggy commit is.

Quoted 2 lines

Anyway, I could give it a try but I must admit I don't know where to start to look for the commits... can somebody give a hand on that?

Sure. Here's the recipe. First, checkout the driver from git with:

git clone git://anongit.freedesktop.org/xorg/driver/xf86-video-intel

After that, compile it with something like:

	cd xf86-video-intel
	./autogen.sh --prefix=/some/directory
        make
        make install

Then, you'll want to edit your xorg.conf file (likely /etc/X11/xorg.conf) to let it know about the new directory to which you installed this driver. So either add to the beginning of the Files section or make a new section named "Files" as follows:

Section "Files"
ModulePath "/some/directory/lib/xorg/modules"
ModulePath "/usr/lib/xorg/modules"
EndSection

[Obviously you might use something other than "/some/directory" in both of the above commands. Use whatever directory you'd like there.]

At that point, you should be able to restart X and run with your custom-compiled driver. And when you're done with all this, you'll be able to undo the changes to return to using your system's driver.

Next, to do the bisect, the first thing to do is to ensure that things work with the 2.11.0 release. That's as follows:

	git checkout 2.11.0
	./autogen.sh --prefix=/some/directory
        make
        make install

And then run the server. If it all works, then you're ready to bisect. Get back to master first with:

git checkout master

And then actually start bisecting with:

git bisect start
git bisect bad master
git bisect good 2.11.0

At that point git will checkout an intermediate commit for you. So then it's time to compile and test, (same commands as before: autogen.sh, make, make install). If it works, run "git bisect good". If it doesn't, run "git bisect bad". Then git will checkout another intermediate commit and you continue the process.

Quoted 8 lines

In the end, if all goes well, git should report "first bad commit is..." and that's the information we'll want to know.

Quoted 1 line

Naaah, we are here to help ;)

Fantastic. Please let me know if any of the above is not clear.

-Carl

Thanks a lot for the explanation Carl. I also would like to give thank to give thanks to Brain0 of the irc on Freenode for his help.

About the time taking... well, the freeze doesn't happen instantly so I have to play around for a while :S

Anyway, I will try to bisect today and report back :)

Thanks again for your explanation,

Regards

Signature · 4 lines
A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?
#14 lore raw
Quoted 1 line · On Wednesday June 16 2010 10:45:34 Marc Deop i Argemí wrote:

Anyway, I will try to bisect today and report back :)

Well, I tried. Does this message:

Bisecting: 0 revisions left to test after this (roughly 0 steps)

[29ba8a84f7cf5c29a5f38688a1ac0ccf41d8e4ec] XvMC: everyone's using execbuffer!

mean I finished the process?

Did that help?

Regards

Signature · 4 lines
A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?
#15 lore raw
Quoted 10 lines · Marc Deop i Argemí
Quoted 1 line · On Wednesday June 16 2010 10:45:34 Marc Deop i Argemí wrote:

Anyway, I will try to bisect today and report back :)

Well, I tried. Does this message:

Bisecting: 0 revisions left to test after this (roughly 0 steps)

[29ba8a84f7cf5c29a5f38688a1ac0ccf41d8e4ec] XvMC: everyone's using execbuffer!

mean I finished the process?

One more step. Test that commit and tell git bisect (good|bad) and it will print a slightly more verbose statement of which commit is triggering the freeze.

Thanks,
-ickle

Signature · 1 line
Chris Wilson, Intel Open Source Technology Centre

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

#16 lore raw
Quoted 3 lines · On Wednesday June 16 2010 15:57:41 Chris Wilson wrote:

One more step. Test that commit and tell git bisect (good|bad) and it will print a slightly more verbose statement of which commit is triggering the freeze.

Sometimes I amaze myself :S It clearly says *after* this.

I'll do that tonight

However, I have to comment that the system has been stable with the 2.12RC for two hours after applying this patch from my distribution (Archlinux):

diff -up xf86-video-intel-2.11.0/src/drmmode_display.c.no-flip xf86-video- intel-2.11.0/src/drmmode_display.c
--- xf86-video-intel-2.11.0/src/drmmode_display.c.no-flip 2010-05-03
15:30:19.000000000 -0400
+++ xf86-video-intel-2.11.0/src/drmmode_display.c 2010-05-03
15:30:59.000000000 -0400
@@ -1504,10 +1504,15 @@ Bool drmmode_pre_init(ScrnInfoPtr scrn,
gp.value = &has_flipping; (void)drmCommandWriteRead(intel->drmSubFD, DRM_I915_GETPARAM, &gp,
sizeof(gp));
+
+ xf86DrvMsg(scrn->scrnIndex, X_INFO,

+                   "Pageflipping %s in kernel, %s disabled in X\n",
+                   has_flipping ? "enabled" : "disabled",
+                   has_flipping ? "but" : "and");

if (has_flipping) {
xf86DrvMsg(scrn->scrnIndex, X_INFO,
- "Kernel page flipping support detected, enabling\n");
- intel->use_pageflipping = TRUE;
+ "Don't panic: https://bugzilla.redhat.com/588421\n");
+ intel->use_pageflipping = FALSE;
drmmode->flip_count = 0; drmmode->event_context.version = DRM_EVENT_CONTEXT_VERSION; drmmode->event_context.vblank_handler = drmmode_vblank_handler;
diff -up xf86-video-intel-2.11.0/src/i830_dri.c.no-flip xf86-video- intel-2.11.0/src/i830_dri.c
--- xf86-video-intel-2.11.0/src/i830_dri.c.no-flip 2010-03-29
14:23:02.000000000 -0400
+++ xf86-video-intel-2.11.0/src/i830_dri.c 2010-05-03 15:30:19.000000000
-0400
@@ -1013,7 +1013,7 @@ Bool I830DRI2ScreenInit(ScreenPtr screen

info.CopyRegion = I830DRI2CopyRegion;
#if DRI2INFOREC_VERSION >= 4
- if (intel->use_pageflipping) {
+ if (intel->use_pageflipping || 1) {
info.version = 4; info.ScheduleSwap = I830DRI2ScheduleSwap; info.GetMSC = I830DRI2GetMSC;

Signature · 4 lines
A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?
#17 lore raw
Quoted 4 lines · On Wednesday June 16 2010 18:12:38 Marc Deop i Argemí wrote:

Sometimes I amaze myself :S It clearly says after this.

I'll do that tonight

Finished the bisect although I don't think it was helpfull because all the steps were negative: all made my computer freeze.

However, the output was:

29ba8a84f7cf5c29a5f38688a1ac0ccf41d8e4ec is the first bad commit

Quoted 2 lines

However, I have to comment that the system has been stable with the 2.12RC for two hours after applying this patch from my distribution (Archlinux):

After being with this config for almost a day, I have to say it was rock
solid: played youtube videos (fullscreen and window), played Opengl games (nexuiz) and I could actually play on my crt without problem (before it would turn off after a few seconds). The system was very responsive specially without composing enabled.

There were some problems though: corruption on some menus (KDE4) and I couldn't not rotate the screen 270º as everything was corrupted on the screen (it was solved going back to normal, it was difficult though: I could not see what I was writing...).

Regards

Signature · 4 lines
A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?
#18 lore raw
In reply to Carl Worth

2010/6/15 Carl Worth <[email protected]>:

Quoted 9 lines · Marc Deop i Argemí
Quoted 4 lines

The system just freezes :(

I'm running archlinux, Mesa 7.8.1, libdrm 2.4.21, Xorg-server 1.8.1.901, kernel 2.6.34.

If the other components remain the same and you change only the driver to an older version does it then work reliably? If so could you please bisect the commits in the driver repository to find the offending one?

Hello! I'm, maybe, getting the same thing here. I haven't tested the release candidate, but with Xorg 1.8 and driver 2.11, and a kms-enabled 2.6.34 kernel, using gentoo, X just freezes. If I downgrade it to 2.9 driver and 1.7 Xorg, it runs (but if I do USE=hal at Xorg, it freezes, too).

I have an Acer 532h (Intel GMA 3150, afaik). I've read your explanation on how to do git bisecting and it sounds easy. Compile times here are somewhat slow, and I don't have time to do it right now (I don't have other computer..), but I will inform you about the results soon.

I should test with a fixed Xorg version, right? Which one? Portage currently has this:

1.6.5-r1 1.7.6 ~1.7.7 ~1.8.0 ~1.8.1-r1

(Versions with an ~ are filtered because they are at 'testing')

I haven't tried driver 2.9 with xorg 1.8, because I supposed newer xorg versions were incompatible with old drivers..

Should I install 1.8.0?

1.7.6 is what I've right now.

Signature · 1 line
Elias Gabriel Amaral da Silva <[email protected]>
author
#19 lore raw
Quoted 6 lines · Elias Gabriel Amaral da Silva

2010/6/15 Carl Worth <[email protected]>:
Hello! I'm, maybe, getting the same thing here. I haven't tested the release candidate, but with Xorg 1.8 and driver 2.11, and a kms-enabled 2.6.34 kernel, using gentoo, X just freezes. If I downgrade it to 2.9 driver and 1.7 Xorg, it runs (but if I do USE=hal at Xorg, it freezes, too).

Hi there, Elias!

Quoted 2 lines

I have an Acer 532h (Intel GMA 3150, afaik). I've read your explanation on how to do git bisecting and it sounds easy.

Good. It really should be quite simple.

Quoted 4 lines

I should test with a fixed Xorg version, right? Which one? Portage currently has this:

1.6.5-r1 1.7.6 ~1.7.7 ~1.8.0 ~1.8.1-r1

Before you can bisect, you'll want to identify a single component that exposes the bug. Above you described a change of two components, (both X server and driver), to switch from a broken to a working configuration.

So the first thing to do is to try keeping the X server version fixed and see if changing the driver alone changes the behavior.

Quoted 2 lines

I haven't tried driver 2.9 with xorg 1.8, because I supposed newer xorg versions were incompatible with old drivers..

I'd actually recommend using the newer X in both cases. I would hope that new X would work fine with an older driver, (but there might be some point at which the driver really is *too old* for a particular X server).

In that case, you might need to bump the driver version up and hope that it still works.

Good luck! And thanks in advance for any future information you might be able to provide.

-Carl

Signature · 1 line
[email protected]

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

#20 lore raw

2010/6/18 Carl Worth <[email protected]>:

Quoted 9 lines · Elias Gabriel Amaral da Silva
Quoted 6 lines

2010/6/15 Carl Worth <[email protected]>:
Hello! I'm, maybe, getting the same thing here. I haven't tested the release candidate, but with Xorg 1.8 and driver 2.11, and a kms-enabled 2.6.34 kernel, using gentoo, X just freezes. If I downgrade it to 2.9 driver and 1.7 Xorg, it runs (but if I do USE=hal at Xorg, it freezes, too).

Hi there, Elias!

I did

$ git checkout 2.9.0
$ ./autogen.sh --prefix=/opt/local/intel-gfx-2.9.0

(The version that actually worked here)

Then

$ make
Making all in uxa
 CC    uxa-render.o
uxa-render.c: In function ‘uxa_acquire_pattern’:
uxa-render.c:455: error: too few arguments to function ‘image_from_pict’
make[2]: *** [uxa-render.lo] Error 1
make[1]: *** [all-recursive] Error 1
make: *** [all] Error 2

Maybe I need to follow the trick at
http://old.nabble.com/Building-from-git-td27489808.html ? [ didn't worked; below my attemps ]

The latest intel driver (at master) just compiled without issues. I have Xorg 1.8.1.901 installed right now.

I installed git://anongit.freedesktop.org/git/xorg/util/macros and did

export ACLOCAL="aclocal -I /opt/local/xorg-macros/share/aclocal"
export PKG_CONFIG_PATH="/opt/local/xorg-macros/lib/pkgconfig:/opt/local/xorg-macros/share/pkgconfig"

And tried again. It seems to be using the new vars during ./autogen.sh , because of this line:

$ ./autogen.sh --prefix=/opt/local/intel-gfx-2.9.0
(..)
autoreconf-2.65: running: aclocal -I /opt/local/xorg-macros/share/aclocal -I m4
(..)

but still, same error

$ make
Making all in uxa
 CC    uxa.o
 CC    uxa-accel.o
 CC    uxa-glyphs.o
 CC    uxa-render.o
uxa-render.c: In function 'uxa_acquire_pattern':
uxa-render.c:455: error: too few arguments to function 'image_from_pict'
make[2]: *** [uxa-render.lo] Error 1
make[1]: *** [all-recursive] Error 1
make: *** [all] Error 2

Maybe I need an older version of this xorg-macros? I tried to copy /usr/share/aclocal/xorg-*.m4 to the m4 directory, too, that is the other workaround Chris Bagwell on that link says, but it didn't worked too; it seems to contain files from the to currently installed Xorg, and it looks like the "right" would be older files, relative to Xorg 1.7

I'm lost

Signature · 1 line
Elias Gabriel Amaral da Silva <[email protected]>
#21 lore raw

2010/6/21 Elias Gabriel Amaral da Silva <[email protected]>:

Quoted 9 lines

2010/6/18 Carl Worth <[email protected]>:

Quoted 8 lines · Elias Gabriel Amaral da Silva
Quoted 6 lines

2010/6/15 Carl Worth <[email protected]>:
Hello! I'm, maybe, getting the same thing here. I haven't tested the release candidate, but with Xorg 1.8 and driver 2.11, and a kms-enabled 2.6.34 kernel, using gentoo, X just freezes. If I downgrade it to 2.9 driver and 1.7 Xorg, it runs (but if I do USE=hal at Xorg, it freezes, too).

Quoted 6 lines
Quoted 1 line

Hi there, Elias!

I did

$ git checkout 2.9.0
$ ./autogen.sh --prefix=/opt/local/intel-gfx-2.9.0

I also compiled the latest code at /opt/local/intel-gfx-master

Running with it and X 1.8, X opens my display manager fine, but freezes

alt+sysrq+e can kill things fine (the kernel itself is responsive); after that, I've saved Xorg.0.log and dmesg. Not sure if this is useful at all, but I'm sending as attachment

my xorg.conf is just

Section "Files"
ModulePath "/opt/local/intel-gfx-master"
ModulePath "/usr/lib/xorg/modules"
EndSection

Signature · 1 line
Elias Gabriel Amaral da Silva <[email protected]>
[   159.258] 

This is a pre-release version of the X server from The X.Org Foundation. It is not supported in any way.
Bugs may be filed in the bugzilla at http://bugs.freedesktop.org/.
Select the "xorg" product for bugs you find in this release. Before reporting bugs in pre-release versions please check the latest version in the X.Org Foundation git repository.
See http://wiki.x.org/wiki/GitPage for git access instructions.

[   159.258] 
X.Org X Server 1.8.1.901 (1.8.2 RC 1)
Release Date: 2010-05-11
[   159.258] X Protocol Version 11, Revision 0
[   159.258] Build Operating System: Linux 2.6.32-3-686 i686 Gentoo
[   159.258] Current Operating System: Linux localhost 2.6.34-ckrisos #8 SMP Thu Jun 10 18:28:35 BRT 2010 i686
[   159.258] Kernel command line: BOOT_IMAGE=/boot/gentoo-2.6.34-ckrisos quiet ro video=i915:modeset=1
[   159.258] Build Date: 20 June 2010  11:29:47AM
[   159.258]  
[   159.259] Current version of pixman: 0.17.2
[   159.259] 	Before reporting problems, check http://wiki.x.org
	to make sure that you have the latest version.
[   159.259] Markers: (--) probed, (**) from config file, (==) default setting,
	(++) from command line, (!!) notice, (II) informational,
	(WW) warning, (EE) error, (NI) not implemented, (??) unknown.
[   159.259] (==) Log file: "/var/log/Xorg.0.log", Time: Mon Jun 21 02:45:41 2010
[   159.260] (==) Using config file: "/etc/X11/xorg.conf"
[   159.260] (==) Using system config directory "/usr/share/X11/xorg.conf.d"
[   159.260] (==) No Layout section.  Using the first Screen section.
[   159.260] (==) No screen section available. Using defaults.
[   159.260] (**) |-->Screen "Default Screen Section" (0)
[   159.260] (**) |   |-->Monitor "<default monitor>"
[   159.261] (==) No monitor specified for screen "Default Screen Section".
	Using a default monitor configuration.
[   159.261] (==) Automatically adding devices
[   159.261] (==) Automatically enabling devices
[   159.262] (WW) The directory "/usr/share/fonts/misc/" does not exist.
[   159.262] 	Entry deleted from font path.
[   159.262] (WW) The directory "/usr/share/fonts/TTF/" does not exist.
[   159.262] 	Entry deleted from font path.
[   159.262] (WW) The directory "/usr/share/fonts/OTF/" does not exist.
[   159.262] 	Entry deleted from font path.
[   159.262] (WW) The directory "/usr/share/fonts/Type1/" does not exist.
[   159.262] 	Entry deleted from font path.
[   159.262] (WW) The directory "/usr/share/fonts/100dpi/" does not exist.
[   159.262] 	Entry deleted from font path.
[   159.262] (WW) The directory "/usr/share/fonts/75dpi/" does not exist.
[   159.262] 	Entry deleted from font path.
[   159.262] (==) FontPath set to:
[   159.262] (**) ModulePath set to "/opt/local/intel-gfx-master,/usr/lib/xorg/modules"
[   159.262] (II) The server relies on udev to provide the list of input devices.
	If no devices become available, reconfigure udev or disable AutoAddDevices.
[   159.262] (II) Loader magic: 0x81ecb20
[   159.262] (II) Module ABI versions:
[   159.262] 	X.Org ANSI C Emulation: 0.4
[   159.263] 	X.Org Video Driver: 7.0
[   159.263] 	X.Org XInput driver : 9.0
[   159.263] 	X.Org Server Extension : 3.0
[   159.682] (--) PCI:*(0:0:2:0) 8086:a011:1025:0349 Intel Corporation Pineview Integrated Graphics Controller rev 0, Mem @ 0x98180000/524288, 0x80000000/268435456, 0x98000000/1048576, I/O @ 0x000060c0/8
[   159.682] (--) PCI: (0:0:2:1) 8086:a012:1025:0349 Intel Corporation Pineview Integrated Graphics Controller rev 0, Mem @ 0x98100000/524288
[   159.682] (II) LoadModule: "extmod"
[   159.726] (II) Loading /usr/lib/xorg/modules/extensions/libextmod.so
[   159.743] (II) Module extmod: vendor="X.Org Foundation"
[   159.743] 	compiled for 1.8.1.901, module version = 1.0.0
[   159.743] 	Module class: X.Org Server Extension
[   159.743] 	ABI class: X.Org Server Extension, version 3.0
[   159.743] (II) Loading extension MIT-SCREEN-SAVER
[   159.743] (II) Loading extension XFree86-VidModeExtension
[   159.743] (II) Loading extension XFree86-DGA
[   159.744] (II) Loading extension DPMS
[   159.744] (II) Loading extension XVideo
[   159.744] (II) Loading extension XVideo-MotionCompensation
[   159.744] (II) Loading extension X-Resource
[   159.744] (II) LoadModule: "dbe"
[   159.769] (II) Loading /usr/lib/xorg/modules/extensions/libdbe.so
[   159.789] (II) Module dbe: vendor="X.Org Foundation"
[   159.789] 	compiled for 1.8.1.901, module version = 1.0.0
[   159.789] 	Module class: X.Org Server Extension
[   159.789] 	ABI class: X.Org Server Extension, version 3.0
[   159.789] (II) Loading extension DOUBLE-BUFFER
[   159.789] (II) LoadModule: "glx"
[   159.791] (II) Loading /usr/lib/xorg/modules/extensions/libglx.so
[   159.819] (II) Module glx: vendor="X.Org Foundation"
[   159.819] 	compiled for 1.8.1.901, module version = 1.0.0
[   159.819] 	ABI class: X.Org Server Extension, version 3.0
[   159.824] (==) AIGLX enabled
[   159.824] (II) Loading extension GLX
[   159.824] (II) LoadModule: "record"
[   159.825] (II) Loading /usr/lib/xorg/modules/extensions/librecord.so
[   159.830] (II) Module record: vendor="X.Org Foundation"
[   159.830] 	compiled for 1.8.1.901, module version = 1.13.0
[   159.830] 	Module class: X.Org Server Extension
[   159.830] 	ABI class: X.Org Server Extension, version 3.0
[   159.830] (II) Loading extension RECORD
[   159.830] (II) LoadModule: "dri"
[   159.831] (II) Loading /usr/lib/xorg/modules/extensions/libdri.so
[   159.834] (II) Module dri: vendor="X.Org Foundation"
[   159.834] 	compiled for 1.8.1.901, module version = 1.0.0
[   159.834] 	ABI class: X.Org Server Extension, version 3.0
[   159.834] (II) Loading extension XFree86-DRI
[   159.835] (II) LoadModule: "dri2"
[   159.836] (II) Loading /usr/lib/xorg/modules/extensions/libdri2.so
[   159.840] (II) Module dri2: vendor="X.Org Foundation"
[   159.840] 	compiled for 1.8.1.901, module version = 1.2.0
[   159.840] 	ABI class: X.Org Server Extension, version 3.0
[   159.840] (II) Loading extension DRI2
[   159.840] (==) Matched intel as autoconfigured driver 0
[   159.840] (==) Matched vesa as autoconfigured driver 1
[   159.840] (==) Matched fbdev as autoconfigured driver 2
[   159.841] (==) Assigned the driver to the xf86ConfigLayout
[   159.841] (II) LoadModule: "intel"
[   159.841] (II) Loading /opt/local/intel-gfx-master/lib/xorg/modules/drivers/intel_drv.so
[   159.842] (II) Module intel: vendor="X.Org Foundation"
[   159.842] 	compiled for 1.8.1.901, module version = 2.11.901
[   159.842] 	Module class: X.Org Video Driver
[   159.842] 	ABI class: X.Org Video Driver, version 7.0
[   159.842] (II) LoadModule: "vesa"
[   159.858] (WW) Warning, couldn't open module vesa
[   159.858] (II) UnloadModule: "vesa"
[   159.858] (EE) Failed to load module "vesa" (module does not exist, 0)
[   159.858] (II) LoadModule: "fbdev"
[   159.860] (WW) Warning, couldn't open module fbdev
[   159.860] (II) UnloadModule: "fbdev"
[   159.860] (EE) Failed to load module "fbdev" (module does not exist, 0)
[   159.860] (II) intel: Driver for Intel Integrated Graphics Chipsets: i810,
	i810-dc100, i810e, i815, i830M, 845G, 852GM/855GM, 865G, 915G,
	E7221 (i915), 915GM, 945G, 945GM, 945GME, Pineview GM, Pineview G,
	965G, G35, 965Q, 946GZ, 965GM, 965GME/GLE, G33, Q35, Q33, GM45,
	4 Series, G45/G43, Q45/Q43, G41, B43, Clarkdale, Arrandale
[   159.861] (++) using VT number 7
[   159.871] (II) Primary Device is: PCI 00@00:02:0
[   159.875] drmOpenDevice: node name is /dev/dri/card0
[   159.876] drmOpenDevice: open result is 8, (OK)
[   159.876] drmOpenByBusid: Searching for BusID pci:0000:00:02.0
[   159.876] drmOpenDevice: node name is /dev/dri/card0
[   159.876] drmOpenDevice: open result is 8, (OK)
[   159.876] drmOpenByBusid: drmOpenMinor returns 8
[   159.876] drmOpenByBusid: drmGetBusid reports pci:0000:00:02.0
[   159.876] (II) intel(0): Creating default Display subsection in Screen section
	"Default Screen Section" for depth/fbbpp 24/32
[   159.876] (==) intel(0): Depth 24, (--) framebuffer bpp 32
[   159.876] (==) intel(0): RGB weight 888
[   159.876] (==) intel(0): Default visual is TrueColor
[   159.876] (II) intel(0): Integrated Graphics Chipset: Intel(R) Pineview GM
[   159.876] (--) intel(0): Chipset: "Pineview GM"
[   159.876] (==) intel(0): video overlay key set to 0x101fe
[   159.992] (II) intel(0): Output VGA1 has no monitor section
[   160.095] (II) intel(0): Output LVDS1 has no monitor section
[   160.096] (II) intel(0): found backlight control interface /sys/class/backlight/acpi_video0
[   160.312] (II) intel(0): EDID for output VGA1
[   160.312] (II) intel(0): Manufacturer: PHL  Model: 1851  Serial#: 52908
[   160.312] (II) intel(0): Year: 2009  Week: 46
[   160.312] (II) intel(0): EDID Version: 1.3
[   160.312] (II) intel(0): Analog Display Input,  Input Voltage Level: 0.700/0.700 V
[   160.312] (II) intel(0): Sync:  Separate
[   160.312] (II) intel(0): Max Image Size [cm]: horiz.: 41  vert.: 23
[   160.312] (II) intel(0): Gamma: 2.20
[   160.312] (II) intel(0): DPMS capabilities: Off; RGB/Color Display
[   160.313] (II) intel(0): First detailed timing is preferred mode
[   160.313] (II) intel(0): redX: 0.647 redY: 0.334   greenX: 0.284 greenY: 0.607
[   160.313] (II) intel(0): blueX: 0.151 blueY: 0.071   whiteX: 0.312 whiteY: 0.329
[   160.313] (II) intel(0): Supported established timings:
[   160.313] (II) intel(0): 720x400@70Hz
[   160.313] (II) intel(0): 640x480@60Hz
[   160.313] (II) intel(0): 640x480@67Hz
[   160.313] (II) intel(0): 640x480@72Hz
[   160.313] (II) intel(0): 640x480@75Hz
[   160.313] (II) intel(0): 800x600@56Hz
[   160.313] (II) intel(0): 800x600@60Hz
[   160.313] (II) intel(0): 800x600@72Hz
[   160.313] (II) intel(0): 800x600@75Hz
[   160.313] (II) intel(0): 832x624@75Hz
[   160.313] (II) intel(0): 1024x768@60Hz
[   160.313] (II) intel(0): 1024x768@70Hz
[   160.313] (II) intel(0): 1024x768@75Hz
[   160.314] (II) intel(0): Manufacturer's mask: 0
[   160.314] (II) intel(0): Supported standard timings:
[   160.314] (II) intel(0): #0: hsize: 640  vsize 400  refresh: 70  vid: 2609
[   160.314] (II) intel(0): #1: hsize: 1024  vsize 768  refresh: 72  vid: 19553
[   160.314] (II) intel(0): Supported detailed timing:
[   160.314] (II) intel(0): clock: 85.5 MHz   Image Size:  413 x 234 mm
[   160.314] (II) intel(0): h_active: 1366  h_sync: 1436  h_sync_end 1579 h_blank_end 1792 h_border: 0
[   160.314] (II) intel(0): v_active: 768  v_sync: 771  v_sync_end 774 v_blanking: 798 v_border: 0
[   160.314] (II) intel(0): Supported detailed timing:
[   160.314] (II) intel(0): clock: 85.5 MHz   Image Size:  34 x 19 mm
[   160.314] (II) intel(0): h_active: 1360  h_sync: 1424  h_sync_end 1536 h_blank_end 1792 h_border: 0
[   160.314] (II) intel(0): v_active: 768  v_sync: 771  v_sync_end 777 v_blanking: 795 v_border: 0
[   160.314] (II) intel(0): Ranges: V min: 55 V max: 75 Hz, H min: 30 H max: 81 kHz, PixClock max 90 MHz
[   160.314] (II) intel(0): Monitor name: Philips 185Vw
[   160.314] (II) intel(0): EDID (in hex):
[   160.315] (II) intel(0): 	00ffffffffffff00410c5118acce0000
[   160.315] (II) intel(0): 	2e130103682917782aeed1a555489b26
[   160.315] (II) intel(0): 	125054bfee00310a614c010101010101
[   160.315] (II) intel(0): 	010101010101662156aa51001e30468f
[   160.315] (II) intel(0): 	33009dea1000001e662150b051001b30
[   160.315] (II) intel(0): 	4070360022130000001e000000fd0037
[   160.315] (II) intel(0): 	4b1e5109000a202020202020000000fc
[   160.315] (II) intel(0): 	005068696c69707320313835567700ce
[   160.315] (II) intel(0): EDID vendor "PHL", prod id 6225
[   160.315] (II) intel(0): Using EDID range info for horizontal sync
[   160.315] (II) intel(0): Using EDID range info for vertical refresh
[   160.315] (II) intel(0): Printing DDC gathered Modelines:
[   160.316] (II) intel(0): Modeline "1366x768"x0.0   85.50  1366 1436 1579 1792  768 771 774 798 +hsync +vsync (47.7 kHz)
[   160.316] (II) intel(0): Modeline "1360x768"x0.0   85.50  1360 1424 1536 1792  768 771 777 795 +hsync +vsync (47.7 kHz)
[   160.316] (II) intel(0): Modeline "800x600"x0.0   40.00  800 840 968 1056  600 601 605 628 +hsync +vsync (37.9 kHz)
[   160.316] (II) intel(0): Modeline "800x600"x0.0   36.00  800 824 896 1024  600 601 603 625 +hsync +vsync (35.2 kHz)
[   160.316] (II) intel(0): Modeline "640x480"x0.0   31.50  640 656 720 840  480 481 484 500 -hsync -vsync (37.5 kHz)
[   160.316] (II) intel(0): Modeline "640x480"x0.0   31.50  640 664 704 832  480 489 492 520 -hsync -vsync (37.9 kHz)
[   160.316] (II) intel(0): Modeline "640x480"x0.0   30.24  640 704 768 864  480 483 486 525 -hsync -vsync (35.0 kHz)
[   160.316] (II) intel(0): Modeline "640x480"x0.0   25.18  640 656 752 800  480 490 492 525 -hsync -vsync (31.5 kHz)
[   160.316] (II) intel(0): Modeline "720x400"x0.0   28.32  720 738 846 900  400 412 414 449 -hsync +vsync (31.5 kHz)
[   160.316] (II) intel(0): Modeline "1024x768"x0.0   78.75  1024 1040 1136 1312  768 769 772 800 +hsync +vsync (60.0 kHz)
[   160.316] (II) intel(0): Modeline "1024x768"x0.0   75.00  1024 1048 1184 1328  768 771 777 806 -hsync -vsync (56.5 kHz)
[   160.316] (II) intel(0): Modeline "1024x768"x0.0   65.00  1024 1048 1184 1344  768 771 777 806 -hsync -vsync (48.4 kHz)
[   160.317] (II) intel(0): Modeline "832x624"x0.0   57.28  832 864 928 1152  624 625 628 667 -hsync -vsync (49.7 kHz)
[   160.317] (II) intel(0): Modeline "800x600"x0.0   49.50  800 816 896 1056  600 601 604 625 +hsync +vsync (46.9 kHz)
[   160.317] (II) intel(0): Modeline "800x600"x0.0   50.00  800 856 976 1040  600 637 643 666 +hsync +vsync (48.1 kHz)
[   160.317] (II) intel(0): Modeline "640x400"x70.0   23.35  640 656 720 800  400 401 404 417 -hsync +vsync (29.2 kHz)
[   160.317] (II) intel(0): Modeline "1024x768"x72.0   78.43  1024 1080 1192 1360  768 769 772 801 -hsync +vsync (57.7 kHz)
[   160.317] (II) intel(0): Printing probed modes for output VGA1
[   160.317] (II) intel(0): Modeline "1366x768"x59.8   85.50  1366 1436 1579 1792  768 771 774 798 +hsync +vsync (47.7 kHz)
[   160.317] (II) intel(0): Modeline "1360x768"x60.0   85.50  1360 1424 1536 1792  768 771 777 795 +hsync +vsync (47.7 kHz)
[   160.317] (II) intel(0): Modeline "1024x768"x75.1   78.80  1024 1040 1136 1312  768 769 772 800 +hsync +vsync (60.1 kHz)
[   160.317] (II) intel(0): Modeline "1024x768"x72.0   78.44  1024 1080 1192 1360  768 769 772 801 -hsync +vsync (57.7 kHz)
[   160.317] (II) intel(0): Modeline "1024x768"x70.1   75.00  1024 1048 1184 1328  768 771 777 806 -hsync -vsync (56.5 kHz)
[   160.318] (II) intel(0): Modeline "1024x768"x60.0   65.00  1024 1048 1184 1344  768 771 777 806 -hsync -vsync (48.4 kHz)
[   160.318] (II) intel(0): Modeline "832x624"x74.6   57.28  832 864 928 1152  624 625 628 667 -hsync -vsync (49.7 kHz)
[   160.318] (II) intel(0): Modeline "800x600"x72.2   50.00  800 856 976 1040  600 637 643 666 +hsync +vsync (48.1 kHz)
[   160.318] (II) intel(0): Modeline "800x600"x75.0   49.50  800 816 896 1056  600 601 604 625 +hsync +vsync (46.9 kHz)
[   160.318] (II) intel(0): Modeline "800x600"x60.3   40.00  800 840 968 1056  600 601 605 628 +hsync +vsync (37.9 kHz)
[   160.318] (II) intel(0): Modeline "800x600"x56.2   36.00  800 824 896 1024  600 601 603 625 +hsync +vsync (35.2 kHz)
[   160.318] (II) intel(0): Modeline "640x480"x72.8   31.50  640 664 704 832  480 489 491 520 -hsync -vsync (37.9 kHz)
[   160.318] (II) intel(0): Modeline "640x480"x75.0   31.50  640 656 720 840  480 481 484 500 -hsync -vsync (37.5 kHz)
[   160.318] (II) intel(0): Modeline "640x480"x66.7   30.24  640 704 768 864  480 483 486 525 -hsync -vsync (35.0 kHz)
[   160.318] (II) intel(0): Modeline "640x480"x60.0   25.20  640 656 752 800  480 490 492 525 -hsync -vsync (31.5 kHz)
[   160.318] (II) intel(0): Modeline "720x400"x70.1   28.32  720 738 846 900  400 412 414 449 -hsync +vsync (31.5 kHz)
[   160.423] (II) intel(0): EDID for output LVDS1
[   160.423] (II) intel(0): Manufacturer: AUO  Model: 30d2  Serial#: 0
[   160.423] (II) intel(0): Year: 2008  Week: 1
[   160.423] (II) intel(0): EDID Version: 1.3
[   160.423] (II) intel(0): Digital Display Input
[   160.423] (II) intel(0): Max Image Size [cm]: horiz.: 22  vert.: 13
[   160.423] (II) intel(0): Gamma: 2.20
[   160.423] (II) intel(0): No DPMS capabilities specified
[   160.423] (II) intel(0): Supported color encodings: RGB 4:4:4 YCrCb 4:4:4 
[   160.423] (II) intel(0): First detailed timing is preferred mode
[   160.423] (II) intel(0): redX: 0.588 redY: 0.351   greenX: 0.342 greenY: 0.567
[   160.423] (II) intel(0): blueX: 0.158 blueY: 0.123   whiteX: 0.313 whiteY: 0.329
[   160.424] (II) intel(0): Manufacturer's mask: 0
[   160.424] (II) intel(0): Supported detailed timing:
[   160.424] (II) intel(0): clock: 52.0 MHz   Image Size:  223 x 125 mm
[   160.424] (II) intel(0): h_active: 1024  h_sync: 1048  h_sync_end 1184 h_blank_end 1344 h_border: 0
[   160.424] (II) intel(0): v_active: 600  v_sync: 603  v_sync_end 604 v_blanking: 644 v_border: 0
[   160.424] (II) intel(0): Unknown vendor-specific block f
[   160.424] (II) intel(0):  AUO
[   160.424] (II) intel(0):  B101AW03 V0
[   160.424] (II) intel(0): EDID (in hex):
[   160.424] (II) intel(0): 	00ffffffffffff0006afd23000000000
[   160.424] (II) intel(0): 	0112010380160d780ab9a59659579128
[   160.424] (II) intel(0): 	1f505400000001010101010101010101
[   160.424] (II) intel(0): 	0101010101015014004041582c201888
[   160.424] (II) intel(0): 	3100df7d000000180000000f00000000
[   160.425] (II) intel(0): 	00000000000000000020000000fe0041
[   160.425] (II) intel(0): 	554f0a202020202020202020000000fe
[   160.425] (II) intel(0): 	004231303141573033205630200a003c
[   160.425] (II) intel(0): Not using default mode "320x240" (doublescan mode not supported)
[   160.425] (II) intel(0): Not using default mode "400x300" (doublescan mode not supported)
[   160.425] (II) intel(0): Not using default mode "400x300" (doublescan mode not supported)
[   160.425] (II) intel(0): Not using default mode "512x384" (doublescan mode not supported)
[   160.426] (II) intel(0): Not using default mode "640x480" (doublescan mode not supported)
[   160.426] (II) intel(0): Not using default mode "640x512" (doublescan mode not supported)
[   160.426] (II) intel(0): Not using default mode "800x600" (doublescan mode not supported)
[   160.426] (II) intel(0): Not using default mode "700x525" (doublescan mode not supported)
[   160.426] (II) intel(0): Printing probed modes for output LVDS1
[   160.426] (II) intel(0): Modeline "1024x600"x60.1   52.00  1024 1048 1184 1344  600 603 604 644 -hsync -vsync (38.7 kHz)
[   160.426] (II) intel(0): Modeline "800x600"x60.3   40.00  800 840 968 1056  600 601 605 628 +hsync +vsync (37.9 kHz)
[   160.426] (II) intel(0): Modeline "800x600"x56.2   36.00  800 824 896 1024  600 601 603 625 +hsync +vsync (35.2 kHz)
[   160.426] (II) intel(0): Modeline "640x480"x59.9   25.18  640 656 752 800  480 490 492 525 -hsync -vsync (31.5 kHz)
[   160.426] (II) intel(0): Output VGA1 connected
[   160.426] (II) intel(0): Output LVDS1 connected
[   160.426] (II) intel(0): Using fuzzy aspect match for initial modes
[   160.426] (II) intel(0): Output VGA1 using initial mode 800x600
[   160.426] (II) intel(0): Output LVDS1 using initial mode 800x600
[   160.427] (II) intel(0): Using default gamma of (1.0, 1.0, 1.0) unless otherwise stated.
[   160.427] (II) intel(0): Kernel page flipping support detected, enabling
[   160.427] (**) intel(0): Display dimensions: (410, 230) mm
[   160.427] (**) intel(0): DPI set to (49, 66)
[   160.427] (II) Loading sub module "fb"
[   160.427] (II) LoadModule: "fb"
[   160.428] (II) Loading /usr/lib/xorg/modules/libfb.so
[   160.459] (II) Module fb: vendor="X.Org Foundation"
[   160.459] 	compiled for 1.8.1.901, module version = 1.0.0
[   160.459] 	ABI class: X.Org ANSI C Emulation, version 0.4
[   160.459] (==) Depth 24 pixmap format is 32 bpp
[   160.459] (II) intel(0): [DRI2] Setup complete
[   160.459] (II) intel(0): [DRI2]   DRI driver: i915
[   160.459] (**) intel(0): Tiling enabled
[   160.459] (**) intel(0): SwapBuffers wait enabled
[   160.460] (==) intel(0): VideoRam: 262144 KB
[   160.460] (II) intel(0): Allocated new frame buffer 832x600 stride 4096, tiled
[   160.486] (II) UXA(0): Driver registered support for the following operations:
[   160.486] (II)         solid
[   160.486] (II)         copy
[   160.486] (II)         composite (RENDER acceleration)
[   160.486] (II)         put_image
[   160.486] (II)         get_image
[   160.486] (==) intel(0): Backing store disabled
[   160.486] (==) intel(0): Silken mouse enabled
[   160.486] (II) intel(0): Initializing HW Cursor
[   161.288] (II) intel(0): RandR 1.2 enabled, ignore the following RandR disabled message.
[   161.291] (==) intel(0): DPMS enabled
[   161.291] (==) intel(0): Intel XvMC decoder disabled
[   161.291] (II) intel(0): Set up textured video
[   161.291] (II) intel(0): Set up overlay video
[   161.291] (II) intel(0): direct rendering: DRI2 Enabled
[   161.292] (--) RandR disabled
[   161.292] (II) Initializing built-in extension Generic Event Extension
[   161.292] (II) Initializing built-in extension SHAPE
[   161.292] (II) Initializing built-in extension MIT-SHM
[   161.292] (II) Initializing built-in extension XInputExtension
[   161.292] (II) Initializing built-in extension XTEST
[   161.292] (II) Initializing built-in extension BIG-REQUESTS
[   161.292] (II) Initializing built-in extension SYNC
[   161.292] (II) Initializing built-in extension XKEYBOARD
[   161.292] (II) Initializing built-in extension XC-MISC
[   161.292] (II) Initializing built-in extension XINERAMA
[   161.292] (II) Initializing built-in extension XFIXES
[   161.293] (II) Initializing built-in extension RENDER
[   161.293] (II) Initializing built-in extension RANDR
[   161.293] (II) Initializing built-in extension COMPOSITE
[   161.293] (II) Initializing built-in extension DAMAGE
[   161.423] (II) AIGLX: enabled GLX_MESA_copy_sub_buffer
[   161.423] (II) AIGLX: enabled GLX_INTEL_swap_event
[   161.423] (II) AIGLX: enabled GLX_SGI_swap_control and GLX_MESA_swap_control
[   161.423] (II) AIGLX: enabled GLX_SGI_make_current_read
[   161.423] (II) AIGLX: GLX_EXT_texture_from_pixmap backed by buffer objects
[   161.423] (II) AIGLX: Loaded and initialized /usr/lib/dri/i915_dri.so
[   161.424] (II) GLX: Initialized DRI2 GL provider for screen 0
[   161.425] (II) intel(0): Setting screen physical size to 211 x 158
[   162.055] (II) config/udev: Adding input device Power Button (/dev/input/event0)
[   162.055] (**) Power Button: Applying InputClass "evdev keyboard catchall"
[   162.055] (II) LoadModule: "evdev"
[   162.056] (II) Loading /usr/lib/xorg/modules/input/evdev_drv.so
[   162.069] (II) Module evdev: vendor="X.Org Foundation"
[   162.069] 	compiled for 1.7.6, module version = 2.3.2
[   162.069] 	Module class: X.Org XInput Driver
[   162.069] 	ABI class: X.Org XInput driver, version 7.0
[   162.069] (EE) module ABI major version (7) doesn't match the server's version (9)
[   162.069] (II) UnloadModule: "evdev"
[   162.070] (II) Unloading /usr/lib/xorg/modules/input/evdev_drv.so
[   162.070] (EE) Failed to load module "evdev" (module requirement mismatch, 0)
[   162.070] (EE) No input driver matching `evdev'
[   162.070] (II) config/udev: Adding input device Sleep Button (/dev/input/event1)
[   162.070] (**) Sleep Button: Applying InputClass "evdev keyboard catchall"
[   162.071] (II) LoadModule: "evdev"
[   162.072] (II) Loading /usr/lib/xorg/modules/input/evdev_drv.so
[   162.072] (II) Module evdev: vendor="X.Org Foundation"
[   162.072] 	compiled for 1.7.6, module version = 2.3.2
[   162.072] 	Module class: X.Org XInput Driver
[   162.072] 	ABI class: X.Org XInput driver, version 7.0
[   162.072] (EE) module ABI major version (7) doesn't match the server's version (9)
[   162.072] (II) UnloadModule: "evdev"
[   162.072] (II) Unloading /usr/lib/xorg/modules/input/evdev_drv.so
[   162.072] (EE) Failed to load module "evdev" (module requirement mismatch, 0)
[   162.072] (EE) No input driver matching `evdev'
[   162.073] (II) config/udev: Adding input device Lid Switch (/dev/input/event2)
[   162.073] (II) No input driver/identifier specified (ignoring)
[   162.074] (II) config/udev: Adding input device Power Button (/dev/input/event3)
[   162.074] (**) Power Button: Applying InputClass "evdev keyboard catchall"
[   162.074] (II) LoadModule: "evdev"
[   162.075] (II) Loading /usr/lib/xorg/modules/input/evdev_drv.so
[   162.075] (II) Module evdev: vendor="X.Org Foundation"
[   162.075] 	compiled for 1.7.6, module version = 2.3.2
[   162.075] 	Module class: X.Org XInput Driver
[   162.075] 	ABI class: X.Org XInput driver, version 7.0
[   162.075] (EE) module ABI major version (7) doesn't match the server's version (9)
[   162.076] (II) UnloadModule: "evdev"
[   162.076] (II) Unloading /usr/lib/xorg/modules/input/evdev_drv.so
[   162.076] (EE) Failed to load module "evdev" (module requirement mismatch, 0)
[   162.076] (EE) No input driver matching `evdev'
[   162.076] (II) config/udev: Adding input device AT Translated Set 2 keyboard (/dev/input/event4)
[   162.076] (**) AT Translated Set 2 keyboard: Applying InputClass "evdev keyboard catchall"
[   162.077] (II) LoadModule: "evdev"
[   162.078] (II) Loading /usr/lib/xorg/modules/input/evdev_drv.so
[   162.078] (II) Module evdev: vendor="X.Org Foundation"
[   162.078] 	compiled for 1.7.6, module version = 2.3.2
[   162.078] 	Module class: X.Org XInput Driver
[   162.078] 	ABI class: X.Org XInput driver, version 7.0
[   162.078] (EE) module ABI major version (7) doesn't match the server's version (9)
[   162.078] (II) UnloadModule: "evdev"
[   162.078] (II) Unloading /usr/lib/xorg/modules/input/evdev_drv.so
[   162.078] (EE) Failed to load module "evdev" (module requirement mismatch, 0)
[   162.078] (EE) No input driver matching `evdev'
[   162.079] (II) config/udev: Adding input device USB USB Keykoard (/dev/input/event5)
[   162.079] (**) USB USB Keykoard: Applying InputClass "evdev keyboard catchall"
[   162.079] (II) LoadModule: "evdev"
[   162.080] (II) Loading /usr/lib/xorg/modules/input/evdev_drv.so
[   162.080] (II) Module evdev: vendor="X.Org Foundation"
[   162.081] 	compiled for 1.7.6, module version = 2.3.2
[   162.081] 	Module class: X.Org XInput Driver
[   162.081] 	ABI class: X.Org XInput driver, version 7.0
[   162.081] (EE) module ABI major version (7) doesn't match the server's version (9)
[   162.081] (II) UnloadModule: "evdev"
[   162.081] (II) Unloading /usr/lib/xorg/modules/input/evdev_drv.so
[   162.081] (EE) Failed to load module "evdev" (module requirement mismatch, 0)
[   162.081] (EE) No input driver matching `evdev'
[   162.082] (II) config/udev: Adding input device ImPS/2 Generic Wheel Mouse (/dev/input/event6)
[   162.082] (**) ImPS/2 Generic Wheel Mouse: Applying InputClass "evdev pointer catchall"
[   162.082] (II) LoadModule: "evdev"
[   162.083] (II) Loading /usr/lib/xorg/modules/input/evdev_drv.so
[   162.083] (II) Module evdev: vendor="X.Org Foundation"
[   162.083] 	compiled for 1.7.6, module version = 2.3.2
[   162.083] 	Module class: X.Org XInput Driver
[   162.083] 	ABI class: X.Org XInput driver, version 7.0
[   162.083] (EE) module ABI major version (7) doesn't match the server's version (9)
[   162.083] (II) UnloadModule: "evdev"
[   162.083] (II) Unloading /usr/lib/xorg/modules/input/evdev_drv.so
[   162.084] (EE) Failed to load module "evdev" (module requirement mismatch, 0)
[   162.084] (EE) No input driver matching `evdev'
[   162.084] (II) config/udev: Adding input device ImPS/2 Generic Wheel Mouse (/dev/input/mouse0)
[   162.084] (II) No input driver/identifier specified (ignoring)
[   162.085] (II) config/udev: Adding input device USB USB Keykoard (/dev/input/event7)
[   162.085] (**) USB USB Keykoard: Applying InputClass "evdev keyboard catchall"
[   162.085] (II) LoadModule: "evdev"
[   162.086] (II) Loading /usr/lib/xorg/modules/input/evdev_drv.so
[   162.086] (II) Module evdev: vendor="X.Org Foundation"
[   162.086] 	compiled for 1.7.6, module version = 2.3.2
[   162.087] 	Module class: X.Org XInput Driver
[   162.087] 	ABI class: X.Org XInput driver, version 7.0
[   162.087] (EE) module ABI major version (7) doesn't match the server's version (9)
[   162.087] (II) UnloadModule: "evdev"
[   162.087] (II) Unloading /usr/lib/xorg/modules/input/evdev_drv.so
[   162.087] (EE) Failed to load module "evdev" (module requirement mismatch, 0)
[   162.087] (EE) No input driver matching `evdev'
[   162.088] (II) config/udev: Adding input device USB Mouse (/dev/input/event8)
[   162.088] (**) USB Mouse: Applying InputClass "evdev pointer catchall"
[   162.088] (II) LoadModule: "evdev"
[   162.089] (II) Loading /usr/lib/xorg/modules/input/evdev_drv.so
[   162.089] (II) Module evdev: vendor="X.Org Foundation"
[   162.089] 	compiled for 1.7.6, module version = 2.3.2
[   162.089] 	Module class: X.Org XInput Driver
[   162.089] 	ABI class: X.Org XInput driver, version 7.0
[   162.089] (EE) module ABI major version (7) doesn't match the server's version (9)
[   162.089] (II) UnloadModule: "evdev"
[   162.089] (II) Unloading /usr/lib/xorg/modules/input/evdev_drv.so
[   162.089] (EE) Failed to load module "evdev" (module requirement mismatch, 0)
[   162.090] (EE) No input driver matching `evdev'
[   162.090] (II) config/udev: Adding input device USB Mouse (/dev/input/mouse1)
[   162.090] (II) No input driver/identifier specified (ignoring)
[   162.091] (II) config/udev: Adding input device Video Bus (/dev/input/event9)
[   162.091] (**) Video Bus: Applying InputClass "evdev keyboard catchall"
[   162.091] (II) LoadModule: "evdev"
[   162.092] (II) Loading /usr/lib/xorg/modules/input/evdev_drv.so
[   162.092] (II) Module evdev: vendor="X.Org Foundation"
[   162.092] 	compiled for 1.7.6, module version = 2.3.2
[   162.092] 	Module class: X.Org XInput Driver
[   162.092] 	ABI class: X.Org XInput driver, version 7.0
[   162.092] (EE) module ABI major version (7) doesn't match the server's version (9)
[   162.093] (II) UnloadModule: "evdev"
[   162.093] (II) Unloading /usr/lib/xorg/modules/input/evdev_drv.so
[   162.093] (EE) Failed to load module "evdev" (module requirement mismatch, 0)
[   162.093] (EE) No input driver matching `evdev'

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

#22 lore raw
Quoted 4 lines · Elias Gabriel Amaral da Silva

...Snip bits of Xorg.0.log...

[   162.075] (EE) module ABI major version (7) doesn't match the
server's version (9)
[   162.076] (II) UnloadModule: "evdev" 

You haven't rebuilt your input drivers against the new Xserver. The intel driver is working, X just doesn't have any input devices so it looks like it's frozen.

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

#23 lore raw

2010/6/21 Christopher James Halse Rogers
<[email protected]>:

Quoted 9 lines · Elias Gabriel Amaral da Silva
Quoted 4 lines

...Snip bits of Xorg.0.log...
[   162.075] (EE) module ABI major version (7) doesn't match the server's version (9)
[   162.076] (II) UnloadModule: "evdev"

You haven't rebuilt your input drivers against the new Xserver.  The intel driver is working, X just doesn't have any input devices so it looks like it's frozen.

Recompiling fixed it, and 2.11.901 is working great here.

Thank you =)

Signature · 1 line
Elias Gabriel Amaral da Silva <[email protected]>
#24 lore raw
In reply to Carl Worth

2010/6/18 Carl Worth <[email protected]>:

Quoted 10 lines
Quoted 2 lines

I haven't tried driver 2.9 with xorg 1.8, because I supposed newer xorg versions were incompatible with old drivers..

I'd actually recommend using the newer X in both cases. I would hope that new X would work fine with an older driver, (but there might be some point at which the driver really is *too old* for a particular X server).

In that case, you might need to bump the driver version up and hope that it still works.

When using an 2.9 driver compiled for 1.7 xorg, and 1.8 xorg, X complains about ABI mismatch. Recompiling the driver is supposed to fix this, but, I get the same error I got while building from git sources. In fact I see it was already reported at gentoo bugzilla,
http://bugs.gentoo.org/317585

Well, I suppose I could just downgrade X..

Is 2.11 and beyond expected to work with Xorg 1.7? [ hmm, every time I do this, I have to wait a while.. I'm using debian testing right now (Xorg 1.7.5, intel driver 2.9.1) and it's working fine. ]

Signature · 1 line
Elias Gabriel Amaral da Silva <[email protected]>
author
#25 lore raw
Quoted 11 lines · Elias Gabriel Amaral da Silva

When using an 2.9 driver compiled for 1.7 xorg, and 1.8 xorg, X complains about ABI mismatch. Recompiling the driver is supposed to fix this, but, I get the same error I got while building from git sources. In fact I see it was already reported at gentoo bugzilla,
http://bugs.gentoo.org/317585

Well, I suppose I could just downgrade X..

Is 2.11 and beyond expected to work with Xorg 1.7? [ hmm, every time I do this, I have to wait a while.. I'm using debian testing right now (Xorg 1.7.5, intel driver 2.9.1) and it's working fine. ]

Sorry about all this.

There's a tight ABI binding between the X server and the xf86-video-intel driver. That means you can't compile the 2.9 driver with the latest X server, nor can you compile the latest driver with the X server that works with the 2.9 driver.

That means it's impossible to hold one of these projects constant while bisecting the other, (or even just trying a couple of versions of the other to pinpoint it as the problematic component).

That's unfortunate since bisecting would be extremely useful here.

For this specific instance, one might be able to construct a patch to the 2.9 driver to make it compile against the latest X server. And then one could carry that patch along while bisecting. That will be a bit of work though.

In the future, I think the right thing to do is to put the driver back into the X server repository. That would eliminate this problem entirely, (for future versions), and bisecting either X server or driver problems would work just fine.

-Carl

Signature · 1 line
[email protected]

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

#26 lore raw

Hi Carl,

Quoted 4 lines

In the future, I think the right thing to do is to put the driver back into the X server repository. That would eliminate this problem entirely, (for future versions), and bisecting either X server or driver problems would work just fine.

I don't see what that would change - in both ways the whole X stack has to be recompiled.

On the other hand, at least I am not willed to recompile the whole X userland just to give driver prereleases a try - as I did the past few releases.
Wasn't that one of the main reasons why drivers were moved out of the xorg distribution?

- Clemens

#27 lore raw
In reply to Marc Deop i Argemí
Quoted 8 lines · Marc Deop i Argemí
Quoted 3 lines · On Tuesday June 15 2010 02:39:54 Carl Worth wrote:

This is the first release candidate in preparation for the upcoming 2.12.0 release. We will appreciate any feedback we can get from testing of this snapshot to improve the 2.12.0 release.

It makes my system really unstable. Sometimes I can't even login :S

The system just freezes :(

After a freeze, can you grab a copy of dmesg, Xorg.log, intel_reg_dump, and /sys/kernel/debug/dri/0/i915_error_state. From my perspective, the last contains the most interesting information (the batchbuffer executing at the time of a gpu hang). If there is nothing in any of those, echo t > /proc/sysrq-trigger and look for the stacktraces in dmesg.

Also what type of freeze is it? A complete hard hang, the system no longer even pingable? The cursor continues to move, but no other screen updates? The screen entirely frozen (or maybe flashing in a hypnotic pattern) but the machine nevertheless is accessible over the network? Does VT switching still work? Etc. 'Freeze' covers a whole multitude of sins, the more information you can give us, the quicker we can pinpoint the cause.

Thanks,
-ickle

Signature · 1 line
Chris Wilson, Intel Open Source Technology Centre

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

#28 lore raw
Quoted 5 lines · On Wednesday June 16 2010 11:20:09 Chris Wilson wrote:

After a freeze, can you grab a copy of dmesg, Xorg.log, intel_reg_dump, and /sys/kernel/debug/dri/0/i915_error_state. From my perspective, the last contains the most interesting information (the batchbuffer executing at the time of a gpu hang). If there is nothing in any of those, echo t > /proc/sysrq-trigger and look for the stacktraces in dmesg.

I checked some of those files (dmesg, Xorg.log) and couldn't fine anything. I'll try again and will check the other files you mention

Quoted 7 lines

Also what type of freeze is it? A complete hard hang, the system no longer even pingable? The cursor continues to move, but no other screen updates? The screen entirely frozen (or maybe flashing in a hypnotic pattern) but the machine nevertheless is accessible over the network? Does VT switching still work? Etc. 'Freeze' covers a whole multitude of sins, the more information you can give us, the quicker we can pinpoint the cause.

Freeze is... freeze :P

Now, seriously, the computer stops responding. I'm not able to switch between VT nor to ping nor login to the machine via my cell phone. If I'm playing sound the machine gets in kind of a loop with the last sound it played repeating again and again.

I usually test the stability with the flash player on youtube because it's the *fastest* way to make the computer freeze. With the latest RC it freezes in about a minute of playing video as well as in the bisects I've been trying to do. That doesn't happen with the 2.11 release.

Could it be a problem of the compiler? I recently upgraded to GCC 4.5 and maybe the driver from the repos is compiled against the old 4.4? I'll give it a try...

I'll report back when I'm done

Regards

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

#29 lore raw
Quoted 18 lines · Marc Deop i Argemí
Quoted 5 lines · On Wednesday June 16 2010 11:20:09 Chris Wilson wrote:

After a freeze, can you grab a copy of dmesg, Xorg.log, intel_reg_dump, and /sys/kernel/debug/dri/0/i915_error_state. From my perspective, the last contains the most interesting information (the batchbuffer executing at the time of a gpu hang). If there is nothing in any of those, echo t > /proc/sysrq-trigger and look for the stacktraces in dmesg.

I checked some of those files (dmesg, Xorg.log) and couldn't fine anything. I'll try again and will check the other files you mention

Quoted 7 lines

Also what type of freeze is it? A complete hard hang, the system no longer even pingable? The cursor continues to move, but no other screen updates? The screen entirely frozen (or maybe flashing in a hypnotic pattern) but the machine nevertheless is accessible over the network? Does VT switching still work? Etc. 'Freeze' covers a whole multitude of sins, the more information you can give us, the quicker we can pinpoint the cause.

Quoted 6 lines

Freeze is... freeze :P

Now, seriously, the computer stops responding. I'm not able to switch between VT nor to ping nor login to the machine via my cell phone. If I'm playing sound the machine gets in kind of a loop with the last sound it played repeating again and again.

No ping? Sounds like a kernel bug to me. You could try netconsole to get a oops dump.

--Andy

#30 lore raw
Quoted 5 lines · Andrew Lutomirski
Quoted 4 lines · Marc Deop i Argemí

On Wednesday June 16 2010 11:20:09 Chris Wilson wrote:

Freeze is... freeze :P

Quoted 7 lines
Quoted 4 lines

Now, seriously, the computer stops responding. I'm not able to switch between VT nor to ping nor login to the machine via my cell phone. If I'm playing sound the machine gets in kind of a loop with the last sound it played repeating again and again.

No ping?  Sounds like a kernel bug to me.  You could try netconsole to get a oops dump.

Hah. I just got a freeze, too. As in, frozen so hard that caps lock and magic sysrq didn't work. This is git master from today (a25573d5c47ebea34c076075e1993233d7db2b4f).

--Andy

#31 lore raw
Quoted 11 lines · On Wed, 16 Jun 2010 11:07:21 -0400 Andrew Lutomirski <[email protected]> wrote:
Quoted 10 lines · Andrew Lutomirski
Quoted 8 lines · Marc Deop i Argemí

On Wednesday June 16 2010 11:20:09 Chris Wilson wrote:

Freeze is... freeze :P

Now, seriously, the computer stops responding. I'm not able to switch between VT nor to ping nor login to the machine via my cell phone. If I'm playing sound the machine gets in kind of a loop with the last sound it played repeating again and again.

Quoted 6 lines
Quoted 2 lines

No ping?  Sounds like a kernel bug to me.  You could try netconsole to get a oops dump.

Hah. I just got a freeze, too. As in, frozen so hard that caps lock and magic sysrq didn't work. This is git master from today (a25573d5c47ebea34c076075e1993233d7db2b4f).

Are these both on 945? If so, can you guys try two of the patches I posted earlier? In particular:
drm/i915: fix page flipping on gen3
drm/i915: don't queue flips during a flip pending event

(attached but may only apply with fuzz).

Signature · 1 line
Jesse Barnes, Intel Open Source Technology Center
Quoted 1 line

From 7153c638e0944af6fd81190a8f1ae3fc66054896 Mon Sep 17 00:00:00 2001

From: Jesse Barnes <[email protected]>
Date: Fri, 26 Mar 2010 10:35:20 -0700
Subject: [PATCH 6/7] drm/i915: fix page flipping on gen3

Gen3 chips have slightly different flip commands, and also contain a bit that indicates whether a "flip pending" interrupt means the flip has been queued or has been completed.

So implement support for the gen3 flip command, and make sure we use the flip pending interrupt correctly depending on the value of ECOSKPD bit 0.

---
 drivers/gpu/drm/i915/i915_dma.c      |    4 ++++
 drivers/gpu/drm/i915/i915_drv.h      |    1 +
 drivers/gpu/drm/i915/i915_irq.c      |   21 +++++++++++++++++----
 drivers/gpu/drm/i915/i915_reg.h      |    4 ++++
 drivers/gpu/drm/i915/intel_display.c |   29 ++++++++++++++++++++++++-----
 drivers/gpu/drm/i915/intel_drv.h     |    1 +
 6 files changed, 51 insertions(+), 9 deletions(-)
drivers/gpu/drm/i915/i915_dma.c +4 -0
index a9f8589..193bf16 100644
@@ -1475,6 +1475,10 @@ static int i915_load_modeset_init(struct drm_device *dev,
 	if (ret)
 		goto destroy_ringbuffer;
 
+	/* IIR "flip pending" bit means done if this bit is set */
+	if (IS_GEN3(dev) && (I915_READ(ECOSKPD) & ECO_FLIP_DONE))
+		dev_priv->flip_pending_is_done = true;
+
 	intel_modeset_init(dev);
 
 	ret = drm_irq_install(dev);
drivers/gpu/drm/i915/i915_drv.h +1 -0
index abf2713..4e41f0f 100644
@@ -608,6 +608,7 @@ typedef struct drm_i915_private {
 	struct drm_crtc *plane_to_crtc_mapping[2];
 	struct drm_crtc *pipe_to_crtc_mapping[2];
 	wait_queue_head_t pending_flip_queue;
+	bool flip_pending_is_done;
 
 	/* Reclocking support */
 	bool render_reclock_avail;
drivers/gpu/drm/i915/i915_irq.c +17 -4
index b16bb0d..d5eeca6 100644
@@ -924,22 +924,35 @@ irqreturn_t i915_driver_irq_handler(DRM_IRQ_ARGS)
 			mod_timer(&dev_priv->hangcheck_timer, jiffies + DRM_I915_HANGCHECK_PERIOD);
 		}
 
-		if (iir & I915_DISPLAY_PLANE_A_FLIP_PENDING_INTERRUPT)
+		if ((iir & I915_DISPLAY_PLANE_B_FLIP_PENDING_INTERRUPT) &&
+		    !(pipeb_stats & pipe_vblank_mask)) {
+			DRM_ERROR("flip complete w/o vblank irq?\n");
+		}
+
+		if (iir & I915_DISPLAY_PLANE_A_FLIP_PENDING_INTERRUPT) {
 			intel_prepare_page_flip(dev, 0);
+			if (dev_priv->flip_pending_is_done)
+				intel_finish_page_flip_plane(dev, 0);
+		}
 
-		if (iir & I915_DISPLAY_PLANE_B_FLIP_PENDING_INTERRUPT)
+		if (iir & I915_DISPLAY_PLANE_B_FLIP_PENDING_INTERRUPT) {
 			intel_prepare_page_flip(dev, 1);
+			if (dev_priv->flip_pending_is_done)
+				intel_finish_page_flip_plane(dev, 1);
+		}
 
 		if (pipea_stats & pipe_vblank_mask) {
 			vblank++;
 			drm_handle_vblank(dev, 0);
-			intel_finish_page_flip(dev, 0);
+			if (!dev_priv->flip_pending_is_done)
+				intel_finish_page_flip(dev, 0);
 		}
 
 		if (pipeb_stats & pipe_vblank_mask) {
 			vblank++;
 			drm_handle_vblank(dev, 1);
-			intel_finish_page_flip(dev, 1);
+			if (!dev_priv->flip_pending_is_done)
+				intel_finish_page_flip(dev, 1);
 		}
 
 		if ((pipeb_stats & PIPE_LEGACY_BLC_EVENT_STATUS) ||
drivers/gpu/drm/i915/i915_reg.h +4 -0
index df20187..9179b38 100644
@@ -178,6 +178,7 @@
 #define   MI_OVERLAY_OFF	(0x2<<21)
 #define MI_LOAD_SCAN_LINES_INCL MI_INSTR(0x12, 0)
 #define MI_DISPLAY_FLIP		MI_INSTR(0x14, 2)
+#define MI_DISPLAY_FLIP_I915	MI_INSTR(0x14, 1)
 #define   MI_DISPLAY_FLIP_PLANE(n) ((n) << 20)
 #define MI_STORE_DWORD_IMM	MI_INSTR(0x20, 1)
 #define   MI_MEM_VIRTUAL	(1 << 22) /* 965+ only */
@@ -357,6 +358,9 @@
 #define   CM0_RC_OP_FLUSH_DISABLE (1<<0)
 #define BB_ADDR		0x02140 /* 8 bytes */
 #define GFX_FLSH_CNTL	0x02170 /* 915+ only */
+#define ECOSKPD		0x021d0
+#define   ECO_GATING_CX_ONLY	(1<<3)
+#define   ECO_FLIP_DONE		(1<<0)
 
 
 /*
drivers/gpu/drm/i915/intel_display.c +24 -5
index 92ca56c..cd6f8b7 100644
@@ -4136,10 +4136,10 @@ static void intel_unpin_work_fn(struct work_struct *__work)
 	kfree(work);
 }
 
-void intel_finish_page_flip(struct drm_device *dev, int pipe)
+static void do_intel_finish_page_flip(struct drm_device *dev,
+				      struct drm_crtc *crtc)
 {
 	drm_i915_private_t *dev_priv = dev->dev_private;
-	struct drm_crtc *crtc = dev_priv->pipe_to_crtc_mapping[pipe];
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct intel_unpin_work *work;
 	struct drm_i915_gem_object *obj_priv;
@@ -4189,6 +4189,22 @@ void intel_finish_page_flip(struct drm_device *dev, int pipe)
 	schedule_work(&work->work);
 }
 
+void intel_finish_page_flip(struct drm_device *dev, int pipe)
+{
+	drm_i915_private_t *dev_priv = dev->dev_private;
+	struct drm_crtc *crtc = dev_priv->pipe_to_crtc_mapping[pipe];
+
+	do_intel_finish_page_flip(dev, crtc);
+}
+
+void intel_finish_page_flip_plane(struct drm_device *dev, int plane)
+{
+	drm_i915_private_t *dev_priv = dev->dev_private;
+	struct drm_crtc *crtc = dev_priv->plane_to_crtc_mapping[plane];
+
+	do_intel_finish_page_flip(dev, crtc);
+}
+
 void intel_prepare_page_flip(struct drm_device *dev, int plane)
 {
 	drm_i915_private_t *dev_priv = dev->dev_private;
@@ -4270,14 +4286,17 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	work->pending_flip_obj = obj;
 
 	BEGIN_LP_RING(4);
-	OUT_RING(MI_DISPLAY_FLIP |
-		 MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
-	OUT_RING(fb->pitch);
 	if (IS_I965G(dev)) {
+		OUT_RING(MI_DISPLAY_FLIP |
+			 MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
+		OUT_RING(fb->pitch);
 		OUT_RING(obj_priv->gtt_offset | obj_priv->tiling_mode);
 		pipesrc = I915_READ(pipesrc_reg); 
 		OUT_RING(pipesrc & 0x0fff0fff);
 	} else {
+		OUT_RING(MI_DISPLAY_FLIP_I915 |
+			 MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
+		OUT_RING(fb->pitch);
 		OUT_RING(obj_priv->gtt_offset);
 		OUT_RING(MI_NOOP);
 	}
drivers/gpu/drm/i915/intel_drv.h +1 -0
index 3a467ca..8323225 100644
@@ -219,6 +219,7 @@ extern int intel_framebuffer_create(struct drm_device *dev,
 
 extern void intel_prepare_page_flip(struct drm_device *dev, int plane);
 extern void intel_finish_page_flip(struct drm_device *dev, int pipe);
+extern void intel_finish_page_flip_plane(struct drm_device *dev, int plane);
 
 extern void intel_setup_overlay(struct drm_device *dev);
 extern void intel_cleanup_overlay(struct drm_device *dev);
Signature · 1 line
1.6.1.3
Quoted 1 line

From a8ddc8b3743abb72b6c7e26ce75ae98d93d92133 Mon Sep 17 00:00:00 2001

From: Jesse Barnes <[email protected]>
Date: Mon, 5 Apr 2010 14:03:51 -0700
Subject: [PATCH 3/3] drm/i915: don't queue flips during a flip pending event

Hardware will set the flip pending ISR bit as soon as it receives the flip instruction, and (supposedly) clear it once the flip completes (e.g. at the next vblank). If we try to send down a flip instruction while the ISR bit is set, the hardware can become very confused, and we may never receive the corresponding flip pending interrupt, effectively hanging the chip.

Fixes fdo bug #27040.

---
 drivers/gpu/drm/i915/intel_display.c |   10 ++++++++++
 1 files changed, 10 insertions(+), 0 deletions(-)
drivers/gpu/drm/i915/intel_display.c +10 -0
index 6f5e7f5..24853e1 100644
@@ -4235,6 +4235,7 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	unsigned long flags;
 	int pipesrc_reg = (intel_crtc->pipe == 0) ? PIPEASRC : PIPEBSRC;
 	int ret, pipesrc;
+	u32 flip_mask;
 	RING_LOCALS;
 
 	work = kzalloc(sizeof *work, GFP_KERNEL);
@@ -4285,6 +4286,15 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	atomic_inc(&obj_priv->pending_flip);
 	work->pending_flip_obj = obj;
 
+	if (intel_crtc->plane)
+		flip_mask = I915_DISPLAY_PLANE_B_FLIP_PENDING_INTERRUPT;
+	else
+		flip_mask = I915_DISPLAY_PLANE_A_FLIP_PENDING_INTERRUPT;
+
+	/* Wait for any previous flip to finish */
+	while (I915_READ(ISR) & flip_mask)
+		;
+
 	BEGIN_LP_RING(4);
 	if (IS_I965G(dev)) {
 		OUT_RING(MI_DISPLAY_FLIP |
Signature · 1 line
1.6.6.1

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

#32 lore raw
Quoted 1 line · On Thursday June 17 2010 22:32:17 Jesse Barnes wrote:

Are these both on 945?

Mine is on a 945GM

Quoted 5 lines

If so, can you guys try two of the patches I posted earlier? In particular:
drm/i915: fix page flipping on gen3
drm/i915: don't queue flips during a flip pending event

Sure I can try. I'm gonna do it right now :)

Regards

Signature · 4 lines
A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?
#33 lore raw
In reply to Jesse Barnes
Quoted 17 lines · Jesse Barnes
Quoted 14 lines · On Wed, 16 Jun 2010 11:07:21 -0400 Andrew Lutomirski <[email protected]> wrote:
Quoted 12 lines · Andrew Lutomirski
Quoted 8 lines · Marc Deop i Argemí

On Wednesday June 16 2010 11:20:09 Chris Wilson wrote:

Freeze is... freeze :P

Now, seriously, the computer stops responding. I'm not able to switch between VT nor to ping nor login to the machine via my cell phone. If I'm playing sound the machine gets in kind of a loop with the last sound it played repeating again and again.

No ping?  Sounds like a kernel bug to me.  You could try netconsole to get a oops dump.

Quoted 6 lines
Quoted 3 lines

Hah.  I just got a freeze, too.  As in, frozen so hard that caps lock and magic sysrq didn't work.  This is git master from today (a25573d5c47ebea34c076075e1993233d7db2b4f).

Are these both on 945?  If so, can you guys try two of the patches I posted earlier?  In particular:

This is GM45.

Quoted 2 lines

 drm/i915: fix page flipping on gen3
 drm/i915: don't queue flips during a flip pending event

Neither patch applies for me.

$ patch -p1 <0003-drm-i915-don-t-queue-flips-during-a-flip-pending-eve.patch
patching file drivers/gpu/drm/i915/intel_display.c
Hunk #1 FAILED at 4235.
[more output about fuzz]
$ git reset --hard origin/master
HEAD is now at 7e27d6e Linux 2.6.35-rc3
$ patch -p1 <0006-drm-i915-fix-page-flipping-on-gen3.patch
[stuff that worked]
patching file drivers/gpu/drm/i915/i915_irq.c
Hunk #1 FAILED at 924.
1 out of 1 hunk FAILED -- saving rejects to file
drivers/gpu/drm/i915/i915_irq.c.rej
[more stuff that worked]

--Andy

#34 lore raw
Quoted 1 line · On Friday June 18 2010 02:17:53 Andrew Lutomirski wrote:

Neither patch applies for me.

One of them do apply for me, the other one doesn't.

Testing done on latest 2.6.35-rc3, the building fails.

Regards

Signature · 4 lines
A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?
#35 lore raw
Quoted 6 lines · On Fri, 18 Jun 2010 02:20:23 +0200 Marc Deop i Argemí <[email protected]>…
Quoted 1 line · On Friday June 18 2010 02:17:53 Andrew Lutomirski wrote:

Neither patch applies for me.

One of them do apply for me, the other one doesn't.

Testing done on latest 2.6.35-rc3, the building fails.

Arg, ok, I'll refresh them and post new ones tomorrow.

Thanks,

Signature · 5 lines
Jesse Barnes, Intel Open Source Technology Center
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
#36 lore raw
Quoted 11 lines · On Thu, 17 Jun 2010 19:44:10 -0700 Jesse Barnes <[email protected]>…
Quoted 6 lines · On Fri, 18 Jun 2010 02:20:23 +0200 Marc Deop i Argemí <[email protected]>…
Quoted 1 line · On Friday June 18 2010 02:17:53 Andrew Lutomirski wrote:

Neither patch applies for me.

One of them do apply for me, the other one doesn't.

Testing done on latest 2.6.35-rc3, the building fails.

Arg, ok, I'll refresh them and post new ones tomorrow.

Ok here are some updated ones.

Signature · 1 line
Jesse Barnes, Intel Open Source Technology Center
Quoted 1 line

From a53925d44f1e8b66cd0aed163af656991c414492 Mon Sep 17 00:00:00 2001

From: Jesse Barnes <[email protected]>
Date: Mon, 5 Apr 2010 14:03:51 -0700
Subject: [PATCH 1/2] drm/i915: don't queue flips during a flip pending event

Hardware will set the flip pending ISR bit as soon as it receives the flip instruction, and (supposedly) clear it once the flip completes (e.g. at the next vblank). If we try to send down a flip instruction while the ISR bit is set, the hardware can become very confused, and we may never receive the corresponding flip pending interrupt, effectively hanging the chip.

---
 drivers/gpu/drm/i915/intel_display.c |   11 +++++++++++
 1 files changed, 11 insertions(+), 0 deletions(-)
drivers/gpu/drm/i915/intel_display.c +11 -0
index 88a1ab7..f460761 100644
@@ -4664,6 +4664,7 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	unsigned long flags;
 	int pipesrc_reg = (intel_crtc->pipe == 0) ? PIPEASRC : PIPEBSRC;
 	int ret, pipesrc;
+	u32 flip_mask;
 
 	work = kzalloc(sizeof *work, GFP_KERNEL);
 	if (work == NULL)
@@ -4717,6 +4718,16 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	atomic_inc(&obj_priv->pending_flip);
 	work->pending_flip_obj = obj;
 
+	if (intel_crtc->plane)
+		flip_mask = I915_DISPLAY_PLANE_B_FLIP_PENDING_INTERRUPT;
+	else
+		flip_mask = I915_DISPLAY_PLANE_A_FLIP_PENDING_INTERRUPT;
+
+	/* Wait for any previous flip to finish */
+	if (IS_GEN3(dev))
+		while (I915_READ(ISR) & flip_mask)
+			;
+
 	BEGIN_LP_RING(4);
 	OUT_RING(MI_DISPLAY_FLIP |
 		 MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
Signature · 1 line
1.6.6.1
Quoted 1 line

From 5e1e40dda5b675e1432a0f9d2d8943196f7a1130 Mon Sep 17 00:00:00 2001

From: Jesse Barnes <[email protected]>
Date: Fri, 26 Mar 2010 10:35:20 -0700
Subject: [PATCH 2/2] drm/i915: gen3 page flipping fixes

Gen3 chips have slightly different flip commands, and also contain a bit that indicates whether a "flip pending" interrupt means the flip has been queued or has been completed.

So implement support for the gen3 flip command, and make sure we use the flip pending interrupt correctly depending on the value of ECOSKPD bit 0.

---
 drivers/gpu/drm/i915/i915_dma.c      |    4 ++++
 drivers/gpu/drm/i915/i915_drv.h      |    1 +
 drivers/gpu/drm/i915/i915_irq.c      |   16 ++++++++++++----
 drivers/gpu/drm/i915/i915_reg.h      |    4 ++++
 drivers/gpu/drm/i915/intel_display.c |   29 ++++++++++++++++++++++++-----
 drivers/gpu/drm/i915/intel_drv.h     |    1 +
 6 files changed, 46 insertions(+), 9 deletions(-)
drivers/gpu/drm/i915/i915_dma.c +4 -0
index 84ce956..4d59710 100644
@@ -1408,6 +1408,10 @@ static int i915_load_modeset_init(struct drm_device *dev,
 	if (ret)
 		goto destroy_ringbuffer;
 
+	/* IIR "flip pending" bit means done if this bit is set */
+	if (IS_GEN3(dev) && (I915_READ(ECOSKPD) & ECO_FLIP_DONE))
+		dev_priv->flip_pending_is_done = true;
+
 	intel_modeset_init(dev);
 
 	ret = drm_irq_install(dev);
drivers/gpu/drm/i915/i915_drv.h +1 -0
index 9ed8ecd..fd62d1e 100644
@@ -595,6 +595,7 @@ typedef struct drm_i915_private {
 	struct drm_crtc *plane_to_crtc_mapping[2];
 	struct drm_crtc *pipe_to_crtc_mapping[2];
 	wait_queue_head_t pending_flip_queue;
+	bool flip_pending_is_done;
 
 	/* Reclocking support */
 	bool render_reclock_avail;
drivers/gpu/drm/i915/i915_irq.c +12 -4
index 2479be0..a846cd8 100644
@@ -940,22 +940,30 @@ irqreturn_t i915_driver_irq_handler(DRM_IRQ_ARGS)
 		if (HAS_BSD(dev) && (iir & I915_BSD_USER_INTERRUPT))
 			DRM_WAKEUP(&dev_priv->bsd_ring.irq_queue);
 
-		if (iir & I915_DISPLAY_PLANE_A_FLIP_PENDING_INTERRUPT)
+		if (iir & I915_DISPLAY_PLANE_A_FLIP_PENDING_INTERRUPT) {
 			intel_prepare_page_flip(dev, 0);
+			if (dev_priv->flip_pending_is_done)
+				intel_finish_page_flip_plane(dev, 0);
+		}
 
-		if (iir & I915_DISPLAY_PLANE_B_FLIP_PENDING_INTERRUPT)
+		if (iir & I915_DISPLAY_PLANE_B_FLIP_PENDING_INTERRUPT) {
+			if (dev_priv->flip_pending_is_done)
+				intel_finish_page_flip_plane(dev, 1);
 			intel_prepare_page_flip(dev, 1);
+		}
 
 		if (pipea_stats & vblank_status) {
 			vblank++;
 			drm_handle_vblank(dev, 0);
-			intel_finish_page_flip(dev, 0);
+			if (!dev_priv->flip_pending_is_done)
+				intel_finish_page_flip(dev, 0);
 		}
 
 		if (pipeb_stats & vblank_status) {
 			vblank++;
 			drm_handle_vblank(dev, 1);
-			intel_finish_page_flip(dev, 1);
+			if (!dev_priv->flip_pending_is_done)
+				intel_finish_page_flip(dev, 1);
 		}
 
 		if ((pipea_stats & I915_LEGACY_BLC_EVENT_STATUS) ||
drivers/gpu/drm/i915/i915_reg.h +4 -0
index 64b0a3a..2cae38a 100644
@@ -178,6 +178,7 @@
 #define   MI_OVERLAY_OFF	(0x2<<21)
 #define MI_LOAD_SCAN_LINES_INCL MI_INSTR(0x12, 0)
 #define MI_DISPLAY_FLIP		MI_INSTR(0x14, 2)
+#define MI_DISPLAY_FLIP_I915	MI_INSTR(0x14, 1)
 #define   MI_DISPLAY_FLIP_PLANE(n) ((n) << 20)
 #define MI_STORE_DWORD_IMM	MI_INSTR(0x20, 1)
 #define   MI_MEM_VIRTUAL	(1 << 22) /* 965+ only */
@@ -368,6 +369,9 @@
 #define   CM0_RC_OP_FLUSH_DISABLE (1<<0)
 #define BB_ADDR		0x02140 /* 8 bytes */
 #define GFX_FLSH_CNTL	0x02170 /* 915+ only */
+#define ECOSKPD		0x021d0
+#define   ECO_GATING_CX_ONLY	(1<<3)
+#define   ECO_FLIP_DONE		(1<<0)
 
 /* GEN6 interrupt control */
 #define GEN6_RENDER_HWSTAM	0x2098
drivers/gpu/drm/i915/intel_display.c +24 -5
index f460761..ac23b29 100644
@@ -4587,10 +4587,10 @@ static void intel_unpin_work_fn(struct work_struct *__work)
 	kfree(work);
 }
 
-void intel_finish_page_flip(struct drm_device *dev, int pipe)
+static void do_intel_finish_page_flip(struct drm_device *dev,
+				      struct drm_crtc *crtc)
 {
 	drm_i915_private_t *dev_priv = dev->dev_private;
-	struct drm_crtc *crtc = dev_priv->pipe_to_crtc_mapping[pipe];
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct intel_unpin_work *work;
 	struct drm_i915_gem_object *obj_priv;
@@ -4634,6 +4634,22 @@ void intel_finish_page_flip(struct drm_device *dev, int pipe)
 	schedule_work(&work->work);
 }
 
+void intel_finish_page_flip(struct drm_device *dev, int pipe)
+{
+	drm_i915_private_t *dev_priv = dev->dev_private;
+	struct drm_crtc *crtc = dev_priv->pipe_to_crtc_mapping[pipe];
+
+	do_intel_finish_page_flip(dev, crtc);
+}
+
+void intel_finish_page_flip_plane(struct drm_device *dev, int plane)
+{
+	drm_i915_private_t *dev_priv = dev->dev_private;
+	struct drm_crtc *crtc = dev_priv->plane_to_crtc_mapping[plane];
+
+	do_intel_finish_page_flip(dev, crtc);
+}
+
 void intel_prepare_page_flip(struct drm_device *dev, int plane)
 {
 	drm_i915_private_t *dev_priv = dev->dev_private;
@@ -4729,14 +4745,17 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 			;
 
 	BEGIN_LP_RING(4);
-	OUT_RING(MI_DISPLAY_FLIP |
-		 MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
-	OUT_RING(fb->pitch);
 	if (IS_I965G(dev)) {
+		OUT_RING(MI_DISPLAY_FLIP |
+			 MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
+		OUT_RING(fb->pitch);
 		OUT_RING(obj_priv->gtt_offset | obj_priv->tiling_mode);
 		pipesrc = I915_READ(pipesrc_reg); 
 		OUT_RING(pipesrc & 0x0fff0fff);
 	} else {
+		OUT_RING(MI_DISPLAY_FLIP_I915 |
+			 MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
+		OUT_RING(fb->pitch);
 		OUT_RING(obj_priv->gtt_offset);
 		OUT_RING(MI_NOOP);
 	}
drivers/gpu/drm/i915/intel_drv.h +1 -0
index df931f7..72206f3 100644
@@ -224,6 +224,7 @@ extern void intel_fbdev_fini(struct drm_device *dev);
 
 extern void intel_prepare_page_flip(struct drm_device *dev, int plane);
 extern void intel_finish_page_flip(struct drm_device *dev, int pipe);
+extern void intel_finish_page_flip_plane(struct drm_device *dev, int plane);
 
 extern void intel_setup_overlay(struct drm_device *dev);
 extern void intel_cleanup_overlay(struct drm_device *dev);
Signature · 1 line
1.6.6.1

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

#37 lore raw
Quoted 9 lines · On Fri, 18 Jun 2010 13:04:50 -0700 Jesse Barnes <[email protected]>…
Quoted 6 lines · On Thu, 17 Jun 2010 19:44:10 -0700 Jesse Barnes <[email protected]>…
Quoted 3 lines · On Fri, 18 Jun 2010 02:20:23 +0200 Marc Deop i Argemí <[email protected]>…
Quoted 1 line · On Friday June 18 2010 02:17:53 Andrew Lutomirski wrote:

Neither patch applies for me.

Quoted 7 lines
Quoted 5 lines
Quoted 3 lines

One of them do apply for me, the other one doesn't.

Testing done on latest 2.6.35-rc3, the building fails.

Arg, ok, I'll refresh them and post new ones tomorrow.

Ok here are some updated ones.

Btw these were against drm-intel-next from a few minutes ago.

Signature · 5 lines
Jesse Barnes, Intel Open Source Technology Center
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
#38 lore raw
In reply to Jesse Barnes
Quoted 1 line · On Friday June 18 2010 22:04:50 Jesse Barnes wrote:

Ok here are some updated ones.

Already tried the patches. Now they both apply correctly on kernel 2.6.35-rc3.

However, this is what I find on the Xorg.0.log:

[   463.790] (EE) intel(0): Detected a hung GPU, disabling acceleration.

If I try to run glxgears the X window system crashes. This happens with both 2.11.0 and with 2.11.901RC

However, I'm trying right now the kernel 2.6.35-rc3 with the 2.11.901 RC and, for now, it seems stable (I still get corruption if I try to rotate the screen as well as in some KDE menus).

Regards

Signature · 4 lines
A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?
#39 lore raw
In reply to Jesse Barnes
Quoted 9 lines · Jesse Barnes
Quoted 6 lines · On Thu, 17 Jun 2010 19:44:10 -0700 Jesse Barnes <[email protected]>…
Quoted 3 lines · On Fri, 18 Jun 2010 02:20:23 +0200 Marc Deop i Argemí <[email protected]>…
Quoted 1 line · On Friday June 18 2010 02:17:53 Andrew Lutomirski wrote:

Neither patch applies for me.

Quoted 7 lines
Quoted 5 lines
Quoted 3 lines

One of them do apply for me, the other one doesn't.

Testing done on latest 2.6.35-rc3, the building fails.

Arg, ok, I'll refresh them and post new ones tomorrow.

Ok here are some updated ones.

Running these patches on 2.6.35-rc3 (plus the brown-paper-bag TCP fix, plus my hotplug_mask hack, plus my CRT regression fix) with xf86-video-intel be55066c6481b4c5e2cd39ef1c0f3be88cae0c93 (which is about a day old) seems stable and I don't have any visible corruption.

That being said, if I rotate the screen, then typing becomes very annoyingly laggy. I have no idea whether that's a regression or not.

--Andy

Quoted 10 lines

--
Jesse Barnes, Intel Open Source Technology Center

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

#40 lore raw
Quoted 14 lines · Andrew Lutomirski
Quoted 13 lines · Jesse Barnes
Quoted 10 lines · On Thu, 17 Jun 2010 19:44:10 -0700 Jesse Barnes <[email protected]>…
Quoted 6 lines · On Fri, 18 Jun 2010 02:20:23 +0200 Marc Deop i Argemí <[email protected]>…
Quoted 1 line · On Friday June 18 2010 02:17:53 Andrew Lutomirski wrote:

Neither patch applies for me.

One of them do apply for me, the other one doesn't.

Testing done on latest 2.6.35-rc3, the building fails.

Quoted 8 lines
Quoted 3 lines
Quoted 1 line

Arg, ok, I'll refresh them and post new ones tomorrow.

Ok here are some updated ones.

Running these patches on 2.6.35-rc3 (plus the brown-paper-bag TCP fix, plus my hotplug_mask hack, plus my CRT regression fix) with xf86-video-intel be55066c6481b4c5e2cd39ef1c0f3be88cae0c93 (which is about a day old) seems stable and I don't have any visible corruption.

Just froze again. I moved my mouse and the screen turned black except for the mouse cursor and a little underscore in the top left that looked like the fbcon cursor.

Again, magic sysrq didn't work, which I'd imagine would help narrow things down (presumably, no matter how hard the graphics hardware gets wedged, magic sysrq should still work).

--Andy

Quoted 17 lines

That being said, if I rotate the screen, then typing becomes very annoyingly laggy.  I have no idea whether that's a regression or not.

--Andy

Quoted 10 lines

--
Jesse Barnes, Intel Open Source Technology Center

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

#41 lore raw
Quoted 24 lines · On Mon, 21 Jun 2010 16:32:48 -0400 Andrew Lutomirski <[email protected]> wrote:
Quoted 14 lines · Andrew Lutomirski
Quoted 13 lines · Jesse Barnes
Quoted 10 lines · On Thu, 17 Jun 2010 19:44:10 -0700 Jesse Barnes <[email protected]>…
Quoted 6 lines · On Fri, 18 Jun 2010 02:20:23 +0200 Marc Deop i Argemí <[email protected]>…
Quoted 1 line · On Friday June 18 2010 02:17:53 Andrew Lutomirski wrote:

Neither patch applies for me.

One of them do apply for me, the other one doesn't.

Testing done on latest 2.6.35-rc3, the building fails.

Quoted 8 lines
Quoted 3 lines
Quoted 1 line

Arg, ok, I'll refresh them and post new ones tomorrow.

Ok here are some updated ones.

Running these patches on 2.6.35-rc3 (plus the brown-paper-bag TCP fix, plus my hotplug_mask hack, plus my CRT regression fix) with xf86-video-intel be55066c6481b4c5e2cd39ef1c0f3be88cae0c93 (which is about a day old) seems stable and I don't have any visible corruption.

Quoted 7 lines

Just froze again. I moved my mouse and the screen turned black except for the mouse cursor and a little underscore in the top left that looked like the fbcon cursor.

Again, magic sysrq didn't work, which I'd imagine would help narrow things down (presumably, no matter how hard the graphics hardware gets wedged, magic sysrq should still work).

Yeah means the GMCH itself probably hung, possibly not responding to memory requests from the CPU.

The display was otherwise idle when it froze? The behavior you describe sounds like a panic; there's a patch available to get some more info in that case: "vt/console: try harder to print output when panicing" that Dave Airlie just posted (attached). Can you apply it as well and see if you can reproduce the problem?

Thanks,

Signature · 1 line
Jesse Barnes, Intel Open Source Technology Center

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

#42 lore raw

[Dave, cc-ing you because your vt panic patch is involved.]

Quoted 35 lines · Jesse Barnes
Quoted 24 lines · On Mon, 21 Jun 2010 16:32:48 -0400 Andrew Lutomirski <[email protected]> wrote:
Quoted 14 lines · Andrew Lutomirski
Quoted 13 lines · Jesse Barnes
Quoted 10 lines · On Thu, 17 Jun 2010 19:44:10 -0700 Jesse Barnes <[email protected]>…
Quoted 6 lines · On Fri, 18 Jun 2010 02:20:23 +0200 Marc Deop i Argemí <[email protected]>…
Quoted 1 line · On Friday June 18 2010 02:17:53 Andrew Lutomirski wrote:

Neither patch applies for me.

One of them do apply for me, the other one doesn't.

Testing done on latest 2.6.35-rc3, the building fails.

Quoted 8 lines
Quoted 3 lines
Quoted 1 line

Arg, ok, I'll refresh them and post new ones tomorrow.

Ok here are some updated ones.

Running these patches on 2.6.35-rc3 (plus the brown-paper-bag TCP fix, plus my hotplug_mask hack, plus my CRT regression fix) with xf86-video-intel be55066c6481b4c5e2cd39ef1c0f3be88cae0c93 (which is about a day old) seems stable and I don't have any visible corruption.

Quoted 7 lines

Just froze again.  I moved my mouse and the screen turned black except for the mouse cursor and a little underscore in the top left that looked like the fbcon cursor.

Again, magic sysrq didn't work, which I'd imagine would help narrow things down (presumably, no matter how hard the graphics hardware gets wedged, magic sysrq should still work).

Quoted 8 lines

Yeah means the GMCH itself probably hung, possibly not responding to memory requests from the CPU.

The display was otherwise idle when it froze?  The behavior you describe sounds like a panic; there's a patch available to get some more info in that case: "vt/console: try harder to print output when panicing" that Dave Airlie just posted (attached).  Can you apply it as well and see if you can reproduce the problem?

Yes. That patch sort of worked, but the main part of the OOPS scrolled off the screen (other subsequent oopsen displaced it) and shift-pgup didn't do anything. (Neither did sysrq.)

Here's a blurry photo.

--Andy

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

#43 lore raw
Quoted 47 lines · On Thu, 1 Jul 2010 16:55:47 -0400 Andrew Lutomirski <[email protected]> wrote:

[Dave, cc-ing you because your vt panic patch is involved.]

Quoted 35 lines · Jesse Barnes
Quoted 24 lines · On Mon, 21 Jun 2010 16:32:48 -0400 Andrew Lutomirski <[email protected]> wrote:
Quoted 14 lines · Andrew Lutomirski
Quoted 13 lines · Jesse Barnes
Quoted 10 lines · On Thu, 17 Jun 2010 19:44:10 -0700 Jesse Barnes <[email protected]>…
Quoted 6 lines · On Fri, 18 Jun 2010 02:20:23 +0200 Marc Deop i Argemí <[email protected]>…
Quoted 1 line · On Friday June 18 2010 02:17:53 Andrew Lutomirski wrote:

Neither patch applies for me.

One of them do apply for me, the other one doesn't.

Testing done on latest 2.6.35-rc3, the building fails.

Quoted 8 lines
Quoted 3 lines
Quoted 1 line

Arg, ok, I'll refresh them and post new ones tomorrow.

Ok here are some updated ones.

Running these patches on 2.6.35-rc3 (plus the brown-paper-bag TCP fix, plus my hotplug_mask hack, plus my CRT regression fix) with xf86-video-intel be55066c6481b4c5e2cd39ef1c0f3be88cae0c93 (which is about a day old) seems stable and I don't have any visible corruption.

Quoted 7 lines

Just froze again.  I moved my mouse and the screen turned black except for the mouse cursor and a little underscore in the top left that looked like the fbcon cursor.

Again, magic sysrq didn't work, which I'd imagine would help narrow things down (presumably, no matter how hard the graphics hardware gets wedged, magic sysrq should still work).

Quoted 8 lines

Yeah means the GMCH itself probably hung, possibly not responding to memory requests from the CPU.

The display was otherwise idle when it froze?  The behavior you describe sounds like a panic; there's a patch available to get some more info in that case: "vt/console: try harder to print output when panicing" that Dave Airlie just posted (attached).  Can you apply it as well and see if you can reproduce the problem?

Quoted 5 lines

Yes. That patch sort of worked, but the main part of the OOPS scrolled off the screen (other subsequent oopsen displaced it) and shift-pgup didn't do anything. (Neither did sysrq.)

Here's a blurry photo.

Ah great, so we're getting some info. But the good stuff scrolled off the screen. Can you try setting up netconsole so you can capture all of it in a terminal on a working machine?

You can also match the call trace info back to the source line. The call trace will look something like
i915_gem_execbuffer 0x123/0x456
where the first number is the offset into the function and the second is the size of the function. Once you have that (just grab the first one, usually up near the top of the register dump) you can use gdb to find it:
gdb /path/to/i915.ko
...
(gdb) list *i915_gem_execbuffer+0x123
...
That should show you exactly where things went bad...

Thanks,

Signature · 5 lines
Jesse Barnes, Intel Open Source Technology Center
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
#44 lore raw
Quoted 54 lines · Jesse Barnes
Quoted 51 lines · On Thu, 1 Jul 2010 16:55:47 -0400 Andrew Lutomirski <[email protected]> wrote:

[Dave, cc-ing you because your vt panic patch is involved.]

Quoted 35 lines · Jesse Barnes
Quoted 24 lines · On Mon, 21 Jun 2010 16:32:48 -0400 Andrew Lutomirski <[email protected]> wrote:
Quoted 14 lines · Andrew Lutomirski
Quoted 13 lines · Jesse Barnes
Quoted 10 lines · On Thu, 17 Jun 2010 19:44:10 -0700 Jesse Barnes <[email protected]>…
Quoted 6 lines · On Fri, 18 Jun 2010 02:20:23 +0200 Marc Deop i Argemí <[email protected]>…
Quoted 1 line · On Friday June 18 2010 02:17:53 Andrew Lutomirski wrote:

Neither patch applies for me.

One of them do apply for me, the other one doesn't.

Testing done on latest 2.6.35-rc3, the building fails.

Quoted 8 lines
Quoted 3 lines
Quoted 1 line

Arg, ok, I'll refresh them and post new ones tomorrow.

Ok here are some updated ones.

Running these patches on 2.6.35-rc3 (plus the brown-paper-bag TCP fix, plus my hotplug_mask hack, plus my CRT regression fix) with xf86-video-intel be55066c6481b4c5e2cd39ef1c0f3be88cae0c93 (which is about a day old) seems stable and I don't have any visible corruption.

Quoted 7 lines

Just froze again.  I moved my mouse and the screen turned black except for the mouse cursor and a little underscore in the top left that looked like the fbcon cursor.

Again, magic sysrq didn't work, which I'd imagine would help narrow things down (presumably, no matter how hard the graphics hardware gets wedged, magic sysrq should still work).

Quoted 8 lines

Yeah means the GMCH itself probably hung, possibly not responding to memory requests from the CPU.

The display was otherwise idle when it froze?  The behavior you describe sounds like a panic; there's a patch available to get some more info in that case: "vt/console: try harder to print output when panicing" that Dave Airlie just posted (attached).  Can you apply it as well and see if you can reproduce the problem?

Yes.  That patch sort of worked, but the main part of the OOPS scrolled off the screen (other subsequent oopsen displaced it) and shift-pgup didn't do anything.  (Neither did sysrq.)

Quoted 5 lines
Quoted 1 line

Here's a blurry photo.

Ah great, so we're getting some info.  But the good stuff scrolled off the screen.  Can you try setting up netconsole so you can capture all of it in a terminal on a working machine?

Um, maybe. I've yet to trigger this when I'm near another computer.

But it just happened again and the attached (terrible) picture might
be interesting. But given the fbcon_switch lines, I bet we're seeing
the wrong thing. I just set pause_on_oops=120 and I'll let you know if I see it again.

--Andy

Quoted 14 lines

You can also match the call trace info back to the source line.  The call trace will look something like
 i915_gem_execbuffer 0x123/0x456
where the first number is the offset into the function and the second is the size of the function.  Once you have that (just grab the first one, usually up near the top of the register dump) you can use gdb to find it:
 gdb /path/to/i915.ko
 ...
 (gdb) list *i915_gem_execbuffer+0x123
 ...
That should show you exactly where things went bad...

Quoted 4 lines

Thanks,
--
Jesse Barnes, Intel Open Source Technology Center

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

#45 lore raw
In reply to Carl Worth

В сообщении от 15 июня 2010 03:39:54 автор Carl Worth написал:

Quoted 5 lines

This is the first release candidate in preparation for the upcoming 2.12.0 release. We will appreciate any feedback we can get from testing of this snapshot to improve the 2.12.0 release.

-Carl

Sorry, but I got artifacts with latest version of libdrm and xf86-video-intel on 945gm hardware (see attached screenshot).

If you need some additional info - just ask.

Regards
Vasily

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

author
#46 lore raw
Quoted 2 lines

Sorry, but I got artifacts with latest version of libdrm and xf86-video-intel on 945gm hardware (see attached screenshot).

Thanks for the report.

Quoted 1 line

If you need some additional info - just ask.

Actually, I do. I'm not familiar enough with the particular theme you're using to know what looks wrong there. Could you supply an imeage showing what things should look like?

And if this behavior started only with this release candidate, would you be willing to bisect the commits in the xf86-video-intel repository to identify the offending commit?

Thanks for your report,

-Carl

Signature · 1 line
[email protected]

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

#47 lore raw

В сообщении от 15 июня 2010 19:49:41 автор Carl Worth написал:

Quoted 10 lines
Quoted 2 lines

Sorry, but I got artifacts with latest version of libdrm and xf86-video-intel on 945gm hardware (see attached screenshot).

Thanks for the report.

Quoted 1 line

If you need some additional info - just ask.

Actually, I do. I'm not familiar enough with the particular theme you're using to know what looks wrong there. Could you supply an imeage showing what things should look like?

It's default kde theme - oxygen, screen how should menu selection look like is in attachment.

Quoted 3 lines

And if this behavior started only with this release candidate, would you be willing to bisect the commits in the xf86-video-intel repository to identify the offending commit?

Ok, I'll do.

Regards
Vasily

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

#48 lore raw

В сообщении от 15 июня 2010 19:55:55 автор Vasily Khoruzhick написал:

Quoted 5 lines
Quoted 3 lines

And if this behavior started only with this release candidate, would you be willing to bisect the commits in the xf86-video-intel repository to identify the offending commit?

Ok, I'll do.

bisected:

First faulty commit is:

commit f74b3f82bab11463a0f4de9f614fc6aa1492ef24

Author: Chris Wilson <[email protected]>
Date:   Tue Jun 1 22:31:35 2010 +0100

i915; Avoid the implicit flush on changing BUF_INFO

Regards
Vasily

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

#49 lore raw
In reply to Vasily Khoruzhick

Hi,

I also get artifacts with kde's oxygen QT theme, however everything else looks great so far.
Probably related to QT's gradient implementation, doing tons of put/getimage :/

- Clemens

2010/6/15 Vasily Khoruzhick <[email protected]>:

Quoted 12 lines

В сообщении от 15 июня 2010 03:39:54 автор Carl Worth написал:

Quoted 5 lines

This is the first release candidate in preparation for the upcoming 2.12.0 release. We will appreciate any feedback we can get from testing of this snapshot to improve the 2.12.0 release.

-Carl

Sorry, but I got artifacts with latest version of libdrm and xf86-video-intel on 945gm hardware (see attached screenshot).

If you need some additional info - just ask.

Quoted 9 lines

Regards
Vasily

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

#50 lore raw
Quoted 2 lines · Clemens Eisserer

I also get artifacts with kde's oxygen QT theme, however everything else looks great so far.

Vasily, Clemens can you both try reverting f74b3f82 on top of 2.11.901 and seeing if that fixes the corruption for you. It did not seem to make a difference here...

Thanks,
-ickle

Signature · 1 line
Chris Wilson, Intel Open Source Technology Centre
#51 lore raw

Hi Chris,

Quoted 3 lines

Vasily, Clemens can you both try reverting f74b3f82 on top of 2.11.901 and seeing if that fixes the corruption for you. It did not seem to make a difference here...

Sorry if that sounds stupid, but how can I do that?

Thanks Clemens

PS: Saw you experimented with shader based trapezoid rasterization. just to be curious, what were the results so far?

author
#52 lore raw
Quoted 7 lines · Clemens Eisserer

Hi Chris,

Quoted 3 lines

Vasily, Clemens can you both try reverting f74b3f82 on top of 2.11.901 and seeing if that fixes the corruption for you. It did not seem to make a difference here...

Sorry if that sounds stupid, but how can I do that?

No problem at all. Just checkout the driver from git with:

git clone git://anongit.freedesktop.org/xorg/driver/xf86-video-intel

Then checkout 2.11.901 specifically with:

	git checkout 2.11.901

And finally, revert the commit of interest with:

	git revert f74b3f82

Then compile, and test the driver, and let us know.

-Carl

Signature · 1 line
[email protected]

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

#53 lore raw
In reply to Clemens Eisserer
Quoted 7 lines · Clemens Eisserer

Hi Chris,

Quoted 3 lines

Vasily, Clemens can you both try reverting f74b3f82 on top of 2.11.901 and seeing if that fixes the corruption for you. It did not seem to make a difference here...

Sorry if that sounds stupid, but how can I do that?

Starting from scratch:

 $ git clone git://anongit.freedesktop.org/git/xorg/driver/xf86-video-intel.git
 $ cd xf86-video-intel
 $ git checkout 2.11.901
 $ git revert f74b3f82

Build and install as usual.

Quoted 4 lines

Thanks Clemens

PS: Saw you experimented with shader based trapezoid rasterization. just to be curious, what were the results so far?

Give-or-take that it is dubious that we can do a simple coverage based accumulation of subpixels trapezoids within the specification of PolyModeImprecise, benchmarking shows a small performance win for most cairo traces [the improvements matching the elimination of pixman_image_rasterize_trapezoid() and co], but in a few cases it can be twice as slow. So I think it is back to a span renderer -- even if I just trade GPU time for CPU time, that should be a win for application performance.

Signature · 1 line
Chris Wilson, Intel Open Source Technology Centre
#54 lore raw

Hi Chris,

Quoted 2 lines
 $ git checkout 2.11.901
 $ git revert f74b3f82

Ah thanks, didn't know the checkout thing.

Hmm, if I revert that commit it no longer builds:
CC i915_render.lo
i915_render.c: In function ‘i915_prepare_composite’:
i915_render.c:836: error: ‘intel_screen_private’ has no member named ‘render_current_dest’

- Clemens

Quoted 6 lines

Give-or-take that it is dubious that we can do a simple coverage based accumulation of subpixels trapezoids within the specification of PolyModeImprecise, benchmarking shows a small performance win for most cairo traces [the improvements matching the elimination of pixman_image_rasterize_trapezoid() and co], but in a few cases it can be twice as slow.

Whatever the result will look like, I am really curious - as some workloads show a lot of time spent in rasterize_edge().

Quoted 3 lines

So I think it is back to a span renderer -- even if I just trade GPU time for CPU time, that should be a win for application performance.

just reminds me at the commit "OMG! spans!" ;)
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

#55 lore raw

Hi again,

I bisected and indeed f74b3f82 is the first bad commit:

f74b3f82bab11463a0f4de9f614fc6aa1492ef24 is the first bad commit commit f74b3f82bab11463a0f4de9f614fc6aa1492ef24

Author: Chris Wilson <[email protected]>
Date:   Tue Jun 1 22:31:35 2010 +0100

i915; Avoid the implicit flush on changing BUF_INFO

Thanks, Clemens

#56 lore raw
In reply to Clemens Eisserer
Quoted 11 lines · Clemens Eisserer

Hi Chris,

Quoted 2 lines
 $ git checkout 2.11.901
 $ git revert f74b3f82

Ah thanks, didn't know the checkout thing.

Hmm, if I revert that commit it no longer builds:
CC i915_render.lo
i915_render.c: In function ‘i915_prepare_composite’:
i915_render.c:836: error: ‘intel_screen_private’ has no member named ‘render_current_dest’

Argh. It has the vital half of the centre sampling fix mixed in.

This hack should be sufficient to force the target buffer to be updated everytime.

src/i915_render.c +1 -1
index 8870019..3505083 100644
@@ -1435,7 +1435,7 @@ static void i915_emit_composite_setup(ScrnInfoPtr scrn)
        intel->last_3d = LAST_3D_RENDER;
 
        /* BUF_INFO is an implicit flush, so avoid if the target has not changed */
-       if (dest != intel->render_current_dest) {
+       if (1 || dest != intel->render_current_dest) {
                uint32_t tiling_bits;
 
                tiling_bits = 0;

That does indeed seem to have an effect on KDE -- can you confirm if it fixes the rendering corruption and if there remains any?
Now to understand why.

Thanks,
-ickle

Signature · 1 line
Chris Wilson, Intel Open Source Technology Centre

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

#57 lore raw

Hi Chris,

Quoted 2 lines

That does indeed seem to have an effect on KDE -- can you confirm if it fixes the rendering corruption and if there remains any?

With that change the corruptions seem to be gone.

- Clemens

#58 lore raw
In reply to Chris Wilson

В сообщении от 15 июня 2010 23:15:02 автор Chris Wilson написал:

Quoted 1 line

On Tue, 15 Jun 2010 21:58:57 +0200, Clemens Eisserer <[email protected]>

Quoted 9 lines · wrote:
Quoted 2 lines

I also get artifacts with kde's oxygen QT theme, however everything else looks great so far.

Vasily, Clemens can you both try reverting f74b3f82 on top of 2.11.901 and seeing if that fixes the corruption for you. It did not seem to make a difference here...

Thanks,
-ickle

Driver doesn't compile with this commit reverted, sorry.

Regards
Vasily

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx