There is a new driver from intel. Is an ebuild available? Reproducible: Always
Needs backports into the kernel to work properly. Will be handled when I have some time. Thanks
I tried it myself, used the ebuild 2.9.1 and changed libdrm (>=2.4.16) and xorgserver (> 1.7). Compiles and seems to work. The kernel patch could be downloaded here: http://gitorious.org/daniel-s-linux-stuff/linux-kernel/commits/intel-kms-overlay-for-2.6.32 furhter information: http://intellinuxgraphics.org/2009Q4.html
btw: I did not patch the kernel... I will try to patch gentoo-sources ... if there is time...
@kernel, could you guys pick up the 5 or so patches needed to get KMS overlay support? -intel 2.10.0 no longer works without KMS so this is needed to support XVideo playback on 8xx through 945 chips. Let me know if you have any questions/concerns. Thanks
In portage under p.mask @herd, starting with 2.10.0, the ebuild should check for KMS support in .config and print something if it's not found (like synpatics or wacom). I haven't done it yet, but feel free to beat me to it. Thanks
Created attachment 215803 [details] Ebuild for xf86-video-intel Something like this?
After installing 2.10.0, kernel 2.6.32-gentoo-r1 and unstable mesa+dependencies (includes xorg-server), I sometimes get this error many times in the log, and graphics freezes: (EE) intel(0): Failed to submit batch buffer, expect rendering corruption or even a frozen display: Input/output error. So far, I think it's only happened when sliding (animated) from one desktop to another in KDE4. At least the last time, there was a new-mail popup window that was fading in/out at the same time that I tried to change desktops (it got stuck half-way). Should I file a separate bug report, and if so, what info should I provide? (II) intel(0): Integrated Graphics Chipset: Intel(R) G33 (--) intel(0): Chipset: "G33"
(In reply to comment #7) > After installing 2.10.0, kernel 2.6.32-gentoo-r1 and unstable mesa+dependencies > (includes xorg-server), I sometimes get this error many times in the log, and > graphics freezes: > > (EE) intel(0): Failed to submit batch buffer, expect rendering corruption or > even a frozen display: Input/output error. > > So far, I think it's only happened when sliding (animated) from one desktop to > another in KDE4. At least the last time, there was a new-mail popup window that > was fading in/out at the same time that I tried to change desktops (it got > stuck half-way). > > Should I file a separate bug report, and if so, what info should I provide? > > (II) intel(0): Integrated Graphics Chipset: Intel(R) G33 > (--) intel(0): Chipset: "G33" I hit exactly the same display freeze, only a reboot can bring the display back. (II) intel(0): Integrated Graphics Chipset: Intel(R) 915GM (--) intel(0): Chipset: "915GM" running lxde+compiz, not kde
(In reply to comment #6) > Something like this? Looks good but a unified diff would be easier for us to review and apply. (In reply to comment #8) (In reply to comment #7) [snip] Please try 2.6.33_rcXX kernels, there's a good chance the bug has already been fixed there. If not, please file _new_ bug reports. Thanks
Created attachment 217224 [details] diff against xf86-video-intel-2.10.0, add info about needed KMS support
(In reply to comment #10) > Created an attachment (id=217224) [details] > diff against xf86-video-intel-2.10.0, add info about needed KMS support > Hey guys, many owners of GMA500 devices are waiting this driver :)
(In reply to comment #11) > Hey guys, many owners of GMA500 devices are waiting this driver :) Again, it won't be unmasked until either gentoo-sources-2.6.33 hits portage or .32 gets the required patches for KMS Overlay support. Until then, just unmask the driver. Thanks
I made a ticket with the i915 KMS Video Overlay patch (bug #303521).
*** Bug 303521 has been marked as a duplicate of this bug. ***
Kernel team is already CCed here, no need to open a new bug. Thanks
(In reply to comment #15) > Kernel team is already CCed here, no need to open a new bug. Ok, sorry. I was thinking that patching the kernel is one problem and unmasking the xf86-video-intel ebuild is another one (that depends on the previous one). That's why I submitted the patch in a separate ticket. Sorry for the inconvenience.
Linux Kernel 2.6.33 have just been released and it is already in portage, so i think this is good time to unmask this driver. But it should depends on 2.6.33 kernel.
Please note that 2.10.901 has been released: http://lists.freedesktop.org/archives/intel-gfx/2010-February/005961.html This is the first release candidate working toward an upcoming 2.11 release. The most significant new feature of this release is support for new DRI2 APIs, allowing page flipping to occur for swaps that are full-screen and not rotated. Some significant bug fixes -------------------------- * Avoid fallback due to failing to extract pixel value from an alpha-only solid, (this fix eliminated 1 second from Mobilin boot time). https://bugs.freedesktop.org/show_bug.cgi?id=26189 * Handle full-height blits without tearing in CopyRegion https://bugs.freedesktop.org/show_bug.cgi?id=22475 Several other fixes and cleanups are included as well, (including the removal of stale utility code that is now maintained in the separate intel-gpu-tools module).
Fixed in portage. Thanks :)
