<html> <head> <base href="https://bugs.freedesktop.org/" /> </head> <body> <div> <a class="bz_bug_link bz_status_NEW " title="NEW - Barts (HD6850): Failure in evergreen_surface_check_2d" href="https://bugs.freedesktop.org/show_bug.cgi?id=75357#c10">Comment # 10</a> on <a class="bz_bug_link bz_status_NEW " title="NEW - Barts (HD6850): Failure in evergreen_surface_check_2d" href="https://bugs.freedesktop.org/show_bug.cgi?id=75357">bug 75357</a> from <a class="email" href="mailto:nix@esperi.org.uk" title="Nix <nix@esperi.org.uk>"> Nix</a> <pre>FWIW, I have just started seeing behaviour very similar to this, also with an HD6850, though it is probably a distinct bug because all the software versions involved are very different. This happened after closely-adjoined upgrades of the kernel (from 3.19 to 4.0.3), the X server (from 1.15.1, yes, I know, it's because of the security holes I finally upgraded, to 1.17.1), the ATI driver (from 7.3.0-26 to 7.5.0.68, all "git describe" output from the unmodified trunk), Mesa (from 10.2.6 to 10.5.5), and LLVM (from 3.3 to 3.5.2), so the fault could, alas, in theory lie in any or all of these. Mesa is using Gallium in both cases. Glamor is not in use. dpm is, and has been working flawlessly thank you very much, cutting my desktop's power consumption from 130W to 90W :) over the last year that's saved at least the price of the beer I owe you for the dpm work, Michel :) With this setup, with an older Chromium (SVN revision 322707) I see very frequent failures to map the interior of its windows at or shortly after startup, sometimes accompanied by a total loss of the ability to switch focus, whether by keyboard or mouse, whether on a desktop where the Chromium window was mapped or not (X still responded, so I could switch out of X and kill Chromium, whereupon X recovered). Chromium screams on its stderr as this happens: [8893:8893:0523/225116:ERROR:gpu_channel_host.cc(175)] GpuChannelHost::CreateViewCommandBuffer failed. [8893:8893:0523/225116:ERROR:webgraphicscontext3d_command_buffer_impl.cc(226)] Failed to initialize command buffer. [8893:8893:0523/225116:ERROR:gpu_channel_host.cc(175)] GpuChannelHost::CreateViewCommandBuffer failed. [8893:8893:0523/225116:ERROR:webgraphicscontext3d_command_buffer_impl.cc(226)] Failed to initialize command buffer. [8893:8893:0523/225116:ERROR:gpu_process_transport_factory.cc(574)] Lost UI shared context. [9929:9929:0523/225118:ERROR:sandbox_linux.cc(345)] InitializeSandbox() called with multiple threads in process gpu-process [15:41:0523/225118:ERROR:gpu_channel_host.cc(175)] GpuChannelHost::CreateViewCommandBuffer failed. [15:41:0523/225118:ERROR:webgraphicscontext3d_command_buffer_impl.cc(226)] Failed to initialize command buffer. [8893:8916:0523/225118:ERROR:channel.cc(300)] RawChannel read error (connection broken) [8893:8916:0523/225118:ERROR:channel.cc(300)] RawChannel read error (connection broken) [322:337:0523/225119:ERROR:gpu_channel_host.cc(175)] GpuChannelHost::CreateViewCommandBuffer failed. [322:337:0523/225119:ERROR:webgraphicscontext3d_command_buffer_impl.cc(226)] Failed to initialize command buffer. [8893:8916:0523/225119:ERROR:channel.cc(300)] RawChannel read error (connection broken) [9929:9929:0523/225119:ERROR:gles2_cmd_decoder.cc(11630)] [GroupMarkerNotSet(crbug.com/242999)!:D088433BA20D0000]GL ERROR :GL_INVALID_OPERATION : glCreateAndConsumeTextureCHROMIUM: invalid mailbox name [9929:9929:0523/225119:ERROR:gles2_cmd_decoder.cc(4431)] [GroupMarkerNotSet(crbug.com/242999)!:D088433BA20D0000]GL ERROR :GL_INVALID_OPERATION : glBindTexture: id not generated by glGenTextures [9929:9929:0523/225119:ERROR:gles2_cmd_decoder.cc(11630)] [GroupMarkerNotSet(crbug.com/242999)!:D088433BA20D0000]GL ERROR :GL_INVALID_OPERATION : glCreateAndConsumeTextureCHROMIUM: invalid mailbox name [9929:9929:0523/225119:ERROR:gles2_cmd_decoder.cc(4431)] [GroupMarkerNotSet(crbug.com/242999)!:D088433BA20D0000]GL ERROR :GL_INVALID_OPERATION : glBindTexture: id not generated by glGenTextures (There was nothing in dmesg when this happened.) So far, so problematic. But I upgraded Chromium to r331233 today and it got *ever* so much worse. The outward symptoms were the same -- until X crashed with an obvious oops-I-didn't-check-errno error: Fatal server error: [354365.137] (EE) failed to map pixmap -22 In dmesg, we see rather a lot of complaining: May 23 21:59:12 mutilate warning: : [353427.133393] radeon 0000:01:00.0: evergreen_surface_check_2d:278 texture pitch 128 invalid must be aligned with 512 May 23 21:59:12 mutilate warning: : [353427.133402] radeon 0000:01:00.0: evergreen_cs_track_validate_texture:827 texture invalid 0x01ec03c1 0x40000013 0x060a0000 0x00000000 0x80000000 0x800304da May 23 21:59:12 mutilate err: : [353427.133407] [drm:radeon_cs_ioctl] *ERROR* Invalid command stream ! May 23 21:59:13 mutilate err: : [353428.036913] [TTM] Could not find buffer object to map May 23 22:02:03 mutilate err: : [353598.745908] [drm:radeon_cs_parser_relocs] *ERROR* gem object lookup failed 0x421 May 23 22:02:03 mutilate err: : [353598.745917] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -2! May 23 22:02:03 mutilate err: : [353598.838717] [drm:radeon_cs_parser_relocs] *ERROR* gem object lookup failed 0x421 May 23 22:02:03 mutilate err: : [353598.838727] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -2! May 23 22:02:03 mutilate err: : [353598.876327] [drm:radeon_cs_parser_relocs] *ERROR* gem object lookup failed 0x421 May 23 22:02:03 mutilate err: : [353598.876336] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -2! May 23 22:02:11 mutilate err: : [353606.091591] [drm:radeon_cs_parser_relocs] *ERROR* gem object lookup failed 0x5f May 23 22:02:11 mutilate err: : [353606.091600] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -2! May 23 22:02:11 mutilate err: : [353606.106477] [drm:radeon_cs_parser_relocs] *ERROR* gem object lookup failed 0x5f May 23 22:02:11 mutilate err: : [353606.106491] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -2! May 23 22:02:11 mutilate err: : [353606.255578] [drm:radeon_cs_parser_relocs] *ERROR* gem object lookup failed 0x5f May 23 22:02:11 mutilate err: : [353606.255586] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -2! May 23 22:02:15 mutilate err: : [353610.061328] [drm:radeon_cs_parser_relocs] *ERROR* gem object lookup failed 0x5f May 23 22:02:15 mutilate err: : [353610.061341] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -2! May 23 22:02:15 mutilate warning: : [353610.346573] radeon 0000:01:00.0: evergreen_cs_track_validate_cb:477 cb[0] bo too small (layer size 1148928, offset 0, max layer 1, bo size 8192, slice 4487) May 23 22:02:15 mutilate warning: : [353610.346581] radeon 0000:01:00.0: evergreen_cs_track_validate_cb:481 problematic surf: (704 408) (0 4 1 0 0 4 3) May 23 22:02:15 mutilate warning: : [353610.346584] radeon 0000:01:00.0: evergreen_packet3_check:1970 invalid cmd stream 454 May 23 22:02:15 mutilate err: : [353610.346589] [drm:radeon_cs_ioctl] *ERROR* Invalid command stream ! May 23 22:02:15 mutilate err: : [353610.667843] [drm:radeon_cs_parser_relocs] *ERROR* gem object lookup failed 0x62 May 23 22:02:15 mutilate err: : [353610.667856] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -2! May 23 22:02:15 mutilate err: : [353610.846176] [drm:radeon_cs_parser_relocs] *ERROR* gem object lookup failed 0x5f May 23 22:02:15 mutilate err: : [353610.846189] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -2! May 23 22:02:15 mutilate err: : [353610.847112] [drm:radeon_cs_parser_relocs] *ERROR* gem object lookup failed 0x5f May 23 22:02:15 mutilate err: : [353610.847118] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -2! May 23 22:02:15 mutilate err: : [353610.863055] [drm:radeon_cs_parser_relocs] *ERROR* gem object lookup failed 0x5f May 23 22:02:15 mutilate err: : [353610.863064] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -2! May 23 22:02:15 mutilate err: : [353610.865096] [drm:radeon_cs_parser_relocs] *ERROR* gem object lookup failed 0x5f May 23 22:02:15 mutilate err: : [353610.865103] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -2! May 23 22:02:15 mutilate err: : [353610.879232] [drm:radeon_cs_parser_relocs] *ERROR* gem object lookup failed 0x5f May 23 22:02:15 mutilate err: : [353610.879240] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -2! May 23 22:02:15 mutilate err: : [353610.879883] [drm:radeon_cs_parser_relocs] *ERROR* gem object lookup failed 0x5f May 23 22:02:15 mutilate err: : [353610.879888] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -2! May 23 22:02:16 mutilate err: : [353610.895706] [drm:radeon_cs_parser_relocs] *ERROR* gem object lookup failed 0x5f May 23 22:02:16 mutilate err: : [353610.895714] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -2! May 23 22:02:16 mutilate err: : [353610.896413] [drm:radeon_cs_parser_relocs] *ERROR* gem object lookup failed 0x5f May 23 22:02:16 mutilate err: : [353610.896418] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -2! May 23 22:04:50 mutilate err: : [353765.784060] [drm:radeon_cs_parser_relocs] *ERROR* gem object lookup failed 0x185 May 23 22:04:50 mutilate err: : [353765.784067] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -2! May 23 22:04:55 mutilate err: : [353770.520155] [drm:radeon_cs_parser_relocs] *ERROR* gem object lookup failed 0x185 May 23 22:04:55 mutilate err: : [353770.520218] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -2! May 23 22:04:57 mutilate err: : [353772.122908] [drm:radeon_cs_parser_relocs] *ERROR* gem object lookup failed 0x50c May 23 22:04:57 mutilate err: : [353772.122916] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -2! May 23 22:04:57 mutilate warning: : [353772.132150] radeon 0000:01:00.0: evergreen_cs_track_validate_cb:477 cb[0] bo too small (layer size 1148928, offset 0, max layer 1, bo size 8192, slice 4487) May 23 22:04:57 mutilate warning: : [353772.132159] radeon 0000:01:00.0: evergreen_cs_track_validate_cb:481 problematic surf: (704 408) (0 4 1 0 0 4 3) May 23 22:04:57 mutilate warning: : [353772.132230] radeon 0000:01:00.0: evergreen_packet3_check:1970 invalid cmd stream 454 May 23 22:04:57 mutilate err: : [353772.132239] [drm:radeon_cs_ioctl] *ERROR* Invalid command stream ! May 23 22:05:07 mutilate warning: : [353782.686843] radeon 0000:01:00.0: evergreen_cs_track_validate_cb:477 cb[0] bo too small (layer size 1148928, offset 0, max layer 1, bo size 8192, slice 4487) May 23 22:05:07 mutilate warning: : [353782.686851] radeon 0000:01:00.0: evergreen_cs_track_validate_cb:481 problematic surf: (704 408) (0 4 1 0 0 4 3) May 23 22:05:07 mutilate warning: : [353782.686854] radeon 0000:01:00.0: evergreen_packet3_check:1970 invalid cmd stream 454 May 23 22:05:07 mutilate err: : [353782.686859] [drm:radeon_cs_ioctl] *ERROR* Invalid command stream ! May 23 22:05:10 mutilate err: : [353784.942883] [drm:radeon_cs_parser_relocs] *ERROR* gem object lookup failed 0x185 May 23 22:05:10 mutilate err: : [353784.942893] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -2! [... lots more of this ...] May 23 22:10:39 mutilate warning: : [354113.922155] radeon 0000:01:00.0: vbo resource seems too big for the bo May 23 22:10:39 mutilate warning: : [354113.922166] radeon 0000:01:00.0: vbo resource seems too big for the bo May 23 22:10:39 mutilate warning: : [354113.922170] radeon 0000:01:00.0: vbo resource seems too big for the bo May 23 22:10:39 mutilate warning: : [354113.922173] radeon 0000:01:00.0: vbo resource seems too big for the bo May 23 22:10:39 mutilate warning: : [354113.922176] radeon 0000:01:00.0: vbo resource seems too big for the bo May 23 22:10:39 mutilate warning: : [354113.922179] radeon 0000:01:00.0: vbo resource seems too big for the bo May 23 22:10:39 mutilate warning: : [354113.922182] radeon 0000:01:00.0: vbo resource seems too big for the bo May 23 22:10:39 mutilate warning: : [354113.922185] radeon 0000:01:00.0: vbo resource seems too big for the bo May 23 22:10:39 mutilate warning: : [354113.922188] radeon 0000:01:00.0: vbo resource seems too big for the bo May 23 22:10:39 mutilate warning: : [354113.922191] radeon 0000:01:00.0: vbo resource seems too big for the bo May 23 22:10:39 mutilate warning: : [354113.922194] radeon 0000:01:00.0: vbo resource seems too big for the bo May 23 22:10:39 mutilate warning: : [354113.922196] radeon 0000:01:00.0: vbo resource seems too big for the bo May 23 22:10:39 mutilate warning: : [354113.922199] radeon 0000:01:00.0: vbo resource seems too big for the bo May 23 22:10:39 mutilate warning: : [354113.922202] radeon 0000:01:00.0: vbo resource seems too big for the bo [... more, followed finally by ...] May 23 21:59:12 mutilate warning: : [353427.133393] radeon 0000:01:00.0: evergreen_surface_check_2d:278 texture pitch 128 invalid must be aligned with 512 May 23 21:59:12 mutilate warning: : [353427.133402] radeon 0000:01:00.0: evergreen_cs_track_validate_texture:827 texture invalid 0x01ec03c1 0x40000013 0x060a0000 0x00000000 0x80000000 0x800304da May 23 21:59:12 mutilate err: : [353427.133407] [drm:radeon_cs_ioctl] *ERROR* Invalid command stream ! May 23 21:59:13 mutilate err: : [353428.036913] [TTM] Could not find buffer object to map Following which the X server crashed, as above. If I start Chromium with --disable-gpu, the problem goes away in both cases, so the fault is likely either in Mesa or in the kernel or in the interaction between the two. (I had the Chromium software rendering blacklist overridden years earlier because it was trying to avoid problems that had been long fixed. I've now reversed that decision, obviously!)</pre> </div> <hr> You are receiving this mail because: <ul> <li>You are the assignee for the bug.</li> </ul> </body> </html>