The kernel commit 167f3a04d7366d65c7fa9a92f0d604cdcf4a11ae titled "drm: read EDID extensions from monitor" makes the kernel unable to read the edid from my monitor, connected via HDMI. Reverting this commit and the following (f23c20c83d523e5f8cda1f8f7ed52fe6afffbe29, to get the kernel to compile), makes the kernel able to read the edid again. A working kernel produces: dagb-home edid-decode # ./edid-decode "/sys/class/drm/card0/card0-HDMI Type A-1/edid" Extracted contents: header: 00 ff ff ff ff ff ff 00 serial number: 09 d1 0a 78 45 54 00 00 2f 11 version: 01 03 basic params: 80 34 20 78 2e chroma info: c5 c4 a3 57 4a 9c 23 12 4f 52 established: a5 6b 80 standard: 71 00 81 00 95 00 81 80 b3 00 a9 40 d1 00 01 01 descriptor 1: 28 3c 80 a0 70 b0 23 40 30 20 36 00 07 44 21 00 00 1a descriptor 2: 00 00 00 ff 00 4c 42 37 30 30 32 35 33 5a 4c 30 0a 20 descriptor 3: 00 00 00 fd 00 32 55 1f 5e 11 00 0a 20 20 20 20 20 20 descriptor 4: 00 00 00 fc 00 42 65 6e 51 20 47 32 34 30 30 57 0a 20 extensions: 01 checksum: cf Manufacturer: BNQ Model 780a Serial Number 21573 Made week 47 of 2007 EDID version: 1.3 Digital display Maximum image size: 52 cm x 32 cm Gamma: 2.20 DPMS levels: Off Supported color formats: RGB 4:4:4, YCrCb 4:2:2 Default (sRGB) color space is primary color space First detailed timing is preferred timing Established timings supported: 720x400@70Hz 640x480@60Hz 640x480@75Hz 800x600@60Hz 800x600@75Hz 832x624@75Hz 1024x768@60Hz 1024x768@75Hz 1280x1024@75Hz 1152x870@75Hz Standard timings supported: 1152x720@60Hz 1280x800@60Hz 1440x900@60Hz 1280x1024@60Hz 1680x1050@60Hz 1600x1200@60Hz 1920x1200@60Hz Detailed mode: Clock 154.000 MHz, 519 mm x 324 mm 1920 1968 2000 2080 hborder 0 1200 1203 1209 1235 vborder 0 +hsync -vsync Serial number: LB700253ZL0 Monitor ranges: 50-85HZ vertical, 31-94kHz horizontal, max dotclock 170MHz Monitor name: BenQ G2400W Has 1 extension blocks Checksum: 0xcf A broken kernel read slots and lots of zeroes. gfx hardware is ASUS P5E-V HDMI, with G35 gfx.
Created attachment 24581 [details] please try the patch on your machine, thanks. based on commit 167f3a04d7366d65c7fa9a92f0d604cdcf4a11ae use the patch, then upload your dmesg after runing xorg. Thanks Ma Ling
Created attachment 24587 [details] dmesg with patch on top of 1055f9ddad093f54dfd708a0f976582034d4ce1a
Created attachment 24588 [details] edid from working kernel
For what it's worth: the motherboard has two connectors. VGA and HDMI. The provided edid is from the HDMIconnector. VGA is currently unconnected.
(In reply to comment #4) > For what it's worth: the motherboard has two connectors. VGA and HDMI. The > provided edid is from the HDMIconnector. VGA is currently unconnected. From dmesg I see the problem look like from i2c driver, could you please try graphics in UMS, then upload log file with debug option on under the comment #2 environment. Thanks Ma Ling
Created attachment 24631 [details] Xorg.0.log from UMS
Created attachment 24632 [details] The xorg.conf I currently use
some noise in dmesg. possibly not related. May have come when I restarted xdm? ------------[ cut here ]------------ WARNING: at drivers/gpu/drm/i915/i915_gem.c:3698 i915_gem_idle+0x17a/0x28c() Hardware name: P5E-V HDMI Modules linked in: msp3400 tuner tea5767 tda8290 tuner_xc2028 xc5000 tda9887 tuner_simple tuner_types mt20xx tea5761 bttv ir_common v4l2_common videodev v4l1_compat videobuf_dma_sg videobuf_core btcx_risc tveeprom snd_bt87x Pid: 5368, comm: X Not tainted 2.6.29-10238-g1055f9d-dirty #2 Call Trace: [<c012974d>] warn_slowpath+0x71/0xa8 [<c03178cc>] ? i915_add_request+0x25/0x180 [<c0116505>] ? default_spin_lock_flags+0x8/0xb [<c047fb85>] ? _spin_lock_irqsave+0x20/0x28 [<c0131159>] ? lock_timer_base+0x1f/0x3e [<c0131438>] ? del_timer+0x4e/0x55 [<c0136d38>] ? __cancel_work_timer+0x124/0x162 [<c0317612>] ? i915_gem_retire_requests+0x113/0x12d [<c0317612>] ? i915_gem_retire_requests+0x113/0x12d [<c0318a5f>] i915_gem_idle+0x17a/0x28c [<c0318bc4>] i915_gem_leavevt_ioctl+0x1f/0x2e [<c0306a1e>] drm_ioctl+0x1f6/0x27c [<c0318ba5>] ? i915_gem_leavevt_ioctl+0x0/0x2e [<c047eb80>] ? mutex_lock+0xe/0x28 [<c02fe77c>] ? agpioc_release_wrap+0x57/0x5b [<c02ff2c1>] ? agp_ioctl+0x3ad/0x3d0 [<c013ba17>] ? hrtimer_forward+0xf4/0x10a [<c0140163>] ? getnstimeofday+0x51/0xda [<c0110603>] ? lapic_next_event+0x16/0x1a [<c0142726>] ? clockevents_program_event+0xdb/0xea [<c0180cb1>] vfs_ioctl+0x50/0x69 [<c0181107>] do_vfs_ioctl+0x43d/0x476 [<c0143670>] ? tick_program_event+0x22/0x29 [<c013c6ee>] ? hrtimer_interrupt+0x134/0x144 [<c018116c>] sys_ioctl+0x2c/0x45 [<c01029d8>] sysenter_do_call+0x12/0x2c ---[ end trace 289b5fdc453c22ff ]---
platform specific issue, lower down priority.
Would you be so kind as to define 'platform' in this context? SDVO-HDMI? G35? ASUS P5E-V?
actully only sdvo-hdmi device will fail to get edid, because sdvo i2c need to additional bus operation to get edid extension.
Created attachment 24657 [details] please try the debug patch on your machine, thanks.
The patch appears to work perfectly, thank you. I now have functional KMS again, with your patch on top of 1055f9ddad093f54dfd708a0f976582034d4ce1a. Dag B
I don't think this patch is upstream yet?
I will test this on my Asus P5E-VM HDMI (intel G35) as soon there is a koji build available (Fedora Rawhide).
Tested kernel-2.6.29.1-104.fc11 from koji, and updated some xorg bits from Fedora Rawhide and was able to start the X server - thanks! However, the X server froze when I tried to play a video. I'll investigate further and file a new bug (or comment if it already exist).
and for anyone reading this at a later time: A revision of the patch that actually works for 2.6.30-c4 is located here: http://lists.freedesktop.org/archives/intel-gfx/2009-April/002084.html And Eric Anholt suggest that the entire patch is redone here: http://lists.freedesktop.org/archives/intel-gfx/2009-April/002214.html Thanks, Dag B
Hi Dag Bakke The issue has been fixed by commit 619ac3b75a1e9b2df66857f6a0fb466f1da5fa9e Author: Ma Ling <ling.ma@intel.com> Date: Mon May 18 16:12:46 2009 +0800 close it now :) Thanks for your help Ma Ling
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.