On 17 May 2011 12:27, Daniel J Blueman <daniel.blueman@xxxxxxxxx> wrote: > With 2.6.39-rc7 on my Sandy Bridge laptop GPU (8086:0126 rev 9), > sometimes I find one of the kworker threads busily running with 15-20% > system time for some minutes, causing terrible interactivity latency. > I've seen it occur when plugging eg a HDMI display, and also when no > display has been plugged (ie only the internal LVDS connection is > active). > > Across multiple kernel task captures, I see the kernel thread > consistently reading one of the connector's EDID data [1]; I guess > either it's having a hard time reading from a disconnected connector > and retrying, or is incorrectly detecting a change when there is none. > > I'll enable DRM debugging to see what connectors it believes it needs > to read from. Anything else that would be handy to capture, or any > thoughts? > > Also, the 100ms connector change polling seems overkill, particularly > when power consumption is important; 1000-2000ms would be sufficient, > do you think? > > Thanks, > Daniel > > --- [1] > > kworker/2:2 R running task 5048 86 2 0x00000000 > 0000000000000002 ffff88021e804040 ffff88021e85f9b0 ffff88021e804040 > ffff88021e85e000 0000000000004000 ffff8802210a4040 ffff88021e804040 > 0000000000000046 ffffffff81c18b20 ffff88022106c000 ffffffff8270b740 > Call Trace: > [<ffffffff8109a460>] ? mark_held_locks+0x70/0xa0 > [<ffffffff81059261>] ? get_parent_ip+0x11/0x50 > [<ffffffff8105933d>] ? sub_preempt_count+0x9d/0xd0 > [<ffffffff81705a35>] schedule_timeout+0x175/0x250 > [<ffffffff8106ec10>] ? run_timer_softirq+0x2a0/0x2a0 > [<ffffffff81705b29>] schedule_timeout_uninterruptible+0x19/0x20 > [<ffffffff8106f878>] msleep+0x18/0x20 > [<ffffffffa017c620>] gmbus_xfer+0x400/0x620 [i915] > [<ffffffff8150c892>] i2c_transfer+0xa2/0xf0 > [<ffffffffa002bc96>] drm_do_probe_ddc_edid+0x66/0xa0 [drm] > [<ffffffffa002c0f9>] drm_get_edid+0x29/0x60 [drm] > [<ffffffffa0176f86>] intel_hdmi_detect+0x56/0xe0 [i915] > [<ffffffffa00d1177>] output_poll_execute+0xd7/0x1a0 [drm_kms_helper] > [<ffffffff81078e14>] process_one_work+0x1a4/0x450 > [<ffffffff81078db6>] ? process_one_work+0x146/0x450 > [<ffffffffa00d10a0>] ? > drm_helper_disable_unused_functions+0x150/0x150 [drm_kms_helper] > [<ffffffff810790ec>] process_scheduled_works+0x2c/0x40 > [<ffffffff8107c384>] worker_thread+0x284/0x350 > [<ffffffff8107c100>] ? manage_workers.clone.23+0x120/0x120 > [<ffffffff81080ea6>] kthread+0xb6/0xc0 > [<ffffffff8109a5cd>] ? trace_hardirqs_on_caller+0x13d/0x180 > [<ffffffff8170a494>] kernel_thread_helper+0x4/0x10 > [<ffffffff8104c64f>] ? finish_task_switch+0x6f/0x100 > [<ffffffff81708bc4>] ? retint_restore_args+0xe/0xe > [<ffffffff81080df0>] ? __init_kthread_worker+0x70/0x70 > [<ffffffff8170a490>] ? gs_change+0xb/0xb Interestingly, I appear to observe this behaviour when the laptop battery is charging. Anyway, we see continual connector update events [1], booted drm.debug=0x4. This reproduces with 2.6.39. Any thoughts? Daniel --- [1] [ 67.035097] [drm:output_poll_execute], [CONNECTOR:22:HDMI-A-2] status updated from 2 to 2 [ 67.046549] [drm:output_poll_execute], [CONNECTOR:25:HDMI-A-3] status updated from 2 to 2 [ 67.047063] [drm:intel_dp_aux_ch], dp_aux_ch timeout status 0x5143003e [ 67.047065] [drm:ironlake_dp_detect], DPCD: 0000 [ 67.047066] [drm:output_poll_execute], [CONNECTOR:27:DP-2] status updated from 2 to 2 [ 67.047578] [drm:intel_dp_aux_ch], dp_aux_ch timeout status 0x5143003e [ 67.047579] [drm:ironlake_dp_detect], DPCD: 0000 [ 67.047581] [drm:output_poll_execute], [CONNECTOR:30:DP-3] status updated from 2 to 2 [ 67.047588] [drm:intel_ironlake_crt_detect_hotplug], ironlake hotplug adpa=0xf40000, result 0 [ 67.047591] [drm:intel_crt_detect], CRT not detected via hotplug [ 67.047593] [drm:output_poll_execute], [CONNECTOR:12:VGA-1] status updated from 2 to 2 [ 67.059062] [drm:output_poll_execute], [CONNECTOR:15:HDMI-A-1] status updated from 2 to 2 [ 67.059573] [drm:intel_dp_aux_ch], dp_aux_ch timeout status 0x5143003e [ 67.059575] [drm:ironlake_dp_detect], DPCD: 0000 [ 67.059576] [drm:output_poll_execute], [CONNECTOR:18:DP-1] status updated from 2 to 2 [ 67.059886] [drm:i915_hotplug_work_func], running encoder hotplug functions <85 lines the same> [ 67.153172] [drm:i915_hotplug_work_func], running encoder hotplug functions [ 67.155109] [drm:output_poll_execute], [CONNECTOR:22:HDMI-A-2] status updated from 2 to 2 [ 67.166569] [drm:output_poll_execute], [CONNECTOR:25:HDMI-A-3] status updated from 2 to 2 [ 67.167082] [drm:intel_dp_aux_ch], dp_aux_ch timeout status 0x5143003e [ 67.167084] [drm:ironlake_dp_detect], DPCD: 0000 [ 67.167086] [drm:output_poll_execute], [CONNECTOR:27:DP-2] status updated from 2 to 2 [ 67.167598] [drm:intel_dp_aux_ch], dp_aux_ch timeout status 0x5143003e [ 67.167600] [drm:ironlake_dp_detect], DPCD: 0000 [ 67.167601] [drm:output_poll_execute], [CONNECTOR:30:DP-3] status updated from 2 to 2 [ 67.167608] [drm:intel_ironlake_crt_detect_hotplug], ironlake hotplug adpa=0xf40000, result 0 [ 67.167610] [drm:intel_crt_detect], CRT not detected via hotplug [ 67.167612] [drm:output_poll_execute], [CONNECTOR:12:VGA-1] status updated from 2 to 2 [ 67.179051] [drm:output_poll_execute], [CONNECTOR:15:HDMI-A-1] status updated from 2 to 2 [ 67.179563] [drm:intel_dp_aux_ch], dp_aux_ch timeout status 0x5143003e [ 67.179564] [drm:ironlake_dp_detect], DPCD: 0000 [ 67.179566] [drm:output_poll_execute], [CONNECTOR:18:DP-1] status updated from 2 to 2 [ 67.179976] [drm:i915_hotplug_work_func], running encoder hotplug functions <73 lines the same> [ 67.275136] [drm:i915_hotplug_work_func], running encoder hotplug functions [ 67.275145] [drm:output_poll_execute], [CONNECTOR:22:HDMI-A-2] status updated from 2 to 2 [ 67.286602] [drm:output_poll_execute], [CONNECTOR:25:HDMI-A-3] status updated from 2 to 2 [ 67.287113] [drm:intel_dp_aux_ch], dp_aux_ch timeout status 0x5143003e [ 67.287115] [drm:ironlake_dp_detect], DPCD: 0000 [ 67.287117] [drm:output_poll_execute], [CONNECTOR:27:DP-2] status updated from 2 to 2 [ 67.287629] [drm:intel_dp_aux_ch], dp_aux_ch timeout status 0x5143003e [ 67.287631] [drm:ironlake_dp_detect], DPCD: 0000 [ 67.287633] [drm:output_poll_execute], [CONNECTOR:30:DP-3] status updated from 2 to 2 [ 67.287640] [drm:intel_ironlake_crt_detect_hotplug], ironlake hotplug adpa=0xf40000, result 0 [ 67.287643] [drm:intel_crt_detect], CRT not detected via hotplug [ 67.287645] [drm:output_poll_execute], [CONNECTOR:12:VGA-1] status updated from 2 to 2 [ 67.299144] [drm:output_poll_execute], [CONNECTOR:15:HDMI-A-1] status updated from 2 to 2 [ 67.299657] [drm:intel_dp_aux_ch], dp_aux_ch timeout status 0x5143003e [ 67.299659] [drm:ironlake_dp_detect], DPCD: 0000 [ 67.299660] [drm:output_poll_execute], [CONNECTOR:18:DP-1] status updated from 2 to 2 [ 67.300809] [drm:i915_hotplug_work_func], running encoder hotplug functions <78 lines the same> [ 67.395164] [drm:i915_hotplug_work_func], running encoder hotplug functions [ 67.395173] [drm:output_poll_execute], [CONNECTOR:22:HDMI-A-2] status updated from 2 to 2 [ 67.406644] [drm:output_poll_execute], [CONNECTOR:25:HDMI-A-3] status updated from 2 to 2 [ 67.407158] [drm:intel_dp_aux_ch], dp_aux_ch timeout status 0x5143003e [ 67.407160] [drm:ironlake_dp_detect], DPCD: 0000 [ 67.407162] [drm:output_poll_execute], [CONNECTOR:27:DP-2] status updated from 2 to 2 [ 67.407673] [drm:intel_dp_aux_ch], dp_aux_ch timeout status 0x5143003e [ 67.407675] [drm:ironlake_dp_detect], DPCD: 0000 [ 67.407676] [drm:output_poll_execute], [CONNECTOR:30:DP-3] status updated from 2 to 2 [ 67.407683] [drm:intel_ironlake_crt_detect_hotplug], ironlake hotplug adpa=0xf40000, result 0 [ 67.407686] [drm:intel_crt_detect], CRT not detected via hotplug [ 67.407688] [drm:output_poll_execute], [CONNECTOR:12:VGA-1] status updated from 2 to 2 [ 67.419136] [drm:output_poll_execute], [CONNECTOR:15:HDMI-A-1] status updated from 2 to 2 [ 67.419648] [drm:intel_dp_aux_ch], dp_aux_ch timeout status 0x5143003e [ 67.419650] [drm:ironlake_dp_detect], DPCD: 0000 [ 67.419651] [drm:output_poll_execute], [CONNECTOR:18:DP-1] status updated from 2 to 2 [ 67.420569] [drm:i915_hotplug_work_func], running encoder hotplug functions -- Daniel J Blueman _______________________________________________ dri-devel mailing list dri-devel@xxxxxxxxxxxxxxxxxxxxx http://lists.freedesktop.org/mailman/listinfo/dri-devel
