Created attachment 68692 [details] dmesg Hello. I notice a significant slow down when booting because of i915. I can boot in three scenarios: Either I put i915 into the initramfs, or I load it via systemd's module-load, or I don't load it explicitely (which only works sometimes). If I don't load it explicitely, Xorg sometimes fails to start with "(EE) no devices detected" which I guess is expected since the module isn't ready, yet. In both cases of loading the i915 module explicitely, I get a boot slow down of about 3-5 seconds. Attached are three systemd-analyze plots: * plot_simple.svg: No explicit loading, booting is very quick but X fails sometimes * plot_initramfs.svg: i915 built into the initramfs, a very clear slowdown occurs in the 'kernel' portion of the boot * plot_module-load.svg: i915 loaded by systemd, a similar slowdown can be seen during module-load. From dmesg with drm.debug=0x06, it is quite obvious what is causing the slowdown. Below's the relevant snippet from the attached file. drm takes about from seconds 1.8 until second 5.2 to load completely. It seems like it's turning the display on and off a bunch of times (I also see this happening while the laptop is booting) and it has too wait around a lot for things to happen. It's not super cool that it's taking this long, while the rest of the boot process is lightning quick, so if it were possible to fix this somehow, that would be great. Thank you! ----------------------------------------------- Chipset and Version information: * chipset: 7 Series / C210 Series (3rd Gen Core processor Graphics Controller) * system architecture: x86_64 * xf86-video-intel version: 2.20.9 * xserver version: 1.13.0 * mesa version: 9.0 * libdrm version: 2.4.39 * kernel version: 3.6.2 * Linux distribution: Archlinux * Machine or mobo model: ASUS Zenbook Prime UX32VD * Display connector: eDP1 (according to xrandr) Attached files: * Xorg.0.log * dmesg with 'panic=10 debug ignore_loglevel initcall_debug log_buf_len=10M drm.debug=0x06' * xrandr_verbose.txt * intel_reg_dumper.txt * vbios.dump Relevant snippet: --------------------------------------------------------- [ 1.834823] Console: switching to colour dummy device 80x25 [ 1.835005] i915 0000:00:02.0: setting latency timer to 64 [ 1.934571] [drm:intel_detect_pch], Found PatherPoint PCH [ 1.938977] [drm:__gen6_gt_force_wake_mt_get] *ERROR* Force wake wait timed out [ 1.938984] [drm:intel_gt_init], Using MT version of forcewake [ 1.939101] [drm:intel_opregion_setup], graphic opregion physical addr: 0xca882018 [ 1.939126] [drm:intel_opregion_setup], Public ACPI methods supported [ 1.939130] [drm:intel_opregion_setup], SWSCI supported [ 1.939134] [drm:intel_opregion_setup], ASLE supported [ 1.939168] i915 0000:00:02.0: irq 43 for MSI/MSI-X [ 1.939188] [drm] Supports vblank timestamp caching Rev 1 (10.10.2010). [ 1.939192] [drm] Driver supports precise vblank timestamp query. [ 1.939196] [drm:init_vbt_defaults], Set default to SSC at 100MHz [ 1.939201] [drm:intel_parse_bios], Using VBT from OpRegion: $VBT SNB/IVB-MOBILE d [ 1.939206] [drm:parse_general_features], BDB_GENERAL_FEATURES int_tv_support 0 int_crt_support 1 lvds_use_ssc 0 lvds_ssc_freq 120 display_clock_mode 0 [ 1.939214] [drm:parse_general_definitions], crt_ddc_bus_pin: 2 [ 1.939221] [drm:parse_lfp_panel_data], Found panel mode in BIOS VBT tables: [ 1.939226] [drm:drm_mode_debug_printmodeline], Modeline 0:"1920x1080" 0 138780 1920 1966 1996 2080 1080 1082 1086 1112 0x8 0xa [ 1.939234] [drm:parse_lfp_panel_data], VBT initial LVDS value 30033c [ 1.939239] [drm:parse_sdvo_panel_data], Found SDVO panel mode in BIOS VBT tables: [ 1.939244] [drm:drm_mode_debug_printmodeline], Modeline 0:"1600x1200" 0 162000 1600 1664 1856 2160 1200 1201 1204 1250 0x8 0xa [ 1.939252] [drm:parse_sdvo_device_mapping], No SDVO device info is found in VBT [ 1.939309] [drm:intel_dsm_pci_probe], failed to get supported _DSM functions [ 1.939326] [drm:intel_modeset_init], 3 display pipes available. [ 1.939346] vgaarb: device changed decodes: PCI:0000:00:02.0,olddecodes=io+mem,decodes=io+mem:owns=io+mem [ 1.939656] [drm:intel_lvds_init], LVDS is not present in VBT [ 1.939701] [drm:intel_dp_init], cur t1_t3 1700 t8 2000 t9 2000 t10 500 t11_t12 6000 [ 1.939707] [drm:intel_dp_init], vbt t1_t3 1700 t8 2000 t9 2000 t10 500 t11_t12 5000 [ 1.939712] [drm:intel_dp_init], panel power up delay 170, power down delay 50, power cycle delay 600 [ 1.939718] [drm:intel_dp_init], backlight on delay 200, off delay 200 [ 1.939723] [drm:intel_dp_i2c_init], i2c_init DPDDC-A [ 1.939727] [drm:ironlake_edp_panel_vdd_on], Turn eDP VDD on [ 1.939738] [drm:ironlake_edp_panel_vdd_on], PCH_PP_STATUS: 0x80000008 PCH_PP_CONTROL: 0xabcd000f [ 1.939948] [drm:intel_dp_i2c_aux_ch], aux_i2c nack [ 1.940158] [drm:intel_dp_i2c_aux_ch], aux_i2c nack [ 1.940193] [drm:ironlake_edp_panel_vdd_off], Turn eDP VDD off 1 [ 1.940199] [drm:ironlake_edp_panel_vdd_on], Turn eDP VDD on [ 1.940204] [drm:ironlake_edp_panel_vdd_on], eDP VDD already on [ 1.940511] [drm:ironlake_edp_panel_vdd_off], Turn eDP VDD off 1 [ 1.940515] [drm:ironlake_edp_panel_vdd_on], Turn eDP VDD on [ 1.940521] [drm:ironlake_edp_panel_vdd_on], eDP VDD already on [ 1.941542] [drm:i2c_algo_dp_aux_xfer], dp_aux_xfer return 2 [ 1.968370] [drm:i2c_algo_dp_aux_xfer], dp_aux_xfer return 2 [ 1.968381] [drm:drm_edid_to_eld], ELD: no CEA Extension found [ 1.968385] [drm:ironlake_edp_panel_vdd_off], Turn eDP VDD off 1 [ 1.968390] [drm:intel_panel_get_backlight], get backlight PWM = 4296 [ 1.968422] [drm:intel_panel_get_backlight], get backlight PWM = 4296 [ 1.968456] [drm:intel_crt_init], pch crt adpa set to 0xf40000 [ 1.968538] [drm:intel_dp_i2c_init], i2c_init DPDDC-C [ 1.971093] [drm:intel_dp_aux_ch], dp_aux_ch timeout status 0x5145003f [ 1.971098] [drm:intel_dp_i2c_aux_ch], aux_ch failed -110 [ 1.973652] [drm:intel_dp_aux_ch], dp_aux_ch timeout status 0x5145003f [ 1.973657] [drm:intel_dp_i2c_aux_ch], aux_ch failed -110 [ 1.973683] [drm:ironlake_edp_panel_vdd_on], Turn eDP VDD on [ 1.973689] [drm:ironlake_edp_panel_vdd_on], eDP VDD already on [ 1.973693] [drm:ironlake_edp_backlight_off], [ 2.175577] [drm:ironlake_edp_panel_off], Turn eDP power off [ 2.175591] [drm:ironlake_wait_panel_off], Wait for panel power off time [ 2.175598] [drm:ironlake_wait_panel_status], mask b000000f value 00000000 status 80000008 control abcd0000 [ 2.541488] tsc: Refined TSC clocksource calibration: 2394.560 MHz [ 2.541501] Switching to clocksource tsc [ 2.987480] [drm:intel_dp_link_down], [ 3.064244] [drm:ironlake_crtc_dpms], crtc 0/0 dpms off [ 3.064258] [drm:i915_get_vblank_timestamp], crtc 0 is disabled [ 3.117283] [drm:ironlake_wait_for_vblank], vblank wait timed out [ 3.223792] [drm:intel_wait_for_pipe_off], pipe_off wait timed out [ 3.224241] [drm:ironlake_crtc_dpms], crtc 1/1 dpms off [ 3.224249] [drm:gm45_get_vblank_counter], trying to get vblank count for disabled pipe B [ 3.224257] [drm:i915_get_vblank_timestamp], crtc 1 is disabled [ 3.224262] [drm:gm45_get_vblank_counter], trying to get vblank count for disabled pipe B [ 3.224701] [drm:ironlake_crtc_dpms], crtc 2/2 dpms off [ 3.224707] [drm:gm45_get_vblank_counter], trying to get vblank count for disabled pipe C [ 3.224714] [drm:i915_get_vblank_timestamp], crtc 2 is disabled [ 3.224719] [drm:gm45_get_vblank_counter], trying to get vblank count for disabled pipe C [ 3.225158] [drm:ironlake_init_pch_refclk], has_panel 1 has_lvds 0 has_pch_edp 0 has_cpu_edp 1 has_ck505 0 [ 3.265141] [drm:init_status_page], render ring hws offset: 0x00000000 [ 3.265472] [drm:init_status_page], bsd ring hws offset: 0x00022000 [ 3.265814] [drm:init_status_page], blitter ring hws offset: 0x00043000 [ 3.266166] [drm:create_default_context], Default HW context loaded [ 3.266170] [drm:i915_gem_context_init], HW context support initialized [ 3.266349] [drm:intel_enable_rc6], RC6 and deep RC6 enabled [ 3.266353] [drm] Enabling RC6 states: RC6 on, RC6p on, RC6pp off [ 3.383594] [drm:ironlake_edp_panel_vdd_on], Turn eDP VDD on [ 3.383607] [drm:ironlake_wait_panel_power_cycle], Wait for panel power cycle [ 3.383615] [drm:ironlake_wait_panel_status], mask b800000f value 00000000 status 00000000 control abcd0000 [ 3.383628] [drm:ironlake_edp_panel_vdd_on], PCH_PP_STATUS: 0x00000000 PCH_PP_CONTROL: 0xabcd0008 [ 3.383636] [drm:ironlake_edp_panel_vdd_on], eDP was not running [ 3.556629] [drm:ironlake_edp_backlight_off], [ 3.759877] [drm:ironlake_edp_panel_off], Turn eDP power off [ 3.759891] [drm:ironlake_wait_panel_off], Wait for panel power off time [ 3.759899] [drm:ironlake_wait_panel_status], mask b000000f value 00000000 status 00000000 control abcd0000 [ 3.760123] [drm:drm_helper_probe_single_connector_modes], [CONNECTOR:9:eDP-1] [ 3.760130] [drm:intel_dp_detect], DPCD: 110a020000000000 [ 3.760155] [drm:drm_edid_to_eld], ELD: no CEA Extension found [ 3.760164] [drm:drm_helper_probe_single_connector_modes], [CONNECTOR:9:eDP-1] probed modes : [ 3.760171] [drm:drm_mode_debug_printmodeline], Modeline 22:"1920x1080" 60 138780 1920 1966 1996 2080 1080 1082 1086 1112 0x48 0xa [ 3.760180] [drm:drm_mode_debug_printmodeline], Modeline 23:"1920x1080" 40 92520 1920 1966 1996 2080 1080 1082 1086 1112 0x40 0xa [ 3.760189] [drm:drm_helper_probe_single_connector_modes], [CONNECTOR:16:VGA-1] [ 3.760197] [drm:intel_ironlake_crt_detect_hotplug], trigger hotplug detect cycle: adpa=0xf40000 [ 3.772977] [drm:intel_ironlake_crt_detect_hotplug], ironlake hotplug adpa=0xf40000, result 0 [ 3.772990] [drm:intel_crt_detect], CRT not detected via hotplug [ 3.779633] [drm:gmbus_xfer], GMBUS [i915 gmbus vga] NAK for addr: 0050 r(1) [ 3.779648] [drm:drm_do_probe_ddc_edid], drm: skipping non-existent adapter i915 gmbus vga [ 3.779655] [drm:intel_crt_get_edid], CRT GMBUS EDID read failed, retry using GPIO bit-banging [ 3.780030] [drm:drm_do_probe_ddc_edid], drm: skipping non-existent adapter i915 gmbus vga [ 3.780035] [drm:intel_crt_detect_ddc], CRT not detected via DDC:0x50 [no valid EDID found] [ 3.780042] [drm:drm_helper_probe_single_connector_modes], [CONNECTOR:16:VGA-1] disconnected [ 3.780050] [drm:drm_helper_probe_single_connector_modes], [CONNECTOR:19:HDMI-A-1] [ 3.786290] [drm:gmbus_xfer], GMBUS [i915 gmbus dpc] NAK for addr: 0050 r(1) [ 3.786304] [drm:drm_do_probe_ddc_edid], drm: skipping non-existent adapter i915 gmbus dpc [ 3.786312] [drm:drm_helper_probe_single_connector_modes], [CONNECTOR:19:HDMI-A-1] disconnected [ 3.786320] [drm:drm_helper_probe_single_connector_modes], [CONNECTOR:20:DP-1] [ 3.788877] [drm:intel_dp_aux_ch], dp_aux_ch timeout status 0x5145003f [ 3.795493] [drm:intel_dp_aux_ch], dp_aux_ch timeout status 0x5145003f [ 3.802149] [drm:intel_dp_aux_ch], dp_aux_ch timeout status 0x5145003f [ 3.806256] [drm:intel_dp_detect], DPCD: 0000000000000000 [ 3.806269] [drm:drm_helper_probe_single_connector_modes], [CONNECTOR:20:DP-1] disconnected [ 3.806277] [drm:drm_setup_crtcs], [ 3.806282] [drm:drm_enable_connectors], connector 9 enabled? yes [ 3.806287] [drm:drm_enable_connectors], connector 16 enabled? no [ 3.806291] [drm:drm_enable_connectors], connector 19 enabled? no [ 3.806296] [drm:drm_enable_connectors], connector 20 enabled? no [ 3.806300] [drm:drm_target_preferred], looking for cmdline mode on connector 9 [ 3.806306] [drm:drm_target_preferred], looking for preferred mode on connector 9 [ 3.806311] [drm:drm_target_preferred], found mode 1920x1080 [ 3.806315] [drm:drm_setup_crtcs], picking CRTCs for 8192x8192 config [ 3.806322] [drm:drm_setup_crtcs], desired mode 1920x1080 set on crtc 3 [ 3.815848] [drm:intelfb_create], allocated 1920x1080 fb: 0x00073000, bo ffff88029fcb2c00 [ 3.815999] fbcon: inteldrmfb (fb0) is primary device [ 3.816046] [drm:drm_crtc_helper_set_config], [ 3.816048] [drm:drm_crtc_helper_set_config], [CRTC:3] [FB:13] #connectors=1 (x y) (0 0) [ 3.816053] [drm:drm_crtc_helper_set_config], crtc has no fb, full mode set [ 3.816054] [drm:drm_crtc_helper_set_config], modes are different, full mode set [ 3.816057] [drm:drm_mode_debug_printmodeline], Modeline 0:"" 0 0 0 0 0 0 0 0 0 0 0x0 0x0 [ 3.816060] [drm:drm_mode_debug_printmodeline], Modeline 12:"1920x1080" 60 138780 1920 1966 1996 2080 1080 1082 1086 1112 0x48 0xa [ 3.816061] [drm:drm_crtc_helper_set_config], encoder changed, full mode switch [ 3.816062] [drm:drm_crtc_helper_set_config], crtc changed, full mode switch [ 3.816063] [drm:drm_crtc_helper_set_config], [CONNECTOR:9:eDP-1] to [CRTC:3] [ 3.816065] [drm:drm_crtc_helper_set_config], attempting to set mode from userspace [ 3.816067] [drm:drm_mode_debug_printmodeline], Modeline 12:"1920x1080" 60 138780 1920 1966 1996 2080 1080 1082 1086 1112 0x48 0xa [ 3.816071] [drm:intel_dp_mode_fixup], DP link computation with max lane count 2 max bw 0a pixel clock 138780KHz [ 3.816072] [drm:intel_dp_mode_fixup], DP link bw 0a lane count 2 clock 270000 bpp 24 [ 3.816073] [drm:intel_dp_mode_fixup], DP link bw required 333072 available 432000 [ 3.816075] [drm:drm_crtc_helper_set_mode], [CRTC:3] [ 3.816077] [drm:ironlake_edp_panel_vdd_on], Turn eDP VDD on [ 3.816081] [drm:ironlake_wait_panel_power_cycle], Wait for panel power cycle [ 3.816085] [drm:ironlake_wait_panel_status], mask b800000f value 00000000 status 00000000 control abcd0000 [ 3.816092] [drm:ironlake_edp_panel_vdd_on], PCH_PP_STATUS: 0x00000000 PCH_PP_CONTROL: 0xabcd0008 [ 3.816094] [drm:ironlake_edp_panel_vdd_on], eDP was not running [ 3.985999] [drm:ironlake_edp_backlight_off], [ 4.189237] [drm:ironlake_edp_panel_off], Turn eDP power off [ 4.189241] [drm:ironlake_wait_panel_off], Wait for panel power off time [ 4.189247] [drm:ironlake_wait_panel_status], mask b000000f value 00000000 status 00000000 control abcd0000 [ 4.189264] [drm:intel_choose_pipe_bpp_dither], clamping display bpc (was -1) to EDID reported max of 8 [ 4.189266] [drm:intel_choose_pipe_bpp_dither], setting pipe bpc to 8 (max display bpc 8) [ 4.189267] [drm:ironlake_crtc_mode_set], Mode for pipe 0: [ 4.189271] [drm:drm_mode_debug_printmodeline], Modeline 12:"1920x1080" 60 138780 1920 1966 1996 2080 1080 1082 1086 1112 0x48 0xa [ 4.189277] [drm:ironlake_set_pll_edp], eDP PLL enable for clock 270000 [ 4.242275] [drm:ironlake_wait_for_vblank], vblank wait timed out [ 4.242284] [drm:ironlake_update_plane], Writing base 00073000 00000000 0 0 7680 [ 4.242288] [drm:sandybridge_update_wm], FIFO watermarks For pipe A - plane 13, cursor: 6 [ 4.242291] [drm:ironlake_check_srwm], watermark 1: display plane 20, fbc lines 3, cursor 6 [ 4.242292] [drm:ironlake_check_srwm], watermark 2: display plane 72, fbc lines 3, cursor 6 [ 4.242294] [drm:ironlake_check_srwm], watermark 3: display plane 141, fbc lines 4, cursor 10 [ 4.242298] [drm:drm_crtc_helper_set_mode], [ENCODER:10:TMDS-10] set [MODE:12:1920x1080] [ 4.242299] [drm:ironlake_edp_pll_on], [ 4.242502] [drm:sandybridge_update_wm], FIFO watermarks For pipe A - plane 13, cursor: 6 [ 4.242503] [drm:ironlake_check_srwm], watermark 1: display plane 20, fbc lines 3, cursor 6 [ 4.242505] [drm:ironlake_check_srwm], watermark 2: display plane 72, fbc lines 3, cursor 6 [ 4.242507] [drm:ironlake_check_srwm], watermark 3: display plane 141, fbc lines 4, cursor 10 [ 4.295528] [drm:ironlake_wait_for_vblank], vblank wait timed out [ 4.348782] [drm:ironlake_wait_for_vblank], vblank wait timed out [ 4.348788] [drm:ironlake_edp_panel_vdd_on], Turn eDP VDD on [ 4.348792] [drm:ironlake_wait_panel_power_cycle], Wait for panel power cycle [ 4.348797] [drm:ironlake_wait_panel_status], mask b800000f value 00000000 status 00000000 control abcd0000 [ 4.348803] [drm:ironlake_edp_panel_vdd_on], PCH_PP_STATUS: 0x00000000 PCH_PP_CONTROL: 0xabcd0008 [ 4.348806] [drm:ironlake_edp_panel_vdd_on], eDP was not running [ 4.526583] [drm:intel_dp_start_link_train], too many voltage retries, give up [ 4.526584] [drm:ironlake_edp_panel_on], Turn eDP power on [ 4.526587] [drm:ironlake_wait_panel_power_cycle], Wait for panel power cycle [ 4.526592] [drm:ironlake_wait_panel_status], mask b800000f value 00000000 status 00000000 control abcd0008 [ 4.526597] [drm:ironlake_wait_panel_on], Wait for panel power on [ 4.526601] [drm:ironlake_wait_panel_status], mask b000000f value 80000008 status 9000000a control abcd000b [ 4.897969] [drm:ironlake_edp_panel_vdd_off], Turn eDP VDD off 1 [ 4.897977] [drm:ironlake_panel_vdd_off_sync], PCH_PP_STATUS: 0x80000008 PCH_PP_CONTROL: 0xabcd0003 [ 4.952436] [drm:ironlake_edp_backlight_on], [ 4.959364] [drm:intel_prepare_page_flip], preparing flip with no unpin work? [ 5.154767] [drm:drm_crtc_helper_set_config], Setting connector DPMS state to on [ 5.154770] [drm:drm_crtc_helper_set_config], [CONNECTOR:9:eDP-1] set DPMS on [ 5.154785] [drm:drm_crtc_helper_set_config], [ 5.154786] [drm:drm_crtc_helper_set_config], [CRTC:5] [NOFB] [ 5.154796] [drm:drm_crtc_helper_set_config], [ 5.154797] [drm:drm_crtc_helper_set_config], [CRTC:7] [NOFB] [ 5.154814] Console: switching to colour frame buffer device 240x67 [ 5.154825] [drm:drm_crtc_helper_set_config], [ 5.154826] [drm:drm_crtc_helper_set_config], [CRTC:3] [FB:13] #connectors=1 (x y) (0 0) [ 5.154832] [drm:drm_crtc_helper_set_config], [CONNECTOR:9:eDP-1] to [CRTC:3] [ 5.205523] fb0: inteldrmfb frame buffer device [ 5.207218] drm: registered panic notifier [ 5.210033] [drm:intel_panel_get_max_backlight], max backlight PWM = 4296 [ 5.210909] acpi device:06: registered as cooling_device0 [ 5.210943] ACPI: Video Device [PEGP] (multi-head: yes rom: yes post: no) [ 5.211011] input: Video Bus as /devices/LNXSYSTM:00/device:00/PNP0A08:00/device:03/LNXVIDEO:00/input/input4 [ 5.217234] [drm:intel_panel_actually_set_backlight], set backlight PWM = 4296 [ 5.218958] [drm:intel_opregion_gse_intr], non asle set request?? [ 5.222209] [drm:intel_opregion_gse_intr], PWM freq is not supported [ 5.223712] ACPI Error: Current brightness invalid (20120711/video-376) [ 5.226328] [drm:intel_opregion_gse_intr], non asle set request?? [ 5.228137] ACPI: Video Device [GFX0] (multi-head: yes rom: no post: no) [ 5.229881] input: Video Bus as /devices/LNXSYSTM:00/device:00/PNP0A08:00/LNXVIDEO:01/input/input5 [ 5.231722] [drm] Initialized i915 1.6.0 20080730 for 0000:00:02.0 on minor 0 ---------------------------------------------------------
Created attachment 68693 [details] inte_reg_dumper
Created attachment 68694 [details] plot initramfs
Created attachment 68695 [details] plot module-load
Created attachment 68696 [details] plot simple
Created attachment 68697 [details] vbios.dump
Created attachment 68698 [details] Xorg.0.log
Created attachment 68699 [details] xrandr
Created attachment 68708 [details] fixed mime type Sorry, I must have made a mistake when uploading the plots. I'm trying to correct it.
Created attachment 68709 [details] plot_simple.svg
Created attachment 68710 [details] plot_module-load.svg
Created attachment 68711 [details] plot_initramfs.svg
The delays seem to be mostly the result of touching the eDP panel, which unfortunately cannot be reduced and only avoided if the BIOS puts the panel into a native mode. (The latter also requires the fastboot work to avoid changing the BIOS mode.)
Created attachment 68799 [details] grub2 videoinfo output (efi gop driver) Well that's unfortunate. As far as I was able to discover, the EFI GOP driver on this laptop only provies 1024x768, even though the native mode is 1920x1080. I guess I'm out of luck (unless Asus is going to fix this). See the attached grub2 screenshot. At least I wasn't able to get grub2 to use a higher mode... I'm still wondering why there are so many duplicate entries in dmesg, such as: ... [ 1.973683] [drm:ironlake_edp_panel_vdd_on], Turn eDP VDD on [ 1.973689] [drm:ironlake_edp_panel_vdd_on], eDP VDD already on [ 1.973693] [drm:ironlake_edp_backlight_off], [ 2.175577] [drm:ironlake_edp_panel_off], Turn eDP power off ... [ 3.383628] [drm:ironlake_edp_panel_vdd_on], PCH_PP_STATUS: 0x00000000 PCH_PP_CONTROL: 0xabcd0008 [ 3.383636] [drm:ironlake_edp_panel_vdd_on], eDP was not running [ 3.556629] [drm:ironlake_edp_backlight_off], [ 3.759877] [drm:ironlake_edp_panel_off], Turn eDP power off ... [ 3.816094] [drm:ironlake_edp_panel_vdd_on], eDP was not running [ 3.985999] [drm:ironlake_edp_backlight_off], [ 4.189237] [drm:ironlake_edp_panel_off], Turn eDP power off ... [ 4.348806] [drm:ironlake_edp_panel_vdd_on], eDP was not running [ 4.526583] [drm:intel_dp_start_link_train], too many voltage retries, give up [ 4.526584] [drm:ironlake_edp_panel_on], Turn eDP power on [ 4.526587] [drm:ironlake_wait_panel_power_cycle], Wait for panel power cycle ... Especially the "too many voltage retries, give up" message seems interesting, after all that switching about. Can this process not be shortened? If not, I believe you and I guess the ticket can be closed as invalid...
I'm working on improving our dp link training sequence, but yeah, for the moment on non-cooperative displays, this seemingly forever retrying is what's expected. I guess you can keep this report around for a while longer ...
Hm, this bug was idling for a long time. And the dp link training improvements turned out to be wishful thinking and had to be curbed again. Can you please retest on latest bits to check whether your issue still exists?
Created attachment 84237 [details] 2013 systemd-analyze plot
Created attachment 84238 [details] 2013 dmesg
The issue still exists. No noticeable change since I first reported it. The screen is still power cycling a bunch of times and dmesg says "too many voltage retries, give up" at some point. This is with xf86-video-intel 2.21.14-1 and linux 3.10.5.
Please retest with latest drm-intel-fixes, specifically commit 2e6efddd203c15ca5c4700511f717c0e9a3ea31a Author: Imre Deak <imre.deak@intel.com> Date: Fri Aug 23 23:50:23 2013 +0300 drm/i915: ivb: fix edp voltage swing reg val
Created attachment 84639 [details] 2013 dmesg with drm-intel-fixes So I compiled the kernel from http://cgit.freedesktop.org/~danvet/drm-intel (drm-intel-fixes branch), using this config https://projects.archlinux.org/svntogit/packages.git/plain/trunk/config.x86_64?h=packages/linux but it doesn't seem to make much of a difference. The boot time is about the same and the messages are similar. "Too many voltage retries" is still present as well. Attached is the new dmesg. I hope I did everything correctly, it has been a while since I compiled my own kernel...
Looking at the logs, the DP clock recovery seems to fail, but luckily it works at channel eq phase... we've got something fishy there. Also seen in https://bugzilla.kernel.org/show_bug.cgi?id=59841 but there channel eq fails too with some bpp values. I think we should get the link training fixed first, although AFAICT the slowness isn't caused so much by the link training, but rather by the panel power sequencing delays.
Possibly related to bug 70117, although that is ivb only.
We've fixed a number of link training and power sequencing issues, please retest current drm-intel-nightly branch from http://cgit.freedesktop.org/drm-intel.
Hm, looks like the reporter unfortunately disappeared. If this is still an issue then please retest on latest kernels and reopen.
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.