Bug 15015
| Summary: | blank screen at random times in laptop when sitting idle | ||
|---|---|---|---|
| Product: | Drivers | Reporter: | Jithin Emmanuel (jithin1987) |
| Component: | Video(DRI - Intel) | Assignee: | Jesse Barnes (jbarnes) |
| Status: | CLOSED CODE_FIX | ||
| Severity: | blocking | CC: | combuster, jbarnes, jcnengel, rjw, yakui.zhao |
| Priority: | P1 | ||
| Hardware: | All | ||
| OS: | Linux | ||
| Kernel Version: | 2.6.32-4 | Subsystem: | |
| Regression: | Yes | Bisected commit-id: | |
| Bug Depends on: | |||
| Bug Blocks: | 14230 | ||
| Attachments: | return early and timeout in FBC disable path | ||
|
Description
Jithin Emmanuel
2010-01-09 16:48:23 UTC
Ther problem was not present in 2.6.31 kernel. With kerner 2.6.32-4 the problem still exists. But its a different stack trace. Jan 21 05:52:44 grainlead-lx kernel: X invoked oom-killer: gfp_mask=0x0, order=0, oom_adj=0 Jan 21 05:52:44 grainlead-lx kernel: Pid: 5294, comm: X Not tainted 2.6.32-ARCH #1 Jan 21 05:52:44 grainlead-lx kernel: Call Trace: Jan 21 05:52:44 grainlead-lx kernel: [<c10b1529>] ? T.484+0x99/0x260 Jan 21 05:52:44 grainlead-lx kernel: [<c10b17cc>] ? __out_of_memory+0xdc/0x140 Jan 21 05:52:44 grainlead-lx kernel: [<c10b1920>] ? pagefault_out_of_memory+0x40/0x80 Jan 21 05:52:44 grainlead-lx kernel: [<c1025cd9>] ? do_page_fault+0x329/0x330 Jan 21 05:52:44 grainlead-lx kernel: [<c10f1f76>] ? sys_ioctl+0x76/0x90 Jan 21 05:52:44 grainlead-lx kernel: [<c10259b0>] ? do_page_fault+0x0/0x330 Jan 21 05:52:44 grainlead-lx kernel: [<c12b9c83>] ? error_code+0x73/0x78 Jan 21 05:52:44 grainlead-lx kernel: Mem-Info: Jan 21 05:52:44 grainlead-lx kernel: DMA per-cpu: Jan 21 05:52:44 grainlead-lx kernel: CPU 0: hi: 0, btch: 1 usd: 0 Jan 21 05:52:44 grainlead-lx kernel: CPU 1: hi: 0, btch: 1 usd: 0 Jan 21 05:52:44 grainlead-lx kernel: Normal per-cpu: Jan 21 05:52:44 grainlead-lx kernel: CPU 0: hi: 186, btch: 31 usd: 123 Jan 21 05:52:44 grainlead-lx kernel: CPU 1: hi: 186, btch: 31 usd: 180 Jan 21 05:52:44 grainlead-lx kernel: HighMem per-cpu: Jan 21 05:52:44 grainlead-lx kernel: CPU 0: hi: 186, btch: 31 usd: 154 Jan 21 05:52:44 grainlead-lx kernel: CPU 1: hi: 186, btch: 31 usd: 163 Jan 21 05:52:44 grainlead-lx kernel: active_anon:185662 inactive_anon:99664 isolated_anon:0 Jan 21 05:52:44 grainlead-lx kernel: active_file:98302 inactive_file:94841 isolated_file:0 Jan 21 05:52:44 grainlead-lx kernel: unevictable:17 dirty:28 writeback:180 unstable:0 Jan 21 05:52:44 grainlead-lx kernel: free:12205 slab_reclaimable:6905 slab_unreclaimable:10409 Jan 21 05:52:44 grainlead-lx kernel: mapped:25916 shmem:55936 pagetables:2027 bounce:0 Jan 21 05:52:44 grainlead-lx kernel: DMA free:8080kB min:64kB low:80kB high:96kB active_anon:420kB inactive_anon:2616kB active_file:1112kB inactive_file:3712kB unevictable:0kB isolated(anon):0kB isolated(file):0kB present:15852kB mlocked:0kB dirty:0kB writeback:0kB mapped:432kB shmem:264kB slab_reclaimable:0kB slab_unreclaimable:8kB kernel_stack:0kB pagetables:0kB unstable:0kB bounce:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? no Jan 21 05:52:44 grainlead-lx kernel: lowmem_reserve[]: 0 865 2007 2007 Jan 21 05:52:44 grainlead-lx kernel: Normal free:40244kB min:3728kB low:4660kB high:5592kB active_anon:196080kB inactive_anon:211296kB active_file:165532kB inactive_file:163676kB unevictable:0kB isolated(anon):0kB isolated(file):0kB present:885944kB mlocked:0kB dirty:48kB writeback:100kB mapped:32580kB shmem:109300kB slab_reclaimable:27620kB slab_unreclaimable:41628kB kernel_stack:3000kB pagetables:8108kB unstable:0kB bounce:0kB writeback_tmp:0kB pages_scanned:228 all_unreclaimable? no Jan 21 05:52:44 grainlead-lx kernel: lowmem_reserve[]: 0 0 9141 9141 Jan 21 05:52:44 grainlead-lx kernel: HighMem free:496kB min:512kB low:1740kB high:2972kB active_anon:546148kB inactive_anon:184744kB active_file:226564kB inactive_file:211976kB unevictable:68kB isolated(anon):0kB isolated(file):0kB present:1170120kB mlocked:68kB dirty:64kB writeback:620kB mapped:70652kB shmem:114180kB slab_reclaimable:0kB slab_unreclaimable:0kB kernel_stack:0kB pagetables:0kB unstable:0kB bounce:0kB writeback_tmp:0kB pages_scanned:549 all_unreclaimable? no Jan 21 05:52:44 grainlead-lx kernel: lowmem_reserve[]: 0 0 0 0 Jan 21 05:52:44 grainlead-lx kernel: DMA: 32*4kB 54*8kB 36*16kB 25*32kB 8*64kB 12*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 1*4096kB = 8080kB Jan 21 05:52:44 grainlead-lx kernel: Normal: 1531*4kB 835*8kB 1439*16kB 10*32kB 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 1*4096kB = 40244kB Jan 21 05:52:44 grainlead-lx kernel: HighMem: 108*4kB 8*8kB 0*16kB 0*32kB 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 496kB Jan 21 05:52:44 grainlead-lx kernel: 250124 total pagecache pages Jan 21 05:52:44 grainlead-lx kernel: 1017 pages in swap cache Jan 21 05:52:44 grainlead-lx kernel: Swap cache stats: add 22441, delete 21424, find 6601/6735 Jan 21 05:52:44 grainlead-lx kernel: Free swap = 1955588kB Jan 21 05:52:44 grainlead-lx kernel: Total swap = 2040244kB Jan 21 05:52:44 grainlead-lx kernel: 522160 pages RAM Jan 21 05:52:44 grainlead-lx kernel: 294834 pages HighMem Jan 21 05:52:44 grainlead-lx kernel: 5659 pages reserved Jan 21 05:52:44 grainlead-lx kernel: 274825 pages shared Jan 21 05:52:44 grainlead-lx kernel: 360471 pages non-shared Jan 21 05:52:44 grainlead-lx kernel: Out of memory: kill process 5506 (kdeinit4) score 20971 or a child Jan 21 05:52:44 grainlead-lx kernel: Killed process 5509 (klauncher) Jan 21 06:42:59 grainlead-lx kernel: SysRq : Keyboard mode set to system default Jan 21 06:42:59 grainlead-lx kernel: SysRq : Emergency Sync Jan 21 06:42:59 grainlead-lx kernel: Emergency Sync complete Jan 21 06:43:00 grainlead-lx kernel: SysRq : Terminate All Tasks If you have noticed the time, I closed the laptop lid at around Jan 20 12:30 PM and freeze occurred at Jan 21 05:52 AM. Is this maybe related to #14933? Yakui, looks like our DPMS on or off path is spinning forever somewhere? Any ideas? On Monday 25 January 2010, Pekka Enberg wrote:
> On Mon, Jan 25, 2010 at 12:23 AM, Rafael J. Wysocki <rjw@sisk.pl> wrote:
> > This message has been generated automatically as a part of a report
> > of regressions introduced between 2.6.31 and 2.6.32.
> >
> > The following bug entry is on the current list of known regressions
> > introduced between 2.6.31 and 2.6.32. Please verify if it still should
> > be listed and let me know (either way).
> >
> >
> > Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=15015
> > Subject : blank screen at random times in laptop when sitting idle
> > Submitter : Jithin Emmanuel <jithin1987@gmail.com>
> > Date : 2010-01-09 16:48 (16 days old)
>
> FWIW, I might have seen this on my Macbook as well (screen went blank
> but I didn't check the logs). I think I managed to unblank the screen
> by hitting the "increase brightness" key. But as you say, the bug
> doesn't trigger very often and I have no idea how to reproduce it at
> will.
On Monday 25 January 2010, Manish Katiyar wrote:
> On Mon, Jan 25, 2010 at 7:40 PM, Pekka Enberg <penberg@cs.helsinki.fi> wrote:
> > On Mon, Jan 25, 2010 at 12:23 AM, Rafael J. Wysocki <rjw@sisk.pl> wrote:
> >> This message has been generated automatically as a part of a report
> >> of regressions introduced between 2.6.31 and 2.6.32.
> >>
> >> The following bug entry is on the current list of known regressions
> >> introduced between 2.6.31 and 2.6.32. Please verify if it still should
> >> be listed and let me know (either way).
> >>
> >>
> >> Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=15015
> >> Subject : blank screen at random times in laptop when sitting idle
> >> Submitter : Jithin Emmanuel <jithin1987@gmail.com>
> >> Date : 2010-01-09 16:48 (16 days old)
> >
> > FWIW, I might have seen this on my Macbook as well (screen went blank
> > but I didn't check the logs). I think I managed to unblank the screen
> > by hitting the "increase brightness" key. But as you say, the bug
> > doesn't trigger very often and I have no idea how to reproduce it at
> > will.
>
> I have seen it too on my lenovo T60 with 2.6.32.2. It happens to me
> frequently and has become famous as "coffee bug" in my colleagues.
> Every morning when I goto office and after putting my laptop on
> docking station go for coffee. Once I come back the laptop screen goes
> blank, but haven't been able to manage how to get screen back. Only
> option is hard reboot. This happens almost every once in two three
> days.
>
> Let me know what information/logs I can provide to help debug this.
> BTW I normally suspend my laptop at home and then resume it in office
> (if that matters).
I am using 2.6.32.5 kernel at the moment. So far I have not seen this issue with this kernel version. But considering the random nature I would like to wait a day or 2 to confirm. I spoke too early. I got my first blank screen in days and with kernel 2.6.32-5 The laptop was just in a locked state lid was not even closed. Log is same as that of one in description Jan 26 07:14:27 grainlead-lx kernel: [drm] LVDS-8: set mode 1280x800 1f Jan 26 08:53:34 grainlead-lx kernel: BUG: soft lockup - CPU#0 stuck for 61s! [X:5377] Jan 26 08:53:34 grainlead-lx kernel: Modules linked in: aes_i586 aes_generic ipv6 ext2 ide_cs ide_core arc4 snd_hda_codec_analog vboxdrv ecb snd_seq_dummy iwl3945 snd_seq_oss snd_seq_midi_event snd_seq snd_seq_device parport_pc snd_hda_intel snd_pcm_oss irtty_sir snd_mixer_oss fan snd_hda_codec snd_hwdep firewire_ohci tpm_infineon sir_dev iwlagn joydev sdhci_pci sdhci snd_pcm snd_timer firewire_core tpm tpm_bios yenta_socket iwlcore mmc_core rsrc_nonstatic container crc_itu_t snd soundcore snd_page_alloc ppdev ricoh_mmc hp_accel lis3lv02d input_polldev irda mac80211 led_class crc_ccitt thermal battery sg wmi ac cfg80211 btusb iTCO_wdt iTCO_vendor_support evdev bluetooth psmouse serio_raw rfkill lp parport e1000e cpufreq_powersave cpufreq_ondemand acpi_cpufreq freq_table processor rtc_cmos rtc_core rtc_lib ext4 mbcache jbd2 crc16 i915 drm_kms_helper drm i2c_algo_bit button i2c_core video output intel_agp agpgart uhci_hcd sr_mod cdrom ehci_hcd usbcore sd_mod ata_piix ahci pata_pcmcia libata scsi_mod pcmcia pcmcia_core Jan 26 08:53:34 grainlead-lx kernel: Jan 26 08:53:34 grainlead-lx kernel: Pid: 5377, comm: X Not tainted (2.6.32-ARCH #1) HP Compaq 6910p (GX978UC#ACJ) Jan 26 08:53:34 grainlead-lx kernel: EIP: 0060:[<f841d121>] EFLAGS: 00203286 CPU: 0 Jan 26 08:53:34 grainlead-lx kernel: EIP is at i8xx_disable_fbc+0x231/0x270 [i915] Jan 26 08:53:34 grainlead-lx kernel: EAX: f68c7c00 EBX: f8680000 ECX: f68ec000 EDX: a0000042 Jan 26 08:53:34 grainlead-lx kernel: ESI: 00071008 EDI: f68ec000 EBP: 00071180 ESP: f6553d84 Jan 26 08:53:34 grainlead-lx kernel: DS: 007b ES: 007b FS: 00d8 GS: 00e0 SS: 0068 Jan 26 08:53:34 grainlead-lx kernel: CR0: 80050033 CR2: b33f5000 CR3: 36900000 CR4: 000006d0 Jan 26 08:53:34 grainlead-lx kernel: DR0: 00000000 DR1: 00000000 DR2: 00000000 DR3: 00000000 Jan 26 08:53:34 grainlead-lx kernel: DR6: ffff0ff0 DR7: 00000400 Jan 26 08:53:34 grainlead-lx kernel: Call Trace: Jan 26 08:53:34 grainlead-lx kernel: [<f841d914>] ? i9xx_crtc_dpms+0x3a4/0x410 [i915] Jan 26 08:53:34 grainlead-lx kernel: [<f841d9ab>] ? intel_crtc_dpms+0x2b/0x100 [i915] Jan 26 08:53:34 grainlead-lx kernel: [<f8423657>] ? intel_lvds_set_power+0xd7/0xf0 [i915] Jan 26 08:53:34 grainlead-lx kernel: [<f8131fd4>] ? drm_helper_connector_dpms+0x144/0x260 [drm_kms_helper] Jan 26 08:53:34 grainlead-lx kernel: [<f8423670>] ? intel_lvds_dpms+0x0/0x20 [i915] Jan 26 08:53:34 grainlead-lx kernel: [<f8197d19>] ? drm_mode_object_find+0x49/0x70 [drm] Jan 26 08:53:34 grainlead-lx kernel: [<f81980cb>] ? drm_mode_connector_property_set_ioctl+0x1eb/0x1f0 [drm] Jan 26 08:53:34 grainlead-lx kernel: [<f819828f>] ? drm_mode_getproperty_ioctl+0x13f/0x2e0 [drm] Jan 26 08:53:34 grainlead-lx kernel: [<f818d298>] ? drm_ioctl+0x158/0x320 [drm] Jan 26 08:53:34 grainlead-lx kernel: [<f8197ee0>] ? drm_mode_connector_property_set_ioctl+0x0/0x1f0 [drm] Jan 26 08:53:34 grainlead-lx kernel: [<c105e117>] ? enqueue_hrtimer+0x67/0xd0 Jan 26 08:53:34 grainlead-lx kernel: [<c10f17e9>] ? vfs_ioctl+0x89/0xa0 Jan 26 08:53:34 grainlead-lx kernel: [<c10f1969>] ? do_vfs_ioctl+0x79/0x5c0 Jan 26 08:53:34 grainlead-lx kernel: [<c105f0c0>] ? hrtimer_start+0x20/0x30 Jan 26 08:53:34 grainlead-lx kernel: [<c1044692>] ? do_setitimer+0x162/0x1f0 Jan 26 08:53:34 grainlead-lx kernel: [<c10649a0>] ? ktime_get_ts+0xd0/0x100 Jan 26 08:53:34 grainlead-lx kernel: [<c10f1f26>] ? sys_ioctl+0x76/0x90 Jan 26 08:53:34 grainlead-lx kernel: [<c10039f3>] ? sysenter_do_call+0x12/0x28 Jan 26 08:54:39 grainlead-lx kernel: BUG: soft lockup - CPU#0 stuck for 61s! [X:5377] Jan 26 08:54:39 grainlead-lx kernel: Modules linked in: aes_i586 aes_generic ipv6 ext2 ide_cs ide_core arc4 snd_hda_codec_analog vboxdrv ecb snd_seq_dummy iwl3945 snd_seq_oss snd_seq_midi_event snd_seq snd_seq_device parport_pc snd_hda_intel snd_pcm_oss irtty_sir snd_mixer_oss fan snd_hda_codec snd_hwdep firewire_ohci tpm_infineon sir_dev iwlagn joydev sdhci_pci sdhci snd_pcm snd_timer firewire_core tpm tpm_bios yenta_socket iwlcore mmc_core rsrc_nonstatic container crc_itu_t snd soundcore snd_page_alloc ppdev ricoh_mmc hp_accel lis3lv02d input_polldev irda mac80211 led_class crc_ccitt thermal battery sg wmi ac cfg80211 btusb iTCO_wdt iTCO_vendor_support evdev bluetooth psmouse serio_raw rfkill lp parport e1000e cpufreq_powersave cpufreq_ondemand acpi_cpufreq freq_table processor rtc_cmos rtc_core rtc_lib ext4 mbcache jbd2 crc16 i915 drm_kms_helper drm i2c_algo_bit button i2c_core video output intel_agp agpgart uhci_hcd sr_mod cdrom ehci_hcd usbcore sd_mod ata_piix ahci pata_pcmcia libata scsi_mod pcmcia pcmcia_core Jan 26 08:54:39 grainlead-lx kernel: Jan 26 08:54:39 grainlead-lx kernel: Pid: 5377, comm: X Not tainted (2.6.32-ARCH #1) HP Compaq 6910p (GX978UC#ACJ) Jan 26 08:54:39 grainlead-lx kernel: EIP: 0060:[<f841d121>] EFLAGS: 00203286 CPU: 0 Jan 26 08:54:39 grainlead-lx kernel: EIP is at i8xx_disable_fbc+0x231/0x270 [i915] Jan 26 08:54:39 grainlead-lx kernel: EAX: f68c7c00 EBX: f8680000 ECX: f68ec000 EDX: a0000042 Jan 26 08:54:39 grainlead-lx kernel: ESI: 00071008 EDI: f68ec000 EBP: 00071180 ESP: f6553d84 Jan 26 08:54:39 grainlead-lx kernel: DS: 007b ES: 007b FS: 00d8 GS: 00e0 SS: 0068 Jan 26 08:54:39 grainlead-lx kernel: CR0: 80050033 CR2: b33f5000 CR3: 36900000 CR4: 000006d0 Jan 26 08:54:39 grainlead-lx kernel: DR0: 00000000 DR1: 00000000 DR2: 00000000 DR3: 00000000 Jan 26 08:54:39 grainlead-lx kernel: DR6: ffff0ff0 DR7: 00000400 Jan 26 08:54:39 grainlead-lx kernel: Call Trace: Jan 26 08:54:39 grainlead-lx kernel: [<f841d914>] ? i9xx_crtc_dpms+0x3a4/0x410 [i915] Jan 26 08:54:39 grainlead-lx kernel: [<f841d9ab>] ? intel_crtc_dpms+0x2b/0x100 [i915] Jan 26 08:54:39 grainlead-lx kernel: [<f8423657>] ? intel_lvds_set_power+0xd7/0xf0 [i915] Jan 26 08:54:39 grainlead-lx kernel: [<f8131fd4>] ? drm_helper_connector_dpms+0x144/0x260 [drm_kms_helper] Jan 26 08:54:39 grainlead-lx kernel: [<f8423670>] ? intel_lvds_dpms+0x0/0x20 [i915] I am now with 2.6.32-6 kernel and the problem is still there. I just cannot leave my laptop idle, it will freeze fore sure. I am also facing a system freeze during shutdown sometimes. Its totally broken atleast for me. That bug with softlockup going endlessly is fixed with 2.6.32.7 (at least for me), what is strange about it that there were no drm/i915 fixes from 2.6.32.6 and browsing to the changelog I couldn't find anything relevant to this kind of behaviour. However with 2.6.33-rc5 this bug is still present, at first I thought it was because of LVDS downclock feature but I'm not so sure anymore. I installed 2.6.32-7 today. Will monitor for a couple of days and report back if its fixed or not. I can confirm that this bug is not fixed with kernel 2.6.32-7. Just now I had a blank screen. The laptop lid was closed and I was away for about 1 hr. I had the same soft lockup logs in /var/log/kernel.log and had to do RSEIUB. Strange that I did not face this bug when I was using it at home and was easily reproducible at office. I use a docking station with my laptop at office. Could that trigger the issue. This is happening every where. NO connection to using docking station. It froze twice yesterday. My system was doing a massive update to kde 4.4 rc3 and some whee in middle it froze. It was 2nd time at home. Printing the same soft bug lock up bugs. 2.6.32 kernel is totally broke. IT also froze just 5 mins earlier when I was doing VT switch. I dont know when I did some 10-15 switches and it froze after that. This 2.6.32 is a great release. It freezes when, 1. sitting idle. 2. doing vt switch 3. suspend 4. restart/shutdown and all random. Great job!! and no one is even looking in to this one. Atleast the suspend one I reported is getting enough attention. Easy friend, I'm sure Jesse and the crew are already working on it. This is not so easy to catch, it appeared againg in both 2.6.32.7 and 2.6.33-rc5, and I can do nothing to reproduce the bug on my own will. It just occur randomly, sometimes more often (2-5 times a day), sometimes 2 days without a freeze. I've suspended, turned off lvds backlight, restarted, did all that numerous times and neither once I was able to reproduce it. Just when I give up it shows up by itself. And what is more strange I don't seem to find soft lockup BUG in kernel messages anymore when the freezing occur. This one seams a bit hard to investigate but if there is anything I would be 100% certain about I will report here. This happenes on a Dell Vostro 1310 laptop with GM965 chipset. Downgraded to 2.6.31.5 kernel. Everything seems to be working fine now. Can you try the drm-intel-next branch from the drm-intel repo (git://git.kernel.org/pub/scm/linux/kernel/git/anholt/drm-intel.git)? It's quite a bit different from 2.6.32, and if the problem is fixed there we'll know we have something to backport. Another option would be to bisect between 2.6.31 and 2.6.32 when the breakage started and figure out which commit broke things for you. Bisecting would take a lot of time because this bug occurs randomly, for 3 days I havent had a lockup and today happened again with the same error when I've tried to reboot:
Feb 5 14:59:01 vostro kernel: [24919.401005] BUG: soft lockup - CPU#0 stuck for 61s! [Xorg:2086]
I'm cloning the drm-intel-next and report here if the bug occurs again. I havent had any printk's because I didn't wait long enough for it to occur so last time I waited a minute or two before I've powered down my laptop and found it this time in /var/log/everything.log
I also got this in .xsession-error
(polkit-gnome-authentication-agent-1:2335): polkit-gnome-1-WARNING **: Error enumerating temporary authorizations: Remote Exception invoking org.freedesktop.PolicyKit1.Authority.EnumerateTemporaryAuthorizations() on /org/freedesktop/PolicyKit1/Authority at name org.freedesktop.PolicyKit1: org.freedesktop.PolicyKit1.Error.Failed: Cannot determine session the caller is in
gnome-settings-daemon: Fatal IO error 11 (Resource temporarily unavailable) on X server :0.0.
gtk-window-decorator: Fatal IO error 11 (Resource temporarily unavailable) on X server :0.0.
gnome-screensaver: Fatal IO error 11 (Resource temporarily unavailable) on X server :0.0.
gdu-notification-daemon: Fatal IO error 11 (Resource temporarily unavailable) on X server :0.0.
polkit-gnome-authentication-agent-1: Fatal IO error 11 (Resource temporarily unavailable) on X server :0.0.
bluetooth-applet: Fatal IO error 11 (Resource temporarily unavailable) on X server :0.0.
fusion-icon: Fatal IO error 11 (Resource temporarily unavailable) on X server :0.0.
XIO: fatal IO error 11 (Resource temporarily unavailable) on X server ":0.0"
after 2371012 requests (2371011 known processed) with 14 events remaining.
XIO: fatal IO error 11 (Resource temporarily unavailable) on X server ":0.0"
after 2808288 requests (2808281 known processed) with 0 events remaining.
* Detected Session: gnome
* Searching for installed applications...
* Intel detected, exporting: INTEL_BATCH=1
* Using the GTK Interface
* Starting Compiz
... executing: compiz --replace --sm-disable --ignore-desktop-hints ccp
It looks like X tried to respawn itself upon reboot and triggered the bug, CPU must be on 100% load because my fan spins like crazy but all I see is a black screen with cursor in the top left corner.
I really don't know if this two errors are related but I've rebooted to init3 upon restart because I didn't want to taint log files with new output.
This is definitly not related to compositing or DE because it happened on KDE with kwin and compositing disabled also.
I hope I'll have some good news in a few days with the drm-intel-next.
Bug is still present. Linux vostro 2.6.33-rc5-TESTING-50523-g0d0e2fe #1 SMP PREEMPT Sat Feb 6 01:45:04 CET 2010 x86_64 Intel(R) Core(TM)2 Duo CPU T5670 @ 1.80GHz GenuineIntel GNU/Linux It froze upon restart again, I've turned on almost every debugging option in kernel hacking section and still there was no output whatsoever. I've waited for 5min before powering off hoping it will provide some information. Good news is that it didn't froze while turning off lvds or waking up from suspend for three days period since I've tested it. There's a patch at the end of https://bugs.freedesktop.org/show_bug.cgi?id=24314, I'm curious if it helps with this issue. If not, it makes sense to add a timeout to our disable routine. I'll attach a patch for that. Created attachment 24958 [details]
return early and timeout in FBC disable path
Check FBC status when disabling, and add a timeout to the disable if we end up spinning.
I've applied both of these patches to the drm-intel-next, even though I can't trigger the bug described in the freedesktop bugtracker. Works fine for me in those situations, playing video after suspend doesn't hang the gpu and there are no desktop glitches. Although I have a different test system. 00:02.0 VGA compatible controller: Intel Corporation Mobile GM965/GL960 Integrated Graphics Controller (rev 0c) 00:02.1 Display controller: Intel Corporation Mobile GM965/GL960 Integrated Graphics Controller (rev 0c) I only have Projection and LOD, Bypass FIFO and Color calculator on 100% but I think that's normal. I'll be on a lookout for that bug upon restart. Can I apply these patches to the 2.6.33-rc7 as well and will they be included in the 2.6.33-rc8? Thanks Jesse. I'm hoping they'll be in -rc8; depends on whether Eric gets back in time to do a batch of patches before Linus decides to drop -rc8 on us. Oh and would it be possible for you to isolate which fix prevents the blank screen for you? I'm hoping the one from comment #21 is sufficient, since the one in #22 is more a workaround... That wouldn't be a problem, I'll patch 2.6.33-rc7 with the one from #21 and leave intel-drm-next with both of them. I can't be certain when I'm gonna have some results because in the last six days it only happened twice. I'll try my best to figure out how to trigger this bug for sure. Ok, with the 2.6.33-rc7 and the patch from comment #21 only I couldn't reproduce the bug for 3 days now, done everything to trigger it but no show. So it looks good. I've compiled 2.6.33-rc8 today (clean) and I'll see if it locks again. I saw the FBC_C3_IDLE patch entered the rc8 but only for 945GM so if it locks with rc8 then I think that FBC_C3_IDLE should be included for 965GM also. *** Bug 13453 has been marked as a duplicate of this bug. *** Sent out the 965 version as well. Ok, 2.6.33-rc8 clean with no patches, works like a charm. No problems at all. 5 days of testing with big uptime, suspending, restarting, playing games, video etc. So it works for me. off: I got 3D performance increase of 30% with rc8 compared to rc7. 22 to 33fps jump on urban terror. No mesa/xf86-video-intel upgrades in the meantime. Great work... Jesse's patch in Comment #22 works for me, too (solved the DPMS latency problem) --- in which mainline release is it included? James, which chipset do you have? (In reply to comment #32) > James, which chipset do you have? Intel X3100. Arg I thought 965 had working FBC, but maybe it's broken on your platform somehow. We disabled it on 945, which fixed this bug for most people, but it looks like the timeout is still needed for some. I'll push it now. Well I haven't had these problems since 2.6.33-rc8, 965 also... (In reply to comment #34) > Arg I thought 965 had working FBC, but maybe it's broken on your platform > somehow. We disabled it on 945, which fixed this bug for most people, but it > looks like the timeout is still needed for some. I'll push it now. Noted. For co-ordination purposes: I've been applying it to all the kernels I've built, up to the most recent (which is based on 2.6.33); things seem to work just great with it. (In reply to comment #34) > Arg I thought 965 had working FBC, but maybe it's broken on your platform > somehow. We disabled it on 945, which fixed this bug for most people, but it > looks like the timeout is still needed for some. I'll push it now. Jesse, in which kernel version was fbc-disable-timeout merged (for 965)? It doesn't seem to be in any of my distro's recent kernels, I still need to rebuild them with this patch... It's in git, so it should be in 2.6.35. If you want it in a stable kernel release for your distro to pick up you can send the commit id to stable@kernel.org and request that it be merged. Sorry to keep bringing this up again. I've been doing some testing of vanilla kernels 2.6.34.1 with and without the fbc-disable-timeout patches applied (had suspicious about the F13 sources). What I have found is that with or without the patch applied, the DPMS-off/screensaver fade latency issues remain (e.g., repeating/skipping audio), IF the kernel is compiled with voluntary pre-emption. On the other hand, if I compile it with full pre-emption, the spike in system CPU activity is still visible but the latency issues do not manifest so badly. [Since Fedora seem to have no intention of building kernels with full pre-emption, I doubt their 2.6.35 series will have this fixed.] So it appears as if there is something left here from Bug 13453; would it do to re-open this old bug? |