Bug 80549 - [IVB,HSW bisected]Resuming form s3 cause Call Trace, with warm boot
Summary: [IVB,HSW bisected]Resuming form s3 cause Call Trace, with warm boot
Status: CLOSED FIXED
Alias: None
Product: DRI
Classification: Unclassified
Component: DRM/Intel (show other bugs)
Version: unspecified
Hardware: Other Linux (All)
: high normal
Assignee: Damien Lespiau
QA Contact: Intel GFX Bugs mailing list
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2014-06-26 05:40 UTC by liulei
Modified: 2017-10-06 14:37 UTC (History)
4 users (show)

See Also:
i915 platform:
i915 features:


Attachments
dmesg (123.75 KB, text/plain)
2014-06-26 05:40 UTC, liulei
no flags Details
Patch to make it WARN-free (3.56 KB, patch)
2014-07-07 22:03 UTC, Paulo Zanoni
no flags Details | Splinter Review
dmesg_with_new_Call_Trace (36.79 KB, text/plain)
2014-07-08 08:21 UTC, liulei
no flags Details

Description liulei 2014-06-26 05:40:41 UTC
Created attachment 101784 [details]
dmesg

Only on latest -testing, I find this Call Trace, I don't find this issue on both -next-queue and -fixes which merged in -testing. 
==kernel==
--------------------------
-testing ac710a93740e609759fa75dacdc96f1dfc34b5c5

origin/drm-intel-next-queued 9c33baa6b3bbb01c1a88dceba986b20e6642cf31(works)
    drivers/i915: Fix unnoticed failure of init_ring_common()
origin/drm-intel-fixes 4a11e398b39b018c7aa3b22267cfdf3b7fae3625 (works)
drm/i915: cache hw power well enabled state

This round testing merged in above two commit.
==Bug detailed description==
-----------------------------
s3 and check dmesg

[  143.337859] call 0000:00:16.0+ returned 0 after 0 usecs
[  143.337970] calling  0000:00:02.0+ @ 4098, parent: pci0000:00
[  143.338101] sd 0:0:0:0: [sda] Stopping disk
[  143.338592] ------------[ cut here ]------------
[  143.338613] WARNING: CPU: 3 PID: 4098 at drivers/gpu/drm/i915/i915_irq.c:154 ironlake_disable_display_irq+0x36/0x77 [i915]()
[  143.338639] Modules linked in: ip6table_filter ip6_tables iptable_filter ip_tables ebtable_nat ebtables x_tables bnep bluetooth rfkill ib_iser rdma_cm iw_cm ib_cm ib
_sa ib_mad ib_core ib_addr joydev nls_utf8 iTCO_wdt iTCO_vendor_support hfsplus snd_hda_codec_hdmi uvcvideo snd_hda_codec_cirrus videobuf2_vmalloc snd_hda_codec_generic
videobuf2_memops applesmc videobuf2_core input_polldev videodev snd_hda_intel snd_hda_controller snd_hda_codec snd_hwdep snd_seq snd_seq_device pcspkr i2c_i801 snd_pcm
lpc_ich snd_timer mfd_core snd soundcore sbs sbshc battery ac acpi_cpufreq kvm_intel kvm uinput i915 drm_kms_helper drm video button dm_mirror dm_region_hash dm_log dm
_mod ipv6
[  143.338642] CPU: 3 PID: 4098 Comm: kworker/u16:20 Not tainted 3.15.0-rc8_drm-intel-testing_ac710a_20140621_+ #55
[  143.338643] Hardware name: Apple Inc. MacBookPro10,2/Mac-AFD8A9D944EA4843, BIOS MBP102.88Z.0106.B03.1211161133 11/16/2012
[  143.338646] Workqueue: events_unbound async_run_entry_fn
[  143.338648]  0000000000000009 ffff88025e1ef9b8 ffffffff81820aea 0000000000000046
[  143.338650]  0000000000000000 ffff88025e1ef9f8 ffffffff8103bf44 0000000000000000
[  143.338651]  ffffffffa0142e41 ffff88025ccb0000 ffff88025ccb0000 ffff88025ccb7f20
[  143.338652] Call Trace:
[  143.338657]  [<ffffffff81820aea>] dump_stack+0x46/0x58
[  143.338660]  [<ffffffff8103bf44>] warn_slowpath_common+0x81/0x9b
[  143.338675]  [<ffffffffa0142e41>] ? ironlake_disable_display_irq+0x36/0x77 [i915]
[  143.338677]  [<ffffffff8103bf78>] warn_slowpath_null+0x1a/0x1c
[  143.338689]  [<ffffffffa0142e41>] ironlake_disable_display_irq+0x36/0x77 [i915]
[  143.338690] call 0000:00:1f.3+ returned 0 after 1353 usecs
[  143.338703]  [<ffffffffa0142ece>] ironlake_disable_vblank+0x4c/0x60 [i915]
[  143.338715]  [<ffffffffa00b6b97>] vblank_disable_and_save+0x43/0x163 [drm]
[  143.338716] calling  0000:00:01.1+ @ 4095, parent: pci0000:00
[  143.338724]  [<ffffffffa00b73db>] drm_vblank_off+0x3e/0x13a [drm]
[  143.338741]  [<ffffffffa01660a7>] intel_crtc_disable_planes+0xeb/0xfa [i915]
[  143.338742] call 0000:00:01.1+ returned 0 after 9 usecs
[  143.338759]  [<ffffffffa01664db>] ironlake_crtc_disable+0x39/0x83d [i915]
[  143.338760] calling  0000:00:01.0+ @ 4095, parent: pci0000:00
[  143.338772]  [<ffffffffa00c9dc6>] ? drm_modeset_lock_all_crtcs+0x32/0x51 [drm]
[  143.338772] call 0000:00:01.0+ returned 0 after 7 usecs
[  143.338781]  [<ffffffffa01192a2>] i915_drm_freeze+0xdf/0x172 [i915]
[  143.338783] calling  0000:00:00.0+ @ 4095, parent: pci0000:00
[  143.338793]  [<ffffffffa01193dc>] i915_pm_suspend+0x4a/0x4f [i915]
[  143.338794] call 0000:00:00.0+ returned 0 after 0 usecs
[  143.338798]  [<ffffffff813a3433>] pci_pm_suspend+0x87/0x109
[  143.338801]  [<ffffffff813a33ac>] ? pci_pm_freeze+0xba/0xba
[  143.338804]  [<ffffffff8143d7cd>] dpm_run_callback+0x39/0x75
[  143.338807]  [<ffffffff8143e541>] __device_suspend+0x1e8/0x269
[  143.338810]  [<ffffffff8143e5e1>] async_suspend+0x1f/0x8a
[  143.338813]  [<ffffffff8105df5b>] async_run_entry_fn+0x61/0x10e
[  143.338816]  [<ffffffff81053222>] process_one_work+0x25a/0x442
[  143.338819]  [<ffffffff8105318f>] ? process_one_work+0x1c7/0x442
[  143.338822]  [<ffffffff81053bac>] worker_thread+0x1d0/0x2cb
[  143.338824]  [<ffffffff810539dc>] ? manage_workers.isra.20+0x1c8/0x1c8
[  143.338826]  [<ffffffff81059228>] kthread+0xed/0xf5
[  143.338828]  [<ffffffff8105913b>] ? __init_kthread_worker+0x5a/0x5a
[  143.338830]  [<ffffffff8182f0ec>] ret_from_fork+0x7c/0xb0
[  143.338832]  [<ffffffff8105913b>] ? __init_kthread_worker+0x5a/0x5a
[  143.338833] ---[ end trace 396cae77903a4ab2 ]---

==Reproduce steps==
---------------------------- 
1. s3 and check dmesg
Comment 1 Paulo Zanoni 2014-07-07 21:39:56 UTC
In order to get rid of all the WARNs on HSW S3, I had to revert the following patches:

85e90679335f56d162f4a0ff525573818e17ce44
drm/i915: disable power wells on suspend

8abdc17941c71b37311bb93876ac83dce58160c8
drm/i915: make sure PC8 is enabled on suspend and disabled on resume v4

e11aa362308f5de467ce355a2a2471321b15a35c
drm/i915: use runtime irq suspend/resume in freeze/thaw
Comment 2 Paulo Zanoni 2014-07-07 22:03:52 UTC
Created attachment 102393 [details] [review]
Patch to make it WARN-free

Hi

This is probably not going to be the patch that is going to be merged, but at least it solves the problems I could reproduce, so if you can confirm it also solves the problem of your machine, then I can know we are on the same page :)

Thanks,
Paulo
Comment 3 liulei 2014-07-08 08:19:15 UTC
(In reply to comment #2)
> Created attachment 102393 [details] [review] [review]
> Patch to make it WARN-free
> 
> Hi
> 
> This is probably not going to be the patch that is going to be merged, but
> at least it solves the problems I could reproduce, so if you can confirm it
> also solves the problem of your machine, then I can know we are on the same
> page :)
> 
> Thanks,
> Paulo
Hi,

Your patch fixes this issue, but, I got a ton of new Call Trace like below:
[   24.668170] sd 7:0:0:0: [sda] Stopping disk
[   24.691978] ------------[ cut here ]------------
[   24.691990] WARNING: CPU: 3 PID: 4123 at drivers/gpu/drm/i915/intel_pm.c:4882 intel_suspend_gt_powersave+0x1f/0x38 [i915]()
[   24.692126] ------------[ cut here ]------------
[   24.692132] WARNING: CPU: 3 PID: 4123 at drivers/gpu/drm/i915/i915_irq.c:423 intel_set_pch_fifo_underrun_reporting+0xde/0x145 [i915]()
[   24.713805] ------------[ cut here ]------------
[   24.713814] WARNING: CPU: 2 PID: 4123 at drivers/gpu/drm/i915/i915_irq.c:423 intel_set_pch_fifo_underrun_reporting+0x12b/0x145 [i915]()
[   25.969939] WARNING: CPU: 2 PID: 6 at drivers/gpu/drm/i915/intel_pm.c:4894 intel_disable_gt_powersave+0x27/0x3e7 [i915]()
[   26.043297] ------------[ cut here ]------------
[   26.043304] WARNING: CPU: 1 PID: 6 at drivers/gpu/drm/i915/i915_irq.c:139 intel_set_cpu_fifo_underrun_reporting+0x249/0x264 [i915]()
[   26.043388] WARNING: CPU: 1 PID: 6 at drivers/gpu/drm/i915/i915_irq.c:423 intel_set_pch_fifo_underrun_reporting+0x12b/0x145 [i915]()
[   26.051902] WARNING: CPU: 1 PID: 6 at drivers/gpu/drm/i915/i915_irq.c:423 ibx_hpd_irq_setup+0xa0/0xd7 [i915]()
[   27.758916] WARNING: CPU: 1 PID: 3793 at drivers/gpu/drm/i915/i915_irq.c:209 gen6_enable_rps_interrupts+0x39/0x5c [i915]()

My testing based on last tesing (ac710a93740e609759fa75dacdc96f1dfc34b5c5), and patch your attachment on it. Was I using right branch to test? If I misunderstood what you said, Please correct me. I attach new Call Trace
Comment 4 liulei 2014-07-08 08:21:15 UTC
Created attachment 102418 [details]
dmesg_with_new_Call_Trace
Comment 5 Paulo Zanoni 2014-07-17 20:24:08 UTC
Can you please retest this on HSW with the current drm-intel-nightly tree?

I believe commit daa390e5ee45cc051d6bf37b296901f2f92b002d "drm/i915: don't warn if IRQs are disabled when shutting down display IRQs" fixed the last problem on HSW.

I am not sure about IVB.
Comment 6 liulei 2014-07-18 03:28:58 UTC
(In reply to comment #5)
> Can you please retest this on HSW with the current drm-intel-nightly tree?
> 
> I believe commit daa390e5ee45cc051d6bf37b296901f2f92b002d "drm/i915: don't
> warn if IRQs are disabled when shutting down display IRQs" fixed the last
> problem on HSW.
> 
> I am not sure about IVB.

We retested on latest drm-intel-nightly and we didn't find this Call Trace. But we found another one still there. I have no idea that if this one is related with the one we talked about.
[  620.126722] ------------[ cut here ]------------
[  620.126747] WARNING: CPU: 2 PID: 3692 at drivers/gpu/drm/i915/i915_irq.c:154 ironlake_disable_vblank+0x41/0x54 [i915]()
[  620.126763] Modules linked in: dm_mod iTCO_wdt iTCO_vendor_support snd_hda_codec_realtek snd_hda_codec_generic snd_hda_codec_hdmi pcspkr i2c_i801 lpc_ich mfd_core snd_hda_intel snd_hda_controller snd_hda_codec snd_hwdep snd_pcm uvcvideo videobuf2_vmalloc videobuf2_memops videobuf2_core v4l2_common videodev snd_timer snd soundcore wmi battery ac acpi_cpufreq i915 button video drm_kms_helper drm
[  620.126766] CPU: 2 PID: 3692 Comm: kworker/u16:61 Tainted: G        W     3.16.0-rc5_drm-intel-nightly_f326cd_20140717+ #574
[  620.126767] Hardware name: LENOVO 20267           /Cherry 3A       , BIOS 7CCN15WW 07/24/2013
[  620.126772] Workqueue: events_unbound async_run_entry_fn
[  620.126774]  0000000000000000 0000000000000009 ffffffff8178053d 0000000000000000
[  620.126776]  ffffffff8103986c ffff88015a3ec000 ffffffffa0096887 ffff88015a3ec000
[  620.126778]  0000000000000001 ffff880154cf0000 ffff880154cf7020 0000000000000002
[  620.126779] Call Trace:
[  620.126785]  [<ffffffff8178053d>] ? dump_stack+0x41/0x51
[  620.126789]  [<ffffffff8103986c>] ? warn_slowpath_common+0x78/0x90
[  620.126802]  [<ffffffffa0096887>] ? ironlake_disable_vblank+0x41/0x54 [i915]
[  620.126813]  [<ffffffffa0096887>] ? ironlake_disable_vblank+0x41/0x54 [i915]
[  620.126821]  [<ffffffffa0005548>] ? vblank_disable_and_save+0x3d/0x152 [drm]
[  620.126828]  [<ffffffffa0005d26>] ? drm_vblank_off+0x3a/0x147 [drm]
[  620.126843]  [<ffffffffa00ba9b5>] ? haswell_crtc_disable+0x3c/0x278 [i915]
[  620.126857]  [<ffffffffa00b456f>] ? intel_crtc_control+0x81/0xc4 [i915]
[  620.126864]  [<ffffffffa006e265>] ? i915_drm_freeze+0xd4/0x166 [i915]
[  620.126868]  [<ffffffff81350d12>] ? pci_pm_suspend+0x7b/0xfb
[  620.126873]  [<ffffffff813de98f>] ? __pm_runtime_barrier+0x10/0xfe
[  620.126875]  [<ffffffff81350c97>] ? pci_pm_freeze+0xb5/0xb5
[  620.126878]  [<ffffffff813e0260>] ? dpm_run_callback+0x57/0xcf
[  620.126881]  [<ffffffff813e105c>] ? __device_suspend+0x1d3/0x25f
[  620.126884]  [<ffffffff813e10fe>] ? async_suspend+0x16/0x7d
[  620.126887]  [<ffffffff810568af>] ? async_run_entry_fn+0x55/0x10b
[  620.126889]  [<ffffffff8104d636>] ? process_one_work+0x1e2/0x363
[  620.126891]  [<ffffffff8104de16>] ? worker_thread+0x2f2/0x3f1
[  620.126893]  [<ffffffff8104db24>] ? pool_mayday_timeout+0xf2/0xf2
[  620.126896]  [<ffffffff8105258e>] ? kthread+0xc5/0xcd
[  620.126899]  [<ffffffff810524c9>] ? kthread_freezable_should_stop+0x40/0x40
[  620.126902]  [<ffffffff8178602c>] ? ret_from_fork+0x7c/0xb0
[  620.126904]  [<ffffffff810524c9>] ? kthread_freezable_should_stop+0x40/0x40
[  620.126906] ---[ end trace 9e41ade7fe768372 ]---
Comment 7 Paulo Zanoni 2014-07-18 13:40:34 UTC
(In reply to comment #6)
> (In reply to comment #5)
> > Can you please retest this on HSW with the current drm-intel-nightly tree?
> > 
> > I believe commit daa390e5ee45cc051d6bf37b296901f2f92b002d "drm/i915: don't
> > warn if IRQs are disabled when shutting down display IRQs" fixed the last
> > problem on HSW.
> > 
> > I am not sure about IVB.
> 
> We retested on latest drm-intel-nightly and we didn't find this Call Trace.
> But we found another one still there. I have no idea that if this one is
> related with the one we talked about.
> [  620.126722] ------------[ cut here ]------------
> [  620.126747] WARNING: CPU: 2 PID: 3692 at
> drivers/gpu/drm/i915/i915_irq.c:154 ironlake_disable_vblank+0x41/0x54
> [i915]()
> [  620.126763] Modules linked in: dm_mod iTCO_wdt iTCO_vendor_support
> snd_hda_codec_realtek snd_hda_codec_generic snd_hda_codec_hdmi pcspkr
> i2c_i801 lpc_ich mfd_core snd_hda_intel snd_hda_controller snd_hda_codec
> snd_hwdep snd_pcm uvcvideo videobuf2_vmalloc videobuf2_memops videobuf2_core
> v4l2_common videodev snd_timer snd soundcore wmi battery ac acpi_cpufreq
> i915 button video drm_kms_helper drm
> [  620.126766] CPU: 2 PID: 3692 Comm: kworker/u16:61 Tainted: G        W    
> 3.16.0-rc5_drm-intel-nightly_f326cd_20140717+ #574
> [  620.126767] Hardware name: LENOVO 20267           /Cherry 3A       , BIOS
> 7CCN15WW 07/24/2013
> [  620.126772] Workqueue: events_unbound async_run_entry_fn
> [  620.126774]  0000000000000000 0000000000000009 ffffffff8178053d
> 0000000000000000
> [  620.126776]  ffffffff8103986c ffff88015a3ec000 ffffffffa0096887
> ffff88015a3ec000
> [  620.126778]  0000000000000001 ffff880154cf0000 ffff880154cf7020
> 0000000000000002
> [  620.126779] Call Trace:
> [  620.126785]  [<ffffffff8178053d>] ? dump_stack+0x41/0x51
> [  620.126789]  [<ffffffff8103986c>] ? warn_slowpath_common+0x78/0x90
> [  620.126802]  [<ffffffffa0096887>] ? ironlake_disable_vblank+0x41/0x54
> [i915]
> [  620.126813]  [<ffffffffa0096887>] ? ironlake_disable_vblank+0x41/0x54
> [i915]
> [  620.126821]  [<ffffffffa0005548>] ? vblank_disable_and_save+0x3d/0x152
> [drm]
> [  620.126828]  [<ffffffffa0005d26>] ? drm_vblank_off+0x3a/0x147 [drm]
> [  620.126843]  [<ffffffffa00ba9b5>] ? haswell_crtc_disable+0x3c/0x278 [i915]
> [  620.126857]  [<ffffffffa00b456f>] ? intel_crtc_control+0x81/0xc4 [i915]
> [  620.126864]  [<ffffffffa006e265>] ? i915_drm_freeze+0xd4/0x166 [i915]
> [  620.126868]  [<ffffffff81350d12>] ? pci_pm_suspend+0x7b/0xfb
> [  620.126873]  [<ffffffff813de98f>] ? __pm_runtime_barrier+0x10/0xfe
> [  620.126875]  [<ffffffff81350c97>] ? pci_pm_freeze+0xb5/0xb5
> [  620.126878]  [<ffffffff813e0260>] ? dpm_run_callback+0x57/0xcf
> [  620.126881]  [<ffffffff813e105c>] ? __device_suspend+0x1d3/0x25f
> [  620.126884]  [<ffffffff813e10fe>] ? async_suspend+0x16/0x7d
> [  620.126887]  [<ffffffff810568af>] ? async_run_entry_fn+0x55/0x10b
> [  620.126889]  [<ffffffff8104d636>] ? process_one_work+0x1e2/0x363
> [  620.126891]  [<ffffffff8104de16>] ? worker_thread+0x2f2/0x3f1
> [  620.126893]  [<ffffffff8104db24>] ? pool_mayday_timeout+0xf2/0xf2
> [  620.126896]  [<ffffffff8105258e>] ? kthread+0xc5/0xcd
> [  620.126899]  [<ffffffff810524c9>] ?
> kthread_freezable_should_stop+0x40/0x40
> [  620.126902]  [<ffffffff8178602c>] ? ret_from_fork+0x7c/0xb0
> [  620.126904]  [<ffffffff810524c9>] ?
> kthread_freezable_should_stop+0x40/0x40
> [  620.126906] ---[ end trace 9e41ade7fe768372 ]---

This WARN should not exist on -nightly anymore. Are you sure the tree you tested contains commit daa390e5ee45cc051d6bf37b296901f2f92b002d "drm/i915: don't warn if IRQs are disabled when shutting down display IRQs"?
Comment 8 liulei 2014-07-21 02:22:24 UTC
> > [  620.126722] ------------[ cut here ]------------
> > [  620.126747] WARNING: CPU: 2 PID: 3692 at
> > drivers/gpu/drm/i915/i915_irq.c:154 ironlake_disable_vblank+0x41/0x54
> > [i915]()

> This WARN should not exist on -nightly anymore. Are you sure the tree you
> tested contains commit daa390e5ee45cc051d6bf37b296901f2f92b002d "drm/i915:
> don't warn if IRQs are disabled when shutting down display IRQs"?

Double checked . This WARN has gone.
Comment 9 Paulo Zanoni 2014-07-21 15:30:59 UTC
(In reply to comment #8)
> > > [  620.126722] ------------[ cut here ]------------
> > > [  620.126747] WARNING: CPU: 2 PID: 3692 at
> > > drivers/gpu/drm/i915/i915_irq.c:154 ironlake_disable_vblank+0x41/0x54
> > > [i915]()
> 
> > This WARN should not exist on -nightly anymore. Are you sure the tree you
> > tested contains commit daa390e5ee45cc051d6bf37b296901f2f92b002d "drm/i915:
> > don't warn if IRQs are disabled when shutting down display IRQs"?
> 
> Double checked . This WARN has gone.

So as far as I understood, all WARNs are gone? Closing bug then. Please reopen if I am wrong.

Thanks for testing,
Paulo
Comment 10 Elizabeth 2017-10-06 14:37:32 UTC
Closing old verified.


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.