Bug 93697 - [BAT SKL] sporadic WARNING backtrace: "DC6 already programmed to be enabled."
Summary: [BAT SKL] sporadic WARNING backtrace: "DC6 already programmed to be enabled."
Status: CLOSED DUPLICATE of bug 93768
Alias: None
Product: DRI
Classification: Unclassified
Component: DRM/Intel (show other bugs)
Version: XOrg git
Hardware: Other All
: highest normal
Assignee: Patrik Jakobsson
QA Contact: Intel GFX Bugs mailing list
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2016-01-13 13:04 UTC by Daniel Vetter
Modified: 2017-07-24 22:43 UTC (History)
3 users (show)

See Also:
i915 platform: SKL
i915 features:


Attachments

Description Daniel Vetter 2016-01-13 13:04:06 UTC
This seems to be a very sporadic failure which happens on bot skl-i5k-2 and skl-i7k-2 with the kms_flip@basic-flip-vs-dpms testcase. For long term results overview see

/archive/results/CI_IGT_test/igt@kms_flip@basic-flip-vs-dpms.html

The thestcase itself completes without hiccup, it's just the backtrace in the kernel:

 	

[  170.702268] ------------[ cut here ]------------
[  170.702285] WARNING: CPU: 0 PID: 5863 at drivers/gpu/drm/i915/intel_runtime_pm.c:578 skl_enable_dc6+0x16a/0x190 [i915]()
[  170.702287] DC6 already programmed to be enabled.
[  170.702288] Modules linked in: ax88179_178a usbnet mii x86_pkg_temp_thermal i915 intel_powerclamp coretemp crct10dif_pclmul crc32_pclmul mei_me mei i2c_hid e1000e ptp pps_core
[  170.702300] CPU: 0 PID: 5863 Comm: kms_flip Tainted: G     U          4.4.0-gfxbench+ #1
[  170.702302] Hardware name: System manufacturer System Product Name/Z170M-PLUS, BIOS 0505 11/16/2015
[  170.702303]  ffffffffa0210718 ffff88022ec5baa0 ffffffff813df90c ffff88022ec5bae8
[  170.702306]  ffff88022ec5bad8 ffffffff810746e1 ffff88022bda0000 ffff880230a08000
[  170.702309]  0000000000000002 ffff88022bda0000 0000000000000800 ffff88022ec5bb38
[  170.702312] Call Trace:
[  170.702316]  [<ffffffff813df90c>] dump_stack+0x4e/0x82
[  170.702318]  [<ffffffff810746e1>] warn_slowpath_common+0x81/0xc0
[  170.702320]  [<ffffffff81074767>] warn_slowpath_fmt+0x47/0x50
[  170.702329]  [<ffffffffa0140cba>] skl_enable_dc6+0x16a/0x190 [i915]
[  170.702338]  [<ffffffffa01411f6>] gen9_dc_off_power_well_disable+0x136/0x240 [i915]
[  170.702347]  [<ffffffffa013ee67>] intel_power_well_disable+0x27/0x50 [i915]
[  170.702356]  [<ffffffffa0141937>] intel_display_power_put+0xb7/0x130 [i915]
[  170.702370]  [<ffffffffa0191fae>] modeset_put_power_domains+0x3e/0x50 [i915]
[  170.702385]  [<ffffffffa01a3aac>] intel_atomic_commit+0x9ac/0x17c0 [i915]
[  170.702389]  [<ffffffff814ea061>] ? __drm_atomic_helper_crtc_duplicate_state+0x51/0x70
[  170.702391]  [<ffffffff8150e192>] drm_atomic_commit+0x32/0x50
[  170.702394]  [<ffffffff814e9949>] drm_atomic_helper_connector_dpms+0xe9/0x1a0
[  170.702396]  [<ffffffff81502f85>] drm_mode_obj_set_property_ioctl+0x235/0x240
[  170.702398]  [<ffffffff81502fbb>] drm_mode_connector_property_set_ioctl+0x2b/0x30
[  170.702400]  [<ffffffff814f3eed>] drm_ioctl+0x13d/0x590
[  170.702403]  [<ffffffff81502f90>] ? drm_mode_obj_set_property_ioctl+0x240/0x240
[  170.702405]  [<ffffffff811c1ff3>] ? __vfs_read+0x23/0xd0
[  170.702407]  [<ffffffff811d4c4c>] do_vfs_ioctl+0x2fc/0x550
[  170.702410]  [<ffffffff811e06ba>] ? __fget_light+0x6a/0x90
[  170.702412]  [<ffffffff811d4edc>] SyS_ioctl+0x3c/0x70
[  170.702414]  [<ffffffff8179a75b>] entry_SYSCALL_64_fastpath+0x16/0x73
[  170.702416] ---[ end trace d8793596364dbb8b ]---
Comment 1 Patrik Jakobsson 2016-01-18 12:18:42 UTC
It seems the DMC is setting the "Allow DC6" bit for us in DC_STATE_EN. It shouldn't so this looks like a firmware bug. We could remove the warning and just fix up the bits but lets see if I can figure out if this is reliably reproducable first. I haven't been able to see any bad effects of this bug aside from the warning.
Comment 2 Mika Kuoppala 2016-01-19 08:00:20 UTC
(In reply to Patrik Jakobsson from comment #1)
> It seems the DMC is setting the "Allow DC6" bit for us in DC_STATE_EN. It
> shouldn't so this looks like a firmware bug. We could remove the warning and
> just fix up the bits but lets see if I can figure out if this is reliably
> reproducable first. I haven't been able to see any bad effects of this bug
> aside from the warning.

Please also note:
https://bugs.freedesktop.org/show_bug.cgi?id=93768

The warning about DC6 being enabled might be also just a symptom of a prior
gpu death, but the problems starts immediately after.
Comment 3 cprigent 2016-01-26 17:08:13 UTC
Bug scrub
----------
Assigned to Patrick.
Is it a duplicate of https://bugs.freedesktop.org/show_bug.cgi?id=93768?
Comment 4 Patrik Jakobsson 2016-01-26 17:23:41 UTC

*** This bug has been marked as a duplicate of bug 93768 ***


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.