[Intel-gfx] hsw edp regression (was: Re: World's worst bug report :)
Jani Nikula
jani.nikula at intel.com
Tue Jun 3 14:24:17 CEST 2014
Kenneth, please file a bug report on DRM/Intel at [1], with the below
information, and attach dmesg from boot with drm.debug=0xe module
parameter set.
Please find a further question below inline.
[1] https://bugs.freedesktop.org/enter_bug.cgi?product=DRI
On Wed, 28 May 2014, "Kenneth R. Crudup" <kenny at panix.com> wrote:
> I tend to run the tip of Linus' master branch, and that's been working
> well. After merging in "master" about 3 weeks ago though, I'd often come
> back to a hard-hung laptop after being gone for 15-20 mins or so, i.e.,
> after the (blank-screen) screensaver had kicked in, but only when I'm
> at the office and I had a higher-res monitor (2560x1440) plugged in.
>
> I started going thru the i915 commit logs and after ruling out stuff
> that doesn't appear to apply to my laptop, I'd reverted:
>
> eeb6324d ("drm/i915: consider the source max DP lane count too")
> and
> 56071a20 ("drm/i915: use lane count and link rate from VBT as minimums for eDP")
>
> ... and when I'd come in this morning, my laptop was still up and came
> up from the screen saver (it stays on when I'm away from it and not
> suspended). Due to the sporadic nature of the hangs (not always
> reproducible) and time constraints on my part, I haven't narrowed down
> which commit seems to have broken things, but I hope this does help
> you guys to perhaps have an "A-ha!" moment.
>
> My laptop's a Toshiba Satellite P75-A7200 with a Haswell CPU and fairly
> standard i915 graphics:
>
> ----
> processor : 0
> vendor_id : GenuineIntel
> cpu family : 6
> model : 60
> model name : Intel(R) Core(TM) i7-4700MQ CPU @ 2.40GHz
> stepping : 3
> microcode : 0x16
> cpu MHz : 886.500
> cache size : 6144 KB
> physical id : 0
> siblings : 8
> core id : 0
> cpu cores : 4
> apicid : 0
> initial apicid : 0
> fpu : yes
> fpu_exception : yes
> cpuid level : 13
> wp : yes
> ----
> 00:02.0 0300: 8086:0416 (rev 06) (prog-if 00 [VGA controller])
> Subsystem: 1179:fa72
> Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx+
> Status: Cap+ 66MHz- UDF- FastB2B+ ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
> Latency: 0
> Interrupt: pin A routed to IRQ 40
> Region 0: Memory at b2000000 (64-bit, non-prefetchable) [size=4M]
> Region 2: Memory at a0000000 (64-bit, prefetchable) [size=256M]
> Region 4: I/O ports at 5000 [size=64]
> Expansion ROM at <unassigned> [disabled]
> Capabilities: [90] MSI: Enable+ Count=1/1 Maskable- 64bit-
> Address: feeff00c Data: 4181
> Capabilities: [d0] Power Management version 2
> Flags: PMEClk- DSI+ D1- D2- AuxCurrent=0mA PME(D0-,D1-,D2-,D3hot-,D3cold-)
> Status: D0 NoSoftRst- PME-Enable- DSel=0 DScale=0 PME-
> Capabilities: [a4] PCI Advanced Features
> AFCap: TP+ FLR+
> AFCtrl: FLR-
> AFStatus: TP-
> Kernel driver in use: i915
> ----
> eDP1 connected primary 1920x1080+0+360 (normal left inverted right x axis y axis) 381mm x 214mm
> 1920x1080 60.5*+ 59.9 40.3
> ----
> HDMI1 connected 2560x1440+1920+0 (normal left inverted right x axis y axis) 597mm x 336mm
> ...
> 2560x1440_40 40.0*
> ----
>
> If you need any other information from me, please don't hesitate to ask.
>
> Thanks,
>
> -Kenny
>
> --
> Kenneth R. Crudup Sr. SW Engineer, Scott County Consulting, Los Angeles
On Thu, 29 May 2014, "Kenneth R. Crudup" <kenny at panix.com> wrote:
> On Wed, 28 May 2014, Kenneth R. Crudup wrote:
>
>> [I'd get] back to a hard-hung laptop after being gone for 15-20 mins
>> ... after the (blank-screen) screensaver had kicked in, but only when I'm
>> at the office and I had a higher-res monitor (2560x1440) plugged in.
> ...
>> 56071a20 ("drm/i915: use lane count and link rate from VBT as minimums for eDP")
>
> It's apparently this commit; reverting it (which also pulls f4cdbc214
> ("drm/i915/dp: force eDP lane count to max available lanes on BDW") as
> a side-effect, which doesn't matter on my non-Broadwell machine anyway)
> seems to have stopped the panic-after-screen-off issue for me.
You refer both to a hard hang and a panic, which one is it? What
happens? Do you get a backtrace? Does the machine stop responding to
ping? Can you ssh in and get dmesg?
BR,
Jani.
>
> HTH someone,
>
> -Kenny
>
> --
> Kenneth R. Crudup Sr. SW Engineer, Scott County Consulting, Los Angeles
--
Jani Nikula, Intel Open Source Technology Center
More information about the Intel-gfx
mailing list