Re: [Intel-gfx] [git pull] drm tree for 4.2

[Jani Nikula <jani.nikula@linux.intel.com>]

On Mon, 29 Jun 2015, Daniel Vetter <daniel@ffwll.ch> wrote:
> On Mon, Jun 29, 2015 at 05:50:09PM +0300, Jani Nikula wrote:
>> On Mon, 29 Jun 2015, Ander Conselvan De Oliveira <conselvan2@gmail.com> wrote:
>> > On Fri, 2015-06-26 at 14:43 -0700, Linus Torvalds wrote:
>> >> On Thu, Jun 25, 2015 at 6:00 PM, Dave Airlie <airlied@linux.ie> wrote:
>> >> >
>> >> > This is the main drm pull request for v4.2.
>> >> 
>> >> It seems to work ok for me, but it causes quite a few new warnings on
>> >> my Sony VAIO Pro laptop. It's (once more) a regular i5-4200U CPU (aka
>> >> Haswell, aka 4th gen Intel Core i5)
>> >> 
>> >> Most of them are in check_crtc_state(), and I currently have 18 of
>> >> these in my log:
>> >> 
>> >>   [drm:check_crtc_state [i915]] *ERROR* mismatch in
>> >> dpll_hw_state.wrpll (expected 0x90280202, found 0x00000000)
>> >>   WARNING: CPU: 0 PID: 115 at
>> >> drivers/gpu/drm/i915/intel_display.c:12319
>> >> check_crtc_state+0x8be/0xf60 [i915]()
>> >>   pipe state doesn't match!
>> >> 
>> >> but there's a few others too:
>> >> 
>> >>   WARNING: CPU: 3 PID: 1871 at
>> >> drivers/gpu/drm/i915/intel_display.c:1362 hsw_disable_ips+0x34/0x160
>> >> [i915]()
>> >>   plane A assertion failure (expected on, current off)
>> >> 
>> >>   WARNING: CPU: 3 PID: 1871 at drivers/gpu/drm/drm_irq.c:1162
>> >> drm_wait_one_vblank+0x148/0x1a0 [drm]()
>> >>   vblank not available on crtc 0, ret=-22
>> >> 
>> >> and the backtraces aren't all that interesting, but I'm attaching the
>> >> cleaned-up dmesg, duplicate callchains and all.
>> >
>> > Please provide a full dmesg with drm.debug=0x1f in the kernel command
>> > line.
>> 
>> Ander, I think I was able to reproduce this on the BDW NUC here. Bisect
>> points at...
>> 
>> commit dd3cd74acf12723045a64f1f2c6298ac7b34a5d5
>> Author: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
>> Date:   Fri May 15 13:34:29 2015 +0300
>> 
>>     drm/i915: Don't overwrite (e)DP PLL selection on SKL
>>     
>>     In the following commit, the place where the contents of dpll_hw_state
>>     in crtc_state where zeroed was changed. Prior to that commit, it
>>     happened when the new state was allocated, but now that happens just
>>     before the call the .crtc_compute_clock() hook. The DP code for SKL,
>>     however, sets up the (private) PLL in the encoder compute config
>>     function that has already run by the time that memset() is reached,
>>     causing the previous value to be lost.
>>     
>>     This patch fixes the issue by moving the memset() down the call chain,
>>     so that it is only called if the values in dpll_hw_state are going to be
>>     updated.
>>     
>>     commit 4978cc93d9ac240b435ce60431aef24239b4c270
>>     Author: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
>>     Date:   Tue Apr 21 17:13:21 2015 +0300
>>     
>>         drm/i915: Preserve shared DPLL information in new pipe_config
>>     
>>     Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=90462
>>     Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
>>     Reviewed-by: Damien Lespiau <damien.lespiau@intel.com>
>>     Reported-and-tested-by: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
>>     Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
>> 
>> This doesn't revert cleanly on Linus' master, and I didn't have the time
>> to look deeper right now. However I confirmed that this commit fails and
>> its parent doesn't.
>
> Note that there seems to be two bugs here: Firs one is display state
> checker getting annoyed, which is probably the one Jani bisected to here
> (please confirm).

Right, the above bisect is for the "[drm:check_crtc_state [i915]]
*ERROR* mismatch in dpll_hw_state.wrpll (expected 0x90280202, found
0x00000000)" warning.

> The other is the two backtraces complaining that the pipe is off (both the
> drm_irq.c and the one in hsw_display_ips amount to that) because we seem
> to call disable_planes on a disable pipe, which is bullocks (with runtime
> pm the hw is dead and will just drop the writes).

The "vblank not available on crtc 0, ret=-22" warning bisects to

commit 8c7b5ccb729870e606321b3703e2c2e698c49a95
Author: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Date:   Tue Apr 21 17:13:19 2015 +0300

    drm/i915: Use atomic helpers for computing changed flags
    
    Replace the drivers own logic for computing mode_changed, active_changed
    and planes_changed flags with the check_modeset() atomic helper. Since
    that function needs to compare the crtc's new mode with the current,
    this patch also moves the set up of crtc_state->mode earlier in the call
    chain.
    
    Note that for the call to check_plane() to work properly, we need to
    check new plane state against new crtc state. But since we still use the
    plane update helper, which doesn't have a full atomic state, we need to
    hack around that in intel_plane_atomic_check().
    
    Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
    Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
    Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>


BR,
Jani.


> -Daniel
> -- 
> Daniel Vetter
> Software Engineer, Intel Corporation
> http://blog.ffwll.ch

-- 
Jani Nikula, Intel Open Source Technology Center