All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Michel Dänzer" <michel@daenzer.net>
To: "Mario Kleiner" <mario.kleiner.de@gmail.com>,
	"Vlastimil Babka" <vbabka@suse.cz>,
	"Ville Syrjälä" <ville.syrjala@linux.intel.com>
Cc: "Daniel Vetter" <daniel.vetter@ffwll.ch>,
	LKML <linux-kernel@vger.kernel.org>,
	dri-devel@lists.freedesktop.org, mgraesslin@kde.org,
	kwin@kde.org, "Alex Deucher" <alexander.deucher@amd.com>,
	"Christian König" <christian.koenig@amd.com>
Subject: Re: linux-4.4 bisected: kwin5 stuck on kde5 loading screen with radeon
Date: Thu, 21 Jan 2016 15:41:27 +0900	[thread overview]
Message-ID: <56A07D97.6030606@daenzer.net> (raw)
In-Reply-To: <56A07CF9.5060506@daenzer.net>

On 21.01.2016 15:38, Michel Dänzer wrote:
> On 21.01.2016 14:31, Mario Kleiner wrote:
>> On 01/21/2016 04:43 AM, Michel Dänzer wrote:
>>> On 21.01.2016 05:32, Mario Kleiner wrote:
>>>>
>>>> So the problem is that AMDs hardware frame counters reset to
>>>> zero during a modeset. The old DRM code dealt with drivers doing that by
>>>> keeping vblank irqs enabled during modesets and incrementing vblank
>>>> count by one during each vblank irq, i think that's what
>>>> drm_vblank_pre_modeset() and drm_vblank_post_modeset() were meant for.
>>>
>>> Right, looks like there's been a regression breaking this. I suspect the
>>> problem is that vblank->last isn't getting updated from
>>> drm_vblank_post_modeset. Not sure which change broke that though, or how
>>> to fix it. Ville?
>>>
>>
>> The whole logic has changed and the software counter updates are now
>> driven all the time by the hw counter.
>>
>>>
>>> BTW, I'm seeing a similar issue with drm_vblank_on/off as well, which
>>> exposed the bug fixed by 209e4dbc ("drm/vblank: Use u32 consistently for
>>> vblank counters"). I've been meaning to track that down since then; one
>>> of these days hopefully, but if anybody has any ideas offhand...
>>
>> I spent the last few hours reading through the drm and radeon code and i
>> think what should probably work is to replace the
>> drm_vblank_pre/post_modeset calls in radeon/amdgpu by drm_vblank_off/on
>> calls. These are apparently meant for drivers whose hw counters reset
>> during modeset, [...]
> 
> ... just like drm_vblank_pre/post_modeset. That those were broken is a
> regression which needs to be fixed anyway. I don't think switching to
> drm_vblank_on/off is suitable for stable trees.

Even more so since as I mentioned, there is (has been since at least
about half a year ago) a counter jumping bug with drm_vblank_on/off as well.


-- 
Earthling Michel Dänzer               |               http://www.amd.com
Libre software enthusiast             |             Mesa and X developer

WARNING: multiple messages have this Message-ID (diff)
From: "Michel Dänzer" <michel@daenzer.net>
To: "Mario Kleiner" <mario.kleiner.de@gmail.com>,
	"Vlastimil Babka" <vbabka@suse.cz>,
	"Ville Syrjälä" <ville.syrjala@linux.intel.com>
Cc: "Daniel Vetter" <daniel.vetter@ffwll.ch>,
	LKML <linux-kernel@vger.kernel.org>,
	dri-devel@lists.freedesktop.org, mgraesslin@kde.org,
	kwin@kde.org, "Alex Deucher" <alexander.deucher@amd.com>,
	"Christian König" <christian.koenig@amd.com>
Subject: Re: linux-4.4 bisected: kwin5 stuck on kde5 loading screen with radeon
Date: Thu, 21 Jan 2016 15:41:27 +0900	[thread overview]
Message-ID: <56A07D97.6030606@daenzer.net> (raw)
In-Reply-To: <56A07CF9.5060506@daenzer.net>

On 21.01.2016 15:38, Michel Dänzer wrote:
> On 21.01.2016 14:31, Mario Kleiner wrote:
>> On 01/21/2016 04:43 AM, Michel Dänzer wrote:
>>> On 21.01.2016 05:32, Mario Kleiner wrote:
>>>>
>>>> So the problem is that AMDs hardware frame counters reset to
>>>> zero during a modeset. The old DRM code dealt with drivers doing that by
>>>> keeping vblank irqs enabled during modesets and incrementing vblank
>>>> count by one during each vblank irq, i think that's what
>>>> drm_vblank_pre_modeset() and drm_vblank_post_modeset() were meant for.
>>>
>>> Right, looks like there's been a regression breaking this. I suspect the
>>> problem is that vblank->last isn't getting updated from
>>> drm_vblank_post_modeset. Not sure which change broke that though, or how
>>> to fix it. Ville?
>>>
>>
>> The whole logic has changed and the software counter updates are now
>> driven all the time by the hw counter.
>>
>>>
>>> BTW, I'm seeing a similar issue with drm_vblank_on/off as well, which
>>> exposed the bug fixed by 209e4dbc ("drm/vblank: Use u32 consistently for
>>> vblank counters"). I've been meaning to track that down since then; one
>>> of these days hopefully, but if anybody has any ideas offhand...
>>
>> I spent the last few hours reading through the drm and radeon code and i
>> think what should probably work is to replace the
>> drm_vblank_pre/post_modeset calls in radeon/amdgpu by drm_vblank_off/on
>> calls. These are apparently meant for drivers whose hw counters reset
>> during modeset, [...]
> 
> ... just like drm_vblank_pre/post_modeset. That those were broken is a
> regression which needs to be fixed anyway. I don't think switching to
> drm_vblank_on/off is suitable for stable trees.

Even more so since as I mentioned, there is (has been since at least
about half a year ago) a counter jumping bug with drm_vblank_on/off as well.


-- 
Earthling Michel Dänzer               |               http://www.amd.com
Libre software enthusiast             |             Mesa and X developer
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/dri-devel

  reply	other threads:[~2016-01-21  6:41 UTC|newest]

Thread overview: 59+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-01-15 10:34 linux-4.4 bisected: kwin5 stuck on kde5 loading screen with radeon Vlastimil Babka
2016-01-15 12:26 ` Ville Syrjälä
2016-01-15 12:40   ` Vlastimil Babka
2016-01-16  4:24   ` Mario Kleiner
2016-01-16  4:24     ` Mario Kleiner
2016-01-18 10:49     ` Vlastimil Babka
2016-01-18 14:06       ` Vlastimil Babka
2016-01-18 14:14         ` Christian König
2016-01-18 14:14           ` Christian König
2016-01-20 20:25       ` Vlastimil Babka
2016-01-20 20:25         ` Vlastimil Babka
2016-01-20 20:32       ` Mario Kleiner
2016-01-20 20:32         ` Mario Kleiner
2016-01-21  3:43         ` Michel Dänzer
2016-01-21  3:43           ` Michel Dänzer
2016-01-21  5:31           ` Mario Kleiner
2016-01-21  5:31             ` Mario Kleiner
2016-01-21  6:38             ` Michel Dänzer
2016-01-21  6:38               ` Michel Dänzer
2016-01-21  6:41               ` Michel Dänzer [this message]
2016-01-21  6:41                 ` Michel Dänzer
2016-01-21  7:58                 ` Daniel Vetter
2016-01-21  7:58                   ` Daniel Vetter
2016-01-21  8:36                   ` Michel Dänzer
2016-01-21  8:36                     ` Michel Dänzer
2016-01-21 10:09                     ` Daniel Vetter
2016-01-21 10:09                       ` Daniel Vetter
2016-01-22  3:06                       ` Michel Dänzer
2016-01-22  3:06                         ` Michel Dänzer
2016-01-22 15:18                         ` Ville Syrjälä
2016-01-22 15:18                           ` Ville Syrjälä
2016-01-22 18:29                           ` Mario Kleiner
2016-01-22 18:29                             ` Mario Kleiner
2016-01-23 18:23                             ` Mario Kleiner
2016-01-23 18:23                               ` Mario Kleiner
2016-01-25  4:15                           ` Michel Dänzer
2016-01-25  4:15                             ` Michel Dänzer
2016-01-25 13:16                             ` Mario Kleiner
2016-01-25 13:16                               ` Mario Kleiner
2016-01-25 13:23                               ` Ville Syrjälä
2016-01-25 13:44                                 ` Mario Kleiner
2016-01-25 13:44                                   ` Mario Kleiner
2016-01-25 14:53                                   ` Ville Syrjälä
2016-01-25 14:53                                     ` Ville Syrjälä
2016-01-25 16:38                                     ` Mario Kleiner
2016-01-25 18:51                                       ` Daniel Vetter
2016-01-25 18:51                                         ` Daniel Vetter
2016-01-25 19:30                                         ` Mario Kleiner
2016-01-25 19:30                                           ` Mario Kleiner
2016-01-25 20:32                                           ` Daniel Vetter
2016-01-25 20:32                                             ` Daniel Vetter
2016-01-25 21:42                                             ` Mario Kleiner
2016-01-25 21:42                                               ` Mario Kleiner
2016-01-25 22:05                                               ` Daniel Vetter
2016-01-25 22:05                                                 ` Daniel Vetter
2016-01-21  8:28               ` Mario Kleiner
2016-01-21  8:28                 ` Mario Kleiner
2016-01-21  9:15                 ` Vlastimil Babka
2016-01-21  9:15                   ` Vlastimil Babka

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=56A07D97.6030606@daenzer.net \
    --to=michel@daenzer.net \
    --cc=alexander.deucher@amd.com \
    --cc=christian.koenig@amd.com \
    --cc=daniel.vetter@ffwll.ch \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=kwin@kde.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mario.kleiner.de@gmail.com \
    --cc=mgraesslin@kde.org \
    --cc=vbabka@suse.cz \
    --cc=ville.syrjala@linux.intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.