All of lore.kernel.org
 help / color / mirror / Atom feed
From: Vlastimil Babka <vbabka@suse.cz>
To: Mario Kleiner <mario.kleiner.de@gmail.com>,
	dri-devel@lists.freedesktop.org
Cc: linux@bernd-steinhauser.de, stable@vger.kernel.org,
	michel@daenzer.net, ville.syrjala@linux.intel.com,
	daniel.vetter@ffwll.ch, alexander.deucher@amd.com,
	christian.koenig@amd.com
Subject: Re: [PATCH 3/6] drm: Fix drm_vblank_pre/post_modeset regression from Linux 4.4
Date: Thu, 11 Feb 2016 14:03:39 +0100	[thread overview]
Message-ID: <56BC86AB.4070004@suse.cz> (raw)
In-Reply-To: <1454894009-15466-4-git-send-email-mario.kleiner.de@gmail.com>

On 02/08/2016 02:13 AM, Mario Kleiner wrote:
> Changes to drm_update_vblank_count() in Linux 4.4 broke the
> behaviour of the pre/post modeset functions as the new update
> code doesn't deal with hw vblank counter resets inbetween calls
> to drm_vblank_pre_modeset an drm_vblank_post_modeset, as it
> should.
>
> This causes mistreatment of such hw counter resets as counter
> wraparound, and thereby large forward jumps of the software
> vblank counter which in turn cause vblank event dispatching
> and vblank waits to fail/hang --> userspace clients hang.
>
> This symptom was reported on radeon-kms to cause a infinite
> hang of KDE Plasma 5 shell's login procedure, preventing users
> from logging in.
>
> Fix this by detecting when drm_update_vblank_count() is called
> inside a pre->post modeset interval. If so, clamp valid vblank
> increments to the safe values 0 and 1, pretty much restoring
> the update behavior of the old update code of Linux 4.3 and
> earlier. Also reset the last recorded hw vblank count at call
> to drm_vblank_post_modeset() to be safe against hw that after
> modesetting, dpms on etc. only fires its first vblank irq after
> drm_vblank_post_modeset() was already called.
>
> Reported-by: Vlastimil Babka <vbabka@suse.cz>

FWIW, I've applied the whole patchset to 4.4 and the kde5 login problem 
didn't occur. I can test the next version too.

Thanks,
Vlastimil


  parent reply	other threads:[~2016-02-11 13:03 UTC|newest]

Thread overview: 55+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-02-08  1:13 drm vblank regression fixes for Linux 4.4+ Mario Kleiner
2016-02-08  1:13 ` [PATCH 1/6] drm: No-Op redundant calls to drm_vblank_off() Mario Kleiner
2016-02-08  1:13   ` Mario Kleiner
2016-02-09  9:54   ` Daniel Vetter
2016-02-09  9:54     ` Daniel Vetter
2016-02-09 13:27     ` Mario Kleiner
2016-02-08  1:13 ` [PATCH 2/6] drm: Prevent vblank counter bumps > 1 with active vblank clients Mario Kleiner
2016-02-08  1:13   ` Mario Kleiner
2016-02-09  9:56   ` Daniel Vetter
2016-02-09  9:56     ` Daniel Vetter
2016-02-09 10:07     ` Ville Syrjälä
2016-02-09 10:07       ` Ville Syrjälä
2016-02-09 10:23       ` Daniel Vetter
2016-02-09 10:23         ` Daniel Vetter
2016-02-09 13:39         ` Mario Kleiner
2016-02-09 13:39           ` Mario Kleiner
2016-02-09 14:29           ` Daniel Vetter
2016-02-09 14:29             ` Daniel Vetter
2016-02-09 16:18             ` Mario Kleiner
2016-02-09 16:18               ` Mario Kleiner
2016-02-08  1:13 ` [PATCH 3/6] drm: Fix drm_vblank_pre/post_modeset regression from Linux 4.4 Mario Kleiner
2016-02-08  1:13   ` Mario Kleiner
2016-02-09 10:00   ` Daniel Vetter
2016-02-11 13:03   ` Vlastimil Babka [this message]
2016-02-08  1:13 ` [PATCH 4/6] drm: Fix treatment of drm_vblank_offdelay in drm_vblank_on() Mario Kleiner
2016-02-08  1:13   ` Mario Kleiner
2016-02-09 10:06   ` Daniel Vetter
2016-02-09 10:06     ` Daniel Vetter
2016-02-09 11:10     ` Ville Syrjälä
2016-02-09 11:10       ` Ville Syrjälä
2016-02-09 13:29       ` Mario Kleiner
2016-02-09 13:29         ` Mario Kleiner
2016-02-09 13:41         ` Ville Syrjälä
2016-02-09 13:41           ` Ville Syrjälä
2016-02-09 14:31           ` Daniel Vetter
2016-02-09 14:31             ` Daniel Vetter
2016-02-08  1:13 ` [PATCH 5/6] drm: Prevent vblank counter jumps with timestamp based update method Mario Kleiner
2016-02-08  1:13   ` Mario Kleiner
2016-02-09 10:09   ` Daniel Vetter
2016-02-09 13:53     ` Mario Kleiner
2016-02-09 14:11       ` Ville Syrjälä
2016-02-09 14:11         ` Ville Syrjälä
2016-02-09 15:03         ` Daniel Vetter
2016-02-09 15:03           ` Daniel Vetter
2016-02-10 16:28           ` Mario Kleiner
2016-02-10 16:28             ` Mario Kleiner
2016-02-10 17:17             ` Daniel Vetter
2016-02-10 18:36               ` Mario Kleiner
2016-02-10 19:34                 ` Daniel Vetter
2016-02-10 19:34                   ` Daniel Vetter
2016-02-08  1:13 ` [PATCH 6/6] drm/radeon/pm: Handle failure of drm_vblank_get Mario Kleiner
2016-02-09 10:10   ` Daniel Vetter
2016-02-12 19:30 Respin: drm vblank regression fixes for Linux 4.4+ (v2) Mario Kleiner
2016-02-12 19:30 ` [PATCH 3/6] drm: Fix drm_vblank_pre/post_modeset regression from Linux 4.4 Mario Kleiner
2016-02-12 19:30   ` Mario Kleiner
2016-02-24  8:59   ` 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=56BC86AB.4070004@suse.cz \
    --to=vbabka@suse.cz \
    --cc=alexander.deucher@amd.com \
    --cc=christian.koenig@amd.com \
    --cc=daniel.vetter@ffwll.ch \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=linux@bernd-steinhauser.de \
    --cc=mario.kleiner.de@gmail.com \
    --cc=michel@daenzer.net \
    --cc=stable@vger.kernel.org \
    --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.