dri-devel.lists.freedesktop.org archive mirror
 help / color / mirror / Atom feed
From: Mario Kleiner <mario.kleiner.de@gmail.com>
To: ville.syrjala@linux.intel.com, dri-devel@lists.freedesktop.org
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>,
	intel-gfx@lists.freedesktop.org, Dave Airlie <airlied@redhat.com>
Subject: Re: [PATCH 14/19] drm: Don't update vblank timestamp when the counter didn't change
Date: Sat, 13 Sep 2014 18:25:54 +0200	[thread overview]
Message-ID: <54147012.40609@gmail.com> (raw)
In-Reply-To: <1407325803-6944-15-git-send-email-ville.syrjala@linux.intel.com>

[-- Attachment #1: Type: text/plain, Size: 1903 bytes --]

The current drm-next misses Ville's original Patch 14/19, the one i 
first objected, then objected to my objection. It is needed to avoid 
actual regressions. Attached a trivially rebased (v2) of Ville's patch 
to go on top of drm-next, also as tgz in case my e-mail client mangles 
the patch again, because it's one of those "email hates me" weeks.

-mario



On 08/06/2014 01:49 PM, ville.syrjala@linux.intel.com wrote:
> From: Ville Syrjälä <ville.syrjala@linux.intel.com>
>
> If we already have a timestamp for the current vblank counter, don't
> update it with a new timestmap. Small errors can creep in between two
> timestamp queries for the same vblank count, which could be confusing to
> userspace when it queries the timestamp for the same vblank sequence
> number twice.
>
> This problem gets exposed when the vblank disable timer is not used
> (or is set to expire quickly) and thus we can get multiple vblank
> disable<->enable transition during the same frame which would all
> attempt to update the timestamp with the latest estimate.
>
> Testcase: igt/kms_flip/flip-vs-expired-vblank
> Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
> ---
>   drivers/gpu/drm/drm_irq.c | 3 +++
>   1 file changed, 3 insertions(+)
>
> diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c
> index af33df1..0523f5b 100644
> --- a/drivers/gpu/drm/drm_irq.c
> +++ b/drivers/gpu/drm/drm_irq.c
> @@ -106,6 +106,9 @@ static void drm_update_vblank_count(struct drm_device *dev, int crtc)
>   	DRM_DEBUG("enabling vblank interrupts on crtc %d, missed %d\n",
>   		  crtc, diff);
>   
> +	if (diff == 0)
> +		return;
> +
>   	/* Reinitialize corresponding vblank timestamp if high-precision query
>   	 * available. Skip this step if query unsupported or failed. Will
>   	 * reinitialize delayed at next vblank interrupt in that case.


[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: 0001-drm-Don-t-update-vblank-timestamp-when-the-counter-d.patch --]
[-- Type: text/x-patch; name="0001-drm-Don-t-update-vblank-timestamp-when-the-counter-d.patch", Size: 1909 bytes --]

>From c0a5228a7fc43d4c3615a471c340b68bcb2caa16 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 6 Aug 2014 14:49:57 +0300
Subject: [PATCH] drm: Don't update vblank timestamp when the counter didn't
 change (v2)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

If we already have a timestamp for the current vblank counter, don't
update it with a new timestmap. Small errors can creep in between two
timestamp queries for the same vblank count, which could be confusing to
userspace when it queries the timestamp for the same vblank sequence
number twice.

This problem gets exposed when the vblank disable timer is not used
(or is set to expire quickly) and thus we can get multiple vblank
disable<->enable transition during the same frame which would all
attempt to update the timestamp with the latest estimate.

Testcase: igt/kms_flip/flip-vs-expired-vblank
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Mario Kleiner <mario.kleiner.de@gmail.com>

v2:Mario: Trivial rebase on top of current drm-next (13-Sep-2014)
Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com>
---
 drivers/gpu/drm/drm_irq.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c
index 80ff94a..e73cbda 100644
--- a/drivers/gpu/drm/drm_irq.c
+++ b/drivers/gpu/drm/drm_irq.c
@@ -126,6 +126,9 @@ static void drm_update_vblank_count(struct drm_device *dev, int crtc)
 	DRM_DEBUG("updating vblank count on crtc %d, missed %d\n",
 		  crtc, diff);
 
+	if (diff == 0)
+		return;
+
 	/* Reinitialize corresponding vblank timestamp if high-precision query
 	 * available. Skip this step if query unsupported or failed. Will
 	 * reinitialize delayed at next vblank interrupt in that case.
-- 
1.9.1


[-- Attachment #3: 0001-drm-Don-t-update-vblank-timestamp-when-the-counter-d.patch.tar.gz --]
[-- Type: application/gzip, Size: 1171 bytes --]

[-- Attachment #4: Type: text/plain, Size: 159 bytes --]

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

  parent reply	other threads:[~2014-09-13 16:25 UTC|newest]

Thread overview: 38+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-08-06 11:49 [PATCH v2 00/19] drm: More vblank on/off work ville.syrjala
2014-08-06 11:49 ` [PATCH v2 01/19] drm: Always reject drm_vblank_get() after drm_vblank_off() ville.syrjala
2014-08-06 11:49 ` [PATCH v2 02/19] drm/i915: Warn if drm_vblank_get() still works " ville.syrjala
2014-08-06 11:49 ` [PATCH 03/19] drm: Don't clear vblank timestamps when vblank interrupt is disabled ville.syrjala
2014-08-06 11:49 ` [PATCH 04/19] drm: Move drm_update_vblank_count() ville.syrjala
2014-08-06 11:49 ` [PATCH 05/19] drm: Have the vblank counter account for the time between vblank irq disable and drm_vblank_off() ville.syrjala
2014-09-02 19:33   ` Mario Kleiner
2014-08-06 11:49 ` [PATCH 06/19] drm: Avoid random vblank counter jumps if the hardware counter has been reset ville.syrjala
2014-08-06 11:49 ` [PATCH v2 07/19] drm: Reduce the amount of dev->vblank[crtc] in the code ville.syrjala
2014-08-06 11:49 ` [PATCH 08/19] drm: Fix deadlock between event_lock and vbl_lock/vblank_time_lock ville.syrjala
2014-08-06 11:49 ` [PATCH 09/19] drm: Fix race between drm_vblank_off() and drm_queue_vblank_event() ville.syrjala
2014-08-06 13:23   ` Daniel Vetter
2014-08-06 13:33     ` [Intel-gfx] " Ville Syrjälä
2014-08-06 11:49 ` [PATCH v2 10/19] drm: Disable vblank interrupt immediately when drm_vblank_offdelay<0 ville.syrjala
2014-08-06 11:49 ` [PATCH v2 11/19] drm: Add dev->vblank_disable_immediate flag ville.syrjala
2014-08-06 11:49 ` [PATCH 12/19] drm/i915: Opt out of vblank disable timer on >gen2 ville.syrjala
2014-08-06 11:49 ` [PATCH v2 13/19] drm: Kick start vblank interrupts at drm_vblank_on() ville.syrjala
2014-08-06 11:49 ` [PATCH 14/19] drm: Don't update vblank timestamp when the counter didn't change ville.syrjala
2014-08-06 12:56   ` Daniel Vetter
2014-08-06 13:09     ` Daniel Vetter
2014-09-04 12:14   ` Mario Kleiner
2014-09-13 16:25   ` Mario Kleiner [this message]
2014-09-15  8:50     ` Daniel Vetter
2014-09-23 12:48       ` [Intel-gfx] " Jani Nikula
2014-09-23 13:51         ` Daniel Vetter
2014-09-23 14:15           ` Mario Kleiner
2014-08-06 11:49 ` [PATCH 15/19] drm: Update vblank->last in drm_update_vblank_count() ville.syrjala
2014-08-06 13:08   ` [Intel-gfx] " Daniel Vetter
2014-08-06 13:30     ` Ville Syrjälä
2014-08-06 14:19       ` Daniel Vetter
2014-08-06 11:49 ` [PATCH 16/19] drm: Store the vblank timestamp when adjusting the counter during disable ville.syrjala
2014-08-06 13:12   ` [Intel-gfx] " Daniel Vetter
2014-08-06 11:50 ` [PATCH 17/19] drm/i915: Clear .last vblank count before drm_vblank_off() when sanitizing crtc state ville.syrjala
2014-08-06 13:30   ` [Intel-gfx] " Daniel Vetter
2014-08-06 13:43     ` Ville Syrjälä
2014-08-06 11:50 ` [PATCH 18/19] drm/i915: Update scanline_offset only for active crtcs ville.syrjala
2014-08-06 11:50 ` [PATCH 19/19] drm: Fix confusing debug message in drm_update_vblank_count() ville.syrjala
2014-08-06 11:50 ` [PATCH igt] tests/kms_flip: Assert that vblank timestamps aren't zeroed ville.syrjala

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=54147012.40609@gmail.com \
    --to=mario.kleiner.de@gmail.com \
    --cc=airlied@redhat.com \
    --cc=daniel.vetter@ffwll.ch \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=intel-gfx@lists.freedesktop.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).