[Bug 93147] [regression bisected] Stuttering in games caused by commit 4dfd6486 "drm: Use vblank timestamps to guesstimate how many vblanks were missed"

[bugzilla-dae...@freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=93147

Alex Deucher  changed:

   What|Removed |Added

 Status|NEW |RESOLVED
 Resolution|--- |FIXED

--- Comment #22 from Alex Deucher  ---
Both are already upstream:
http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=5b5561b3660db734652fbd02b4b6cbe00434d96b
http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=8e36f9d33c134d5c6448ad65b423a9fd94e045cf

-- 
You are receiving this mail because:
You are the assignee for the bug.
-- next part --
An HTML attachment was scrubbed...
URL: 

[Bug 93147] [regression bisected] Stuttering in games caused by commit 4dfd6486 "drm: Use vblank timestamps to guesstimate how many vblanks were missed"

[bugzilla-dae...@freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=93147

--- Comment #21 from Ernst Sjöstrand  ---
Don't forget to merge the AMDGPU version also.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-- next part --
An HTML attachment was scrubbed...
URL: 

[Bug 93147] [regression bisected] Stuttering in games caused by commit 4dfd6486 "drm: Use vblank timestamps to guesstimate how many vblanks were missed"

[bugzilla-dae...@freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=93147

--- Comment #20 from Dave Witbrodt  ---
Just brought my local tree up to date with drm-fixes.  The patch from comment
13 is now present there (and has been picked up by Linus), and updating with
this patch and the others relevant to my system produces a nicely working
kernel.  Thanks for getting this fixed!

I'm not sure what the protocol is for closing this bug.  Usually if a bug is in
a released kernel, the bug is not closed until the patch appears in a released
kernel.  This bug never appeared in a released kernel, though, so it seems like
it can be closed immediately?

-- 
You are receiving this mail because:
You are the assignee for the bug.
-- next part --
An HTML attachment was scrubbed...
URL: 

[Bug 93147] [regression bisected] Stuttering in games caused by commit 4dfd6486 "drm: Use vblank timestamps to guesstimate how many vblanks were missed"

[bugzilla-dae...@freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=93147

--- Comment #19 from Ernst Sjöstrand  ---
Tested the amdgpu v3 patch on top of agd5f/amdgpu-powerplay, seems to work
fine.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-- next part --
An HTML attachment was scrubbed...
URL: 

[Bug 93147] [regression bisected] Stuttering in games caused by commit 4dfd6486 "drm: Use vblank timestamps to guesstimate how many vblanks were missed"

[bugzilla-dae...@freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=93147

--- Comment #18 from Dave Witbrodt  ---
(In reply to Mario Kleiner from comment #13)
> Created attachment 120350 [details] [review]
> v2 of the radeon-kms fix: Slightly more efficient.

Works very well on my hardware.  Thanks for "Tested By:"

Will wait to see if agd5f picks this up (then airlied), or if others will be
commenting/changing it first.

DW

-- 
You are receiving this mail because:
You are the assignee for the bug.
-- next part --
An HTML attachment was scrubbed...
URL: 

[Bug 93147] [regression bisected] Stuttering in games caused by commit 4dfd6486 "drm: Use vblank timestamps to guesstimate how many vblanks were missed"

[bugzilla-dae...@freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=93147

Alex Deucher  changed:

   What|Removed |Added

 Attachment #120310|0   |1
is obsolete||
 Attachment #120351|0   |1
is obsolete||

--- Comment #17 from Alex Deucher  ---
Created attachment 120356
  --> https://bugs.freedesktop.org/attachment.cgi?id=120356=edit
v3 for amdgpu

Fix a crash in the crtc disabled case.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-- next part --
An HTML attachment was scrubbed...
URL: 

[Bug 93147] [regression bisected] Stuttering in games caused by commit 4dfd6486 "drm: Use vblank timestamps to guesstimate how many vblanks were missed"

[bugzilla-dae...@freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=93147

--- Comment #16 from Ernst Sjöstrand  ---
I tried the v2 amdgpu patch on top of agd5f/amdgpu-powerplay and my screens
never turn on with it.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-- next part --
An HTML attachment was scrubbed...
URL: 

[Bug 93147] [regression bisected] Stuttering in games caused by commit 4dfd6486 "drm: Use vblank timestamps to guesstimate how many vblanks were missed"

[bugzilla-dae...@freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=93147

--- Comment #15 from Mario Kleiner  ---
Thanks Dave for testing the v1 of the patch.

I think i'm done with these, v2 should be as good as it gets from my side.
-mario

-- 
You are receiving this mail because:
You are the assignee for the bug.
-- next part --
An HTML attachment was scrubbed...
URL: 

[Bug 93147] [regression bisected] Stuttering in games caused by commit 4dfd6486 "drm: Use vblank timestamps to guesstimate how many vblanks were missed"

[bugzilla-dae...@freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=93147

--- Comment #14 from Mario Kleiner  ---
Created attachment 120351
  --> https://bugs.freedesktop.org/attachment.cgi?id=120351=edit
v2 of the amdgpu-kms fix: Slightly more efficient.

Like v2 of radeon-kms. Alex original port is R-b'ed by me.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-- next part --
An HTML attachment was scrubbed...
URL: 

[Bug 93147] [regression bisected] Stuttering in games caused by commit 4dfd6486 "drm: Use vblank timestamps to guesstimate how many vblanks were missed"

[bugzilla-dae...@freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=93147

--- Comment #13 from Mario Kleiner  ---
Created attachment 120350
  --> https://bugs.freedesktop.org/attachment.cgi?id=120350=edit
v2 of the radeon-kms fix: Slightly more efficient.

Same as original, but busy wait in flip_work_func replaced by hr-timer sleep
with event lock released, for more efficiency. Ditto for the following amdgpu
fix v2.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-- next part --
An HTML attachment was scrubbed...
URL: 

[Bug 93147] [regression bisected] Stuttering in games caused by commit 4dfd6486 "drm: Use vblank timestamps to guesstimate how many vblanks were missed"

[bugzilla-dae...@freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=93147

--- Comment #12 from Dave Witbrodt  ---
(In reply to Mario Kleiner from comment #10)
> Created attachment 120301 [details] [review]
> First proposed patch to fix this on radeon-kms

Works good on my hardware.  Thanks.

Will test any proposed fixes that follow, as well.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-- next part --
An HTML attachment was scrubbed...
URL: 

[Bug 93147] [regression bisected] Stuttering in games caused by commit 4dfd6486 "drm: Use vblank timestamps to guesstimate how many vblanks were missed"

[bugzilla-dae...@freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=93147

--- Comment #11 from Alex Deucher  ---
Created attachment 120310
  --> https://bugs.freedesktop.org/attachment.cgi?id=120310=edit
port fix to amdgpu

-- 
You are receiving this mail because:
You are the assignee for the bug.
-- next part --
An HTML attachment was scrubbed...
URL: 

[Bug 93147] [regression bisected] Stuttering in games caused by commit 4dfd6486 "drm: Use vblank timestamps to guesstimate how many vblanks were missed"

[bugzilla-dae...@freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=93147

--- Comment #10 from Mario Kleiner  ---
Created attachment 120301
  --> https://bugs.freedesktop.org/attachment.cgi?id=120301=edit
First proposed patch to fix this on radeon-kms

Ok, the attached patch works on my test systems with my timing tests and
hardware timing measurement equipment and i think it hopefully avoids the races
Ville and me could come up with so far.

One thing which might make sense for improved efficiency is to replace the
udelay(5) in the radeon_flip_work_func() with a usleep_range() that avoids
polling and sleeps the right minimum amount of time. I'll try to try that
today.

For the line buffer sizes for < DCE4 i just guessed values which are hopefully
big enough to cover earlier asics. I think too big is not a big problem for
correctness, but potentially for performance under higher graphics load.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-- next part --
An HTML attachment was scrubbed...
URL: 

[Bug 93147] [regression bisected] Stuttering in games caused by commit 4dfd6486 "drm: Use vblank timestamps to guesstimate how many vblanks were missed"

[bugzilla-dae...@freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=93147

--- Comment #9 from Mario Kleiner  ---
(In reply to Michel Dänzer from comment #7)
> Mario, can you submit your fix(es) for inclusion?

Still on it. The proper patch for radeon-kms will hopefully be finished and
tested within a day or so. But ideally i'd need some feedback on the line
buffer sizes/partitions/watermarks - see my last e-mail on that. Otherwise i'll
have to use very conservative guesstimates for the fudge constant in the
current patch to be on the safe side.

amdgpu-kms will have the same problem and solution, but i haven't looked into
how to hook that up, with the difference in how the display engine code is
hooked up.

thanks,
-mario

-- 
You are receiving this mail because:
You are the assignee for the bug.
-- next part --
An HTML attachment was scrubbed...
URL: 

[Bug 93147] [regression bisected] Stuttering in games caused by commit 4dfd6486 "drm: Use vblank timestamps to guesstimate how many vblanks were missed"

[bugzilla-dae...@freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=93147

--- Comment #8 from Dave Witbrodt  ---
(In reply to Michel Dänzer from comment #6)
> (In reply to Ville Syrjala from comment #5)
> > I'm guessing this is the stuff Mario was trying to fix.
> 
> You're right, the patch from
> http://lists.freedesktop.org/archives/dri-devel/2015-November/095679.html
> fixes it for me.

Applying the patch in question to my local 4.3 + DRM 4.4 kernel (with no
reverts) makes the symptoms go away for me also.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-- next part --
An HTML attachment was scrubbed...
URL: 

[Bug 93147] [regression bisected] Stuttering in games caused by commit 4dfd6486 "drm: Use vblank timestamps to guesstimate how many vblanks were missed"

[bugzilla-dae...@freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=93147

Michel Dänzer  changed:

   What|Removed |Added

 CC||mario.kleiner at tuebingen.mpg
   ||.de

--- Comment #7 from Michel Dänzer  ---
Mario, can you submit your fix(es) for inclusion?

-- 
You are receiving this mail because:
You are the assignee for the bug.
-- next part --
An HTML attachment was scrubbed...
URL: 

[Bug 93147] [regression bisected] Stuttering in games caused by commit 4dfd6486 "drm: Use vblank timestamps to guesstimate how many vblanks were missed"

[bugzilla-dae...@freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=93147

Michel Dänzer  changed:

   What|Removed |Added

  Component|DRM/other   |DRM/Radeon

--- Comment #6 from Michel Dänzer  ---
(In reply to Ville Syrjala from comment #5)
> I'm guessing this is the stuff Mario was trying to fix.

You're right, the patch from
http://lists.freedesktop.org/archives/dri-devel/2015-November/095679.html fixes
it for me.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-- next part --
An HTML attachment was scrubbed...
URL: 

[Bug 93147] [regression bisected] Stuttering in games caused by commit 4dfd6486 "drm: Use vblank timestamps to guesstimate how many vblanks were missed"

[bugzilla-dae...@freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=93147

--- Comment #5 from Ville Syrjala  ---
(In reply to Michel Dänzer from comment #4)
> I recently ran into this with armagetronad. I've now confirmed that
> reverting these changes fixes that as well, thanks Dave for saving me the
> time to track this down!
> 
> Ville, any ideas what's going on / how to fix it? If not, please revert
> these changes.

I'm guessing this is the stuff Mario was trying to fix. Oh FFS, apparently he
didn't cc dri-devel so the entire discussion has all been in private :( Though
you were cc:d on most of it I think. Maybe I can bounce the entire thing onto
the list...

-- 
You are receiving this mail because:
You are the assignee for the bug.
-- next part --
An HTML attachment was scrubbed...
URL: 

[Bug 93147] [regression bisected] Stuttering in games caused by commit 4dfd6486 "drm: Use vblank timestamps to guesstimate how many vblanks were missed"

[bugzilla-dae...@freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=93147

Michel Dänzer  changed:

   What|Removed |Added

  Component|DRM/Radeon  |DRM/other

--- Comment #4 from Michel Dänzer  ---
I recently ran into this with armagetronad. I've now confirmed that reverting
these changes fixes that as well, thanks Dave for saving me the time to track
this down!

Ville, any ideas what's going on / how to fix it? If not, please revert these
changes.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-- next part --
An HTML attachment was scrubbed...
URL: 

[Bug 93147] [regression bisected] Stuttering in games caused by commit 4dfd6486 "drm: Use vblank timestamps to guesstimate how many vblanks were missed"

[bugzilla-dae...@freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=93147

--- Comment #3 from Dave Witbrodt  ---
(In reply to Michel Dänzer from comment #2)
> Does running the affected apps with the environment variable
> LIBGL_DRI3_DISABLE=1 avoid the problem?

Yes, symptoms are completely eliminated in prboom-plus when using that
variable.

(I am currently using a kernel built from a local branch with fa4270d8 and
4dfd6486 reverted, but I have left my previous kernel installed.  I tested by
rebooting to the previous kernel, and running prboom-plus with and w/o the
variable set.)

-- 
You are receiving this mail because:
You are the assignee for the bug.
-- next part --
An HTML attachment was scrubbed...
URL: 

[Bug 93147] [regression bisected] Stuttering in games caused by commit 4dfd6486 "drm: Use vblank timestamps to guesstimate how many vblanks were missed"

[bugzilla-dae...@freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=93147

--- Comment #2 from Michel Dänzer  ---
Does running the affected apps with the environment variable
LIBGL_DRI3_DISABLE=1 avoid the problem?

-- 
You are receiving this mail because:
You are the assignee for the bug.
-- next part --
An HTML attachment was scrubbed...
URL: 

[Bug 93147] [regression bisected] Stuttering in games caused by commit 4dfd6486 "drm: Use vblank timestamps to guesstimate how many vblanks were missed"

[bugzilla-dae...@freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=93147

--- Comment #1 from Dave Witbrodt  ---
Created attachment 120188
  --> https://bugs.freedesktop.org/attachment.cgi?id=120188=edit
Xorg.0.log when running kernel built at "bad" commit

-- 
You are receiving this mail because:
You are the assignee for the bug.
-- next part --
An HTML attachment was scrubbed...
URL: 

[Bug 93147] [regression bisected] Stuttering in games caused by commit 4dfd6486 "drm: Use vblank timestamps to guesstimate how many vblanks were missed"

[bugzilla-dae...@freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=93147

Bug ID: 93147
   Summary: [regression bisected] Stuttering in games caused by
commit 4dfd6486 "drm: Use vblank timestamps to
guesstimate how many vblanks were missed"
   Product: DRI
   Version: DRI git
  Hardware: x86-64 (AMD64)
OS: Linux (All)
Status: NEW
  Severity: minor
  Priority: medium
 Component: DRM/Radeon
  Assignee: dri-devel at lists.freedesktop.org
  Reporter: dawitbro at sbcglobal.net
CC: ville.syrjala at linux.intel.com

Created attachment 120187
  --> https://bugs.freedesktop.org/attachment.cgi?id=120187=edit
dmesg from kernel built at "bad" commit

Building Linux from drm-next or drm-fixes gives me a kernel which causes what
appears to be stuttering/hesitations in test applications (particularly severe
in 'prboom-plus') which were not present in Linux 4.3.


HARDWARE
GPU:  Radeon HD 7850 (Pitcairn)
CPU:  64-bit AMD FX-8320E Eight-Core Processor
chipset:  AMD 990FX
 mboard:  ASUS M5A99FX PRO R2.0

SOFTWARE
xf86-video-ati:  7.6.99+ (git master at commit 10b7c3de, Nov 17 2015)
   xorg-server:  1.18.0+ (git master at commit 51984ddd, Nov 18 2015)
  mesa:  11.1.0+ (git master at commit d8c26969, Nov 20 2015)
libdrm:  2.4.65+ (git master at commit c3deddd9, Nov 10 2015)
kernel:  drm-fixes (at commit 2d591ab1, Nov 20 2015)
  distribution:  Debian unstable


Reproducing steps:

I apply future Radeon DRM and core DRM code to latest stable Linux kernels, and
I use several applications to test the results.  I use a less demanding game,
'prboom-plus', as a kind of canary; I use more demanding games like
'alien-arena' and 'torcs' as well.  All 3 of the games mentioned show a kind of
stuttering or hesitation if using the current git HEAD of drm-next or
drm-fixes, but 'prboom-plus' is affected to the point of being extremely
annoying, almost unplayable.

To reproduce, boot Linux 4.3 and play a few moments of 'prboom-plus'; then boot
a kernel built from drm-next or drm-fixes, then play 'prboom-plus' again.  The
effects are quite pronounced and obvious.  (The effects on 'alien-arena' and
'torcs' are much more subtle, and might easily go unnoticed, though someone
looking for a difference should easily be able to notice.)  None of the games
crash, but they are simply affected (to varying degrees) by a kind of
stuttering lag caused by the commit identified below by bisection.


Additional info:

I have reported this against DRI/Radeon but problem commit is touching code in
'drivers/gpu/drm/drm_irq.c', which does not seem limited to affecting only
Radeon.  (I am not a developer, so any comment I make about code can probably
be ignored.)  I would appreciate if the developers reading this would consider
whether I have reported the bug against the wrong product and/or component, and
make changes as appropriate!

In 'xorg.conf' I usually have "SwapbuffersWait" disabled with a stanza like
this:

Section "Device"
  Identifier"Configured Video Device"
  Driver"radeon"
  Option"DRI"  "3"
  Option"SwapbuffersWait"  "off"
EndSection

Changing the "SwapbuffersWait" setting has no effect on the symptoms of this
bug, however.


Bisection:

[I test future Radeon code in advance by cherry picking from drm-next and
drm-fixes changesets which touch Radeon DRM and core DRM.  Because of the
likelihood of error on my part, I never report bugs on one of these
Frankenstein kernels of my own making.  If I bisect a bug, I confirm that the
same buggy behavior is present upstream before reporting it.]

I first noticed this problem in early October, when I had made a local branch
from Linux 4.2.3 and cherry picked from DRM 4.3 and DRM 4.4 (up to cf648305). 
I noticed the bug described earlier immediately.  I didn't have much time to
troubleshoot, and assumed the error was my own, so I made a second local branch
from 4.2.3 and only cherry-picked from DRM 4.3 (up to 30c64664, my previously
working list of commits).

After Linux 4.3 was released, I tried again.  I made a local branch based on
4.3 and applied DRM 4.4 up to commit 5481c8fb.  The process went very smooth,
but the buggy behavior of my test programs (particularly 'prboom-plus') was
back again.  I decided to bisect my 4.3 + DRM 4.4 tree to identify where things
were going wrong.

The HEAD of my tree was clearly "bad" so I had a place to start.  I built Linux
4.3 without any cherry picks, and it was "good".  After 6 more builds, 'git
bisect' pointed at the following commit:

commit 4dfd64862ff852df7b1198d667dda778715ee88f
Author: Ville Syrjälä 
Date:   Mon Sep 14 22:43:51 2015 +0300

drm: Use vblank timestamps to guesstimate how many vblanks were missed

Since the problem could still have been only in my Frankenstein local tree, I
built