* [Piketon bisected]TIMEOUT when running gem_tiled_blits
@ 2011-02-11 8:02 Meng, Mengmeng
2011-02-11 9:15 ` Chris Wilson
0 siblings, 1 reply; 3+ messages in thread
From: Meng, Mengmeng @ 2011-02-11 8:02 UTC (permalink / raw)
To: Intel-gfx@lists.freedesktop.org; +Cc: Meng, Mengmeng
[-- Attachment #1.1: Type: text/plain, Size: 1472 bytes --]
Hi,
There is something wrong about gem_tiled_blits. Timeout when running Intel_gpu_tools/gem_tiled_blits.
It's case regression. By bisected,371f87f2d825900484e34a8ec78272b9e62cbc02 is the first bad commit.
Information about this commit:
Commit 371f87f2d825900484e34a8ec78272b9e62cbc02
Author: Chris Wilson <chris@chris-wilson.co.uk<mailto:chris@chris-wilson.co.uk>>
Date: Tue Feb 1 10:53:57 2011 +0000
Exercise the whole aperture with tiled blits
After full-gtt, gem_tiled_blits doesn't allocate enough to force
eviction. So query the total aperture and accommodate.
Also introduce a similar test that utilizes fences rather than
use the BLT to perform the tiling and detiling.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk<mailto:chris@chris-wilson.co.uk>>
System Environment:
--------------------------------------------------------------------------
Libdrm: (master)2.4.23-6-g550fe2ca3b29ad2191eab4fdfbed9ed21e25492d
Mesa: (master)fc9170d0cf1c7a7dcd580e9da17742a6ab68bd1b
Xserver: (master)xorg-server-1.9.99.901-197-gea1ffd3e60bdcedbec5a6f28929f8677bf45d450
Xf86_video_intel: (master)2.14.0-27-g90c56ced00f8b951a1a72885cac8eba956a006bf
Libva: (master)2b7850b35a51c59a6692781d5e40e4bb608b6344
Kerenl: (drm-intel-next)9db4a9c7b2a3bd5b4952846bc0c2f58daa80ddd7
Meng
[-- Attachment #1.2: Type: text/html, Size: 6408 bytes --]
[-- Attachment #2: Type: text/plain, Size: 159 bytes --]
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [Piketon bisected]TIMEOUT when running gem_tiled_blits
2011-02-11 8:02 [Piketon bisected]TIMEOUT when running gem_tiled_blits Meng, Mengmeng
@ 2011-02-11 9:15 ` Chris Wilson
2011-02-11 10:47 ` Meng, Mengmeng
0 siblings, 1 reply; 3+ messages in thread
From: Chris Wilson @ 2011-02-11 9:15 UTC (permalink / raw)
To: Intel-gfx@lists.freedesktop.org; +Cc: Meng, Mengmeng
On Fri, 11 Feb 2011 16:02:45 +0800, "Meng, Mengmeng" <mengmeng.meng@intel.com> wrote:
> Hi,
> There is something wrong about gem_tiled_blits. Timeout when running Intel_gpu_tools/gem_tiled_blits.
> It's case regression. By bisected,371f87f2d825900484e34a8ec78272b9e62cbc02 is the first bad commit.
Takes around 30s to run the exercise on my i3. The fun part is that it
tries to allocate 3+ GiB of memory in order to cause the aperture
thrashing and so test evicting tiled buffers. If you don't have that much
memory available, then you get swap thrashing or even the dreaded OOM.
[I thought about limiting it to total available memory...]
Is your machine making any progress? Did the kernel send any warnings over
netconsole? How much memory do you have and what is your timeout?
-Chris
--
Chris Wilson, Intel Open Source Technology Centre
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [Piketon bisected]TIMEOUT when running gem_tiled_blits
2011-02-11 9:15 ` Chris Wilson
@ 2011-02-11 10:47 ` Meng, Mengmeng
0 siblings, 0 replies; 3+ messages in thread
From: Meng, Mengmeng @ 2011-02-11 10:47 UTC (permalink / raw)
To: Chris Wilson, Intel-gfx@lists.freedesktop.org; +Cc: Meng, Mengmeng
My machine does not made progress. There is no warnings over netconsole. MemTotal in Piketon is about 1.8G. It takes about
5s when running gem_tiled_blits normally. Timeout means 90s,even exist in process continuously.
-----Original Message-----
From: Chris Wilson [mailto:chris@chris-wilson.co.uk]
Sent: Friday, February 11, 2011 5:15 PM
To: Meng, Mengmeng; Intel-gfx@lists.freedesktop.org
Cc: Meng, Mengmeng
Subject: Re: [Intel-gfx] [Piketon bisected]TIMEOUT when running gem_tiled_blits
On Fri, 11 Feb 2011 16:02:45 +0800, "Meng, Mengmeng" <mengmeng.meng@intel.com> wrote:
> Hi,
> There is something wrong about gem_tiled_blits. Timeout when running Intel_gpu_tools/gem_tiled_blits.
> It's case regression. By bisected,371f87f2d825900484e34a8ec78272b9e62cbc02 is the first bad commit.
Takes around 30s to run the exercise on my i3. The fun part is that it
tries to allocate 3+ GiB of memory in order to cause the aperture
thrashing and so test evicting tiled buffers. If you don't have that much
memory available, then you get swap thrashing or even the dreaded OOM.
[I thought about limiting it to total available memory...]
Is your machine making any progress? Did the kernel send any warnings over
netconsole? How much memory do you have and what is your timeout?
-Chris
--
Chris Wilson, Intel Open Source Technology Centre
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2011-02-11 10:47 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2011-02-11 8:02 [Piketon bisected]TIMEOUT when running gem_tiled_blits Meng, Mengmeng
2011-02-11 9:15 ` Chris Wilson
2011-02-11 10:47 ` Meng, Mengmeng
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.