From: bugzilla-daemon@freedesktop.org
To: dri-devel@lists.freedesktop.org
Subject: [Bug 97879] [amdgpu] Rocket League: long hangs (several seconds) when loading assets (models/textures/shaders?)
Date: Wed, 25 Jan 2017 17:02:08 +0000 [thread overview]
Message-ID: <bug-97879-502-qILaZksh6q@http.bugs.freedesktop.org/> (raw)
In-Reply-To: <bug-97879-502@http.bugs.freedesktop.org/>
[-- Attachment #1.1: Type: text/plain, Size: 1777 bytes --]
https://bugs.freedesktop.org/show_bug.cgi?id=97879
--- Comment #55 from Timothee Besset <ttimo@ttimo.net> ---
Rocket League uses an async I/O thread to stream in content in the background.
That thread works off of a queue and only does disk reads and decompression (it
doesn't issue any OpenGL calls for instance, that happens on a dedicated render
thread).
The CPU usage you are seeing out of Rocket League is this thread being busy
with decompression. This is generally normal and not an indication of a
problem.
However, when running with radeonsi, the async I/O thread is performing very
poorly. Much, much worse than with amdgpupro. To the point that the main thread
has to wait for async and causes a significant hitch (up to 4 seconds in some
logs I captured).
I tend to agree that this problem is not related to shader compilation. The
bumps from shader compilation are barely showing up in profiles. But it's
definitely made a lot worse by running on radeonsi. On the same system with
amdgpupro those same hitches can happen but they are in the 100/200ms range.
At this point of the investigation, and due to scheduling constraints, I am
likely only going to come back to this once Timothy Arceri has a shader cache
implementation available. I suspect this will not help significantly but let's
hope I'm wrong there.
In the meantime there are some log files from the async thread compared between
radeonsi and amdgpu-pro if someone feels like extracting some possibly useful
data:
https://www.dropbox.com/s/d73jk66uhfrhv8l/amdgpupro-compressedread-timings.log?dl=0
https://www.dropbox.com/s/1xfl3l24f5buzs0/radeonsi-compressedread-timings.log?dl=0
--
You are receiving this mail because:
You are on the CC list for the bug.
[-- Attachment #1.2: Type: text/html, Size: 2853 bytes --]
[-- Attachment #2: Type: text/plain, Size: 160 bytes --]
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel
next prev parent reply other threads:[~2017-01-25 17:02 UTC|newest]
Thread overview: 94+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <bug-97879-502@http.bugs.freedesktop.org/>
2016-09-20 23:28 ` [Bug 97879] [amdgpu] Rocket League: long hangs (several seconds) when loading assets (models/textures/shaders?) bugzilla-daemon
2016-09-21 2:05 ` bugzilla-daemon
2016-09-21 8:05 ` bugzilla-daemon
2016-09-21 14:59 ` bugzilla-daemon
2016-09-21 18:04 ` bugzilla-daemon
2016-09-21 19:01 ` bugzilla-daemon
2016-09-21 19:02 ` bugzilla-daemon
2016-09-21 19:02 ` bugzilla-daemon
2016-09-21 19:07 ` bugzilla-daemon
2016-09-21 19:15 ` bugzilla-daemon
2016-09-22 12:27 ` bugzilla-daemon
2016-09-22 17:47 ` bugzilla-daemon
2016-09-23 1:15 ` bugzilla-daemon
2016-09-24 15:19 ` bugzilla-daemon
2016-09-24 23:58 ` bugzilla-daemon
2016-09-24 23:58 ` bugzilla-daemon
2016-09-25 0:00 ` bugzilla-daemon
2016-09-25 0:09 ` bugzilla-daemon
2016-09-25 0:10 ` bugzilla-daemon
2016-09-25 0:11 ` bugzilla-daemon
2016-09-26 11:12 ` bugzilla-daemon
2016-09-26 18:07 ` bugzilla-daemon
2016-09-26 18:09 ` bugzilla-daemon
2016-09-27 8:49 ` bugzilla-daemon
2016-09-27 9:11 ` bugzilla-daemon
2016-09-27 15:23 ` bugzilla-daemon
2016-09-27 15:26 ` bugzilla-daemon
2016-09-27 15:47 ` bugzilla-daemon
2016-09-27 16:07 ` bugzilla-daemon
2016-09-27 18:45 ` bugzilla-daemon
2016-09-28 2:40 ` bugzilla-daemon
2016-10-15 11:16 ` bugzilla-daemon
2016-10-15 14:51 ` bugzilla-daemon
2016-10-15 15:35 ` bugzilla-daemon
2016-10-15 16:00 ` bugzilla-daemon
2016-10-16 11:55 ` bugzilla-daemon
2016-10-16 12:41 ` bugzilla-daemon
2016-10-16 14:29 ` bugzilla-daemon
2016-10-25 14:52 ` bugzilla-daemon
2016-10-28 12:49 ` bugzilla-daemon
2016-10-31 3:32 ` bugzilla-daemon
2016-10-31 11:51 ` bugzilla-daemon
2016-11-02 19:13 ` bugzilla-daemon
2016-11-11 18:27 ` bugzilla-daemon
2016-11-11 23:49 ` bugzilla-daemon
2016-11-15 17:25 ` bugzilla-daemon
2016-11-15 20:38 ` bugzilla-daemon
2017-01-06 22:23 ` bugzilla-daemon
2017-01-09 7:12 ` bugzilla-daemon
2017-01-10 4:50 ` bugzilla-daemon
2017-01-10 17:02 ` bugzilla-daemon
2017-01-13 22:22 ` bugzilla-daemon
2017-01-24 17:59 ` bugzilla-daemon
2017-01-24 18:20 ` bugzilla-daemon
2017-01-25 1:08 ` bugzilla-daemon
2017-01-25 17:02 ` bugzilla-daemon [this message]
2017-01-25 17:34 ` bugzilla-daemon
2017-01-26 2:28 ` bugzilla-daemon
2017-01-26 4:35 ` bugzilla-daemon
2017-01-26 14:17 ` bugzilla-daemon
2017-01-26 14:22 ` bugzilla-daemon
2017-02-07 10:40 ` bugzilla-daemon
2017-02-07 10:41 ` bugzilla-daemon
2017-02-07 12:21 ` bugzilla-daemon
2017-02-07 13:32 ` bugzilla-daemon
2017-02-07 17:07 ` bugzilla-daemon
2017-02-07 18:30 ` bugzilla-daemon
2017-02-07 21:34 ` bugzilla-daemon
2017-02-08 1:08 ` bugzilla-daemon
2017-02-08 12:39 ` bugzilla-daemon
2017-02-08 18:16 ` bugzilla-daemon
2017-02-08 18:23 ` bugzilla-daemon
2017-02-08 18:36 ` bugzilla-daemon
2017-02-08 18:37 ` bugzilla-daemon
2017-02-09 12:48 ` bugzilla-daemon
2017-02-10 0:59 ` bugzilla-daemon
2017-02-10 1:05 ` bugzilla-daemon
2017-02-10 2:08 ` bugzilla-daemon
2017-02-10 12:38 ` bugzilla-daemon
2017-02-10 18:10 ` bugzilla-daemon
2017-02-10 20:13 ` bugzilla-daemon
2017-02-10 22:23 ` bugzilla-daemon
2017-02-12 15:08 ` bugzilla-daemon
2017-02-12 16:45 ` bugzilla-daemon
2017-02-13 3:23 ` bugzilla-daemon
2017-02-13 9:22 ` bugzilla-daemon
2017-02-13 13:13 ` bugzilla-daemon
2017-02-13 13:19 ` bugzilla-daemon
2017-02-13 14:34 ` bugzilla-daemon
2017-02-13 19:53 ` bugzilla-daemon
2017-02-14 0:08 ` bugzilla-daemon
2017-02-14 1:12 ` bugzilla-daemon
2017-02-14 9:51 ` bugzilla-daemon
2017-02-16 20:28 ` bugzilla-daemon
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=bug-97879-502-qILaZksh6q@http.bugs.freedesktop.org/ \
--to=bugzilla-daemon@freedesktop.org \
--cc=dri-devel@lists.freedesktop.org \
/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.