* [Bug 103697] [regression] shared memory size 64k -> 32k?
[not found] <bug-103697-502@http.bugs.freedesktop.org/>
@ 2017-11-12 15:09 ` bugzilla-daemon
2017-11-12 22:39 ` bugzilla-daemon
` (4 subsequent siblings)
5 siblings, 0 replies; 6+ messages in thread
From: bugzilla-daemon @ 2017-11-12 15:09 UTC (permalink / raw)
To: dri-devel
[-- Attachment #1.1: Type: text/plain, Size: 660 bytes --]
https://bugs.freedesktop.org/show_bug.cgi?id=103697
Kenneth Graunke <kenneth@whitecape.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Assignee|mesa-dev@lists.freedesktop. |dri-devel@lists.freedesktop
|org |.org
Component|glsl-compiler |Drivers/Gallium/radeonsi
QA Contact|intel-3d-bugs@lists.freedes |dri-devel@lists.freedesktop
|ktop.org |.org
--
You are receiving this mail because:
You are the assignee for the bug.
[-- Attachment #1.2: Type: text/html, Size: 1560 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
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug 103697] [regression] shared memory size 64k -> 32k?
[not found] <bug-103697-502@http.bugs.freedesktop.org/>
2017-11-12 15:09 ` [Bug 103697] [regression] shared memory size 64k -> 32k? bugzilla-daemon
@ 2017-11-12 22:39 ` bugzilla-daemon
2017-11-13 15:33 ` bugzilla-daemon
` (3 subsequent siblings)
5 siblings, 0 replies; 6+ messages in thread
From: bugzilla-daemon @ 2017-11-12 22:39 UTC (permalink / raw)
To: dri-devel
[-- Attachment #1.1: Type: text/plain, Size: 728 bytes --]
https://bugs.freedesktop.org/show_bug.cgi?id=103697
Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl> changed:
What |Removed |Added
----------------------------------------------------------------------------
Resolution|--- |NOTABUG
Status|NEW |RESOLVED
--- Comment #1 from Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl> ---
This is working as intended, the limit has been 32k for a while now.
What changed was that we added error checking in
https://cgit.freedesktop.org/mesa/mesa/commit/?id=a2c8812f919c59933605c5942d6613e14ec8b3d1
--
You are receiving this mail because:
You are the assignee for the bug.
[-- Attachment #1.2: Type: text/html, Size: 2348 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
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug 103697] [regression] shared memory size 64k -> 32k?
[not found] <bug-103697-502@http.bugs.freedesktop.org/>
2017-11-12 15:09 ` [Bug 103697] [regression] shared memory size 64k -> 32k? bugzilla-daemon
2017-11-12 22:39 ` bugzilla-daemon
@ 2017-11-13 15:33 ` bugzilla-daemon
2017-11-28 14:54 ` bugzilla-daemon
` (2 subsequent siblings)
5 siblings, 0 replies; 6+ messages in thread
From: bugzilla-daemon @ 2017-11-13 15:33 UTC (permalink / raw)
To: dri-devel
[-- Attachment #1.1: Type: text/plain, Size: 667 bytes --]
https://bugs.freedesktop.org/show_bug.cgi?id=103697
--- Comment #2 from O Heid <oliver.heid@h-next.de> ---
I checked back with AMD and they say that there is hardware support for 64kB
shared memory. This actually gets exposed in OpenCL and ROCm/HC.
The 2016 AMD GCN3 manual wrongly states it is only 32kB, but the AMD Vega ISA
reference guide has the correct 64kB LDS size.
So it seems the restriction is due to the mesa software stack, not due to
hardware. Could you please raise the bar back to 64kB? I wouldn't ask if it
wouldn't hit me with with a 2x performance penalty...
--
You are receiving this mail because:
You are the assignee for the bug.
[-- Attachment #1.2: Type: text/html, Size: 1496 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
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug 103697] [regression] shared memory size 64k -> 32k?
[not found] <bug-103697-502@http.bugs.freedesktop.org/>
` (2 preceding siblings ...)
2017-11-13 15:33 ` bugzilla-daemon
@ 2017-11-28 14:54 ` bugzilla-daemon
2017-11-28 16:41 ` bugzilla-daemon
2017-12-19 8:45 ` bugzilla-daemon
5 siblings, 0 replies; 6+ messages in thread
From: bugzilla-daemon @ 2017-11-28 14:54 UTC (permalink / raw)
To: dri-devel
[-- Attachment #1.1: Type: text/plain, Size: 333 bytes --]
https://bugs.freedesktop.org/show_bug.cgi?id=103697
--- Comment #3 from Nicolai Hähnle <nhaehnle@gmail.com> ---
Bas explained what's going on correctly. We could consider raising the limit
for compute shaders though. May ask what your use case is?
--
You are receiving this mail because:
You are the assignee for the bug.
[-- Attachment #1.2: Type: text/html, Size: 1168 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
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug 103697] [regression] shared memory size 64k -> 32k?
[not found] <bug-103697-502@http.bugs.freedesktop.org/>
` (3 preceding siblings ...)
2017-11-28 14:54 ` bugzilla-daemon
@ 2017-11-28 16:41 ` bugzilla-daemon
2017-12-19 8:45 ` bugzilla-daemon
5 siblings, 0 replies; 6+ messages in thread
From: bugzilla-daemon @ 2017-11-28 16:41 UTC (permalink / raw)
To: dri-devel
[-- Attachment #1.1: Type: text/plain, Size: 737 bytes --]
https://bugs.freedesktop.org/show_bug.cgi?id=103697
--- Comment #4 from O Heid <oliver.heid@h-next.de> ---
Many thanks for your consideration - That would be great!
Our application is real time image processing and display. The main computing
effort is delay-and-sum beamforming.
Source and destination data sets are much larger than LDS so we need to do
block processing. This causes additional LDS <-> GPURAM traffic essentially
inversely proportional to block (LDS) size. A 32k vs 64k limit (and thus twice
the RAM traffic) therefore causes quite a performance hit.
Anyway: Mesa's OpenGL performance on AMD Radeons is quite amazing. Great job!
--
You are receiving this mail because:
You are the assignee for the bug.
[-- Attachment #1.2: Type: text/html, Size: 1572 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
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug 103697] [regression] shared memory size 64k -> 32k?
[not found] <bug-103697-502@http.bugs.freedesktop.org/>
` (4 preceding siblings ...)
2017-11-28 16:41 ` bugzilla-daemon
@ 2017-12-19 8:45 ` bugzilla-daemon
5 siblings, 0 replies; 6+ messages in thread
From: bugzilla-daemon @ 2017-12-19 8:45 UTC (permalink / raw)
To: dri-devel
[-- Attachment #1.1: Type: text/plain, Size: 378 bytes --]
https://bugs.freedesktop.org/show_bug.cgi?id=103697
--- Comment #5 from O Heid <oliver.heid@h-next.de> ---
Created attachment 136272
--> https://bugs.freedesktop.org/attachment.cgi?id=136272&action=edit
Enable 64kB LDS on AMD Radeon GPUs
applies to mesa-17.3.0 and its release candidates
--
You are receiving this mail because:
You are the assignee for the bug.
[-- Attachment #1.2: Type: text/html, Size: 1453 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
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2017-12-19 8:45 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
[not found] <bug-103697-502@http.bugs.freedesktop.org/>
2017-11-12 15:09 ` [Bug 103697] [regression] shared memory size 64k -> 32k? bugzilla-daemon
2017-11-12 22:39 ` bugzilla-daemon
2017-11-13 15:33 ` bugzilla-daemon
2017-11-28 14:54 ` bugzilla-daemon
2017-11-28 16:41 ` bugzilla-daemon
2017-12-19 8:45 ` bugzilla-daemon
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).