we
would need to account for the worst case.
Hopefully the requirements and approach we described make
sense, we're
looking forward to your feedback and suggestions.
Thanks!
- Pierre-Loup
Sincerely yours,
Serguei Sagalovitch
From: Andres Rodriguez
Sent: December 16, 2016 10:00 PM
To:
. Griffais;
Zhang, Hawking
Subject: Re: [RFC] Mechanism for high priority scheduling in amdgpu
Hey Christian,
But yes, in general you don't want another compositor in the way, so we'll be
acquiring the HMD display directly, separate from any desktop or display server.
Assuming that the
ng
Subject: Re: [RFC] Mechanism for high priority scheduling in amdgpu
John,
One comment: When Andres is talking about compute he is talking about Vulkan
compute not OpenCL one and it means that it is not HSA path.
Sincerely yours,
Serguei Sagalovitch
From: Bridgman, John
Sent: December 23,
gt;>>>>>> On 2016年12月18日 06:05, Pierre-Loup A. Griffais wrote:
>>>>>>>>>>>>>>
>>>>>>>>>>>>>>> Hi Serguei,
>>>>>>>>>>>>>>>
>>>>>>>
, David(ChunMing); Huan, Alvin; Mao, David;
Sagalovitch, Serguei; amd-gfx@lists.freedesktop.org; Andres Rodriguez;
Pierre-Loup A. Griffais; Zhang, Hawking
Subject: Re: [RFC] Mechanism for high priority scheduling in amdgpu
Excellent, thanks. Agree that it is not a complete solution, just a good
edesktop.org; Andres Rodriguez;
Pierre-Loup A. Griffais; Zhang, Hawking
Subject: Re: [RFC] Mechanism for high priority scheduling in amdgpu
I'm actually testing that out that today.
Example prototype patch:
https://github.com/lostgoat/linux/commit/c9d88d409d8655d63aa8386edc66687a2ba64a12
My goal i
;>>>>> sharing,
>>>>>>>>>>>>>>> since it's achieved by using whatever is the latest, most
>>>>>>>>>>>>>>> up-to-date
>>>>>>>>>>>>>>> eye images that have alr
variance then that would be
troublesome and we
would need to account for the worst case.
Hopefully the requirements and approach we described make
sense, we're
looking forward to your feedback and suggestions.
Thanks!
- Pierre-Loup
Sincerely yours,
Serguei Sagalovitch
From: Andres Rodriguez
;>>>>>>>>>>> On 12/16/2016 09:05 PM, Sagalovitch, Serguei wrote:
>>>>>>>>>>>>>>
>>>>>>>>>>>>>>> Andres,
>>>>>>>>>>>>>>>
>>>>>>>>>>>
res
____________
From: Sagalovitch, Serguei [serguei.sagalovi...@amd.com]
Sent: Friday, December 16, 2016 9:13 PM
To: Andres Rodriguez; amd-gfx@lists.freedesktop.org
Subject: Re: [RFC] Mechanism for high priority scheduling
in amdgpu
Hi Andres,
Please see inline (as [S
feedback and suggestions.
Thanks!
- Pierre-Loup
Sincerely yours,
Serguei Sagalovitch
From: Andres Rodriguez
Sent: December 16, 2016 10:00 PM
To: Sagalovitch, Serguei; amd-gfx@lists.freedesktop.org
Subject: RE: [RFC] Mechanism for high priority scheduling
in amdgpu
Hey Serguei,
[Serguei] No.
hanks!
- Pierre-Loup
Sincerely yours,
Serguei Sagalovitch
From: Andres Rodriguez
Sent: December 16, 2016 10:00 PM
To: Sagalovitch, Serguei; amd-gfx@lists.freedesktop.org
Subject: RE: [RFC] Mechanism for high priority scheduling
in amdgpu
Hey Serguei,
[Serguei] No. I mean pipe :-) as
sense, we're
looking forward to your feedback and suggestions.
Thanks!
- Pierre-Loup
Sincerely yours,
Serguei Sagalovitch
From: Andres Rodriguez
Sent: December 16, 2016 10:00 PM
To: Sagalovitch, Serguei; amd-gfx@lists.freedesktop.org
Subject: RE: [RFC] Mechanism for high priority sch
: Andres Rodriguez
Sent: December 16, 2016 10:00 PM
To: Sagalovitch, Serguei; amd-gfx@lists.freedesktop.org
Subject: RE: [RFC] Mechanism for high priority scheduling in
amdgpu
Hey Serguei,
[Serguei] No. I mean pipe :-) as MEC define it. As far as I
understand (by simplifying)
some scheduling is
tch
From: Andres Rodriguez
Sent: December 16, 2016 10:00 PM
To: Sagalovitch, Serguei; amd-gfx@lists.freedesktop.org
Subject: RE: [RFC] Mechanism for high priority scheduling in
amdgpu
Hey Serguei,
[Serguei] No. I mean pipe :-) as MEC define it. As far as I
understand (by simplifying)
some scheduling i
- Pierre-Loup
Sincerely yours,
Serguei Sagalovitch
From: Andres Rodriguez
Sent: December 16, 2016 10:00 PM
To: Sagalovitch, Serguei; amd-gfx@lists.freedesktop.org
Subject: RE: [RFC] Mechanism for high priority scheduling in
amdgpu
Hey Serguei,
[Serguei] No. I mean pipe :-) as MEC de
Sagalovitch
From: Andres Rodriguez
Sent: December 16, 2016 10:00 PM
To: Sagalovitch, Serguei; amd-gfx@lists.freedesktop.org
Subject: RE: [RFC] Mechanism for high priority scheduling in
amdgpu
Hey Serguei,
[Serguei] No. I mean pipe :-) as MEC define it. As far as I
understand (by simplifying)
Hey Guys,
One particular piece I'd like to discuss is how to get around the below issue:
> Known current obstacles:
>
>
> The SQ is currently programmed to disregard the HQD priorities, and instead
> it picks
> jobs at random. Settings from the shader itself are also di
tever is the latest, most up-to-date
>>>>>> eye images that have already been sent by the client application,
>>>>>> which are already available to use without additional
>>>>>> synchronization.
>>>>>>
>>>
ierre-Loup
Sincerely yours,
Serguei Sagalovitch
From: Andres Rodriguez
Sent: December 16, 2016 10:00 PM
To: Sagalovitch, Serguei; amd-gfx@lists.freedesktop.org
Subject: RE: [RFC] Mechanism for high priority scheduling in amdgpu
Hey Serguei,
[Serguei] No. I mean pipe :-) as MEC define it. As fa
erguei [serguei.sagalovi...@amd.com]
Sent: Friday, December 16, 2016 9:13 PM
To: Andres Rodriguez; amd-gfx@lists.freedesktop.org
Subject: Re: [RFC] Mechanism for high priority scheduling in amdgpu
Hi Andres,
Please see inline (as [Serguei])
Sincerely yours,
Serguei Sagalovitch
From: Andres Ro
looking forward to your feedback and suggestions.
Thanks!
- Pierre-Loup
Sincerely yours,
Serguei Sagalovitch
From: Andres Rodriguez
Sent: December 16, 2016 10:00 PM
To: Sagalovitch, Serguei; amd-gfx@lists.freedesktop.org
Subject: RE: [RFC] Mechanism for high priority scheduling in amdgpu
Hey Serguei
sense, we're
looking forward to your feedback and suggestions.
Thanks!
- Pierre-Loup
Sincerely yours,
Serguei Sagalovitch
From: Andres Rodriguez
Sent: December 16, 2016 10:00 PM
To: Sagalovitch, Serguei; amd-gfx@lists.freedesktop.org
Subject: RE: [RFC] Mechanism for high priority schedu
rd to your feedback and suggestions.
Thanks!
- Pierre-Loup
Sincerely yours,
Serguei Sagalovitch
From: Andres Rodriguez
Sent: December 16, 2016 10:00 PM
To: Sagalovitch, Serguei; amd-gfx@lists.freedesktop.org
Subject: RE: [RFC] Mechanism for high priority scheduling in amdgpu
Hey Serguei,
[
Sent: December 16, 2016 10:00 PM
To: Sagalovitch, Serguei; amd-gfx@lists.freedesktop.org
Subject: RE: [RFC] Mechanism for high priority scheduling in amdgpu
Hey Serguei,
[Serguei] No. I mean pipe :-) as MEC define it. As far as I
understand (by simplifying)
some scheduling is per pipe. I know about
iguez
Sent: December 16, 2016 10:00 PM
To: Sagalovitch, Serguei; amd-gfx@lists.freedesktop.org
Subject: RE: [RFC] Mechanism for high priority scheduling in amdgpu
Hey Serguei,
[Serguei] No. I mean pipe :-) as MEC define it. As far as I
understand (by simplifying)
some scheduling is per pipe. I know
tions.
Thanks!
- Pierre-Loup
Sincerely yours,
Serguei Sagalovitch
From: Andres Rodriguez
Sent: December 16, 2016 10:00 PM
To: Sagalovitch, Serguei; amd-gfx@lists.freedesktop.org
Subject: RE: [RFC] Mechanism for high priority scheduling in amdgpu
Hey Serguei,
[Serguei] No. I mean pipe
ation).
Sincerely yours,
Serguei Sagalovitch
From: Andres Rodriguez
Sent: December 16, 2016 10:00 PM
To: Sagalovitch, Serguei; amd-gfx@lists.freedesktop.org
Subject: RE: [RFC] Mechanism for high priority scheduling in amdgpu
Hey Serguei,
> [Serguei] No. I mean pipe :-) as MEC d
Hi Andres,
Please see inline (as [Serguei])
Sincerely yours,
Serguei Sagalovitch
From: Andres Rodriguez
Sent: December 16, 2016 8:29 PM
To: Sagalovitch, Serguei; amd-gfx@lists.freedesktop.org
Subject: RE: [RFC] Mechanism for high priority scheduling in amdgpu
Hi Serguei,
Thanks for the
: Friday, December 16, 2016 9:13 PM
To: Andres Rodriguez; amd-gfx@lists.freedesktop.org
Subject: Re: [RFC] Mechanism for high priority scheduling in amdgpu
Hi Andres,
Please see inline (as [Serguei])
Sincerely yours,
Serguei Sagalovitch
From: Andres Rodriguez
Sent: December 16, 2016 8:29 PM
To:
Andres,
Quick comments:
1) To minimize "bubbles", etc. we need to "force" CU assignments/binding
to high-priority queue when it will be in use and "free" them later
(we do not want forever take CUs from e.g. graphic task to degrade graphics
performance).
Otherwise we could have scenario whe
for high priority scheduling in amdgpu
Andres,
Quick comments:
1) To minimize "bubbles", etc. we need to "force" CU assignments/binding
to high-priority queue when it will be in use and "free" them later
(we do not want forever take CUs from e.g. graphic task t
32 matches
Mail list logo