AMD-GFX Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
To: "Christian König" <christian.koenig@amd.com>,
	"Li, Yunxiang (Teddy)" <Yunxiang.Li@amd.com>,
	"dri-devel@lists.freedesktop.org"
	<dri-devel@lists.freedesktop.org>,
	"amd-gfx@lists.freedesktop.org" <amd-gfx@lists.freedesktop.org>
Cc: "Deucher, Alexander" <Alexander.Deucher@amd.com>
Subject: Re: [PATCH v5 4/4] drm/amdgpu: track bo memory stats at runtime
Date: Thu, 24 Oct 2024 09:29:21 +0100	[thread overview]
Message-ID: <50803e40-eaaf-4301-8459-e1b465b2d6f7@igalia.com> (raw)
In-Reply-To: <c80eb8ff-3965-4036-b763-20c4c2550e04@amd.com>


On 23/10/2024 13:56, Christian König wrote:
> Am 23.10.24 um 14:24 schrieb Tvrtko Ursulin:
>> [SNIP]
>>>> To fold or not the special placements (GWS, GDS & co) is also 
>>>> tangential. In my patch I just preserved the legacy behaviour so it 
>>>> can easily be tweaked on top.
>>>
>>> Yeah, but again the original behavior is completely broken.
>>>
>>> GWS, GDS and OA are counted in blocks of HW units (multiplied by 
>>> PAGE_SIZE IIRC to avoid some GEM&TTM warnings).
>>>
>>> When you accumulate that anywhere in the memory stats then that is 
>>> just completely off.
>>
>> Ooops. :) Are they backed by some memory though, be it system or VRAM?
> 
> GDS is an internal 4 or 64KiB memory block which is only valid while 
> shaders are running. It is used to communicate stuff between different 
> shader stages and not even CPU accessible.
> 
> GWS and OA are not even memory, those are just HW blocks which implement 
> a fixed function.
> 
> IIRC most HW generation have 16 of each and when setting up the 
> application virtual address space you can specify how many will be used 
> by the application.

I see, thank you! Though I could have bothered to look in the code or 
even instrument at runtime too.

I agree removing it from system is correct. If wanted and/or desirable 
some or all could be exported as different memory regions even. DRM 
fdinfo specs already allows that. Like:

drm-total-vram: ...
drm-total-gds: ...
drm-total-oa: ...

Etc.

Regards,

Tvrtko

  reply	other threads:[~2024-10-25 12:48 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-10-18 13:33 [PATCH v5 0/4] rework bo mem stats tracking Yunxiang Li
2024-10-18 13:33 ` [PATCH v5 1/4] drm/amdgpu: remove unused function parameter Yunxiang Li
2024-10-22  6:58   ` Christian König
2024-10-18 13:33 ` [PATCH v5 2/4] drm/amdgpu: make drm-memory-* report resident memory Yunxiang Li
2024-10-18 15:39   ` Tvrtko Ursulin
2024-10-22  7:00   ` Christian König
2024-10-18 13:33 ` [PATCH v5 3/4] drm/amdgpu: stop tracking visible memory stats Yunxiang Li
2024-10-22  7:22   ` Christian König
2024-10-18 13:33 ` [PATCH v5 4/4] drm/amdgpu: track bo memory stats at runtime Yunxiang Li
2024-10-22  7:43   ` Christian König
2024-10-22 15:17     ` Li, Yunxiang (Teddy)
2024-10-22 16:24       ` Christian König
2024-10-22 16:46         ` Li, Yunxiang (Teddy)
2024-10-22 17:06           ` Christian König
2024-10-22 17:09             ` Li, Yunxiang (Teddy)
2024-10-23  6:34               ` Christian König
2024-10-23  7:33             ` Tvrtko Ursulin
2024-10-23  7:38         ` Tvrtko Ursulin
2024-10-23  9:14           ` Christian König
2024-10-23 11:37             ` Tvrtko Ursulin
2024-10-23 12:12               ` Christian König
2024-10-23 12:24                 ` Tvrtko Ursulin
2024-10-23 12:56                   ` Christian König
2024-10-24  8:29                     ` Tvrtko Ursulin [this message]
2024-10-23 13:31                   ` Li, Yunxiang (Teddy)
2024-10-23 13:40                     ` Li, Yunxiang (Teddy)
2024-10-23 14:27                     ` Tvrtko Ursulin

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=50803e40-eaaf-4301-8459-e1b465b2d6f7@igalia.com \
    --to=tvrtko.ursulin@igalia.com \
    --cc=Alexander.Deucher@amd.com \
    --cc=Yunxiang.Li@amd.com \
    --cc=amd-gfx@lists.freedesktop.org \
    --cc=christian.koenig@amd.com \
    --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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox