All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Christian König" <christian.koenig@amd.com>
To: Friedrich Vock <friedrich.vock@gmx.de>,
	amd-gfx@lists.freedesktop.org, dri-devel@lists.freedesktop.org,
	"Sharma, Shashank" <Shashank.Sharma@amd.com>,
	Alex Deucher <Alexander.Deucher@amd.com>
Cc: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>,
	Tatsuyuki Ishi <ishitatsuyuki@gmail.com>
Subject: Re: [PATCH v3 0/3] drm/amdgpu: Explicit sync for GEM VA operations
Date: Tue, 4 Feb 2025 14:36:13 +0100	[thread overview]
Message-ID: <2217a9b5-894a-4f1c-95f8-dbd6f18fc8b4@amd.com> (raw)
In-Reply-To: <e6330f10-a736-4ccd-8a0e-95e1899648d6@gmx.de>

Hi Friedrich,

adding Alex.

Am 04.02.25 um 13:32 schrieb Friedrich Vock:
> Hi,
>
> On 19.08.24 13:21, Christian König wrote:
>> Am 19.08.24 um 09:21 schrieb Friedrich Vock:
>>> In Vulkan, it is the application's responsibility to perform adequate
>>> synchronization before a sparse unmap, replace or BO destroy operation.
>>> This adds an option to AMDGPU_VA_OPs to disable redundant implicit sync
>>> that happens on sparse unmap or replace operations.
>>>
>>> This has seen a significant improvement in stutter in Forza Horizon 5
>>> and Forza Horizon 4. (As games that had significant issues in sparse
>>> binding related stutter).
>>
>> Looks pretty good, I have Shashank and his team working on the similar
>> patches ever since Bas initially came up with it since we need it for
>> user queues as well.
>>
>> Shashank can you take a look at the UAPI? Of hand looks pretty similar
>> to what we have done as well, doesn't it?
>>
>> For the internal implementation in the VM I'm currently working on a bug
>> fix for the KFD team, so this is subject to change anyway. Going to keep
>> this requirement here in mind while doing that, whatever implementation
>> we end up with we probably need to re-base it anyway.
>
> Bumping this again - it's been quite a while, what became of that KFD
> bugfix and the userqueue stuff? It'd be nice to finally make progress
> here, whether it's using the user queue interface you worked on or a
> re-spin of this. Maybe it's possible to split this off from the rest of
> the userqueue stuff and merge it beforehand if you're reasonably certain
> about how the uapi should look? Let me know.

That is merged into amd-staging-drm-next for quite a while now, but we 
only defined the interface and dropped all optimizations to initially 
get it upstream.

@Alex IIRC we pushed the KFD part upstream already, but the userqueue 
part is still waiting for the final firmware release, right?

Regards,
Christian.

>
> Thanks,
> Friedrich
>
>>
>> Regards,
>> Christian.
>>
>>>
>>> Userspace changes for this new version can be found at [1][2], and a
>>> kernel
>>> branch containing these patches can be found at [3].
>>>
>>> [1] https://gitlab.freedesktop.org/pixelcluster/drm/-/commits/vm-
>>> explicit-sync
>>> [2] https://gitlab.freedesktop.org/pixelcluster/mesa/-/commits/vm-
>>> explicit-sync
>>> [3] https://gitlab.freedesktop.org/pixelcluster/linux/-/commits/
>>> amdgpu-vm-explicit-sync
>>>
>>> v3 changes:
>>> - Rebased onto current amd-staging-drm-next
>>> - Added option to wait for drm_syncobjs instead of executing 
>>> immediately
>>>
>>> Tatsuyuki Ishi (3):
>>>    drm/amdgpu: Don't implicit sync PRT maps.
>>>    drm/amdgpu: Add optional explicit sync fences for GEM operations.
>>>    drm/amdgpu: Bump amdgpu driver version.
>>>
>>>   .../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c  |  2 +-
>>>   drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c       |  2 +-
>>>   drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c       |  3 +-
>>>   drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c       | 76 
>>> ++++++++++++++++---
>>>   drivers/gpu/drm/amd/amdgpu/amdgpu_object.h    | 23 +++++-
>>>   drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h     |  6 +-
>>>   drivers/gpu/drm/amd/amdgpu/amdgpu_umsch_mm.c  |  2 +-
>>>   drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c        | 68 +++++++++++------
>>>   drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h        | 30 ++++----
>>>   drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c    | 12 ++-
>>>   drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c     |  2 +-
>>>   drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c   |  9 +++
>>>   drivers/gpu/drm/amd/amdkfd/kfd_svm.c          | 18 ++---
>>>   include/uapi/drm/amdgpu_drm.h                 |  7 ++
>>>   14 files changed, 194 insertions(+), 66 deletions(-)
>>>
>>> -- 
>>> 2.46.0
>>>
>


  reply	other threads:[~2025-02-04 13:36 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-08-19  7:21 [PATCH v3 0/3] drm/amdgpu: Explicit sync for GEM VA operations Friedrich Vock
2024-08-19  7:21 ` [PATCH v3 1/3] drm/amdgpu: Don't implicit sync PRT maps Friedrich Vock
2024-08-19  7:21 ` [PATCH v3 2/3] drm/amdgpu: Add optional explicit sync fences for GEM operations Friedrich Vock
2024-08-19  7:21 ` [PATCH v3 3/3] drm/amdgpu: Bump amdgpu driver version Friedrich Vock
2024-08-19 11:21 ` [PATCH v3 0/3] drm/amdgpu: Explicit sync for GEM VA operations Christian König
2024-08-20 13:36   ` Sharma, Shashank
2025-02-04 12:32   ` Friedrich Vock
2025-02-04 13:36     ` Christian König [this message]
2025-02-04 13:58       ` Alex Deucher
2025-02-06 17:33         ` Friedrich Vock

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=2217a9b5-894a-4f1c-95f8-dbd6f18fc8b4@amd.com \
    --to=christian.koenig@amd.com \
    --cc=Alexander.Deucher@amd.com \
    --cc=Shashank.Sharma@amd.com \
    --cc=amd-gfx@lists.freedesktop.org \
    --cc=bas@basnieuwenhuizen.nl \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=friedrich.vock@gmx.de \
    --cc=ishitatsuyuki@gmail.com \
    /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.