AMD-GFX Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: John Olender <john.olender@gmail.com>
To: "Christian König" <christian.koenig@amd.com>,
	"Alex Deucher" <alexdeucher@gmail.com>
Cc: amd-gfx@lists.freedesktop.org, arunpravin.paneerselvam@amd.com
Subject: Re: [RFC PATCH 2/2] drm/amdgpu/uvd: Ensure vcpu bos are within the uvd segment
Date: Thu, 5 Jun 2025 05:21:47 -0400	[thread overview]
Message-ID: <450fa7e8-5d91-46e8-8b01-1195fe437a7a@gmail.com> (raw)
In-Reply-To: <b09012e2-f361-46b5-afbf-313334fad69a@amd.com>

On 5/2/25 4:36 AM, Christian König wrote:
>> I think we can just make this VRAM only.  Or something like:
>> adev->uvd.address_64_bit ? AMDGPU_GEM_DOMAIN_GTT : AMDGPU_GEM_DOMAIN_VRAM
> Yeah completely agree. It's a good catch, but the solution is incorrect.
> 
> On the older UVD MC interface the FW needs to be in VRAM or the validation fails. If it's inside the window for the message and fence is actually irrelevant.
> 
> So something like AMDGPU_GEM_DOMAIN_VRAM | (adev->uvd.address_64_bit ? AMDGPU_GEM_DOMAIN_GTT : 0) would be correct I think.
> 
>> If that fixes it, this should be tagged with:
>> Fixes: 58ab2c08d708 ("drm/amdgpu: use VRAM|GTT for a bunch of kernel
>> allocations")
> And CC stable I think.
> 
> Regards,
> Christian.
> 
>> Alex

The simplified patch needs this change to the vcpu bo domain.  Would you
prefer this change as a separate patch?

Thanks,
John

  parent reply	other threads:[~2025-06-05  9:22 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-04-29 11:24 [RFC PATCH 0/2] drm/amdgpu: CIK UVD initialization fixes John Olender
2025-04-29 11:24 ` [RFC PATCH 1/2] drm/amdgpu: amdgpu_vram_mgr_new(): Clamp lpfn to total vram John Olender
2025-04-30 21:20   ` Alex Deucher
2025-04-30 21:44     ` Paneer Selvam, Arunpravin
2025-05-02 15:32       ` John Olender
2025-05-03 12:23         ` Paneer Selvam, Arunpravin
2025-05-11 20:33           ` Paneer Selvam, Arunpravin
2025-05-11 20:37             ` Paneer Selvam, Arunpravin
2025-05-12  7:09               ` Christian König
2025-05-12  7:11                 ` Paneer Selvam, Arunpravin
2025-05-15 15:49                   ` Paneer Selvam, Arunpravin
2025-05-23 11:31                     ` Paneer Selvam, Arunpravin
2025-05-02  8:28     ` Christian König
2025-04-29 11:24 ` [RFC PATCH 2/2] drm/amdgpu/uvd: Ensure vcpu bos are within the uvd segment John Olender
2025-04-30 21:39   ` Alex Deucher
2025-05-02  8:36     ` Christian König
2025-05-03 20:31       ` John Olender
2025-05-05  9:02         ` Christian König
2025-05-05 16:06           ` John Olender
2025-05-07 11:31             ` John Olender
2025-05-29 23:15               ` John Olender
2025-06-02 10:00                 ` Christian König
2025-06-02 13:03                   ` John Olender
2025-06-03 14:34                     ` John Olender
2025-06-03 16:26                       ` Christian König
2025-06-03 17:52                         ` John Olender
2025-06-05  9:21       ` John Olender [this message]
2025-06-05  9:54         ` Christian König

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=450fa7e8-5d91-46e8-8b01-1195fe437a7a@gmail.com \
    --to=john.olender@gmail.com \
    --cc=alexdeucher@gmail.com \
    --cc=amd-gfx@lists.freedesktop.org \
    --cc=arunpravin.paneerselvam@amd.com \
    --cc=christian.koenig@amd.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox