mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: "Christian König" <christian.koenig@amd.com>
To: phasta@kernel.org, Alice Ryhl <aliceryhl@google.com>,
	Boris Brezillon <boris.brezillon@collabora.com>
Cc: Danilo Krummrich <dakr@kernel.org>,
	David Airlie <airlied@gmail.com>, Simona Vetter <simona@ffwll.ch>,
	Gary Guo <gary@garyguo.net>, Benno Lossin <lossin@kernel.org>,
	Daniel Almeida <daniel.almeida@collabora.com>,
	Joel Fernandes <joelagnelf@nvidia.com>,
	linux-kernel@vger.kernel.org, dri-devel@lists.freedesktop.org,
	rust-for-linux@vger.kernel.org
Subject: Re: [RFC PATCH 2/4] rust: sync: Add dma_fence abstractions
Date: Tue, 10 Feb 2026 15:06:38 +0100	[thread overview]
Message-ID: <c1d4ad2a-44a0-4c04-b0b4-94d6242d4ae8@amd.com> (raw)
In-Reply-To: <fe2f97ae271116e7544be778a1d1aaace529dfec.camel@mailbox.org>

On 2/10/26 15:00, Philipp Stanner wrote:
> On Tue, 2026-02-10 at 14:56 +0100, Christian König wrote:
>> On 2/10/26 14:49, Alice Ryhl wrote:
>>> On Tue, Feb 10, 2026 at 02:26:31PM +0100, Boris Brezillon wrote:
>>>> On Tue, 10 Feb 2026 13:15:31 +0000
>>>> Alice Ryhl <aliceryhl@google.com> wrote:
>>>>
>>>>
>>>>
> 
> […]
> 
>>>> I mean, the timeout handler should also be considered a DMA-signalling
>>>> path, and the same rules should apply to it.
>>>
>>> I guess that's fair. Even with a timeout you want both to be signalling
>>> path.
>>>
>>> I guess more generally, if a fence is signalled by mechanism A or B,
>>> whichever happens first, you have the choice between:
>>
>> That doesn't happen in practice.
>>
>> For each fence you only have one signaling path you need to guarantee forward progress for.
>>
>> All other signaling paths are just opportunistically optimizations which *can* signal the fence, but there is no guarantee that they will.
> 
> Are you now referring to the fast-path callbacks like fence->ops-
>> is_signaled()? Or are you talking about different reference holders
> which might want to signal?

Yes, I'm referring to the is_signaled() callback.

When you have multiple reference holders which can all signal the fence by calling dma_fence_signal then there is clearly something going wrong.

What can be is that you have something like a fallback timer for buddy HW which swallows IRQs (e.g. like radeon or amdgpu have), but then you have something like a high level spinlock which makes sure that your IRQ handler is neither re-entrant nor multiple instances running at the same time on different CPUs.

>>
>> We used to have some exceptions to that, especially around aborting submissions, but those turned out to be a really bad idea as well.  
>>
>> Thinking more about it you should probably enforce that there is only one signaling path for each fence signaling.
> 
> An idea that is floating around is to move the entire fence signaling
> functionality into the dma fence context. That would have exclusive
> access, and could also finally guarantee that fences must be signaled
> in order.

That sounds like a sane idea to me.

Regards,
Christian.

> 
> 
> P.


  reply	other threads:[~2026-02-10 14:06 UTC|newest]

Thread overview: 103+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-02-03  8:13 [RFC PATCH 0/4] Add dma_fence abstractions and DRM Jobqueue Philipp Stanner
2026-02-03  8:14 ` [RFC PATCH 1/4] rust: list: Add unsafe for container_of Philipp Stanner
2026-02-03 15:25   ` Gary Guo
2026-02-04 10:30   ` Alice Ryhl
2026-02-03  8:14 ` [RFC PATCH 2/4] rust: sync: Add dma_fence abstractions Philipp Stanner
2026-02-05  8:57   ` Boris Brezillon
2026-02-06 10:23     ` Danilo Krummrich
2026-02-09  8:19       ` Philipp Stanner
2026-02-09 14:58         ` Boris Brezillon
2026-02-10  8:16           ` Christian König
2026-02-10  8:38             ` Alice Ryhl
2026-02-10  9:06               ` Philipp Stanner
2026-02-10  9:54                 ` Christian König
2026-02-10  9:15               ` Boris Brezillon
2026-02-10 10:15                 ` Alice Ryhl
2026-02-10 10:36                   ` Danilo Krummrich
2026-02-10 10:46                     ` Christian König
2026-02-10 11:40                       ` Alice Ryhl
2026-02-10 12:28                         ` Boris Brezillon
2026-02-11  9:57                         ` Danilo Krummrich
2026-02-11 10:08                           ` Philipp Stanner
2026-02-11 10:28                             ` Boris Brezillon
2026-02-11 10:20                           ` Boris Brezillon
2026-02-11 11:00                             ` Danilo Krummrich
2026-02-11 11:12                               ` Boris Brezillon
2026-02-11 14:38                                 ` Danilo Krummrich
2026-02-11 15:00                                   ` Boris Brezillon
2026-02-11 15:05                                     ` Danilo Krummrich
2026-02-11 15:14                                       ` Boris Brezillon
2026-02-11 15:16                                         ` Danilo Krummrich
2026-03-13 17:27                                     ` Matthew Brost
2026-02-10 10:46                   ` Boris Brezillon
2026-02-10 11:34                   ` Boris Brezillon
2026-02-10 11:45                     ` Alice Ryhl
2026-02-10 12:21                       ` Boris Brezillon
2026-02-10 13:34                         ` Alice Ryhl
2026-02-10 12:36                   ` Boris Brezillon
2026-02-10 13:15                     ` Alice Ryhl
2026-02-10 13:26                       ` Boris Brezillon
2026-02-10 13:49                         ` Alice Ryhl
2026-02-10 13:56                           ` Christian König
2026-02-10 14:00                             ` Philipp Stanner
2026-02-10 14:06                               ` Christian König [this message]
2026-02-10 15:32                                 ` Philipp Stanner
2026-02-10 15:50                                   ` Christian König
2026-02-10 15:07                             ` Alice Ryhl
2026-02-10 15:45                               ` Christian König
2026-02-11  8:16                                 ` Philipp Stanner
2026-02-17 14:03                                 ` Philipp Stanner
2026-02-17 14:09                                   ` Alice Ryhl
2026-02-17 14:22                                     ` Christian König
2026-02-17 14:28                                       ` Philipp Stanner
2026-02-17 14:44                                         ` Danilo Krummrich
2026-03-13 23:20                                           ` Matthew Brost
2026-02-17 15:01                                         ` Christian König
2026-02-18  9:50                                         ` Alice Ryhl
2026-02-18 10:48                                           ` Boris Brezillon
2026-02-10 12:49                   ` Boris Brezillon
2026-02-10 12:56                     ` Boris Brezillon
2026-02-10 13:26                     ` Alice Ryhl
2026-02-10 13:51                       ` Boris Brezillon
2026-02-10 14:11                         ` Alice Ryhl
2026-02-10 14:50                           ` Boris Brezillon
2026-02-11  8:16                             ` Alice Ryhl
2026-02-11  9:20                               ` Boris Brezillon
2026-02-10  9:26               ` Christian König
2026-02-05 10:16   ` Boris Brezillon
2026-02-05 13:16     ` Gary Guo
2026-02-06  9:32       ` Philipp Stanner
2026-02-06 10:16         ` Danilo Krummrich
2026-02-06 13:24           ` Philipp Stanner
2026-02-06 11:04         ` Boris Brezillon
2026-02-09  8:21           ` Philipp Stanner
2026-02-06 11:23         ` Boris Brezillon
2026-02-09 11:30   ` Alice Ryhl
2026-02-03  8:14 ` [RFC PATCH 3/4] rust/drm: Add DRM Jobqueue Philipp Stanner
2026-02-10 14:57   ` Boris Brezillon
2026-02-11 10:47     ` Philipp Stanner
2026-02-11 11:07       ` Boris Brezillon
2026-02-11 11:19         ` Danilo Krummrich
2026-02-11 12:10           ` Boris Brezillon
2026-02-11 12:32             ` Danilo Krummrich
2026-02-11 12:51               ` Boris Brezillon
2026-02-11 11:19         ` Philipp Stanner
2026-02-11 11:59           ` Boris Brezillon
2026-02-11 12:14             ` Philipp Stanner
2026-02-11 12:24               ` Boris Brezillon
2026-02-11 12:22           ` Alice Ryhl
2026-02-11 12:44             ` Philipp Stanner
2026-02-11 12:52               ` Alice Ryhl
2026-02-11 13:53                 ` Philipp Stanner
2026-02-11 15:28                   ` Alice Ryhl
2026-02-11 12:45             ` Danilo Krummrich
2026-02-11 13:45             ` Gary Guo
2026-02-11 14:07               ` Boris Brezillon
2026-02-11 15:17                 ` Alice Ryhl
2026-02-11 15:20                   ` Philipp Stanner
2026-02-11 15:51                     ` Boris Brezillon
2026-02-11 15:53                     ` Alice Ryhl
2026-02-11 15:54                     ` Danilo Krummrich
2026-02-11 15:33               ` Alice Ryhl
2026-02-03  8:14 ` [RFC PATCH 4/4] samples: rust: Add jobqueue tester Philipp Stanner
2026-02-03 16:46 ` [RFC PATCH 0/4] Add dma_fence abstractions and DRM Jobqueue Daniel Almeida

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=c1d4ad2a-44a0-4c04-b0b4-94d6242d4ae8@amd.com \
    --to=christian.koenig@amd.com \
    --cc=airlied@gmail.com \
    --cc=aliceryhl@google.com \
    --cc=boris.brezillon@collabora.com \
    --cc=dakr@kernel.org \
    --cc=daniel.almeida@collabora.com \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=gary@garyguo.net \
    --cc=joelagnelf@nvidia.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=lossin@kernel.org \
    --cc=phasta@kernel.org \
    --cc=rust-for-linux@vger.kernel.org \
    --cc=simona@ffwll.ch \
    /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

all inboxes | Powered by JetHome®