From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-1.web.codeaurora.org [10.30.226.201]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 180F6350D73; Fri, 17 Oct 2025 13:47:54 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=10.30.226.201 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1760708875; cv=none; b=eX8DlQqkGcD9e0rW03g1L5i4nrSnBl4KE3k3vJ0sCqrl419YOVNxwo8qVhNL57GL1FytzfzWKejzjnxMvqnow5PVq93XMZHdJ1PmdLhgCHTNZD1n4XrpjUUaYk/yKYvBGr/192Ifja63CI88KoUIQZw2IDOyWA+Xpt+EgyG6OZ4= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1760708875; c=relaxed/simple; bh=vrRFEKTK4AWiLuQSasLbMw065+46eQtPuGvNJqTLrH4=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version:Content-Type; b=KaJhJ5ejGzgtUoCQfde3wHA+vGDHFRcGPQF1o7AN7PZwXDczyrXfvjH/ZAuk4f7vItJ6D8lIvAv433A8S/wSQCXFSjNCeiQvJp3GNYQIEFWM2jdm8NWXKqMuzZrMRasFsDfQUWCqvr6iy3c2faQs/ZysHialBv4/yE+XOWR4nso= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b=MgP3WEzH; arc=none smtp.client-ip=10.30.226.201 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b="MgP3WEzH" Received: by smtp.kernel.org (Postfix) with ESMTPSA id 55AA1C4CEFE; Fri, 17 Oct 2025 13:47:52 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=k20201202; t=1760708874; bh=vrRFEKTK4AWiLuQSasLbMw065+46eQtPuGvNJqTLrH4=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=MgP3WEzH4xzjfD6A6/bW0TBNZZNnElvkbi1wUWqZ12I36MqZyEv/PJVoBsBNuZH/7 WlLvZgFOdDrpj0Ck0ql7eSo1Mj9JDHTnFvT5q3rHiMJt1sij5uxRAOILDaQKd5PI92 x78WkVedghK5peiINGn5Hw+ps6c6cplNXiyaT+Xz9ZtJf93/dmkKnjkvxlSpO7EaIX tV3YLDte3ZSdrNepHn+D7fr5M6gL4DKG83FpGZkiw6VCF7C5duPZD6zQNhV6qVjFJ4 ED6fHIBBZVbwWQtC3kr9kPQD+rdJwNkG2G+qFVHTYg7mBGaV2doQg7xwIc8j7dopeB FkVzV32lxWT6w== From: Philipp Stanner To: Matthew Brost , Danilo Krummrich , Philipp Stanner , =?UTF-8?q?Christian=20K=C3=B6nig?= , David Airlie , Simona Vetter , Sumit Semwal Cc: dri-devel@lists.freedesktop.org, linux-kernel@vger.kernel.org, linux-media@vger.kernel.org Subject: [PATCH 2/3] drm/sched: Add TODO file with first entry Date: Fri, 17 Oct 2025 15:47:02 +0200 Message-ID: <20251017134716.187723-3-phasta@kernel.org> X-Mailer: git-send-email 2.49.0 In-Reply-To: <20251017134716.187723-2-phasta@kernel.org> References: <20251017134716.187723-2-phasta@kernel.org> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Add a drm_sched TODO file with open tasks, contact info, difficulty level and a job description. Add the missing successor of drm_sched_resubmit_jobs() as a first task. Signed-off-by: Philipp Stanner --- drivers/gpu/drm/scheduler/TODO | 27 +++++++++++++++++++++++++++ 1 file changed, 27 insertions(+) create mode 100644 drivers/gpu/drm/scheduler/TODO diff --git a/drivers/gpu/drm/scheduler/TODO b/drivers/gpu/drm/scheduler/TODO new file mode 100644 index 000000000000..6a06e2858dd6 --- /dev/null +++ b/drivers/gpu/drm/scheduler/TODO @@ -0,0 +1,27 @@ +=== drm_sched TODO list === + +* GPU job resubmits + - Difficulty: hard + - Contact: + - Christian König + - Philipp Stanner + - Description: + drm_sched_resubmit_jobs() is deprecated. Main reason being that it leads to + reinitializing dma_fences. See that function's docu for details. The better + approach for valid resubmissions by amdgpu and Xe is (apparently) to figure + out which job (and, through association: which entity) caused the hang. Then, + the job's buffer data, together with all other jobs' buffer data currently + in the same hardware ring, must be invalidated. This can for example be done + by overwriting it. + amdgpu currently determines which jobs are in the ring and need to be + overwritten by keeping copies of the job. Xe obtains that information by + directly accessing drm_sched's pending_list. + - Tasks: + 1. implement scheduler functionality through which + the driver can obtain the information which *broken* jobs are currently in + the hardware ring. + 2. Such infrastructure would then typically be used in + drm_sched_backend_ops.timedout_job(). Document that. + 3. Port a driver as first user. + 3. Document the new alternative in the docu of deprecated + drm_sched_resubmit_jobs(). -- 2.49.0