From: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
To: amd-gfx@lists.freedesktop.org, dri-devel@lists.freedesktop.org
Cc: kernel-dev@igalia.com, intel-xe@lists.freedesktop.org,
Danilo Krummrich <dakr@kernel.org>,
Philipp Stanner <phasta@kernel.org>,
Tvrtko Ursulin <tvrtko.ursulin@igalia.com>,
Christian König <christian.koenig@amd.com>,
Matthew Brost <matthew.brost@intel.com>,
Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Subject: [PATCH v7 12/29] drm/sched: Switch default policy to fair
Date: Fri, 6 Mar 2026 16:34:28 +0000 [thread overview]
Message-ID: <20260306163445.97243-13-tvrtko.ursulin@igalia.com> (raw)
In-Reply-To: <20260306163445.97243-1-tvrtko.ursulin@igalia.com>
FAIR policy works better than FIFO for all known use cases and either
matches or gets close to RR. Lets make it a default to improve the user
experience especially with interactive workloads competing with heavy
clients.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Cc: Christian König <christian.koenig@amd.com>
Cc: Danilo Krummrich <dakr@kernel.org>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Philipp Stanner <phasta@kernel.org>
Cc: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Acked-by: Danilo Krummrich <dakr@kernel.org>
---
drivers/gpu/drm/scheduler/sched_main.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/scheduler/sched_main.c b/drivers/gpu/drm/scheduler/sched_main.c
index 337db7b1e688..8127c93a1718 100644
--- a/drivers/gpu/drm/scheduler/sched_main.c
+++ b/drivers/gpu/drm/scheduler/sched_main.c
@@ -84,13 +84,13 @@
#define CREATE_TRACE_POINTS
#include "gpu_scheduler_trace.h"
-int drm_sched_policy = DRM_SCHED_POLICY_FIFO;
+int drm_sched_policy = DRM_SCHED_POLICY_FAIR;
/**
* DOC: sched_policy (int)
* Used to override default entities scheduling policy in a run queue.
*/
-MODULE_PARM_DESC(sched_policy, "Specify the scheduling policy for entities on a run-queue, " __stringify(DRM_SCHED_POLICY_RR) " = Round Robin, " __stringify(DRM_SCHED_POLICY_FIFO) " = FIFO (default), " __stringify(DRM_SCHED_POLICY_FAIR) " = Fair.");
+MODULE_PARM_DESC(sched_policy, "Specify the scheduling policy for entities on a run-queue, " __stringify(DRM_SCHED_POLICY_RR) " = Round Robin, " __stringify(DRM_SCHED_POLICY_FIFO) " = FIFO, " __stringify(DRM_SCHED_POLICY_FAIR) " = Fair (default).");
module_param_named(sched_policy, drm_sched_policy, int, 0444);
static u32 drm_sched_available_credits(struct drm_gpu_scheduler *sched)
--
2.52.0
next prev parent reply other threads:[~2026-03-06 16:35 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-03-06 16:34 [PATCH v7 00/29] Fair(er) DRM scheduler Tvrtko Ursulin
2026-03-06 16:34 ` [PATCH v7 01/29] drm/sched: Disallow initializing entities with no schedulers Tvrtko Ursulin
2026-03-06 16:34 ` [PATCH v7 02/29] drm/sched: Consolidate entity run queue management Tvrtko Ursulin
2026-03-06 16:34 ` [PATCH v7 03/29] drm/sched: Move run queue related code into a separate file Tvrtko Ursulin
2026-03-06 16:34 ` [PATCH v7 04/29] drm/sched: Add some scheduling quality unit tests Tvrtko Ursulin
2026-03-08 10:10 ` kernel test robot
2026-03-06 16:34 ` [PATCH v7 05/29] drm/sched: Add some more " Tvrtko Ursulin
2026-03-06 16:34 ` [PATCH v7 06/29] drm/sched: Implement RR via FIFO Tvrtko Ursulin
2026-03-06 16:34 ` [PATCH v7 07/29] drm/sched: Free all finished jobs at once Tvrtko Ursulin
2026-03-06 16:34 ` [PATCH v7 08/29] drm/sched: Account entity GPU time Tvrtko Ursulin
2026-03-06 16:34 ` [PATCH v7 09/29] drm/sched: Remove idle entity from tree Tvrtko Ursulin
2026-03-06 16:34 ` [PATCH v7 10/29] drm/sched: Add fair scheduling policy Tvrtko Ursulin
2026-03-06 16:34 ` [PATCH v7 11/29] drm/sched: Favour interactive clients slightly Tvrtko Ursulin
2026-03-06 16:34 ` Tvrtko Ursulin [this message]
2026-03-06 16:34 ` [PATCH v7 13/29] drm/sched: Remove FIFO and RR and simplify to a single run queue Tvrtko Ursulin
2026-03-06 16:34 ` [PATCH v7 14/29] drm/sched: Embed run queue singleton into the scheduler Tvrtko Ursulin
2026-03-06 16:34 ` [PATCH v7 15/29] accel/amdxdna: Remove drm_sched_init_args->num_rqs usage Tvrtko Ursulin
2026-03-06 16:34 ` [PATCH v7 16/29] accel/rocket: " Tvrtko Ursulin
2026-03-06 16:34 ` [PATCH v7 17/29] accel/ethosu: " Tvrtko Ursulin
2026-03-06 16:34 ` [PATCH v7 18/29] drm/amdgpu: " Tvrtko Ursulin
2026-03-06 16:34 ` [PATCH v7 19/29] drm/etnaviv: " Tvrtko Ursulin
2026-03-06 16:34 ` [PATCH v7 20/29] drm/imagination: " Tvrtko Ursulin
2026-03-06 16:34 ` [PATCH v7 21/29] drm/lima: " Tvrtko Ursulin
2026-03-06 16:34 ` [PATCH v7 22/29] drm/msm: " Tvrtko Ursulin
2026-03-06 16:34 ` [PATCH v7 23/29] drm/nouveau: " Tvrtko Ursulin
2026-03-06 16:34 ` [PATCH v7 24/29] drm/panfrost: " Tvrtko Ursulin
2026-03-06 16:34 ` [PATCH v7 25/29] drm/panthor: " Tvrtko Ursulin
2026-03-06 16:34 ` [PATCH v7 26/29] drm/sched: " Tvrtko Ursulin
2026-03-06 16:34 ` [PATCH v7 27/29] drm/v3d: " Tvrtko Ursulin
2026-03-06 16:34 ` [PATCH v7 28/29] drm/xe: " Tvrtko Ursulin
2026-03-06 16:34 ` [PATCH v7 29/29] drm/sched: Remove drm_sched_init_args->num_rqs Tvrtko Ursulin
2026-03-08 22:37 ` Claude review: Fair(er) DRM scheduler Claude Code Review Bot
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=20260306163445.97243-13-tvrtko.ursulin@igalia.com \
--to=tvrtko.ursulin@igalia.com \
--cc=amd-gfx@lists.freedesktop.org \
--cc=christian.koenig@amd.com \
--cc=dakr@kernel.org \
--cc=dri-devel@lists.freedesktop.org \
--cc=intel-xe@lists.freedesktop.org \
--cc=kernel-dev@igalia.com \
--cc=matthew.brost@intel.com \
--cc=phasta@kernel.org \
--cc=pierre-eric.pelloux-prayer@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