From: Tejun Heo <tj@kernel.org>
To: void@manifault.com, arighi@nvidia.com, multics69@gmail.com
Cc: linux-kernel@vger.kernel.org, sched-ext@meta.com
Subject: [PATCHSET v2 sched_ext/for-6.16] sched_ext: Introduce scx_sched
Date: Fri, 25 Apr 2025 11:58:15 -1000 [thread overview]
Message-ID: <20250425215840.2334972-1-tj@kernel.org> (raw)
v[1] -> v2:
- 0003-sched_ext-Use-dynamic-allocation-for-scx_sched.patch updated to store
scx_sched pointer in sched_ext_ops->priv so that bpf_scx_unreg() can
determine the scx_sched to destroy regardless of where initialization
failed. This will also help with supporting multiple schedulers.
- Minor updates to 0010-sched_ext-Move-event_stats_cpu-into-scx_sched.patch.
In preparation of supporting multiple hierarchical schedulers, this patchset
packages the states which are currently global but need to become per
scheduler instance into the new struct scx_sched.
Currently, the only supported scheduler instance is the system-wide root one
(scx_root) and the code assumes scx_root in many places. A follow-up
patchset will introduce multiple scheduler support scrutinizing and updating
each such occurrence appropriately.
There's significant amount of churning but most changes are straightforward
if not trivial, and no user visible changes are expected.
0001-sched_ext-Introduce-scx_sched.patch
0002-sched_ext-Avoid-NULL-scx_root-deref-through-SCX_HAS_.patch
0003-sched_ext-Use-dynamic-allocation-for-scx_sched.patch
0004-sched_ext-Inline-create_dsq-into-scx_bpf_create_dsq.patch
0005-sched_ext-Factor-out-scx_alloc_and_add_sched.patch
0006-sched_ext-Move-dsq_hash-into-scx_sched.patch
0007-sched_ext-Move-global_dsqs-into-scx_sched.patch
0008-sched_ext-Relocate-scx_event_stats-definition.patch
0009-sched_ext-Factor-out-scx_read_events.patch
0010-sched_ext-Move-event_stats_cpu-into-scx_sched.patch
0011-sched_ext-Move-disable-machinery-into-scx_sched.patch
0012-sched_ext-Clean-up-SCX_EXIT_NONE-handling-in-scx_dis.patch
and is also available in the following git branch:
git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext.git scx_sched-v2
diffstat follows. Thanks.
kernel/sched/ext.c | 892 ++++++++++++++++++++++++++----------------------
kernel/sched/ext_idle.c | 3
2 files changed, 496 insertions(+), 399 deletions(-)
--
tejun
[1] http://lkml.kernel.org/r/20250423234542.1890867-1-tj@kernel.org
next reply other threads:[~2025-04-25 21:58 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-04-25 21:58 Tejun Heo [this message]
2025-04-25 21:58 ` [PATCH 01/12] sched_ext: Introduce scx_sched Tejun Heo
2025-04-25 21:58 ` [PATCH 02/12] sched_ext: Avoid NULL scx_root deref through SCX_HAS_OP() Tejun Heo
2025-04-25 21:58 ` [PATCH 03/12] sched_ext: Use dynamic allocation for scx_sched Tejun Heo
2025-04-25 21:58 ` [PATCH 04/12] sched_ext: Inline create_dsq() into scx_bpf_create_dsq() Tejun Heo
2025-04-25 21:58 ` [PATCH 05/12] sched_ext: Factor out scx_alloc_and_add_sched() Tejun Heo
2025-04-25 21:58 ` [PATCH 06/12] sched_ext: Move dsq_hash into scx_sched Tejun Heo
2025-04-26 20:25 ` Andrea Righi
2025-04-28 20:43 ` [PATCH v2 " Tejun Heo
2025-04-28 23:34 ` Changwoo Min
2025-04-25 21:58 ` [PATCH 07/12] sched_ext: Move global_dsqs " Tejun Heo
2025-04-25 21:58 ` [PATCH 08/12] sched_ext: Relocate scx_event_stats definition Tejun Heo
2025-04-25 21:58 ` [PATCH 09/12] sched_ext: Factor out scx_read_events() Tejun Heo
2025-04-25 21:58 ` [PATCH 10/12] sched_ext: Move event_stats_cpu into scx_sched Tejun Heo
2025-04-25 21:58 ` [PATCH 11/12] sched_ext: Move disable machinery " Tejun Heo
2025-04-25 21:58 ` [PATCH 12/12] sched_ext: Clean up SCX_EXIT_NONE handling in scx_disable_workfn() Tejun Heo
2025-04-27 7:04 ` [PATCHSET v2 sched_ext/for-6.16] sched_ext: Introduce scx_sched Changwoo Min
2025-04-28 20:58 ` Andrea Righi
2025-04-29 18:41 ` Tejun Heo
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=20250425215840.2334972-1-tj@kernel.org \
--to=tj@kernel.org \
--cc=arighi@nvidia.com \
--cc=linux-kernel@vger.kernel.org \
--cc=multics69@gmail.com \
--cc=sched-ext@meta.com \
--cc=void@manifault.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.