From: Lai Jiangshan <jiangshanlai@gmail.com>
To: linux-kernel@vger.kernel.org
Cc: Tejun Heo <tj@kernel.org>,
Naohiro.Aota@wdc.com, Lai Jiangshan <jiangshan.ljs@antgroup.com>,
Lai Jiangshan <jiangshanlai@gmail.com>
Subject: [PATCH 1/7] workqueue: Reuse the default PWQ as much as possible
Date: Wed, 27 Dec 2023 22:51:37 +0800 [thread overview]
Message-ID: <20231227145143.2399-2-jiangshanlai@gmail.com> (raw)
In-Reply-To: <20231227145143.2399-1-jiangshanlai@gmail.com>
From: Lai Jiangshan <jiangshan.ljs@antgroup.com>
If the PWQ to be allocated has the same __pod_cpumask as the
default one, just reuse the default one.
No functionality changes intend.
Signed-off-by: Lai Jiangshan <jiangshan.ljs@antgroup.com>
---
kernel/workqueue.c | 16 +++++++++-------
1 file changed, 9 insertions(+), 7 deletions(-)
diff --git a/kernel/workqueue.c b/kernel/workqueue.c
index 2989b57e154a..e734625fc8ce 100644
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -4270,7 +4270,7 @@ static void wq_calc_pod_cpumask(struct workqueue_attrs *attrs, int cpu,
if (cpu_going_down >= 0)
cpumask_clear_cpu(cpu_going_down, attrs->__pod_cpumask);
- if (cpumask_empty(attrs->__pod_cpumask)) {
+ if (attrs->ordered || cpumask_empty(attrs->__pod_cpumask)) {
cpumask_copy(attrs->__pod_cpumask, attrs->cpumask);
return;
}
@@ -4360,15 +4360,15 @@ apply_wqattrs_prepare(struct workqueue_struct *wq,
goto out_free;
for_each_possible_cpu(cpu) {
- if (new_attrs->ordered) {
+ wq_calc_pod_cpumask(new_attrs, cpu, -1);
+ if (cpumask_equal(new_attrs->cpumask, new_attrs->__pod_cpumask)) {
ctx->dfl_pwq->refcnt++;
ctx->pwq_tbl[cpu] = ctx->dfl_pwq;
- } else {
- wq_calc_pod_cpumask(new_attrs, cpu, -1);
- ctx->pwq_tbl[cpu] = alloc_unbound_pwq(wq, new_attrs);
- if (!ctx->pwq_tbl[cpu])
- goto out_free;
+ continue;
}
+ ctx->pwq_tbl[cpu] = alloc_unbound_pwq(wq, new_attrs);
+ if (!ctx->pwq_tbl[cpu])
+ goto out_free;
}
/* save the user configured attrs and sanitize it. */
@@ -4530,6 +4530,8 @@ static void wq_update_pod(struct workqueue_struct *wq, int cpu,
lockdep_is_held(&wq_pool_mutex));
if (wqattrs_equal(target_attrs, pwq->pool->attrs))
return;
+ if (cpumask_equal(target_attrs->cpumask, target_attrs->__pod_cpumask))
+ goto use_dfl_pwq;
/* create a new pwq */
pwq = alloc_unbound_pwq(wq, target_attrs);
--
2.19.1.6.gb485710b
next prev parent reply other threads:[~2023-12-27 14:49 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-12-27 14:51 [PATCH 0/7] workqueue: Share the same PWQ for the CPUs of a pod and distribute max_active across pods Lai Jiangshan
2023-12-27 14:51 ` Lai Jiangshan [this message]
2023-12-27 14:51 ` [PATCH 2/7] workqueue: Share the same PWQ for the CPUs of a pod Lai Jiangshan
2024-01-03 2:55 ` kernel test robot
2024-01-03 9:01 ` Lai Jiangshan
2023-12-27 14:51 ` [PATCH 3/7] workqueue: Add pwq_calculate_max_active() Lai Jiangshan
2023-12-27 14:51 ` [PATCH 4/7] workqueue: Wrap common code into wq_adjust_pwqs_max_active() Lai Jiangshan
2023-12-27 14:51 ` [PATCH 5/7] workqueue: Addjust pwq's max_active when CPU online/offine Lai Jiangshan
2023-12-27 14:51 ` [PATCH 6/7] workqueue: Implement system-wide max_active enforcement for unbound workqueues Lai Jiangshan
2023-12-27 23:06 ` Tejun Heo
2023-12-27 14:51 ` [PATCH 7/7] workqueue: Rename wq->saved_max_active to wq->max_active Lai Jiangshan
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=20231227145143.2399-2-jiangshanlai@gmail.com \
--to=jiangshanlai@gmail.com \
--cc=Naohiro.Aota@wdc.com \
--cc=jiangshan.ljs@antgroup.com \
--cc=linux-kernel@vger.kernel.org \
--cc=tj@kernel.org \
/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®