* [PATCH v2] mm/damon/core: Optimize kdamod.%d thread creation code
@ 2021-10-16 16:59 Xin Hao
2021-10-19 10:03 ` SeongJae Park
0 siblings, 1 reply; 2+ messages in thread
From: Xin Hao @ 2021-10-16 16:59 UTC (permalink / raw)
To: sjpark; +Cc: xhao, akpm, linux-mm, linux-kernel
When the ctx->adaptive_targets list is empty,
i did some test on monitor_on interface like this.
echo on > /sys/kernel/debug/damon/monitor_on
[ 851.988307] damon: kdamond (5390) starts
Though the ctx->adaptive_targets list is empty, but the
kthread_run still be called, and the kdamond.x thread still
be created, this is meaningless.
So there adds a judgment. only if the ctx->adaptive_targets
list is not empty, and ctx->kdamond pointer is NULL, then call
the __damon_start function.
Signed-off-by: Xin Hao <xhao@linux.alibaba.com>
---
mm/damon/core.c | 30 ++++++++++++++++++------------
1 file changed, 18 insertions(+), 12 deletions(-)
diff --git a/mm/damon/core.c b/mm/damon/core.c
index 30e9211f494a..998c707fdca2 100644
--- a/mm/damon/core.c
+++ b/mm/damon/core.c
@@ -107,6 +107,11 @@ void damon_add_target(struct damon_ctx *ctx, struct damon_target *t)
list_add_tail(&t->list, &ctx->adaptive_targets);
}
+static int damon_target_empty(struct damon_ctx *ctx)
+{
+ return list_empty(&ctx->adaptive_targets);
+}
+
static void damon_del_target(struct damon_target *t)
{
list_del(&t->list);
@@ -307,15 +312,14 @@ static int __damon_start(struct damon_ctx *ctx)
int err = -EBUSY;
mutex_lock(&ctx->kdamond_lock);
- if (!ctx->kdamond) {
+ ctx->kdamond_stop = false;
+ ctx->kdamond = kthread_run(kdamond_fn, ctx, "kdamond.%d",
+ nr_running_ctxs);
+ if (IS_ERR(ctx->kdamond)) {
+ err = PTR_ERR(ctx->kdamond);
+ ctx->kdamond = 0;
+ } else {
err = 0;
- ctx->kdamond_stop = false;
- ctx->kdamond = kthread_run(kdamond_fn, ctx, "kdamond.%d",
- nr_running_ctxs);
- if (IS_ERR(ctx->kdamond)) {
- err = PTR_ERR(ctx->kdamond);
- ctx->kdamond = 0;
- }
}
mutex_unlock(&ctx->kdamond_lock);
@@ -347,10 +351,12 @@ int damon_start(struct damon_ctx **ctxs, int nr_ctxs)
}
for (i = 0; i < nr_ctxs; i++) {
- err = __damon_start(ctxs[i]);
- if (err)
- break;
- nr_running_ctxs++;
+ if (!damon_target_empty(ctxs[i]) && !ctxs[i]->kdamond) {
+ err = __damon_start(ctxs[i]);
+ if (err)
+ break;
+ nr_running_ctxs++;
+ }
}
mutex_unlock(&damon_lock);
--
2.31.0
^ permalink raw reply related [flat|nested] 2+ messages in thread
* Re: [PATCH v2] mm/damon/core: Optimize kdamod.%d thread creation code
2021-10-16 16:59 [PATCH v2] mm/damon/core: Optimize kdamod.%d thread creation code Xin Hao
@ 2021-10-19 10:03 ` SeongJae Park
0 siblings, 0 replies; 2+ messages in thread
From: SeongJae Park @ 2021-10-19 10:03 UTC (permalink / raw)
To: Xin Hao; +Cc: sjpark, akpm, linux-mm, linux-kernel
Hello Xin,
On Sun, 17 Oct 2021 00:59:14 +0800 Xin Hao <xhao@linux.alibaba.com> wrote:
> When the ctx->adaptive_targets list is empty,
> i did some test on monitor_on interface like this.
>
> echo on > /sys/kernel/debug/damon/monitor_on
> [ 851.988307] damon: kdamond (5390) starts
>
> Though the ctx->adaptive_targets list is empty, but the
> kthread_run still be called, and the kdamond.x thread still
> be created, this is meaningless.
Good finding, thank you!
>
> So there adds a judgment. only if the ctx->adaptive_targets
> list is not empty, and ctx->kdamond pointer is NULL, then call
> the __damon_start function.
I think we cannot believe user-space users, but kernel-space API users. How
about adding the check in 'dbgfs_monitor_on_write()' rather than
'__damon_start()'?
Also, apparently this patch is based on latest 'damon/next' branch of my tree.
Could you please rebase this on latest -mm tree?
Thanks,
SJ
[...]
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2021-10-19 10:03 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-10-16 16:59 [PATCH v2] mm/damon/core: Optimize kdamod.%d thread creation code Xin Hao
2021-10-19 10:03 ` SeongJae Park
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).