This is a note to let you know that I've just added the patch titled
io-wq: fix max-workers not correctly set on multi-node system
to the 5.15-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary
The filename of the patch is:
io-wq-fix-max-workers-not-correctly-set-on-multi-nod.patch
and it can be found in the queue-5.15 subdirectory.
If you, or anyone else, feels it should not be added to the stable tree,
please let <stable@xxxxxxxxxxxxxxx> know about it.
commit bd6d5078102f2606b74ecfff0b03b556a4a83406
Author: Beld Zhang <beldzhang@xxxxxxxxx>
Date: Tue Nov 2 12:32:08 2021 -0600
io-wq: fix max-workers not correctly set on multi-node system
[ Upstream commit 71c9ce27bb57c59d8d7f5298e730c8096eef3d1f ]
In io-wq.c:io_wq_max_workers(), new_count[] was changed right after each
node's value was set. This caused the following node getting the setting
of the previous one.
Returned values are copied from node 0.
Fixes: 2e480058ddc2 ("io-wq: provide a way to limit max number of workers")
Signed-off-by: Beld Zhang <beldzhang@xxxxxxxxx>
[axboe: minor fixups]
Signed-off-by: Jens Axboe <axboe@xxxxxxxxx>
Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx>
diff --git a/fs/io-wq.c b/fs/io-wq.c
index 5d189b24a8d4b..8c61315657546 100644
--- a/fs/io-wq.c
+++ b/fs/io-wq.c
@@ -1318,7 +1318,9 @@ int io_wq_cpu_affinity(struct io_wq *wq, cpumask_var_t mask)
*/
int io_wq_max_workers(struct io_wq *wq, int *new_count)
{
- int i, node, prev = 0;
+ int prev[IO_WQ_ACCT_NR];
+ bool first_node = true;
+ int i, node;
BUILD_BUG_ON((int) IO_WQ_ACCT_BOUND != (int) IO_WQ_BOUND);
BUILD_BUG_ON((int) IO_WQ_ACCT_UNBOUND != (int) IO_WQ_UNBOUND);
@@ -1329,6 +1331,9 @@ int io_wq_max_workers(struct io_wq *wq, int *new_count)
new_count[i] = task_rlimit(current, RLIMIT_NPROC);
}
+ for (i = 0; i < IO_WQ_ACCT_NR; i++)
+ prev[i] = 0;
+
rcu_read_lock();
for_each_node(node) {
struct io_wqe *wqe = wq->wqes[node];
@@ -1337,14 +1342,19 @@ int io_wq_max_workers(struct io_wq *wq, int *new_count)
raw_spin_lock(&wqe->lock);
for (i = 0; i < IO_WQ_ACCT_NR; i++) {
acct = &wqe->acct[i];
- prev = max_t(int, acct->max_workers, prev);
+ if (first_node)
+ prev[i] = max_t(int, acct->max_workers, prev[i]);
if (new_count[i])
acct->max_workers = new_count[i];
- new_count[i] = prev;
}
raw_spin_unlock(&wqe->lock);
+ first_node = false;
}
rcu_read_unlock();
+
+ for (i = 0; i < IO_WQ_ACCT_NR; i++)
+ new_count[i] = prev[i];
+
return 0;
}
