From: "Paul E. McKenney" <paulmck@kernel.org>
To: rcu@vger.kernel.org
Cc: linux-kernel@vger.kernel.org, kernel-team@meta.com,
rostedt@goodmis.org, "Paul E. McKenney" <paulmck@kernel.org>,
Vlastimil Babka <vbabka@suse.cz>
Subject: [PATCH rcu 11/11] rcuscale: Count outstanding callbacks per-task rather than per-CPU
Date: Thu, 1 Aug 2024 17:43:08 -0700 [thread overview]
Message-ID: <20240802004308.4134731-11-paulmck@kernel.org> (raw)
In-Reply-To: <917e8cc8-8688-428a-9122-25544c5cc101@paulmck-laptop>
The current rcu_scale_writer() asynchronous grace-period testing uses a
per-CPU counter to track the number of outstanding callbacks. This is
subject to CPU-imbalance errors when tasks migrate from one CPU to another
between the time that the counter is incremented and the callback is
queued, and additionally in kernels configured such that callbacks can
be invoked on some CPU other than the one that queued it.
This commit therefore arranges for per-task callback counts, thus avoiding
any issues with migration of either tasks or callbacks.
Reported-by: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
---
kernel/rcu/rcuscale.c | 10 ++++++----
1 file changed, 6 insertions(+), 4 deletions(-)
diff --git a/kernel/rcu/rcuscale.c b/kernel/rcu/rcuscale.c
index 13d379c179248..b1c50df142eba 100644
--- a/kernel/rcu/rcuscale.c
+++ b/kernel/rcu/rcuscale.c
@@ -114,6 +114,7 @@ struct writer_mblock {
struct writer_freelist {
struct llist_head ws_lhg;
+ atomic_t ws_inflight;
struct llist_head ____cacheline_internodealigned_in_smp ws_lhp;
struct writer_mblock *ws_mblocks;
};
@@ -136,7 +137,6 @@ static u64 t_rcu_scale_writer_started;
static u64 t_rcu_scale_writer_finished;
static unsigned long b_rcu_gp_test_started;
static unsigned long b_rcu_gp_test_finished;
-static DEFINE_PER_CPU(atomic_t, n_async_inflight);
#define MAX_MEAS 10000
#define MIN_MEAS 100
@@ -520,8 +520,9 @@ static void rcu_scale_free(struct writer_mblock *wmbp)
static void rcu_scale_async_cb(struct rcu_head *rhp)
{
struct writer_mblock *wmbp = container_of(rhp, struct writer_mblock, wmb_rh);
+ struct writer_freelist *wflp = wmbp->wmb_wfl;
- atomic_dec(this_cpu_ptr(&n_async_inflight));
+ atomic_dec(&wflp->ws_inflight);
rcu_scale_free(wmbp);
}
@@ -541,6 +542,7 @@ rcu_scale_writer(void *arg)
DEFINE_TORTURE_RANDOM(tr);
u64 *wdp;
u64 *wdpp = writer_durations[me];
+ struct writer_freelist *wflp = &writer_freelists[me];
struct writer_mblock *wmbp = NULL;
VERBOSE_SCALEOUT_STRING("rcu_scale_writer task started");
@@ -584,8 +586,8 @@ rcu_scale_writer(void *arg)
if (gp_async && !WARN_ON_ONCE(!cur_ops->async)) {
if (!wmbp)
wmbp = rcu_scale_alloc(me);
- if (wmbp && atomic_read(this_cpu_ptr(&n_async_inflight)) < gp_async_max) {
- atomic_inc(this_cpu_ptr(&n_async_inflight));
+ if (wmbp && atomic_read(&wflp->ws_inflight) < gp_async_max) {
+ atomic_inc(&wflp->ws_inflight);
cur_ops->async(&wmbp->wmb_rh, rcu_scale_async_cb);
wmbp = NULL;
gp_succeeded = true;
--
2.40.1
prev parent reply other threads:[~2024-08-02 0:43 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-08-02 0:43 [PATCH rcu 0/11] RCU update-side scalability update test Paul E. McKenney
2024-08-02 0:42 ` [PATCH rcu 02/11] rcuscale: Dump stacks of stalled rcu_scale_writer() instances Paul E. McKenney
2024-08-02 0:43 ` [PATCH rcu 03/11] rcuscale: Dump grace-period statistics when rcu_scale_writer() stalls Paul E. McKenney
2024-08-02 0:43 ` [PATCH rcu 04/11] rcu: Mark callbacks not currently participating in barrier operation Paul E. McKenney
2024-08-02 0:43 ` [PATCH rcu 05/11] rcuscale: Print detailed grace-period and barrier diagnostics Paul E. McKenney
2024-08-02 0:43 ` [PATCH rcu 06/11] rcuscale: Provide clear error when async specified without primitives Paul E. McKenney
2024-08-14 12:49 ` Neeraj Upadhyay
2024-08-14 15:09 ` Paul E. McKenney
2024-08-02 0:43 ` [PATCH rcu 07/11] rcuscale: Make all writer tasks report upon hang Paul E. McKenney
2024-08-02 0:43 ` [PATCH rcu 08/11] rcuscale: Make rcu_scale_writer() tolerate repeated GFP_KERNEL failure Paul E. McKenney
2024-08-02 0:43 ` [PATCH rcu 09/11] rcuscale: Use special allocator for rcu_scale_writer() Paul E. McKenney
2024-08-02 0:43 ` [PATCH rcu 10/11] rcuscale: NULL out top-level pointers to heap memory Paul E. McKenney
2024-08-02 0:43 ` Paul E. McKenney [this message]
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=20240802004308.4134731-11-paulmck@kernel.org \
--to=paulmck@kernel.org \
--cc=kernel-team@meta.com \
--cc=linux-kernel@vger.kernel.org \
--cc=rcu@vger.kernel.org \
--cc=rostedt@goodmis.org \
--cc=vbabka@suse.cz \
/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