> From: Paul E. McKenney <paulmck@xxxxxxxxxx>
> [...]
> Subject: Re: [PATCH 1/1] rcu/rcuscale: Stop kfree_scale_thread thread(s)
> after unloading rcuscale
>
> On Mon, Mar 13, 2023 at 04:04:03PM +0800, Qiuxu Zhuo wrote:
> > When running the 'kfree_rcu_test' test case with commands [1] the call
> > trace [2] was thrown. This was because the kfree_scale_thread
> > thread(s) still run after unloading rcuscale and torture modules. Fix
> > the call trace by invoking kfree_scale_cleanup() when removing the
> rcuscale module.
>
> Good catch, thank you!
>
> > [1] modprobe torture
> > modprobe rcuscale kfree_rcu_test=1
>
> Given that loading the rcuscale kernel module automatically pulls in the
> rcuscale kernel module, why the "modprobe torture"? Is doing the
Oops ...
I forgot that the system could automatically figure out and load dependent modules.
Thank you for pointing it out.
> modprobes separately like this necessary to reproduce this bug?
No. It isn't.
> If it reproduces without manually loading and unloading the "torture"
> kernel module, could you please update the commit log to show that smaller
> reproducer?
To reproduce the bug, it needs to manually unload the "torture" but doesn't need
to manually load "torture".
I'll remove the unnecessary step "modprobe torture" from the commit message in the v2 patch.
> > // After some time
> > rmmod rcuscale
> > rmmod torture
> >
[...]
> > ---
> > kernel/rcu/rcuscale.c | 7 +++++++
> > 1 file changed, 7 insertions(+)
> >
> > diff --git a/kernel/rcu/rcuscale.c b/kernel/rcu/rcuscale.c index
> > 91fb5905a008..5e580cd08c58 100644
> > --- a/kernel/rcu/rcuscale.c
> > +++ b/kernel/rcu/rcuscale.c
> > @@ -522,6 +522,8 @@ rcu_scale_print_module_parms(struct
> rcu_scale_ops *cur_ops, const char *tag)
> > scale_type, tag, nrealreaders, nrealwriters, verbose,
> shutdown);
> > }
> >
> > +static void kfree_scale_cleanup(void);
> > +
>
> I do applaud minmimizing the size of the patch, but in this case could you
> please pull the kfree_scale_cleanup() function ahead of its first use?
>
I thought about it, but was also concerned that would make the patch bigger
while the effective changes were just only a few lines ...
Pulling the kfree_scale_cleanup() function ahead of rcu_scale_cleanup() also needs
to pull another two kfree_* variables ahead used by kfree_scale_cleanup().
This looks like will mess up kfree_* and rcu_scale_* functions/variables in the source file.
How about to pull the rcu_scale_cleanup() function after kfree_scale_cleanup().
This groups kfree_* functions and groups rcu_scale_* functions.
Then the code would look cleaner.
So, do you think the changes below are better?
---
kernel/rcu/rcuscale.c | 201 ++++++++++++++++++++++--------------------
1 file changed, 103 insertions(+), 98 deletions(-)
diff --git a/kernel/rcu/rcuscale.c b/kernel/rcu/rcuscale.c
index 91fb5905a008..5a000d26f03e 100644
--- a/kernel/rcu/rcuscale.c
+++ b/kernel/rcu/rcuscale.c
@@ -522,89 +522,6 @@ rcu_scale_print_module_parms(struct rcu_scale_ops *cur_ops, const char *tag)
scale_type, tag, nrealreaders, nrealwriters, verbose, shutdown);
}
-static void
-rcu_scale_cleanup(void)
-{
- int i;
- int j;
- int ngps = 0;
- u64 *wdp;
- u64 *wdpp;
-
- /*
- * Would like warning at start, but everything is expedited
- * during the mid-boot phase, so have to wait till the end.
- */
- if (rcu_gp_is_expedited() && !rcu_gp_is_normal() && !gp_exp)
- SCALEOUT_ERRSTRING("All grace periods expedited, no normal ones to measure!");
- if (rcu_gp_is_normal() && gp_exp)
- SCALEOUT_ERRSTRING("All grace periods normal, no expedited ones to measure!");
- if (gp_exp && gp_async)
- SCALEOUT_ERRSTRING("No expedited async GPs, so went with async!");
-
- if (torture_cleanup_begin())
- return;
- if (!cur_ops) {
- torture_cleanup_end();
- return;
- }
-
- if (reader_tasks) {
- for (i = 0; i < nrealreaders; i++)
- torture_stop_kthread(rcu_scale_reader,
- reader_tasks[i]);
- kfree(reader_tasks);
- }
-
- if (writer_tasks) {
- for (i = 0; i < nrealwriters; i++) {
- torture_stop_kthread(rcu_scale_writer,
- writer_tasks[i]);
- if (!writer_n_durations)
- continue;
- j = writer_n_durations[i];
- pr_alert("%s%s writer %d gps: %d\n",
- scale_type, SCALE_FLAG, i, j);
- ngps += j;
- }
- pr_alert("%s%s start: %llu end: %llu duration: %llu gps: %d batches: %ld\n",
- scale_type, SCALE_FLAG,
- t_rcu_scale_writer_started, t_rcu_scale_writer_finished,
- t_rcu_scale_writer_finished -
- t_rcu_scale_writer_started,
- ngps,
- rcuscale_seq_diff(b_rcu_gp_test_finished,
- b_rcu_gp_test_started));
- for (i = 0; i < nrealwriters; i++) {
- if (!writer_durations)
- break;
- if (!writer_n_durations)
- continue;
- wdpp = writer_durations[i];
- if (!wdpp)
- continue;
- for (j = 0; j < writer_n_durations[i]; j++) {
- wdp = &wdpp[j];
- pr_alert("%s%s %4d writer-duration: %5d %llu\n",
- scale_type, SCALE_FLAG,
- i, j, *wdp);
- if (j % 100 == 0)
- schedule_timeout_uninterruptible(1);
- }
- kfree(writer_durations[i]);
- }
- kfree(writer_tasks);
- kfree(writer_durations);
- kfree(writer_n_durations);
- }
-
- /* Do torture-type-specific cleanup operations. */
- if (cur_ops->cleanup != NULL)
- cur_ops->cleanup();
-
- torture_cleanup_end();
-}
-
/*
* Return the number if non-negative. If -1, the number of CPUs.
* If less than -1, that much less than the number of CPUs, but
@@ -624,21 +541,6 @@ static int compute_real(int n)
return nr;
}
-/*
- * RCU scalability shutdown kthread. Just waits to be awakened, then shuts
- * down system.
- */
-static int
-rcu_scale_shutdown(void *arg)
-{
- wait_event(shutdown_wq,
- atomic_read(&n_rcu_scale_writer_finished) >= nrealwriters);
- smp_mb(); /* Wake before output. */
- rcu_scale_cleanup();
- kernel_power_off();
- return -EINVAL;
-}
-
/*
* kfree_rcu() scalability tests: Start a kfree_rcu() loop on all CPUs for number
* of iterations and measure total time and number of GP for all iterations to complete.
@@ -875,6 +777,109 @@ kfree_scale_init(void)
return firsterr;
}
+static void
+rcu_scale_cleanup(void)
+{
+ int i;
+ int j;
+ int ngps = 0;
+ u64 *wdp;
+ u64 *wdpp;
+
+ /*
+ * Would like warning at start, but everything is expedited
+ * during the mid-boot phase, so have to wait till the end.
+ */
+ if (rcu_gp_is_expedited() && !rcu_gp_is_normal() && !gp_exp)
+ SCALEOUT_ERRSTRING("All grace periods expedited, no normal ones to measure!");
+ if (rcu_gp_is_normal() && gp_exp)
+ SCALEOUT_ERRSTRING("All grace periods normal, no expedited ones to measure!");
+ if (gp_exp && gp_async)
+ SCALEOUT_ERRSTRING("No expedited async GPs, so went with async!");
+
+ if (kfree_rcu_test) {
+ kfree_scale_cleanup();
+ return;
+ }
+
+ if (torture_cleanup_begin())
+ return;
+ if (!cur_ops) {
+ torture_cleanup_end();
+ return;
+ }
+
+ if (reader_tasks) {
+ for (i = 0; i < nrealreaders; i++)
+ torture_stop_kthread(rcu_scale_reader,
+ reader_tasks[i]);
+ kfree(reader_tasks);
+ }
+
+ if (writer_tasks) {
+ for (i = 0; i < nrealwriters; i++) {
+ torture_stop_kthread(rcu_scale_writer,
+ writer_tasks[i]);
+ if (!writer_n_durations)
+ continue;
+ j = writer_n_durations[i];
+ pr_alert("%s%s writer %d gps: %d\n",
+ scale_type, SCALE_FLAG, i, j);
+ ngps += j;
+ }
+ pr_alert("%s%s start: %llu end: %llu duration: %llu gps: %d batches: %ld\n",
+ scale_type, SCALE_FLAG,
+ t_rcu_scale_writer_started, t_rcu_scale_writer_finished,
+ t_rcu_scale_writer_finished -
+ t_rcu_scale_writer_started,
+ ngps,
+ rcuscale_seq_diff(b_rcu_gp_test_finished,
+ b_rcu_gp_test_started));
+ for (i = 0; i < nrealwriters; i++) {
+ if (!writer_durations)
+ break;
+ if (!writer_n_durations)
+ continue;
+ wdpp = writer_durations[i];
+ if (!wdpp)
+ continue;
+ for (j = 0; j < writer_n_durations[i]; j++) {
+ wdp = &wdpp[j];
+ pr_alert("%s%s %4d writer-duration: %5d %llu\n",
+ scale_type, SCALE_FLAG,
+ i, j, *wdp);
+ if (j % 100 == 0)
+ schedule_timeout_uninterruptible(1);
+ }
+ kfree(writer_durations[i]);
+ }
+ kfree(writer_tasks);
+ kfree(writer_durations);
+ kfree(writer_n_durations);
+ }
+
+ /* Do torture-type-specific cleanup operations. */
+ if (cur_ops->cleanup != NULL)
+ cur_ops->cleanup();
+
+ torture_cleanup_end();
+}
+
+/*
+ * RCU scalability shutdown kthread. Just waits to be awakened, then shuts
+ * down system.
+ */
+static int
+rcu_scale_shutdown(void *arg)
+{
+ wait_event(shutdown_wq,
+ atomic_read(&n_rcu_scale_writer_finished) >= nrealwriters);
+ smp_mb(); /* Wake before output. */
+ rcu_scale_cleanup();
+ kernel_power_off();
+ return -EINVAL;
+}
+
static int __init
rcu_scale_init(void)
{
--
2.17.1
> [...]
