[<prev] [next>] [day] [month] [year] [list]
Message-ID: <160222533003.7002.8090034918982279919.tip-bot2@tip-bot2>
Date: Fri, 09 Oct 2020 06:35:30 -0000
From: "tip-bot2 for Paul E. McKenney" <tip-bot2@...utronix.de>
To: linux-tip-commits@...r.kernel.org
Cc: "Paul E. McKenney" <paulmck@...nel.org>, x86 <x86@...nel.org>,
LKML <linux-kernel@...r.kernel.org>
Subject: [tip: core/rcu] scftorture: Add smp_call_function_many()
memory-ordering checks
The following commit has been merged into the core/rcu branch of tip:
Commit-ID: 980205ee8489d53c4380f7762debac87312b0fb3
Gitweb: https://git.kernel.org/tip/980205ee8489d53c4380f7762debac87312b0fb3
Author: Paul E. McKenney <paulmck@...nel.org>
AuthorDate: Wed, 01 Jul 2020 12:30:02 -07:00
Committer: Paul E. McKenney <paulmck@...nel.org>
CommitterDate: Mon, 24 Aug 2020 18:38:34 -07:00
scftorture: Add smp_call_function_many() memory-ordering checks
This commit adds checks for memory misordering across calls to and
returns from smp_call_function_many() in the case where the caller waits.
Misordering results in a splat.
Note that in contrast to smp_call_function_single(), this code does not
test memory ordering into the handler in the no-wait case because none
of the handlers would be able to free the scf_check structure without
introducing heavy synchronization to work out which was last.
[ paulmck: s/GFP_KERNEL/GFP_ATOMIC/ per kernel test robot feedback. ]
Signed-off-by: Paul E. McKenney <paulmck@...nel.org>
---
kernel/scftorture.c | 26 +++++++++++++++++++++++---
1 file changed, 23 insertions(+), 3 deletions(-)
diff --git a/kernel/scftorture.c b/kernel/scftorture.c
index 9b42271..3519ad1 100644
--- a/kernel/scftorture.c
+++ b/kernel/scftorture.c
@@ -240,8 +240,11 @@ static void scf_handler(void *scfc_in)
unsigned long r = torture_random(this_cpu_ptr(&scf_torture_rand));
struct scf_check *scfcp = scfc_in;
- if (likely(scfcp) && WARN_ON_ONCE(unlikely(!READ_ONCE(scfcp->scfc_in))))
- atomic_inc(&n_mb_in_errs);
+ if (likely(scfcp)) {
+ WRITE_ONCE(scfcp->scfc_out, false); // For multiple receivers.
+ if (WARN_ON_ONCE(unlikely(!READ_ONCE(scfcp->scfc_in))))
+ atomic_inc(&n_mb_in_errs);
+ }
this_cpu_inc(scf_invoked_count);
if (longwait <= 0) {
if (!(r & 0xffc0))
@@ -325,11 +328,28 @@ static void scftorture_invoke_one(struct scf_statistics *scfp, struct torture_ra
}
break;
case SCF_PRIM_MANY:
+ if (scfsp->scfs_wait) {
+ scfcp = kmalloc(sizeof(*scfcp), GFP_ATOMIC);
+ if (WARN_ON_ONCE(!scfcp))
+ atomic_inc(&n_alloc_errs);
+ }
if (scfsp->scfs_wait)
scfp->n_many_wait++;
else
scfp->n_many++;
- smp_call_function_many(cpu_online_mask, scf_handler, NULL, scfsp->scfs_wait);
+ if (scfcp) {
+ scfcp->scfc_cpu = -1;
+ scfcp->scfc_wait = true;
+ scfcp->scfc_out = false;
+ scfcp->scfc_in = true;
+ }
+ smp_call_function_many(cpu_online_mask, scf_handler, scfcp, scfsp->scfs_wait);
+ if (scfcp) {
+ if (WARN_ON_ONCE(!scfcp->scfc_out))
+ atomic_inc(&n_mb_out_errs); // Leak rather than trash!
+ else
+ kfree(scfcp);
+ }
break;
case SCF_PRIM_ALL:
if (scfsp->scfs_wait)
Powered by blists - more mailing lists