Use unbound workqueue so that work is automatically ballanced between
available CPUs.
Note: workqueue implementation is somehow buggy, so it still does not yield
full parallelization across multiple CPUs.
Signed-off-by: Mikulas Patocka <mpatocka@xxxxxxxxxx>
---
drivers/md/dm-crypt.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c
index 7a3e8eb..4c8b4e0 100644
--- a/drivers/md/dm-crypt.c
+++ b/drivers/md/dm-crypt.c
@@ -1649,8 +1649,9 @@ static int crypt_ctr(struct dm_target *ti, unsigned int argc, char **argv)
cc->crypt_queue = alloc_workqueue("kcryptd",
WQ_NON_REENTRANT|
WQ_CPU_INTENSIVE|
- WQ_MEM_RECLAIM,
- 1);
+ WQ_MEM_RECLAIM|
+ WQ_UNBOUND,
+ num_online_cpus());
if (!cc->crypt_queue) {
ti->error = "Couldn't create kcryptd queue";
goto bad;
--
1.7.10.4
--
dm-devel mailing list
dm-devel@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/dm-devel
