summaryrefslogtreecommitdiff
path: root/net/core/dev.c
diff options
context:
space:
mode:
authorSebastian Andrzej Siewior <bigeasy@linutronix.de>2021-12-13 13:53:29 +0300
committerDavid S. Miller <davem@davemloft.net>2021-12-13 17:57:03 +0300
commit64445dda9d8384975eca54e3f01886fca61e1db6 (patch)
tree16dda030b630cf2f363e0ab71e08322f6a994393 /net/core/dev.c
parent93d576f54e0f697c7b54f4ddd718e68ccc52c4f2 (diff)
downloadlinux-64445dda9d8384975eca54e3f01886fca61e1db6.tar.xz
net: dev: Always serialize on Qdisc::busylock in __dev_xmit_skb() on PREEMPT_RT.
The root-lock is dropped before dev_hard_start_xmit() is invoked and after setting the __QDISC___STATE_RUNNING bit. If the Qdisc owner is preempted by another sender/task with a higher priority then this new sender won't be able to submit packets to the NIC directly instead they will be enqueued into the Qdisc. The NIC will remain idle until the Qdisc owner is scheduled again and finishes the job. By serializing every task on the ->busylock then the task will be preempted by a sender only after the Qdisc has no owner. Always serialize on the busylock on PREEMPT_RT. Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'net/core/dev.c')
-rw-r--r--net/core/dev.c6
1 files changed, 5 insertions, 1 deletions
diff --git a/net/core/dev.c b/net/core/dev.c
index 4420086f3aeb..53bef2aae378 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -3719,8 +3719,12 @@ no_lock_out:
* separate lock before trying to get qdisc main lock.
* This permits qdisc->running owner to get the lock more
* often and dequeue packets faster.
+ * On PREEMPT_RT it is possible to preempt the qdisc owner during xmit
+ * and then other tasks will only enqueue packets. The packets will be
+ * sent after the qdisc owner is scheduled again. To prevent this
+ * scenario the task always serialize on the lock.
*/
- contended = qdisc_is_running(q);
+ contended = IS_ENABLED(CONFIG_PREEMPT_RT) || qdisc_is_running(q);
if (unlikely(contended))
spin_lock(&q->busylock);