From: Johannes Berg <johannes.berg@xxxxxxxxx>
[ Upstream commit ebef8ea2ba967026192a26f4529890893919bc57 ]
As the comment here indicates, we need to do the polling in the
idle loop without blocking interrupts, since interrupts can be
vhost-user messages that we must process even while in our idle
loop.
I don't know why I explained one thing and implemented another,
but we have indeed observed random hangs due to this, depending
on the timing of the messages.
Fixes: 88ce64249233 ("um: Implement time-travel=ext")
Signed-off-by: Johannes Berg <johannes.berg@xxxxxxxxx>
Acked-By: Anton Ivanov <anton.ivanov@xxxxxxxxxxxxxxxxxx>
Signed-off-by: Richard Weinberger <richard@xxxxxx>
Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx>
---
arch/um/kernel/time.c | 14 +++++++++-----
1 file changed, 9 insertions(+), 5 deletions(-)
diff --git a/arch/um/kernel/time.c b/arch/um/kernel/time.c
index 25eaa6a0c6583..c07436e89e599 100644
--- a/arch/um/kernel/time.c
+++ b/arch/um/kernel/time.c
@@ -70,13 +70,17 @@ static void time_travel_handle_message(struct um_timetravel_msg *msg,
* read of the message and write of the ACK.
*/
if (mode != TTMH_READ) {
+ bool disabled = irqs_disabled();
+
+ BUG_ON(mode == TTMH_IDLE && !disabled);
+
+ if (disabled)
+ local_irq_enable();
while (os_poll(1, &time_travel_ext_fd) != 0) {
- if (mode == TTMH_IDLE) {
- BUG_ON(!irqs_disabled());
- local_irq_enable();
- local_irq_disable();
- }
+ /* nothing */
}
+ if (disabled)
+ local_irq_disable();
}
ret = os_read_file(time_travel_ext_fd, msg, sizeof(*msg));
--
2.25.1
