[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Qemu-block] [PULL 6/8] aio-posix: ensure poll mode is left when aio_not
From: |
Stefan Hajnoczi |
Subject: |
[Qemu-block] [PULL 6/8] aio-posix: ensure poll mode is left when aio_notify is called |
Date: |
Fri, 10 May 2019 14:02:41 +0100 |
From: Paolo Bonzini <address@hidden>
With aio=thread, adaptive polling makes latency worse rather than
better, because it delays the execution of the ThreadPool's
completion bottom half.
event_notifier_poll() does run while polling, detecting that
a bottom half was scheduled by a worker thread, but because
ctx->notifier is explicitly ignored in run_poll_handlers_once(),
scheduling the BH does not count as making progress and
run_poll_handlers() keeps running. Fix this by recomputing
the deadline after *timeout could have changed.
With this change, ThreadPool still cannot participate in polling
but at least it does not suffer from extra latency.
Reported-by: Sergio Lopez <address@hidden>
Signed-off-by: Paolo Bonzini <address@hidden>
Message-id: address@hidden
Cc: Stefan Hajnoczi <address@hidden>
Cc: Kevin Wolf <address@hidden>
Cc: address@hidden
Signed-off-by: Paolo Bonzini <address@hidden>
Message-Id: <address@hidden>
Signed-off-by: Paolo Bonzini <address@hidden>
Message-Id: <address@hidden>
Signed-off-by: Stefan Hajnoczi <address@hidden>
---
util/aio-posix.c | 12 ++++++++----
1 file changed, 8 insertions(+), 4 deletions(-)
diff --git a/util/aio-posix.c b/util/aio-posix.c
index 6fbfa7924f..db11021287 100644
--- a/util/aio-posix.c
+++ b/util/aio-posix.c
@@ -519,6 +519,10 @@ static bool run_poll_handlers_once(AioContext *ctx,
int64_t *timeout)
if (!node->deleted && node->io_poll &&
aio_node_check(ctx, node->is_external) &&
node->io_poll(node->opaque)) {
+ /*
+ * Polling was successful, exit try_poll_mode immediately
+ * to adjust the next polling time.
+ */
*timeout = 0;
if (node->opaque != &ctx->notifier) {
progress = true;
@@ -558,8 +562,9 @@ static bool run_poll_handlers(AioContext *ctx, int64_t
max_ns, int64_t *timeout)
do {
progress = run_poll_handlers_once(ctx, timeout);
elapsed_time = qemu_clock_get_ns(QEMU_CLOCK_REALTIME) - start_time;
- } while (!progress && elapsed_time < max_ns
- && !atomic_read(&ctx->poll_disable_cnt));
+ max_ns = qemu_soonest_timeout(*timeout, max_ns);
+ assert(!(max_ns && progress));
+ } while (elapsed_time < max_ns && !atomic_read(&ctx->poll_disable_cnt));
/* If time has passed with no successful polling, adjust *timeout to
* keep the same ending time.
@@ -585,8 +590,7 @@ static bool run_poll_handlers(AioContext *ctx, int64_t
max_ns, int64_t *timeout)
*/
static bool try_poll_mode(AioContext *ctx, int64_t *timeout)
{
- /* See qemu_soonest_timeout() uint64_t hack */
- int64_t max_ns = MIN((uint64_t)*timeout, (uint64_t)ctx->poll_ns);
+ int64_t max_ns = qemu_soonest_timeout(*timeout, ctx->poll_ns);
if (max_ns && !atomic_read(&ctx->poll_disable_cnt)) {
poll_set_started(ctx, true);
--
2.21.0
- [Qemu-block] [PULL 0/8] Block patches, Stefan Hajnoczi, 2019/05/10
- [Qemu-block] [PULL 1/8] util/readline: add a space to fix errors by checkpatch tool, Stefan Hajnoczi, 2019/05/10
- [Qemu-block] [PULL 2/8] util: readline: replace tab indent by four spaces to fix checkpatch errors, Stefan Hajnoczi, 2019/05/10
- [Qemu-block] [PULL 3/8] util/readline: Add braces to fix checkpatch errors, Stefan Hajnoczi, 2019/05/10
- [Qemu-block] [PULL 4/8] block: Add coroutine_fn to bdrv_check_co_entry, Stefan Hajnoczi, 2019/05/10
- [Qemu-block] [PULL 5/8] block/io.c: fix for the allocation failure, Stefan Hajnoczi, 2019/05/10
- [Qemu-block] [PULL 6/8] aio-posix: ensure poll mode is left when aio_notify is called,
Stefan Hajnoczi <=
- [Qemu-block] [PULL 8/8] docs: add Security chapter to the documentation, Stefan Hajnoczi, 2019/05/10
- [Qemu-block] [PULL 7/8] docs: add Secure Coding Practices to developer docs, Stefan Hajnoczi, 2019/05/10
- Re: [Qemu-block] [PULL 0/8] Block patches, Peter Maydell, 2019/05/10