From: Stefan Hajnoczi <stefanha@redhat.com>
To: qemu-devel@nongnu.org
Cc: "Fam Zheng" <fam@euphon.net>, "Kevin Wolf" <kwolf@redhat.com>,
qemu-block@nongnu.org, "Max Reitz" <mreitz@redhat.com>,
"Stefan Hajnoczi" <stefanha@redhat.com>,
"Marc-André Lureau" <marcandre.lureau@redhat.com>,
"Paolo Bonzini" <pbonzini@redhat.com>
Subject: [PATCH 5/5] aio-posix: make AioHandler dispatch O(1) with epoll
Date: Fri, 14 Feb 2020 17:17:12 +0000 [thread overview]
Message-ID: <20200214171712.541358-6-stefanha@redhat.com> (raw)
In-Reply-To: <20200214171712.541358-1-stefanha@redhat.com>
File descriptor monitoring is O(1) with epoll(7), but
aio_dispatch_handlers() still scans all AioHandlers instead of
dispatching just those that are ready. This makes aio_poll() O(n) with
respect to the total number of registered handlers.
Add a local ready_list to aio_poll() so that each nested aio_poll()
builds a list of handlers ready to be dispatched. Since file descriptor
polling is level-triggered, nested aio_poll() calls also see fds that
were ready in the parent but not yet dispatched. This guarantees that
nested aio_poll() invocations will dispatch all fds, even those that
became ready before the nested invocation.
Since only handlers ready to be dispatched are placed onto the
ready_list, the new aio_dispatch_ready_handlers() function provides O(1)
dispatch.
Note that AioContext polling is still O(n) and currently cannot be fully
disabled. This still needs to be fixed before aio_poll() is fully O(1).
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
---
util/aio-posix.c | 106 +++++++++++++++++++++++++++++++++--------------
1 file changed, 76 insertions(+), 30 deletions(-)
diff --git a/util/aio-posix.c b/util/aio-posix.c
index 3a98a2acb9..dc33ca08a6 100644
--- a/util/aio-posix.c
+++ b/util/aio-posix.c
@@ -34,6 +34,7 @@ struct AioHandler
void *opaque;
bool is_external;
QLIST_ENTRY(AioHandler) node;
+ QLIST_ENTRY(AioHandler) node_ready; /* only used during aio_poll() */
QLIST_ENTRY(AioHandler) node_deleted;
};
@@ -104,7 +105,18 @@ static void aio_epoll_update(AioContext *ctx, AioHandler *node, bool is_new)
}
}
-static int aio_epoll(AioContext *ctx, int64_t timeout)
+/* Add a handler to a ready list */
+static void add_ready_handler(AioHandlerList *ready_list,
+ AioHandler *node,
+ int revents)
+{
+ QLIST_SAFE_REMOVE(node, node_ready); /* remove from nested parent's list */
+ node->pfd.revents = revents;
+ QLIST_INSERT_HEAD(ready_list, node, node_ready);
+}
+
+static int aio_epoll(AioContext *ctx, AioHandlerList *ready_list,
+ int64_t timeout)
{
GPollFD pfd = {
.fd = ctx->epollfd,
@@ -129,11 +141,13 @@ static int aio_epoll(AioContext *ctx, int64_t timeout)
}
for (i = 0; i < ret; i++) {
int ev = events[i].events;
+ int revents = (ev & EPOLLIN ? G_IO_IN : 0) |
+ (ev & EPOLLOUT ? G_IO_OUT : 0) |
+ (ev & EPOLLHUP ? G_IO_HUP : 0) |
+ (ev & EPOLLERR ? G_IO_ERR : 0);
+
node = events[i].data.ptr;
- node->pfd.revents = (ev & EPOLLIN ? G_IO_IN : 0) |
- (ev & EPOLLOUT ? G_IO_OUT : 0) |
- (ev & EPOLLHUP ? G_IO_HUP : 0) |
- (ev & EPOLLERR ? G_IO_ERR : 0);
+ add_ready_handler(ready_list, node, revents);
}
}
out:
@@ -437,36 +451,63 @@ static void aio_free_deleted_handlers(AioContext *ctx)
qemu_lockcnt_inc_and_unlock(&ctx->list_lock);
}
-static bool aio_dispatch_handlers(AioContext *ctx)
+static bool aio_dispatch_handler(AioContext *ctx, AioHandler *node)
{
- AioHandler *node, *tmp;
bool progress = false;
+ int revents;
- QLIST_FOREACH_SAFE_RCU(node, &ctx->aio_handlers, node, tmp) {
- int revents;
+ revents = node->pfd.revents & node->pfd.events;
+ node->pfd.revents = 0;
- revents = node->pfd.revents & node->pfd.events;
- node->pfd.revents = 0;
+ if (!QLIST_IS_INSERTED(node, node_deleted) &&
+ (revents & (G_IO_IN | G_IO_HUP | G_IO_ERR)) &&
+ aio_node_check(ctx, node->is_external) &&
+ node->io_read) {
+ node->io_read(node->opaque);
- if (!QLIST_IS_INSERTED(node, node_deleted) &&
- (revents & (G_IO_IN | G_IO_HUP | G_IO_ERR)) &&
- aio_node_check(ctx, node->is_external) &&
- node->io_read) {
- node->io_read(node->opaque);
-
- /* aio_notify() does not count as progress */
- if (node->opaque != &ctx->notifier) {
- progress = true;
- }
- }
- if (!QLIST_IS_INSERTED(node, node_deleted) &&
- (revents & (G_IO_OUT | G_IO_ERR)) &&
- aio_node_check(ctx, node->is_external) &&
- node->io_write) {
- node->io_write(node->opaque);
+ /* aio_notify() does not count as progress */
+ if (node->opaque != &ctx->notifier) {
progress = true;
}
}
+ if (!QLIST_IS_INSERTED(node, node_deleted) &&
+ (revents & (G_IO_OUT | G_IO_ERR)) &&
+ aio_node_check(ctx, node->is_external) &&
+ node->io_write) {
+ node->io_write(node->opaque);
+ progress = true;
+ }
+
+ return progress;
+}
+
+/*
+ * If we have a list of ready handlers then this is more efficient than
+ * scanning all handlers with aio_dispatch_handlers().
+ */
+static bool aio_dispatch_ready_handlers(AioContext *ctx,
+ AioHandlerList *ready_list)
+{
+ bool progress = false;
+ AioHandler *node;
+
+ while ((node = QLIST_FIRST(ready_list))) {
+ QLIST_SAFE_REMOVE(node, node_ready);
+ progress = aio_dispatch_handler(ctx, node) || progress;
+ }
+
+ return progress;
+}
+
+/* Slower than aio_dispatch_ready_handlers() but only used via glib */
+static bool aio_dispatch_handlers(AioContext *ctx)
+{
+ AioHandler *node, *tmp;
+ bool progress = false;
+
+ QLIST_FOREACH_SAFE_RCU(node, &ctx->aio_handlers, node, tmp) {
+ progress = aio_dispatch_handler(ctx, node) || progress;
+ }
return progress;
}
@@ -628,6 +669,7 @@ static bool try_poll_mode(AioContext *ctx, int64_t *timeout)
bool aio_poll(AioContext *ctx, bool blocking)
{
+ AioHandlerList ready_list = QLIST_HEAD_INITIALIZER(ready_list);
AioHandler *node;
int i;
int ret = 0;
@@ -678,7 +720,7 @@ bool aio_poll(AioContext *ctx, bool blocking)
/* wait until next event */
if (aio_epoll_check_poll(ctx, pollfds, npfd, timeout)) {
npfd = 0; /* pollfds[] is not being used */
- ret = aio_epoll(ctx, timeout);
+ ret = aio_epoll(ctx, &ready_list, timeout);
} else {
ret = qemu_poll_ns(pollfds, npfd, timeout);
}
@@ -733,7 +775,11 @@ bool aio_poll(AioContext *ctx, bool blocking)
/* if we have any readable fds, dispatch event */
if (ret > 0) {
for (i = 0; i < npfd; i++) {
- nodes[i]->pfd.revents = pollfds[i].revents;
+ int revents = pollfds[i].revents;
+
+ if (revents) {
+ add_ready_handler(&ready_list, nodes[i], revents);
+ }
}
}
@@ -742,7 +788,7 @@ bool aio_poll(AioContext *ctx, bool blocking)
progress |= aio_bh_poll(ctx);
if (ret > 0) {
- progress |= aio_dispatch_handlers(ctx);
+ progress |= aio_dispatch_ready_handlers(ctx, &ready_list);
}
aio_free_deleted_handlers(ctx);
--
2.24.1
next prev parent reply other threads:[~2020-02-14 17:23 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-02-14 17:17 [PATCH 0/5] aio-posix: towards an O(1) event loop Stefan Hajnoczi
2020-02-14 17:17 ` [PATCH 1/5] aio-posix: fix use after leaving scope in aio_poll() Stefan Hajnoczi
2020-02-19 7:02 ` Sergio Lopez
2020-02-14 17:17 ` [PATCH 2/5] aio-posix: don't pass ns timeout to epoll_wait() Stefan Hajnoczi
2020-02-19 10:12 ` Sergio Lopez
2020-02-14 17:17 ` [PATCH 3/5] qemu/queue.h: add QLIST_SAFE_REMOVE() Stefan Hajnoczi
2020-02-19 10:30 ` Sergio Lopez
2020-02-14 17:17 ` [PATCH 4/5] aio-posix: make AioHandler deletion O(1) Stefan Hajnoczi
2020-02-19 10:41 ` Sergio Lopez
2020-02-14 17:17 ` Stefan Hajnoczi [this message]
2020-02-19 11:00 ` [PATCH 5/5] aio-posix: make AioHandler dispatch O(1) with epoll Sergio Lopez
2020-02-19 11:13 ` Paolo Bonzini
2020-02-21 12:59 ` Stefan Hajnoczi
2020-02-21 13:06 ` Paolo Bonzini
2020-02-21 14:44 ` Stefan Hajnoczi
2020-02-21 14:47 ` Stefan Hajnoczi
2020-02-21 15:04 ` Paolo Bonzini
2020-02-21 15:29 ` Stefan Hajnoczi
2020-02-21 15:37 ` Paolo Bonzini
2020-02-21 15:29 ` [PATCH 0/5] aio-posix: towards an O(1) event loop Stefan Hajnoczi
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20200214171712.541358-6-stefanha@redhat.com \
--to=stefanha@redhat.com \
--cc=fam@euphon.net \
--cc=kwolf@redhat.com \
--cc=marcandre.lureau@redhat.com \
--cc=mreitz@redhat.com \
--cc=pbonzini@redhat.com \
--cc=qemu-block@nongnu.org \
--cc=qemu-devel@nongnu.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).