From: Stefan Hajnoczi <stefanha@redhat.com>
To: linux-fsdevel@vger.kernel.org
Cc: kvm@vger.kernel.org, linux-kernel@vger.kernel.org,
Avi Kivity <avi@scylladb.com>,
Davide Libenzi <davidel@xmailserver.org>,
Alexander Viro <viro@zeniv.linux.org.uk>,
Masatake YAMATO <yamato@redhat.com>
Subject: Re: [RFC] eventfd: add EFD_AUTORESET flag
Date: Tue, 4 Feb 2020 15:40:35 +0000 [thread overview]
Message-ID: <20200204154035.GA47059@stefanha-x1.localdomain> (raw)
In-Reply-To: <20200129172010.162215-1-stefanha@redhat.com>
[-- Attachment #1: Type: text/plain, Size: 7000 bytes --]
On Wed, Jan 29, 2020 at 05:20:10PM +0000, Stefan Hajnoczi wrote:
> Some applications simply use eventfd for inter-thread notifications
> without requiring counter or semaphore semantics. They wait for the
> eventfd to become readable using poll(2)/select(2) and then call read(2)
> to reset the counter.
>
> This patch adds the EFD_AUTORESET flag to reset the counter when
> f_ops->poll() finds the eventfd is readable, eliminating the need to
> call read(2) to reset the counter.
>
> This results in a small but measurable 1% performance improvement with
> QEMU virtio-blk emulation. Each read(2) takes 1 microsecond execution
> time in the event loop according to perf.
>
> Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
> ---
> Does this look like a reasonable thing to do? I'm not very familiar
> with f_ops->poll() or the eventfd internals, so maybe I'm overlooking a
> design flaw.
Ping?
> I've tested this with QEMU and it works fine:
> https://github.com/stefanha/qemu/commits/eventfd-autoreset
> ---
> fs/eventfd.c | 99 +++++++++++++++++++++++++----------------
> include/linux/eventfd.h | 3 +-
> 2 files changed, 62 insertions(+), 40 deletions(-)
>
> diff --git a/fs/eventfd.c b/fs/eventfd.c
> index 8aa0ea8c55e8..208f6b9e2234 100644
> --- a/fs/eventfd.c
> +++ b/fs/eventfd.c
> @@ -116,45 +116,62 @@ static __poll_t eventfd_poll(struct file *file, poll_table *wait)
>
> poll_wait(file, &ctx->wqh, wait);
>
> - /*
> - * All writes to ctx->count occur within ctx->wqh.lock. This read
> - * can be done outside ctx->wqh.lock because we know that poll_wait
> - * takes that lock (through add_wait_queue) if our caller will sleep.
> - *
> - * The read _can_ therefore seep into add_wait_queue's critical
> - * section, but cannot move above it! add_wait_queue's spin_lock acts
> - * as an acquire barrier and ensures that the read be ordered properly
> - * against the writes. The following CAN happen and is safe:
> - *
> - * poll write
> - * ----------------- ------------
> - * lock ctx->wqh.lock (in poll_wait)
> - * count = ctx->count
> - * __add_wait_queue
> - * unlock ctx->wqh.lock
> - * lock ctx->qwh.lock
> - * ctx->count += n
> - * if (waitqueue_active)
> - * wake_up_locked_poll
> - * unlock ctx->qwh.lock
> - * eventfd_poll returns 0
> - *
> - * but the following, which would miss a wakeup, cannot happen:
> - *
> - * poll write
> - * ----------------- ------------
> - * count = ctx->count (INVALID!)
> - * lock ctx->qwh.lock
> - * ctx->count += n
> - * **waitqueue_active is false**
> - * **no wake_up_locked_poll!**
> - * unlock ctx->qwh.lock
> - * lock ctx->wqh.lock (in poll_wait)
> - * __add_wait_queue
> - * unlock ctx->wqh.lock
> - * eventfd_poll returns 0
> - */
> - count = READ_ONCE(ctx->count);
> + if (ctx->flags & EFD_AUTORESET) {
> + unsigned long flags;
> + __poll_t requested = poll_requested_events(wait);
> +
> + spin_lock_irqsave(&ctx->wqh.lock, flags);
> + count = ctx->count;
> +
> + /* Reset counter if caller is polling for read */
> + if (count != 0 && (requested & EPOLLIN)) {
> + ctx->count = 0;
> + events |= EPOLLOUT;
> + /* TODO is a EPOLLOUT wakeup necessary here? */
> + }
> +
> + spin_unlock_irqrestore(&ctx->wqh.lock, flags);
> + } else {
> + /*
> + * All writes to ctx->count occur within ctx->wqh.lock. This read
> + * can be done outside ctx->wqh.lock because we know that poll_wait
> + * takes that lock (through add_wait_queue) if our caller will sleep.
> + *
> + * The read _can_ therefore seep into add_wait_queue's critical
> + * section, but cannot move above it! add_wait_queue's spin_lock acts
> + * as an acquire barrier and ensures that the read be ordered properly
> + * against the writes. The following CAN happen and is safe:
> + *
> + * poll write
> + * ----------------- ------------
> + * lock ctx->wqh.lock (in poll_wait)
> + * count = ctx->count
> + * __add_wait_queue
> + * unlock ctx->wqh.lock
> + * lock ctx->qwh.lock
> + * ctx->count += n
> + * if (waitqueue_active)
> + * wake_up_locked_poll
> + * unlock ctx->qwh.lock
> + * eventfd_poll returns 0
> + *
> + * but the following, which would miss a wakeup, cannot happen:
> + *
> + * poll write
> + * ----------------- ------------
> + * count = ctx->count (INVALID!)
> + * lock ctx->qwh.lock
> + * ctx->count += n
> + * **waitqueue_active is false**
> + * **no wake_up_locked_poll!**
> + * unlock ctx->qwh.lock
> + * lock ctx->wqh.lock (in poll_wait)
> + * __add_wait_queue
> + * unlock ctx->wqh.lock
> + * eventfd_poll returns 0
> + */
> + count = READ_ONCE(ctx->count);
> + }
>
> if (count > 0)
> events |= EPOLLIN;
> @@ -400,6 +417,10 @@ static int do_eventfd(unsigned int count, int flags)
> if (flags & ~EFD_FLAGS_SET)
> return -EINVAL;
>
> + /* Semaphore semantics don't make sense when autoreset is enabled */
> + if ((flags & EFD_SEMAPHORE) && (flags & EFD_AUTORESET))
> + return -EINVAL;
> +
> ctx = kmalloc(sizeof(*ctx), GFP_KERNEL);
> if (!ctx)
> return -ENOMEM;
> diff --git a/include/linux/eventfd.h b/include/linux/eventfd.h
> index ffcc7724ca21..27577fafc553 100644
> --- a/include/linux/eventfd.h
> +++ b/include/linux/eventfd.h
> @@ -21,11 +21,12 @@
> * shared O_* flags.
> */
> #define EFD_SEMAPHORE (1 << 0)
> +#define EFD_AUTORESET (1 << 6) /* aliases O_CREAT */
> #define EFD_CLOEXEC O_CLOEXEC
> #define EFD_NONBLOCK O_NONBLOCK
>
> #define EFD_SHARED_FCNTL_FLAGS (O_CLOEXEC | O_NONBLOCK)
> -#define EFD_FLAGS_SET (EFD_SHARED_FCNTL_FLAGS | EFD_SEMAPHORE)
> +#define EFD_FLAGS_SET (EFD_SHARED_FCNTL_FLAGS | EFD_SEMAPHORE | EFD_AUTORESET)
>
> struct eventfd_ctx;
> struct file;
> --
> 2.24.1
>
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]
next prev parent reply other threads:[~2020-02-04 15:40 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-01-29 17:20 [RFC] eventfd: add EFD_AUTORESET flag Stefan Hajnoczi
2020-02-04 15:40 ` Stefan Hajnoczi [this message]
2020-02-11 9:32 ` Stefan Hajnoczi
2020-02-12 8:31 ` Paolo Bonzini
2020-02-12 10:10 ` Avi Kivity
2020-02-12 10:29 ` Stefan Hajnoczi
2020-02-12 10:47 ` Paolo Bonzini
2020-02-12 10:54 ` Avi Kivity
2020-02-19 10:37 ` Stefan Hajnoczi
2020-02-19 10:43 ` Avi Kivity
2020-02-19 11:10 ` Paolo Bonzini
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=20200204154035.GA47059@stefanha-x1.localdomain \
--to=stefanha@redhat.com \
--cc=avi@scylladb.com \
--cc=davidel@xmailserver.org \
--cc=kvm@vger.kernel.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=viro@zeniv.linux.org.uk \
--cc=yamato@redhat.com \
/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).