From: Stefano Brivio <sbrivio@redhat.com>
To: David Gibson <david@gibson.dropbear.id.au>
Cc: passt-dev@passt.top
Subject: Re: [PATCH 5/6] tap: Re-introduce EPOLLET for tap connections
Date: Tue, 3 Sep 2024 21:25:50 +0200 [thread overview]
Message-ID: <20240903212550.05900f14@elisabeth> (raw)
In-Reply-To: <20240903120235.1688429-6-david@gibson.dropbear.id.au>
On Tue, 3 Sep 2024 22:02:34 +1000
David Gibson <david@gibson.dropbear.id.au> wrote:
> Since 4684f603446b ("tap: Don't use EPOLLET on Qemu sockets") we've only
> used level-triggered events for the tap device. Prior to that we used it
> inconsistently which caused some problems.
It didn't actually cause any issue according to your commit message for
4684f603446b itself.
> We want to add support for EPOLLOUT events on the tap connection, and
> without EPOLLET that would require toggling EPOLLOUT on and off, which is
> awkward. So, re-introduce EPOLLET, but now use it uniformly for all tap
> modes. The main change this requires is making sure on EPOLLIN we loop
> until all there's no more data to process.
>
> Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
> ---
> tap.c | 28 +++++++++++++++++++---------
> 1 file changed, 19 insertions(+), 9 deletions(-)
>
> diff --git a/tap.c b/tap.c
> index 2fbcef04..d7d3fc19 100644
> --- a/tap.c
> +++ b/tap.c
> @@ -985,8 +985,10 @@ static void tap_sock_reset(struct ctx *c)
> * tap_passt_input() - Handler for new data on the socket to qemu
> * @c: Execution context
> * @now: Current timestamp
> + *
> + * Return: true if there may be additional data to read, false otherwise
> */
> -static void tap_passt_input(struct ctx *c, const struct timespec *now)
> +static bool tap_passt_input(struct ctx *c, const struct timespec *now)
> {
> static const char *partial_frame;
> static ssize_t partial_len = 0;
> @@ -1013,7 +1015,7 @@ static void tap_passt_input(struct ctx *c, const struct timespec *now)
> err_perror("Receive error on guest connection, reset");
> tap_sock_reset(c);
> }
> - return;
> + return false;
> }
>
> p = pkt_buf;
> @@ -1025,7 +1027,7 @@ static void tap_passt_input(struct ctx *c, const struct timespec *now)
> if (l2len < sizeof(struct ethhdr) || l2len > ETH_MAX_MTU) {
> err("Bad frame size from guest, resetting connection");
> tap_sock_reset(c);
> - return;
> + return false;
> }
>
> if (l2len + sizeof(uint32_t) > (size_t)n)
> @@ -1045,6 +1047,8 @@ static void tap_passt_input(struct ctx *c, const struct timespec *now)
> partial_frame = p;
>
> tap_handler(c, now);
> +
> + return true;
> }
>
> /**
> @@ -1062,15 +1066,18 @@ void tap_handler_passt(struct ctx *c, uint32_t events,
> }
>
> if (events & EPOLLIN)
> - tap_passt_input(c, now);
> + while (tap_passt_input(c, now))
> + ;
Nit (same below): use curly brackets for multi-line block. Or just use:
while (tap_passt_input(c, now));
> }
>
> /**
> * tap_passt_input() - Handler for new data on the socket to qemu
> * @c: Execution context
> * @now: Current timestamp
> + *
> + * Return: true if there may be additional data to read, false otherwise
> */
> -static void tap_pasta_input(struct ctx *c, const struct timespec *now)
> +static bool tap_pasta_input(struct ctx *c, const struct timespec *now)
> {
> ssize_t n, len;
>
> @@ -1101,6 +1108,8 @@ static void tap_pasta_input(struct ctx *c, const struct timespec *now)
> die("EOF on tap device, exiting");
>
> tap_handler(c, now);
> +
> + return len > 0;
> }
>
> /**
> @@ -1116,7 +1125,8 @@ void tap_handler_pasta(struct ctx *c, uint32_t events,
> die("Disconnect event on /dev/net/tun device, exiting");
>
> if (events & EPOLLIN)
> - tap_pasta_input(c, now);
> + while (tap_pasta_input(c, now))
> + ;
> }
>
> /**
> @@ -1250,7 +1260,7 @@ void tap_listen_handler(struct ctx *c, uint32_t events)
> trace("tap: failed to set SO_SNDBUF to %i", v);
>
> ref.fd = c->fd_tap;
> - ev.events = EPOLLIN | EPOLLRDHUP;
> + ev.events = EPOLLIN | EPOLLRDHUP | EPOLLET;
> ev.data.u64 = ref.u64;
> epoll_ctl(c->epollfd, EPOLL_CTL_ADD, c->fd_tap, &ev);
> }
> @@ -1306,7 +1316,7 @@ static void tap_sock_tun_init(struct ctx *c)
> pasta_ns_conf(c);
>
> ref.fd = c->fd_tap;
> - ev.events = EPOLLIN | EPOLLRDHUP;
> + ev.events = EPOLLIN | EPOLLRDHUP | EPOLLET;
> ev.data.u64 = ref.u64;
> epoll_ctl(c->epollfd, EPOLL_CTL_ADD, c->fd_tap, &ev);
> }
> @@ -1339,7 +1349,7 @@ void tap_sock_init(struct ctx *c)
> else
> ref.type = EPOLL_TYPE_TAP_PASTA;
>
> - ev.events = EPOLLIN | EPOLLRDHUP;
> + ev.events = EPOLLIN | EPOLLRDHUP | EPOLLET;
> ev.data.u64 = ref.u64;
> epoll_ctl(c->epollfd, EPOLL_CTL_ADD, c->fd_tap, &ev);
> return;
--
Stefano
next prev parent reply other threads:[~2024-09-03 19:25 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-09-03 12:02 [PATCH 0/6] RFC: Clean up tap-side event handling David Gibson
2024-09-03 12:02 ` [PATCH 1/6] tap: Split out handling of EPOLLIN events David Gibson
2024-09-03 19:25 ` Stefano Brivio
2024-09-04 1:17 ` David Gibson
2024-09-03 12:02 ` [PATCH 2/6] tap: Improve handling of EINTR in tap_passt_input() David Gibson
2024-09-03 19:25 ` Stefano Brivio
2024-09-04 1:30 ` David Gibson
2024-09-03 12:02 ` [PATCH 3/6] tap: Restructure in tap_pasta_input() David Gibson
2024-09-03 19:25 ` Stefano Brivio
2024-09-04 1:33 ` David Gibson
2024-09-03 12:02 ` [PATCH 4/6] tap: Don't risk truncating frames on full buffer " David Gibson
2024-09-03 19:25 ` Stefano Brivio
2024-09-04 1:33 ` David Gibson
2024-09-03 12:02 ` [PATCH 5/6] tap: Re-introduce EPOLLET for tap connections David Gibson
2024-09-03 19:25 ` Stefano Brivio [this message]
2024-09-04 1:36 ` David Gibson
2024-09-03 12:02 ` [PATCH 6/6] tap: Stub EPOLLOUT handling David Gibson
2024-09-03 19:25 ` [PATCH 0/6] RFC: Clean up tap-side event handling Stefano Brivio
2024-09-04 3:17 ` David Gibson
2024-09-04 17:19 ` Stefano Brivio
2024-09-05 0:35 ` David Gibson
2024-09-05 8:32 ` Stefano Brivio
2024-09-05 11:33 ` Stefano Brivio
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=20240903212550.05900f14@elisabeth \
--to=sbrivio@redhat.com \
--cc=david@gibson.dropbear.id.au \
--cc=passt-dev@passt.top \
/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.
Code repositories for project(s) associated with this public inbox
https://passt.top/passt
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 IMAP folder(s).