public inbox for passt-dev@passt.top
 help / color / mirror / code / Atom feed
From: David Gibson <david@gibson.dropbear.id.au>
To: passt-dev@passt.top, Stefano Brivio <sbrivio@redhat.com>
Cc: David Gibson <david@gibson.dropbear.id.au>
Subject: [PATCH v2 03/11] tcp_splice: Don't handle EPOLL_CTL_DEL as part of tcp_splice_epoll_ctl()
Date: Tue,  7 Nov 2023 13:42:42 +1100	[thread overview]
Message-ID: <20231107024250.2290959-4-david@gibson.dropbear.id.au> (raw)
In-Reply-To: <20231107024250.2290959-1-david@gibson.dropbear.id.au>

tcp_splice_epoll_ctl() removes both sockets from the epoll set if called
when conn->flags & CLOSING.  This will always happen immediately after
setting that flag, since conn_flag_do() makes the call itself.  That's also
the _only_ time it can happen: we perform the EPOLL_CTL_DEL without
clearing the conn->in_epoll flag, meaning that any further calls to
tcp_splice_epoll_ctl() would attempt EPOLL_CTL_MOD, which would necessarily
fail since the fds are no longer in the epoll.

The EPOLL_CTL_DEL path in tcp_splice_epoll_ctl() has essentially zero
overlap with anything else the function does, so just move them to be
open coded in conn_flag_do().

This does require kernel 2.6.9 or later, in order to pass NULL as the
event structure for epoll_ctl().  However, we already require at least
3.13 to allow unprivileged user namespaces.

Given that, simply directly perform the EPOLL_CTL_DEL operations from
conn_flag_do() rather than unnecessarily multiplexini

Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
---
 tcp_splice.c | 12 ++++--------
 1 file changed, 4 insertions(+), 8 deletions(-)

diff --git a/tcp_splice.c b/tcp_splice.c
index 6c4c664..570a8c6 100644
--- a/tcp_splice.c
+++ b/tcp_splice.c
@@ -152,8 +152,10 @@ static void conn_flag_do(const struct ctx *c, struct tcp_splice_conn *conn,
 		}
 	}
 
-	if (flag == CLOSING)
-		tcp_splice_epoll_ctl(c, conn);
+	if (flag == CLOSING) {
+		epoll_ctl(c->epollfd, EPOLL_CTL_DEL, conn->a, NULL);
+		epoll_ctl(c->epollfd, EPOLL_CTL_DEL, conn->b, NULL);
+	}
 }
 
 #define conn_flag(c, conn, flag)					\
@@ -182,12 +184,6 @@ static int tcp_splice_epoll_ctl(const struct ctx *c,
 	struct epoll_event ev_b = { .data.u64 = ref_b.u64 };
 	uint32_t events_a, events_b;
 
-	if (conn->flags & CLOSING) {
-		epoll_ctl(c->epollfd, EPOLL_CTL_DEL, conn->a, &ev_a);
-		epoll_ctl(c->epollfd, EPOLL_CTL_DEL, conn->b, &ev_b);
-		return 0;
-	}
-
 	tcp_splice_conn_epoll_events(conn->events, &events_a, &events_b);
 	ev_a.events = events_a;
 	ev_b.events = events_b;
-- 
@@ -152,8 +152,10 @@ static void conn_flag_do(const struct ctx *c, struct tcp_splice_conn *conn,
 		}
 	}
 
-	if (flag == CLOSING)
-		tcp_splice_epoll_ctl(c, conn);
+	if (flag == CLOSING) {
+		epoll_ctl(c->epollfd, EPOLL_CTL_DEL, conn->a, NULL);
+		epoll_ctl(c->epollfd, EPOLL_CTL_DEL, conn->b, NULL);
+	}
 }
 
 #define conn_flag(c, conn, flag)					\
@@ -182,12 +184,6 @@ static int tcp_splice_epoll_ctl(const struct ctx *c,
 	struct epoll_event ev_b = { .data.u64 = ref_b.u64 };
 	uint32_t events_a, events_b;
 
-	if (conn->flags & CLOSING) {
-		epoll_ctl(c->epollfd, EPOLL_CTL_DEL, conn->a, &ev_a);
-		epoll_ctl(c->epollfd, EPOLL_CTL_DEL, conn->b, &ev_b);
-		return 0;
-	}
-
 	tcp_splice_conn_epoll_events(conn->events, &events_a, &events_b);
 	ev_a.events = events_a;
 	ev_b.events = events_b;
-- 
2.41.0


  parent reply	other threads:[~2023-11-07  2:43 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-11-07  2:42 [PATCH v2 00/11] tcp_splice: Better exploit symmetry between sides of connection David Gibson
2023-11-07  2:42 ` [PATCH v2 01/11] tcp_splice: Remove redundant tcp_splice_epoll_ctl() David Gibson
2023-11-07  2:42 ` [PATCH v2 02/11] tcp_splice: Correct error handling in tcp_splice_epoll_ctl() David Gibson
2023-11-07  2:42 ` David Gibson [this message]
2023-11-07  2:42 ` [PATCH v2 04/11] tcp_splice: Remove unnecessary forward declaration David Gibson
2023-11-07  2:42 ` [PATCH v2 05/11] tcp_splice: Avoid awkward temporaries in tcp_splice_epoll_ctl() David Gibson
2023-11-07  2:42 ` [PATCH v2 06/11] tcp_splice: Don't pool pipes in pairs David Gibson
2023-11-07  2:42 ` [PATCH v2 07/11] tcp_splice: Rename sides of connection from a/b to 0/1 David Gibson
2023-11-07  2:42 ` [PATCH v2 08/11] tcp_splice: Exploit side symmetry in tcp_splice_timer() David Gibson
2023-11-07  2:42 ` [PATCH v2 09/11] tcp_splice: Exploit side symmetry in tcp_splice_connect_finish() David Gibson
2023-11-07  2:42 ` [PATCH v2 10/11] tcp_splice: Exploit side symmetry in tcp_splice_destroy() David Gibson
2023-11-07  2:42 ` [PATCH v2 11/11] tcp_splice: Simplify selection of socket and pipe sides in socket handler David Gibson
2023-11-07 12:45 ` [PATCH v2 00/11] tcp_splice: Better exploit symmetry between sides of connection 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=20231107024250.2290959-4-david@gibson.dropbear.id.au \
    --to=david@gibson.dropbear.id.au \
    --cc=passt-dev@passt.top \
    --cc=sbrivio@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.
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).