From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Message-Id: <20130217225008.842367297@decadent.org.uk> Date: Sun, 17 Feb 2013 22:51:03 +0000 From: Ben Hutchings To: linux-kernel@vger.kernel.org, stable@vger.kernel.org Cc: akpm@linux-foundation.org, Eric Dumazet , Willy Tarreau , "David S. Miller" Subject: [ 62/66] tcp: fix MSG_SENDPAGE_NOTLAST logic In-Reply-To: <20130217225001.621306883@decadent.org.uk> Sender: linux-kernel-owner@vger.kernel.org List-ID: 3.2-stable review patch. If anyone has any objections, please let me know. ------------------ From: Eric Dumazet [ Upstream commit ae62ca7b03217be5e74759dc6d7698c95df498b3 ] commit 35f9c09fe9c72e (tcp: tcp_sendpages() should call tcp_push() once) added an internal flag : MSG_SENDPAGE_NOTLAST meant to be set on all frags but the last one for a splice() call. The condition used to set the flag in pipe_to_sendpage() relied on splice() user passing the exact number of bytes present in the pipe, or a smaller one. But some programs pass an arbitrary high value, and the test fails. The effect of this bug is a lack of tcp_push() at the end of a splice(pipe -> socket) call, and possibly very slow or erratic TCP sessions. We should both test sd->total_len and fact that another fragment is in the pipe (pipe->nrbufs > 1) Many thanks to Willy for providing very clear bug report, bisection and test programs. Reported-by: Willy Tarreau Bisected-by: Willy Tarreau Tested-by: Willy Tarreau Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller Signed-off-by: Ben Hutchings --- fs/splice.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/fs/splice.c b/fs/splice.c index 014fcb4..58ab918 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -697,8 +697,10 @@ static int pipe_to_sendpage(struct pipe_inode_info *pipe, return -EINVAL; more = (sd->flags & SPLICE_F_MORE) ? MSG_MORE : 0; - if (sd->len < sd->total_len) + + if (sd->len < sd->total_len && pipe->nrbufs > 1) more |= MSG_SENDPAGE_NOTLAST; + return file->f_op->sendpage(file, buf->page, buf->offset, sd->len, &pos, more); }