kernel_optimize_test/net
Eric Dumazet 42324c6270 net: splice() from tcp to pipe should take into account O_NONBLOCK
tcp_splice_read() doesnt take into account socket's O_NONBLOCK flag

Before this patch :

splice(socket,0,pipe,0,128*1024,SPLICE_F_MOVE);
causes a random endless block (if pipe is full) and
splice(socket,0,pipe,0,128*1024,SPLICE_F_MOVE | SPLICE_F_NONBLOCK);
will return 0 immediately if the TCP buffer is empty.

User application has no way to instruct splice() that socket should be in blocking mode
but pipe in nonblock more.

Many projects cannot use splice(tcp -> pipe) because of this flaw.

http://git.samba.org/?p=samba.git;a=history;f=source3/lib/recvfile.c;h=ea0159642137390a0f7e57a123684e6e63e47581;hb=HEAD
http://lkml.indiana.edu/hypermail/linux/kernel/0807.2/0687.html

Linus introduced  SPLICE_F_NONBLOCK in commit 29e350944f
(splice: add SPLICE_F_NONBLOCK flag )

  It doesn't make the splice itself necessarily nonblocking (because the
  actual file descriptors that are spliced from/to may block unless they
  have the O_NONBLOCK flag set), but it makes the splice pipe operations
  nonblocking.

Linus intention was clear : let SPLICE_F_NONBLOCK control the splice pipe mode only

This patch instruct tcp_splice_read() to use the underlying file O_NONBLOCK
flag, as other socket operations do.

Users will then call :

splice(socket,0,pipe,0,128*1024,SPLICE_F_MOVE | SPLICE_F_NONBLOCK );

to block on data coming from socket (if file is in blocking mode),
and not block on pipe output (to avoid deadlock)

First version of this patch was submitted by Octavian Purdila

Reported-by: Volker Lendecke <vl@samba.org>
Reported-by: Jason Gunthorpe <jgunthorpe@obsidianresearch.com>
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: Octavian Purdila <opurdila@ixiacom.com>
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Acked-by: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-02 09:46:05 -07:00
..
9p virtio: add virtio IDs file 2009-09-23 22:26:32 +09:30
802
8021q net: fix vlan_get_size to include vlan_flags size 2009-09-26 20:16:07 -07:00
appletalk Have atalk_route_packet() return NET_RX_SUCCESS not NET_XMIT_SUCCESS 2009-09-14 17:02:47 -07:00
atm net: Make setsockopt() optlen be unsigned. 2009-09-30 16:12:20 -07:00
ax25 ax25: Fix possible oops in ax25_make_new 2009-09-30 16:44:12 -07:00
bluetooth net: Make setsockopt() optlen be unsigned. 2009-09-30 16:12:20 -07:00
bridge bridge: Fix double-free in br_add_if. 2009-09-28 12:54:25 -07:00
can net: Make setsockopt() optlen be unsigned. 2009-09-30 16:12:20 -07:00
core pktgen: Fix delay handling 2009-10-01 09:29:45 -07:00
dcb net: fix double skb free in dcbnl 2009-09-26 20:16:15 -07:00
dccp net: Make setsockopt() optlen be unsigned. 2009-09-30 16:12:20 -07:00
decnet net: Make setsockopt() optlen be unsigned. 2009-09-30 16:12:20 -07:00
dsa netdev: convert pseudo-devices to netdev_tx_t 2009-09-01 01:13:07 -07:00
econet
ethernet
ieee802154 net: Make setsockopt() optlen be unsigned. 2009-09-30 16:12:20 -07:00
ipv4 net: splice() from tcp to pipe should take into account O_NONBLOCK 2009-10-02 09:46:05 -07:00
ipv6 sit: fix off-by-one in ipip6_tunnel_get_prl 2009-09-30 16:39:27 -07:00
ipx net: Make setsockopt() optlen be unsigned. 2009-09-30 16:12:20 -07:00
irda net: Make setsockopt() optlen be unsigned. 2009-09-30 16:12:20 -07:00
iucv net: Make setsockopt() optlen be unsigned. 2009-09-30 16:12:20 -07:00
key net: file_operations should be const 2009-09-02 01:03:53 -07:00
lapb
llc net: Make setsockopt() optlen be unsigned. 2009-09-30 16:12:20 -07:00
mac80211 mac80211: Fix [re]association power saving issue on AP side 2009-09-29 17:25:15 -04:00
netfilter net: Make setsockopt() optlen be unsigned. 2009-09-30 16:12:20 -07:00
netlabel
netlink net: Make setsockopt() optlen be unsigned. 2009-09-30 16:12:20 -07:00
netrom net: Make setsockopt() optlen be unsigned. 2009-09-30 16:12:20 -07:00
packet net: Make setsockopt() optlen be unsigned. 2009-09-30 16:12:20 -07:00
phonet Phonet: fix mutex imbalance 2009-09-30 16:41:34 -07:00
rds net: Make setsockopt() optlen be unsigned. 2009-09-30 16:12:20 -07:00
rfkill
rose net: Make setsockopt() optlen be unsigned. 2009-09-30 16:12:20 -07:00
rxrpc net: Make setsockopt() optlen be unsigned. 2009-09-30 16:12:20 -07:00
sched trivial: fix typo "to to" in multiple files 2009-09-21 15:14:55 +02:00
sctp net: Make setsockopt() optlen be unsigned. 2009-09-30 16:12:20 -07:00
sunrpc net: fix htmldocs sunrpc, clnt.c 2009-09-24 15:39:14 -07:00
tipc net: Make setsockopt() optlen be unsigned. 2009-09-30 16:12:20 -07:00
unix net: unix: fix sending fds in multiple buffers 2009-09-11 11:31:45 -07:00
wanrouter
wimax
wireless wext: add back wireless/ dir in sysfs for cfg80211 interfaces 2009-09-28 16:55:07 -04:00
x25 net: Make setsockopt() optlen be unsigned. 2009-09-30 16:12:20 -07:00
xfrm net: file_operations should be const 2009-09-02 01:03:53 -07:00
compat.c net: Make setsockopt() optlen be unsigned. 2009-09-30 16:12:20 -07:00
Kconfig
Makefile
nonet.c
socket.c net: Make setsockopt() optlen be unsigned. 2009-09-30 16:12:20 -07:00
sysctl_net.c
TUNABLE