Eric Dumazet eb9344781a tcp: add a force_schedule argument to sk_stream_alloc_skb() ... In commit 8e4d980ac2 ("tcp: fix behavior for epoll edge trigger") we fixed a possible hang of TCP sockets under memory pressure, by allowing sk_stream_alloc_skb() to use sk_forced_mem_schedule() if no packet is in socket write queue. It turns out there are other cases where we want to force memory schedule : tcp_fragment() & tso_fragment() need to split a big TSO packet into two smaller ones. If we block here because of TCP memory pressure, we can effectively block TCP socket from sending new data. If no further ACK is coming, this hang would be definitive, and socket has no chance to effectively reduce its memory usage. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>		2015-05-21 16:56:40 -04:00
..
6lowpan
9p
802
8021q
appletalk
atm
ax25
batman-adv
bluetooth
bridge	bridge_netfilter: No ICMP packet on IPv4 fragmentation error	2015-05-19 00:15:39 -04:00
caif
can
ceph
core	neigh: Better handling of transition to NUD_PROBE state	2015-05-21 16:52:17 -04:00
dcb
dccp
decnet
dns_resolver
dsa
ethernet
hsr
ieee802154
ipv4	tcp: add a force_schedule argument to sk_stream_alloc_skb()	2015-05-21 16:56:40 -04:00
ipv6
ipx
irda
iucv
key
l2tp
lapb
llc
mac80211	This just has a few fixes:	2015-05-19 16:43:17 -04:00
mac802154
mpls
netfilter	Merge git://git.kernel.org/pub/scm/linux/kernel/git/pablo/nf-next	2015-05-18 14:47:36 -04:00
netlabel
netlink	netlink: Use random autobind rover	2015-05-17 23:43:31 -04:00
netrom
nfc
openvswitch
packet
phonet
rds
rfkill
rose
rxrpc
sched
sctp
sunrpc
switchdev	switchdev: add support for fdb add/del/dump via switchdev_port_obj ops.	2015-05-17 22:49:09 -04:00
tipc
unix
vmw_vsock
wimax
wireless
x25
xfrm
compat.c
Kconfig
Makefile
socket.c
sysctl_net.c