lwn.git/net/ipv4/tcp_output.c, branch v3.7.4

tcp: bug fix Fast Open client retransmission

2012-12-07T19:39:28+00:00

If SYN-ACK partially acks SYN-data, the client retransmits the remaining data by tcp_retransmit_skb(). This increments lost recovery state variables like tp->retrans_out in Open state. If loss recovery happens before the retransmission is acked, it triggers the WARN_ON check in tcp_fastretrans_alert(). For example: the client sends SYN-data, gets SYN-ACK acking only ISN, retransmits data, sends another 4 data packets and get 3 dupacks. Since the retransmission is not caused by network drop it should not update the recovery state variables. Further the server may return a smaller MSS than the cached MSS used for SYN-data, so the retranmission needs a loop. Otherwise some data will not be retransmitted until timeout or other loss recovery events. Signed-off-by: Yuchung Cheng Acked-by: Neal Cardwell Signed-off-by: David S. Miller

tcp: fix retransmission in repair mode

2012-11-15T22:44:58+00:00

Currently if a socket was repaired with a few packet in a write queue, a kernel bug may be triggered: kernel BUG at net/ipv4/tcp_output.c:2330! RIP: 0010:[] tcp_retransmit_skb+0x5ff/0x610 According to the initial realization v3.4-rc2-963-gc0e88ff, all skb-s should look like already posted. This patch fixes code according with this sentence. Here are three points, which were not done in the initial patch: 1. A tcp send head should not be changed 2. Initialize TSO state of a skb 3. Reset the retransmission time This patch moves logic from tcp_sendmsg to tcp_write_xmit. A packet passes the ussual way, but isn't sent to network. This patch solves all described problems and handles tcp_sendpages. Cc: Pavel Emelyanov Cc: "David S. Miller" Cc: Alexey Kuznetsov Cc: James Morris Cc: Hideaki YOSHIFUJI Cc: Patrick McHardy Signed-off-by: Andrey Vagin Acked-by: Pavel Emelyanov Signed-off-by: David S. Miller

tcp: use PRR to reduce cwin in CWR state

2012-09-03T18:34:02+00:00

Use proportional rate reduction (PRR) algorithm to reduce cwnd in CWR state, in addition to Recovery state. Retire the current rate-halving in CWR. When losses are detected via ACKs in CWR state, the sender enters Recovery state but the cwnd reduction continues and does not restart. Rename and refactor cwnd reduction functions since both CWR and Recovery use the same algorithm: tcp_init_cwnd_reduction() is new and initiates reduction state variables. tcp_cwnd_reduction() is previously tcp_update_cwnd_in_recovery(). tcp_ends_cwnd_reduction() is previously tcp_complete_cwr(). The rate halving functions and logic such as tcp_cwnd_down(), tcp_min_cwnd(), and the cwnd moderation inside tcp_enter_cwr() are removed. The unused parameter, flag, in tcp_cwnd_reduction() is also removed. Signed-off-by: Yuchung Cheng Acked-by: Neal Cardwell Signed-off-by: David S. Miller

tcp: TCP Fast Open Server - support TFO listeners

2012-09-01T00:02:19+00:00

This patch builds on top of the previous patch to add the support for TFO listeners. This includes - 1. allocating, properly initializing, and managing the per listener fastopen_queue structure when TFO is enabled 2. changes to the inet_csk_accept code to support TFO. E.g., the request_sock can no longer be freed upon accept(), not until 3WHS finishes 3. allowing a TCP_SYN_RECV socket to properly poll() and sendmsg() if it's a TFO socket 4. properly closing a TFO listener, and a TFO socket before 3WHS finishes 5. supporting TCP_FASTOPEN socket option 6. modifying tcp_check_req() to use to check a TFO socket as well as request_sock 7. supporting TCP's TFO cookie option 8. adding a new SYN-ACK retransmit handler to use the timer directly off the TFO socket rather than the listener socket. Note that TFO server side will not retransmit anything other than SYN-ACK until the 3WHS is completed. The patch also contains an important function "reqsk_fastopen_remove()" to manage the somewhat complex relation between a listener, its request_sock, and the corresponding child socket. See the comment above the function for the detail. Signed-off-by: H.K. Jerry Chu Cc: Yuchung Cheng Cc: Neal Cardwell Cc: Eric Dumazet Cc: Tom Herbert Signed-off-by: David S. Miller

tcp: fix possible socket refcount problem

2012-08-21T21:42:23+00:00

Commit 6f458dfb40 (tcp: improve latencies of timer triggered events) added bug leading to following trace : [ 2866.131281] IPv4: Attempt to release TCP socket in state 1 ffff880019ec0000 [ 2866.131726] [ 2866.132188] ========================= [ 2866.132281] [ BUG: held lock freed! ] [ 2866.132281] 3.6.0-rc1+ #622 Not tainted [ 2866.132281] ------------------------- [ 2866.132281] kworker/0:1/652 is freeing memory ffff880019ec0000-ffff880019ec0a1f, with a lock still held there! [ 2866.132281] (sk_lock-AF_INET-RPC){+.+...}, at: [] tcp_sendmsg+0x29/0xcc6 [ 2866.132281] 4 locks held by kworker/0:1/652: [ 2866.132281] #0: (rpciod){.+.+.+}, at: [] process_one_work+0x1de/0x47f [ 2866.132281] #1: ((&task->u.tk_work)){+.+.+.}, at: [] process_one_work+0x1de/0x47f [ 2866.132281] #2: (sk_lock-AF_INET-RPC){+.+...}, at: [] tcp_sendmsg+0x29/0xcc6 [ 2866.132281] #3: (&icsk->icsk_retransmit_timer){+.-...}, at: [] run_timer_softirq+0x1ad/0x35f [ 2866.132281] [ 2866.132281] stack backtrace: [ 2866.132281] Pid: 652, comm: kworker/0:1 Not tainted 3.6.0-rc1+ #622 [ 2866.132281] Call Trace: [ 2866.132281] [] debug_check_no_locks_freed+0x112/0x159 [ 2866.132281] [] ? __sk_free+0xfd/0x114 [ 2866.132281] [] kmem_cache_free+0x6b/0x13a [ 2866.132281] [] __sk_free+0xfd/0x114 [ 2866.132281] [] sk_free+0x1c/0x1e [ 2866.132281] [] tcp_write_timer+0x51/0x56 [ 2866.132281] [] run_timer_softirq+0x218/0x35f [ 2866.132281] [] ? run_timer_softirq+0x1ad/0x35f [ 2866.132281] [] ? rb_commit+0x58/0x85 [ 2866.132281] [] ? tcp_write_timer_handler+0x148/0x148 [ 2866.132281] [] __do_softirq+0xcb/0x1f9 [ 2866.132281] [] ? _raw_spin_unlock+0x29/0x2e [ 2866.132281] [] call_softirq+0x1c/0x30 [ 2866.132281] [] do_softirq+0x4a/0xa6 [ 2866.132281] [] irq_exit+0x51/0xad [ 2866.132281] [] do_IRQ+0x9d/0xb4 [ 2866.132281] [] common_interrupt+0x6f/0x6f [ 2866.132281] [] ? sched_clock_cpu+0x58/0xd1 [ 2866.132281] [] ? _raw_spin_unlock_irqrestore+0x4c/0x56 [ 2866.132281] [] mod_timer+0x178/0x1a9 [ 2866.132281] [] sk_reset_timer+0x19/0x26 [ 2866.132281] [] tcp_rearm_rto+0x99/0xa4 [ 2866.132281] [] tcp_event_new_data_sent+0x6e/0x70 [ 2866.132281] [] tcp_write_xmit+0x7de/0x8e4 [ 2866.132281] [] ? __alloc_skb+0xa0/0x1a1 [ 2866.132281] [] __tcp_push_pending_frames+0x2e/0x8a [ 2866.132281] [] tcp_sendmsg+0xb32/0xcc6 [ 2866.132281] [] inet_sendmsg+0xaa/0xd5 [ 2866.132281] [] ? inet_autobind+0x5f/0x5f [ 2866.132281] [] ? trace_clock_local+0x9/0xb [ 2866.132281] [] sock_sendmsg+0xa3/0xc4 [ 2866.132281] [] ? rb_reserve_next_event+0x26f/0x2d5 [ 2866.132281] [] ? native_sched_clock+0x29/0x6f [ 2866.132281] [] ? sched_clock+0x9/0xd [ 2866.132281] [] ? trace_clock_local+0x9/0xb [ 2866.132281] [] kernel_sendmsg+0x37/0x43 [ 2866.132281] [] xs_send_kvec+0x77/0x80 [ 2866.132281] [] xs_sendpages+0x6f/0x1a0 [ 2866.132281] [] ? try_to_del_timer_sync+0x55/0x61 [ 2866.132281] [] xs_tcp_send_request+0x55/0xf1 [ 2866.132281] [] xprt_transmit+0x89/0x1db [ 2866.132281] [] ? call_connect+0x3c/0x3c [ 2866.132281] [] call_transmit+0x1c5/0x20e [ 2866.132281] [] __rpc_execute+0x6f/0x225 [ 2866.132281] [] ? call_connect+0x3c/0x3c [ 2866.132281] [] rpc_async_schedule+0x28/0x34 [ 2866.132281] [] process_one_work+0x24d/0x47f [ 2866.132281] [] ? process_one_work+0x1de/0x47f [ 2866.132281] [] ? __rpc_execute+0x225/0x225 [ 2866.132281] [] worker_thread+0x236/0x317 [ 2866.132281] [] ? process_scheduled_works+0x2f/0x2f [ 2866.132281] [] kthread+0x9a/0xa2 [ 2866.132281] [] kernel_thread_helper+0x4/0x10 [ 2866.132281] [] ? retint_restore_args+0x13/0x13 [ 2866.132281] [] ? __init_kthread_worker+0x5a/0x5a [ 2866.132281] [] ? gs_change+0x13/0x13 [ 2866.308506] IPv4: Attempt to release TCP socket in state 1 ffff880019ec0000 [ 2866.309689] ============================================================================= [ 2866.310254] BUG TCP (Not tainted): Object already free [ 2866.310254] ----------------------------------------------------------------------------- [ 2866.310254] The bug comes from the fact that timer set in sk_reset_timer() can run before we actually do the sock_hold(). socket refcount reaches zero and we free the socket too soon. timer handler is not allowed to reduce socket refcnt if socket is owned by the user, or we need to change sk_reset_timer() implementation. We should take a reference on the socket in case TCP_DELACK_TIMER_DEFERRED or TCP_DELACK_TIMER_DEFERRED bit are set in tsq_flags Also fix a typo in tcp_delack_timer(), where TCP_WRITE_TIMER_DEFERRED was used instead of TCP_DELACK_TIMER_DEFERRED. For consistency, use same socket refcount change for TCP_MTU_REDUCED_DEFERRED, even if not fired from a timer. Reported-by: Fengguang Wu Tested-by: Fengguang Wu Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller

tcp_output: fix sparse warning for tcp_wfree

2012-08-06T20:29:56+00:00

Fix sparse warning: * symbol 'tcp_wfree' was not declared. Should it be static? Signed-off-by: Silviu-Mihai Popescu Signed-off-by: David S. Miller

tcp: Apply device TSO segment limit earlier

2012-08-02T07:19:17+00:00

Cache the device gso_max_segs in sock::sk_gso_max_segs and use it to limit the size of TSO skbs. This avoids the need to fall back to software GSO for local TCP senders. Signed-off-by: Ben Hutchings Signed-off-by: David S. Miller

net: introduce sk_gfp_atomic() to allow addition of GFP flags depending on the individual socket

2012-08-01T01:42:46+00:00

Introduce sk_gfp_atomic(), this function allows to inject sock specific flags to each sock related allocation. It is only used on allocation paths that may be required for writing pages back to network storage. [davem@davemloft.net: Use sk_gfp_atomic only when necessary] Signed-off-by: Peter Zijlstra Signed-off-by: Mel Gorman Acked-by: David S. Miller Cc: Neil Brown Cc: Mike Christie Cc: Eric B Munson Cc: Eric Dumazet Cc: Sebastian Andrzej Siewior Cc: Mel Gorman Cc: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

tcp: dont drop MTU reduction indications

2012-07-23T07:58:46+00:00

ICMP messages generated in output path if frame length is bigger than mtu are actually lost because socket is owned by user (doing the xmit) One example is the ipgre_tunnel_xmit() calling icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED, htonl(mtu)); We had a similar case fixed in commit a34a101e1e6 (ipv6: disable GSO on sockets hitting dst_allfrag). Problem of such fix is that it relied on retransmit timers, so short tcp sessions paid a too big latency increase price. This patch uses the tcp_release_cb() infrastructure so that MTU reduction messages (ICMP messages) are not lost, and no extra delay is added in TCP transmits. Reported-by: Maciej Żenczykowski Diagnosed-by: Neal Cardwell Signed-off-by: Eric Dumazet Cc: Nandita Dukkipati Cc: Tom Herbert Cc: Tore Anderson Signed-off-by: David S. Miller

tcp: improve latencies of timer triggered events

2012-07-20T17:59:41+00:00

Modern TCP stack highly depends on tcp_write_timer() having a small latency, but current implementation doesn't exactly meet the expectations. When a timer fires but finds the socket is owned by the user, it rearms itself for an additional delay hoping next run will be more successful. tcp_write_timer() for example uses a 50ms delay for next try, and it defeats many attempts to get predictable TCP behavior in term of latencies. Use the recently introduced tcp_release_cb(), so that the user owning the socket will call various handlers right before socket release. This will permit us to post a followup patch to address the tcp_tso_should_defer() syndrome (some deferred packets have to wait RTO timer to be transmitted, while cwnd should allow us to send them sooner) Signed-off-by: Eric Dumazet Cc: Tom Herbert Cc: Yuchung Cheng Cc: Neal Cardwell Cc: Nandita Dukkipati Cc: H.K. Jerry Chu Cc: John Heffner Signed-off-by: David S. Miller