Since recent changes on TCP splicing (starting with commits 2f533844
"tcp: allow splice() to build full TSO packets" and 35f9c09f "tcp:
tcp_sendpages() should call tcp_push() once"), I started seeing
massive stalls when forwarding traffic between two sockets using
splice() when pipe buffers were larger than socket buffers.

Latest changes (net: netdev_alloc_skb() use build_skb()) made the
problem even more apparent.

The reason seems to be that if do_tcp_sendpages() fails on out of memory
condition without being able to send at least one byte, tcp_push() is not
called and the buffers cannot be flushed.

After applying the attached patch, I cannot reproduce the stalls at all
and the data rate it perfectly stable and steady under any condition
which previously caused the problem to be permanent.

The issue seems to have been there since before the kernel migrated to
git, which makes me think that the stalls I occasionally experienced
with tux during stress-tests years ago were probably related to the
same issue.

This issue was first encountered on 3.0.31 and 3.2.17, so please backport
to -stable.
Signed-off-by: NWilly Tarreau <w@1wt.eu>
Acked-by: NEric Dumazet <edumazet@google.com>
Cc: <stable@vger.kernel.org>

tcp_adv_win_scale default value is 2, meaning we expect a good citizen
skb to have skb->len / skb->truesize ratio of 75% (3/4)

In 2.6 kernels we (mis)accounted for typical MSS=1460 frame :
1536 + 64 + 256 = 1856 'estimated truesize', and 1856 * 3/4 = 1392.
So these skbs were considered as not bloated.

With recent truesize fixes, a typical MSS=1460 frame truesize is now the
more precise :
2048 + 256 = 2304. But 2304 * 3/4 = 1728.
So these skb are not good citizen anymore, because 1460 < 1728

(GRO can escape this problem because it build skbs with a too low
truesize.)

This also means tcp advertises a too optimistic window for a given
allocated rcvspace : When receiving frames, sk_rmem_alloc can hit
sk_rcvbuf limit and we call tcp_prune_queue()/tcp_collapse() too often,
especially when application is slow to drain its receive queue or in
case of losses (netperf is fast, scp is slow). This is a major latency
source.

We should adjust the len/truesize ratio to 50% instead of 75%

This patch :

1) changes tcp_adv_win_scale default to 1 instead of 2

2) increase tcp_rmem[2] limit from 4MB to 6MB to take into account
better truesize tracking and to allow autotuning tcp receive window to
reach same value than before. Note that same amount of kernel memory is
consumed compared to 2.6 kernels.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Cc: Neal Cardwell <ncardwell@google.com>
Cc: Tom Herbert <therbert@google.com>
Cc: Yuchung Cheng <ycheng@google.com>
Acked-by: NNeal Cardwell <ncardwell@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Marc Merlin reported many order-1 allocations failures in TX path on its
wireless setup, that dont make any sense with MTU=1500 network, and non
SG capable hardware.

After investigation, it turns out TCP uses sk_stream_alloc_skb() and
used as a convention skb_tailroom(skb) to know how many bytes of data
payload could be put in this skb (for non SG capable devices)

Note : these skb used kmalloc-4096 (MTU=1500 + MAX_HEADER +
sizeof(struct skb_shared_info) being above 2048)

Later, mac80211 layer need to add some bytes at the tail of skb
(IEEE80211_ENCRYPT_TAILROOM = 18 bytes) and since no more tailroom is
available has to call pskb_expand_head() and request order-1
allocations.

This patch changes sk_stream_alloc_skb() so that only
sk->sk_prot->max_header bytes of headroom are reserved, and use a new
skb field, avail_size to hold the data payload limit.

This way, order-0 allocations done by TCP stack can leave more than 2 KB
of tailroom and no more allocation is performed in mac80211 layer (or
any layer needing some tailroom)

avail_size is unioned with mark/dropcount, since mark will be set later
in IP stack for output packets. Therefore, skb size is unchanged.
Reported-by: NMarc MERLIN <marc@merlins.org>
Tested-by: NMarc MERLIN <marc@merlins.org>
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Commit c43b874d (tcp: properly initialize tcp memory limits) tried
to fix a regression added in commits 4acb4190 & 3dc43e3e,
but still get it wrong.

Result is machines with low amount of memory have too small tcp_rmem[2]
value and slow tcp receives : Per socket limit being 1/1024 of memory
instead of 1/128 in old kernels, so rcv window is capped to small
values.

Fix this to match comment and previous behavior.
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Cc: Jason Wang <jasowang@redhat.com>
Cc: Glauber Costa <glommer@parallels.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

commit 2f533844 (tcp: allow splice() to build full TSO packets) added
a regression for splice() calls using SPLICE_F_MORE.

We need to call tcp_flush() at the end of the last page processed in
tcp_sendpages(), or else transmits can be deferred and future sends
stall.

Add a new internal flag, MSG_SENDPAGE_NOTLAST, acting like MSG_MORE, but
with different semantic.

For all sendpage() providers, its a transparent change. Only
sock_sendpage() and tcp_sendpages() can differentiate the two different
flags provided by pipe_to_sendpage()
Reported-by: NTom Herbert <therbert@google.com>
Cc: Nandita Dukkipati <nanditad@google.com>
Cc: Neal Cardwell <ncardwell@google.com>
Cc: Tom Herbert <therbert@google.com>
Cc: Yuchung Cheng <ycheng@google.com>
Cc: H.K. Jerry Chu <hkchu@google.com>
Cc: Maciej Żenczykowski <maze@google.com>
Cc: Mahesh Bandewar <maheshb@google.com>
Cc: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: NEric Dumazet <eric.dumazet@gmail&gt;com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This is the fallout from adding memcpy alignment workaround for certain
IOATDMA hardware. NetDMA will only use DMA engine that can handle byte align
ops.
Acked-by: NDavid S. Miller <davem@davemloft.net>
Signed-off-by: NDave Jiang <dave.jiang@intel.com>
Signed-off-by: NDan Williams <dan.j.williams@intel.com>

vmsplice()/splice(pipe, socket) call do_tcp_sendpages() one page at a
time, adding at most 4096 bytes to an skb. (assuming PAGE_SIZE=4096)

The call to tcp_push() at the end of do_tcp_sendpages() forces an
immediate xmit when pipe is not already filled, and tso_fragment() try
to split these skb to MSS multiples.

4096 bytes are usually split in a skb with 2 MSS, and a remaining
sub-mss skb (assuming MTU=1500)

This makes slow start suboptimal because many small frames are sent to
qdisc/driver layers instead of big ones (constrained by cwnd and packets
in flight of course)

In fact, applications using sendmsg() (adding an additional memory copy)
instead of vmsplice()/splice()/sendfile() are a bit faster because of
this anomaly, especially if serving small files in environments with
large initial [c]wnd.

Call tcp_push() only if MSG_MORE is not set in the flags parameter.

This bit is automatically provided by splice() internals but for the
last page, or on all pages if user specified SPLICE_F_MORE splice()
flag.

In some workloads, this can reduce number of sent logical packets by an
order of magnitude, making zero-copy TCP actually faster than
one-copy :)
Reported-by: NTom Herbert <therbert@google.com>
Cc: Nandita Dukkipati <nanditad@google.com>
Cc: Neal Cardwell <ncardwell@google.com>
Cc: Tom Herbert <therbert@google.com>
Cc: Yuchung Cheng <ycheng@google.com>
Cc: H.K. Jerry Chu <hkchu@google.com>
Cc: Maciej Żenczykowski <maze@google.com>
Cc: Mahesh Bandewar <maheshb@google.com>
Cc: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: NEric Dumazet <eric.dumazet@gmail&gt;com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add #define pr_fmt(fmt) as appropriate.

Add "IPv4: ", "TCP: ", and "IPsec: " to appropriate files.
Standardize on "UDPLite: " for appropriate uses.
Some prefixes were previously "UDPLITE: " and "UDP-Lite: ".

Add KBUILD_MODNAME ": " to icmp and gre.
Remove embedded prefixes as appropriate.

Add missing "\n" to pr_info in gre.c.
Signed-off-by: NJoe Perches <joe@perches.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Use a more current kernel messaging style.

Convert a printk block to print_hex_dump.
Coalesce formats, align arguments.
Use %s, __func__ instead of embedding function names.

Some messages that were prefixed with <foo>_close are
now prefixed with <foo>_fini.  Some ah4 and esp messages
are now not prefixed with "ip ".

The intent of this patch is to later add something like
  #define pr_fmt(fmt) "IPv4: " fmt.
to standardize the output messages.

Text size is trivially reduced. (x86-32 allyesconfig)

$ size net/ipv4/built-in.o*
   text	   data	    bss	    dec	    hex	filename
 887888	  31558	 249696	1169142	 11d6f6	net/ipv4/built-in.o.new
 887934	  31558	 249800	1169292	 11d78c	net/ipv4/built-in.o.old
Signed-off-by: NJoe Perches <joe@perches.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When the number of dentry cache hash table entries gets too high
(2147483648 entries), as happens by default on a 16TB system, use of a
signed integer in the dcache_init() initialization loop prevents the
dentry_hashtable from getting initialized, causing a panic in
__d_lookup().  Fix this in dcache_init() and similar areas.
Signed-off-by: NDimitri Sivanich <sivanich@sgi.com>
Acked-by: NDavid S. Miller <davem@davemloft.net>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Commit 4acb4190 tries to fix the using uninitialized value
introduced by commit 3dc43e3e,  but it would make the
per-socket memory limits too small.

This patch fixes this and also remove the redundant codes
introduced in 4acb4190.
Signed-off-by: NJason Wang <jasowang@redhat.com>
Acked-by: NGlauber Costa <glommer@parallels.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Some of our machines were reporting:

TCP: too many of orphaned sockets

even when the number of orphaned sockets was well below the
limit.

We print a different message depending on whether we're out
of TCP memory or there are too many orphaned sockets.

Also move the check out of line and cleanup the messages
that were printed.
Signed-off-by: NArun Sharma <asharma@fb.com>
Suggested-by: NMohan Srinivasan <mohan@fb.com>
Cc: netdev@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Cc: David Miller <davem@davemloft.net>
Cc: Glauber Costa <glommer@parallels.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Joe Perches <joe@perches.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

sysctl_tcp_mem() initialization was moved to sysctl_tcp_ipv4.c
in commit 3dc43e3e, since it
became a per-ns value.

That code, however, will never run when CONFIG_SYSCTL is
disabled, leading to bogus values on those fields - causing hung
TCP sockets.

This patch fixes it by keeping an initialization code in
tcp_init(). It will be overwritten by the first net namespace
init if CONFIG_SYSCTL is compiled in, and do the right thing if
it is compiled out.

It is also named properly as tcp_init_mem(), to properly signal
its non-sysctl side effect on TCP limits.
Reported-by: NIngo Molnar <mingo@elte.hu>
Signed-off-by: NGlauber Costa <glommer@parallels.com>
Cc: David S. Miller <davem@davemloft.net>
Link: http://lkml.kernel.org/r/4F22D05A.8030604@parallels.com
[ renamed the function, tidied up the changelog a bit ]
Signed-off-by: NIngo Molnar <mingo@elte.hu>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch allows each namespace to independently set up
its levels for tcp memory pressure thresholds. This patch
alone does not buy much: we need to make this values
per group of process somehow. This is achieved in the
patches that follows in this patchset.
Signed-off-by: NGlauber Costa <glommer@parallels.com>
Reviewed-by: NKAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
CC: David S. Miller <davem@davemloft.net>
CC: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

As mentioned by Joe Perches, TCP_OFF() and TCP_PAGE() macros are
useless.
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

If our TCP_PAGE(sk) is not shared (page_count() == 1), we can set page
offset to 0.

This permits better filling of the pages on small to medium tcp writes.

"tbench 16" results on my dev server (2x4x2 machine) :

Before : 3072 MB/s
After  : 3146 MB/s  (2.4 % gain)
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

tcp_sendmsg() uses select_size() helper to choose skb head size when a
new skb must be allocated.

If GSO is enabled for the socket, current strategy is to force all
payload data to be outside of headroom, in PAGE fragments.

This strategy is not welcome for small packets, wasting memory.

Experiments show that best results are obtained when using 2048 bytes
for skb head (This includes the skb overhead and various headers)

This patch provides better len/truesize ratios for packets sent to
loopback device, and reduce memory needs for in-flight loopback packets,
particularly on arches with big pages.

If a sender sends many 1-byte packets to an unresponsive application,
receiver rmem_alloc will grow faster and will stop queuing these packets
sooner, or will collapse its receive queue to free excess memory.

netperf -t TCP_RR results are improved by ~4 %, and many workloads are
improved as well (tbench, mysql...)
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Now sk_route_caps is u64, its dangerous to use an integer to store
result of an AND operator. It wont work if NETIF_F_SG is moved on the
upper part of u64.
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
CC: Michał Mirosław <mirq-linux@rere.qmqm.pl>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

v2:	add couple missing conversions in drivers
	split unexporting netdev_fix_features()
	implemented %pNF
	convert sock::sk_route_(no?)caps
Signed-off-by: NMichał Mirosław <mirq-linux@rere.qmqm.pl>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

It was enabled by default and the messages guarded
by the define are useful.
Signed-off-by: NFlavio Leitner <fbl@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

tcp_md5_hash_header() writes into skb header a temporary zero value,
this might confuse other users of this area.

Since tcphdr is small (20 bytes), copy it in a temporary variable and
make the change in the copy.
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Adding const qualifiers to pointers can ease code review, and spot some
bugs. It might allow compiler to optimize code further.

For example, is it legal to temporary write a null cksum into tcphdr
in tcp_md5_hash_header() ? I am afraid a sniffer could catch the
temporary null value...
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

To ease skb->truesize sanitization, its better to be able to localize
all references to skb frags size.

Define accessors : skb_frag_size() to fetch frag size, and
skb_frag_size_{set|add|sub}() to manipulate it.
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Allows ss command (iproute2) to display "ecnseen" if at least one packet
with ECT(0) or ECT(1) or ECN was received by this socket.

"ecn" means ECN was negotiated at session establishment (TCP level)

"ecnseen" means we received at least one packet with ECT fields set (IP
level)

ss -i
...
ESTAB      0      0   192.168.20.110:22  192.168.20.144:38016
ino:5950 sk:f178e400
	 mem:(r0,w0,f0,t0) ts sack ecn ecnseen bic wscale:7,8 rto:210
rtt:12.5/7.5 cwnd:10 send 9.3Mbps rcv_space:14480
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Rename struct tcp_skb_cb "flags" to "tcp_flags" to ease code review and
maintenance.

Its content is a combination of FIN/SYN/RST/PSH/ACK/URG/ECE/CWR flags
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

tcp_md5sig_pool is currently an 'array' (a percpu object) of pointers to
struct tcp_md5sig_pool. Only the pointers are NUMA aware, but objects
themselves are all allocated on a single node.

Remove this extra indirection to get proper percpu memory (NUMA aware)
and make code simpler.
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NIan Campbell <ian.campbell@citrix.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
Cc: "Pekka Savola (ipv6)" <pekkas@netcore.fi>
Cc: James Morris <jmorris@namei.org>
Cc: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
Cc: Patrick McHardy <kaber@trash.net>
Cc: netdev@vger.kernel.org
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Current tcp/udp/sctp global memory limits are not taking into account
hugepages allocations, and allow 50% of ram to be used by buffers of a
single protocol [ not counting space used by sockets / inodes ...]

Lets use nr_free_buffer_pages() and allow a default of 1/8 of kernel ram
per protocol, and a minimum of 128 pages.
Heavy duty machines sysadmins probably need to tweak limits anyway.


References: https://bugzilla.stlinux.com/show_bug.cgi?id=38032Reported-by: Nstarlight <starlight@binnacle.cx>
Suggested-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch uses __copy_from_user_nocache on transmit to bypass data
cache for a performance improvement.  skb_add_data_nocache and
skb_copy_to_page_nocache can be called by sendmsg functions to use
this feature, initial support is in tcp_sendmsg.  This functionality is
configurable per device using ethtool.

Presumably, this feature would only be useful when the driver does
not touch the data.  The feature is turned on by default if a device
indicates that it does some form of checksum offload; it is off by
default for devices that do no checksum offload or indicate no checksum
is necessary.  For the former case copy-checksum is probably done
anyway, in the latter case the device is likely loopback in which case
the no cache copy is probably not beneficial.

This patch was tested using 200 instances of netperf TCP_RR with
1400 byte request and one byte reply.  Platform is 16 core AMD x86.

No-cache copy disabled:
   672703 tps, 97.13% utilization
   50/90/99% latency:244.31 484.205 1028.41

No-cache copy enabled:
   702113 tps, 96.16% utilization,
   50/90/99% latency 238.56 467.56 956.955

Using 14000 byte request and response sizes demonstrate the
effects more dramatically:

No-cache copy disabled:
   79571 tps, 34.34 %utlization
   50/90/95% latency 1584.46 2319.59 5001.76

No-cache copy enabled:
   83856 tps, 34.81% utilization
   50/90/95% latency 2508.42 2622.62 2735.88

Note especially the effect on latency tail (95th percentile).

This seems to provide a nice performance improvement and is
consistent in the tests I ran.  Presumably, this would provide
the greatest benfits in the presence of an application workload
stressing the cache and a lot of transmit data happening.
Signed-off-by: NTom Herbert <therbert@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

In contrast to SIOCOUTQ which returns the amount of data sent
but not yet acknowledged plus data not yet sent this patch only
returns the data not sent.

For various methods of live streaming bitrate control it may
be helpful to know how much data are in the tcp outqueue are
not sent yet.
Signed-off-by: NMario Schuknecht <m.schuknecht@dresearch.de>
Signed-off-by: NSteffen Sledz <sledz@dresearch.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Now, TCP_CHECK_TIMER is not used for debuging, it does nothing.
And, it has been there for several years, maybe 6 years.

Remove it to keep code clearer.
Signed-off-by: NShan Wei <shanwei@cn.fujitsu.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Quoting Ben Hutchings: we presumably won't be defining features that
can only be enabled on 64-bit architectures.

Occurences found by `grep -r` on net/, drivers/net, include/

[ Move features and vlan_features next to each other in
  struct netdev, as per Eric Dumazet's suggestion -DaveM ]
Signed-off-by: NMichał Mirosław <mirq-linux@rere.qmqm.pl>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Use TCP_MIN_MSS instead of constant 64.
Reported-by: NMin Zhang <mzhang@mvista.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

As noted by Steve Chen, since commit
f5fff5dc ("tcp: advertise MSS
requested by user") we can end up with a situation where
tcp_select_initial_window() does a divide by a zero (or
even negative) mss value.

The problem is that sometimes we effectively subtract
TCPOLEN_TSTAMP_ALIGNED and/or TCPOLEN_MD5SIG_ALIGNED from the mss.

Fix this by increasing the minimum from 8 to 64.
Reported-by: NSteve Chen <schen@mvista.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Robin Holt tried to boot a 16TB machine and found some limits were
reached : sysctl_tcp_mem[2], sysctl_udp_mem[2]

We can switch infrastructure to use long "instead" of "int", now
atomic_long_t primitives are available for free.
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Reported-by: NRobin Holt <holt@sgi.com>
Reviewed-by: NRobin Holt <holt@sgi.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Coalesce long formats.
Align arguments.
Remove KERN_<level>.
Signed-off-by: NJoe Perches <joe@perches.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Fixes kernel bugzilla #16603

tcp_sendmsg() truncates iov_len to an 'int' which a 4GB write to write
zero bytes, for example.

There is also the problem higher up of how verify_iovec() works.  It
wants to prevent the total length from looking like an error return
value.

However it does this using 'int', but syscalls return 'long' (and
thus signed 64-bit on 64-bit machines).  So it could trigger
false-positives on 64-bit as written.  So fix it to use 'long'.
Reported-by: NOlaf Bonorden <bono@onlinehome.de>
Reported-by: NDaniel Büse <dbuese@gmx.de>
Reported-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

If a RST comes in immediately after checking sk->sk_err, tcp_poll will
return POLLIN but not POLLOUT.  Fix this by checking sk->sk_err at the end
of tcp_poll.  Additionally, ensure the correct order of operations on SMP
machines with memory barriers.
Signed-off-by: NTom Marshall <tdm.code@gmail.com>
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch provides a "user timeout" support as described in RFC793. The
socket option is also needed for the the local half of RFC5482 "TCP User
Timeout Option".

TCP_USER_TIMEOUT is a TCP level socket option that takes an unsigned int,
when > 0, to specify the maximum amount of time in ms that transmitted
data may remain unacknowledged before TCP will forcefully close the
corresponding connection and return ETIMEDOUT to the application. If
0 is given, TCP will continue to use the system default.

Increasing the user timeouts allows a TCP connection to survive extended
periods without end-to-end connectivity. Decreasing the user timeouts
allows applications to "fail fast" if so desired. Otherwise it may take
upto 20 minutes with the current system defaults in a normal WAN
environment.

The socket option can be made during any state of a TCP connection, but
is only effective during the synchronized states of a connection
(ESTABLISHED, FIN-WAIT-1, FIN-WAIT-2, CLOSE-WAIT, CLOSING, or LAST-ACK).
Moreover, when used with the TCP keepalive (SO_KEEPALIVE) option,
TCP_USER_TIMEOUT will overtake keepalive to determine when to close a
connection due to keepalive failure.

The option does not change in anyway when TCP retransmits a packet, nor
when a keepalive probe will be sent.

This option, like many others, will be inherited by an acceptor from its
listener.
Signed-off-by: NH.K. Jerry Chu <hkchu@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This issue come from ruby language community. Below test program
hang up when only run on Linux.

	% uname -mrsv
	Linux 2.6.26-2-486 #1 Sat Dec 26 08:37:39 UTC 2009 i686
	% ruby -rsocket -ve '
	BasicSocket.do_not_reverse_lookup = true
	serv = TCPServer.open("127.0.0.1", 0)
	s1 = TCPSocket.open("127.0.0.1", serv.addr[1])
	s2 = serv.accept
	s2.close
	s1.write("a") rescue p $!
	s1.write("a") rescue p $!
	Thread.new {
	  s1.write("a")
	}.join'
	ruby 1.9.3dev (2010-07-06 trunk 28554) [i686-linux]
	#<Errno::EPIPE: Broken pipe>
	[Hang Here]

FreeBSD, Solaris, Mac doesn't. because Ruby's write() method call
select() internally. and tcp_poll has a bug.

SUS defined 'ready for writing' of select() as following.

|  A descriptor shall be considered ready for writing when a call to an output
|  function with O_NONBLOCK clear would not block, whether or not the function
|  would transfer data successfully.

That said, EPIPE situation is clearly one of 'ready for writing'.

We don't have read-side issue because tcp_poll() already has read side
shutdown care.

|        if (sk->sk_shutdown & RCV_SHUTDOWN)
|                mask |= POLLIN | POLLRDNORM | POLLRDHUP;

So, Let's insert same logic in write side.

- reference url
  http://blade.nagaokaut.ac.jp/cgi-bin/scat.rb/ruby/ruby-core/31065
  http://blade.nagaokaut.ac.jp/cgi-bin/scat.rb/ruby/ruby-core/31068Signed-off-by: NKOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>