[PATCH net-next] sock: do not set sk_err in sock_dequeue_err_skb

From: Soheil Hassas Yeganeh <soh...@google.com> Do not set sk_err when dequeuing errors from the error queue. Doing so results in: a) Bugs: By overwriting existing sk_err values, it possibly hides legitimate errors. It is also incorrect when local errors are queued with ip_local

Re: [PATCH net] tcp: fix return value for partial writes

> Issue was diagnosed by Soheil and Willem, big kudos to them ! > > Fixes: d41a69f1d390f ("tcp: make tcp_sendmsg() aware of socket backlog") > Signed-off-by: Eric Dumazet <eduma...@google.com> > Cc: Willem de Bruijn <will...@google.com> > Cc: Soheil Hassas Y

Re: [PATCH net] net: clear sk_err_soft in sk_clone_lock()

ue in the child, as it > makes future getsockopt(SO_ERROR) calls quite unreliable. > > Signed-off-by: Eric Dumazet <eduma...@google.com> Acked-by: Soheil Hassas Yeganeh <soh...@google.com> > --- > net/core/sock.c |1 + > 1 file changed, 1 insertion(+

Re: [PATCH net] inet: Fix missing return value in inet6_hash

in the IPv6 case. > > Fixes: 496611d7b5ea ("inet: create IPv6-equivalent inet_hash function") > Reported-by: Soheil Hassas Yeganeh <soh...@google.com> > Signed-off-by: Craig Gallek <kr...@google.com> Acked-by: Soheil Hassas Yeganeh <soh...@google.com> >

Re: [PATCH net-next 1/2] tcp: measure rwnd-limited time

nt) in tcp_sock to carefully handle tcp_info's reading of > rwnd_limited_ts and rwnd_limited in order to get a consistent snapshot > of both variables together. > > Signed-off-by: Francis Y. Yan <francisy...@gmail.com> > Signed-off-by: Yuchung Cheng <ych...@google.com> Acked-by: Soheil Hassas Yeganeh <soh...@google.com>

Re: [PATCH net-next 2/2] tcp: put a TLV list of TCP stats in error queue

sage with a cmsg_type of > SCM_OPT_STATS containing the list of TLVs in its cmsg_data. > > Signed-off-by: Francis Y. Yan <francisy...@gmail.com> > Signed-off-by: Yuchung Cheng <ych...@google.com> Acked-by: Soheil Hassas Yeganeh <soh...@google.com>

[PATCH net] tun: fix transmit timestamp support

From: Soheil Hassas Yeganeh <soh...@google.com> Instead of using sock_tx_timestamp, use skb_tx_timestamp to record software transmit timestamp of a packet. sock_tx_timestamp resets and overrides the tx_flags of the skb. The function is intended to be called from within the protocol laye

Re: [PATCH net-next] tcp: refine tcp_prune_ofo_queue() to not drop all packets

esources. > > This patch cleans only a part of the out of order queue in order > to meet the memory constraints. > > Signed-off-by: Eric Dumazet <eduma...@google.com> > Cc: Neal Cardwell <ncardw...@google.com> > Cc: Yuchung Cheng <ych...@google.com> > C

[PATCH v2 net] tcp: consider recv buf for the initial window scale

From: Soheil Hassas Yeganeh <soh...@google.com> tcp_select_initial_window() intends to advertise a window scaling for the maximum possible window size. To do so, it considers the maximum of net.ipv4.tcp_rmem[2] and net.core.rmem_max as the only possible upper-bounds. However,

Re: [PATCH net] tcp: consider recv buf for the initial window scale

On Fri, Jul 29, 2016 at 9:21 AM, Neal Cardwell <ncardw...@google.com> wrote: > > On Thu, Jul 28, 2016 at 11:11 PM, Soheil Hassas Yeganeh > <soheil.k...@gmail.com> wrote: > > > > From: Soheil Hassas Yeganeh <soh...@google.com> > > > > tcp_se

[PATCH net] tcp: consider recv buf for the initial window scale

From: Soheil Hassas Yeganeh <soh...@google.com> tcp_select_initial_window() intends to advertise a window scaling for the maximum possible window size. To do so, it considers the maximum of net.ipv4.tcp_rmem[2] and net.core.rmem_max as the only possible upper-bounds. However,

[PATCH net] packet: propagate sock_cmsg_send() error

From: Soheil Hassas Yeganeh <soh...@google.com> sock_cmsg_send() can return different error codes and not only -EINVAL, and we should properly propagate them. Fixes: c14ac9451c34 ("sock: enable timestamping using control messages") Signed-off-by: Soheil Hassas Yeganeh <so

Re: [PATCH v3] packet: fix second argument of sock_tx_timestamp()

y: Kazuya Mizuguchi <kazuya.mizuguchi...@renesas.com> > Reported-by: Keita Kobayashi <keita.kobayashi...@renesas.com> > Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda...@renesas.com> Acked-by: Soheil Hassas Yeganeh <soh...@google.com> > --- > Changes from v2:

Re: [PATCH net] sock: ignore SCM_RIGHTS and SCM_CREDENTIALS in __sock_cmsg_send

On Mon, Jul 11, 2016 at 4:39 PM, David Miller <da...@davemloft.net> wrote: > > From: Soheil Hassas Yeganeh <soheil.k...@gmail.com> > Date: Sun, 10 Jul 2016 12:51:46 -0400 > > > From: Soheil Hassas Yeganeh <soh...@google.com> > > > > Se

[PATCH v2 net] sock: ignore SCM_RIGHTS and SCM_CREDENTIALS in __sock_cmsg_send

From: Soheil Hassas Yeganeh <soh...@google.com> Sergei Trofimovich reported that pulse audio sends SCM_CREDENTIALS as a control message to TCP. Since __sock_cmsg_send does not support SCM_RIGHTS and SCM_CREDENTIALS, it returns an error and hence breaks pulse audio over TCP. SCM_

[PATCH net] sock: ignore SCM_RIGHTS and SCM_CREDENTIALS in __sock_cmsg_send

From: Soheil Hassas Yeganeh <soh...@google.com> Sergei Trofimovich reported that pulse audio sends SCM_CREDENTIALS as a control message to TCP. Since __sock_cmsg_send does not support SCM_RIGHTS and SCM_CREDENTIALS, it returns an error and hence breaks pulse audio over TCP. SCM_

Re: [BUG?] tcp regression in v4.7-r1: c14ac9451c34832554db33386a4393be8bba3a7b breaks pulseaudio over TCP

On Sun, Jul 10, 2016 at 12:25 PM, Sergei Trofimovich <sly...@gentoo.org> wrote: > On Sun, 10 Jul 2016 11:15:01 -0400 > Soheil Hassas Yeganeh <soh...@google.com> wrote: > >> On Sun, Jul 10, 2016 at 7:42 AM, Sergei Trofimovich <sly...@gentoo.org> >> wrote

Re: [BUG?] tcp regression in v4.7-r1: c14ac9451c34832554db33386a4393be8bba3a7b breaks pulseaudio over TCP

On Sun, Jul 10, 2016 at 7:42 AM, Sergei Trofimovich wrote: > Hi netdev folk! > > Commit c14ac9451c34832554db33386a4393be8bba3a7b > broke pulseaudio (PA) over TCP. Sorry that my patch broke your app and thanks for the bug report. Breaking PA was certainly not my intention. >

Re: [PATCH net-next] sock: propagate __sock_cmsg_send() error

ntrol messages in IPv4") > Fixes: ad1e46a83716 ("ipv6: process socket-level control messages in IPv6") > Signed-off-by: Eric Dumazet <eduma...@google.com> > Cc: Soheil Hassas Yeganeh <soh...@google.com> Acked-by: Soheil Hassas Yeganeh <soh...@google.com> > --- >

Re: [PATCH net-next] sock: ignore TIMESTAMP, RXQ_OVFL, WIFI_STATUS in sock_cmsg_send

On Fri, May 13, 2016 at 2:03 AM, David Miller <da...@davemloft.net> wrote: > From: Soheil Hassas Yeganeh <soheil.k...@gmail.com> > Date: Fri, 13 May 2016 00:47:10 -0400 > >> From: Soheil Hassas Yeganeh <soh...@google.com> >> >> SO_TIMESTAMP(N

[PATCH net-next] sock: ignore TIMESTAMP, RXQ_OVFL, WIFI_STATUS in sock_cmsg_send

From: Soheil Hassas Yeganeh <soh...@google.com> SO_TIMESTAMP(NS), RXQ_OVFL, and WIFI_STATUS can be returned as receive-side control messages from recvmsg(). Although invalid, some applications may reflect those receive-side control messages back to sendmsg(). Since socket-level control me

Re: [PATCH net-next] tcp: guarantee forward progress in tcp_sendmsg()

always return true. > > Fix this by calling sk_flush_backlog() only if one skb had been > allocated and filled before last backlog check. > > Fixes: d41a69f1d390 ("tcp: make tcp_sendmsg() aware of socket backlog") > Signed-off-by: Eric Dumazet <eduma...@google.com> Ack

Re: [PATCH v2 net-next 1/7] tcp: do not assume TCP code is non preemptible

On Fri, Apr 29, 2016 at 10:37 AM, Eric Dumazet <eric.duma...@gmail.com> wrote: > On Fri, 2016-04-29 at 09:18 -0400, Soheil Hassas Yeganeh wrote: >> On Thu, Apr 28, 2016 at 11:10 PM, Eric Dumazet <eduma...@google.com> wrote: > >> > +

Re: [PATCH v2 net-next 6/7] net: do not block BH while processing socket backlog

> once ring buffer is filled. > > All users are now ready to be called from process context, > we can unblock BH and let interrupts be serviced faster. > > cond_resched_softirq() could be removed, as it has no more user. > > Signed-off-by: Eric Dumazet <eduma...@goo

Re: [PATCH v2 net-next 4/7] udp: prepare for non BH masking at backlog processing

-by: Eric Dumazet <eduma...@google.com> Acked-by: Soheil Hassas Yeganeh <soh...@google.com> > --- > net/ipv4/udp.c | 4 ++-- > net/ipv6/udp.c | 4 ++-- > 2 files changed, 4 insertions(+), 4 deletions(-) > > diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c > index 093284c5c03b..f6

Re: [PATCH v2 net-next 3/7] dccp: do not assume DCCP code is non preemptible

On Thu, Apr 28, 2016 at 11:10 PM, Eric Dumazet <eduma...@google.com> wrote: > DCCP uses the generic backlog code, and this will soon > be changed to not disable BH when protocol is called back. > > Signed-off-by: Eric Dumazet <eduma...@google.com> Acked-by: Soheil Hassas Ye

Re: [PATCH v2 net-next 2/7] tcp: do not block bh during prequeue processing

ve prequeue mode some care"), > processing a batch of packets might take time, better not block BH > at all. > > Signed-off-by: Eric Dumazet <eduma...@google.com> Acked-by: Soheil Hassas Yeganeh <soh...@google.com> > --- > net/ipv4/tcp.c | 4 > net/ipv4/tcp_i

Re: [PATCH v2 net-next 1/7] tcp: do not assume TCP code is non preemptible

On Thu, Apr 28, 2016 at 11:10 PM, Eric Dumazet wrote: > We want to to make TCP stack preemptible, as draining prequeue > and backlog queues can take lot of time. > > Many SNMP updates were assuming that BH (and preemption) was disabled. > > Need to convert some

Re: [PATCH v2 net-next 7/7] tcp: make tcp_sendmsg() aware of socket backlog

e to not call tcp_release_cb(), > but we might consider this later. > > Signed-off-by: Eric Dumazet <eduma...@google.com> > Cc: Soheil Hassas Yeganeh <soh...@google.com> > Cc: Alexei Starovoitov <a...@fb.com> > Cc: Marcelo Ricardo Leitner <marcelo.leit

[PATCH v2 net-next 2/2] tcp: remove SKBTX_ACK_TSTAMP since it is redundant

From: Soheil Hassas Yeganeh <soh...@google.com> The SKBTX_ACK_TSTAMP flag is set in skb_shinfo->tx_flags when the timestamp of the TCP acknowledgement should be reported on error queue. Since accessing skb_shinfo is likely to incur a cache-line miss at the time of receivin

[PATCH v2 net-next 1/2] tcp: remove an unnecessary check in tcp_tx_timestamp

From: Soheil Hassas Yeganeh <soh...@google.com> Remove the redundant check for sk->sk_tsflags in tcp_tx_timestamp. tcp_tx_timestamp() receives the tsflags as a parameter. As a result the "sk->sk_tsflags || tsflags" is redundant, since tsflags already includes sk->sk_ts

[PATCH v2 net-next 0/2] tcp: simplify ack tx timestamps

From: Soheil Hassas Yeganeh <soh...@google.com> v2: - Fully remove SKBTX_ACK_TSTAMP, as suggested by Willem de Bruijn. This patch series aims at removing redundant checks and fields for ack timestamps for TCP. Soheil Hassas Yeganeh (2): tcp: remove an unnecessary check in tcp_tx_tim

Re: [PATCH net-next 2/2] tcp: remove a redundant check for SKBTX_ACK_TSTAMP

On Wed, Apr 27, 2016 at 2:19 PM, Martin KaFai Lau <ka...@fb.com> wrote: > On Mon, Apr 25, 2016 at 04:51:13PM -0400, Soheil Hassas Yeganeh wrote: >> From: Soheil Hassas Yeganeh <soh...@google.com> >> >> txstamp_ack in tcp_skb_cb is set iff the SKBTX_ACK_TSTAM

Re: [PATCH v4 net-next 0/3] tcp: Make use of MSG_EOR in tcp_sendmsg

; tcp_sendpage because there could be >1 threads doing > sendmsg. > ~ Thanks to Eric Dumazet's suggestions on v2. > ~ The TCP timestamp bug fixes are separated into other threads. > > v2: > ~ Rework based on the recent work > "add TX timestamping via cmsg"

Re: [PATCH v4 net-next 3/3] tcp: Handle eor bit when fragmenting a skb

F. 1:1(0) ack 16062 win 257 > 0.400 > . 16062:16062(0) ack 2 > > Signed-off-by: Martin KaFai Lau <ka...@fb.com> > Cc: Eric Dumazet <eduma...@google.com> > Cc: Neal Cardwell <ncardw...@google.com> > Cc: Soheil Hassas Yeganeh <soh...@google.com> > Cc:

Re: [PATCH v4 net-next 2/3] tcp: Handle eor bit when coalescing skb

730) ack 1 > 0.300 > P. 731:1461(730) ack 1 > 0.400 < . 1:1(0) ack 13141 win 257 > > 0.400 close(4) = 0 > 0.400 > F. 13141:13141(0) ack 1 > 0.500 < F. 1:1(0) ack 13142 win 257 > 0.500 > . 13142:13142(0) ack 2 > > Signed-off-by: Martin KaFai Lau <ka...@fb

Re: [PATCH v4 net-next 1/3] tcp: Make use of MSG_EOR in tcp_sendmsg

F. 16061:16061(0) ack 1 > 0.400 < F. 1:1(0) ack 16062 win 257 > 0.400 > . 16062:16062(0) ack 2 > > Signed-off-by: Martin KaFai Lau <ka...@fb.com> > Cc: Eric Dumazet <eduma...@google.com> > Cc: Neal Cardwell <ncardw...@google.com> > Cc: Soheil Hassa

[PATCH net-next 2/2] tcp: remove a redundant check for SKBTX_ACK_TSTAMP

From: Soheil Hassas Yeganeh <soh...@google.com> txstamp_ack in tcp_skb_cb is set iff the SKBTX_ACK_TSTAMP flag is set for an skb. Thus, it is not required to check shinfo->tx_flags if the txstamp_ack bit is checked. Remove the check on shinfo->tx_flags & SKBTX_ACK_TSTAMP, sinc

[PATCH net-next 1/2] tcp: remove an unnecessary check in tcp_tx_timestamp

From: Soheil Hassas Yeganeh <soh...@google.com> tcp_tx_timestamp() receives tsflags as a parameter. tsflags is initialized to sk->sk_tsflags and is overridden by control messages. As a result the "sk->sk_tsflags || tsflags" is the same expression as "tsflags" R

Re: [RFC PATCH v3 net-next 2/3] tcp: Handle eor bit when coalescing skb

On Thu, Apr 21, 2016 at 12:56 PM, Martin KaFai Lau <ka...@fb.com> wrote: > On Wed, Apr 20, 2016 at 04:04:54PM -0400, Soheil Hassas Yeganeh wrote: >> > diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c >> > index a6e4a83..96bdf98 100644 >> > --- a/net/i

Re: [RFC PATCH v3 net-next 2/3] tcp: Handle eor bit when coalescing skb

gned-off-by: Martin KaFai Lau <ka...@fb.com> > Cc: Eric Dumazet <eduma...@google.com> > Cc: Neal Cardwell <ncardw...@google.com> > Cc: Soheil Hassas Yeganeh <soh...@google.com> > Cc: Willem de Bruijn <will...@google.com> > Cc: Yuchung Cheng <ych...@google.co

Re: [PATCH net-next 2/2] tcp: Merge txstamp_ack in tcp_skb_collapse_tstamp

1 win 257 > 0.300 > P. 1:1461(1460) ack 1 > 0.400 < . 1:1(0) ack 14601 win 257 > > BPF Output Before: > ~ > > > BPF Output After: > ~ > <...>-2049 [007] d.s.89.185538: : ee_data:14599 > > Signed-off-by: Martin KaFai Lau <ka...@fb.c

Re: [PATCH net-next 1/2] tcp: Carry txstamp_ack in tcp_fragment_tstamp

00 < F. 1:1(0) ack 16062 win 257 > 0.400 > . 14602:14602(0) ack 2 > > Signed-off-by: Martin KaFai Lau <ka...@fb.com> > Cc: Eric Dumazet <eduma...@google.com> > Cc: Neal Cardwell <ncardw...@google.com> > Cc: Soheil Hassas Yeganeh <soh...@google.com> >

Re: [PATCH net 2/2] tcp: Merge tx_flags and tskey in tcp_shifted_skb

:14601(5840) ack 1 > > 0.300 < . 1:1(0) ack 1 win 257 > 0.300 > P. 1:1461(1460) ack 1 > 0.400 < . 1:1(0) ack 14601 win 257 > > 0.400 close(4) = 0 > 0.400 > F. 14601:14601(0) ack 1 > 0.500 < F. 1:1(0) ack 14602 win 257 > 0.500 > . 14602:14602(0) ack 2 &

Re: [PATCH net 1/2] tcp: Merge tx_flags and tskey in tcp_collapse_retrans

:1(0) ack 13142 win 257 > 0.500 > . 13142:13142(0) ack 2 > > Signed-off-by: Martin KaFai Lau <ka...@fb.com> > Cc: Eric Dumazet <eduma...@google.com> > Cc: Neal Cardwell <ncardw...@google.com> > Cc: Soheil Hassas Yeganeh <soh...@google.com> > Cc: Wille

Re: [PATCH net] tcp: Fix SOF_TIMESTAMPING_TX_ACK when handling dup acks

On Tue, Apr 19, 2016 at 9:54 AM, Soheil Hassas Yeganeh <soh...@google.com> wrote: > On Mon, Apr 18, 2016 at 6:39 PM, Martin KaFai Lau <ka...@fb.com> wrote: >> Assuming SOF_TIMESTAMPING_TX_ACK is on. When dup acks are received, >> it could incorrectly think that a sk

Re: [RFC PATCH v2 net-next 2/7] tcp: Merge tx_flags/tskey/txstamp_ack in tcp_collapse_retrans

On Tue, Apr 19, 2016 at 2:18 PM, Martin KaFai Lau wrote: > On Tue, Apr 19, 2016 at 10:35:52AM -0700, Eric Dumazet wrote: >> On Tue, Apr 19, 2016 at 10:28 AM, Martin KaFai Lau wrote: >> >> > A bit off topic, I feel like the SKBTX_ACK_TSTAMP and txstamp_ack are sort >>

Re: [RFC PATCH v2 net-next 1/7] tcp: Carry txstamp_ack in tcp_fragment_tstamp

On Tue, Apr 19, 2016 at 1:39 PM, Martin KaFai Lau <ka...@fb.com> wrote: > On Tue, Apr 19, 2016 at 01:21:04AM -0400, Soheil Hassas Yeganeh wrote: >> Could you please submit the timestamping patches separately as non RFCs? >> Thanks! > Agree. I will re-spin. Great, thank you very much!

Re: [RFC PATCH v2 net-next 2/7] tcp: Merge tx_flags/tskey/txstamp_ack in tcp_collapse_retrans

On Tue, Apr 19, 2016 at 1:28 PM, Martin KaFai Lau <ka...@fb.com> wrote: > On Tue, Apr 19, 2016 at 01:32:14AM -0400, Soheil Hassas Yeganeh wrote: >> > + TCP_SKB_CB(skb)->txstamp_ack = >> > + !!(shinfo->tx_flags & SKBTX

Re: [PATCH net] tcp: Fix SOF_TIMESTAMPING_TX_ACK when handling dup acks

tsockopt(4, SOL_SOCKET, 37, [2688], 4) = 0 > 0.200 write(4, ..., 1460) = 1460 > 0.200 write(4, ..., 13140) = 13140 > > 0.200 > P. 1:1461(1460) ack 1 > 0.200 > . 1461:8761(7300) ack 1 > 0.200 > P. 8761:14601(5840) ack 1 > > 0.300 < . 1:1(0) ack 1 win 257 >

Re: [RFC PATCH v2 net-next 3/7] tcp: Merge tx_flags/tskey/txstamp_ack in tcp_shifted_skb

t; F. 1:1(0) ack 14602 win 257 > 0.500 > . 14602:14602(0) ack 2 > > Signed-off-by: Martin KaFai Lau <ka...@fb.com> > Cc: Eric Dumazet <eduma...@google.com> > Cc: Neal Cardwell <ncardw...@google.com> > Cc: Soheil Hassas Yeganeh <soheil.k...@gmail.com> Ac

Re: [RFC PATCH v2 net-next 2/7] tcp: Merge tx_flags/tskey/txstamp_ack in tcp_collapse_retrans

close(4) = 0 > 0.400 > F. 13141:13141(0) ack 1 > 0.500 < F. 1:1(0) ack 13142 win 257 > 0.500 > . 13142:13142(0) ack 2 > > Signed-off-by: Martin KaFai Lau <ka...@fb.com> > Cc: Eric Dumazet <eduma...@google.com> > Cc: Neal Cardwell <ncardw...@google.com> > C

Re: [RFC PATCH v2 net-next 1/7] tcp: Carry txstamp_ack in tcp_fragment_tstamp

k 1 > 0.400 < F. 1:1(0) ack 16062 win 257 > 0.400 > . 14602:14602(0) ack 2 > > Signed-off-by: Martin KaFai Lau <ka...@fb.com> > Cc: Eric Dumazet <eduma...@google.com> > Cc: Neal Cardwell <ncardw...@google.com> > Cc: Soheil Hassas Yeganeh <soheil.k...

[PATCH v3 net-next 6/8] ipv6: process socket-level control messages in IPv6

From: Soheil Hassas Yeganeh <soh...@google.com> Process socket-level control messages by invoking __sock_cmsg_send in ip6_datagram_send_ctl for control messages on the SOL_SOCKET layer. This makes sure whenever ip6_datagram_send_ctl is called for udp and raw, we also process socket-level c

[PATCH v3 net-next 8/8] sock: document timestamping via cmsg in Documentation

From: Soheil Hassas Yeganeh <soh...@google.com> Update docs and add code snippet for using cmsg for timestamping. Signed-off-by: Soheil Hassas Yeganeh <soh...@google.com> Acked-by: Willem de Bruijn <will...@google.com> --- Documentation/networking/tim

[PATCH v3 net-next 5/8] ipv4: process socket-level control messages in IPv4

From: Soheil Hassas Yeganeh <soh...@google.com> Process socket-level control messages by invoking __sock_cmsg_send in ip_cmsg_send for control messages on the SOL_SOCKET layer. This makes sure whenever ip_cmsg_send is called in udp, icmp, and raw, we also process socket-level control me

[PATCH v3 net-next 7/8] sock: enable timestamping using control messages

From: Soheil Hassas Yeganeh <soh...@google.com> Currently, SOL_TIMESTAMPING can only be enabled using setsockopt. This is very costly when users want to sample writes to gather tx timestamps. Add support for enabling SO_TIMESTAMPING via control messages by using tsflags added in `

[PATCH v3 net-next 4/8] sock: accept SO_TIMESTAMPING flags in socket cmsg

From: Soheil Hassas Yeganeh <soh...@google.com> Accept SO_TIMESTAMPING in control messages of the SOL_SOCKET level as a basis to accept timestamping requests per write. This implementation only accepts TX recording flags (i.e., SOF_TIMESTAMPING_TX_HARDWARE, SOF_TIMESTAMPING_TX_SO

[PATCH v3 net-next 2/8] tcp: accept SOF_TIMESTAMPING_OPT_ID for passive TFO

From: Soheil Hassas Yeganeh <soh...@google.com> SOF_TIMESTAMPING_OPT_ID is set to get data-independent IDs to associate timestamps with send calls. For TCP connections, tp->snd_una is used as the starting point to calculate relative IDs. This socket option will fail if set before the

[PATCH v3 net-next 3/8] tcp: use one bit in TCP_SKB_CB to mark ACK timestamps

From: Soheil Hassas Yeganeh <soh...@google.com> Currently, to avoid a cache line miss for accessing skb_shinfo, tcp_ack_tstamp skips socket that do not have SOF_TIMESTAMPING_TX_ACK bit set in sk_tsflags. This is implemented based on an implicit assumption that the SOF_TIMESTAMPING_TX_ACK

[PATCH v3 net-next 1/8] sock: break up sock_cmsg_snd into __sock_cmsg_snd and loop

logic from the loop that walks the list, to allow having this called directly from a walker in the protocol specific code. Signed-off-by: Willem de Bruijn <will...@google.com> Signed-off-by: Soheil Hassas Yeganeh <soh...@google.com> --- include/net/sock.h | 2 ++ net/c

[PATCH v3 net-next 0/8] add TX timestamping via cmsg

From: Soheil Hassas Yeganeh <soh...@google.com> This patch series aim at enabling TX timestamping via cmsg. Currently, to occasionally sample TX timestamping on a socket, applications need to call setsockopt twice: first for enabling timestamps and then for disabling them. This is an unnec

Re: [PATCH v2 net-next 0/8] add TX timestamping via cmsg

On Sat, Apr 2, 2016 at 9:27 PM, David Miller wrote: > From: David Miller > Date: Sat, 02 Apr 2016 21:19:42 -0400 (EDT) > >> Series applied, thanks. > > I had to revert, this breaks the build: > > net/l2tp/l2tp_ip6.c: In function ‘l2tp_ip6_sendmsg’: >

[PATCH v2 net-next 0/8] add TX timestamping via cmsg

From: Soheil Hassas Yeganeh <soh...@google.com> This patch series aim at enabling TX timestamping via cmsg. Currently, to occasionally sample TX timestamping on a socket, applications need to call setsockopt twice: first for enabling timestamps and then for disabling them. This is an unnec

[PATCH v2 net-next 2/8] tcp: accept SOF_TIMESTAMPING_OPT_ID for passive TFO

From: Soheil Hassas Yeganeh <soh...@google.com> SOF_TIMESTAMPING_OPT_ID is set to get data-independent IDs to associate timestamps with send calls. For TCP connections, tp->snd_una is used as the starting point to calculate relative IDs. This socket option will fail if set before the

[PATCH v2 net-next 3/8] tcp: use one bit in TCP_SKB_CB to mark ACK timestamps

From: Soheil Hassas Yeganeh <soh...@google.com> Currently, to avoid a cache line miss for accessing skb_shinfo, tcp_ack_tstamp skips socket that do not have SOF_TIMESTAMPING_TX_ACK bit set in sk_tsflags. This is implemented based on an implicit assumption that the SOF_TIMESTAMPING_TX_ACK

[PATCH v2 net-next 1/8] sock: break up sock_cmsg_snd into __sock_cmsg_snd and loop

logic from the loop that walks the list, to allow having this called directly from a walker in the protocol specific code. Signed-off-by: Willem de Bruijn <will...@google.com> Signed-off-by: Soheil Hassas Yeganeh <soh...@google.com> --- include/net/sock.h | 2 ++ net/c

[PATCH v2 net-next 8/8] sock: document timestamping via cmsg in Documentation

From: Soheil Hassas Yeganeh <soh...@google.com> Update docs and add code snippet for using cmsg for timestamping. Signed-off-by: Soheil Hassas Yeganeh <soh...@google.com> Acked-by: Willem de Bruijn <will...@google.com> --- Documentation/networking/tim

[PATCH v2 net-next 7/8] sock: enable timestamping using control messages

From: Soheil Hassas Yeganeh <soh...@google.com> Currently, SOL_TIMESTAMPING can only be enabled using setsockopt. This is very costly when users want to sample writes to gather tx timestamps. Add support for enabling SO_TIMESTAMPING via control messages by using tsflags added in `

[PATCH v2 net-next 6/8] ipv6: process socket-level control messages in IPv6

From: Soheil Hassas Yeganeh <soh...@google.com> Process socket-level control messages by invoking __sock_cmsg_send in ip6_datagram_send_ctl for control messages on the SOL_SOCKET layer. This makes sure whenever ip6_datagram_send_ctl is called for udp and raw, we also process socket-level c

[PATCH v2 net-next 5/8] ipv4: process socket-level control messages in IPv4

From: Soheil Hassas Yeganeh <soh...@google.com> Process socket-level control messages by invoking __sock_cmsg_send in ip_cmsg_send for control messages on the SOL_SOCKET layer. This makes sure whenever ip_cmsg_send is called in udp, icmp, and raw, we also process socket-level control me

[PATCH v2 net-next 4/8] sock: accept SO_TIMESTAMPING flags in socket cmsg

From: Soheil Hassas Yeganeh <soh...@google.com> Accept SO_TIMESTAMPING in control messages of the SOL_SOCKET level as a basis to accept timestamping requests per write. This implementation only accepts TX recording flags (i.e., SOF_TIMESTAMPING_TX_HARDWARE, SOF_TIMESTAMPING_TX_SO

Re: [PATCH net-next 8/8] sock: document timestamping via cmsg in Documentation

Hi Lino, On Thu, Mar 31, 2016 at 5:57 AM, Lino Sanfilippo <lsan...@marvell.com> wrote: > > Hi, > > > On 31.03.2016 00:37, Soheil Hassas Yeganeh wrote: > >> + >> +Moreover, applications must still enable timestamp reporting via >> +setsockopt

Re: [PATCH net-next 0/8] add TX timestamping via cmsg

On Wed, Mar 30, 2016 at 8:38 PM, Martin KaFai Lau <ka...@fb.com> wrote: > On Wed, Mar 30, 2016 at 06:37:20PM -0400, Soheil Hassas Yeganeh wrote: >> I will follow up with another patch to enable timestamping for >> active TFO (client-side TCP Fast Open) and also setting pac

[PATCH net-next 6/8] ipv6: process socket-level control messages in IPv6

From: Soheil Hassas Yeganeh <soh...@google.com> Process socket-level control messages by invoking __sock_cmsg_send in ip6_datagram_send_ctl for control messages on the SOL_SOCKET layer. This makes sure whenever ip6_datagram_send_ctl is called for udp and raw, we also process socket-level c

[PATCH net-next 2/8] tcp: accept SOF_TIMESTAMPING_OPT_ID for passive TFO

From: Soheil Hassas Yeganeh <soh...@google.com> SOF_TIMESTAMPING_OPT_ID is set to get data-independent IDs to associate timestamps with send calls. For TCP connections, tp->snd_una is used as the starting point to calculate relative IDs. This socket option will fail if set before the

[PATCH net-next 3/8] tcp: use one bit in TCP_SKB_CB to mark ACK timestamps

From: Soheil Hassas Yeganeh <soh...@google.com> Currently, to avoid a cache line miss for accessing skb_shinfo, tcp_ack_tstamp skips socket that do not have SOF_TIMESTAMPING_TX_ACK bit set in sk_tsflags. This is implemented based on an implicit assumption that the SOF_TIMESTAMPING_TX_ACK

[PATCH net-next 5/8] ipv4: process socket-level control messages in IPv4

From: Soheil Hassas Yeganeh <soh...@google.com> Process socket-level control messages by invoking __sock_cmsg_send in ip_cmsg_send for control messages on the SOL_SOCKET layer. This makes sure whenever ip_cmsg_send is called in udp, icmp, and raw, we also process socket-level control me

[PATCH net-next 1/8] sock: break up sock_cmsg_snd into __sock_cmsg_snd and loop

logic from the loop that walks the list, to allow having this called directly from a walker in the protocol specific code. Signed-off-by: Willem de Bruijn <will...@google.com> Signed-off-by: Soheil Hassas Yeganeh <soh...@google.com> --- include/net/sock.h | 2 ++ net/c

[PATCH net-next 0/8] add TX timestamping via cmsg

From: Soheil Hassas Yeganeh <soh...@google.com> This patch series aim at enabling TX timestamping via cmsg. Currently, to occasionally sample TX timestamping on a socket, applications need to call setsockopt twice: first for enabling timestamps and then for disabling them. This is an unnec

[PATCH net-next 7/8] sock: enable timestamping using control messages

From: Soheil Hassas Yeganeh <soh...@google.com> Currently, SOL_TIMESTAMPING can only be enabled using setsockopt. This is very costly when users want to sample writes to gather tx timestamps. Add support for enabling SO_TIMESTAMPING via control messages by using tsflags added in `

[PATCH net-next 4/8] sock: accept SO_TIMESTAMPING flags in socket cmsg

From: Soheil Hassas Yeganeh <soh...@google.com> Accept SO_TIMESTAMPING in control messages of the SOL_SOCKET level as a basis to accept timestamping requests per write. This implementation only accepts TX recording flags (i.e., SOF_TIMESTAMPING_TX_HARDWARE, SOF_TIMESTAMPING_TX_SO

[PATCH net-next 8/8] sock: document timestamping via cmsg in Documentation

From: Soheil Hassas Yeganeh <soh...@google.com> Update docs and add code snippet for using cmsg for timestamping. Signed-off-by: Soheil Hassas Yeganeh <soh...@google.com> --- Documentation/networking/timestamping.txt | 48 +-- 1 file changed, 45 inse

<    1   2