refactor push pending
Subsystems (2)
v2 cover refactor push pending
v2:
- add snd_burst check in dfrags loop as Mat suggested.
Refactor __mptcp_push_pending() and __mptcp_subflow_push_pending() to remove duplicate code and support redundant scheduler more easily in __mptcp_subflow_push_pending().
Geliang Tang (4):
mptcp: update __mptcp_push_pending
mptcp: add do_push_pending helper
mptcp: update __mptcp_subflow_push_pending
mptcp: simplify __mptcp_subflow_push_pending
net/mptcp/protocol.c | 153 ++++++++++++++++--------------------------- 1 file changed, 58 insertions(+), 95 deletions(-)
Signature · 1 line
2.35.3
PATCH v2 1/4 mptcp: update __mptcp_push_pending
To support redundant package schedulers more easily, this patch moves the packet scheduler out of the dfrags loop in __mptcp_push_pending(), invoke mptcp_sched_get_send() only once. And update the socket locks correspondingly.
--- net/mptcp/protocol.c | 50 ++++++++++++++++++-------------------------- 1 file changed, 20 insertions(+), 30 deletions(-)
net/mptcp/protocol.c +20 -30
@@ -1417,14 +1417,6 @@ struct sock *mptcp_subflow_get_send(struct mptcp_sock *msk)u64 linger_time;long tout =0 ; -/* re-use last subflow, if the burst allow that */ -if (msk->last_snd && msk->snd_burst >0 && -sk_stream_memory_free (msk->last_snd) && -mptcp_subflow_active (mptcp_subflow_ctx (msk->last_snd))) { -mptcp_set_timeout (sk); -return msk->last_snd; - } -/* pick the subflow with the lower wmem/wspace ratio */ for (i =0 ; i <SSK_MODE_MAX ; ++i) { send_info[i].ssk =NULL ;
@@ -1530,15 +1522,21 @@ void mptcp_check_and_set_pending(struct sock *sk)void __mptcp_push_pending (struct sock *sk,unsigned int flags) { -struct sock *prev_ssk =NULL , *ssk =NULL ;struct mptcp_sock *msk =mptcp_sk (sk);struct mptcp_sendmsg_info info = { .flags = flags, };bool do_check_data_fin = false;struct mptcp_data_frag *dfrag; +struct sock *ssk;int len; + ssk =mptcp_sched_get_send (msk); +if (!ssk) + goto out; + +lock_sock (ssk); +while ((dfrag =mptcp_send_head (sk))) { info.sent = dfrag->already_sent; info.limit = dfrag->data_len;
@@ -1546,24 +1544,6 @@ void __mptcp_push_pending(struct sock *sk, unsigned int flags)while (len >0 ) {int ret =0 ; - prev_ssk = ssk; - ssk =mptcp_sched_get_send (msk); - -/* First check. If the ssk has changed since -* the last round, release prev_ssk -*/ -if (ssk != prev_ssk && prev_ssk) -mptcp_push_release (prev_ssk, &info); -if (!ssk) - goto out; - -/* Need to lock the new subflow only if different -* from the previous one, otherwise we are still -* helding the relevant lock -*/ -if (ssk != prev_ssk) -lock_sock (ssk); - ret =mptcp_sendmsg_frag (sk, ssk, dfrag, &info);if (ret <=0 ) {if (ret == -EAGAIN )
@@ -1579,11 +1559,21 @@ void __mptcp_push_pending(struct sock *sk, unsigned int flags)mptcp_update_post_push (msk, dfrag, ret); }WRITE_ONCE (msk->first_pending,mptcp_send_next (sk)); + +if (msk->snd_burst >0 && +sk_stream_memory_free (ssk) && +mptcp_subflow_active (mptcp_subflow_ctx (ssk))) { +mptcp_set_timeout (sk); + }else { +break ; + } } -/* at this point we held the socket lock for the last subflow we used */ -if (ssk) -mptcp_push_release (ssk, &info); +if (do_check_data_fin) { +tcp_push (ssk,0 , info.mss_now,tcp_sk (ssk)->nonagle, + info.size_goal); + } +release_sock (ssk); out:/* ensure the rtx timer is running */
Signature · 1 line
2.35.3
Re: [PATCH mptcp-next v2 1/4] mptcp: update __mptcp_push_pending
Quoted 16 lines · Geliang Tang
To support redundant package schedulers more easily, this patch moves the packet scheduler out of the dfrags loop in
__mptcp_push_pending(), invokemptcp_sched_get_send()only once. And update the socket locks correspondingly.Signed-off-by Geliang Tang <[email protected]>---
net/mptcp/protocol.c| 50 ++++++++++++++++++--------------------------1 file changed, 20 insertions(+), 30 deletions(-)net/mptcp/protocol.c +0
-0@@ -1417,14 +1417,6 @@ struct sock *mptcp_subflow_get_send(struct mptcp_sock *msk)u64 linger_time;
net/mptcp/protocol.c struct sock *mptcp_subflow_get_send(struct mptcp_sock *msk) · line 1427
Quoted 10 lines
net/mptcp/protocol.c +0
-8long tout =0 ; -/* re-use last subflow, if the burst allow that */ -if (msk->last_snd && msk->snd_burst >0 && -sk_stream_memory_free (msk->last_snd) && -mptcp_subflow_active (mptcp_subflow_ctx (msk->last_snd))) { -mptcp_set_timeout (sk); -return msk->last_snd; - } -
With msk->last_snd not used here any more, it seems like it can be removed.
Quoted 75 lines
net/mptcp/protocol.c +20
-22/* pick the subflow with the lower wmem/wspace ratio */ for (i =0 ; i <SSK_MODE_MAX ; ++i) { send_info[i].ssk =NULL ;@@ -1530,15 +1522,21 @@ void mptcp_check_and_set_pending(struct sock *sk)void __mptcp_push_pending (struct sock *sk,unsigned int flags) { -struct sock *prev_ssk =NULL , *ssk =NULL ;struct mptcp_sock *msk =mptcp_sk (sk);struct mptcp_sendmsg_info info = { .flags = flags, };bool do_check_data_fin = false;struct mptcp_data_frag *dfrag; +struct sock *ssk;int len; + ssk =mptcp_sched_get_send (msk); +if (!ssk) + goto out; + +lock_sock (ssk); +while ((dfrag =mptcp_send_head (sk))) { info.sent = dfrag->already_sent; info.limit = dfrag->data_len;@@ -1546,24 +1544,6 @@ void __mptcp_push_pending(struct sock *sk, unsigned int flags)while (len >0 ) {int ret =0 ; - prev_ssk = ssk; - ssk =mptcp_sched_get_send (msk); - -/* First check. If the ssk has changed since -* the last round, release prev_ssk -*/ -if (ssk != prev_ssk && prev_ssk) -mptcp_push_release (prev_ssk, &info); -if (!ssk) - goto out; - -/* Need to lock the new subflow only if different -* from the previous one, otherwise we are still -* helding the relevant lock -*/ -if (ssk != prev_ssk) -lock_sock (ssk); - ret =mptcp_sendmsg_frag (sk, ssk, dfrag, &info);if (ret <=0 ) {if (ret == -EAGAIN )@@ -1579,11 +1559,21 @@ void __mptcp_push_pending(struct sock *sk, unsigned int flags)mptcp_update_post_push (msk, dfrag, ret); }WRITE_ONCE (msk->first_pending,mptcp_send_next (sk)); + +if (msk->snd_burst >0 && +sk_stream_memory_free (ssk) && +mptcp_subflow_active (mptcp_subflow_ctx (ssk))) { +mptcp_set_timeout (sk); + }else { +break ; + } } -/* at this point we held the socket lock for the last subflow we used */ -if (ssk) -mptcp_push_release (ssk, &info); +if (do_check_data_fin) { +tcp_push (ssk,0 , info.mss_now,tcp_sk (ssk)->nonagle, + info.size_goal); + } +release_sock (ssk);out:
Quoted 6 lines
/* ensure the rtx timer is running */Signature · 1 line
2.35.3
--
Mat Martineau
Intel
PATCH v2 2/4 mptcp: add do_push_pending helper
In reply to Geliang TangThis patch moves the duplicate code from __mptcp_push_pending() and __mptcp_subflow_push_pending() into a new helper function, named __do_push_pending(). And simplify __mptcp_push_pending() by invoking this helper.
--- net/mptcp/protocol.c | 62 ++++++++++++++++++++++++-------------------- 1 file changed, 34 insertions(+), 28 deletions(-)
net/mptcp/protocol.c +34 -28
@@ -1483,12 +1483,6 @@ struct sock *mptcp_subflow_get_send(struct mptcp_sock *msk)return ssk; } -static void mptcp_push_release (struct sock *ssk,struct mptcp_sendmsg_info *info) -{ -tcp_push (ssk,0 , info->mss_now,tcp_sk (ssk)->nonagle, info->size_goal); -release_sock (ssk); -} -static void mptcp_update_post_push (struct mptcp_sock *msk,struct mptcp_data_frag *dfrag,u32 sent)
@@ -1520,40 +1514,29 @@ void mptcp_check_and_set_pending(struct sock *sk)mptcp_sk (sk)->push_pending |=BIT (MPTCP_PUSH_PENDING ); } -void __mptcp_push_pending (struct sock *sk,unsigned int flags) +static int __do_push_pending (struct sock *sk,struct sock *ssk, +struct mptcp_sendmsg_info *info) {struct mptcp_sock *msk =mptcp_sk (sk); -struct mptcp_sendmsg_info info = { - .flags = flags, - }; -bool do_check_data_fin = false;struct mptcp_data_frag *dfrag; -struct sock *ssk; -int len; - - ssk =mptcp_sched_get_send (msk); -if (!ssk) - goto out; - -lock_sock (ssk); +int len, copied =0 ;while ((dfrag =mptcp_send_head (sk))) { - info.sent = dfrag->already_sent; - info.limit = dfrag->data_len; + info->sent = dfrag->already_sent; + info->limit = dfrag->data_len; len = dfrag->data_len - dfrag->already_sent;while (len >0 ) {int ret =0 ; - ret =mptcp_sendmsg_frag (sk, ssk, dfrag, &info); + ret =mptcp_sendmsg_frag (sk, ssk, dfrag, info);if (ret <=0 ) {if (ret == -EAGAIN )continue ; -mptcp_push_release (ssk, &info); goto out; } - do_check_data_fin = true; - info.sent += ret; + info->sent += ret; + copied += ret; len -= ret;mptcp_update_post_push (msk, dfrag, ret);
@@ -1569,10 +1552,33 @@ void __mptcp_push_pending(struct sock *sk, unsigned int flags) } } -if (do_check_data_fin) { -tcp_push (ssk,0 , info.mss_now,tcp_sk (ssk)->nonagle, - info.size_goal); +out: +if (copied) { +tcp_push (ssk,0 , info->mss_now,tcp_sk (ssk)->nonagle, + info->size_goal); } + +return copied; +} + +void __mptcp_push_pending (struct sock *sk,unsigned int flags) +{ +struct mptcp_sock *msk =mptcp_sk (sk); +struct mptcp_sendmsg_info info = { + .flags = flags, + }; +bool do_check_data_fin = false; +struct sock *ssk; + + ssk =mptcp_sched_get_send (msk); +if (!ssk) + goto out; + +if (!mptcp_send_head (sk)) + goto out; + +lock_sock (ssk); + do_check_data_fin =__do_push_pending (sk, ssk, &info);release_sock (ssk); out:
Signature · 1 line
2.35.3
Re: [PATCH mptcp-next v2 2/4] mptcp: add do_push_pending helper
Quoted 112 lines · Geliang Tang
This patch moves the duplicate code from
__mptcp_push_pending()and__mptcp_subflow_push_pending()into a new helper function, named__do_push_pending(). And simplify__mptcp_push_pending()by invoking this helper.Signed-off-by Geliang Tang <[email protected]>---
net/mptcp/protocol.c| 62 ++++++++++++++++++++++++--------------------1 file changed, 34 insertions(+), 28 deletions(-)net/mptcp/protocol.c +34
-28@@ -1483,12 +1483,6 @@ struct sock *mptcp_subflow_get_send(struct mptcp_sock *msk)return ssk; } -static void mptcp_push_release (struct sock *ssk,struct mptcp_sendmsg_info *info) -{ -tcp_push (ssk,0 , info->mss_now,tcp_sk (ssk)->nonagle, info->size_goal); -release_sock (ssk); -} -static void mptcp_update_post_push (struct mptcp_sock *msk,struct mptcp_data_frag *dfrag,u32 sent)@@ -1520,40 +1514,29 @@ void mptcp_check_and_set_pending(struct sock *sk)mptcp_sk (sk)->push_pending |=BIT (MPTCP_PUSH_PENDING ); } -void __mptcp_push_pending (struct sock *sk,unsigned int flags) +static int __do_push_pending (struct sock *sk,struct sock *ssk, +struct mptcp_sendmsg_info *info) {struct mptcp_sock *msk =mptcp_sk (sk); -struct mptcp_sendmsg_info info = { - .flags = flags, - }; -bool do_check_data_fin = false;struct mptcp_data_frag *dfrag; -struct sock *ssk; -int len; - - ssk =mptcp_sched_get_send (msk); -if (!ssk) - goto out; - -lock_sock (ssk); +int len, copied =0 ;while ((dfrag =mptcp_send_head (sk))) { - info.sent = dfrag->already_sent; - info.limit = dfrag->data_len; + info->sent = dfrag->already_sent; + info->limit = dfrag->data_len; len = dfrag->data_len - dfrag->already_sent;while (len >0 ) {int ret =0 ; - ret =mptcp_sendmsg_frag (sk, ssk, dfrag, &info); + ret =mptcp_sendmsg_frag (sk, ssk, dfrag, info);if (ret <=0 ) {if (ret == -EAGAIN )continue ; -mptcp_push_release (ssk, &info); goto out; } - do_check_data_fin = true; - info.sent += ret; + info->sent += ret; + copied += ret; len -= ret;mptcp_update_post_push (msk, dfrag, ret);@@ -1569,10 +1552,33 @@ void __mptcp_push_pending(struct sock *sk, unsigned int flags) } } -if (do_check_data_fin) { -tcp_push (ssk,0 , info.mss_now,tcp_sk (ssk)->nonagle, - info.size_goal); +out: +if (copied) { +tcp_push (ssk,0 , info->mss_now,tcp_sk (ssk)->nonagle, + info->size_goal); } + +return copied; +} + +void __mptcp_push_pending (struct sock *sk,unsigned int flags) +{ +struct mptcp_sock *msk =mptcp_sk (sk); +struct mptcp_sendmsg_info info = { + .flags = flags, + }; +bool do_check_data_fin = false; +struct sock *ssk; + + ssk =mptcp_sched_get_send (msk); +if (!ssk) + goto out; + +if (!mptcp_send_head (sk)) + goto out; + +lock_sock (ssk); + do_check_data_fin =__do_push_pending (sk, ssk, &info);release_sock (ssk);
Quoted 1 line
out:
I think this is getting close to what I was talking about, but does not have the loop that will call the scheduler again. I tried this code and got slightly better performance in simult_flows.sh:
void __mptcp_push_pending(struct sock *sk, unsigned int flags)
{
struct mptcp_sock *msk = mptcp_sk(sk);
struct mptcp_sendmsg_info info = {
.flags = flags,
};
bool do_check_data_fin = false;
struct sock *ssk;
while (mptcp_send_head(sk) && (ssk = mptcp_sched_get_send(msk))) {
lock_sock(ssk);
do_check_data_fin = __do_push_pending(sk, ssk, &info);
release_sock(ssk);
}
/* ensure the rtx timer is running */
if (!mptcp_timer_pending(sk))
mptcp_reset_timer(sk);
if (do_check_data_fin)
__mptcp_check_send_data_fin(sk);
}--
Mat Martineau
Intel
PATCH v2 3/4 mptcp: update __mptcp_subflow_push_pending
In reply to Geliang TangMove the packet scheduler out of the dfrags loop, invoke it only once in __mptcp_subflow_push_pending().
--- net/mptcp/protocol.c | 44 ++++++++++++++++++++------------------------ 1 file changed, 20 insertions(+), 24 deletions(-)
net/mptcp/protocol.c +20 -24
@@ -1598,7 +1598,15 @@ static void __mptcp_subflow_push_pending(struct sock *sk, struct sock *ssk)struct mptcp_data_frag *dfrag;struct sock *xmit_ssk;int len, copied =0 ; -bool first = true; + + xmit_ssk =mptcp_sched_get_send (mptcp_sk (sk)); +if (!xmit_ssk) + goto out; +if (xmit_ssk != ssk) { +mptcp_subflow_delegate (mptcp_subflow_ctx (xmit_ssk), +MPTCP_DELEGATE_SEND ); + goto out; + } info.flags =0 ;while ((dfrag =mptcp_send_head (sk))) {
@@ -1608,19 +1616,6 @@ static void __mptcp_subflow_push_pending(struct sock *sk, struct sock *ssk)while (len >0 ) {int ret =0 ; -/* the caller already invoked the packet scheduler, -* check for a different subflow usage only after -* spooling the first chunk of data -*/ - xmit_ssk = first ? ssk :mptcp_sched_get_send (mptcp_sk (sk)); -if (!xmit_ssk) - goto out; -if (xmit_ssk != ssk) { -mptcp_subflow_delegate (mptcp_subflow_ctx (xmit_ssk), -MPTCP_DELEGATE_SEND ); - goto out; - } - ret =mptcp_sendmsg_frag (sk, ssk, dfrag, &info);if (ret <=0 ) goto out;
@@ -1628,11 +1623,18 @@ static void __mptcp_subflow_push_pending(struct sock *sk, struct sock *ssk) info.sent += ret; copied += ret; len -= ret; - first = false;mptcp_update_post_push (msk, dfrag, ret); }WRITE_ONCE (msk->first_pending,mptcp_send_next (sk)); + +if (msk->snd_burst >0 && +sk_stream_memory_free (ssk) && +mptcp_subflow_active (mptcp_subflow_ctx (ssk))) { +mptcp_set_timeout (sk); + }else { +break ; + } } out:
@@ -3191,16 +3193,10 @@ void __mptcp_check_push(struct sock *sk, struct sock *ssk)if (!mptcp_send_head (sk))return ; -if (!sock_owned_by_user (sk)) { -struct sock *xmit_ssk =mptcp_sched_get_send (mptcp_sk (sk)); - -if (xmit_ssk == ssk) -__mptcp_subflow_push_pending (sk, ssk); -else if (xmit_ssk) -mptcp_subflow_delegate (mptcp_subflow_ctx (xmit_ssk),MPTCP_DELEGATE_SEND ); - }else { +if (!sock_owned_by_user (sk)) +__mptcp_subflow_push_pending (sk, ssk); +else __set_bit (MPTCP_PUSH_PENDING , &mptcp_sk (sk)->cb_flags); - } }#define MPTCP_FLAGS_PROCESS_CTX_NEED (BIT(MPTCP_PUSH_PENDING) | \
Signature · 1 line
2.35.3
PATCH v2 4/4 mptcp: simplify __mptcp_subflow_push_pending
In reply to Geliang TangThis patch simplifies __mptcp_subflow_push_pending() by invoking __do_push_pending() helper.
--- net/mptcp/protocol.c | 33 ++------------------------------- 1 file changed, 2 insertions(+), 31 deletions(-)
net/mptcp/protocol.c +2 -31
@@ -1595,9 +1595,8 @@ static void __mptcp_subflow_push_pending(struct sock *sk, struct sock *ssk)struct mptcp_sendmsg_info info = { .data_lock_held = true, }; -struct mptcp_data_frag *dfrag;struct sock *xmit_ssk; -int len, copied =0 ; +int copied =0 ; xmit_ssk =mptcp_sched_get_send (mptcp_sk (sk));if (!xmit_ssk)
@@ -1609,41 +1608,13 @@ static void __mptcp_subflow_push_pending(struct sock *sk, struct sock *ssk) } info.flags =0 ; -while ((dfrag =mptcp_send_head (sk))) { - info.sent = dfrag->already_sent; - info.limit = dfrag->data_len; - len = dfrag->data_len - dfrag->already_sent; -while (len >0 ) { -int ret =0 ; - - ret =mptcp_sendmsg_frag (sk, ssk, dfrag, &info); -if (ret <=0 ) - goto out; - - info.sent += ret; - copied += ret; - len -= ret; - -mptcp_update_post_push (msk, dfrag, ret); - } -WRITE_ONCE (msk->first_pending,mptcp_send_next (sk)); - -if (msk->snd_burst >0 && -sk_stream_memory_free (ssk) && -mptcp_subflow_active (mptcp_subflow_ctx (ssk))) { -mptcp_set_timeout (sk); - }else { -break ; - } - } + copied =__do_push_pending (sk, ssk, &info); out:/* __mptcp_alloc_tx_skb could have released some wmem and we are * not going to flush it via release_sock() */ if (copied) { -tcp_push (ssk,0 , info.mss_now,tcp_sk (ssk)->nonagle, - info.size_goal);if (!mptcp_timer_pending (sk))mptcp_reset_timer (sk);
Signature · 1 line
2.35.3
Re: mptcp: simplify __mptcp_subflow_push_pending: Tests Results
Hi Geliang,
Thank you for your modifications, that's great!
Our CI did some validations and here is its report:
- KVM Validation: normal:
- Success! ✅:
- Task: https://cirrus-ci.com/task/5897416976105472
- Summary: https://api.cirrus-ci.com/v1/artifact/task/5897416976105472/summary/summary.txt
- KVM Validation: debug:
- Success! ✅:
- Task: https://cirrus-ci.com/task/5334467022684160
- Summary: https://api.cirrus-ci.com/v1/artifact/task/5334467022684160/summary/summary.txt
Initiator: Patchew Applier
Commits: https://github.com/multipath-tcp/mptcp_net-next/commits/95535e05c593
If there are some issues, you can reproduce them using the same environment as the one used by the CI thanks to a docker image, e.g.:
$ cd [kernel source code]
$ docker run -v "${PWD}:${PWD}:rw" -w "${PWD}" --privileged --rm -it \
--pull always mptcp/mptcp-upstream-virtme-docker:latest \
auto-debugFor more details:
https://github.com/multipath-tcp/mptcp-upstream-virtme-docker
Please note that despite all the efforts that have been already done to have a stable tests suite when executed on a public CI like here, it is possible some reported issues are not due to your modifications. Still, do not hesitate to help us improve that ;-)
Cheers,
MPTCP GH Action bot
Bot operated by Matthieu Baerts (Tessares)