refactor push pending
Subsystems (5)
v8 cover refactor push pending
v8:
- merge "register default scheduler" into this series.
- Move "multi-subflow scheduling" out of this series.
- This series should be merged between the commit "mptcp: add get_subflow wrappers" and "bpf: Add bpf_mptcp_sched_ops", except the squash-to patches.
v7:
- update delegate sending in patch 10
- rebase to export/20221011T063543.
v6:
- drop all msk->last_snd, add last_snd variable instead in patch 13
- fix lock_sock issue in patch 3
- merge squash-to patches
v5:
- address Mat's comments in v4.
v4:
- update __mptcp_subflow_push_pending as Mat suggested.
- add more patches from "BPF redundant scheduler" series.
v3:
- add a cleanup patch.
- remove msk->last_snd in mptcp_subflow_get_send().
- add the loop that calls the scheduler again in __mptcp_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 (15):
Revert "mptcp: add get_subflow wrappers" Squash to "mptcp: add struct mptcp_sched_ops"
mptcp: use msk instead of mptcp_sk(sk)
mptcp: change 'first' as a parameter
mptcp: drop snd_burst in mptcp_sock
mptcp: refactor push_pending logic
mptcp: drop last_snd in mptcp_sock
mptcp: add sched_data_set_contexts helper
mptcp: register default scheduler
mptcp: add scheduler wrappers
Squash to "bpf: Add bpf_mptcp_sched_ops" Squash to "bpf: Add bpf_mptcp_sched_kfunc_set" Squash to "selftests/bpf: Add bpf_first scheduler" Squash to "selftests/bpf: Add bpf_bkup scheduler" Squash to "selftests/bpf: Add bpf_rr scheduler"
include/net/mptcp.h | 7 +- net/mptcp/bpf.c | 1 + net/mptcp/pm.c | 9 +- net/mptcp/pm_netlink.c | 3 - net/mptcp/protocol.c | 208 ++++++++---------- net/mptcp/protocol.h | 15 +- net/mptcp/sched.c | 117 ++++++---- tools/testing/selftests/bpf/bpf_tcp_helpers.h | 10 +- .../selftests/bpf/progs/mptcp_bpf_bkup.c | 10 +- .../selftests/bpf/progs/mptcp_bpf_first.c | 10 +- .../selftests/bpf/progs/mptcp_bpf_rr.c | 16 +- 11 files changed, 212 insertions(+), 194 deletions(-)
Signature · 1 line
2.35.3
PATCH v8 1/15 Revert "mptcp: add get_subflow wrappers"
This reverts commit 517ab13fd55a51e77c1dd39d6023fe568aa250a5.
--- net/mptcp/protocol.c | 26 ++++++++++---- net/mptcp/protocol.h | 4 --- net/mptcp/sched.c | 85 -------------------------------------------- 3 files changed, 20 insertions(+), 95 deletions(-)
net/mptcp/protocol.c +20 -6
@@ -1406,7 +1406,7 @@ bool mptcp_subflow_active(struct mptcp_subflow_context *subflow)* returns the subflow that will transmit the next DSS * additionally updates the rtx timeout */ -struct sock *mptcp_subflow_get_send (struct mptcp_sock *msk) +static struct sock *mptcp_subflow_get_send (struct mptcp_sock *msk) {struct subflow_send_info send_info[SSK_MODE_MAX ];struct mptcp_subflow_context *subflow;
@@ -1417,6 +1417,15 @@ struct sock *mptcp_subflow_get_send(struct mptcp_sock *msk)u64 linger_time;long tout =0 ; +sock_owned_by_me (sk); + +if (__mptcp_check_fallback (msk)) { +if (!msk->first) +return NULL ; +return __tcp_can_send (msk->first) && +sk_stream_memory_free (msk->first) ? msk->first :NULL ; + } +/* re-use last subflow, if the burst allow that */ if (msk->last_snd && msk->snd_burst > 0 && sk_stream_memory_free(msk->last_snd) &&
@@ -1547,7 +1556,7 @@ void __mptcp_push_pending(struct sock *sk, unsigned int flags)int ret =0 ; prev_ssk = ssk; - ssk =mptcp_sched_get_send (msk); + ssk =mptcp_subflow_get_send (msk);/* First check. If the ssk has changed since * the last round, release prev_ssk
@@ -1616,7 +1625,7 @@ static void __mptcp_subflow_push_pending(struct sock *sk, struct sock *ssk)* 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)); + xmit_ssk = first ? ssk :mptcp_subflow_get_send (mptcp_sk (sk));if (!xmit_ssk) goto out;if (xmit_ssk != ssk) {
@@ -2193,12 +2202,17 @@ static void mptcp_timeout_timer(struct timer_list *t)* * A backup subflow is returned only if that is the only kind available. */ -struct sock *mptcp_subflow_get_retrans (struct mptcp_sock *msk) +static struct sock *mptcp_subflow_get_retrans (struct mptcp_sock *msk) {struct sock *backup =NULL , *pick =NULL ;struct mptcp_subflow_context *subflow;int min_stale_count =INT_MAX ; +sock_owned_by_me ((const struct sock *)msk); + +if (__mptcp_check_fallback (msk)) +return NULL ; +mptcp_for_each_subflow (msk, subflow) {struct sock *ssk =mptcp_subflow_tcp_sock (subflow);
@@ -2481,7 +2495,7 @@ static void __mptcp_retrans(struct sock *sk)mptcp_clean_una_wakeup (sk);/* first check ssk: need to kick "stale" logic */ - ssk =mptcp_sched_get_retrans (msk); + ssk =mptcp_subflow_get_retrans (msk); dfrag =mptcp_rtx_head (sk);if (!dfrag) {if (mptcp_data_fin_enabled (msk)) {
@@ -3199,7 +3213,7 @@ void __mptcp_check_push(struct sock *sk, struct sock *ssk)return ;if (!sock_owned_by_user (sk)) { -struct sock *xmit_ssk =mptcp_sched_get_send (mptcp_sk (sk)); +struct sock *xmit_ssk =mptcp_subflow_get_send (mptcp_sk (sk));if (xmit_ssk == ssk)__mptcp_subflow_push_pending (sk, ssk);
net/mptcp/protocol.h +0 -4
@@ -641,10 +641,6 @@ int mptcp_init_sched(struct mptcp_sock *msk,void mptcp_release_sched (struct mptcp_sock *msk);void mptcp_subflow_set_scheduled (struct mptcp_subflow_context *subflow,bool scheduled); -struct sock *mptcp_subflow_get_send (struct mptcp_sock *msk); -struct sock *mptcp_subflow_get_retrans (struct mptcp_sock *msk); -struct sock *mptcp_sched_get_send (struct mptcp_sock *msk); -struct sock *mptcp_sched_get_retrans (struct mptcp_sock *msk);static inline bool __tcp_can_send (const struct sock *ssk) {
net/mptcp/sched.c +0 -85
88 lines not on this page · raw message
Signature · 1 line
2.35.3
PATCH v8 2/15 Squash to "mptcp: add struct mptcp_sched_ops"
In reply to Geliang TangNew api:
- add new data_init
- add snd_burst in mptcp_sched_data
- add an int return value for get_subflow
--- include/net/mptcp.h | 7 +++++-- net/mptcp/sched.c | 2 +- 2 files changed, 6 insertions(+), 3 deletions(-)
include/net/mptcp.h +5 -2
@@ -101,12 +101,15 @@ struct mptcp_out_options {struct mptcp_sched_data {bool reinject; +int snd_burst;struct mptcp_subflow_context *contexts[MPTCP_SUBFLOWS_MAX ]; };struct mptcp_sched_ops { -void (*get_subflow)(const struct mptcp_sock *msk, -struct mptcp_sched_data *data); +void (*data_init)(const struct mptcp_sock *msk, +struct mptcp_sched_data *data); +int (*get_subflow)(const struct mptcp_sock *msk, +struct mptcp_sched_data *data);char name[MPTCP_SCHED_NAME_MAX ];struct module *owner;
net/mptcp/sched.c +1 -1
@@ -33,7 +33,7 @@ struct mptcp_sched_ops *mptcp_sched_find(const char *name)int mptcp_register_scheduler (struct mptcp_sched_ops *sched) { -if (!sched->get_subflow) +if (!sched->data_init || !sched->get_subflow)return -EINVAL ;spin_lock (&mptcp_sched_list_lock);
Signature · 1 line
2.35.3
PATCH v8 3/15 mptcp: use msk instead of mptcp_sk(sk)
In reply to Geliang TangUse msk instead of mptcp_sk(sk) in the functions where the variable "msk = mptcp_sk(sk)" has been defined.
--- net/mptcp/protocol.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-)
net/mptcp/protocol.c +5 -5
@@ -1625,7 +1625,7 @@ static void __mptcp_subflow_push_pending(struct sock *sk, struct sock *ssk)* check for a different subflow usage only after * spooling the first chunk of data */ - xmit_ssk = first ? ssk :mptcp_subflow_get_send (mptcp_sk (sk)); + xmit_ssk = first ? ssk :mptcp_subflow_get_send (msk);if (!xmit_ssk) goto out;if (xmit_ssk != ssk) {
@@ -2256,7 +2256,7 @@ bool __mptcp_retransmit_pending_data(struct sock *sk)struct mptcp_data_frag *cur, *rtx_head;struct mptcp_sock *msk =mptcp_sk (sk); -if (__mptcp_check_fallback (mptcp_sk (sk))) +if (__mptcp_check_fallback (msk))return false;if (tcp_rtx_and_write_queues_empty (sk))
@@ -2938,7 +2938,7 @@ bool __mptcp_close(struct sock *sk, long timeout)sock_hold (sk);pr_debug ("msk=%p state=%d" , sk, sk->sk_state); -if (mptcp_sk (sk)->token) +if (msk->token)mptcp_event (MPTCP_EVENT_CLOSED , msk,NULL ,GFP_KERNEL );if (sk->sk_state ==TCP_CLOSE ) {
@@ -2997,8 +2997,8 @@ static int mptcp_disconnect(struct sock *sk, int flags)mptcp_stop_timer (sk);sk_stop_timer (sk, &sk->sk_timer); -if (mptcp_sk (sk)->token) -mptcp_event (MPTCP_EVENT_CLOSED ,mptcp_sk (sk),NULL ,GFP_KERNEL ); +if (msk->token) +mptcp_event (MPTCP_EVENT_CLOSED , msk,NULL ,GFP_KERNEL );/* msk->subflow is still intact, the following will not free the first * subflow
Signature · 1 line
2.35.3
PATCH v8 4/15 mptcp: change 'first' as a parameter
In reply to Geliang TangThe function mptcp_subflow_process_delegated() uses the input ssk first, while __mptcp_check_push() invokes the packet scheduler first.
So this patch adds a new parameter named 'first' for the function __mptcp_subflow_push_pending() to deal with these two cases separately.
With this change, the code that invokes the packet scheduler in the function __mptcp_check_push() can be removed, and replaced by invoking __mptcp_subflow_push_pending() directly.
--- net/mptcp/protocol.c | 20 ++++++-------------- 1 file changed, 6 insertions(+), 14 deletions(-)
net/mptcp/protocol.c +6 -14
@@ -1602,7 +1602,7 @@ void __mptcp_push_pending(struct sock *sk, unsigned int flags)__mptcp_check_send_data_fin (sk); } -static void __mptcp_subflow_push_pending (struct sock *sk,struct sock *ssk) +static void __mptcp_subflow_push_pending (struct sock *sk,struct sock *ssk,bool first) {struct mptcp_sock *msk =mptcp_sk (sk);struct mptcp_sendmsg_info info = {
@@ -1611,7 +1611,6 @@ 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; info.flags =0 ;while ((dfrag =mptcp_send_head (sk))) {
@@ -1621,8 +1620,7 @@ 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 +/* check for a different subflow usage only after * spooling the first chunk of data */ xmit_ssk = first ? ssk :mptcp_subflow_get_send (msk);
@@ -3212,16 +3210,10 @@ 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_subflow_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, false); +else __set_bit (MPTCP_PUSH_PENDING , &mptcp_sk (sk)->cb_flags); - } }#define MPTCP_FLAGS_PROCESS_CTX_NEED (BIT(MPTCP_PUSH_PENDING) | \
@@ -3312,7 +3304,7 @@ void mptcp_subflow_process_delegated(struct sock *ssk)if (test_bit (MPTCP_DELEGATE_SEND , &subflow->delegated_status)) {mptcp_data_lock (sk);if (!sock_owned_by_user (sk)) -__mptcp_subflow_push_pending (sk, ssk); +__mptcp_subflow_push_pending (sk, ssk, true);else __set_bit (MPTCP_PUSH_PENDING , &mptcp_sk (sk)->cb_flags);mptcp_data_unlock (sk);
Signature · 1 line
2.35.3
PATCH v8 5/15 mptcp: drop snd_burst in mptcp_sock
In reply to Geliang TangDrop snd_burst of struct mptcp_sock, use data->snd_burst instead of msk->snd_burst.
--- net/mptcp/protocol.c | 18 ++++++++++-------- net/mptcp/protocol.h | 1 - 2 files changed, 10 insertions(+), 9 deletions(-)
net/mptcp/protocol.c +10 -8
@@ -1406,7 +1406,8 @@ bool mptcp_subflow_active(struct mptcp_subflow_context *subflow)* returns the subflow that will transmit the next DSS * additionally updates the rtx timeout */ -static struct sock *mptcp_subflow_get_send (struct mptcp_sock *msk) +static struct sock *mptcp_subflow_get_send (struct mptcp_sock *msk, +struct mptcp_sched_data *data) {struct subflow_send_info send_info[SSK_MODE_MAX ];struct mptcp_subflow_context *subflow;
@@ -1427,7 +1428,7 @@ static struct sock *mptcp_subflow_get_send(struct mptcp_sock *msk) }/* re-use last subflow, if the burst allow that */ -if (msk->last_snd && msk->snd_burst >0 && +if (msk->last_snd && data->snd_burst >0 &&sk_stream_memory_free (msk->last_snd) &&mptcp_subflow_active (mptcp_subflow_ctx (msk->last_snd))) {mptcp_set_timeout (sk);
@@ -1496,7 +1497,7 @@ static struct sock *mptcp_subflow_get_send(struct mptcp_sock *msk)READ_ONCE (ssk->sk_pacing_rate) * burst, burst + wmem); msk->last_snd = ssk; - msk->snd_burst = burst; + data->snd_burst = burst;return ssk; }
@@ -1514,8 +1515,6 @@ static void mptcp_update_post_push(struct mptcp_sock *msk, dfrag->already_sent += sent; - msk->snd_burst -= sent; - snd_nxt_new += dfrag->already_sent;/* snd_nxt_new can be smaller than snd_nxt in case mptcp
@@ -1541,6 +1540,7 @@ 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_sched_data data = {0 };struct mptcp_sendmsg_info info = { .flags = flags, };
@@ -1556,7 +1556,7 @@ void __mptcp_push_pending(struct sock *sk, unsigned int flags)int ret =0 ; prev_ssk = ssk; - ssk =mptcp_subflow_get_send (msk); + ssk =mptcp_subflow_get_send (msk, &data);/* First check. If the ssk has changed since * the last round, release prev_ssk
@@ -1584,6 +1584,7 @@ void __mptcp_push_pending(struct sock *sk, unsigned int flags) do_check_data_fin = true; info.sent += ret; len -= ret; + data.snd_burst -= ret;mptcp_update_post_push (msk, dfrag, ret); }
@@ -1605,6 +1606,7 @@ void __mptcp_push_pending(struct sock *sk, unsigned int flags)static void __mptcp_subflow_push_pending (struct sock *sk,struct sock *ssk,bool first) {struct mptcp_sock *msk =mptcp_sk (sk); +struct mptcp_sched_data data = {0 };struct mptcp_sendmsg_info info = { .data_lock_held = true, };
@@ -1623,7 +1625,7 @@ static void __mptcp_subflow_push_pending(struct sock *sk, struct sock *ssk, bool/* check for a different subflow usage only after * spooling the first chunk of data */ - xmit_ssk = first ? ssk :mptcp_subflow_get_send (msk); + xmit_ssk = first ? ssk :mptcp_subflow_get_send (msk, &data);if (!xmit_ssk) goto out;if (xmit_ssk != ssk) {
@@ -1639,6 +1641,7 @@ static void __mptcp_subflow_push_pending(struct sock *sk, struct sock *ssk, bool info.sent += ret; copied += ret; len -= ret; + data.snd_burst -= ret; first = false;mptcp_update_post_push (msk, dfrag, ret);
@@ -2277,7 +2280,6 @@ bool __mptcp_retransmit_pending_data(struct sock *sk)mptcp_data_unlock (sk); msk->first_pending = rtx_head; - msk->snd_burst =0 ;/* be sure to clear the "sent status" on all re-injected fragments */ list_for_each_entry (cur, &msk->rtx_queue, list) {
net/mptcp/protocol.h +0 -1
@@ -259,7 +259,6 @@ struct mptcp_sock {u64 rcv_data_fin_seq;int rmem_fwd_alloc;struct sock *last_snd; -int snd_burst;int old_wspace;u64 recovery_snd_nxt;/* in recovery mode accept up to this seq; * recovery related fields are under data_lock
Signature · 1 line
2.35.3
PATCH v8 6/15 mptcp: refactor push_pending logic
In reply to Geliang TangTo support redundant package schedulers more easily, this patch refactors __mptcp_push_pending() logic from:
For each dfrag:
While sends succeed:
Call the scheduler (selects subflow and msk->snd_burst) Update subflow locks (push/release/acquire as needed) Send the dfrag data with mptcp_sendmsg_frag() Update already_sent, snd_nxt, snd_burst
Update msk->first_pending
Push/release on final subflow
->
While the scheduler selects one subflow:
Lock the subflow
For each pending dfrag:
While sends succeed:
Send the dfrag data with mptcp_sendmsg_frag() Update already_sent, snd_nxt, snd_burst
Update msk->first_pending
Break if required by msk->snd_burst / etc
Push and release the subflow
Refactors __mptcp_subflow_push_pending logic from:
For each dfrag:
While sends succeed: Call the scheduler (selects subflow and msk->snd_burst) Send the dfrag data with mptcp_subflow_delegate(), break Send the dfrag data with mptcp_sendmsg_frag() Update dfrag->already_sent, msk->snd_nxt, msk->snd_burst Update msk->first_pending
->
While first_pending isn't empty:
Call the scheduler (selects subflow and msk->snd_burst) Send the dfrag data with mptcp_subflow_delegate(), break Send the dfrag data with mptcp_sendmsg_frag()
For each pending dfrag:
While sends succeed:
Send the dfrag data with mptcp_sendmsg_frag() Update already_sent, snd_nxt, snd_burst
Update msk->first_pending
Break if required by msk->snd_burst / etc
Move the duplicate code from __mptcp_push_pending() and __mptcp_subflow_push_pending() into a new helper function, named __subflow_push_pending(). Simplify __mptcp_push_pending() and __mptcp_subflow_push_pending() by invoking this helper.
Also move the burst check conditions out of the function mptcp_subflow_get_send(), check them in __mptcp_push_pending() and __mptcp_subflow_push_pending() in the inner "for each pending dfrag" loop.
--- net/mptcp/protocol.c | 160 +++++++++++++++++++------------------------ 1 file changed, 72 insertions(+), 88 deletions(-)
net/mptcp/protocol.c +72 -88
@@ -1427,14 +1427,6 @@ static struct sock *mptcp_subflow_get_send(struct mptcp_sock *msk,sk_stream_memory_free (msk->first) ? msk->first :NULL ; } -/* re-use last subflow, if the burst allow that */ -if (msk->last_snd && data->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 ;
@@ -1501,12 +1493,6 @@ static 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)
@@ -1536,70 +1522,83 @@ 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 __subflow_push_pending (struct sock *sk,struct sock *ssk, +struct mptcp_sendmsg_info *info, +struct mptcp_sched_data *data) { -struct sock *prev_ssk =NULL , *ssk =NULL ;struct mptcp_sock *msk =mptcp_sk (sk); -struct mptcp_sched_data data = {0 }; -struct mptcp_sendmsg_info info = { - .flags = flags, - }; -bool do_check_data_fin = false;struct mptcp_data_frag *dfrag; -int len; +int len, copied =0 , err =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 ; - prev_ssk = ssk; - ssk =mptcp_subflow_get_send (msk, &data); - -/* 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); + ret =mptcp_sendmsg_frag (sk, ssk, dfrag, info);if (ret <=0 ) { -if (ret == -EAGAIN ) -continue ; -mptcp_push_release (ssk, &info); + err = copied ? : ret; goto out; } - do_check_data_fin = true; - info.sent += ret; + info->sent += ret; + copied += ret; len -= ret; - data.snd_burst -= ret; + data->snd_burst -= ret;mptcp_update_post_push (msk, dfrag, ret); }WRITE_ONCE (msk->first_pending,mptcp_send_next (sk)); + +if (data->snd_burst <=0 || + !sk_stream_memory_free (ssk) || + !mptcp_subflow_active (mptcp_subflow_ctx (ssk))) { + err = copied ? : -EAGAIN ; + goto out; + } +mptcp_set_timeout (sk); + } + err = copied; + +out: +if (copied) { +tcp_push (ssk,0 , info->mss_now,tcp_sk (ssk)->nonagle, + info->size_goal); } -/* at this point we held the socket lock for the last subflow we used */ -if (ssk) -mptcp_push_release (ssk, &info); +return err; +} + +void __mptcp_push_pending (struct sock *sk,unsigned int flags) +{ +struct mptcp_sock *msk =mptcp_sk (sk); +struct mptcp_sched_data data = {0 }; +struct mptcp_sendmsg_info info = { + .flags = flags, + }; +struct sock *ssk; +int ret =0 ; + +again: +while (mptcp_send_head (sk) && (ssk =mptcp_subflow_get_send (msk, &data))) { +lock_sock (ssk); + ret =__subflow_push_pending (sk, ssk, &info, &data); +release_sock (ssk); + +if (ret <=0 ) { +if (ret == -EAGAIN ) + goto again; + goto out; + } + } out:/* ensure the rtx timer is running */ if (!mptcp_timer_pending (sk))mptcp_reset_timer (sk); -if (do_check_data_fin) +if (ret >0 )__mptcp_check_send_data_fin (sk); }
@@ -1610,52 +1609,37 @@ static void __mptcp_subflow_push_pending(struct sock *sk, struct sock *ssk, boolstruct mptcp_sendmsg_info info = { .data_lock_held = true, }; -struct mptcp_data_frag *dfrag;struct sock *xmit_ssk; -int len, copied =0 ; +int ret =0 ; 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 ; - -/* check for a different subflow usage only after -* spooling the first chunk of data -*/ - xmit_ssk = first ? ssk :mptcp_subflow_get_send (msk, &data); -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; - - info.sent += ret; - copied += ret; - len -= ret; - data.snd_burst -= ret; - first = false; +again: +while (mptcp_send_head (sk)) { +/* check for a different subflow usage only after +* spooling the first chunk of data +*/ + xmit_ssk = first ? ssk :mptcp_subflow_get_send (msk, &data); +if (!xmit_ssk) + goto out; +if (xmit_ssk != ssk) { +mptcp_subflow_delegate (mptcp_subflow_ctx (xmit_ssk), +MPTCP_DELEGATE_SEND ); + goto out; + } -mptcp_update_post_push (msk, dfrag, ret); + ret =__subflow_push_pending (sk, ssk, &info, &data); +if (ret <=0 ) { +if (ret == -EAGAIN ) + goto again; +break ; } -WRITE_ONCE (msk->first_pending,mptcp_send_next (sk)); } 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 (ret >0 ) {if (!mptcp_timer_pending (sk))mptcp_reset_timer (sk);
Signature · 1 line
2.35.3
PATCH v8 7/15 mptcp: drop last_snd in mptcp_sock
In reply to Geliang Tangmsk->last_snd is no longer used anymore, drop it as well as the macro MPTCP_RESET_SCHEDULER.
--- net/mptcp/pm.c | 9 +-------- net/mptcp/pm_netlink.c | 3 --- net/mptcp/protocol.c | 11 +---------- net/mptcp/protocol.h | 2 -- 4 files changed, 2 insertions(+), 23 deletions(-)
net/mptcp/pm.c +1 -8
@@ -282,15 +282,8 @@ void mptcp_pm_mp_prio_received(struct sock *ssk, u8 bkup)pr_debug ("subflow->backup=%d, bkup=%d\n" , subflow->backup, bkup); msk =mptcp_sk (sk); -if (subflow->backup != bkup) { +if (subflow->backup != bkup) subflow->backup = bkup; -mptcp_data_lock (sk); -if (!sock_owned_by_user (sk)) - msk->last_snd =NULL ; -else -__set_bit (MPTCP_RESET_SCHEDULER , &msk->cb_flags); -mptcp_data_unlock (sk); - }mptcp_event (MPTCP_EVENT_SUB_PRIORITY , msk, ssk,GFP_ATOMIC ); }
net/mptcp/pm_netlink.c +0 -3
@@ -475,9 +475,6 @@ static void __mptcp_pm_send_ack(struct mptcp_sock *msk, struct mptcp_subflow_con slow =lock_sock_fast (ssk);if (prio) { -if (subflow->backup != backup) - msk->last_snd =NULL ; - subflow->send_mp_prio =1 ; subflow->backup = backup; subflow->request_bkup = backup;
net/mptcp/protocol.c +1 -10
@@ -1479,16 +1479,13 @@ static struct sock *mptcp_subflow_get_send(struct mptcp_sock *msk, burst =min_t (int,MPTCP_SEND_BURST_SIZE ,mptcp_wnd_end (msk) - msk->snd_nxt); wmem =READ_ONCE (ssk->sk_wmem_queued); -if (!burst) { - msk->last_snd =NULL ; +if (!burst)return ssk; - } subflow =mptcp_subflow_ctx (ssk); subflow->avg_pacing_rate =div_u64 ((u64 )subflow->avg_pacing_rate * wmem +READ_ONCE (ssk->sk_pacing_rate) * burst, burst + wmem); - msk->last_snd = ssk; data->snd_burst = burst;return ssk; }
@@ -2353,9 +2350,6 @@ static void __mptcp_close_ssk(struct sock *sk, struct sock *ssk, msk->first =NULL ; out: -if (ssk == msk->last_snd) - msk->last_snd =NULL ; -if (need_push)__mptcp_push_pending (sk,0 ); }
@@ -2988,7 +2982,6 @@ static int mptcp_disconnect(struct sock *sk, int flags)* subflow */ mptcp_destroy_common (msk,MPTCP_CF_FASTCLOSE ); - msk->last_snd =NULL ;WRITE_ONCE (msk->flags,0 ); msk->cb_flags =0 ; msk->push_pending =0 ;
@@ -3249,8 +3242,6 @@ static void mptcp_release_cb(struct sock *sk)__mptcp_set_connected (sk);if (__test_and_clear_bit (MPTCP_ERROR_REPORT , &msk->cb_flags))__mptcp_error_report (sk); -if (__test_and_clear_bit (MPTCP_RESET_SCHEDULER , &msk->cb_flags)) - msk->last_snd =NULL ; }__mptcp_update_rmem (sk);
net/mptcp/protocol.h +0 -2
@@ -124,7 +124,6 @@#define MPTCP_RETRANSMIT 4#define MPTCP_FLUSH_JOIN_LIST 5#define MPTCP_CONNECTED 6 -#define MPTCP_RESET_SCHEDULER 7static inline bool before64 (__u64 seq1,__u64 seq2) {
@@ -258,7 +257,6 @@ struct mptcp_sock {atomic64_t rcv_wnd_sent;u64 rcv_data_fin_seq;int rmem_fwd_alloc; -struct sock *last_snd;int old_wspace;u64 recovery_snd_nxt;/* in recovery mode accept up to this seq; * recovery related fields are under data_lock
Signature · 1 line
2.35.3
PATCH v8 8/15 mptcp: add sched_data_set_contexts helper
In reply to Geliang TangAdd a new helper mptcp_sched_data_set_contexts(), it will be used in the BPF contexts.
--- net/mptcp/sched.c | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+)
net/mptcp/sched.c +19 -0
@@ -93,3 +93,22 @@ void mptcp_subflow_set_scheduled(struct mptcp_subflow_context *subflow, {WRITE_ONCE (subflow->scheduled, scheduled); } + +void mptcp_sched_data_set_contexts (const struct mptcp_sock *msk, +struct mptcp_sched_data *data) +{ +struct mptcp_subflow_context *subflow; +int i =0 ; + +mptcp_for_each_subflow (msk, subflow) { +if (i ==MPTCP_SUBFLOWS_MAX ) { +pr_warn_once ("too many subflows" ); +break ; + } +mptcp_subflow_set_scheduled (subflow, false); + data->contexts[i++] = subflow; + } + +for (; i <MPTCP_SUBFLOWS_MAX ; i++) + data->contexts[i] =NULL ; +}
Signature · 1 line
2.35.3
PATCH v8 9/15 mptcp: register default scheduler
In reply to Geliang TangThis patch defines the default packet scheduler mptcp_sched_default. Register it in mptcp_sched_init(), which is invoked in mptcp_proto_init(). Skip deleting this default scheduler in mptcp_unregister_scheduler().
Set msk->sched to the default scheduler when the input parameter of mptcp_init_sched() is NULL.
--- net/mptcp/protocol.c | 7 ++++--- net/mptcp/protocol.h | 4 ++++ net/mptcp/sched.c | 38 ++++++++++++++++++++++++++++++++++++-- 3 files changed, 44 insertions(+), 5 deletions(-)
net/mptcp/protocol.c +4 -3
@@ -1406,8 +1406,8 @@ bool mptcp_subflow_active(struct mptcp_subflow_context *subflow)* returns the subflow that will transmit the next DSS * additionally updates the rtx timeout */ -static struct sock *mptcp_subflow_get_send (struct mptcp_sock *msk, -struct mptcp_sched_data *data) +struct sock *mptcp_subflow_get_send (const struct mptcp_sock *msk, +struct mptcp_sched_data *data) {struct subflow_send_info send_info[SSK_MODE_MAX ];struct mptcp_subflow_context *subflow;
@@ -2184,7 +2184,7 @@ static void mptcp_timeout_timer(struct timer_list *t)* * A backup subflow is returned only if that is the only kind available. */ -static struct sock *mptcp_subflow_get_retrans (struct mptcp_sock *msk) +struct sock *mptcp_subflow_get_retrans (const struct mptcp_sock *msk) {struct sock *backup =NULL , *pick =NULL ;struct mptcp_subflow_context *subflow;
@@ -3851,6 +3851,7 @@ void __init mptcp_proto_init(void)mptcp_subflow_init ();mptcp_pm_init (); +mptcp_sched_init ();mptcp_token_init ();if (proto_register (&mptcp_prot,MPTCP_USE_SLAB ) !=0 )
net/mptcp/protocol.h +4 -0
@@ -633,11 +633,15 @@ void mptcp_info2sockaddr(const struct mptcp_addr_info *info,struct mptcp_sched_ops *mptcp_sched_find (const char *name);int mptcp_register_scheduler (struct mptcp_sched_ops *sched);void mptcp_unregister_scheduler (struct mptcp_sched_ops *sched); +void mptcp_sched_init (void );int mptcp_init_sched (struct mptcp_sock *msk,struct mptcp_sched_ops *sched);void mptcp_release_sched (struct mptcp_sock *msk);void mptcp_subflow_set_scheduled (struct mptcp_subflow_context *subflow,bool scheduled); +struct sock *mptcp_subflow_get_send (const struct mptcp_sock *msk, +struct mptcp_sched_data *data); +struct sock *mptcp_subflow_get_retrans (const struct mptcp_sock *msk);static inline bool __tcp_can_send (const struct sock *ssk) {
net/mptcp/sched.c +36 -2
@@ -16,6 +16,33 @@static DEFINE_SPINLOCK (mptcp_sched_list_lock);static LIST_HEAD (mptcp_sched_list); +static void mptcp_sched_default_data_init (const struct mptcp_sock *msk, +struct mptcp_sched_data *data) +{ + data->snd_burst =0 ; +} + +static int mptcp_sched_default_get_subflow (const struct mptcp_sock *msk, +struct mptcp_sched_data *data) +{ +struct sock *ssk; + + ssk = data->reinject ?mptcp_subflow_get_retrans (msk) : +mptcp_subflow_get_send (msk, data); +if (!ssk) +return -EINVAL ; + +mptcp_subflow_set_scheduled (mptcp_subflow_ctx (ssk), true); +return 0 ; +} + +static struct mptcp_sched_ops mptcp_sched_default = { + .data_init = mptcp_sched_default_data_init, + .get_subflow = mptcp_sched_default_get_subflow, + .name ="default" , + .owner =THIS_MODULE , +}; +/* Must be called with rcu read lock held */ struct mptcp_sched_ops *mptcp_sched_find (const char *name) {
@@ -50,16 +77,24 @@ int mptcp_register_scheduler(struct mptcp_sched_ops *sched)void mptcp_unregister_scheduler (struct mptcp_sched_ops *sched) { +if (sched == &mptcp_sched_default) +return ; +spin_lock (&mptcp_sched_list_lock);list_del_rcu (&sched->list);spin_unlock (&mptcp_sched_list_lock); } +void mptcp_sched_init (void ) +{ +mptcp_register_scheduler (&mptcp_sched_default); +} +int mptcp_init_sched (struct mptcp_sock *msk,struct mptcp_sched_ops *sched) {if (!sched) - goto out; + sched = &mptcp_sched_default;if (!bpf_try_module_get (sched, sched->owner))return -EBUSY ;
@@ -70,7 +105,6 @@ int mptcp_init_sched(struct mptcp_sock *msk,pr_debug ("sched=%s" , msk->sched->name); -out:return 0 ; }
Signature · 1 line
2.35.3
PATCH v8 10/15 mptcp: add scheduler wrappers
In reply to Geliang TangThis patch defines two packet scheduler wrappers mptcp_sched_get_send() and mptcp_sched_get_retrans(), invoke data_init() and get_subflow() of msk->sched in them.
Set data->reinject to true in mptcp_sched_get_retrans(), set it false in mptcp_sched_get_send().
If msk->sched is NULL, use default functions mptcp_subflow_get_send() and mptcp_subflow_get_retrans() to send data.
Move sock_owned_by_me() check and fallback check into the wrappers from mptcp_subflow_get_send() and mptcp_subflow_get_retrans().
--- net/mptcp/protocol.c | 14 ----------- net/mptcp/protocol.h | 4 ++++ net/mptcp/sched.c | 55 ++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 59 insertions(+), 14 deletions(-)
net/mptcp/protocol.c +0 -14
@@ -1418,15 +1418,6 @@ struct sock *mptcp_subflow_get_send(const struct mptcp_sock *msk,u64 linger_time;long tout =0 ; -sock_owned_by_me (sk); - -if (__mptcp_check_fallback (msk)) { -if (!msk->first) -return NULL ; -return __tcp_can_send (msk->first) && -sk_stream_memory_free (msk->first) ? msk->first :NULL ; - } -/* pick the subflow with the lower wmem/wspace ratio */ for (i =0 ; i <SSK_MODE_MAX ; ++i) { send_info[i].ssk =NULL ;
@@ -2190,11 +2181,6 @@ struct sock *mptcp_subflow_get_retrans(const struct mptcp_sock *msk)struct mptcp_subflow_context *subflow;int min_stale_count =INT_MAX ; -sock_owned_by_me ((const struct sock *)msk); - -if (__mptcp_check_fallback (msk)) -return NULL ; -mptcp_for_each_subflow (msk, subflow) {struct sock *ssk =mptcp_subflow_tcp_sock (subflow);
net/mptcp/protocol.h +4 -0
@@ -642,6 +642,10 @@ void mptcp_subflow_set_scheduled(struct mptcp_subflow_context *subflow,struct sock *mptcp_subflow_get_send (const struct mptcp_sock *msk,struct mptcp_sched_data *data);struct sock *mptcp_subflow_get_retrans (const struct mptcp_sock *msk); +int mptcp_sched_get_send (struct mptcp_sock *msk, +struct mptcp_sched_data *data); +int mptcp_sched_get_retrans (struct mptcp_sock *msk, +struct mptcp_sched_data *data);static inline bool __tcp_can_send (const struct sock *ssk) {
net/mptcp/sched.c +55 -0
58 lines not on this page · raw message
Signature · 1 line
2.35.3
PATCH v8 11/15 Squash to "bpf: Add bpf_mptcp_sched_ops"
In reply to Geliang TangUse new API.
--- tools/testing/selftests/bpf/bpf_tcp_helpers.h | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-)
tools/testing/selftests/bpf/bpf_tcp_helpers.h +7 -2
@@ -240,6 +240,7 @@ struct mptcp_subflow_context {struct mptcp_sched_data {bool reinject; +int snd_burst;struct mptcp_subflow_context *contexts[MPTCP_SUBFLOWS_MAX ]; } __attribute__((preserve_access_index));
@@ -249,8 +250,10 @@ struct mptcp_sched_ops {void (*init)(const struct mptcp_sock *msk);void (*release)(const struct mptcp_sock *msk); -void (*get_subflow)(const struct mptcp_sock *msk, -struct mptcp_sched_data *data); +void (*data_init)(const struct mptcp_sock *msk, +struct mptcp_sched_data *data); +int (*get_subflow)(const struct mptcp_sock *msk, +struct mptcp_sched_data *data);void *owner; };
@@ -265,5 +268,7 @@ struct mptcp_sock { extern void mptcp_subflow_set_scheduled(struct mptcp_subflow_context *subflow, bool scheduled) __ksym; +extern void mptcp_sched_data_set_contexts(const struct mptcp_sock *msk, + struct mptcp_sched_data *data) __ksym;#endif
Signature · 1 line
2.35.3
PATCH v8 12/15 Squash to "bpf: Add bpf_mptcp_sched_kfunc_set"
In reply to Geliang TangAdd mptcp_sched_data_set_contexts in kfunc_set.
--- net/mptcp/bpf.c | 1 + 1 file changed, 1 insertion(+)
net/mptcp/bpf.c +1 -0
@@ -164,6 +164,7 @@ struct bpf_struct_ops bpf_mptcp_sched_ops = {BTF_SET8_START (bpf_mptcp_sched_kfunc_ids)BTF_ID_FLAGS (func ,mptcp_subflow_set_scheduled ) +BTF_ID_FLAGS (func, mptcp_sched_data_set_contexts)BTF_SET8_END (bpf_mptcp_sched_kfunc_ids) static const struct btf_kfunc_id_set bpf_mptcp_sched_kfunc_set = {
Signature · 1 line
2.35.3
PATCH v8 13/15 Squash to "selftests/bpf: Add bpf_first scheduler"
In reply to Geliang TangUse new API.
--- tools/testing/selftests/bpf/progs/mptcp_bpf_first.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-)
tools/testing/selftests/bpf/progs/mptcp_bpf_first.c +9 -1
@@ -16,16 +16,24 @@ void BPF_PROG(mptcp_sched_first_release, const struct mptcp_sock *msk) { } -void BPF_STRUCT_OPS (bpf_first_get_subflow ,const struct mptcp_sock *msk, +void BPF_STRUCT_OPS (bpf_first_data_init ,const struct mptcp_sock *msk,struct mptcp_sched_data *data) +{ +mptcp_sched_data_set_contexts (msk, data); +} + +int BPF_STRUCT_OPS (bpf_first_get_subflow ,const struct mptcp_sock *msk, +struct mptcp_sched_data *data) {mptcp_subflow_set_scheduled (data->contexts[0 ], true); +return 0 ; } SEC(".struct_ops")struct mptcp_sched_ops first = { .init = (void *)mptcp_sched_first_init, .release = (void *)mptcp_sched_first_release, + .data_init = (void *)bpf_first_data_init, .get_subflow = (void *)bpf_first_get_subflow, .name ="bpf_first" , };
Signature · 1 line
2.35.3
PATCH v8 14/15 Squash to "selftests/bpf: Add bpf_bkup scheduler"
In reply to Geliang TangUse new API.
--- tools/testing/selftests/bpf/progs/mptcp_bpf_bkup.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-)
tools/testing/selftests/bpf/progs/mptcp_bpf_bkup.c +9 -1
@@ -16,8 +16,14 @@ void BPF_PROG(mptcp_sched_bkup_release, const struct mptcp_sock *msk) { } -void BPF_STRUCT_OPS (bpf_bkup_get_subflow ,const struct mptcp_sock *msk, +void BPF_STRUCT_OPS (bpf_bkup_data_init ,const struct mptcp_sock *msk,struct mptcp_sched_data *data) +{ +mptcp_sched_data_set_contexts (msk, data); +} + +int BPF_STRUCT_OPS (bpf_bkup_get_subflow ,const struct mptcp_sock *msk, +struct mptcp_sched_data *data) {int nr =0 ;
@@ -32,12 +38,14 @@ void BPF_STRUCT_OPS(bpf_bkup_get_subflow, const struct mptcp_sock *msk, }mptcp_subflow_set_scheduled (data->contexts[nr], true); +return 0 ; } SEC(".struct_ops")struct mptcp_sched_ops bkup = { .init = (void *)mptcp_sched_bkup_init, .release = (void *)mptcp_sched_bkup_release, + .data_init = (void *)bpf_bkup_data_init, .get_subflow = (void *)bpf_bkup_get_subflow, .name ="bpf_bkup" , };
Signature · 1 line
2.35.3
PATCH v8 15/15 Squash to "selftests/bpf: Add bpf_rr scheduler"
In reply to Geliang TangUse last_snd instead of msk->last_snd, then last_snd of struct mptcp_sock could be removed.
Use new API.
--- tools/testing/selftests/bpf/bpf_tcp_helpers.h | 1 - tools/testing/selftests/bpf/progs/mptcp_bpf_rr.c | 16 +++++++++++++--- 2 files changed, 13 insertions(+), 4 deletions(-)
tools/testing/selftests/bpf/bpf_tcp_helpers.h +0 -1
@@ -260,7 +260,6 @@ struct mptcp_sched_ops {struct mptcp_sock {struct inet_connection_sock sk; -struct sock *last_snd;__u32 token;struct sock *first;char ca_name[TCP_CA_NAME_MAX ];
tools/testing/selftests/bpf/progs/mptcp_bpf_rr.c +13 -3
@@ -5,6 +5,7 @@#include "bpf_tcp_helpers.h" char _license[] SEC("license") = "GPL"; +struct sock *last_snd =NULL ; SEC("struct_ops/mptcp_sched_rr_init")void BPF_PROG (mptcp_sched_rr_init ,const struct mptcp_sock *msk)
@@ -16,16 +17,22 @@ void BPF_PROG(mptcp_sched_rr_release, const struct mptcp_sock *msk) { } -void BPF_STRUCT_OPS (bpf_rr_get_subflow ,const struct mptcp_sock *msk, +void BPF_STRUCT_OPS (bpf_rr_data_init ,const struct mptcp_sock *msk,struct mptcp_sched_data *data) +{ +mptcp_sched_data_set_contexts (msk, data); +} + +int BPF_STRUCT_OPS (bpf_rr_get_subflow ,const struct mptcp_sock *msk, +struct mptcp_sched_data *data) {int nr =0 ;for (int i =0 ; i <MPTCP_SUBFLOWS_MAX ; i++) { -if (!msk->last_snd || !data->contexts[i]) +if (!last_snd || !data->contexts[i])break ; -if (data->contexts[i]->tcp_sock == msk->last_snd) { +if (data->contexts[i]->tcp_sock == last_snd) {if (i +1 ==MPTCP_SUBFLOWS_MAX || !data->contexts[i +1 ])break ;
@@ -35,12 +42,15 @@ void BPF_STRUCT_OPS(bpf_rr_get_subflow, const struct mptcp_sock *msk, }mptcp_subflow_set_scheduled (data->contexts[nr], true); + last_snd = data->contexts[nr]->tcp_sock; +return 0 ; } SEC(".struct_ops")struct mptcp_sched_ops rr = { .init = (void *)mptcp_sched_rr_init, .release = (void *)mptcp_sched_rr_release, + .data_init = (void *)bpf_rr_data_init, .get_subflow = (void *)bpf_rr_get_subflow, .name ="bpf_rr" , };
Signature · 1 line
2.35.3
Re: Squash to "selftests/bpf: Add bpf_rr scheduler": Build Failure
Hi Geliang,
Thank you for your modifications, that's great!
But sadly, our CI spotted some issues with it when trying to build it.
You can find more details there:
https://patchwork.kernel.org/project/mptcp/patch/a03e67a73655bf701cd26be3dc58480372fdf8f3.1665753926.git.geliang.tang@suse.com/
https://github.com/multipath-tcp/mptcp_net-next/actions/runs/3250457897
Status: failure
Initiator: MPTCPimporter
Commits: https://github.com/multipath-tcp/mptcp_net-next/commits/7d412455802e
Feel free to reply to this email if you cannot access logs, if you need some support to fix the error, if this doesn't seem to be caused by your modifications or if the error is a false positive one.
Cheers,
MPTCP GH Action bot
Bot operated by Matthieu Baerts (Tessares)
Re: Squash to "selftests/bpf: Add bpf_rr scheduler": Tests Results
In reply to Geliang TangHi Geliang,
Thank you for your modifications, that's great!
Our CI did some validations and here is its report:
- KVM Validation: normal:
- Unstable: 1 failed test(s): packetdrill_add_addr 🔴:
- Task: https://cirrus-ci.com/task/5124182068428800
- Summary: https://api.cirrus-ci.com/v1/artifact/task/5124182068428800/summary/summary.txt
- {"code":404,"message":
- "Can't find artifacts containing file conclusion.txt"}:
- Task: https://cirrus-ci.com/task/6250081975271424
- Summary: https://api.cirrus-ci.com/v1/artifact/task/6250081975271424/summary/summary.txt
Initiator: Patchew Applier
Commits: https://github.com/multipath-tcp/mptcp_net-next/commits/7d412455802e
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)
Re: [PATCH mptcp-next v8 02/15] Squash to "mptcp: add struct mptcp_sched_ops"
In reply to Geliang TangQuoted 16 lines · Geliang Tang
New api:
- add newdata_init
- addsnd_burstinmptcp_sched_data
- add an int return value forget_subflowSigned-off-by Geliang Tang <[email protected]>---
include/net/mptcp.h| 7 +++++--net/mptcp/sched.c| 2 +-2 files changed, 6 insertions(+), 3 deletions(-)include/net/mptcp.h +0
-0
include/net/mptcp.h struct mptcp_out_options { · line 104
Quoted 4 lines
include/net/mptcp.h +1
-0@@ -101,12 +101,15 @@ struct mptcp_out_options {struct mptcp_sched_data {bool reinject; +int snd_burst;
I suggest moving this addition to "mptcp: drop snd_burst in mptcp_sock", since it isn't used until then.
Quoted 31 lines
include/net/mptcp.h +4
-2struct mptcp_subflow_context *contexts[MPTCP_SUBFLOWS_MAX ]; };struct mptcp_sched_ops { -void (*get_subflow)(const struct mptcp_sock *msk, -struct mptcp_sched_data *data); +void (*data_init)(const struct mptcp_sock *msk, +struct mptcp_sched_data *data); +int (*get_subflow)(const struct mptcp_sock *msk, +struct mptcp_sched_data *data);char name[MPTCP_SCHED_NAME_MAX ];struct module *owner;net/mptcp/sched.c +1
-1@@ -33,7 +33,7 @@ struct mptcp_sched_ops *mptcp_sched_find(const char *name)int mptcp_register_scheduler (struct mptcp_sched_ops *sched) { -if (!sched->get_subflow) +if (!sched->data_init || !sched->get_subflow)return -EINVAL ;spin_lock (&mptcp_sched_list_lock);Signature · 1 line
2.35.3
--
Mat Martineau
Intel
Re: [PATCH mptcp-next v8 05/15] mptcp: drop snd_burst in mptcp_sock
In reply to Geliang TangQuoted 51 lines · Geliang Tang
Drop
snd_burstof structmptcp_sock, usedata->snd_burstinstead ofmsk->snd_burst.Signed-off-by Geliang Tang <[email protected]>---
net/mptcp/protocol.c| 18 ++++++++++--------net/mptcp/protocol.h| 1 -2 files changed, 10 insertions(+), 9 deletions(-)net/mptcp/protocol.c +4
-5@@ -1406,7 +1406,8 @@ bool mptcp_subflow_active(struct mptcp_subflow_context *subflow)* returns the subflow that will transmit the next DSS * additionally updates the rtx timeout */ -static struct sock *mptcp_subflow_get_send (struct mptcp_sock *msk) +static struct sock *mptcp_subflow_get_send (struct mptcp_sock *msk, +struct mptcp_sched_data *data) {struct subflow_send_info send_info[SSK_MODE_MAX ];struct mptcp_subflow_context *subflow;@@ -1427,7 +1428,7 @@ static struct sock *mptcp_subflow_get_send(struct mptcp_sock *msk) }/* re-use last subflow, if the burst allow that */ -if (msk->last_snd && msk->snd_burst >0 && +if (msk->last_snd && data->snd_burst >0 &&sk_stream_memory_free (msk->last_snd) &&mptcp_subflow_active (mptcp_subflow_ctx (msk->last_snd))) {mptcp_set_timeout (sk);@@ -1496,7 +1497,7 @@ static struct sock *mptcp_subflow_get_send(struct mptcp_sock *msk)READ_ONCE (ssk->sk_pacing_rate) * burst, burst + wmem); msk->last_snd = ssk; - msk->snd_burst = burst; + data->snd_burst = burst;return ssk; }@@ -1514,8 +1515,6 @@ static void mptcp_update_post_push(struct mptcp_sock *msk, dfrag->already_sent += sent; - msk->snd_burst -= sent; - snd_nxt_new += dfrag->already_sent;/* snd_nxt_new can be smaller than snd_nxt in case mptcp
net/mptcp/protocol.c void __mptcp_push_pending(struct sock *sk, unsigned int flags) · line 1543
Quoted 4 lines
net/mptcp/protocol.c +1
-0@@ -1541,6 +1540,7 @@ 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_sched_data data = {0 };
I think moving snd_burst to struct mptcp_sched_data will make the default scheduler behave differently. The old location (in msk) would use the burst length even across multiple pushes. Clearing snd_burst more frequently can cause the scheduler to switch subflows more often when multiple subflows are available.
The patch after this (removing last_snd) also contributes to that behavior.
Paolo, do you think it's important to use a "burst" across multiple mptcp_push_pending() calls, or is it mostly relevant within a single push?
- Mat
Quoted 70 lines
net/mptcp/protocol.c +5
-3struct mptcp_sendmsg_info info = { .flags = flags, };@@ -1556,7 +1556,7 @@ void __mptcp_push_pending(struct sock *sk, unsigned int flags)int ret =0 ; prev_ssk = ssk; - ssk =mptcp_subflow_get_send (msk); + ssk =mptcp_subflow_get_send (msk, &data);/* First check. If the ssk has changed since * the last round, release prev_ssk @@ -1584,6 +1584,7 @@ void __mptcp_push_pending(struct sock *sk, unsigned int flags) do_check_data_fin = true; info.sent += ret; len -= ret; + data.snd_burst -= ret;mptcp_update_post_push (msk, dfrag, ret); }@@ -1605,6 +1606,7 @@ void __mptcp_push_pending(struct sock *sk, unsigned int flags)static void __mptcp_subflow_push_pending (struct sock *sk,struct sock *ssk,bool first) {struct mptcp_sock *msk =mptcp_sk (sk); +struct mptcp_sched_data data = {0 };struct mptcp_sendmsg_info info = { .data_lock_held = true, };@@ -1623,7 +1625,7 @@ static void __mptcp_subflow_push_pending(struct sock *sk, struct sock *ssk, bool/* check for a different subflow usage only after * spooling the first chunk of data */ - xmit_ssk = first ? ssk :mptcp_subflow_get_send (msk); + xmit_ssk = first ? ssk :mptcp_subflow_get_send (msk, &data);if (!xmit_ssk) goto out;if (xmit_ssk != ssk) {@@ -1639,6 +1641,7 @@ static void __mptcp_subflow_push_pending(struct sock *sk, struct sock *ssk, bool info.sent += ret; copied += ret; len -= ret; + data.snd_burst -= ret; first = false;mptcp_update_post_push (msk, dfrag, ret);@@ -2277,7 +2280,6 @@ bool __mptcp_retransmit_pending_data(struct sock *sk)mptcp_data_unlock (sk); msk->first_pending = rtx_head; - msk->snd_burst =0 ;/* be sure to clear the "sent status" on all re-injected fragments */ list_for_each_entry (cur, &msk->rtx_queue, list) {net/mptcp/protocol.h +0
-1@@ -259,7 +259,6 @@ struct mptcp_sock {u64 rcv_data_fin_seq;int rmem_fwd_alloc;struct sock *last_snd; -int snd_burst;int old_wspace;u64 recovery_snd_nxt;/* in recovery mode accept up to this seq; * recovery related fields are under data_lock Signature · 1 line
2.35.3
--
Mat Martineau
Intel
Re: [PATCH mptcp-next v8 06/15] mptcp: refactor push_pending logic
In reply to Geliang TangQuoted 261 lines · Geliang Tang
To support redundant package schedulers more easily, this patch refactors
__mptcp_push_pending()logic from:For each dfrag:
While sends succeed:
Call the scheduler (selects subflow andmsk->snd_burst) Update subflow locks (push/release/acquire as needed) Send the dfrag data withmptcp_sendmsg_frag()Updatealready_sent,snd_nxt,snd_burst
Updatemsk->first_pending
Push/release on final subflow->
While the scheduler selects one subflow:
Lock the subflow
For each pending dfrag:
While sends succeed:
Send the dfrag data withmptcp_sendmsg_frag()Updatealready_sent,snd_nxt,snd_burst
Updatemsk->first_pending
Break if required bymsk->snd_burst/ etc
Push and release the subflowRefactors
__mptcp_subflow_push_pendinglogic from:For each dfrag:
While sends succeed: Call the scheduler (selects subflow and msk->snd_burst) Send the dfrag data with mptcp_subflow_delegate(), break Send the dfrag data with mptcp_sendmsg_frag() Update dfrag->already_sent, msk->snd_nxt, msk->snd_burst Update msk->first_pending->
While
first_pendingisn't empty:
Call the scheduler (selects subflow andmsk->snd_burst) Send the dfrag data withmptcp_subflow_delegate(), break Send the dfrag data withmptcp_sendmsg_frag()
For each pending dfrag:
While sends succeed:
Send the dfrag data withmptcp_sendmsg_frag()Updatealready_sent,snd_nxt,snd_burst
Updatemsk->first_pending
Break if required bymsk->snd_burst/ etcMove the duplicate code from
__mptcp_push_pending()and__mptcp_subflow_push_pending()into a new helper function, named__subflow_push_pending(). Simplify__mptcp_push_pending()and__mptcp_subflow_push_pending()by invoking this helper.Also move the burst check conditions out of the function
mptcp_subflow_get_send(), check them in__mptcp_push_pending()and__mptcp_subflow_push_pending()in the inner "for each pending dfrag" loop.Signed-off-by Geliang Tang <[email protected]>---
net/mptcp/protocol.c| 160 +++++++++++++++++++------------------------1 file changed, 72 insertions(+), 88 deletions(-)net/mptcp/protocol.c +51
-52@@ -1427,14 +1427,6 @@ static struct sock *mptcp_subflow_get_send(struct mptcp_sock *msk,sk_stream_memory_free (msk->first) ? msk->first :NULL ; } -/* re-use last subflow, if the burst allow that */ -if (msk->last_snd && data->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 ;@@ -1501,12 +1493,6 @@ static 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)@@ -1536,70 +1522,83 @@ 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 __subflow_push_pending (struct sock *sk,struct sock *ssk, +struct mptcp_sendmsg_info *info, +struct mptcp_sched_data *data) { -struct sock *prev_ssk =NULL , *ssk =NULL ;struct mptcp_sock *msk =mptcp_sk (sk); -struct mptcp_sched_data data = {0 }; -struct mptcp_sendmsg_info info = { - .flags = flags, - }; -bool do_check_data_fin = false;struct mptcp_data_frag *dfrag; -int len; +int len, copied =0 , err =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 ; - prev_ssk = ssk; - ssk =mptcp_subflow_get_send (msk, &data); - -/* 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); + ret =mptcp_sendmsg_frag (sk, ssk, dfrag, info);if (ret <=0 ) { -if (ret == -EAGAIN ) -continue ; -mptcp_push_release (ssk, &info); + err = copied ? : ret; goto out; } - do_check_data_fin = true; - info.sent += ret; + info->sent += ret; + copied += ret; len -= ret; - data.snd_burst -= ret; + data->snd_burst -= ret;mptcp_update_post_push (msk, dfrag, ret); }WRITE_ONCE (msk->first_pending,mptcp_send_next (sk)); + +if (data->snd_burst <=0 || + !sk_stream_memory_free (ssk) || + !mptcp_subflow_active (mptcp_subflow_ctx (ssk))) { + err = copied ? : -EAGAIN ; + goto out; + } +mptcp_set_timeout (sk); + } + err = copied; + +out: +if (copied) { +tcp_push (ssk,0 , info->mss_now,tcp_sk (ssk)->nonagle, + info->size_goal); } -/* at this point we held the socket lock for the last subflow we used */ -if (ssk) -mptcp_push_release (ssk, &info); +return err; +} + +void __mptcp_push_pending (struct sock *sk,unsigned int flags) +{ +struct mptcp_sock *msk =mptcp_sk (sk); +struct mptcp_sched_data data = {0 }; +struct mptcp_sendmsg_info info = { + .flags = flags, + }; +struct sock *ssk; +int ret =0 ; + +again: +while (mptcp_send_head (sk) && (ssk =mptcp_subflow_get_send (msk, &data))) { +lock_sock (ssk); + ret =__subflow_push_pending (sk, ssk, &info, &data); +release_sock (ssk); + +if (ret <=0 ) { +if (ret == -EAGAIN ) + goto again; + goto out; + } + }out:
/* ensure the rtx timer is running */ if (!mptcp_timer_pending(sk)) mptcp_reset_timer(sk);net/mptcp/protocol.c +1
-1-if (do_check_data_fin) +if (ret >0 )__mptcp_check_send_data_fin(sk); }net/mptcp/protocol.c +7
-30@@ -1610,52 +1609,37 @@ static void __mptcp_subflow_push_pending(struct sock *sk, struct sock *ssk, boolstruct mptcp_sendmsg_info info = { .data_lock_held = true, }; -struct mptcp_data_frag *dfrag;struct sock *xmit_ssk; -int len, copied =0 ; +int ret =0 ; 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 ; - -/* check for a different subflow usage only after -* spooling the first chunk of data -*/ - xmit_ssk = first ? ssk :mptcp_subflow_get_send (msk, &data); -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; - - info.sent += ret; - copied += ret; - len -= ret; - data.snd_burst -= ret; - first = false; +again: +while (mptcp_send_head (sk)) { +/* check for a different subflow usage only after +* spooling the first chunk of data +*/ + xmit_ssk = first ? ssk :mptcp_subflow_get_send (msk, &data);
net/mptcp/protocol.c static void __mptcp_subflow_push_pending(struct sock *sk, struct sock *ssk, bool · line 1630
Quoted 10 lines
net/mptcp/protocol.c +8
-1+if (!xmit_ssk) + goto out; +if (xmit_ssk != ssk) { +mptcp_subflow_delegate (mptcp_subflow_ctx (xmit_ssk), +MPTCP_DELEGATE_SEND ); + goto out; + } -mptcp_update_post_push (msk, dfrag, ret); + ret =__subflow_push_pending (sk, ssk, &info, &data);
I think there should still be a "first = false;" before the loop runs again so that the scheduler will get called on the next iteration. This will help avoid a tight loop if EAGAIN is returned on the line above.
- Mat
Quoted 16 lines
net/mptcp/protocol.c +4
-1+if (ret <=0 ) { +if (ret == -EAGAIN ) + goto again; +break ; } -WRITE_ONCE (msk->first_pending,mptcp_send_next (sk)); }out:
/* __mptcp_alloc_tx_skb could have released some wmem and we are * not going to flush it via release_sock() */net/mptcp/protocol.c +1
-3-if (copied) { -tcp_push (ssk,0 , info.mss_now,tcp_sk (ssk)->nonagle, - info.size_goal); +if (ret >0 ) {
Quoted 8 lines
if (!mptcp_timer_pending(sk)) mptcp_reset_timer(sk);Signature · 1 line
2.35.3
--
Mat Martineau
Intel
Re: [PATCH mptcp-next v8 09/15] mptcp: register default scheduler
In reply to Geliang TangQuoted 8 lines · Geliang Tang
This patch defines the default packet scheduler
mptcp_sched_default. Register it inmptcp_sched_init(), which is invoked inmptcp_proto_init(). Skip deleting this default scheduler inmptcp_unregister_scheduler().Set
msk->schedto the default scheduler when the input parameter ofmptcp_init_sched()is NULL.Signed-off-by Geliang Tang <[email protected]>
Hi Geliang -
Is there a benefit to registering the default scheduler? It adds an indirect call for the common case of using the default scheduler.
The scheduler wrappers can make direct calls to the default scheduler if msk->sched is NULL. It looks like this is partially handled already in "mptcp: add scheduler wrappers", so it doesn't seem necessary to handle it both ways. Some changes to setting the "default" scheduler and handling that name as a special case would still be needed.
- Mat
Quoted 135 lines
---
net/mptcp/protocol.c | 7 ++++--- net/mptcp/protocol.h | 4 ++++ net/mptcp/sched.c | 38 ++++++++++++++++++++++++++++++++++++-- 3 files changed, 44 insertions(+), 5 deletions(-)net/mptcp/protocol.c +4
-3@@ -1406,8 +1406,8 @@ bool mptcp_subflow_active(struct mptcp_subflow_context *subflow)* returns the subflow that will transmit the next DSS * additionally updates the rtx timeout */ -static struct sock *mptcp_subflow_get_send (struct mptcp_sock *msk, -struct mptcp_sched_data *data) +struct sock *mptcp_subflow_get_send (const struct mptcp_sock *msk, +struct mptcp_sched_data *data) {struct subflow_send_info send_info[SSK_MODE_MAX ];struct mptcp_subflow_context *subflow;@@ -2184,7 +2184,7 @@ static void mptcp_timeout_timer(struct timer_list *t)* * A backup subflow is returned only if that is the only kind available. */ -static struct sock *mptcp_subflow_get_retrans (struct mptcp_sock *msk) +struct sock *mptcp_subflow_get_retrans (const struct mptcp_sock *msk) {struct sock *backup =NULL , *pick =NULL ;struct mptcp_subflow_context *subflow;@@ -3851,6 +3851,7 @@ void __init mptcp_proto_init(void)mptcp_subflow_init ();mptcp_pm_init (); +mptcp_sched_init ();mptcp_token_init ();if (proto_register (&mptcp_prot,MPTCP_USE_SLAB ) !=0 )net/mptcp/protocol.h +4
-0@@ -633,11 +633,15 @@ void mptcp_info2sockaddr(const struct mptcp_addr_info *info,struct mptcp_sched_ops *mptcp_sched_find (const char *name);int mptcp_register_scheduler (struct mptcp_sched_ops *sched);void mptcp_unregister_scheduler (struct mptcp_sched_ops *sched); +void mptcp_sched_init (void );int mptcp_init_sched (struct mptcp_sock *msk,struct mptcp_sched_ops *sched);void mptcp_release_sched (struct mptcp_sock *msk);void mptcp_subflow_set_scheduled (struct mptcp_subflow_context *subflow,bool scheduled); +struct sock *mptcp_subflow_get_send (const struct mptcp_sock *msk, +struct mptcp_sched_data *data); +struct sock *mptcp_subflow_get_retrans (const struct mptcp_sock *msk);static inline bool __tcp_can_send (const struct sock *ssk) {net/mptcp/sched.c +36
-2@@ -16,6 +16,33 @@static DEFINE_SPINLOCK (mptcp_sched_list_lock);static LIST_HEAD (mptcp_sched_list); +static void mptcp_sched_default_data_init (const struct mptcp_sock *msk, +struct mptcp_sched_data *data) +{ + data->snd_burst =0 ; +} + +static int mptcp_sched_default_get_subflow (const struct mptcp_sock *msk, +struct mptcp_sched_data *data) +{ +struct sock *ssk; + + ssk = data->reinject ?mptcp_subflow_get_retrans (msk) : +mptcp_subflow_get_send (msk, data); +if (!ssk) +return -EINVAL ; + +mptcp_subflow_set_scheduled (mptcp_subflow_ctx (ssk), true); +return 0 ; +} + +static struct mptcp_sched_ops mptcp_sched_default = { + .data_init = mptcp_sched_default_data_init, + .get_subflow = mptcp_sched_default_get_subflow, + .name ="default" , + .owner =THIS_MODULE , +}; +/* Must be called with rcu read lock held */ struct mptcp_sched_ops *mptcp_sched_find (const char *name) {@@ -50,16 +77,24 @@ int mptcp_register_scheduler(struct mptcp_sched_ops *sched)void mptcp_unregister_scheduler (struct mptcp_sched_ops *sched) { +if (sched == &mptcp_sched_default) +return ; +spin_lock (&mptcp_sched_list_lock);list_del_rcu (&sched->list);spin_unlock (&mptcp_sched_list_lock); } +void mptcp_sched_init (void ) +{ +mptcp_register_scheduler (&mptcp_sched_default); +} +int mptcp_init_sched (struct mptcp_sock *msk,struct mptcp_sched_ops *sched) {if (!sched) - goto out; + sched = &mptcp_sched_default;if (!bpf_try_module_get (sched, sched->owner))return -EBUSY ;@@ -70,7 +105,6 @@ int mptcp_init_sched(struct mptcp_sock *msk,pr_debug ("sched=%s" , msk->sched->name); -out:return 0 ; }Signature · 1 line
2.35.3
--
Mat Martineau
Intel