BPF redundant scheduler
Subsystems (5)
v8 cover BPF redundant scheduler
v8:
- update __mptcp_push_pending(), send the same data on each subflow.
- update __mptcp_retrans, track the max sent data.
= add a new patch.
v7:
- drop redundant flag in v6
- drop __mptcp_subflows_push_pending in v6
- update redundant subflows support in __mptcp_push_pending
- update redundant subflows support in __mptcp_retrans
v6:
- Add redundant flag for struct mptcp_sched_ops.
- add a dedicated function __mptcp_subflows_push_pending() to deal with redundat subflows push pending.
v5:
- address to Paolo's comment, keep the optimization tomptcp_subflow_get_send() for the non eBPF case.
- merge mptcp_sched_get_send() and __mptcp_sched_get_send() in v4 into one.
- depends on "cleanups for bpf sched selftests".
v4:
- small cleanups in patch 1, 2.
- add TODO in patch 3.
- rebase patch 5 on 'cleanups for bpf sched selftests'.
v3:
- use new API.
- fix the link failure tests issue mentioned in ("https://patchwork.kernel.org/project/mptcp/cover/[email protected]/").
v2:
- add MPTCP_SUBFLOWS_MAX limit to avoid infinite loops when the scheduler always sets call_again to true.
- track the largest copied amount.
- deal with __mptcp_subflow_push_pending() and the retransmit loop.
- depends on "BPF round-robin scheduler" v14.
v1:
Implements the redundant BPF MPTCP scheduler, which sends all packets redundantly on all available subflows.
Geliang Tang (6):
Squash to "mptcp: add get_subflow wrappers"
mptcp: accept right edge of mapping validating
mptcp: redundant subflows push pending
mptcp: redundant subflows retrans support
selftests/bpf: Add bpf_red scheduler
selftests/bpf: Add bpf_red test
net/mptcp/protocol.c | 169 +++++++++++------- net/mptcp/protocol.h | 4 +- net/mptcp/sched.c | 63 ++++--- net/mptcp/subflow.c | 4 +- .../testing/selftests/bpf/prog_tests/mptcp.c | 34 ++++ .../selftests/bpf/progs/mptcp_bpf_red.c | 36 ++++ 6 files changed, 214 insertions(+), 96 deletions(-) create mode 100644 tools/testing/selftests/bpf/progs/mptcp_bpf_red.c
Signature · 1 line
2.35.3
PATCH v8 1/6 Squash to "mptcp: add get_subflow wrappers"
Please update the commit log:
'''
This patch defines two new wrappers mptcp_sched_get_send() and mptcp_sched_get_retrans(), invoke get_subflow() of msk->sched in them. Use them instead of using mptcp_subflow_get_send() or mptcp_subflow_get_retrans() directly.
Set the subflow pointers array in struct mptcp_sched_data before invoking get_subflow(), then it can be used in get_subflow() in the BPF contexts.
Check the subflow scheduled flags to test which subflow or subflows are picked by the scheduler.
Move sock_owned_by_me() and the fallback check code from mptcp_subflow_get_send/retrans() into the wrappers.
Redundant subflows are not supported in __mptcp_subflow_push_pending() yet. This patch adds a placeholder in mptcp_sched_get_send() to pick the first subflow for the redundant subflows case.
'''
--- net/mptcp/protocol.c | 12 +++++---- net/mptcp/protocol.h | 4 +-- net/mptcp/sched.c | 63 ++++++++++++++++++++++++++------------------ 3 files changed, 47 insertions(+), 32 deletions(-)
net/mptcp/protocol.c +7 -5
@@ -1532,7 +1532,7 @@ 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
@@ -1593,13 +1593,13 @@ static void __mptcp_subflow_push_pending(struct sock *sk, struct sock *ssk) info.limit = dfrag->data_len; len = dfrag->data_len - dfrag->already_sent;while (len >0 ) { -int ret =0 ; +int ret =0 , err =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)); + xmit_ssk = first ? ssk :mptcp_sched_get_send (mptcp_sk (sk), &err);if (!xmit_ssk) goto out;if (xmit_ssk != ssk) {
@@ -2408,7 +2408,7 @@ 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)) {
@@ -3090,11 +3090,13 @@ void __mptcp_data_acked(struct sock *sk)void __mptcp_check_push (struct sock *sk,struct sock *ssk) { +int err =0 ; +if (!mptcp_send_head (sk))return ;if (!sock_owned_by_user (sk)) { -struct sock *xmit_ssk =mptcp_sched_get_send (mptcp_sk (sk)); +struct sock *xmit_ssk =mptcp_sched_get_send (mptcp_sk (sk), &err);if (xmit_ssk == ssk)__mptcp_subflow_push_pending (sk, ssk);
net/mptcp/protocol.h +2 -2
@@ -635,8 +635,8 @@ 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); +struct sock *mptcp_sched_get_send (struct mptcp_sock *msk,int *err); +int mptcp_sched_get_retrans (struct mptcp_sock *msk);static inline bool __mptcp_subflow_active (struct mptcp_subflow_context *subflow) {
net/mptcp/sched.c +38 -25
103 lines not on this page · raw message
Signature · 1 line
2.35.3
Re: [PATCH mptcp-next v8 1/6] Squash to "mptcp: add get_subflow wrappers"
net/mptcp/sched.c static int mptcp_sched_data_init(struct mptcp_sock *msk, bool reinject, · line 123
Quoted 3 lines · Geliang Tang
net/mptcp/sched.c +1
-1-struct sock *mptcp_sched_get_send (struct mptcp_sock *msk) +struct sock *mptcp_sched_get_send (struct mptcp_sock *msk,int *err) {
Why the extra *err?
Almost everyone else uses ERR_PTR, PTR_ERR, IS_ERR, etc. to return the error code directly, why not follow that model?
PATCH v8 2/6 mptcp: accept right edge of mapping validating
In reply to Geliang TangIf ssn is at the right edge of DSS mapping (ssn == subflow->map_subflow_seq
+ subflow->map_data_len), it shouldn't be handled as a bad map. This
happens in the redundant subflows test cases.
--- net/mptcp/subflow.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
net/mptcp/subflow.c +2 -2
@@ -878,8 +878,8 @@ static bool validate_mapping(struct sock *ssk, struct sk_buff *skb)dbg_bad_map (subflow, ssn);return false; } -if (unlikely (!before (ssn, subflow->map_subflow_seq + - subflow->map_data_len))) { +if (unlikely (after (ssn, subflow->map_subflow_seq + + subflow->map_data_len))) {/* Mapping does covers past subflow data, invalid */ dbg_bad_map (subflow, ssn);return false;
Signature · 1 line
2.35.3
PATCH v8 3/6 mptcp: redundant subflows push pending
In reply to Geliang TangThis patch adds the redundant subflows support for __mptcp_push_pending(). Use mptcp_sched_get_send() wrapper instead of mptcp_subflow_get_send() in it.
Check the subflow scheduled flags to test which subflow or subflows are picked by the scheduler, use them to send data.
--- net/mptcp/protocol.c | 99 ++++++++++++++++++++++++++------------------ 1 file changed, 58 insertions(+), 41 deletions(-)
net/mptcp/protocol.c +58 -41
@@ -1516,63 +1516,80 @@ 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, - }; +struct mptcp_subflow_context *subflow;struct mptcp_data_frag *dfrag; -int len, copied =0 ; +int len, max =0 , err =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 ; +mptcp_sched_get_send (msk, &err); +if (err) + goto out; - prev_ssk = ssk; - ssk =mptcp_subflow_get_send (msk); + dfrag =mptcp_send_head (sk); +if (!dfrag) + goto out; -/* 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; +mptcp_for_each_subflow (msk, subflow) { +if (READ_ONCE (subflow->scheduled)) { +struct sock *prev_ssk =NULL , *ssk =NULL ; +struct mptcp_sendmsg_info info = { + .flags = flags, + }; +int copied =0 ; -/* 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); + 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 ) { -mptcp_push_release (ssk, &info); - goto out; - } + prev_ssk = ssk; + ssk =mptcp_subflow_tcp_sock (subflow); - info.sent += ret; - copied += ret; - len -= ret; +/* 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 ) { +mptcp_push_release (ssk, &info); + goto out; + } + + info.sent += ret; + copied += ret; + len -= ret; + } + max =max (copied, max); -mptcp_update_post_push (msk, dfrag, ret); +/* at this point we held the socket lock for the last subflow we used */ +if (ssk) { +mptcp_push_release (ssk, &info); + msk->last_snd = ssk; +mptcp_subflow_set_scheduled (subflow, false); + } } -WRITE_ONCE (msk->first_pending,mptcp_send_next (sk)); } - -/* at this point we held the socket lock for the last subflow we used */ -if (ssk) -mptcp_push_release (ssk, &info); +if (max) +mptcp_update_post_push (msk, dfrag, max); +WRITE_ONCE (msk->first_pending,mptcp_send_next (sk)); out:/* ensure the rtx timer is running */ if (!mptcp_timer_pending (sk))mptcp_reset_timer (sk); -if (copied) +if (max)__mptcp_check_send_data_fin (sk); }
Signature · 1 line
2.35.3
PATCH v8 4/6 mptcp: redundant subflows retrans support
In reply to Geliang TangThis patch adds the redundant subflows support for __mptcp_retrans(). In it, use sched_get_retrans() wrapper instead of mptcp_subflow_get_retrans().
Iterate each subflow of msk, check the scheduled flag to test if it is picked by the scheduler. If so, use it to send data.
--- net/mptcp/protocol.c | 62 ++++++++++++++++++++++++++++---------------- 1 file changed, 39 insertions(+), 23 deletions(-)
net/mptcp/protocol.c +39 -23
@@ -2416,16 +2416,17 @@ static void mptcp_check_fastclose(struct mptcp_sock *msk)static void __mptcp_retrans (struct sock *sk) {struct mptcp_sock *msk =mptcp_sk (sk); +struct mptcp_subflow_context *subflow;struct mptcp_sendmsg_info info = {};struct mptcp_data_frag *dfrag; -size_t copied =0 ;struct sock *ssk; -int ret; +int ret, err; +u16 max =0 ;mptcp_clean_una_wakeup (sk);/* first check ssk: need to kick "stale" logic */ - ssk =mptcp_subflow_get_retrans (msk); + err =mptcp_sched_get_retrans (msk); dfrag =mptcp_rtx_head (sk);if (!dfrag) {if (mptcp_data_fin_enabled (msk)) {
@@ -2444,31 +2445,46 @@ static void __mptcp_retrans(struct sock *sk) goto reset_timer; } -if (!ssk) +if (err) goto reset_timer; -lock_sock (ssk); +mptcp_for_each_subflow (msk, subflow) { +if (READ_ONCE (subflow->scheduled)) { +u16 copied =0 ; -/* limit retransmission to the bytes already sent on some subflows */ - info.sent =0 ; - info.limit =READ_ONCE (msk->csum_enabled) ? dfrag->data_len : dfrag->already_sent; -while (info.sent < info.limit) { - ret =mptcp_sendmsg_frag (sk, ssk, dfrag, &info); -if (ret <=0 ) -break ; + ssk =mptcp_subflow_tcp_sock (subflow); +if (!ssk) + goto reset_timer; -MPTCP_INC_STATS (sock_net (sk),MPTCP_MIB_RETRANSSEGS ); - copied += ret; - info.sent += ret; - } -if (copied) { - dfrag->already_sent =max (dfrag->already_sent, info.sent); -tcp_push (ssk,0 , info.mss_now,tcp_sk (ssk)->nonagle, - info.size_goal); -WRITE_ONCE (msk->allow_infinite_fallback, false); - } +lock_sock (ssk); -release_sock (ssk); +/* limit retransmission to the bytes already sent on some subflows */ + info.sent =0 ; + info.limit =READ_ONCE (msk->csum_enabled) ? dfrag->data_len : + dfrag->already_sent; +while (info.sent < info.limit) { + ret =mptcp_sendmsg_frag (sk, ssk, dfrag, &info); +if (ret <=0 ) +break ; + +MPTCP_INC_STATS (sock_net (sk),MPTCP_MIB_RETRANSSEGS ); + copied += ret; + info.sent += ret; + } +if (copied) { + max =max (copied, max); +tcp_push (ssk,0 , info.mss_now,tcp_sk (ssk)->nonagle, + info.size_goal); +WRITE_ONCE (msk->allow_infinite_fallback, false); + } + +release_sock (ssk); + + msk->last_snd = ssk; +mptcp_subflow_set_scheduled (subflow, false); + } + } + dfrag->already_sent =max (dfrag->already_sent, max); reset_timer:mptcp_check_and_set_pending (sk);
Signature · 1 line
2.35.3
PATCH v8 5/6 selftests/bpf: Add bpf_red scheduler
In reply to Geliang TangThis patch implements the redundant BPF MPTCP scheduler, named bpf_red, which sends all packets redundantly on all available subflows.
--- .../selftests/bpf/progs/mptcp_bpf_red.c | 36 +++++++++++++++++++ 1 file changed, 36 insertions(+) create mode 100644 tools/testing/selftests/bpf/progs/mptcp_bpf_red.c
tools/testing/selftests/bpf/progs/mptcp_bpf_red.c new file +36 -0
@@ -0,0 +1,36 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2022, SUSE. */ + +#include <linux/bpf.h> +#include "bpf_tcp_helpers.h" + +char _license[] SEC("license") = "GPL"; + +SEC("struct_ops/mptcp_sched_red_init") +void BPF_PROG (mptcp_sched_red_init ,const struct mptcp_sock *msk) +{ +} + +SEC("struct_ops/mptcp_sched_red_release") +void BPF_PROG (mptcp_sched_red_release ,const struct mptcp_sock *msk) +{ +} + +void BPF_STRUCT_OPS (bpf_red_get_subflow ,const struct mptcp_sock *msk, +struct mptcp_sched_data *data) +{ +for (int i =0 ; i <MPTCP_SUBFLOWS_MAX ; i++) { +if (!data->contexts[i]) +break ; + +mptcp_subflow_set_scheduled (data->contexts[i], true); + } +} + +SEC(".struct_ops") +struct mptcp_sched_ops red = { + .init = (void *)mptcp_sched_red_init, + .release = (void *)mptcp_sched_red_release, + .get_subflow = (void *)bpf_red_get_subflow, + .name ="bpf_red" , +};
Signature · 1 line
2.35.3
PATCH v8 6/6 selftests/bpf: Add bpf_red test
In reply to Geliang TangThis patch adds the redundant BPF MPTCP scheduler test: test_red(). Use sysctl to set net.mptcp.scheduler to use this sched. Add two veth net devices to simulate the multiple addresses case. Use 'ip mptcp endpoint' command to add the new endpoint ADDR_2 to PM netlink. Send data and check bytes_sent of 'ss' output after it to make sure the data has been redundantly sent on both net devices.
--- .../testing/selftests/bpf/prog_tests/mptcp.c | 34 +++++++++++++++++++ 1 file changed, 34 insertions(+)
tools/testing/selftests/bpf/prog_tests/mptcp.c +34 -0
@@ -9,6 +9,7 @@#include "mptcp_bpf_first.skel.h" #include "mptcp_bpf_bkup.skel.h" #include "mptcp_bpf_rr.skel.h" +#include "mptcp_bpf_red.skel.h" #ifndef TCP_CA_NAME_MAX #define TCP_CA_NAME_MAX 16
@@ -381,6 +382,37 @@ static void test_rr(void)mptcp_bpf_rr__destroy (rr_skel); } +static void test_red (void ) +{ +struct mptcp_bpf_red *red_skel; +int server_fd, client_fd; +struct bpf_link *link; + + red_skel =mptcp_bpf_red__open_and_load (); +if (!ASSERT_OK_PTR (red_skel,"bpf_red__open_and_load" )) +return ; + + link =bpf_map__attach_struct_ops (red_skel->maps.red); +if (!ASSERT_OK_PTR (link,"bpf_map__attach_struct_ops" )) { +mptcp_bpf_red__destroy (red_skel); +return ; + } + +sched_init ("subflow" ,"bpf_red" ); + server_fd =start_mptcp_server (AF_INET ,ADDR_1 ,0 ,0 ); + client_fd =connect_to_fd (server_fd,0 ); + +send_data (server_fd, client_fd); +ASSERT_OK (has_bytes_sent (ADDR_1 ),"has_bytes_sent addr 1" ); +ASSERT_OK (has_bytes_sent (ADDR_2 ),"has_bytes_sent addr 2" ); + +close (client_fd); +close (server_fd); +sched_cleanup (); +bpf_link__destroy (link); +mptcp_bpf_red__destroy (red_skel); +} +void test_mptcp (void ) {if (test__start_subtest ("base" ))
@@ -391,4 +423,6 @@ void test_mptcp(void)test_bkup ();if (test__start_subtest ("rr" ))test_rr (); +if (test__start_subtest ("red" )) +test_red (); }
Signature · 1 line
2.35.3
Re: selftests/bpf: Add bpf_red test: 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:
- Unstable: 3 failed test(s): mptcp_connect_mmap selftest_mptcp_connect selftest_simult_flows - Critical: 2 Call Trace(s) ❌:
- Task: https://cirrus-ci.com/task/6077066394206208
- Summary: https://api.cirrus-ci.com/v1/artifact/task/6077066394206208/summary/summary.txt
- KVM Validation: debug:
- Unstable: 3 failed test(s): selftest_mptcp_connect selftest_mptcp_join selftest_simult_flows - Critical: 2 Call Trace(s) ❌:
- Task: https://cirrus-ci.com/task/5514116440784896
- Summary: https://api.cirrus-ci.com/v1/artifact/task/5514116440784896/summary/summary.txt
Initiator: Patchew Applier
Commits: https://github.com/multipath-tcp/mptcp_net-next/commits/450c64653a9a
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)