Re: rq lock contention due to commit af7f588d8f73

From: Aaron Lu
Date: Tue Mar 28 2023 - 02:58:41 EST


On Mon, Mar 27, 2023 at 03:57:43PM -0400, Mathieu Desnoyers wrote:

> I've just resuscitated my per-runqueue concurrency ID cache patch from an older
> patchset, and posted it as RFC. So far it passed one round of rseq selftests. Can
> you test it in your environment to see if I'm on the right track ?
>
> https://lore.kernel.org/lkml/20230327195318.137094-1-mathieu.desnoyers@xxxxxxxxxxxx/

There are improvements with this patch.

When running the client side sysbench with nr_thread=56, the lock contention
is gone%; with nr_thread=224(=nr_cpu of this machine), the lock contention
dropped from 75% to 27%.

v6.3.0-rc4:

75.21% 75.20% [kernel.vmlinux] [k] native_queued_spin_lock_slowpath
37.30% native_queued_spin_lock_slowpath;_raw_spin_lock;__schedule;schedule;schedule_hrtimeout_range_clock;schedule_hrtimeout_range;do_sys_poll;__x64_sys_poll;do_syscall_64;entry_SYSCALL_64_after_hwframe;__poll;0x7f943d6fcff8;PQgetResult;0x7f943d6f9a2b;0x55c7f9bde88b
26.01% native_queued_spin_lock_slowpath;_raw_spin_lock;__schedule;schedule_idle;do_idle;cpu_startup_entry;start_secondary;secondary_startup_64_no_verify
11.36% native_queued_spin_lock_slowpath;_raw_spin_lock;__schedule;schedule;schedule_hrtimeout_range_clock;schedule_hrtimeout_range;do_epoll_wait;__x64_sys_epoll_wait;do_syscall_64;entry_SYSCALL_64_after_hwframe;epoll_wait;secure_read;0x55c4d1363867;pq_getbyte;PostgresMain;0x55c4d140c828;PostmasterMain;main;__libc_start_main;0x5541d68949564100

v6.3.0-rc4+the_above_patch:

27.86% 27.85% [kernel.vmlinux] [k] native_queued_spin_lock_slowpath
7.31% native_queued_spin_lock_slowpath;_raw_spin_lock;try_to_wake_up;default_wake_function;pollwake;__wake_up_common;__wake_up_common_lock;__wake_up_sync_key;sock_def_readable;tcp_data_ready;tcp_rcv_established;tcp_v4_do_rcv;tcp_v4_rcv;ip_protocol_deliver_rcu;ip_local_deliver_finish;ip_local_deliver;ip_rcv;__netif_receive_skb_one_core;__netif_receive_skb;process_backlog;__napi_poll;net_rx_action;__do_softirq;do_softirq.part.0;__local_bh_enable_ip;ip_finish_output2;__ip_finish_output;ip_finish_output;ip_output;ip_local_out;__ip_queue_xmit;ip_queue_xmit;__tcp_transmit_skb;tcp_write_xmit;__tcp_push_pending_frames;tcp_push;tcp_sendmsg_locked;tcp_sendmsg;inet_sendmsg;sock_sendmsg;__sys_sendto;__x64_sys_sendto;do_syscall_64;entry_SYSCALL_64;__libc_send;0x5637a602e4cd;0x5637a602e98d;PostgresMain;0x5637a60d7828;PostmasterMain;main;__libc_start_main;0x5541d68949564100
4.62% native_queued_spin_lock_slowpath;_raw_spin_lock;__schedule;schedule_idle;do_idle;cpu_startup_entry;start_secondary;secondary_startup_64_no_verify
4.20% native_queued_spin_lock_slowpath;_raw_spin_lock;__schedule;schedule;schedule_hrtimeout_range_clock;schedule_hrtimeout_range;do_epoll_wait;__x64_sys_epoll_wait;do_syscall_64;entry_SYSCALL_64;epoll_wait;secure_read;0x5637a602e867;pq_getbyte;PostgresMain;0x5637a60d7828;PostmasterMain;main;__libc_start_main;0x5541d68949564100
1.66% native_queued_spin_lock_slowpath;_raw_spin_lock;raw_spin_rq_lock_nested;_raw_spin_rq_lock_irqsave;try_to_wake_up;default_wake_function;ep_autoremove_wake_function;__wake_up_common;__wake_up_common_lock;__wake_up;ep_poll_callback;__wake_up_common;__wake_up_common_lock;__wake_up_sync_key;sock_def_readable;tcp_data_ready;tcp_rcv_established;tcp_v4_do_rcv;tcp_v4_rcv;ip_protocol_deliver_rcu;ip_local_deliver_finish;ip_local_deliver;ip_rcv;__netif_receive_skb_one_core;__netif_receive_skb;process_backlog;__napi_poll;net_rx_action;__do_softirq;do_softirq.part.0;__local_bh_enable_ip;ip_finish_output2;__ip_finish_output;ip_finish_output;ip_output;ip_local_out;__ip_queue_xmit;ip_queue_xmit;__tcp_transmit_skb;tcp_write_xmit;__tcp_push_pending_frames;tcp_push;tcp_sendmsg_locked;tcp_sendmsg;inet_sendmsg;sock_sendmsg;__sys_sendto;__x64_sys_sendto;do_syscall_64;entry_SYSCALL_64;__libc_send
1.65% native_queued_spin_lock_slowpath;_raw_spin_lock;try_to_wake_up;default_wake_function;pollwake;__wake_up_common;__wake_up_common_lock;__wake_up_sync_key;sock_def_readable;tcp_data_ready;tcp_data_queue;tcp_rcv_established;tcp_v4_do_rcv;tcp_v4_rcv;ip_protocol_deliver_rcu;ip_local_deliver_finish;ip_local_deliver;ip_rcv;__netif_receive_skb_one_core;__netif_receive_skb;process_backlog;__napi_poll;net_rx_action;__do_softirq;do_softirq.part.0;__local_bh_enable_ip;ip_finish_output2;__ip_finish_output;ip_finish_output;ip_output;ip_local_out;__ip_queue_xmit;ip_queue_xmit;__tcp_transmit_skb;tcp_write_xmit;__tcp_push_pending_frames;tcp_push;tcp_sendmsg_locked;tcp_sendmsg;inet_sendmsg;sock_sendmsg;__sys_sendto;__x64_sys_sendto;do_syscall_64;entry_SYSCALL_64;__libc_send;0x5637a602e4cd;0x5637a602e69e;0x5637a602e7ae;0x5637a5e5f62e;standard_ExecutorRun;0x5637a615abeb;PortalRun;PostgresMain;0x5637a60d7828;PostmasterMain;main;__libc_start_main;0x5541d68949564100
1.63% native_queued_spin_lock_slowpath;_raw_spin_lock;raw_spin_rq_lock_nested;__schedule;schedule;schedule_hrtimeout_range_clock;schedule_hrtimeout_range;do_epoll_wait;__x64_sys_epoll_wait;do_syscall_64;entry_SYSCALL_64;epoll_wait;secure_read;0x5637a602e867;pq_getbyte;PostgresMain;0x5637a60d7828;PostmasterMain;main;__libc_start_main;0x5541d68949564100
1.40% native_queued_spin_lock_slowpath;_raw_spin_lock;raw_spin_rq_lock_nested;try_to_wake_up;default_wake_function;pollwake;__wake_up_common;__wake_up_common_lock;__wake_up_sync_key;sock_def_readable;tcp_data_ready;tcp_rcv_established;tcp_v4_do_rcv;tcp_v4_rcv;ip_protocol_deliver_rcu;ip_local_deliver_finish;ip_local_deliver;ip_rcv;__netif_receive_skb_one_core;__netif_receive_skb;process_backlog;__napi_poll;net_rx_action;__do_softirq;do_softirq.part.0;__local_bh_enable_ip;ip_finish_output2;__ip_finish_output;ip_finish_output;ip_output;ip_local_out;__ip_queue_xmit;ip_queue_xmit;__tcp_transmit_skb;tcp_write_xmit;__tcp_push_pending_frames;tcp_push;tcp_sendmsg_locked;tcp_sendmsg;inet_sendmsg;sock_sendmsg;__sys_sendto;__x64_sys_sendto;do_syscall_64;entry_SYSCALL_64;__libc_send;0x5637a602e4cd;0x5637a602e98d;PostgresMain;0x5637a60d7828;PostmasterMain;main;__libc_start_main;0x5541d68949564100
1.20% native_queued_spin_lock_slowpath;_raw_spin_lock;raw_spin_rq_lock_nested;_raw_spin_rq_lock_irqsave;try_to_wake_up;default_wake_function;pollwake;__wake_up_common;__wake_up_common_lock;__wake_up_sync_key;sock_def_readable;tcp_data_ready;tcp_rcv_established;tcp_v4_do_rcv;tcp_v4_rcv;ip_protocol_deliver_rcu;ip_local_deliver_finish;ip_local_deliver;ip_rcv;__netif_receive_skb_one_core;__netif_receive_skb;process_backlog;__napi_poll;net_rx_action;__do_softirq;do_softirq.part.0;__local_bh_enable_ip;ip_finish_output2;__ip_finish_output;ip_finish_output;ip_output;ip_local_out;__ip_queue_xmit;ip_queue_xmit;__tcp_transmit_skb;tcp_write_xmit;__tcp_push_pending_frames;tcp_push;tcp_sendmsg_locked;tcp_sendmsg;inet_sendmsg;sock_sendmsg;__sys_sendto;__x64_sys_sendto;do_syscall_64;entry_SYSCALL_64;__libc_send;0x5637a602e4cd;0x5637a602e98d;PostgresMain;0x5637a60d7828;PostmasterMain;main;__libc_start_main;0x5541d68949564100
0.83% native_queued_spin_lock_slowpath;_raw_spin_lock;raw_spin_rq_lock_nested;__task_rq_lock;try_to_wake_up;default_wake_function;ep_autoremove_wake_function;__wake_up_common;__wake_up_common_lock;__wake_up;ep_poll_callback;__wake_up_common;__wake_up_common_lock;__wake_up_sync_key;sock_def_readable;tcp_data_ready;tcp_rcv_established;tcp_v4_do_rcv;tcp_v4_rcv;ip_protocol_deliver_rcu;ip_local_deliver_finish;ip_local_deliver;ip_rcv;__netif_receive_skb_one_core;__netif_receive_skb;process_backlog;__napi_poll;net_rx_action;__do_softirq;do_softirq.part.0;__local_bh_enable_ip;ip_finish_output2;__ip_finish_output;ip_finish_output;ip_output;ip_local_out;__ip_queue_xmit;ip_queue_xmit;__tcp_transmit_skb;tcp_write_xmit;__tcp_push_pending_frames;tcp_push;tcp_sendmsg_locked;tcp_sendmsg;inet_sendmsg;sock_sendmsg;__sys_sendto;__x64_sys_sendto;do_syscall_64;entry_SYSCALL_64;__libc_send
0.65% native_queued_spin_lock_slowpath;_raw_spin_lock;raw_spin_rq_lock_nested;__schedule;schedule_idle;do_idle;cpu_startup_entry;start_secondary;secondary_startup_64_no_verify