Use of percpu_counter structure to track count of orphaned
sockets is causing problems on modern hosts with 256 cpus
or more.
Stefan Bach reported a serious spinlock contention in real workloads,
that I was able to reproduce with a netfilter rule dropping
incoming FIN packets.
53.56% server [kernel.kallsyms] [k] queued_spin_lock_slowpath
|
---queued_spin_lock_slowpath
|
--53.51%--_raw_spin_lock_irqsave
|
--53.51%--__percpu_counter_sum
tcp_check_oom
|
|--39.03%--__tcp_close
| tcp_close
| inet_release
| inet6_release
| sock_close
| __fput
| ____fput
| task_work_run
| exit_to_usermode_loop
| do_syscall_64
| entry_SYSCALL_64_after_hwframe
| __GI___libc_close
|
--14.48%--tcp_out_of_resources
tcp_write_timeout
tcp_retransmit_timer
tcp_write_timer_handler
tcp_write_timer
call_timer_fn
expire_timers
__run_timers
run_timer_softirq
__softirqentry_text_start
As explained in commit cf86a086a1 ("net/dst: use a smaller percpu_counter
batch for dst entries accounting"), default batch size is too big
for the default value of tcp_max_orphans (262144).
But even if we reduce batch sizes, there would still be cases
where the estimated count of orphans is beyond the limit,
and where tcp_too_many_orphans() has to call the expensive
percpu_counter_sum_positive().
One solution is to use plain per-cpu counters, and have
a timer to periodically refresh this cache.
Updating this cache every 100ms seems about right, tcp pressure
state is not radically changing over shorter periods.
percpu_counter was nice 15 years ago while hosts had less
than 16 cpus, not anymore by current standards.
v2: Fix the build issue for CONFIG_CRYPTO_DEV_CHELSIO_TLS=m,
reported by kernel test robot <lkp@intel.com>
Remove unused socket argument from tcp_too_many_orphans()
Fixes: dd24c00191 ("net: Use a percpu_counter for orphan_count")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Stefan Bach <sfb@google.com>
Cc: Neal Cardwell <ncardwell@google.com>
Acked-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The Max imm data size in cxgb4 is not similar to the max imm data size
in the chtls. This caused an mismatch in output of is_ofld_imm() of
cxgb4 and chtls. So fixed this by keeping the max wreq size of imm data
same in both chtls and cxgb4 as MAX_IMM_OFLD_TX_DATA_WR_LEN.
As cxgb4's max imm. data value for ofld packets is changed to
MAX_IMM_OFLD_TX_DATA_WR_LEN. Using the same in cxgbit also.
Fixes: 36bedb3f2e ("crypto: chtls - Inline TLS record Tx")
Signed-off-by: Ayush Sawal <ayush.sawal@chelsio.com>
Acked-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
chelsio inline tls driver(chtls) is mostly overlaps with NIC drivers
but currenty it is part of crypto driver, so move it out to appropriate
directory for better maintenance.
Signed-off-by: Vinay Kumar Yadav <vinay.yadav@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2020-08-21 14:15:15 -07:00
Renamed from drivers/crypto/chelsio/chtls/chtls_cm.h (Browse further)
