forked from Minki/linux
29b4433d99
We tried very hard to remove all possible dev_hold()/dev_put() pairs in
network stack, using RCU conversions.
There is still an unavoidable device refcount change for every dst we
create/destroy, and this can slow down some workloads (routers or some
app servers, mmap af_packet)
We can switch to a percpu refcount implementation, now dynamic per_cpu
infrastructure is mature. On a 64 cpus machine, this consumes 256 bytes
per device.
On x86, dev_hold(dev) code :
before
lock incl 0x280(%ebx)
after:
movl 0x260(%ebx),%eax
incl fs:(%eax)
Stress bench :
(Sending 160.000.000 UDP frames,
IP route cache disabled, dual E5540 @2.53GHz,
32bit kernel, FIB_TRIE)
Before:
real 1m1.662s
user 0m14.373s
sys 12m55.960s
After:
real 0m51.179s
user 0m15.329s
sys 10m15.942s
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
||
|---|---|---|
| .. | ||
| Kconfig | ||
| Makefile | ||
| nes_cm.c | ||
| nes_cm.h | ||
| nes_context.h | ||
| nes_hw.c | ||
| nes_hw.h | ||
| nes_nic.c | ||
| nes_user.h | ||
| nes_utils.c | ||
| nes_verbs.c | ||
| nes_verbs.h | ||
| nes.c | ||
| nes.h | ||