From: zhangliping
Max delat_t should be the full_bucket/rate instead of the full_bucket.
Also report EINVAL if the rate is zero.
Fixes: 96fbc13d7e77 ("openvswitch: Add meter infrastructure")
Cc: Andy Zhou
Signed-off-by: zhangliping
---
V2: report EINVAL if the rate is 0 to avoid
From: zhangliping
Max delat_t should be the full_bucket/rate instead of the full_bucket.
Fixes: 96fbc13d7e77 ("openvswitch: Add meter infrastructure")
Cc: Andy Zhou
Signed-off-by: zhangliping
---
net/openvswitch/meter.c | 7 ---
1 file changed, 4 insertions(+), 3 deletions(-)
Hi,
At 2017-12-18 22:45:30, "Paolo Abeni" wrote:
>Understood, thanks. Still the time spent in 'udp4_lib_lookup2' looks
>quite different/higher than what I observe in my tests. Are you using
>x86_64? if not, do you see many cache misses in udp4_lib_lookup2?
Yes, x86_64. Here is the host's lscpu o
Hi,
At 2017-12-18 18:26:28, "Paolo Abeni" wrote:
>Hi,
>
>On Mon, 2017-12-18 at 12:11 +0800, zhangliping wrote:
>> From: zhangliping
>>
>> Under our udp pressure performance test, after gro is disabled, rx rate
>> will be improved from ~2500kpp
From: zhangliping
Under our udp pressure performance test, after gro is disabled, rx rate
will be improved from ~2500kpps to ~2800kpps. We can find some difference
from perf report:
1. gro is enabled:
24.23% [kernel] [k] udp4_lib_lookup2
5.42% [kernel] [k] __memcpy
3.87
From: zhangliping
If we want to add a datapath flow, which has more than 500 vxlan outputs'
action, we will get the following error reports:
openvswitch: netlink: Flow action size 32832 bytes exceeds max
openvswitch: netlink: Flow action size 32832 bytes exceeds max
openvswitch: ne
