diff options
| author |   Simon Zhang <yuwei1.zhang@intel.com> | 2018-11-26 17:15:24 +0800 |
|---|---|---|
| committer |   Damjan Marion <dmarion@me.com> | 2018-11-28 08:43:46 +0000 |
| commit | 5a5a86987b1b5bd792e9615c3f26471abd4babb4 (patch) | |
| tree | fc376ca9d216a8b3dd19c54217b0d3d2a91a29fc /src/vnet | |
| parent | 7e713f133a62efdc9d3019aeeee15cce774087e2 (diff) | |
Remove useless prefetch in ip4-rewrite node
Prefetching first 2 packets' header is useless cause of the prefetching
action is not done before using the packets.
There's no performance drop in Xeon platform and slightly performance
gain in Atom platform after rmoving the prefetch.
Change-Id: Ib4b074af20d7cd5053aecc7147b162141aec31f5
Signed-off-by: Simon Zhang <yuwei1.zhang@intel.com>
Diffstat (limited to 'src/vnet')
| -rw-r--r-- | src/vnet/ip/ip4_forward.c | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/src/vnet/ip/ip4_forward.c b/src/vnet/ip/ip4_forward.c index 9963f6b1e26..1911f088008 100644 --- a/src/vnet/ip/ip4_forward.c +++ b/src/vnet/ip/ip4_forward.c @@ -2197,7 +2197,7 @@ ip4_rewrite_inline (vlib_main_t * vm, if (n_left_from >= 6) { int i; - for (i = 0; i < 6; i++) + for (i = 2; i < 6; i++) vlib_prefetch_buffer_header (bufs[i], LOAD); } |