summaryrefslogtreecommitdiff
path: root/net/xdp
diff options
context:
space:
mode:
authorMagnus Karlsson <magnus.karlsson@intel.com>2020-11-16 12:12:45 +0100
committerDaniel Borkmann <daniel@iogearbox.net>2020-11-17 22:07:40 +0100
commitb8c7aece29bc06e1e63efeefb9e31ab259e84ea2 (patch)
tree7f9f19040af3ec4d73d0b21878886ae1a84afc67 /net/xdp
parentf320460b9489d80355821829069fdefa6c698cb4 (diff)
xsk: Introduce padding between more ring pointers
Introduce one cache line worth of padding between the consumer pointer and the flags field as well as between the flags field and the start of the descriptors in all the lockless rings. This so that the x86 HW adjacency prefetcher will not prefetch the adjacent pointer/field when only one pointer/field is going to be used. This improves throughput performance for the l2fwd sample app with 1% on my machine with HW prefetching turned on in the BIOS. Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Acked-by: John Fastabend <john.fastabend@gmail.com> Link: https://lore.kernel.org/bpf/1605525167-14450-4-git-send-email-magnus.karlsson@gmail.com
Diffstat (limited to 'net/xdp')
-rw-r--r--net/xdp/xsk_queue.h4
1 files changed, 3 insertions, 1 deletions
diff --git a/net/xdp/xsk_queue.h b/net/xdp/xsk_queue.h
index cdb9cf3cd136..74fac802cce1 100644
--- a/net/xdp/xsk_queue.h
+++ b/net/xdp/xsk_queue.h
@@ -18,9 +18,11 @@ struct xdp_ring {
/* Hinder the adjacent cache prefetcher to prefetch the consumer
* pointer if the producer pointer is touched and vice versa.
*/
- u32 pad ____cacheline_aligned_in_smp;
+ u32 pad1 ____cacheline_aligned_in_smp;
u32 consumer ____cacheline_aligned_in_smp;
+ u32 pad2 ____cacheline_aligned_in_smp;
u32 flags;
+ u32 pad3 ____cacheline_aligned_in_smp;
};
/* Used for the RX and TX queues for packets */