diff options
author | Eric Dumazet <eric.dumazet@gmail.com> | 2011-11-14 06:05:34 +0000 |
---|---|---|
committer | David S. Miller <davem@davemloft.net> | 2011-11-14 14:13:30 -0500 |
commit | e52fcb2462ac484e6dd6e68869536609f0216938 (patch) | |
tree | 794407ff242481d5d09e3e669cabfe129d80f6df /drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c | |
parent | b2b5ce9d1ccf1c45f8ac68e5d901112ab76ba199 (diff) | |
download | op-kernel-dev-e52fcb2462ac484e6dd6e68869536609f0216938.zip op-kernel-dev-e52fcb2462ac484e6dd6e68869536609f0216938.tar.gz |
bnx2x: uses build_skb() in receive path
bnx2x uses following formula to compute its rx_buf_sz :
dev->mtu + 2*L1_CACHE_BYTES + 14 + 8 + 8 + 2
Then core network adds NET_SKB_PAD and SKB_DATA_ALIGN(sizeof(struct
skb_shared_info))
Final allocated size for skb head on x86_64 (L1_CACHE_BYTES = 64,
MTU=1500) : 2112 bytes : SLUB/SLAB round this to 4096 bytes.
Since skb truesize is then bigger than SK_MEM_QUANTUM, we have lot of
false sharing because of mem_reclaim in UDP stack.
One possible way to half truesize is to reduce the need by 64 bytes
(2112 -> 2048 bytes)
Instead of allocating a full cache line at the end of packet for
alignment, we can use the fact that skb_shared_info sits at the end of
skb->head, and we can use this room, if we convert bnx2x to new
build_skb() infrastructure.
skb_shared_info will be initialized after hardware finished its
transfert, so we can eventually overwrite the final padding.
Using build_skb() also reduces cache line misses in the driver, since we
use cache hot skb instead of cold ones. Number of in-flight sk_buff
structures is lower, they are recycled while still hot.
Performance results :
(820.000 pps on a rx UDP monothread benchmark, instead of 720.000 pps)
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
CC: Eilon Greenstein <eilong@broadcom.com>
CC: Ben Hutchings <bhutchings@solarflare.com>
CC: Tom Herbert <therbert@google.com>
CC: Jamal Hadi Salim <hadi@mojatatu.com>
CC: Stephen Hemminger <shemminger@vyatta.com>
CC: Thomas Graf <tgraf@infradead.org>
CC: Herbert Xu <herbert@gondor.apana.org.au>
CC: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Acked-by: Eilon Greenstein <eilong@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c')
-rw-r--r-- | drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c | 6 |
1 files changed, 3 insertions, 3 deletions
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c index f6402fa..ec31871 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c @@ -1740,6 +1740,7 @@ static int bnx2x_run_loopback(struct bnx2x *bp, int loopback_mode) struct sw_rx_bd *rx_buf; u16 len; int rc = -ENODEV; + u8 *data; /* check the loopback mode */ switch (loopback_mode) { @@ -1865,10 +1866,9 @@ static int bnx2x_run_loopback(struct bnx2x *bp, int loopback_mode) dma_sync_single_for_cpu(&bp->pdev->dev, dma_unmap_addr(rx_buf, mapping), fp_rx->rx_buf_size, DMA_FROM_DEVICE); - skb = rx_buf->skb; - skb_reserve(skb, cqe->fast_path_cqe.placement_offset); + data = rx_buf->data + NET_SKB_PAD + cqe->fast_path_cqe.placement_offset; for (i = ETH_HLEN; i < pkt_size; i++) - if (*(skb->data + i) != (unsigned char) (i & 0xff)) + if (*(data + i) != (unsigned char) (i & 0xff)) goto test_loopback_rx_exit; rc = 0; |