diff options
author | Ian Campbell <Ian.Campbell@citrix.com> | 2013-01-07 05:32:06 (GMT) |
---|---|---|
committer | David S. Miller <davem@davemloft.net> | 2013-01-08 03:51:19 (GMT) |
commit | d9a58a782e396a0f04e8445b7ba3763c8a48c7fe (patch) | |
tree | 319af7e412885f2c7c9c7f45d5d7bfbc68d762cd /drivers/net/xen-netfront.c | |
parent | 32fa10b24ef64b41e8ca17068fa9dc625427a05e (diff) | |
download | linux-d9a58a782e396a0f04e8445b7ba3763c8a48c7fe.tar.xz |
xen/netfront: improve truesize tracking
Using RX_COPY_THRESHOLD is incorrect if the SKB is actually smaller
than that. We have already accounted for this in
NETFRONT_SKB_CB(skb)->pull_to so use that instead.
Fixes WARN_ON from skb_try_coalesce.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Cc: Sander Eikelenboom <linux@eikelenboom.it>
Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Cc: annie li <annie.li@oracle.com>
Cc: xen-devel@lists.xen.org
Cc: netdev@vger.kernel.org
Cc: stable@kernel.org # 3.7.x only
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'drivers/net/xen-netfront.c')
-rw-r--r-- | drivers/net/xen-netfront.c | 27 |
1 files changed, 4 insertions, 23 deletions
diff --git a/drivers/net/xen-netfront.c b/drivers/net/xen-netfront.c index c26e28b..7ffa43b 100644 --- a/drivers/net/xen-netfront.c +++ b/drivers/net/xen-netfront.c @@ -1015,29 +1015,10 @@ err: i = xennet_fill_frags(np, skb, &tmpq); /* - * Truesize approximates the size of true data plus - * any supervisor overheads. Adding hypervisor - * overheads has been shown to significantly reduce - * achievable bandwidth with the default receive - * buffer size. It is therefore not wise to account - * for it here. - * - * After alloc_skb(RX_COPY_THRESHOLD), truesize is set - * to RX_COPY_THRESHOLD + the supervisor - * overheads. Here, we add the size of the data pulled - * in xennet_fill_frags(). - * - * We also adjust for any unused space in the main - * data area by subtracting (RX_COPY_THRESHOLD - - * len). This is especially important with drivers - * which split incoming packets into header and data, - * using only 66 bytes of the main data area (see the - * e1000 driver for example.) On such systems, - * without this last adjustement, our achievable - * receive throughout using the standard receive - * buffer size was cut by 25%(!!!). - */ - skb->truesize += skb->data_len - RX_COPY_THRESHOLD; + * Truesize is the actual allocation size, even if the + * allocation is only partially used. + */ + skb->truesize += PAGE_SIZE * skb_shinfo(skb)->nr_frags; skb->len += skb->data_len; if (rx->flags & XEN_NETRXF_csum_blank) |