Prev: [PATCH v6 3/7] tcp: harmonize tcp_vx_rcv header length assumptions
Next: [PATCH v3 5/7] TCPCT part 2e: accept SYNACK data
From: William Allen Simpson on 11 Mar 2010 07:30 Fix incorrect header prediction flags documentation. Relieve register pressure in (the i386) fast path by accessing skb->len directly, instead of carrying a rarely used len parameter. Eliminate unused len parameters in two other functions. Don't use output calculated tp->tcp_header_len for input decisions. While the output header is usually the same as the input (same options in both directions), that's a poor assumption. In particular, Sack will be different. Newer options are not guaranteed. Moreover, in the fast path, that only saved a shift or two. The other efficiencies in this patch more than make up the difference. Instead, use tp->rx_opt.tstamp_ok to accurately predict header length. Likewise, use tp->rx_opt.tstamp_ok for received MSS calculations. Don't use "sizeof(struct tcphdr) + TCPOLEN_TSTAMP_ALIGNED" to guess that the timestamp is present. This may have been OK in the days with fewer possible options, but various combinations of newer options may yield the same header length. (This bug is in 3 places.) Instead, use tp->rx_opt.saw_tstamp to determine a timestamp is present. There's no need to test buffer length against header length, already checked by tcp_v[4,6]_rcv(). Straighten code for minor efficiency gain. Stand-alone patch, originally developed for TCPCT. Requires: net: tcp_header_len_th and tcp_option_len_th tcp: harmonize tcp_vx_rcv header length assumptions Signed-off-by: William.Allen.Simpson(a)gmail.com Reviewed-by: Andi Kleen <andi(a)firstfloor.org> --- include/linux/tcp.h | 6 ++- include/net/tcp.h | 18 ++++++-- net/ipv4/tcp_input.c | 96 +++++++++++++++++++--------------------------- net/ipv4/tcp_ipv4.c | 4 +- net/ipv4/tcp_minisocks.c | 3 +- net/ipv4/tcp_probe.c | 2 +- net/ipv6/tcp_ipv6.c | 4 +- 7 files changed, 63 insertions(+), 70 deletions(-) |