summaryrefslogtreecommitdiffstats
path: root/kernel
diff options
context:
space:
mode:
authorEric Dumazet <edumazet@google.com>2015-09-09 21:55:07 -0700
committerDavid S. Miller <davem@davemloft.net>2015-09-10 10:58:33 -0700
commit30927520dbae297182990bb21d08762bcc35ce1d (patch)
tree05082e3d589a52aea0384feafb9d60ae4195ce3a /kernel
parent05c5a46d71f621df620fbabbd7758ee1b44575ad (diff)
downloadlinux-30927520dbae297182990bb21d08762bcc35ce1d.tar.bz2
tcp_cubic: better follow cubic curve after idle period
Jana Iyengar found an interesting issue on CUBIC : The epoch is only updated/reset initially and when experiencing losses. The delta "t" of now - epoch_start can be arbitrary large after app idle as well as the bic_target. Consequentially the slope (inverse of ca->cnt) would be really large, and eventually ca->cnt would be lower-bounded in the end to 2 to have delayed-ACK slow-start behavior. This particularly shows up when slow_start_after_idle is disabled as a dangerous cwnd inflation (1.5 x RTT) after few seconds of idle time. Jana initial fix was to reset epoch_start if app limited, but Neal pointed out it would ask the CUBIC algorithm to recalculate the curve so that we again start growing steeply upward from where cwnd is now (as CUBIC does just after a loss). Ideally we'd want the cwnd growth curve to be the same shape, just shifted later in time by the amount of the idle period. Reported-by: Jana Iyengar <jri@google.com> Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Yuchung Cheng <ycheng@google.com> Signed-off-by: Neal Cardwell <ncardwell@google.com> Cc: Stephen Hemminger <stephen@networkplumber.org> Cc: Sangtae Ha <sangtae.ha@gmail.com> Cc: Lawrence Brakmo <lawrence@brakmo.org> Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'kernel')
0 files changed, 0 insertions, 0 deletions