linux-toradex.git/net/ipv4/tcp_bic.c, branch v3.0.97 Linux kernel for Apalis and Colibri modules tcp: mark tcp_congestion_ops read_mostly 2011-03-10T08:40:17+00:00 Stephen Hemminger shemminger@vyatta.com 2011-03-10T08:40:17+00:00 a252bebe22155313ccdadc20b79f67a239dc9ecb Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net> 
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
tcp: add helper for AI algorithm 2009-03-02T11:00:15+00:00 Ilpo Järvinen ilpo.jarvinen@helsinki.fi 2009-02-28T04:44:37+00:00 758ce5c8d11d6fc57fe5f1dbc237aa8ff6386eac It seems that implementation in yeah was inconsistent to what other did as it would increase cwnd one ack earlier than the others do. Size benefits: bictcp_cong_avoid | -36 tcp_cong_avoid_ai | +52 bictcp_cong_avoid | -34 tcp_scalable_cong_avoid | -36 tcp_veno_cong_avoid | -12 tcp_yeah_cong_avoid | -38 = -104 bytes total Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi> Signed-off-by: David S. Miller <davem@davemloft.net> 
It seems that implementation in yeah was inconsistent to what
other did as it would increase cwnd one ack earlier than the
others do.

Size benefits:

  bictcp_cong_avoid |  -36
  tcp_cong_avoid_ai |  +52
  bictcp_cong_avoid |  -34
  tcp_scalable_cong_avoid |  -36
  tcp_veno_cong_avoid |  -12
  tcp_yeah_cong_avoid |  -38

= -104 bytes total

Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
[TCP]: BIC web page link is corrected. 2008-02-29T06:14:32+00:00 Sangtae Ha sha2@ncsu.edu 2008-02-29T06:14:32+00:00 0bc8c7bf9e7440500fe04f95d6cddc58ea023cd2 Signed-off-by: Sangtae Ha <sha2@ncsu.edu> Signed-off-by: David S. Miller <davem@davemloft.net> 
Signed-off-by: Sangtae Ha <sha2@ncsu.edu>
Signed-off-by: David S. Miller <davem@davemloft.net>
[TCP]: Cong.ctrl modules: remove unused good_ack from cong_avoid 2008-01-28T22:55:41+00:00 Ilpo Järvinen ilpo.jarvinen@helsinki.fi 2007-12-01T22:47:59+00:00 c3a05c6050a339c92e49fae0ba77dbba0d41fd99 Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net> 
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
[TCP]: Remove num_acked>0 checks from cong.ctrl mods pkts_acked 2007-10-10T23:47:55+00:00 Ilpo Järvinen ilpo.jarvinen@helsinki.fi 2007-05-31T07:16:47+00:00 35e8694198ba94b62df8aa35fa6e52a1cfb86df2 There is no need for such check in pkts_acked because the callback is not invoked unless at least one segment got fully ACKed (i.e., the snd_una moved past skb's end_seq) by the cumulative ACK's snd_una advancement. Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi> Signed-off-by: David S. Miller <davem@davemloft.net> 
There is no need for such check in pkts_acked because the
callback is not invoked unless at least one segment got fully
ACKed (i.e., the snd_una moved past skb's end_seq) by the
cumulative ACK's snd_una advancement.

Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
[TCP]: congestion control API pass RTT in microseconds 2007-07-31T09:27:57+00:00 Stephen Hemminger shemminger@linux-foundation.org 2007-07-26T06:49:34+00:00 30cfd0baf0a0c4329fff1ef4b622919297969ec8 This patch changes the API for the callback that is done after an ACK is received. It solves a couple of issues: * Some congestion controls want higher resolution value of RTT (controlled by TCP_CONG_RTT_SAMPLE flag). These don't really want a ktime, but all compute a RTT in microseconds. * Other congestion control could use RTT at jiffies resolution. To keep API consistent the units should be the same for both cases, just the resolution should change. Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org> Signed-off-by: David S. Miller <davem@davemloft.net> 
This patch changes the API for the callback that is done after an ACK is
received. It solves a couple of issues:

  * Some congestion controls want higher resolution value of RTT
    (controlled by TCP_CONG_RTT_SAMPLE flag). These don't really want a ktime, but
    all compute a RTT in microseconds.

  * Other congestion control could use RTT at jiffies resolution.

To keep API consistent the units should be the same for both cases, just the
resolution should change.

Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
[TCP]: remove unused argument to cong_avoid op 2007-07-18T08:46:58+00:00 Stephen Hemminger shemminger@linux-foundation.org 2007-07-17T01:35:52+00:00 16751347a060a10c09b11593bb179fd5b0240c04 None of the existing TCP congestion controls use the rtt value pased in the ca_ops->cong_avoid interface. Which is lucky because seq_rtt could have been -1 when handling a duplicate ack. Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org> Signed-off-by: David S. Miller <davem@davemloft.net> 
None of the existing TCP congestion controls use the rtt value pased
in the ca_ops->cong_avoid interface.  Which is lucky because seq_rtt
could have been -1 when handling a duplicate ack.

Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
[TCP]: Set initial_ssthresh default to zero in Cubic and BIC. 2007-06-13T08:03:53+00:00 David S. Miller davem@sunset.davemloft.net 2007-06-13T08:03:53+00:00 66e1e3b20cbbf99da63e6c1af0fc6d39c2ed099a Because of the current default of 100, Cubic and BIC perform very poorly compared to standard Reno. In the worst case, this change makes Cubic and BIC as aggressive as Reno. So this change should be very safe. Signed-off-by: David S. Miller <davem@davemloft.net> 
Because of the current default of 100, Cubic and BIC perform very
poorly compared to standard Reno.

In the worst case, this change makes Cubic and BIC as aggressive as
Reno.  So this change should be very safe.

Signed-off-by: David S. Miller <davem@davemloft.net>
[TCP]: Congestion control API update. 2007-04-26T05:29:45+00:00 Stephen Hemminger shemminger@linux-foundation.org 2007-04-24T05:26:16+00:00 164891aadf1721fca4dce473bb0e0998181537c6 Do some simple changes to make congestion control API faster/cleaner. * use ktime_t rather than timeval * merge rtt sampling into existing ack callback this means one indirect call versus two per ack. * use flags bits to store options/settings Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org> Signed-off-by: David S. Miller <davem@davemloft.net> 
Do some simple changes to make congestion control API faster/cleaner.
* use ktime_t rather than timeval
* merge rtt sampling into existing ack callback
  this means one indirect call versus two per ack.
* use flags bits to store options/settings

Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
[TCP] tcp_bic: use BUILD_BUG_ON 2006-09-22T22:18:04+00:00 Alexey Dobriyan adobriyan@gmail.com 2006-08-25T07:38:03+00:00 65e3d72654d9a33cdccd5c19777a5515ae9dd37d Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net> 
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>