aboutsummaryrefslogtreecommitdiffstats
path: root/net/ipv4/tcp_input.c
Commit message (Expand)AuthorAge
* [TCP]: high_seq parameter removed (all callers use tp->high_seq)Ilpo Järvinen2007-10-15
* [TCP]: Limit processing lost_retrans loop to work-to-do casesIlpo Järvinen2007-10-11
* [TCP]: Fix lost_retrans loop vs fastpath problemsIlpo Järvinen2007-10-11
* [TCP]: No need to re-count fackets_out/sacked_out at RTOIlpo Järvinen2007-10-11
* [TCP]: Extract tcp_match_queue_to_sack from sacktag codeIlpo Järvinen2007-10-11
* [TCP]: Kill almost unused variable pcount from sacktagIlpo Järvinen2007-10-11
* [TCP]: Fix mark_head_lost to ignore R-bit when trying to mark LIlpo Järvinen2007-10-11
* [TCP]: Add bytes_acked (ABC) clearing to FRTO tooIlpo Järvinen2007-10-11
* [TCP]: Separate lost_retrans loop into own functionIlpo Järvinen2007-10-10
* [NET]: sparse warning fixesStephen Hemminger2007-10-10
* [TCP]: "Annotate" another fackets_out state resetIlpo Järvinen2007-10-10
* [TCP]: Wrap-safed reordering detection FRTO checkIlpo Järvinen2007-10-10
* [TCP]: Update comment of SACK block validatorIlpo Järvinen2007-10-10
* [TCP]: fix comments that got messed up during code moveIlpo Järvinen2007-10-10
* [TCP] MIB: Count FRTO's successfully detected spurious RTOsIlpo Järvinen2007-10-10
* [TCP]: Reordered ACK's (old) SACKs not included to discarded MIBIlpo Järvinen2007-10-10
* [TCP]: Avoid clearing sacktag hint in trivial situationsIlpo Järvinen2007-10-10
* [TCP]: Enable SACK enhanced FRTO (RFC4138) by defaultIlpo Järvinen2007-10-10
* [TCP] FRTO: Improve interoperability with other undo_marker usersIlpo Järvinen2007-10-10
* [TCP]: Cleanup tcp_tso_acked and tcp_clean_rtx_queueIlpo Järvinen2007-10-10
* [TCP]: Move accounting from tso_acked to clean_rtx_queueIlpo Järvinen2007-10-10
* [TCP]: clear_all_retrans_hints prefixed by tcp_Ilpo Järvinen2007-10-10
* [TCP]: Make fackets_out accurateIlpo Järvinen2007-10-10
* [TCP] MIB: Add counters for discarded SACK blocksIlpo Järvinen2007-10-10
* [TCP]: Discard fuzzy SACK blocksIlpo Järvinen2007-10-10
* [TCP]: Rename tcp_ack_packets_out -> tcp_rearm_rtoIlpo Järvinen2007-10-10
* [TCP]: Remove unnecessary wrapper tcp_packets_out_decIlpo Järvinen2007-10-10
* [TCP]: Move sack_ok access to obviously named funcs & cleanupIlpo Järvinen2007-10-10
* [TCP]: Reduce sacked_out with reno when purging write_queueIlpo Järvinen2007-10-10
* [TCP]: Keep state in Disorder also if only lost_out > 0Ilpo Järvinen2007-10-10
* [TCP]: Restore over-zealous tcp_sync_left_out-like removalsIlpo Järvinen2007-10-10
* [TCP]: Left out sync->verify (the new meaning of it) & definifyIlpo Järvinen2007-10-10
* [TCP]: Add tcp_left_out(tp) "back" to get cleaner looking linesIlpo Järvinen2007-10-10
* [TCP]: Tighten tcp_sock's belt, drop left_outIlpo Järvinen2007-10-10
* [TCP]: Move code from tcp_ecn.h to tcp*.c and tcp.h & remove itIlpo Järvinen2007-10-10
* [TCP] FRTO: remove unnecessary fackets/sacked_out recountingIlpo Järvinen2007-10-10
* [TCP]: Move Reno SACKed_out counter functions earlierIlpo Järvinen2007-10-10
* [TCP]: Extract DSACK detection code from tcp_sacktag_write_queue().David S. Miller2007-10-10
* [TCP]: Rexmit hint must be cleared instead of setting itIlpo Järvinen2007-10-10
* [TCP]: Extracted rexmit hint clearing from the LOST marking codeIlpo Järvinen2007-10-10
* [TCP]: Add highest_sack seqno, points to globally highest SACKIlpo Järvinen2007-10-10
* [TCP]: Fix fastpath_cnt_hint when GSO skb is partially ACKedIlpo Järvinen2007-10-08
* [TCP]: 'dst' can be NULL in tcp_rto_min()David S. Miller2007-08-31
* [TCP]: Allow minimum RTO to be configurable via routing metrics.David S. Miller2007-08-31
* [TCP]: Describe tcp_init_cwnd() thoroughly in a comment.David S. Miller2007-08-26
* [TCP]: DSACK signals data receival, be conservativeIlpo Järvinen2007-08-02
* [TCP]: Also handle snd_una changes in tcp_cwnd_downIlpo Järvinen2007-08-02
* [TCP]: Bidir flow must not disregard SACK blocks for lost markingIlpo Järvinen2007-07-31
* [TCP]: Fix ratehalving with bidirectional flowsIlpo Järvinen2007-07-31
* [TCP]: congestion control API pass RTT in microsecondsStephen Hemminger2007-07-31