summaryrefslogtreecommitdiffstats
path: root/sys/netinet/tcp_hostcache.c
Commit message (Collapse)AuthorAgeFilesLines
* White space cleanup for netinet before branch:rwatson2004-08-161-11/+11
| | | | | | | | | | | - Trailing tab/space cleanup - Remove spurious spaces between or before tabs This change avoids touching files that Andre likely has in his working set for PFIL hooks changes for IPFW/DUMMYNET. Approved by: re (scottl) Submitted by: Xin LI <delphij@frontfree.net>
* Remove the UMA_ZONE_NOFREE flag to all uma_zcreate() calls in the IP andandre2004-08-111-1/+1
| | | | | TCP code. This flag would have prevented giving back excessive free slabs to the global pool after a transient peak usage.
* Correct the displayed bandwidth calculation for a readout via sysctl. Theandre2004-08-111-1/+1
| | | | | saved value does not have to be scaled with HZ; it is already in bytes per second. Only the multiply by eight remains to show bits per second (bps).
* Fix a potential race when purging expired hostcache entries.andre2004-04-231-3/+3
| | | | Spotted by: luigi
* Swap destination and source arguments of two bcopy() calls.andre2003-12-021-2/+2
| | | | | | | | | | | | Before committing the initial tcp_hostcache I changed them from memcpy() to conform with FreeBSD style without realizing the difference in argument definition. This fixes hostcache operation for IPv6 (in general and explicitly IPv6 path mtu discovery) and T/TCP (RFC1644). Submitted by: Taku YAMAMOTO <taku@cent.saitama-u.ac.jp> Approved by: re (rwatson)
* Fix an optimization where I made an ifdef'd out section to broad.andre2003-11-281-4/+4
| | | | | | | | | | | When the hostcache bucket limit is reached the last bucket wasn't removed from the bucket row but inserted a few lines later at the bucket row head again. This leads to infinite loop when the same bucket row is accessed the next time for a lookup/insert or purge action. Tested by: imp, Matt Smith Approved by: re (rwatson)
* Introduce tcp_hostcache and remove the tcp specific metrics fromandre2003-11-201-0/+728
the routing table. Move all usage and references in the tcp stack from the routing table metrics to the tcp hostcache. It caches measured parameters of past tcp sessions to provide better initial start values for following connections from or to the same source or destination. Depending on the network parameters to/from the remote host this can lead to significant speedups for new tcp connections after the first one because they inherit and shortcut the learning curve. tcp_hostcache is designed for multiple concurrent access in SMP environments with high contention and is hash indexed by remote ip address. It removes significant locking requirements from the tcp stack with regard to the routing table. Reviewed by: sam (mentor), bms Reviewed by: -net, -current, core@kame.net (IPv6 parts) Approved by: re (scottl)
OpenPOWER on IntegriCloud