summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2005-05-28bye bye VOP_WHITEOUTPedro Martelletto
2005-05-28Don't use vid/pid from the CIS if vid is 0 when reporting unmatchedJonathan Gray
devices. Several devices have been known to have this behaviour. ok deraadt@
2005-05-28ufs_whiteout() is now pointless, remove it, okay millert@Pedro Martelletto
2005-05-28import regression tests for gzsig.marius eriksen
2005-05-28import gzsig by Dug Song. marius eriksen
a bunch of lumberjack work, zapped support for keynote and SSH1 and added support for SSH2 keys. ok millert@ deraadt@
2005-05-28Dired now prompts in the minibuffer using the current directory nameChad Loder
or the name of the directory of the current buffer's file. OK kjell, jason
2005-05-28"log-all" is gone.Marco Pfatschbacher
ok dhartmei@
2005-05-28don't print the "[ Inserted: uid pid ]" line when -g is used, so theDaniel Hartmeier
regress tests don't have to deal with it (and it's a useless thing to check from there).
2005-05-28Add sasyncd. deraadt@ okHakan Olsson
2005-05-28result not used; ok hoTheo de Raadt
2005-05-28make sure we check the Entry file when we need to.Joris Vink
ok jfb@
2005-05-28Optionally prevent syncing failover node-node SA/SPD info (master side).Hakan Olsson
2005-05-28fixups due to the careful eyes of pedro@marius eriksen
ok pedro@
2005-05-28Set default test shell to /bin/sh.Marco Pfatschbacher
Also destroy lo1000010 at interrupt. ok henning@ mcbride@
2005-05-28Move the va_start()/va_end() pair such that it directly backets the callTodd C. Miller
to vfprintf() like the rest of the *printf functions. This is clearer and makes the error case in asprintf() simpler. From Andrey Matveev.
2005-05-28regression tests for pf checksum.Marco Pfatschbacher
ok mcbride@
2005-05-28make sure to enable socket on resume.Robert Nagy
in order for the enable to work disable the socket on suspend. ok jsg@, mickey@
2005-05-28If the user gets into a minibuffer, they will not know how to hitChad Loder
Ctrl+g to get out of it. Let them hit enter, like in emacs. OK kjell, jason
2005-05-28Step 3 in fixing bioctl commands intermixed with regular io.Marco Peereboom
Enhance performance by using two sleep channels instead of one. All mickey, but I'll take credit for typing it. ok mickey@
2005-05-28Step 2 in fixing bioctl commands intermixed with regular io.Marco Peereboom
Add tsleep/wakeup magic to drain IO when a RAID mgmt command wants to run. Issue RAID mgmt command. Resume IO. ok mickey@
2005-05-27Don't bother issuing a MODE SELECT in cd_set_pa_immed() if we aren'tKenneth R Westerback
going to change anything. From NetBSD. Fixes playing music cd in ATAPI drive in USB enclosure pascoe@ is using. ok pascoe@.
2005-05-27oops, need to get the boundary from the map when allocating gart addressesJason Wright
(fixes borked pciide chipsets)
2005-05-27remove some dead code.marius eriksen
ok pedro@
2005-05-27add missing free on error. thanks to Andrey Matveev.Reyk Floeter
2005-05-27if the user specified a revision to diff, stop caring if the file isJoris Vink
up to date.
2005-05-27Use rtm_fmask instead of rtm_useRyan Thomas McBride
ok marius@ claudio@
2005-05-27Use rtm_fmask instead of rtm_use.Ryan Thomas McBride
ok marius@ claudio@
2005-05-27missing handler for the diff commandJoris Vink
ok jfb@
2005-05-27add vnsubr(9) from NetBSD. add the more detailed vn_lock(9) text to itmarius eriksen
and remove vn_lock. ok tedu@
2005-05-27Hide Hostid and Checksum in pfctl -si output unless the -v flag is used.Ryan Thomas McBride
Prodded by henning@
2005-05-27 - Use the direct map for mapping the PTEs.Artur Grabowski
- Actually allocate the right number of softcs.
2005-05-27Calculate an MD5 checksum over the main pf ruleset.Marco Pfatschbacher
This is the basis for further pfsync improvements, to ensure that pf rules are in sync with the master. "get it in" mcbride@
2005-05-27sync pf table test with new ksh behaviour.Marco Pfatschbacher
ok henning@
2005-05-27Kill __HAVE_NWSCONS harder.Miod Vallat
2005-05-27Back out last commit, it was correct as is.Todd C. Miller
2005-05-27Update, also mention pfsync integrationHakan Olsson
2005-05-27add missing prototypesUwe Stuehler
2005-05-27Make monitor.c use unsigned lengths in messages. Makes this compileChad Loder
with -Wsign-compare. OK and a little testing by hshoexer, OK moritz Now it's anil's turn to do some of this somewhere else
2005-05-27Keep sockaddr in syncpeer struct.Hakan Olsson
2005-05-27- add missing optionsXavier Santolaria
- correctly handle -a and -u flags ok joris jfbeers
2005-05-27add a __mp_lock_try implementation. ok art@Niklas Hallqvist
2005-05-27If _POSIX_SOURCE is defined but _POSIX_C_SOURCE is not justTodd C. Miller
define _POSIX_C_SOURCE to 198808.
2005-05-27step 1 in fixing bioctl commands intermixed with regular io.Marco Peereboom
- Added the magical sequence for polling IO - Created new polling function per hardware generation Lots and lots of constructive yelling mickey@ ok mickey@
2005-05-27allow reception of Jumbo frames by default without having to bumpBrad Smith
the MTU up. tested by marious@, ok mcbride@
2005-05-27filtering on ruleset name is already implemented, document it.Daniel Hartmeier
2005-05-27Dale's lost diff for IRQ sharing, turned on by default.Uwe Stuehler
2005-05-27remove references of VOP_WHITEOUT from the kernel, okay millert@Pedro Martelletto
2005-05-27document new fieldsDaniel Hartmeier
2005-05-27pass UID_MAX/NO_PID when the socket lookup failed, so tcpdump canDaniel Hartmeier
suppress output in this case.
2005-05-27typo in commentUwe Stuehler