Age | Commit message (Collapse) | Author | |
---|---|---|---|
2004-09-08 | typo, wrong type; cd@sentia.nl | Theo de Raadt | |
2004-09-08 | fill out more on console and boot devices | Michael Shalayeff | |
2004-09-08 | routed assumes it alone controls the routing table and tinkers with | Henning Brauer | |
routes it is not supposed to touch trying to aggregate them, which has unwanted effects when other routing daemons (bgpd) are running. note that bgpd behaves and only touches its own routes. workaround by ignoring routes with RTF_PROTO1 set, which bgpd sets for its routes. from a discussion with Arvid Grotting <arvidg@netfonds.no>, who tested a diff of same functionality (basically only formatting and comment different). ok deraadt | |||
2004-09-08 | security fix: | Henning Brauer | |
Apache's mod_rewrite module can be made to write one zero byte in an arbitrary memory position outside of a char array, causing DoS or possibly buffer overflows. The function lookup_map_dbmfile() in modules/mappers/mod_rewrite.c copies data from a DBM file to the char array buf in a _secure_ manner, but it zero-terminates the array afterwards in an _insecure_ manner. If the key that is looked up has an n bytes long value, a zero byte will be written in the memory position n bytes from the start of the char array buf. exploiting would require enabling dbm for mod_rewrite and getting it to use a malicious dbm file. reported by Ulf.Harnhammar.9485@student.uu.se fix by me ok otto, deraadt | |||
2004-09-08 | after openpty() do not close() slave and reopen it. that is just crazy | Theo de Raadt | |
talk. instead, save the descriptor and reuse it. millert ok | |||
2004-09-07 | cleanup multiplex control socket on SIGHUP too, spotted by sturm@ | Damien Miller | |
ok markus@ deraadt@ | |||
2004-09-07 | ignore ntp_sendmsg()s return value in server_dispatch. could result in | Henning Brauer | |
ntpd exiting on sendmsg() failures, which is not desired. | |||
2004-09-07 | 32 -> 64 bit cleanups but no optimizations yet | Per Fogelstrom | |
2004-09-07 | mips is now 64 bits so use a 64 bit config | Per Fogelstrom | |
2004-09-07 | make -netmask work again; ok henning, mcbride, deraadt | Markus Friedl | |
2004-09-07 | Properly put ktrsyscall (and scdebug_foo) into biglock, inspired from NetBSD | Niklas Hallqvist | |
2004-09-07 | Do not incorrectly sign-extend a mode_t value. | Miod Vallat | |
ok deraadt@ | |||
2004-09-06 | set the xflags; form netbsd; millert@ deraadt@ otto@ ok | Michael Shalayeff | |
2004-09-06 | an(4), atw(4) and wi(4) now X-ref each other. Requested by deraadt@ | Todd C. Miller | |
2004-09-06 | rl fits on install media again | Theo de Raadt | |
2004-09-06 | Fix a previous merge error. | Per Fogelstrom | |
ok miod@ and espie@ | |||
2004-09-06 | sync | Theo de Raadt | |
2004-09-06 | put apm (zzz) and apmd on at least some ramdisk media. this lets people | Theo de Raadt | |
wander around akikhabara with a usb cd drive and an openbsd cd, testing for laptops that support apm as god intended. | |||
2004-09-06 | Add t-openbsd64 to handle some libgcc.a build issues (needed functions) | Per Fogelstrom | |
Change configuration to mips64 with default ABI = N64 (enhanced 64 bit ABI) we also support the older O32 and N32 abi for building boot stuff. | |||
2004-09-06 | add INTEL PRO_1000_GT; ok deraadt | Markus Friedl | |
2004-09-06 | sync | Markus Friedl | |
2004-09-06 | add intel PRO_1000_GT; ok deraadt | Markus Friedl | |
2004-09-06 | Make /upgrade mode +x in the ramdisks. | Miod Vallat | |
2004-09-06 | Reset signal mask in the rate limiting case; check for interrupted | Otto Moerbeek | |
waits. Fixes zombies getting reaped late and other ignored signals. Reported by mpech@; fix by millert@ with help from me. Commit requested by deraadt@. ok mpech@ | |||
2004-09-06 | i *SAID* this could go in, now it is days later, I will do it myself | Theo de Raadt | |
2004-09-06 | Jumbo pack of fixes: | Miod Vallat | |
- do not leak memory when polling; - bring LUN support back - Motorola documentation says LUNs are not supported, but it's a SysV/m88k limitation, not a hardware one. - honour request timeout while polling (instead of using a fixed value) - do not program the scsi command length if the hardware knows it from the scsi command group (as advised in the manual) - various minor fixes, especially better error recovery. tested by nick@ and I; ok deraadt@. | |||
2004-09-06 | Correctly account for the %a5 register in function prologues; fixes | Miod Vallat | |
inter-library function calls where the callee would change the GOT register but not restore it when returning to its caller. Helps immensely libpthread, as well as dynamically-linked X11 clients. Fromm gcc 3; tested by matthieu@, nick@ and todd@; ok deraadt@ | |||
2004-09-06 | - Spelling fix. | Brad Smith | |
- Add mention of the fact that timeouts are silently limited to 24 hours. - Document the fact that kqueue will immediately return and not timeout when nevents is 0. From FreeBSD | |||
2004-09-05 | Use ${OSREV} in path names, rather than hardcoded '3.5'. | Kenneth R Westerback | |
Tested by drahn@. ok drahn@ millert@ henning@ | |||
2004-09-04 | password databases -> group databases; | Jason McIntyre | |
from netbsd -r 1.20; ok millert@ | |||
2004-09-03 | sleep instead of busy waiting on ENOBUFS. reduces cpu time. | Ted Unangst | |
pr3900. ok canacar@ deraadt@ matthieu@ millert@ | |||
2004-09-02 | Fix an occurence of "==" that should be "="; no functional change. | Todd C. Miller | |
2004-09-02 | Document IPL_SCHED. | Alexander Yurchenko | |
ok millert@ | |||
2004-09-02 | minor mdoc fixes | Jared Yanovich | |
consistency in wording fix ioctl listing ok jmc | |||
2004-09-02 | propolice -> ProPolice | Brad Smith | |
Mentioned by: Jonathan Gray <khalek at linuxgamers dot net> | |||
2004-09-02 | propolice -> ProPolice | Brad Smith | |
2004-09-02 | add a few more models that had been recently added to intro | Michael Shalayeff | |
2004-09-02 | Prevent mixer from resetting volume back to blead-from-the-ears. | Marco Peereboom | |
From NetBSD via Antoine Jacoutot <ajacoutot at lphp.org> Ok millert@ mickey@ deraadt@ | |||
2004-09-01 | regen for BCM4401-B0 | Todd C. Miller | |
2004-09-01 | Support Broadcom BCM4401-B0; from FreeBSD. OK deraadt@ and henning@ | Todd C. Miller | |
2004-09-01 | Make sure we don't call kernfs_xread() with a negative offset. | Todd C. Miller | |
Problem found by Christer Oberg. OK henning@, deraadt@ | |||
2004-09-01 | Document DST option, and fix grammar nits in TIMEZONE description. | Tom Cosgrove | |
Closes PR 3906. ok jmc@, miod@, millert@ | |||
2004-09-01 | update | Michael Shalayeff | |
2004-09-01 | uppercase `id' and `nvram' for consistency w/ other docs; | Jason McIntyre | |
ok miod@ | |||
2004-08-31 | Buglet in cachectl(), do not use curproc anymore. | Miod Vallat | |
2004-08-31 | changes for gcc3 migration. | Per Fogelstrom | |
2004-08-31 | use the correct Entries path to store in cef_path; | Joris Vink | |
ok jfb | |||
2004-08-31 | Do not crash when there is no user specified in the CVSROOT. Simplified | Jean-Francois Brousseau | |
version of a diff by Joris Vink. | |||
2004-08-31 | check return value of strdup(3); | Joris Vink | |
ok jfb | |||
2004-08-31 | Prevent backup CARP hosts from replying to arp requests, fixes strangeness | Ryan Thomas McBride | |
with some layer-3 switches. From Bill Marquette. ok henning@ mickey@ dhartmei@ and deraadt@ |