Age | Commit message (Collapse) | Author | |
---|---|---|---|
2018-06-28 | calling memset() after calloc() is redudant, since | Gleydson Soares | |
calloc() already filled all the memory block to 0. OK deraadt@ cheloha@ millert@ | |||
2018-04-10 | Schedule events against the monotonic clock. | cheloha | |
So that events fire punctually, even if the system clock is changed. ok jca@ | |||
2017-08-31 | enum can be signed or unsigned depending on the compiler; so make sure we | Otto Moerbeek | |
cast to unsigned and test for > X to avoid warnings that comparing an unsigned < 0 make no sense; ok deraadt@ | |||
2017-07-03 | no need to generate y.tab.h if nothing uses it, set YFLAGS to nothing | Marc Espie | |
instead of CLEANFILES += y.tab.h okay millert@ | |||
2017-05-21 | A few more freezero() uses | Theo de Raadt | |
ok yasuoka mikeb | |||
2017-04-18 | use freezero() | Theo de Raadt | |
2017-04-10 | Check early if the mandatory carp "interface" was configured. | Reyk Floeter | |
This allows the new -n config check to work correctly. OK jsg@ deraadt@ | |||
2017-04-09 | Fix multiple cases of reading past the end of a buffer in the sasyncd(8) | Jonathan Gray | |
config parser found with afl. feedback and ok millert@ ok deraadt@ | |||
2017-04-04 | add a config test flag to sasyncd | Jonathan Gray | |
ok reyk@ deraadt@ jmc@ | |||
2017-04-04 | Add missing newline in fprintf error message. | Reyk Floeter | |
OK jsg@ | |||
2016-08-27 | Pull in <sys/select.h> for fd_set | Philip Guenther | |
ok deraadt@ | |||
2016-08-27 | Use %zu for size_t arguments | Philip Guenther | |
2016-07-18 | add format attributes to log functions and fix some errors | Sebastian Benoit | |
ok claudio@ florian@ | |||
2015-12-12 | Remove NULL-checks before free(). | mmcc | |
2015-11-12 | an invalid carp link state is not a fatal error, since a while newly created | Otto Moerbeek | |
carp interfaces have this state for a while. ok mpi@ phessler@ | |||
2015-10-18 | Use explicit_bzero() when the memory is freed directly afterward. | mmcc | |
ok deraadt@ | |||
2015-08-20 | stdlib.h is in scope; do not cast malloc/calloc/realloc* | Theo de Raadt | |
ok millert krw | |||
2015-01-16 | Replace <sys/param.h> with <limits.h> and other less dirty headers where | Theo de Raadt | |
possible. Annotate <sys/param.h> lines with their current reasons. Switch to PATH_MAX, NGROUPS_MAX, HOST_NAME_MAX+1, LOGIN_NAME_MAX, etc. Change MIN() and MAX() to local definitions of MINIMUM() and MAXIMUM() where sensible to avoid pulling in the pollution. These are the files confirmed through binary verification. ok guenther, millert, doug (helped with the verification protocol) | |||
2014-10-26 | fix use of uninitialized variable in error path | Charles Longeau | |
ok deraadt@ | |||
2014-10-16 | simple reallocarray() | Theo de Raadt | |
2014-07-04 | malloc+memset -> calloc | Philip Guenther | |
from Benjamin Baier (programmer (at) netzbasis.de) | |||
2014-06-29 | If you want <sys/queue.h>, you need to include it. Don't assume that | Theo de Raadt | |
<sys/sysctl.h> will pull in the universe (I am working on breaking that assumption in a gentle fashion) | |||
2013-03-11 | handle ECONNABORTED errors from accept(). In many code blocks they can be | Theo de Raadt | |
ignored silently and without aborting, much like EINTR and EWOULDBLOCK are. ok's from various maintainers of these directories... | |||
2012-12-21 | valid port interval 1 through 65535, not 1 - 65534 | Gleydson Soares | |
check interval with yacc/$5 variable instead of cfgstate.listen_port that is an in_port_t type, for avoid warnings, due to limited range of data type; use IPPORT_HILASTAUTO for more consistence (input from millert@) "yeah, I like that. ok" henning@ | |||
2012-09-15 | Don't depend on sa_family_t being typedefed by <sys/types.h>; pull in | Philip Guenthe | |
<netinet/in.h> to reliably get it ok matthew@, millert@, deraadt@ | |||
2012-09-12 | Fix a race condition which would cause segfault due to the kernel | Christiano F. Haesbaert | |
sending less (or more) data than expected. We do a sysctl to know how much data should be read, and then we try to read that amount, but there is a window between this two calls that things can change, this makes sure we have an "atomic view" of data. From Patrick Wildt, tested with over 7000 SAs, thanks. ok deraadt | |||
2012-09-04 | fix log_err() calls (from Patrick Wildt) and remove extra space. | Okan Demirmen | |
ok haesbaert@ | |||
2012-08-17 | Don't need this anymore, we have it in bsd.own.mk now. | Christiano F. Haesbaert | |
2012-04-15 | fix a leak | Jonathan Gray | |
ok krw@ | |||
2012-04-14 | Kill Boehms GC dead. | Christiano F. Haesbaert | |
ok mikeb@ sure deraadt@ | |||
2012-04-14 | read returns ssize_t not int. | Christiano F. Haesbaert | |
ok mikeb@ | |||
2012-04-14 | Kill useless memset. | Christiano F. Haesbaert | |
ok mikeb@ "looks good" deraadt@ | |||
2012-04-14 | Don't return from main with 2 uppon issuing double -c option, print | Christiano F. Haesbaert | |
proper usage() message. ok mikeb@ "looks good" deraadt@ | |||
2012-04-14 | Add CDIAGFLAGS infrastructure. | Christiano F. Haesbaert | |
Fix signed vs unsigned and dup symbol shadow. ok mikeb@ "looks good" deraadt@ | |||
2012-04-02 | Fix memory leak in error path, originally based on a diff from jsg@. | Todd C. Miller | |
OK deraadt@ | |||
2011-06-23 | Use a common text explaining how the various configuration parsers using | Stuart Henderson | |
the standard OpenBSD-style parse.y handle continuing lines with backslashes, paying particular attention to how comments are handled (which can cause nasty side-effects if you're not expecting it). Most wording from jmc@, with suggestions from fgsch@, marc@, Richard Toohey, patrick keshishian and Florian Obser, ok jmc@. | |||
2010-06-30 | clarify why carp demotion is necessary, and who is doing it. | Kjell Wooding | |
ok reyk@, sthen@ | |||
2010-06-29 | Fix an obviously wrong comment. ok reyk@ | Kjell Wooding | |
2010-06-29 | Having more than 2 log levels is silly. Use only 3 verbosity levels: | Kjell Wooding | |
None (), Important (-v), All (-vv) ok deraadt@ | |||
2010-06-16 | Add support to use sasyncd(8) with iked(8) instead of isakmpd(8). The | Reyk Floeter | |
new config option "control isakmpd|iked|all|none" in sasyncd.conf(5) is used to set the mode; the default is to control isakmpd like before. When controlling iked, sasyncd connects to the daemon using the imsg socket and sends imsgs - this is easy because the imsg framework has been imported into libutil. iked(8) already includes the important bits to work with sasyncd(8) (the active/passive mode option in ikectl). manpage bits ok jsg@ ok jsg@ | |||
2009-11-11 | memory leak found by parfait | Theo de Raadt | |
2009-11-11 | spacing | Theo de Raadt | |
2009-06-26 | use standard safe C | Theo de Raadt | |
2009-06-25 | sasyncd isn't interested in messages on the routing socket other than | Stuart Henderson | |
RTM_IFINFO to monitor the state of carp(4) interfaces; programme the filter on the socket accordingly so it doesn't have to handle other message types. ok henning@ | |||
2008-04-04 | document the bug that sasyncd peers should share the same | Otto Moerbeek | |
platform. it would be really nice if somebody would fix this. ok mpf@ todd@ jmc@ | |||
2008-03-17 | sync the synopsis and usage; "usage:" is lowercase | Igor Sobrado | |
ok jmc@ | |||
2007-09-02 | use calloc() to avoid malloc(n * m) overflows; checked by djm canacar jsg | Theo de Raadt | |
2007-05-31 | convert to new .Dd format; | Jason McIntyre | |
2007-05-17 | Check getpwnam() return value for NULL before dereferencing it. | Moritz Jodeit | |
ok ray@ millert@ | |||
2007-02-26 | Zero out struct before using, not after. | Ray Lai | |
From <sthen at symphytum dot spacehopper dot org>, PR 5388. OK hshoexer@ and mpf@. |