summaryrefslogtreecommitdiff
path: root/usr.bin
AgeCommit message (Collapse)Author
2011-08-16recommit, now that release is done (handle sysv if bsd is incorrect, mayMarc Espie
make more extensive plans later)
2011-08-16Add a -r flag to switch-client to toggle the client read-only flag. FromNicholas Marriott
Johan Commelin.
2011-08-16Add up/down/left/right keys in vi edit mode. From "livibetter" SF bugNicholas Marriott
3385651.
2011-08-16Use key_bindings_remove for unbind-key -a to allow it to work from keyNicholas Marriott
bindings. From "miaout17" SF bug 3392063.
2011-08-16Correctly skip existing numbers when generating the name for a newNicholas Marriott
session.
2011-08-16Make this build for hppa64.Mark Kettenis
2011-08-07typo, fix from Laurent GautrotDarren Tucker
2011-08-03save/restore tty state on SIGTSTP/SIGCONT. ok deraadt@ nicm@Otto Moerbeek
2011-08-02typo in commentDamien Miller
2011-08-02crank now, release laterDamien Miller
2011-08-02Add new SHA256 and SHA512 based HMAC modes fromDamien Miller
http://www.ietf.org/id/draft-dbider-sha2-mac-for-ssh-02.txt feedback and ok markus@
2011-08-01prevent post-auth resource exhaustion (int overflow leading to 4GB malloc);Markus Friedl
report Adam Zabrock; ok djm@, deraadt@
2011-08-01Change Gnu to GNU in man page (also, while here, in README and in a comment)lum
ok jmc@
2011-07-31Actually, by merging individual patches, we are already in syncIngo Schwarze
with release 1.11.4. Correct this for release.
2011-07-31Workaround to prevent misrendering of \*(-- as "O-" in pod2man(1)-Ingo Schwarze
generated manuals; this fixes more than 500 manuals in base alone. As a real fix, .tr will be supported after unlock. OK kristaps@ to put in the workaround for now
2011-07-31Mention that make(1) is the program using .depend files.Ingo Schwarze
Useful because their format and usage is also documented there. feedback and ok jmc@
2011-07-30deemed unsafe by theo, will wait after 5.0Marc Espie
2011-07-30fix error in previous;Jason McIntyre
2011-07-30Extend the mode-mouse option to add a third choice which means the mouseNicholas Marriott
does not enter copy mode. Patch from SF bug 3374493. In future the mode-mouse option is likely to die and be broken into several smaller options.
2011-07-30Do not require a client here, or pipe-pane will not work from theNicholas Marriott
command line.
2011-07-29fail open(2) with EPERM rather than SIGKILLing the whole process. libcDamien Miller
will call open() to do strerror() when NLS is enabled; feedback and ok markus@
2011-07-29systemV trumps our modifiers.Marc Espie
fixes a case encountered by jeremy@ full fix not prudent at this time. okay millert@, otto@
2011-07-29document vi/ex regular expressions, and where they differ from thoseJason McIntyre
documented in re_format(7); diff from alexis fouilhe, and verified/tweaked by sobrado; ok millert otto
2011-07-28Resolve sysctl numbers, original diff from nicm@, man page bits fromOtto Moerbeek
guenther@; ok guenther@ millert@
2011-07-28Adjust for machdep.consdev -> kern.consdev sysctl change. This should restoreMark Kettenis
the functionality to figure out the real console device when /dev/console is used. ok millert@, deraadt@
2011-07-26Backout previous, naddy@ found the following regression:Ingo Schwarze
When the input does not end in a trailing newline character and there is an empty match at the end, the new code adds a spurious '\0' character. I have a fix, but otto@ prefers backout and full re-evaluation after release.
2011-07-25- Use strtof instead of casting strtod back to float (apprentice.c),Martynas Venckus
- Don't reimplement strndup (softmagic.c). OK millert@.
2011-07-25Sort options alphabetically, from Tiago Cunha.Nicholas Marriott
2011-07-24Rewrite the main loop of the "sed s/..." command, shortening it by tenIngo Schwarze
lines and simplifying it by removing the switch statement implementing /g, /1, and /2 separately and repetitively. The idea to make the loop control variable slen, i.e. the length of the string remaining to be processed, signed, and stay in the loop even when slen == 0 (i.e. at the end of the string), lifted from FreeBSD by otto@. On i386, process.o shrinks by 440 bytes, and the sed binary by 23 bytes. This fixes multiple aspects of the replacement of multiple (/g) or specific (e.g. /2) instances of zero-length matches, both with BREs and EREs, both with and without a trailing newline character on the input. Feedback and OK otto@.
2011-07-24AT&T recently released a new version of UNIX (System III).Ingo Schwarze
This will soon be needed for the HISTORY of string(3).
2011-07-22Use dirfd(dirp) instead of dirp->dd_fd.Matthew Dempsky
No binary change.
2011-07-22Speed up igetnext() by passing the record we want to parse in toTodd C. Miller
getent() as toprec. Makesd cap_mkdb faster when building termino.db. OK nicm@
2011-07-19Add fancy kdump support for the openat(2) system calls.Matthew Dempsky
ok otto@
2011-07-17Fix an integer overflow for very long lines by replacing the datatype of 2 ↵aschrijver
offsets from int to regoff_t. Bail if the given size_t line length doesn't fit into the new regoff_t. "I don't think you will ever be able to get a string longer than SSIZE_MAX into memory, but that looks good." tedu@ "Agreed" otto@ regoff_t suggested by otto@
2011-07-17Fix spacingaschrijver
OK otto@
2011-07-17handle files produced by ktrace -a better by making a distinctionOtto Moerbeek
between default and current emulation; ok guenther@
2011-07-17since we're treating native emuls different (more fancy) thanOtto Moerbeek
non-native we should do a better job of tracking the emul corresponding to a pid; ok guenther@; also tested by pirofti@
2011-07-15When killing lines using ^U <n> ^K, count the characters exactly (forTheo de Raadt
both forwards and backwards cases, though the forward case is better tested. This is required because the actual character deleter function (ldelete) requires an exact count. If it runs short, it will not put the deletion into the kill buffer. This is complicated by how mg internals consider newline's as counted characters even though they do not occur in the buffers... and then there is the no newline at EOF fiasco....
2011-07-14spacing, no binary change.Igor Sobrado
2011-07-14str is an optional argument to -t; if no argument is given, thisIgor Sobrado
option is ignored (for compatibility reasons). ok jmc@
2011-07-14Import mandocdb(8) from bsd.lv, coded by kristaps@, for in-tree development.Ingo Schwarze
Will need adjustments to user interface and functionality, and will be linked into the mandoc(1) binary and hardlinked from /usr/libexec/mandocdb to /usr/bin/mandoc when ready for general consumption. The ultimate goal is to call it from OpenBSD::Makewhatis to replace OpenBSD::Makewhatis::Unformated with enhanced functionality.
2011-07-14Revert build system complication, for two reasons:Ingo Schwarze
(1) Simply linking the future utilities (mandoc, mandocdb, apropos, whatis) into a common binary inspecting its process name to find out what it should do (as suggested by deraadt@) does not require the large code changes i feared at first, but can be done in a few lines. (2) As reported by dhill@, we already had the first build system problem before even enabling this: Building without "make obj", the mandoc binary name clashes with the new mandoc subdirectory, breaking the build. Please use "cvs up -dP" when updating!
2011-07-12argument to -t is optional.Igor Sobrado
ok jmc@
2011-07-12print out the amount of memory used for cache. ok ottoTed Unangst
2011-07-11the matchall magic shortcut requires we set c=1 to print nowTed Unangst
fixes libpqxx build
2011-07-11Only complain about zero-sized structures/unions when in strictMartynas Venckus
ANSI C mode. Based on a diff from Theo.
2011-07-10Rename O_DIRECTORY to O_TMP_DIRECTORY to avoid a namespace collisionTodd C. Miller
with sys/fcntl.h. OK deraadt@
2011-07-10Make -m 0 work as expected.Otto Moerbeek
2011-07-10The mandoc directory will soon grow subdirs mandocdb/ and apropos/Ingo Schwarze
and these new utilities will use the mandoc libraries, so prepare the build system to build the parsing and utility libraries separately from the main program and formatters. Not yet linked to the build. No objections from deraadt@ on adding directories for this purpose.
2011-07-09Document default write speed of tao subcommand.Stefan Sperling
generated by cgit v1.2.3 (git 2.25.1) at 2025-01-16 23:59:11 +0000