summaryrefslogtreecommitdiff
path: root/lib
AgeCommit message (Collapse)Author
2016-02-29delete the kern.emul/KERN_EMUL sysctl bits since there are noChristian Weisgerber
emulations left; ok millert@ deraadt@, jmc@ (man pages)
2016-02-29remove NULL checks for pqueue_free()mmcc
ok doug@
2016-02-28fruncate() returns EFBIG when length exceeds the max file size.natano
ok mpi@
2016-02-27Fix STANDARDS: use the appropriate .St argument and add the informationIngo Schwarze
that setting errno is required by POSIX, but not by ISO C.
2016-02-27If an incomplete character is passed to mbtowc(3), set errno to EILSEQ.Ingo Schwarze
This is unambiguously required both by POSIX and by our own manual page. It also makes a lot of sense because having a function that can fail and that sets errno in some failure modes but does not set errno in other failure modes would be a terrible idea. Such a function would be ridiculously complicated to use. To detect the reason for failure, you would have to: - save errno - reset errno to zero - call the function - inspect the return value to detect failure - inspect errno to decide about the reason for failure - if errno is zero, restore the saved errno That is completely unreasonable, in particular for a seemingly innocous function like mbtowc(3). Next to no programmer would get that right in any real-world program. Note that this bug is very widespread, it also affects NetBSD, FreeBSD, Solaris 11, and glibc. I will also send a note around to the other systems. There may be fallout from programs using the interface incorrectly. If you run into any, please report that to me. OK millert@
2016-02-26Fix negation of POSIX character classes; passed new regress test.Todd C. Miller
OK tb@
2016-02-25refactor option letter parsing into a subfunction, to increase clarityTheo de Raadt
about which options are turned on/off by 's' and 'S' ok tedu
2016-02-24Avoid a possible double-free if the "search" keyword is used multiple times.Eric Faurot
ok jca@ gilles@
2016-02-22Document sendsyslog2(2) in man page.Alexander Bluhm
from Rafael Neves; input jmc@; OK deraadt@
2016-02-19Remove the "error" variable; we no longer fall back to writing toTodd C. Miller
/dev/console so the value of "error" is never read. Also mention that sendsyslog2 can fail due to ENOBUFS in the comment. OK deraadt@
2016-02-17Sync some root certificates with Mozilla's cert store. ok bcook@Stuart Henderson
- Add new root certificates present in Mozilla cert store from CA organizations who are already in cert.pem (AddTrust, Comodo, DigiCert, Entrust, GeoTrust, USERTrust). - Replace Startcom's root with their updated sha256 version present in Mozilla cert store. (They maintained serial# etc so this is still valid for existing signed certificates). - Add two root certificates from CA not previously present: "C=US, O=Network Solutions L.L.C., CN=Network Solutions Certificate Authority" "C=PL, O=Unizeto Sp. z o.o., CN=Certum CA" (the latter used by yandex.ru) We are still listing some certificates that have been removed from Mozilla's store (1024-bit etc) however these cannot be removed until cert validation is improved (we don't currently accept a certificate as valid unless the CA is at the end of a chain).
2016-02-13Dv fix;Jason McIntyre
2016-02-13Fix an obvious cut-and-paste error:Ingo Schwarze
This function doesn't even have an "n" argument, so that non-existent argument clearly cannot restrict the return value.
2016-02-11Document how to use sndio in pledged programs.Tim van der Molen
Input and OK jmc@ ratchov@ semarie@
2016-02-08new sentence, new line;Jason McIntyre
2016-02-08mention mktime, since that is where struct tm is described. ok schwarzeTed Unangst
2016-02-08Align with reality, and with POSIX: For this particular function,Ingo Schwarze
incomplete characters are not an error, do not throw EILSEQ, and are not supposed to. OK millert@
2016-02-07compare pointer to NULL in example codemmcc
2016-02-05be more forceful about not using these.Marc Espie
improvements sthen@, jmc@. okay millert@, jca@ jmc@
2016-02-05Use the correct structure size when allocating the array for nodes.Can Erkin Acar
ok tb@ millert@
2016-02-05Fix err(3) calls after allocation failures in examples.Ingo Schwarze
There is long-standing consensus that err(1, NULL) is the best idiom after failure of malloc(3) and friends. Quirk in the manual noticed by tb@.
2016-02-05avoid an uninitialised value in a return path added in rev 1.5Jonathan Gray
ok tobias@
2016-02-02Get rid of an ugly #ifdef in the middle of el_init() byIngo Schwarze
hiding the system dependencies away in "sys.h" where they belong. No binary change on OpenBSD. Seems reasonable to christos at NetBSD, too.
2016-02-01ansify amd64_iopl()Jeremie Courreges-Anglas
2016-02-01ansify; ok sthen@ kettenis@Jeremie Courreges-Anglas
2016-02-01Sort cert.pem alphabetically, first by organisation, then by CA nameStuart Henderson
(CN if available, otherwise OU). Add a comment identifying the org. Now to get an easy-to-read list of certificates in the file you can use "grep ^[#=] cert.pem". Prepared with https://spacehopper.org/format-pem.20160201. If you would like to verify this commit to ensure that I didn't sneak in any other changes, it will be easier to use the script rather than do it by hand.
2016-01-31Revamp cert.pem certificate information formatting. Skip headers whichStuart Henderson
aren't really useful (the information can be obtained by feeding the cert into "openssl x509 -in filename -text") and add a separator between certs showing the CA's CN or OU (similar to the display format in web browsers). Include both SHA1 and SHA256 fingerprints for all certificates. ok beck@ zhuk@ jung@
2016-01-31remove some whitespace differences with NetBSD; no change with diff -bIngo Schwarze
2016-01-31trivial partial sync to NetBSD, no functional change:Ingo Schwarze
* remove unused ifdef'ed header junk * remove pointless cast of NULL * wrap two excessively long lines
2016-01-30Fifth step in synching with NetBSD:Ingo Schwarze
Delete the silly ptr_t and ioctl_t typedefs and delete some "#ifdef notdef" code from "sys.h". No functional change. This makes hist.h identical to the NetBSD version. It reduces the remaining diff from +1526 -734 to +1430 -592. OK czarkoff@
2016-01-30Fourth step in synching with NetBSD:Ingo Schwarze
KNF: Remove parentheses from return lines. No object change. This makes emacs.c and prompt.c identical to the NetBSD versions. It reduces the remaining diff from +2053 -1261 to +1526 -734. OK czarkoff@
2016-01-30delete "#ifdef notdef" code that is no longer present in NetBSD;Ingo Schwarze
no change in the generated source files
2016-01-30Third step in synching with NetBSD:Ingo Schwarze
* rename fkey_t to funckey_t and el_term_t to el_terminal_t * rename struct editline member el_term to el_terminal * rename many functions in terminal.c from term_*() to terminal_*(), for consistency with the file name and to not look related to <term.h> No functional change. This makes refresh.c and sig.c almost identical to the NetBSD versions. It reduces the remaining diff from +2446 -1805 to +2053 -1420. OK czarkoff@
2016-01-29Second step in synching with NetBSD:Ingo Schwarze
* Rename some types from *key*_t to *keymacro*_t. * Rename struct editline member el_key to el_keymacro. * Rename some functions in keymacro.c from key*() to keymacro*(). This removes the conflict of key_clear(), key_end(), and key_print() with macros in <term.h>. No functional change. This makes keymacro.h identical to the NetBSD version. It reduces the remaining diff from +2640 -1998 to +2446 -1805. OK czarkoff@
2016-01-29Start synching with NetBSD:Ingo Schwarze
Rename the files key.[hc] to keymacro.[hc] and term.[hc] to terminal.[hc]. The change makes sense because "term.h" conflicts with <term.h> and the functions key_clear(), key_end(), and key_print() in "key.h" conflict with macros in <term.h>. No content change yet, no binary change in *.o after "strip -d". This reduces the remaining diff from +4634 -3992 to +2640 -1998. OK czarkoff@, and mmcc@ agrees with the direction.
2016-01-29Therefor -> Therefore (where appropriate)tb
from ray@, ok jmc@
2016-01-28sigsuspend is section 2, not 3.Jason McIntyre
2016-01-28The CSRG declared alarm(3) obsolete 30 years ago (before 4.2BSD),Ingo Schwarze
but it is likely to remain in POSIX for the forseeable future, and there is nothing wrong with using it, so remove the marker. By contrast, POSIX 2008 now declared setitimer(3) obsolescent. While here, remove a few cross references pointing to obsolete or non-standard functions and update the POSIX reference. Triggered by a question from <ChrisBennett at bennettconstruction dot us>. OK millert@ tb@
2016-01-27Replace the malloc spinlock with a mutex. This lock is held over system callsMark Kettenis
which run for many cycles and may even sleep. This leads to other threads spinning for a long time waiting on the lock. Using a mutex means those threads go to sleep and get woken up when the lock is released, which results in a lot less CPU usage. More work is needed to improve the performance of threaded code that suffers from malloc lock contention, but this diff makes ports like Firefox significantly more usable. Tested by many. ok mpi@, guenther@, tedu@, jca@
2016-01-27deprecate SSL_OP_SINGLE_DH_USEBob Beck
ok jsing@
2016-01-26Don't crash dumping malloc stats if malloc_init hasn't been called, noted byOtto Moerbeek
David CARLIER
2016-01-26When encoding fails in fputwc(3), set the error indicator as requiredIngo Schwarze
by POSIX and as FreeBSD, SunOS 10/11, and glibc also do it. Note that an enquiry to the Austin Group led to the conclusion that this change probably violates the C standard: C and POSIX unintentionally conflict. But the POSIX behaviour makes more sense (easier to write correct error handling code for it, and a lower risk that programs miss errors) and is much more widespread, and the Austin Group intends to approach the C committee in order to adjust the C standard. See: http://austingroupbugs.net/view.php?id=1022 While here, do not set errno a second time, wcrtomb(3) already did that, and it is required to do it by the standard. OK millert@ and tedu@, and jca@ no longer objects
2016-01-19replace (void *)0 with NULLmmcc
2016-01-19remove a nop assignment that has been #if 0'd since 1996mmcc
ok millert@
2016-01-18adjust some syscalls in proc & id; from Carlin BinghamTheo de Raadt
2016-01-18Call BIO_sock_init() from tls_init() to ensure sockets are enabled on Windows.Brent Cook
This is of course a no-op on other platforms. Noted by equalsraf from github.
2016-01-09Replace "function call" with "function" when referring to a functionmmcc
(well, a syscall, but this is the established practice) in the general sense. ok jmc@
2016-01-09Make code fit in 80 columns, remove trailing spaces & tabs.Alexandre Ratchov
2016-01-09drop "abort" promise, and make it the default behaviour.Sebastien Marie
The current code has already setted it by default since 1.74 any pledge failure tries to make a coredump (default rules for coredump still applies: so setuid binaries don't create them locally). ok deraadt@
2016-01-09revert manual additions of hid.h now that usbhid.h brings it inJoshua Stein
generated by cgit v1.2.3 (git 2.25.1) at 2024-12-12 02:16:01 +0000