aboutsummaryrefslogtreecommitdiff
path: root/lib/libc
Commit message (Collapse)AuthorAgeFilesLines
* tzcode: Fix TZ for non-setugid programsDag-Erling Smørgrav44 hours1-19/+42
| | | | | | | | | | | The previous commit had the desired effect for setugid programs, but broke TZ for everyone else. I didn't notice because my test cases swap out /etc/localtime instead of setting TZ, so add a test case that sets TZ. Fixes: b6ea2513f776 ("tzcode: Limit TZ for setugid programs") Reviewed by: cy Differential Revision: https://reviews.freebsd.org/D52108
* libc: preserve errno in gai_strerror()Konstantin Belousov4 days1-0/+4
| | | | | | | | PR: 288931 Reviewed by: emaste, glebius Sponsored by: The FreeBSD Foundation MFC after: 1 week Differential revision: https://reviews.freebsd.org/D52010
* libc: some style in gai_strerror.cKonstantin Belousov4 days1-4/+4
| | | | | | | Reviewed by: emaste, glebius Sponsored by: The FreeBSD Foundation MFC after: 1 week Differential revision: https://reviews.freebsd.org/D52010
* libc: mark ai_errlist as constKonstantin Belousov4 days1-1/+1
| | | | | | | Reviewed by: emaste, glebius Sponsored by: The FreeBSD Foundation MFC after: 1 week Differential revision: https://reviews.freebsd.org/D52010
* libc: convert ai_errlist array to designated initializers syntaxKonstantin Belousov4 days1-15/+15
| | | | | | | Reviewed by: emaste, glebius Sponsored by: The FreeBSD Foundation MFC after: 1 week Differential revision: https://reviews.freebsd.org/D52010
* jemalloc: don't error out on initializer warning in gccWarner Losh6 days1-0/+2
| | | | | | | | gcc doesn't like something about the initializer that comes with jemalloc. Since it's vendor code, make this warning not an error for -Werror purposes. Sponsored by: Netflix
* libc: Fix a typo in a source code commentGordon Bergling7 days1-1/+1
| | | | | | - s/aguments/arguments/ MFC after: 3 days
* jemalloc: Merge from jemalloc 5.3.0 vendor branchWarner Losh8 days8-128/+648
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This brings in a merge from jemalloc 5.3.0. It's almost fully scripted, except: Three conflicts resolved by hand: include/jemalloc/internal/test_hooks.h Use the new name src/extent.c Use the new code src/jemalloc.c Use the new code since je_realloc has moved The script is recorded in FREEBSD-upgrade. The old script did svn commands that were basically a rebase of our changes. This update has a series of diff reduction changes before this. Note: I'd planned on fixing the above three conflicts with commits, but ran out of time when I did this work in January. I got discouraged when jemalloc was EOL'd and didn't pick this back up. I did the above by hand to get this into FreeBSD 15.0 This work is a repeat of the work by Minsoo Choo who did all these changes and created a pull request. Given the importance of jemalloc, I audited these changes by redoing them in this series of commits (and with the script that was checked in). I did this to confince myself and anybody else in doubt that there was no supply chain attack. The diffs between this series of commits and Minsoo's work are minor (though the version skew makes adds some noise). Interested parties can independent audit each step, I hope. I've listed Minsoo as a co-author since without his pull request to test again, this wouldn't have been possible. Thanks to brooks@ for help with getting the jemalloc 3 ABI compat symbols right. Co-authored-by: Minsoo Choo <minsoochoo0122@proton.me> Pull Request: https://github.com/freebsd/freebsd-src/pull/1337 Sponsored by: Netflix
* Revert "Remove Secure RPC DES authentication"Lexi Winter8 days10-76/+1595
| | | | | | This reverts commit 7ac276298b72982189ac1a5b17461936dc00163e. Requested by: kib
* libc: Drop incorrect qsort optimizationDag-Erling Smørgrav9 days3-13/+114
| | | | | | | | | | | | | | | | As pointed out in the PR and the article linked below, the switch to insertion sort in the BSD qsort code is based on a misunderstanding of Knuth's TAOCP and is actually a pessimization. As demonstrated by the added test, it is trivially easy to construct pathological input which results in quadratic runtime. Without that misguided optimization, the same input runs in nearly linearithmic time. https://www.raygard.net/2022/02/26/Re-engineering-a-qsort-part-3 PR: 287089 MFC after: 1 week Reviewed by: imp Differential Revision: https://reviews.freebsd.org/D51907
* kern: fix setgroups(2) and getgroups(2) to match other platformsKyle Evans9 days1-0/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | On most other platforms observed, including OpenBSD, NetBSD, and Linux, these system calls have long since been converted to only touching the supplementary groups of the process. This poses both portability and security concerns in porting software to and from FreeBSD, as this subtle difference is a landmine waiting to happen. Bugs have been discovered even in FreeBSD-local sources, since this behavior is somewhat unintuitive (see, e.g., fix 48fd05999b0f for chroot(8)). Now that the egid is tracked outside of cr_groups in our ucred, convert the syscalls to deal with only supplementary groups. Some remaining stragglers in base that had baked in assumptions about these syscalls are fixed in the process to avoid heartburn in conversion. For relnotes: application developers should audit their use of both setgroups(2) and getgroups(2) for signs that they had assumed the previous FreeBSD behavior of using the first element for the egid. Any calls to setgroups() to clear groups that used a single array of the now or soon-to-be egid can be converted to setgroups(0, NULL) calls to clear the supplementary groups entirely on all FreeBSD versions. Co-authored-by: olce (but bugs are likely mine) Relnotes: yes (see last paragraph) Reviewed by: kib Differential Revision: https://reviews.freebsd.org/D51648
* Remove Secure RPC DES authenticationLexi Winter13 days10-1595/+76
| | | | | | | | | | | | | | | | | | | Following the earlier removal of keyserv, none of this functionality works since it requires keyserv. Remove the relevant symbols from libc's Symbol.map. Leave compatibility symbols for existing applications, but since the functions don't work without keyserv, stub them out to return an error. Remove some private symbols that were only used by keyserv; these don't get compatibility symbols. Remove the documentation for the old functions. Remove rpc.ypupdated since it requires DES authentication. Reviewed by: manu, des, emaste Differential Revision: https://reviews.freebsd.org/D50442
* libc/amd64: rewrite memrchr() scalar impl. to read the string from the backRobert Clausecker2025-08-091-38/+34
| | | | | | | | | | | A very simple implementation as I don't have the patience right now to write a full SWAR kernel. Should still do the trick if you wish to opt out of SSE for some reason. Reported by: Mikael Simonsson <m@mikaelsimonsson.com> Reviewed by: strajabot PR: 288321 MFC after: 1 month
* libc/amd64: rewrite memrchr() baseline impl. to read the string from the backRobert Clausecker2025-08-091-78/+74
| | | | | | | | | | This ensures O(1) behaviour if the character is a constant offset from the end of the string, regardless of how long the string is. Reported by: Mikael Simonsson <m@mikaelsimonsson.com> Reviewed by: benni PR: 288321 MFC after: 1 month
* stdio: Fix bug in integer-parsing FSMDag-Erling Smørgrav2025-08-084-6/+54
| | | | | | | | | | | If we encounter a zero in the havezero state, we should assume octal, just like we would if we encountered any other digit below 8. MFC after: 1 week PR: 288440 Fixes: d9dc1603d6e4 ("libc: Implement N2630.") Reviewed by: mandree Differential Revision: https://reviews.freebsd.org/D51832
* libregex: fix our mapping for \wKyle Evans2025-08-081-0/+1
| | | | | | | | | | A small oversight in our implementation of \w is that it's actually not strictly [[:alnum:]]. According to the GNU documentation, it's actually [[:alnum:]] + underscore. The fix is rather trivial: just add it to our set explicitly, and amend our test set to be sure that _ is actually included. PR: 287396
* libc: Fix style nits in flushlbuf regression testDag-Erling Smørgrav2025-08-071-5/+5
| | | | Sponsored by: Klara, Inc.
* libc: stop trying to create __<syscall> symbolsBrooks Davis2025-08-0634-62/+0
| | | | | | | | | | | | | | | When we switched to an interposing table (commit 8495e8b1e9e1) for cancelation points we stopped having concrete implementations of __<syscall> and instead use __sys_<syscall> and __thr_<syscall>. These entries of the form: __weak_reference(__sys_<syscall>, __<syscall>); seem to be intended to preserve these symbols, but they have no effect as __sys_<syscall> isn't defined in the translation units in question. Reviewed by: imp Differential Revision: https://reviews.freebsd.org/D51668
* libc: gen: refactor execvPe() for readabilityKyle Evans2025-08-032-80/+197
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The current incarnation of execvPe() is a bit messy, and it can be rather difficult to reason about whether we're actually doing the right thing with our errors. We have two cases in which we may enter the loop: 1.) We have a name that has no slashes in it, and we enter the loop normally through our strsep() logic to process $PATH 2.) We have a name with at least one slash, in which case we jump into the middle of the loop then bail after precisely the one iteration if we failed Both paths will exit the loop if we failed, either via jumping to the `done` label to preserve an errno or into the path that clobbers errno. Clobbering errno for case #2 above would seem to be wrong, as we did not actually search -- this would seem to be what POSIX expects, as well, based on expectations of the conformance test suite. Simplify reasoning about the two paths by splitting out an execvPe_prog that does the execve(2) call specifically, and returns based on whether the error would be fatal in a PATH search or not. For the relative/absolute case, we can just ignore the return value and keep errno intact. The search case gets simplified to return early if we hit a fatal error, or continue until the end and clobber errno if we did not find a suitable candidate. Another posix_spawnp() test is added to confirm that we didn't break our EACCES behavior in the process. Reviewed by: des, markj Sponsored by: Klara, Inc. Differential Revision: https://reviews.freebsd.org/D51629
* libc: Drop gets() from symbol mapDag-Erling Smørgrav2025-08-021-1/+0
| | | | | | | | We have an explicit __sym_compat() entry for the legacy implementation. Fixes: 7381dcc9eeb7 Reviewed by: emaste Differential Revision: https://reviews.freebsd.org/D51701
* libc: Deprecate readdir_r()Dag-Erling Smørgrav2025-08-011-0/+2
| | | | | | | | | | This function was never safe to use. We marked it deprecated in the manual page in 2016, and it is marked obsolete in POSIX 2024. Add a linker warning and annotate the prototype. Sponsored by: Klara, Inc. Reviewed by: imp, markj Differential Revision: https://reviews.freebsd.org/D51681
* sysctl.3: Reference kinfo_getvmmap(3), kinfo_getfile(3), and kinfo_getproc(3)Mateusz Piotrowski2025-08-011-1/+6
| | | | | | | | | | It seems to be a good place to help the user discover those useful wrappers. Reviewed by: christos MFC after: 3 days Sponsored by: Klara, Inc. Differential Revision: https://reviews.freebsd.org/D51654
* db/hash.c: Fix flag check in hash_seqBojan Novković2025-08-011-1/+1
| | | | Fixes: 3a686b851f8f
* dbm_nextkey: Always return an error if we've reached the end of the databaseBojan Novković2025-07-313-2/+58
| | | | | | | | | | | | | POSIX.1 states that `dbm_nextkey` must return an invalid key (i.e., `key.dptr == NULL`) after the end of the database was reached. The current implementation of `hash_seq` will incorrectly restart the key sequence after the end of the database is reached. Fix this by checking the "current bucket" index when R_NEXT is passed. Sponsored by: Klara, Inc. Differential Revision: https://reviews.freebsd.org/D51635 Reviewed by: markj
* db/hash.c: Allow O_WRONLY in dbm_openBojan Novković2025-07-306-21/+126
| | | | | | | | | | | | | | The dbm(3) manpage explicitly states that O_WRONLY is not allowed in dbm_open, but a more recent comment in ` __hash_open` suggests otherwise. Furthermore, POSIX.1 allows O_WRONLY in dbm_open and states that the underlying file must be opened for both reading and writing. Fix this by correcting the O_WRONLY check and moving it further into the function to make sure that the original flags are stored in hashp. Sponsored by: Klara, Inc. Reviewed by: markj Differential Revision: https://reviews.freebsd.org/D51514
* memchr.3: Operates on objects, not stringsEd Maste2025-07-251-3/+3
| | | | | | | Suggested by: avg Reviewed by: avg Sponsored by: The FreeBSD Foundation Differential Revision: https://reviews.freebsd.org/D51523
* dbm: Add tests for dbm_openBojan Novković2025-07-252-0/+45
| | | | | | Sponsored by: Klara, Inc. Reviewed by: markj Differential Revision: https://reviews.freebsd.org/D51492
* db/hash.c: Do not return an error when opening a missing database with O_CREATBojan Novković2025-07-251-1/+2
| | | | | | | | | | | | | | | dbm_open currently returns an error when opening a missing database with O_CREAT but creates the request database file. This is caused by a buggy check in __hash_open which attempts to detect a new (or empty) database file, but fails to take the O_CREAT flag into account. Fix this by expanding the check to include O_CREAT. Fixes: edcdc752ecdd Sponsored by: Klara, Inc. Reviewed by: markj Differential Revision: https://reviews.freebsd.org/D51491
* libc: Remove unused pselect() implementation.Dag-Erling Smørgrav2025-07-241-31/+0
| | | | | | | This hasn't been needed since FreeBSD 5.0. Reviewed by: kevans Differential Revision: https://reviews.freebsd.org/D51434
* libc: Disable debugging code in the resolver.Dag-Erling Smørgrav2025-07-248-31/+19
| | | | | Reviewed by: kevans Differential Revision: https://reviews.freebsd.org/D37319
* memchr.3: mention the len argument in the functions descriptionKonstantin Belousov2025-07-221-5/+11
| | | | | | | | | | | Otherwise, its use is only specified in the error case, which is somewhat obscure. Also properly typeset NULL. Reviewed by: pauamma_gundo.com Discussed with: emaste Sponsored by: The FreeBSD Foundation MFC after: 3 days Differential revision: https://reviews.freebsd.org/D51424
* libc: Test time zone change detection.Dag-Erling Smørgrav2025-07-184-1/+292
| | | | | | | | | While here, clean the detection code up a bit. Sponsored by: Klara, Inc. Sponsored by: NetApp, Inc. Reviewed by: markj Differential Revision: https://reviews.freebsd.org/D51343
* lib: Fix calls that naively set F_SETFD.Ricardo Branco2025-07-173-8/+30
| | | | | | | | | | | With the recent inclusion of the FD_CLOFORK and FD_RESOLVE_BENEATH flags, we must avoid clearing them when setting only FD_CLOEXEC. Signed-off-by: Ricardo Branco <rbranco@suse.de> Reviewed by: kib, markj MFC after: 1 month Pull Request: https://github.com/freebsd/freebsd-src/pull/1766
* libc/powerpc64: Fix swapcontext(3)Timothy Pearson2025-07-134-3/+81
| | | | | | | | | | | | | | | | | | | | | On PowerPC platforms a valid link to the Table of Contents (TOC) is required for PLT lookups to function. This TOC pointer is stored in a dedicated register, and is used along with the stack pointer by both C prologue and PLT lookup code. When calling swapcontext() with uc_link != NULL, a PLT lookup to setcontext(3) is attempted from within the _ctx_done context. The exiting process has usually trashed both r1 and r2 at this point, leading to a crash within the PLT lookup before setcontext(2) is reached to restore the linked context. Save and restore r2 as in a regular function. This ensures the subsequent PLT lookup to setcontext(3) succeeds. Signed-off-by: Timothy Pearson <tpearson@raptorengineering.com> MFC after: 1 week Pull Request: https://github.com/freebsd/freebsd-src/pull/1759
* readdir: Fix error check.Dag-Erling Smørgrav2025-07-113-9/+11
| | | | | | | | | | | | | Now that dd_size is unsigned, we need to check if the return value from getdirentries() was negative before assigning it to dd_size. While here, simplify the scandir_error test case slightly, and verify that calling readdir() again after EOF still returns NULL. Fixes: 42e613018da5 Sponsored by: Klara, Inc. Reviewed by: kevans Differential Revision: https://reviews.freebsd.org/D51266
* libc,libthr: Remove __pthread_distribute_static_tlsJessica Clarke2025-07-104-31/+1
| | | | | | | This private API is no longer used by rtld-elf so can be removed. Reviewed by: kib Differential Revision: https://reviews.freebsd.org/D50921
* opendir, readdir, telldir: Use the correct types.Dag-Erling Smørgrav2025-07-095-11/+13
| | | | | | | | Use either size_t or off_t (as appropriate) instead of long. Sponsored by: Klara, Inc. Reviewed by: kevans Differential Revision: https://reviews.freebsd.org/D51210
* getdirentries: Return ENOTDIR if not a directory.Dag-Erling Smørgrav2025-07-091-4/+1
| | | | | | | | | | | | | | | This is both more logical and more useful than EINVAL. While here, also check for VBAD and return EBADF in that case. This can happen if the underlying filesystem got forcibly unmounted after the directory was opened. Previously, this would also have returned EINVAL, which wasn't right but wasn't wrong either; however, ENOTDIR would not be appropriate. MFC after: never Sponsored by: Klara, Inc. Reviewed by: kevans, kib Differential Revision: https://reviews.freebsd.org/D51209
* libc: Finish removing fscandir{,_b}().Dag-Erling Smørgrav2025-07-092-8/+0
| | | | | | | | | | These only existed for a few days before being renamed, so there's no reason to continue to carry compatibility shims for them. Fixes: deeebfdecab5 Sponsored by: Klara, Inc. Reviewed by: kevans Differential Revision: https://reviews.freebsd.org/D50981
* fts: Fix option list indentation.Dag-Erling Smørgrav2025-07-091-1/+1
| | | | | | | Fixes: da2025a0e894 Sponsored by: Klara, Inc. Reviewed by: bcr Differential Revision: https://reviews.freebsd.org/D51208
* wordexp(3): Handle ECHILD from waitpidKenny Levinsen2025-07-082-1/+35
| | | | | | | | | | | | | | If the calling process has used SIG_IGN as handler or set the SA_NOCLDWAIT flag for SIGCHLD, processes will be automatically reaped on exit and calls to waitpid(3) will therefore fail with ECHILD. We waitpid primarily to reap our child so that the caller does not have to worry about it. ECHILD indicates that there is no child to reap, so we can just treat that as a success and move on. Signed-off-by: Kenny Levinsen <kl@kl.wtf> Tested by: Jan Beich Pull Request: https://github.com/freebsd/freebsd-src/pull/1675
* opendir, fdopendir: Add tests, clean up.Dag-Erling Smørgrav2025-07-085-14/+151
| | | | | | | | | | | | * Add test cases for opendir() and fdopendir(). * Drop O_NONBLOCK from opendir(); it was added a long time ago to avoid blocking if given a closed named pipe, but now we use O_DIRECTORY, which ensures that we get ENOTDIR in that case. * While here, remove unused #includes left over from the split. Sponsored by: Klara, Inc. Reviewed by: kevans, markj Differential Revision: https://reviews.freebsd.org/D51126
* libc: only append exterr herald and text if exterr was recordedKonstantin Belousov2025-07-082-2/+2
| | | | | | | | | | __uexterr_format: return empty string on no error, instead of "No error" err(): check for the first symbol of the exterr to decide if there is anything to print. Requested and reviewed by: markj Sponsored by: The FreeBSD Foundation Differential revision: https://reviews.freebsd.org/D51204
* libc: mkostemp(3) should support O_CLOFORKRicardo Branco2025-07-082-4/+5
| | | | | | Reviewed by: kib, markj MFC after: 1 month Pull Request: https://github.com/freebsd/freebsd-src/pull/1758
* err(3): print extended error if availableKonstantin Belousov2025-07-071-12/+39
| | | | | | Reviewed by: asomers, brooks Sponsored by: The FreeBSD Foundation Differential revision: https://reviews.freebsd.org/D51141
* uexterr_format(): simplify output when ext error string is availableKonstantin Belousov2025-07-071-4/+8
| | | | | | | | | | | If the extended error string is provided by kernel, return only the string, which is supposedly enough to identify exact cause of the error. If the string is not provided, print the technically looking gibberish which still allows to identify location with kernel sources. Reviewed by: asomers, brooks Sponsored by: The FreeBSD Foundation Differential revision: https://reviews.freebsd.org/D51141
* Add manpages for O_CLOFORK flag and othersRicardo Branco2025-07-061-2/+13
| | | | | | Reviewed by: kib MFC after: 1 month Pull Request: https://github.com/freebsd/freebsd-src/pull/1698
* kern: Make dup3() support O_CLOFORKRicardo Branco2025-07-061-4/+5
| | | | | | Reviewed by: kib MFC after: 1 month Pull Request: https://github.com/freebsd/freebsd-src/pull/1698
* libc: Add inotify supportMark Johnston2025-07-043-0/+52
| | | | | | | | | | inotify_init() and inotify_init1() are implemented using __specialfd(2). inotify_add_watch() is implemented in terms of inotify_add_watch_at(2). Reviewed by: kib MFC after: 3 months Sponsored by: Klara, Inc. Differential Revision: https://reviews.freebsd.org/D50315
* opendir: Simplify is_unionstack().Dag-Erling Smørgrav2025-07-021-9/+7
| | | | | | Sponsored by: Klara, Inc. Reviewed by: imp Differential Revision: https://reviews.freebsd.org/D51118
generated by cgit v1.2.3 (git 2.25.1) at 2025-08-24 03:10:42 +0000