Pull up following revision(s) (requested by riastradh in ticket #1384):
external/bsd/file/dist/src/fsmagic.c: revision 1.19
external/bsd/file/include/config.h: revision 1.22
tests/dev/fss/t_fss.sh: revision 1.8
lib/libutil/stat_flags.c: revision 1.4
lib/libutil/stat_flags.c: revision 1.5
external/bsd/file/dist/configure.ac: revision 1.24
(all via patch)
string_to_flags(3): Support `snap' (SF_SNAPSHOT).
PR bin/60380: find(1) can't find snapshots
string_to_flags(3): Put `snap' under #ifdef SF_SNAPSHOT.
Same as in flags_to_string(3). Not really sure why, but let's keep it
consistent so there's only one question to ponder and not two.
[13 lines not shown]
Pull up following revision(s) (requested by riastradh in ticket #538):
external/bsd/file/dist/src/fsmagic.c: revision 1.19
external/bsd/file/include/config.h: revision 1.22
tests/dev/fss/t_fss.sh: revision 1.8
lib/libutil/stat_flags.c: revision 1.4
lib/libutil/stat_flags.c: revision 1.5
external/bsd/file/dist/configure.ac: revision 1.24
(all via patch)
string_to_flags(3): Support `snap' (SF_SNAPSHOT).
PR bin/60380: find(1) can't find snapshots
string_to_flags(3): Put `snap' under #ifdef SF_SNAPSHOT.
Same as in flags_to_string(3). Not really sure why, but let's keep it
consistent so there's only one question to ponder and not two.
[13 lines not shown]
Pull up following revision(s) (requested by riastradh in ticket #1380):
lib/libc/gen/pthread_atfork.c: revision 1.19
PR/59112: Martin Husemann: switch to using mmap instead of malloc
Pull up following revision(s) (requested by riastradh in ticket #1379):
lib/libc/arch/arm/misc/arm_initfini.c: revision 1.8
lib/libc/arch/arm/misc/arm_initfini.c: revision 1.9
lib/libc/arch/arm/misc/arm_initfini.c: revision 1.10
PR/59063: Martin Husemann: Avoid using malloc through sysctlbyname by doing
the sysctl name to mib translation directly. Thanks martin@ for testing!
use the latest version that does not abort on error.
don't abort, just fail.
Pull up following revision(s) (requested by riastradh in ticket #1378):
external/bsd/nvi/dist/regex/regcomp.c: revision 1.9
PR/59103: RVP: Search of non-ASCII chars. broken when `:set extended' is set
Pull up following revision(s) (requested by riastradh in ticket #1377):
share/man/man9/module.9: revision 1.55
PR 59102: fix signature of module callback function.
revert ohci 1.330:
---
use PR_ZERO on on pool allcoated memory and get rid of memset.
currently, xfer is zeroed using the wrong size.
remove it and simplify code.
reviewed by christos@
---
Taylor pointed out that pool_cache*() does not document or honour
PR_ZERO, so this actually no longer is returning zeroed ohci_xfer
structures. it also did not have the "wrong size", but that is an
understandable mistake -- the cost assignes to "struct usb_xfer *",
but is really a "struct ohci_xfer", which is usb_xfer + more, so
while it appears to be memset()ing beyond the buffer, it isn't.
wrt pool_cache and PR_ZERO, there's a strange conflict between the
[11 lines not shown]
Import unbound-1-26-2 (previous was 1-26-1)
Unbound 1.26.2
This release has a number of security fixes.
The release is signed with the OpenPGP software signing key that is
in use since Jan 1st 2026:
User ID: NLnet Labs releases signing key G2 <releases at nlnetlabs.nl>
Key ID: A144 323D EAAC DF45
Fingerprint: 2310 1869 0C4D 903E F419 146A A144 323D EAAC DF45
The key is available from https://nlnetlabs.nl/signing-keys .
This release consolidates security fixes for issues reported over
a period of time. There are fixes for CVE-2026-35501, CVE-2026-50106,
CVE-2026-96405, CVE-2026-97206 and CVE-2026-109125.
Bug Fixes:
[25 lines not shown]
use PR_ZERO on on pool allcoated memory and get rid of memset.
currently, xfer is zeroed using the wrong size.
remove it and simplify code.
reviewed by christos@
alpha: Align kernel stack pointer to 16-byte boundaries
as ABI requires. With this fix, full ATF run successfully completed
without any regression.
PR port-alpha/60865 NetBSD-current/alpha panics running getcwd test
OK thorpej@
