Build date: 1786485603 - Tue Aug 11 22:00:03 UTC 2026 Build cvs date: 1786472742 - Tue Aug 11 18:25:42 UTC 2026 Build id: 2026-08-12.1 Build tags: amd64-regress ports sysupgrade Applied the following diff(s): /home/anton/tmp/robsd/src-sys-em.diff /home/anton/tmp/robsd/src-sys-uhidev-sispm.diff /home/anton/tmp/robsd/src-sysupgrade.diff P lib/libc/stdlib/hcreate.3 P lib/libc/stdlib/hcreate.c P regress/lib/libcrypto/wycheproof/wycheproof.go P regress/usr.bin/ssh/keyscan.sh P sbin/iked/ca.c P share/man/man7/mdoc.7 P sys/dev/bio.c M sys/dev/usb/uhidev.c P sys/net/if.c P sys/net/pf_norm.c P sys/netinet/ip_input.c P sys/netinet/ip_var.h P sys/netinet6/frag6.c P sys/netinet6/ip6_var.h P usr.bin/rsync/flist.c P usr.bin/ssh/myproposal.h P usr.bin/ssh/ssh-keyscan.c P usr.bin/ssh/version.h M usr.sbin/bgpd/session.c commit 7xPovWqIXi7YOSEi Author: tb Date: 2026/08/11 18:25:42 wycheproof: recognize Cobblestone-{128,256} regress/lib/libcrypto/wycheproof/wycheproof.go commit cSsOoGTxSeW8BvsN Author: deraadt Date: 2026/08/11 16:25:29 Put back the whitelist approach previous to 1.18, because without a specific whitelist, arbitrary *TINY* requests can flow through here which lack the required structure header. Issue found by gnezdo. ok gnezdo jmatthew sys/dev/bio.c commit TY1W6Ytt8iOAxpee Author: deraadt Date: 2026/08/11 16:19:02 SIOCGIFDATA (on most architectures) leaked 4 bytes of kernel stack. from Acts1631 sys/net/if.c commit DZtKWrBoE3poRPoI Author: bluhm Date: 2026/08/11 14:28:59 Reassemble IP fragments in correct routing domain. Fragment reassembly in pf(4) and inet(4) and inet6(4) did not respect the routing domain. Fragments from different routing domains could be merged together. Keep track of rdomain in pf and network stack. Convert routing table to routing domain where necessary. Sort struct fields by size. report and diff from Acts1631; fixes by bluhm@; OK sashan@ sys/net/pf_norm.c sys/netinet/ip_input.c sys/netinet/ip_var.h sys/netinet6/frag6.c sys/netinet6/ip6_var.h commit ydUk6vIi3fQwGx4W Author: schwarze Date: 2026/08/11 13:39:44 remove a stale .Sx internal link; similar to a patch sent in by share/man/man7/mdoc.7 commit UXAwO2qAKmiMM2vP Author: tb Date: 2026/08/11 13:14:30 iked: remove pointless call to X509_get_subject_name() A deserialized certificate always has a subject, so this check doesn't do anything useful. This was added in r1.7 (2010) as a mostly cosmetic return check, and it has been useless since r1.30 (2014) when the subject variable became unused. A subsequent removal of subject in r1.66 (2020) made this look weird (why would you check subject but not issuer, SPKI, ...?), so garbage collect it. ok hshoexer sbin/iked/ca.c commit u5h1cUR1ldnlJaQ3 Author: schwarze Date: 2026/08/11 13:09:24 Overhaul documentation of the -offset and -width arguments to make the macro reference section easier to read and less repetitive, and to document these arguments more precisely in a less obtrusive place. * Move the gory details out of the descriptions of .Bd and .Bl; only state the defaults and recommended arguments there. * Stop documenting -offset right and -offset center. Neither is implemented in mandoc(1), both are handled very poorly by groff and almost never occur in the wild. * Add a new WIDTH SPECIFICATIONS section after the macro reference. share/man/man7/mdoc.7 commit 7kCVBQZut2LjkgYv Author: djm Date: 2026/08/10 23:28:15 add regress test for servers that hang durig banner write; from Thomas Yiu regress/usr.bin/ssh/keyscan.sh commit dNteVKgYZN8pIRjE Author: djm Date: 2026/08/10 23:27:30 openssh-10.5 usr.bin/ssh/version.h commit LqCL3cSiqDJbCdiO Author: djm Date: 2026/08/10 23:27:04 remove ml-dsa/x25519 from the default algorithm list for release (it will be back soon) usr.bin/ssh/myproposal.h commit xsxRxdLIQRtIbW9b Author: djm Date: 2026/08/10 23:24:03 make reading server banner non-blocking; prevents one stuck server from blocking a many-host keyscan; from Thomas Yiu ok dtucker@ markus@ usr.bin/ssh/ssh-keyscan.c commit 1aK0S96feW91GJkb Author: daniel Date: 2026/08/10 22:02:41 hdestroy(3): stop freeing key values POSIX leaves ownership of ENTRY.key with the caller. Our current behaviour came from NetBSD when this code was imported in 2004. NetBSD and FreeBSD both stopped freeing key values in July 2014. NetBSD in commit fe0a1a6 and FreeBSD in commit e19f362. Remove the call to free(3) to align ourselves with NetBSD, FreeBSD and POSIX. Also adjust rsync for the change in semantics. Flagged by Sortix os-test. ok millert@, deraadt@ lib/libc/stdlib/hcreate.3 lib/libc/stdlib/hcreate.c usr.bin/rsync/flist.c P MODULES P README.md P lib/libXfont2/ChangeLog P lib/libXfont2/Makefile.am P lib/libXfont2/Makefile.in P lib/libXfont2/README.md P lib/libXfont2/configure P lib/libXfont2/configure.ac P lib/libXfont2/src/fc/fserve.c P lib/libXfont2/src/fc/fservestr.h U lib/libXfont2/test/test-fserve-read-glyphs.c commit whbR1i6gRQ7kgeNK Author: sthen Date: 2026/08/11 11:48:32 small fixes, from Avon Robertson README.md commit Q6CnC9NmiXHpwSws Author: matthieu Date: 2026/08/11 07:49:09 update MODULES commit vPCIucceo6OGiQJK Author: matthieu Date: 2026/08/11 07:27:18 Update to libXfont2 2.0.9 font server code is disabled in xenocara, so OpenBSD is not affected by the security issues fixed in this release. lib/libXfont2/ChangeLog lib/libXfont2/Makefile.am lib/libXfont2/Makefile.in lib/libXfont2/README.md lib/libXfont2/configure lib/libXfont2/configure.ac lib/libXfont2/src/fc/fserve.c lib/libXfont2/src/fc/fservestr.h lib/libXfont2/test/test-fserve-read-glyphs.c