BSD Commit Log Search

   Update to py3-libvirt-12.0.0.

   Update to libvirt-12.0.0.

   Update to babl-0.1.120.

   Update to gnome-text-editor-49.1.

   Adding the re2 feature I did not add it to WANTLIB

   Ops forget to commit new patch

   Spotted by Josh

   add upstream fix for inode % checks reported by danj
   add upstream fix for summaries

   update to py3-trove-classifiers-2026.1.14.14

   update to py3-alembic-1.18.1

   update to py3-regex-2026.1.15, use MODPY_TEST_LINK_SO

   Missing armv7 adaptation to upstream API changes

   update to php-8.4.17

   update to php-8.3.30

   sync change needed after Makefile.inc change

   sync PHP Makefile.inc changes from -current

   Update to rust-analyzer 2026-01-12, from maintainer

   Unbreaks build after the mdbook update

   In SEV-ES mode, guest userland is allowed to execute the vmgexit
   instruction, although it has no control over the GHCB.  Therefore,
   it is important that the GHCB does not contain a valid request after
   use.

   In all "vmgexit paths" the GHCB is cleared by ghcb_sync_in() (it
   calls ghcb_clear()) after returning from the hypervisor back into
   the guest.

   However, in _ghcb_mem_rw() I missed this when requesting MMIO writes
   from the hypervisor.  The diff below corrects this.

   I want to keep this pattern in all vmgexit paths:

           ghcb_sync_out
           vmgexit
           ghcb_verify_bm
           ghcb_sync_in


    [4 lines not shown]

   As vmd(8) direct kernel launch now uses 32-bit legacy mode (with
   paging disabled) we do not need the 64-bit #VC handling in locore0
   anymore.

   ok mlarkin@

   Update to gdb-17.1

   Tested on amd64 arm64 i386 mips64 riscv64 and sparc64 by tb@ kirill@ and me
   ok tb@

   update to dav1d-1.5.3, from Brad (tests run on BTI amd64)

   update to wireshark-4.4.13

   update to wireshark-4.4.13

   stop trying to subst in ${PREFIX}/bin/pip${VERSION} which doesn't exist
   anyway (we use a separate package for pip)

   symlink egfortran -> gfortran, to simplify use in cases where software
   defaults to using gfortran as the compiler name (noticed when trying to
   build some software in Python venvs). ok tb@ pascal@ (been through i386
   bulk).

   update to php-8.5.2, tweak capstone deps

   update to php-8.4.17

   update to php-8.3.30

   tweak whitespace

   update license notes, tweak dep setup for capstone

   remove build loop avoidance, no longer needed now that cmake was split
   with docs in a separate port (previously there was a loop with cmake ->
   sphinx -> ... -> this -> ... -> lapack -> cmake)