Build date: 1773702003 - Mon Mar 16 23:00:03 UTC 2026
Build cvs date: 1773699572 - Mon Mar 16 22:19:32 UTC 2026
Build id: 2026-03-17.1
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/sys/unveil.2
P lib/libcrypto/ecdsa/ecdsa.c
P lib/libcrypto/ecdsa/ecdsa_local.h
P regress/sbin/disklabel/12000.ok
P regress/sbin/disklabel/18000.ok
P regress/sbin/disklabel/20000.ok
P regress/sbin/disklabel/300000.ok
P regress/sbin/disklabel/40000.ok
P regress/sbin/disklabel/90000.ok
P regress/sbin/disklabel/900000.ok
P share/man/man5/port-modules.5
P sys/conf/files
P sys/dev/pci/drm/amd/amdgpu/amdgpu.h
P sys/dev/pci/drm/amd/amdgpu/amdgpu_drv.c
P sys/dev/pci/drm/ttm/ttm_device.c
M sys/dev/usb/uhidev.c
P sys/kern/kern_pledge.c
P usr.bin/calendar/calendar.c
M usr.sbin/bgpd/session.c

commit 6GxpqTmBQ1ldlsjg
Author: tb
Date: 2026/03/16 22:19:32

  Move ECDSA_SIG_st definition to its only consumer, ecdsa.c

  lib/libcrypto/ecdsa/ecdsa.c
  lib/libcrypto/ecdsa/ecdsa_local.h

commit K5FVXpgL3CQFzYjO
Author: kettenis
Date: 2026/03/16 21:19:20

  Our framebuffer might be in "stolen" memory that is lost (and filled with
  garbage) after a hibernate.  So clear it and redraw the framebuffer
  contents.

  ok jsg@

  sys/dev/pci/drm/amd/amdgpu/amdgpu.h
  sys/dev/pci/drm/amd/amdgpu/amdgpu_drv.c

commit PoqrhXEOiZ7f1bSH
Author: deraadt
Date: 2026/03/16 19:54:27

  clarify unveil usage
  with job

  lib/libc/sys/unveil.2

commit R9WFyR173mwZlRbE
Author: sthen
Date: 2026/03/16 17:19:27

  sendmail/cpp only need "x" no "rx" unveil; ok deraadt

  usr.bin/calendar/calendar.c

commit tsWFhYIEd2at3VwT
Author: sthen
Date: 2026/03/16 16:58:03

  calendar also needs to unveil cpp / sendmail, I forgot to make install
  when I was testing :(  from James J. Lippard

  usr.bin/calendar/calendar.c

commit RNKnVV15dPde6XKB
Author: stu
Date: 2026/03/16 15:41:19

  Remove MODTCL_BIN and MODTK_BIN from SUBST_VARS.
  The Tcl and Tk port modules no longer add to SUBST_VARS.


  ok sthen@

  share/man/man5/port-modules.5

commit CkJpyAY38qRc3RoX
Author: sthen
Date: 2026/03/16 15:39:46

  /dev/null is no longer implicitly permitted by some pledges, so explicitly
  unveil it. fixes calendar -a. ok deraadt

  usr.bin/calendar/calendar.c

commit rhRPOyHwHU7kJK1y
Author: claudio
Date: 2026/03/16 14:57:34

  fuse no longer (ab)uses ufs_ihash.c so adjust build config

  OK miod@ deraadt@ helg@ tb@

  sys/conf/files

commit FYdl4nsWTfTigsjj
Author: anton
Date: 2026/03/16 06:14:09

  Sync expected sizes after recent /usr/obj bump.

  regress/sbin/disklabel/12000.ok
  regress/sbin/disklabel/18000.ok
  regress/sbin/disklabel/20000.ok
  regress/sbin/disklabel/300000.ok
  regress/sbin/disklabel/40000.ok
  regress/sbin/disklabel/90000.ok
  regress/sbin/disklabel/900000.ok

commit TlDkmINrS1C7imG0
Author: jsg
Date: 2026/03/16 03:57:39

  make ttm_device_prepare_hibernation() return early

  kettenis reports the ttm_tm_swapout() change is not enough to fix
  hibernate with amdgpu when X is running.

  The call to ttm_device_prepare_hibernation()/ttm_tm_swapout() in the
  hibernate path was added in a linux commit from 2025:
  924dda024f3bea64be5f3ac067a075e466739dc9
  drm/amdgpu: move GTT to shmem after eviction for hibernation

  The linux reason for doing this (server cards with lots of VRAM) isn't
  something that concerns us.

  ok kettenis@

  sys/dev/pci/drm/ttm/ttm_device.c

commit b9QAh4gtGVV2baDr
Author: deraadt
Date: 2026/03/16 03:45:20

  When pledged, the special access(2) inspection for "/etc/localtime" has not
  been needed since changes made in lib/libc/time/localtime.c 1.58 (2019)
  pointed out by david leadbeater

  sys/kern/kern_pledge.c