Build date: 1776981602 - Thu Apr 23 22:00:02 UTC 2026
Build cvs date: 1776973897 - Thu Apr 23 19:51:37 UTC 2026
Build id: 2026-04-24.1
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/time/difftime.c
P regress/usr.bin/ssh/unittests/Makefile.inc
P sys/arch/riscv64/dev/simplebus.c
P sys/arch/riscv64/include/bus.h
P sys/arch/riscv64/riscv64/autoconf.c
P sys/arch/riscv64/riscv64/bus_dma.c
P sys/ddb/db_input.c
P sys/dev/pci/if_ix.c
P sys/dev/pci/drm/amd/amdgpu/amdgpu_vm.c
P sys/dev/pci/drm/amd/amdkfd/kfd_queue.c
P sys/dev/pci/drm/i915/i915_gem.c
P sys/dev/pci/drm/include/linux/uaccess.h
M sys/dev/usb/uhidev.c
P sys/kern/uipc_mbuf.c
P sys/net/art.c
P sys/net/art.h
P sys/net/if_pfsync.c
P sys/net/pf_ioctl.c
P usr.bin/netstat/mbuf.c
P usr.bin/tmux/file.c
P usr.bin/tmux/screen-redraw.c
P usr.bin/tmux/server-client.c
P usr.bin/tmux/tmux.h
M usr.sbin/bgpd/session.c
P usr.sbin/rpki-client/parser.c

commit iLue6llmFOsLtacI
Author: kettenis
Date: 2026/04/23 19:51:37

  Implement bounce buffers for riscv64.

  ok jca@

  sys/arch/riscv64/dev/simplebus.c
  sys/arch/riscv64/include/bus.h
  sys/arch/riscv64/riscv64/autoconf.c
  sys/arch/riscv64/riscv64/bus_dma.c

commit S92BCFVwoD8uaLA0
Author: claudio
Date: 2026/04/23 14:15:53

  Move repo_tree_free() up to where the other repo functions live.
  OK tb@

  usr.sbin/rpki-client/parser.c

commit tqN17waH89WyYdrX
Author: claudio
Date: 2026/04/23 12:57:47

  Drop -Winline from CDIAGFLAGS it breaks on sparc64

  On sparc64 ssh/unittests/kex fails to build with:
  usr.bin/ssh/libcrux_mlkem768_sha3.h:8196: warning: inlining failed
  in call to 'libcrux_ml_kem_polynomial_ZERO_89_ea':

  regress/usr.bin/ssh/unittests/Makefile.inc

commit tmpLQISVDoNa7KTk
Author: nicm
Date: 2026/04/23 12:36:15

  Kill client rather than fatalx on bad file handling messages, reported
  by Tim Zheng.

  usr.bin/tmux/file.c
  usr.bin/tmux/server-client.c
  usr.bin/tmux/tmux.h

commit LE1dhNocCIUigtD5
Author: nicm
Date: 2026/04/23 11:29:23

  Redraw correctly when a popup is present above pane borders, GitHub
  issue 4997 from Manuel Einfalt.

  usr.bin/tmux/screen-redraw.c

commit gN7NvjZIUKM5eXDC
Author: jsg
Date: 2026/04/23 01:33:01

  remove unused pf_statelim_clr()
  ok dlg@

  sys/net/pf_ioctl.c

commit NF8D01QUZ3JOAFUx
Author: jsg
Date: 2026/04/23 01:28:03

  remove unused art_walk()
  ok dlg@

  sys/net/art.c
  sys/net/art.h

commit z79OVe6UEAkuc2yn
Author: jsg
Date: 2026/04/23 01:16:55

  rename __copy_from_user_inatomic_nocache() to
  copy_from_user_inatomic_nontemporal() to follow changes in linux 6.18.24

  sys/dev/pci/drm/include/linux/uaccess.h

commit klwznksESbiPIzkp
Author: jsg
Date: 2026/04/23 01:16:32

  x86: rename and clean up __copy_from_user_inatomic_nocache()

  From Linus Torvalds
  03fd014cd9f3a3d173740ab9c5cbede82fd6322c in linux-6.18.y/6.18.24
  5de7bcaadf160c1716b20a263cf8f5b06f658959 in mainline linux

  sys/dev/pci/drm/i915/i915_gem.c

commit BWU6b4x1bcqsWWd5
Author: dlg
Date: 2026/04/23 01:15:07

  make ctrl-w remove trailing space from words too

  this makes it more consistent with what i experience with ctrl-w
  in the shell.

  ok deraadt@ claudio@

  sys/ddb/db_input.c

commit 2V62DzqyvMfOIG3l
Author: jsg
Date: 2026/04/23 01:10:33

  drm/amdkfd: Fix queue preemption/eviction failures by aligning control stack size to GPU page size

  From Donet Tom
  647fb0dc3818733024fc96c1df1ec3af806b0256 in linux-6.18.y/6.18.24
  78746a474e92fc7aaed12219bec7c78ae1bd6156 in mainline linux

  sys/dev/pci/drm/amd/amdkfd/kfd_queue.c

commit RkS7KTKACTE3iuod
Author: millert
Date: 2026/04/23 01:08:47

  Fix difftime() result when it is passed a negative value

  We need to cast the result of bitwise AND to time_t before the cast
  to double in the HI and LO macros.  Otherwise, we get a very large
  positive floating point value instead of a negative value.

  Reported by Xuntao Chi

  lib/libc/time/difftime.c

commit CqHbubX7R4OgUWj7
Author: jsg
Date: 2026/04/23 01:08:46

  drm/amdgpu: Handle GPU page faults correctly on non-4K page systems

  From Donet Tom
  6a9f2683c66dc54d3598589684c0b3c5cb2862ad in linux-6.18.y/6.18.24
  4e9597f22a3cb8600c72fc266eaac57981d834c8 in mainline linux

  sys/dev/pci/drm/amd/amdgpu/amdgpu_vm.c

commit MI8BCf3dkTh1Gbqx
Author: sashan
Date: 2026/04/22 23:06:01

  pf_purge_states() may trip assert(st->timeout == PFTM_UNLINKED)
  in pf_free_state(). Once member ->timeout in pf_state structure
  reaches PFTM_UNLINKED value, then the ->timeout member must not
  not be updated. This diff reminds pfsync(4) to follow PFTM_UNLINKED
  rule too. The pfsync(4) currently may accidentally update ->timeout
  member while state is being purged, causing pf_purge_states() to
  trip the assert.

  Issue was kindly reported by Stuart Henderson.

  OK @bluhm

  sys/net/if_pfsync.c

commit kx5HD1kCMMhT7AE9
Author: dlg
Date: 2026/04/22 22:12:49

  fix tx dma segment size

  i thought it was weird that an MI driver like ix(4) is using an MD
  number like PAGE_SIZE for the maximum segment size in its tx dma
  maps. the manual says tx segments can be 16k (and goes into a 16bit
  field), so let's try using the documented value here.

  the rx descs also use 16 * 1024 as a magic number here, so it's in
  keeping with other code in the same driver.

  tested by me on arm64 and sparc64 (which has 8k pages)
  tested by and ok jan@

  sys/dev/pci/if_ix.c

commit wz9wWmFRhazqi8Vv
Author: dlg
Date: 2026/04/22 22:09:18

  fix rss hashing on big endian archs

  i noticed that packets seem to be hashed differently by the network
  stack and this hardware on sparc64. mucking around with how we
  handle the endianness of the rss key programmed via registers makes
  it consistent on both big and little archs.

  tested by me on sparc64 and arm64
  ok claudio@

  sys/dev/pci/if_ix.c

commit VCepx3mfNYYn5JjE
Author: dlg
Date: 2026/04/22 21:58:53

  increase the 9k mbuf clusters to 9k + 128 bytes

  pools try to amortise the cost of items against the underlying
  kernel memory allocator by rounding the "page" size up to fit at
  least 8 items, and then rounding that up to the next power of 2.
  the 9k clusters are 9 * 1024 bytes, which is 72k after being
  multiplied by 8, which becomes 128k cos it's the next power of 2.
  if you divide 128k by 9k, you get 14 and some change. there's enough
  change that we can raise the cluster size by 128 bytes without
  affecting the page size or the number of items on the page. ie,
  it's still going to use 128k "pages" and fit 14 clusters.

  i can take advantage of this in some drivers for stupid hardware,
  so given the above it seems like a plan without any drawbacks apart
  from the pool name getting a bit bigger.

  ok claudio@

  sys/kern/uipc_mbuf.c
  usr.bin/netstat/mbuf.c