NetBSD/src q6IGoac — lib/libc/regex engine.c

   PR/61017: Juan Manuel Mendez Rey: regexec reads past the string after a
   multibyte character. Provide a prevchar() function that fetches the previous
   multibyte character.
VersionDeltaFile
1.31+41-10lib/libc/regex/engine.c
+41-101 files

NetBSD/src IW6ukhU — distrib/sets/lists/debug mi, distrib/sets/lists/tests mi

   New regex rest for PR/61017
VersionDeltaFile
1.1436+2-1distrib/sets/lists/tests/mi
1.526+2-1distrib/sets/lists/debug/mi
+4-22 files

NetBSD/src w6aFJEA — tests/lib/libc/regex Makefile t_regex.sh

   PR/61017: Juan Manuel Mendez Rey: regexec reads past the string after a
   multibyte character
VersionDeltaFile
1.1+57-0tests/lib/libc/regex/h_match.c
1.2+33-1tests/lib/libc/regex/t_regex.sh
1.14+2-2tests/lib/libc/regex/Makefile
+92-33 files

NetBSD/src ZkI5BUM — usr.sbin/npf/npfctl npf_data.c, usr.sbin/npf/npftest npftest.conf

   implement /32 IPv4 parsing in NPF.

   Timo Burhmester (timo@) requested in a recent tech-net thread on NPF.

   during the extraction of address/mask data, if mask is equal to max mask,
   both v4 and v6, it is assumed to be no net mask set.

   reviewed by martin@ and tests also included.
VersionDeltaFile
1.35+7-2usr.sbin/npf/npfctl/npf_data.c
1.15+9-0usr.sbin/npf/npftest/libnpftest/npf_nat_test.c
1.20+6-1usr.sbin/npf/npftest/npftest.conf
1.27+2-0usr.sbin/npf/npftest/libnpftest/npf_test.h
+24-34 files

NetBSD/src aDlIWsF — sys/arch/m68k/m68k genassym.cf

   Only define CI_LEV1MAP when TBL40_L1_SIZE is available.
VersionDeltaFile
1.12+3-1sys/arch/m68k/m68k/genassym.cf
+3-11 files

NetBSD/src nvJQzhq — sbin/newfs_lfs Makefile extern.h

   newfs_lfs(8): teach it to use findspecial so it can take NAME= syntax

   PR bin/61018 newfs_lfs(8) cannot take a wedge NAME=abc
VersionDeltaFile
1.31+6-4sbin/newfs_lfs/newfs.c
1.15+3-3sbin/newfs_lfs/extern.h
1.17+2-2sbin/newfs_lfs/Makefile
+11-93 files

NetBSD/src NYDw6Xx — sys/compat/linux/common linux_signal.c

   PR/61005: Bobby Bingham: compat_linux fails to set si_pid when delivering
   SI_QUEUE signal to Linux process because of cut-n-paste mistake
VersionDeltaFile
1.93+3-3sys/compat/linux/common/linux_signal.c
+3-31 files

NetBSD/src poVzNqB — usr.bin/kdump kdump.c

   PR/61007: tuxifan at posteo.de: Don't attempt to translate syscall arguments to
   names for non-native emulations
VersionDeltaFile
1.153+13-2usr.bin/kdump/kdump.c
+13-21 files

NetBSD/src BQTXhKg — lib/libc/gen shquote.c

   libc/shquote: remove obsolete lint comments

   Lint does not complain about 'do stmt while (0)' anymore.
VersionDeltaFile
1.9+5-5lib/libc/gen/shquote.c
+5-51 files

NetBSD/src DJbCD5O — sys/arch/hppa/hppa machdep.c

   hppa: add D210/D200 board ids to the hpcxl list in cpu_model_cpuid

   Fixes boot on D210 as reported by Charles Blythe on port-hppa.
VersionDeltaFile
1.29+4-2sys/arch/hppa/hppa/machdep.c
+4-21 files

NetBSD/src DuhHou5 — sys/arch/m68k/m68k pmap_68k.c

   Oops, need to use PMAP_WRITE_THROUGH in pmap_percpu_init().
VersionDeltaFile
1.96+8-5sys/arch/m68k/m68k/pmap_68k.c
+8-51 files

NetBSD/src NvQtmLU — sys/arch/m68k/include pmap_68k.h, sys/arch/m68k/m68k genassym.cf pmap_68k.c

   First step towards MP support in the new pmap:

   Add new functions that allocate and initialize (as separate steps) a
   per-CPU page that contains its cpu_info as well as lev1, lev2, and lev3
   page tables.  The lev1 table is a copy of the kernel_lev1map for that
   CPU, except the slot corresponding to CPUINFO_VA points to that CPU's
   private lev2 table, whose slot corresponding to CPUINFO_VA points to
   that CPU's private lev3 table, whose slot corresponding to CPUINFO_VA
   points to this page.  Thus, CPUINFO_VA always maps to that specific
   CPU's cpu_info page.

   If new level 2 tables need to be allowed during pmap_growkernel(),
   each secondary CPU's private lev1 table will be updated.  Happily,
   this is an extremely rare event.

   Decorate all of the __read_mostly vars.
VersionDeltaFile
1.95+234-27sys/arch/m68k/m68k/pmap_68k.c
1.28+9-1sys/arch/m68k/include/pmap_68k.h
1.11+2-1sys/arch/m68k/m68k/genassym.cf
+245-293 files

NetBSD/src 94qTTuV — sys/arch/hb68k/pg68k pmap_pgmmu.c, sys/arch/m68k/include cpu.h

   More per-cpu data work:
   - curcpu() needs to evaluate to a globally-accessible pointer, not a
     cpu-local pointer, so do what the sparc port does here - provide a
     ci_self that always points to the global pointer, and use that in
     curcpu().  cpu_number() and curlwp reference through the cpu-local
     alias.
   - Rename that alias from cpu_info_store to cpuinfo, again mirroring
     what the sparc port does here.
   - Add the CPU_INFO_FOREACH() infrastructure, lifted from alpha port.
   - Make the temp VAs used by the pmap code + the vmmap temp VA per-cpu.
VersionDeltaFile
1.94+59-41sys/arch/m68k/m68k/pmap_68k.c
1.5+18-47sys/arch/hb68k/pg68k/pmap_pgmmu.c
1.8+19-23sys/arch/sun68k/sun68k/pmap_sunmmu.c
1.67+17-17sys/arch/sun2/sun2/pmap.c
1.36+21-5sys/arch/m68k/include/cpu.h
1.135+10-15sys/arch/sun3/sun3x/pmap.c
+144-14824 files not shown
+259-27930 files

NetBSD/src J8xMLZS — doc CHANGES-11.2

   Tickets #566 - #578
VersionDeltaFile
1.1.2.2+75-1doc/CHANGES-11.2
+75-11 files

NetBSD/src aQanD8k — doc CHANGES-11.1

   Ooops, the welcome to 11.1 entry did not make it to 11.1 accidently.
VersionDeltaFile
1.1.2.37+6-1doc/CHANGES-11.1
+6-11 files

NetBSD/src 39aJO1k — doc CHANGES-10.3

   Tickets #1409 - #1421
VersionDeltaFile
1.1.2.29+71-1doc/CHANGES-10.3
+71-11 files

NetBSD/src gdpg76X — sys/dev/usb usb_quirks.c uts.c

   Pull up following revision(s) (requested by riastradh in ticket #578):

        sys/dev/usb/usb_quirks.c: revision 1.109
        sys/dev/usb/usbdevs: revision 1.829
        sys/dev/usb/uts.c: revision 1.19

   Add product id for ELAN touchscreen controller.

   Part of PR/60883.

   Add support for ALWAYSON quirk to uts(4).

   Add quirk for ELAN touchscreens.

   Addresses PR/60883.
VersionDeltaFile
1.18.2.1+31-6sys/dev/usb/uts.c
1.108.4.1+4-2sys/dev/usb/usb_quirks.c
+35-82 files

NetBSD/src nxrP3rN — sys/dev/usb usbdevs.h usbdevs_data.h

   regen for ticket #578
VersionDeltaFile
1.815.2.1+2,813-2,809sys/dev/usb/usbdevs_data.h
1.815.2.1+2-1sys/dev/usb/usbdevs.h
+2,815-2,8102 files

NetBSD/src Py3We9t — sys/dev/usb usbdevs

   Pull up the following, requested by nat in ticket #578:

        sys/dev/usb/usbdevs                             1.829

   Add product id for ELAN touchscreen controller.
VersionDeltaFile
1.823.2.1+2-1sys/dev/usb/usbdevs
+2-11 files

NetBSD/src OOpDsFO — sys/external/bsd/drm2/dist/drm/i915/gt intel_ggtt.c

   Pull up following revision(s) (requested by riastradh in ticket #1421):

        sys/external/bsd/drm2/dist/drm/i915/gt/intel_ggtt.c: revision 1.20

   i915: Fix partial page view indexing error.

   PR kern/60907: i915: partial page view indexing error
VersionDeltaFile
1.15.4.2+7-3sys/external/bsd/drm2/dist/drm/i915/gt/intel_ggtt.c
+7-31 files

NetBSD/src hzj212g — sys/external/bsd/drm2/dist/drm/i915/gt intel_ggtt.c

   Pull up following revision(s) (requested by riastradh in ticket #577):

        sys/external/bsd/drm2/dist/drm/i915/gt/intel_ggtt.c: revision 1.20

   i915: Fix partial page view indexing error.

   PR kern/60907: i915: partial page view indexing error
VersionDeltaFile
1.19.2.1+7-3sys/external/bsd/drm2/dist/drm/i915/gt/intel_ggtt.c
+7-31 files

NetBSD/src 6wu67Tc — sys/external/bsd/drm2/dist/drm/nouveau nouveau_fence.c

   Pull up following revision(s) (requested by riastradh in ticket #1420):

        sys/external/bsd/drm2/dist/drm/nouveau/nouveau_fence.c: revision 1.18

   nouveau(4): Fix return value of fence wait.

   kpause(9) returns EAGAIN/EWOULDBLOCK when it times out, which is
   perfectly normal and expected here with an internal one-tick timeout
   which should just go through another loop iteration.  This function
   is expected to return the number of ticks left to wait when done, so
   it reports timeout by returning zero, and is forbidden to fail with
   EAGAIN/EWOULDBLOCK.

   This is also forbidden to fail with EINTR -- on interruption it must
   fail with the Linux ERESTARTSYS instead.  So map EINTR to that too
   while here.

   PR kern/59772: i386 11.0_BETA system crash (Xorg nouveau ?)
VersionDeltaFile
1.17.4.2+4-4sys/external/bsd/drm2/dist/drm/nouveau/nouveau_fence.c
+4-41 files

NetBSD/src nCNUcw8 — doc CHANGES-11.0

   Fix PR reference for ticket #105
VersionDeltaFile
1.1.2.142+2-2doc/CHANGES-11.0
+2-21 files

NetBSD/src cNWwQGR — sys/arch/x86/x86 efi_machdep.c

   Pull up following revision(s) (requested by riastradh in ticket #1419):

        sys/arch/x86/x86/efi_machdep.c: revision 1.7
        sys/arch/x86/x86/efi_machdep.c: revision 1.8

   x86/efi: Bypass kASAN memcpy when copying the efi runtime.

   Fixes crash at boot with `options KASAN' enabled on EFI systems.

   We could try to find some way to exclude the EFI runtime mappings
   from kASAN, e.g. by hacking kasan_md_unsupported.  But there's a lot
   of detailed exclusions to pass through to that (every entry in the
   EFI memory map), and this is the only memcpy in the whole kernel that
   all those details are likely to be applicable.

   PR kern/60391: kernel panic in efi_init() with "options KASAN"


   x86/efi_machdep.c: Nix extra blank line.

    [2 lines not shown]
VersionDeltaFile
1.3.4.2+9-4sys/arch/x86/x86/efi_machdep.c
+9-41 files

NetBSD/src SvB6zYl — sys/arch/x86/x86 efi_machdep.c

   Pull up following revision(s) (requested by riastradh in ticket #576):

        sys/arch/x86/x86/efi_machdep.c: revision 1.7
        sys/arch/x86/x86/efi_machdep.c: revision 1.8

   x86/efi: Bypass kASAN memcpy when copying the efi runtime.

   Fixes crash at boot with `options KASAN' enabled on EFI systems.

   We could try to find some way to exclude the EFI runtime mappings
   from kASAN, e.g. by hacking kasan_md_unsupported.  But there's a lot
   of detailed exclusions to pass through to that (every entry in the
   EFI memory map), and this is the only memcpy in the whole kernel that
   all those details are likely to be applicable.

   PR kern/60391: kernel panic in efi_init() with "options KASAN"


   x86/efi_machdep.c: Nix extra blank line.

    [2 lines not shown]
VersionDeltaFile
1.6.8.1+9-4sys/arch/x86/x86/efi_machdep.c
+9-41 files

NetBSD/src aTmSKVn — sys/kern exec_elf.c, sys/sys exec_elf.h

   Pull up following revision(s) (requested by riastradh in ticket #1418):

        sys/kern/exec_elf.c: revision 1.108
        sys/kern/exec_elf.c: revision 1.109
        sys/sys/exec_elf.h: revision 1.187
        sys/sys/exec_elf.h: revision 1.188

   exec_elf: Ignore freedesktop.org dlopen ELF notes.

   PR kern/60901: Unknown elf note type 1081871370
   exec_elf: Nix diagnostic messages for all (unknown) fd.o notes.

   Same as for unknown GNU notes.  Shaves off a tiny bit of code in
   non-DIAGNOSTIC builds, saves us some trouble when fd.o invariably
   adds more metadata not relevant to execution semantics.

   PR kern/60901: Unknown elf note type 1081871370
VersionDeltaFile
1.170.4.1+27-4sys/sys/exec_elf.h
1.104.2.1+7-5sys/kern/exec_elf.c
+34-92 files

NetBSD/src 2OZgiXw — sys/kern exec_elf.c, sys/sys exec_elf.h

   Pull up following revision(s) (requested by riastradh in ticket #575):

        sys/kern/exec_elf.c: revision 1.108
        sys/kern/exec_elf.c: revision 1.109
        sys/sys/exec_elf.h: revision 1.187
        sys/sys/exec_elf.h: revision 1.188

   exec_elf: Ignore freedesktop.org dlopen ELF notes.

   PR kern/60901: Unknown elf note type 1081871370
   exec_elf: Nix diagnostic messages for all (unknown) fd.o notes.

   Same as for unknown GNU notes.  Shaves off a tiny bit of code in
   non-DIAGNOSTIC builds, saves us some trouble when fd.o invariably
   adds more metadata not relevant to execution semantics.

   PR kern/60901: Unknown elf note type 1081871370
VersionDeltaFile
1.177.2.1+27-4sys/sys/exec_elf.h
1.107.2.1+7-5sys/kern/exec_elf.c
+34-92 files

NetBSD/src 9zdqJlW — sys/dev/i2c i2c.c

   Pull up following revision(s) (requested by riastradh in ticket #1417):

        sys/dev/i2c/i2c.c: revision 1.109

   i2c(4): Avoid leaking kernel heap memory to userland.

   PR kern/60913: iic(4): I2C_IOCTL_EXEC can copy out an uninitialised
   buffer
VersionDeltaFile
1.90.2.2+11-3sys/dev/i2c/i2c.c
+11-31 files

NetBSD/src iDnAIhh — sys/dev/i2c i2c.c

   Pull up following revision(s) (requested by riastradh in ticket #574):

        sys/dev/i2c/i2c.c: revision 1.109

   i2c(4): Avoid leaking kernel heap memory to userland.

   PR kern/60913: iic(4): I2C_IOCTL_EXEC can copy out an uninitialised
   buffer
VersionDeltaFile
1.92.2.1+11-3sys/dev/i2c/i2c.c
+11-31 files

NetBSD/src Tfd8KxT — etc/rc.d network

   Pull up following revision(s) (requested by riastradh in ticket #1416):

        etc/rc.d/network: revision 1.86

   PR misc/60836 - Be more careful with /etc/ifconfig.* cloner configs

   Only config cloner interfaces when an /etc/ifconfig.${cloner}N file
   exists, if N is completely numeric.
VersionDeltaFile
1.85.6.1+4-1etc/rc.d/network
+4-11 files