lint: allow 'thread_local' to be combined with 'static'

(rillig)

Propoerly handle 4k sector size backends:
- report the backend's sector size to upper layers, not DEV_BSIZE.
  Adjust the number of sectors accordingly.
- Use sc_secsize instead of XEN_BSIZE where appropriate. The sectors numbers
  in I/O requests are still in XEN_BSIZE units, but must be a multiple
  of sc_secsize/XEN_BSIZE.
- As a consequence of previous, the buffer has to be aligned to sc_secsize,
  aligned to XEN_BSIZE may not be enough. This means that we may have to
  xbd_map_align() more buffer, including some without B_PHYS set.
- Add some more DPRINTF lines, related to I/O requests

Tested with a linux dom0.
thanks to Christian Kujau for providing access to his hardware for testing
and debugging.

(bouyer)

Ticket #1666

(martin)

Pull up following revision(s) (requested by riastradh in ticket #1666):

sys/dev/nvmm/x86/nvmm_x86_svm.c: revision 1.85

nvmm: Filter CR4 bits on x86 SVM (AMD).

In particular, prohibit PKE, Protection Key Enable, which requires
some additional management of CPU state by nvmm.

(martin)

Ticket #246

(martin)

Pull up following revision(s) (requested by riastradh in ticket #246):

sys/dev/nvmm/x86/nvmm_x86_svm.c: revision 1.85

nvmm: Filter CR4 bits on x86 SVM (AMD).

In particular, prohibit PKE, Protection Key Enable, which requires
some additional management of CPU state by nvmm.

(martin)

Ticket #244

(martin)

Pull up following revision(s) (requested by riastradh in ticket #244):

sys/arch/x86/x86/fpu.c: revision 1.80
sys/arch/x86/include/cpu_extended_state.h: revision 1.18

x86: Mitigate MXCSR Configuration Dependent Timing in kernel FPU use.

In fpu_kern_enter, make sure all the MXCSR exception status bits are
set when we start using the FPU, so that instructions which exhibit
MCDT are unaffected by it.

While here, zero all the other FPU registers in fpu_kern_enter.
In principle we could skip this step on future CPUs that fix the MCDT
bug, but there's probably not much benefit -- workloads that do a lot
of crypto in the kernel are probably better off using
kthread_fpu_enter or WQ_FPU to skip the fpu_kern_enter/leave cycles
in the first place.

For details, see:
https://www.intel.com/content/www/us/en/developer/articles/technical/software-security-guidance/best-practices/mxcsr-configuration-dependent-timing.html

(martin)

Ticket #1665

(martin)

Pull up following revision(s) (requested by riastradh in ticket #1665):

sys/arch/x86/x86/fpu.c: revision 1.86

x86/fpu: Align savefpu to 64 bytes in fpuinit_mxcsr_mask.
16 bytes is not enough.

(Is this why it never worked on Xen some years back?  Got lucky and
accidentally had 64-byte alignment on native x86, but not in the call
stack in Xen?)

(martin)

Ticket #245

(martin)

Pull up following revision(s) (requested by riastradh in ticket #245):

sys/arch/x86/x86/fpu.c: revision 1.86

x86/fpu: Align savefpu to 64 bytes in fpuinit_mxcsr_mask.
16 bytes is not enough.

(Is this why it never worked on Xen some years back?  Got lucky and
accidentally had 64-byte alignment on native x86, but not in the call
stack in Xen?)

(martin)

Ticket #1851

(martin)

Pull up following revision(s) (requested by mrg in ticket #1851):

sys/arch/x86/include/specialreg.h: revision 1.207
sys/arch/x86/x86/errata.c: revision 1.31

x86: turn off zenbleed chicken bit on Zen2 cpus.

this is based upon Taylor's original work.  i just made the list
of CPUs to run on correct as i could determine.  (also, add some
Zen3 and Zen4 cpuids not yet used by any errata.)

(might be nice to have a better way to expression revision ranges
rather than specific cpuid matches, eg, 0x30-0x4f models in a cpu
family, etc.)

tested on ryzen 3600, and a ported zenbleed PoC that no longer
shows any obtained text.  (a similar module-version of it stopped
the PoC on a ryzen 3950x without having to reboot.)

https://www.amd.com/en/resources/product-security/bulletin/amd-sb-7008.html
https://lock.cmpxchg8b.com/zenbleed.html

(martin)

Ticket #1664

(martin)

Pull up following revision(s) (requested by mrg in ticket #1664):

sys/arch/x86/include/specialreg.h: revision 1.207
sys/arch/x86/x86/errata.c: revision 1.31

x86: turn off zenbleed chicken bit on Zen2 cpus.

this is based upon Taylor's original work.  i just made the list
of CPUs to run on correct as i could determine.  (also, add some
Zen3 and Zen4 cpuids not yet used by any errata.)

(might be nice to have a better way to expression revision ranges
rather than specific cpuid matches, eg, 0x30-0x4f models in a cpu
family, etc.)

tested on ryzen 3600, and a ported zenbleed PoC that no longer
shows any obtained text.  (a similar module-version of it stopped
the PoC on a ryzen 3950x without having to reboot.)

https://www.amd.com/en/resources/product-security/bulletin/amd-sb-7008.html
https://lock.cmpxchg8b.com/zenbleed.html

(martin)

Ticket #243

(martin)

Pull up following revision(s) (requested by mrg in ticket #243):

sys/arch/x86/include/specialreg.h: revision 1.207
sys/arch/x86/x86/errata.c: revision 1.31

x86: turn off zenbleed chicken bit on Zen2 cpus.

this is based upon Taylor's original work.  i just made the list
of CPUs to run on correct as i could determine.  (also, add some
Zen3 and Zen4 cpuids not yet used by any errata.)

(might be nice to have a better way to expression revision ranges
rather than specific cpuid matches, eg, 0x30-0x4f models in a cpu
family, etc.)

tested on ryzen 3600, and a ported zenbleed PoC that no longer
shows any obtained text.  (a similar module-version of it stopped
the PoC on a ryzen 3950x without having to reboot.)

https://www.amd.com/en/resources/product-security/bulletin/amd-sb-7008.html
https://lock.cmpxchg8b.com/zenbleed.html

(martin)

Add dkctl(8) and gpt(8) to the evbmips ramdisk

The installer provided by octeon.img creates GPT wedges, which may
later need to be recovered via the standard sysinst installer provided
by netbsd-INSTALL_OCTEON.

(gutteridge)

Add dk[0-3] to the device nodes provided in the evbmips ramdisk

The installer provided by octeon.img creates wedges, which may later
need to be recovered via the standard sysinst installer provided by
netbsd-INSTALL_OCTEON.

(gutteridge)

x86: turn off zenbleed chicken bit on Zen2 cpus.

this is based upon Taylor's original work.  i just made the list
of CPUs to run on correct as i could determine.  (also, add some
Zen3 and Zen4 cpuids not yet used by any errata.)

(might be nice to have a better way to expression revision ranges
rather than specific cpuid matches, eg, 0x30-0x4f models in a cpu
family, etc.)

tested on ryzen 3600, and a ported zenbleed PoC that no longer
shows any obtained text.  (a similar module-version of it stopped
the PoC on a ryzen 3950x without having to reboot.)

https://www.amd.com/en/resources/product-security/bulletin/amd-sb-7008.html
https://lock.cmpxchg8b.com/zenbleed.html

(mrg)

x86/errata.c: Only say the errata revision search for cpu0.

(riastradh)

x86/errata.c: Say what revision we're searching for.

(riastradh)

x86/errata.c: Link to original AMD errata guide.

This one is no longer updated; need to link to newer ones for
individual families too.  That's where all the cryptic nomenclature
comes from here.

(riastradh)

Tickets #240 - #242

(martin)

Pull up following revision(s) (requested by riastradh in ticket #242):

lib/libc/time/localtime.c: revision 1.137
lib/libc/time/zdump.c: revision 1.61
lib/libc/time/NEWS: revision 1.41
lib/libc/time/zic.c: revision 1.88
lib/libc/time/zic.c: revision 1.89
lib/libc/time/private.h: revision 1.65

Apply 9cfe9507fcc22cd4a0c4da486ea1c7f0de6b075f for C23 attribute compliance.
Requested by Jan-Benedict Glaw.

put attributes first for c23 compliance.

(martin)

Pull up following revision(s) (requested by nat in ticket #240):

usr.sbin/bta2dpd/bta2dpd/sbc_encode.c: revision 1.12
usr.sbin/bta2dpd/bta2dpd/sbc_encode.c: revision 1.13

Add thottling when playing from file.
This avoids rapid playback when playing from file with affected devices.

Playback using pad(4) is still preferred ad gives a better result.

Playback from pad(4) is unaffected by this change.

Reorder for readability.
No functional change intended.

(martin)

Pull up following revision(s) (requested by nat in ticket #241):

usr.sbin/bta2dpd/bta2dpd/bta2dpd.8: revision 1.7
usr.sbin/bta2dpd/bta2dpd/bta2dpd.8: revision 1.8
usr.sbin/bta2dpd/bta2dpd/bta2dpd.8: revision 1.9
usr.sbin/bta2dpd/bta2dpd/bta2dpd.8: revision 1.10

Manual page fixups.

Fix EXAMPLES section.

Use correct flags when used with pad(4).
Use correct encoding format (sinear_le).

Bump date for previous commit.

Add BUGS section to manual page.
This notes the problems with playback from file on certain devices.

Fix formatting.

(martin)

MI and x86 efiboot: Fix device path type for PXE boot device.

It should be Messaging, not Media Device Path. Taken from OpenBSD, for
which this code was originally written:

https://cvsweb.openbsd.org/src/sys/arch/amd64/stand/efiboot/efipxe.c#rev1.6

> Network-based device paths use Messaging and not Media types.  Thus
> in reality the depth was always -1 which made the compare function
> a No-Op.  Properly check the device path depth value and look for
> the Messaging type instead to find the correct NIC.  This check
> never worked before and was uncovered by the last change.
>
> Regression noticed by bluhm@

(rin)

macppc/floppies: Use x_ variants of disklabel, fsck_ffs, and newfs.
Fix ramdisk overflow with clang.

- For fsck_ffs(8) and newfs(8), Apple UFS is still supported.
- 2.88MB of ramdisk size can be bumped if really required.

(rin)

efiboot/x86: Add serial console support via raw I/O port access

Unfortunately, some (most?) UEFI implementations do not support
com ports by ``Serial I/O Protocol''.

``PNP0501-0'' and friends are not recognized also.

In this case, if user explicitly requires to switch to serial
console by ``consdev'' command, try to use raw I/O port access.

Ugly, but what FreeBSD does, at least.

Proposed as PR port-amd64/57523

(rin)

s/probaby/probably/ in comments.

(andvar)

bsd.own.mk: force MKSTRIPSYM=no if MKDEBUG=yes

Local symbols are desirable with MKDEBUG=yes, so don't strip them.
Per request from mrg@

(lukem)

bsd.own.mk: style: block comments, move CC_flag vars

Add #{ .. #} block comments to a long conditional.

Move the CC_flag variables closer to other compiler-related
variables instead of being in the middle of some arch-specific
overrides.

(lukem)

PR/57261: hppa should be converted to __HAVE_SYSCALL_INTERN

Provide syscall_intern and use the md_syscall in syscall trap handling.

(skrll)

Use __BIT

(skrll)

port-arm/57388: Minor bug fix in bcopy.S

Use correct register to check alignment of destination in backwards copy.

Patch from Antoni Pokusinski. Thanks.

(skrll)

KASSERT -> KASSERTMSG

(skrll)

ENTRY / END indentation. NFCI.

(skrll)

Indentation consistency. NFCI.

(skrll)

USE __BIT() for CPUF_* flags. NFCI.

(skrll)

Add support Intel 700 series chipset and Alder Lake-N devices.

(msaitoh)

Regen.

(msaitoh)

Add Intel 700 series chipset devices.

(msaitoh)

mark gl_mangle.h and glx_mangle.h as mesa_ver=19.

(mrg)

xen: declare 'default_consinfo' as extern in a header

this makes pvh_consinit.c actually compile with CONS_OVERRIDE set.
i didn't see any good header to add to (bootinfo.h and cpu.h both
seem to be poor choices but were considered), hence the new one
with just this definition.

(mrg)

bsd.README: sync with mk.conf(5), deduping entries

Remove most duplication of variables already documented in mk.conf(5).
Add lists of supported and obsolete mk.conf(5) variables
so that searches in this document at least find them
(as per BUILDING).

Sync text from mk.conf(5) for entries that remain.
Explicitly document DESTDIR and RELEASEDIR.

Replace MKMAN=no antipattern with NOMAN=.

Consistently use "option" versus "flag"; option is more generic,
and may include a flag with an argument.

(lukem)

BUILDING: list known mk.conf vars. tweak section headers

List known supported and obsolete mk.conf(5) variables,
so that searches in this document at least find them.
This partially reverts my revision 1.144 on 2023-06-12
which completely removed the duplication with mk.conf(5).

Remove quote characters from some section headings;
it makes it harder to search for a section by known name
when rendering in markup variations that use smart quotes.

(lukem)

mk.conf(5): tweak BUILDID and option. sort MKKDEBUG

Clarify BUILDID behaviour with OBJMACHINE, and remove unnecessary sentence.
Use option not flag; option may include a flag with an argument.
Sort MKKDEBUG before MKKERBEROS4.

(lukem)

i386/XEN3PAE_DOMU: Pass -g to build debug data like GENERIC.

Needed for CTF data by dtrace when MKDEBUG=no MKDEBUGKERNEL=no.

XXX pullup-10

(riastradh)

i386/XEN3PAE_DOM0: Pass -g to build debug data like GENERIC.

Needed for CTF data by dtrace when MKDEBUG=no MKDEBUGKERNEL=no.

XXX pullup-10

(riastradh)

amd64/XEN3_DOMU: Pass -g to build debug data like GENERIC.

Needed for CTF data by dtrace when MKDEBUG=no MKDEBUGKERNEL=no.

XXX pullup-10

(riastradh)

Don't specify a volume alias if none is given instead of using a default
name. Otherwise iscsictl fails for using non-unique names if more than
one volume is used.

(mlelstv)

regen.

(msaitoh)

Add Alder Lake-N and Raptor Lake devices.

(msaitoh)

Remove backwards compat.

(kim)

BUILDING: clarify environ vs mk.conf behaviour

Document that variables set in the environment or the
nbmake-MACHINE wrapper script don't override variable
assignments in mk.conf(5), and that using ?= in mk.conf
allows environment / nbmake-MACHINE variables to override
mk.conf.

Consistently refer to "nbmake-MACHINE wrapper script".

(lukem)

tools/compat/compat_defs.h: improve NAME_MAX compat

Move the NAME_MAX workaround to the <limits.h> section,
and don't assume that _XOPEN_NAME_MAX is available.

(lukem)

bsd.own.mk: Use MACHINE_ARCH for default MKCTF/MKDTRACE=yes x86.

The substantive impact of this is that it enables MKCTF=yes for Xen
kernels.  This is a change because, when building a Xen kernel
(XEN3_DOM0, XEN3_DOMU), MACHINE is set to `xen', not to `i386' or
`amd64', so the conditional never took effect.

(The side effect of setting MKDTRACE=yes when building Xen kernels is
unlikely to matter; that affects module and userland builds.)

PR port-xen/57535

XXX pullup-10

(riastradh)

lint: use TOOL_SED instead of sed

Use ${TOOL_SED} instead of (the host) sed, as the
latter doesn't necessarily support the non-POSIX -E.

PR bin/57533

(lukem)

Also use XEN_BSIZE when computing the number of bytes for format_bytes().
While there note in a comment that sc_sectors is in XEN_BSIZE units

(bouyer)

rbtree(3): No (rb_tree_t) for constant initializer.

This is a compound literal, which is not formally valid as a constant
initializer.

XXX pullup-10

(riastradh)

The disk size reported in the xenstore is always in XEN_BSIZE units,
not sector-size. Should fix the issue reported by Christian Kujau
on netbsd-users and port-xen.

(bouyer)

Sort by number. No functional change.

(msaitoh)

actually re-gen with GCC 10.5 in gcc.old and earmv4*.

fixes armv4* builds.

(mrg)

lint: reword message about non-constant initializer

(rillig)

tests/lint: document an example of a non-constant initializer

(rillig)

Add missing dot.

(nat)

Mention changes to bta2dpd(8).

(nat)

Reorder for readability.

No functional change intended.

XXX pullup-10.

(nat)

i386/XEN3PAE_DOMU: Enable KDTRACE_HOOKS.

XXX pullup-10

(riastradh)

i386/XEN3PAE_DOM0: Enable KDTRACE_HOOKS.

XXX pullup-10

(riastradh)

amd64/XEN3_DOMU: Enable KDTRACE_HOOKS.

XXX pullup-10

(riastradh)

Use kmem_free instead of kmem_intr_free, as key_freesaval() is not called in softint after key.c:r1.223.

E.g. key_freesaval() was called the following call path before SAD MP-ify.
  esp_input_cb()
    KEY_FREESAV()
      key_freesav()
        key_delsav()
          key_freesaval()

ok'ed by ozaki-r@n.o.

(knakahara)

octeon_gmx.c: fix spelling in (default disabled) error messages

(gutteridge)

socket.2: simplify and improve some phrasing

(gutteridge)

amdgpu: Mark float-dependent variables volatile.

This way they are computed -- using FP instructions -- before
DC_FP_END, after which point FP instructions will trap or behave
unpredictably.

This is a workaround, not a proper solution -- really, the
DC_FP_START/END calls should happen in a separate compilation unit
built without -mhard-float or whatever, but that's a lot more work to
make happen for now.

PR kern/57059

XXX pullup-10

(riastradh)

pax: fix -Wformat for host tool

(lukem)

pax: need <strings.h> for ffs()

Fixes warning as host tool on NetBSD 9.99

(lukem)

various debug updates for some usb drivers

- several new *_DEBUG_DEFAULT options that allow usb debug values to
  be set to a default that is non-zero:
  EHCI_DEBUG_DEFAULT, UGEN_DEBUG_DEFAULT, URTWN_DEBUG_DEFAULT,
  UMS_DEBUG_DEFAULT, and USB_DEBUG_DEFAULT
- ugen debug uses fewer usbhist lines for the same info
- ums.c converted from printf() to usbhist

(mrg)

ctags: fix pointer-sign issues

Refactor init() to avoid -Wpointer-sign for host builds.
Uses same cast pattern used in ctags.h.

(lukem)

avoid calling drm_sched_fini() for AMDGPU_RING_TYPE_KIQ rings.

drm_sched_init() is not called for these rings, and we'd panic trying
to mutex_destroy() a mutex that wasn't initialised.

this seems like a general bug, not a bug in netbsd port.

(mrg)

tools/compat: regen

(lukem)

tools: binstall can use posix_spawn() instead of vfork()

(lukem)

tools/make: regen

(lukem)

tools/compat: regen

(lukem)

tools/make: build with -Wno-deprecated-declarations

Use NB_CHECK_CC_FLAG() from ../tools/compat to determine
if the compiler supports -Wno-deprecated-declarations.
Avoids sooking about vfork() on macOS.

(lukem)

tools/compat: provide nb_check_cc_flag.m4

Implement m4 macro NB_CHECK_CC_FLAG(FLAG, [VAR=FLAG_DERIVED])
Determine if the C compiler supports FLAG,
and sets output variable VAR to FLAG if FLAG is supported.

If VAR is not provided, default to FLAG_DERIVED, which is
FLAG converted to upper-case and all special characters
replaced with "_", and the result prepended with "CC_".
FLAG_DERIVED is appended to the m4 macro NB_CHECK_CC_FLAG_VARS.
E.g., if FLAG is "-Wexample=yes", FLAG_DERIVED is "CC_WEXAMPLE_YES".

This is implemented in separate buildaux/nb_check_cc_flag.m4,
based on configure's NB_CC_FLAG(), but supports VAR override.

Adapt configure from internal NB_CC_FLAG() to NB_CHECK_CC_FLAG().

(Note: AX_CHECK_COMPILE_FLAG() from autoconf-archive could be enhanced
to support the clang and gcc workarounds I've implemented here.)

(lukem)

tools/make/buildmake.sh: fix quoting

autoconf 2.69 generates @DEFS@ with definitions with
quoted embedded spaces, so rework how docmd() is invoked
to avoid weird quoting issues.

As part of this, remove all the shell variables containing @var@
and just use @var@ directly.

(lukem)

tools/make/configure.ac: style tweaks

Consistently use () in zero argument macro calls.
Display results at end.

(lukem)

Fix formatting.

XXX pullup-10.

(nat)

Add BUGS section to manual page.

This notes the problems with playback from file on certain devices.

XXX pullup-10

(nat)

Add thottling when playing from file.

This avoids rapid playback when playing from file with affected devices.

Playback using pad(4) is still preferred ad gives a better result.
Playback from pad(4) is unaffected by this change.

XXX pullup-10.

(nat)

xhci(4): Don't panic on suspend if previous suspend/resume failed.

Trying to resume again probably won't make the situation much worse,
but panicking can definitely make it worse.

XXX pullup-10

(riastradh)

Ticket #239

(martin)

Apply patch, requested by mrg in ticket #239:

lib/libgcc/libgcov/arch/earmv4/gcov-iov.h
lib/libgcc/libgcov/arch/earmv4eb/gcov-iov.h
lib/libstdc++-v3/arch/earmv4/c++config.h
lib/libstdc++-v3/arch/earmv4/gstdint.h
lib/libstdc++-v3/arch/earmv4/symver-config.h
lib/libstdc++-v3/arch/earmv4eb/c++config.h
lib/libstdc++-v3/arch/earmv4eb/gstdint.h
lib/libstdc++-v3/arch/earmv4eb/symver-config.h
usr.bin/gcc/arch/earmv4/bversion.h
usr.bin/gcc/arch/earmv4/configargs.h
usr.bin/gcc/arch/earmv4/defs.mk
usr.bin/gcc/arch/earmv4/plugin-version.h
usr.bin/gcc/arch/earmv4eb/bversion.h
usr.bin/gcc/arch/earmv4eb/configargs.h
usr.bin/gcc/arch/earmv4eb/defs.mk
usr.bin/gcc/arch/earmv4eb/plugin-version.h

regen, missed in previous mknative run.
Cleanup old (no longer used) configs and remove:

lib/libbacktrace/arch/coldfire/backtrace-supported.h
lib/libbacktrace/arch/coldfire/config.h
lib/libbacktrace/arch/or1k/backtrace-supported.h
lib/libbacktrace/arch/or1k/config.h
lib/libgcc/arch/coldfire/auto-target.h
lib/libgcc/arch/coldfire/defs.mk
lib/libgcc/arch/or1k/auto-target.h
lib/libgcc/arch/or1k/defs.mk
lib/libgcc/libgcov/arch/coldfire/defs.mk
lib/libgcc/libgcov/arch/coldfire/gcov-iov.h
lib/libgcc/libgcov/arch/or1k/defs.mk
lib/libgcc/libgcov/arch/or1k/gcov-iov.h
lib/libgomp/arch/coldfire/config.h
lib/libgomp/arch/coldfire/libgomp.spec
lib/libgomp/arch/coldfire/libgomp_f.h
lib/libgomp/arch/coldfire/omp.h lib/libgomp/arch/or1k/config.h
lib/libgomp/arch/or1k/libgomp.spec
lib/libgomp/arch/or1k/libgomp_f.h lib/libgomp/arch/or1k/omp.h
lib/libiberty/arch/coldfire/config.h
lib/libiberty/arch/or1k/config.h
lib/libobjc/arch/coldfire/config.h
lib/libobjc/arch/coldfire/defs.mk
lib/libobjc/arch/or1k/config.h lib/libobjc/arch/or1k/defs.mk
lib/libstdc++-v3/arch/coldfire/c++config.h
lib/libstdc++-v3/arch/coldfire/cxxabi_tweaks.h
lib/libstdc++-v3/arch/coldfire/defs.mk
lib/libstdc++-v3/arch/coldfire/gstdint.h
lib/libstdc++-v3/arch/or1k/c++config.h
lib/libstdc++-v3/arch/or1k/cxxabi_tweaks.h
lib/libstdc++-v3/arch/or1k/defs.mk
lib/libstdc++-v3/arch/or1k/gstdint.h
usr.bin/gcc/arch/coldfire/all-tree.def
usr.bin/gcc/arch/coldfire/auto-host.h
usr.bin/gcc/arch/coldfire/bconfig.h
usr.bin/gcc/arch/coldfire/bversion.h
usr.bin/gcc/arch/coldfire/config.h
usr.bin/gcc/arch/coldfire/configargs.h
usr.bin/gcc/arch/coldfire/defs.mk
usr.bin/gcc/arch/coldfire/gthr-default.h
usr.bin/gcc/arch/coldfire/gtyp-input.list
usr.bin/gcc/arch/coldfire/insn-modes.h
usr.bin/gcc/arch/coldfire/multilib.h
usr.bin/gcc/arch/coldfire/plugin-version.h
usr.bin/gcc/arch/coldfire/tm.h
usr.bin/gcc/arch/or1k/all-tree.def
usr.bin/gcc/arch/or1k/auto-host.h
usr.bin/gcc/arch/or1k/bconfig.h
usr.bin/gcc/arch/or1k/bversion.h
usr.bin/gcc/arch/or1k/config.h
usr.bin/gcc/arch/or1k/configargs.h
usr.bin/gcc/arch/or1k/defs.mk
usr.bin/gcc/arch/or1k/gthr-default.h
usr.bin/gcc/arch/or1k/gtyp-input.list
usr.bin/gcc/arch/or1k/multilib.h
usr.bin/gcc/arch/or1k/plugin-version.h
usr.bin/gcc/arch/or1k/tm.h
usr.bin/libcpp/arch/coldfire/config.h
usr.bin/libcpp/arch/or1k/config.h
usr.bin/libdecnumber/arch/coldfire/config.h
usr.bin/libdecnumber/arch/or1k/config.h

(martin)

regen for earmv4/earmv4eb.

(mrg)

entropy.7: add a missing word to a sentence

While here, bump the date to reflect when substantive changes were last
made.

(gutteridge)

lint: work around a possible compiler error on arm and powerpc

(rillig)

mknative-gcc for earmv4*.

delete obsolete or1k and coldfire configurations.

(mrg)

Warn to clean gcc objectdirs for updated compiler.

(martin)

BUILDING: consistency fix

(lukem)

openssh-9.3p2 out

(wiz)

Tickets #236 - #238

(martin)

Pull up following revision(s) (requested by bouyer in ticket #238):

sys/arch/xen/xen/xen_clock.c: revision 1.10
sys/arch/xen/xen/xen_clock.c: revision 1.12

Unmask event after VCPUOP_stop_periodic_timer and
initializing ci->ci_xen_hardclock_systime_ns, to avoid a possible race with
xen_timer_handler()

Unmask event after arming the one-shot timer in clock initialisation,
to avoid a possible race with xen_timer_handler() updating
ci_xen_hardclock_systime_ns while we're reading it.

Pointed out by Taylor R Campbell

(martin)

Pull up following revision(s) (requested by rin in ticket #237):

usr.bin/xlint/lint1/decl.c: revision 1.361
doc/HACKS: revision 1.239

lint1:initdecl(): Remove hack for now-fixed PR port-sh3/56311
doc/HACKS: PR port-sh3/56311 has been fixed.

(martin)

Pull up following revision(s) (requested by rin in ticket #236):

external/gpl3/gcc/dist/gcc/config/sh/sh.md: revision 1.3

gcc: Fix peephole optimization bug for sh3 (PR port-sh3/56311)

Cherry-picked from upstream:
https://gcc.gnu.org/git/gitweb.cgi?p=gcc.git;h=4dbb3af1efe55174a714d15c2994cf2842ef8c28
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101469

Now, the PR confirmed fixed. No regression observed for full ATF run on
landisk (compared with GCC built with this peephole optimization being
removed by hand).

We thank Oleg Endo, the author of the upstream commit.
Also thanks hgutch@ and uwe@ for analysis.

(martin)

Note import of dhcpcd-10.0.2

(roy)

Sync with dhcpcd-10.0.2

(roy)

- make colour depth in fb mode configurable, default to 8bit
- support ioctl(FBIO*CMAP)

(macallan)

build.sh: use mk.conf (if present) as MAKECONF

Set MAKECONF to mk.conf in the same directory as build.sh
(i.e., the top of the source tree) if mk.conf is present.
This means unprivileged users can benefit from mk.conf(5)
semantics without write privileges to /etc/mk.conf, and
also simplifies per-source-directory configuration.

Fail early if $MAKECONF is empty, since build.sh will fail
anyway after rebuilding nbmake.

Ensure that the computed MAKECONF isn't empty, and then
always set MAKECONF in the makewrapper (nbmake-$MACHINE).

Improve some formatting consistency in BUILDING.mdoc,
(even if BUILDING is generated without markup).

(lukem)

hdafg(4): Do hotplug detection in kthread, not callout.

This can sometimes take a while (~1ms), and the logic to suspend the
callout on device suspend/resume was racy (PR kern/57322).

XXX pullup-8
XXX pullup-9
XXX pullup-10

(riastradh)

x86/fpu: In kernel mode fpu traps, print the instruction pointer.

(riastradh)

autoconf(9): Print `waiting for devices' normally once a minute.

(riastradh)

rbtree(3): Fix RBDEBUG build with RB_TREE_INITIALIZER.

XXX pullup-10

(riastradh)

libc: Use RB_TREE_INITIALIZER to nix initfini.c/_env.c coupling.

Doesn't actually help to remove _env.c from programs that don't use
it, though, because various internal things like __diagassert13 still
call getenv.

(riastradh)

rbtree(3): New RB_TREE_INITIALIZER macro.

Allows static initialization of an rbtree.

XXX pullup-10

(riastradh)

acpiec(4): One more debug message about read/write polling timeout.

(riastradh)

acpiec(4): Take a lock around acpiec_cold updates.

Otherwise we race with readers -- probably harmlessly, but let's
avoid the appearance of problems.

XXX Maybe acpiec_suspend and acpiec_shutdown should interrupt
transactions and force them to fail promptly?

XXX This looks bad because acpiec_cold is global and sc->sc_mtx
doesn't look like it's global, but we expect to have only one
acpiec(4) device anyway from what I understand.  Maybe we should move
acpiec_cold into the softc?

(riastradh)

acpiec(4): Factor out if (state == FREE) cv_signal(sc_cv).

In principle this could have a functional change, but at worst, it is
to signal more wakeups than needed, which should always be safe.

(riastradh)

acpiec(4): Pass softc, not device_t, to acpiec_space_handler.

Better to keep the device_t isolated to public interfaces.  Simpler
internally this way.

No functional change intended.

(riastradh)

acpiec(4): Pass softc, not device_t, to acpiec_gpe_query thread.

Simpler.

No functional change intended.

(riastradh)

acpiec(4): Pass softc, not device_t, to acpiec_read/write.

Simpler, type-safer.

No functional change intended.

(riastradh)

acpiec(4): Pass softc, not device_t, to acpiec_lock/unlock.

Simpler, type-safer.

No functional change intended.

(riastradh)

acpiec(4): Pass softc, not device_t, to acpiec_gpe_handler.

Simpler.

No functional change intended.

(riastradh)

acpiec(4): Pass softc, not device_t, to acpiec_callout.

Simpler.

No functional change intended.

(riastradh)

acpiec(4): Pass softc, not device_t, to acpiec_gpe_state_machine.

Simpler, type-safer.

No functional change intended.

(riastradh)

acpiec(4): Factor wait logic out.

No functional change intended.

(riastradh)

acpiec(4): Merge returns in acpiec_read/write.

No functional change intended.

(riastradh)

acpiec(4): Don't touch sc->sc_state outside sc->sc_mtx.

(riastradh)

acpiec(4): Fix cv_timedwait abuse in acpiec_read/write.

(riastradh)

acpiec(4): Fix interrupt wait loop in acpiec_gpe_query thread.

(riastradh)

acpiec(4): Fix cv_wait loop around sc->sc_got_sci.

That is, make it actually loop as required, so it gracefully handles
spurious wakeups instead of barging into invalid states.

(riastradh)

acpiec(4): Set sc_got_sci only when a transaction is over.

Before, when the acpiec thread noticed an SCI had been requested and
entered acpiec_gpe_state_machine to send the query command, it would
see the SCI is still requested -- because it had yet to acknowledge
it by setting the query command! -- and think the EC was asking for a
_second_ SCI.

So once the first SCI transaction was over, it would start a second
one, even though the EC hadn't asked for another -- and this would
wedge on some ECs.

Now, acpiec_gpe_state_machine waits to see what state we transition
to before taking the SCI bit to mean we need to notify the acpiec
thread to handle another query.

That way, when the acpiec thread enters acpiec_gpe_state_machine with
EC_STATE_QUERY, it can send the query command first, with the side
effect of clearing the SCI bit in subsequent reads of the status
register, and it won't think another SCI has been requested until it
returns to EC_STATE_FREE and sees the SCI bit set again in the status
register.

Possibly relevant PRs:

PR kern/53135
PR kern/52763
PR kern/57162

(riastradh)

acpiec(4): Assert state is free when we start a transaction.

No functional change intended.

(riastradh)

acpiec(4): Sprinkle comments.

Note where this code is abusing cv_wait and needs a loop to handle
spurious wakeups.

No functional change intended.

(riastradh)

acpiec(4): Clarify lock order and sprinkle lock assertions.

No functional change intended.

(riastradh)

acpiec(4): New ACPIEC_DEBUG option.

Value is bit mask of debug messages to enable.

Enable in x86/ALL kernels.

No functional change intended when the option is off.

(riastradh)

acpiec(4): Record device_t self.

Not used yet, to be used soon for device_printf and to allow making
some of the internal functions a little more type-safe later.

(riastradh)

provide more things needed by new subr_prf.c.

(mrg)

doc/HACKS: PR port-sh3/56311 has been fixed.

(rin)

lint1:initdecl(): Remove hack for now-fixed PR port-sh3/56311

(rin)

Sync with external/gpl3/gcc:
https://mail-index.netbsd.org/source-changes/2023/07/18/msg146078.html
----
gcc: Fix peephole optimization bug for sh3 (PR port-sh3/56311)

Cherry-picked from upstream:
https://gcc.gnu.org/git/gitweb.cgi?p=gcc.git;h=4dbb3af1efe55174a714d15c2994cf2842ef8c28
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101469

Now, the PR confirmed fixed. No regression observed for full ATF run on
landisk (compared with GCC built with this peephole optimization being
removed by hand).

We thank Oleg Endo, the author of the upstream commit.

Also thanks hgutch@ and uwe@ for analysis.

(rin)

gcc: Fix peephole optimization bug for sh3 (PR port-sh3/56311)

Cherry-picked from upstream:
https://gcc.gnu.org/git/gitweb.cgi?p=gcc.git;h=4dbb3af1efe55174a714d15c2994cf2842ef8c28
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101469

Now, the PR confirmed fixed. No regression observed for full ATF run on
landisk (compared with GCC built with this peephole optimization being
removed by hand).

We thank Oleg Endo, the author of the upstream commit.

Also thanks hgutch@ and uwe@ for analysis.

(rin)

device_printf(9): Lock to avoid interleaving output.

XXX pullup-9
XXX pullup-10

(riastradh)

timecounter(9): Sprinkle membar_consumer around th->th_generation.

This code was apparently written under the misapprehension that
membar_producer on one side is good enough, but that doesn't
accomplish anything other than making the code unnecessarily obscure.

For semantics, you always always always need memory barriers to come
in pairs, with membar_consumer on the reading side if you want the
membar_producer to have on the writing side to have any useful
effect.

It is unfortunate that this might hurt performance, but that's an
unfortunate consequence of the design made without understanding
memory barriers, not an unfortunate consequence of the memory
barriers.

If it is really critical for the read side to avoid memory barriers,
then the write side needs to issue an IPI or xcall to effect memory
barriers -- at higher cost to the write side, of course.

(riastradh)

timecounter(9): Use atomic_store_release/load_consume for timehands.

This probably fixes real bugs on Alpha and makes the synchronization
pattern clearer everywhere.

(riastradh)

timecounter(9): Use seqlock for atomic snapshots of timebase.

(riastradh)

i386/LEGACY: Enable ATA_DOWNGRADE_MODE.

PR kern/57362

(riastradh)

ata(4): Add ATA_DOWNGRADE_MODE to opt_ata.h.

This way adding it to kernel config will trigger recompilation.

(riastradh)

Revert "timecounter(9): Use an ipi barrier on time_second/uptime rollover."

Evidently rump doesn't have ipi, so this won't work unless we have an
alternate approach for rump.

(riastradh)

signal(7): Clarify semantics of SIGCHLD with SIG_IGN or SA_NOCLDWAIT.

The semantics is not just a nonportable hack for SysV compatibility;
it is enshrined in POSIX.

Related: PR 57527

(riastradh)

timecounter(9): No static; committed wrong version of patch.

(riastradh)

timecounter(9): Limit scope of time__second/uptime.

Relevant only if __HAVE_ATOMIC64_LOADSTORE -- not updated otherwise.

(riastradh)

timecounter(9): Use an ipi barrier on time_second/uptime rollover.

This way we only need __insn_barrier, not membar_consumer, on the
read side.

(riastradh)

timecounter(9): Revert last -- timecounter_lock is already IPL_HIGH.

(riastradh)

timecounter(9): Ward off interrupts during time_second/uptime update.

Only relevant during 32-bit wraparound, so the potential performance
impact of using splhigh here is negligible; indeed, we would have to
go out of our way to exercise this in tests before it will ever
happen in the next century.

(riastradh)

timecounter(9): Fix thinko in previous.

Swapped the wrong variable in this mental macro expansion!

(riastradh)

sys/param.h: Welcome to NetBSD 10.99.6!

Courtesy of:

- one uvmfault rndsource, not one per CPU
- removal of non-atomic time_second and time_uptime symbols
- struct syncobj::sobj_name

(riastradh)

uvm(9): One rndsource for faults -- not one per CPU.

All relevant state is per-CPU anyway; the only substantive difference
this makes is how many entries appear in `rndctl -l' output and what
they are called -- formerly the somewhat confusing `cpuN', meaning
`page faults on cpuN', and now just `uvmfault'.  I don't think
there's any real value in being able to enable or disable measurement
or counting of page faults on one CPU vs others, so although this
could be a minor compatibility change, it's hard to imagine it
matters much.

XXX kernel ABI change in struct cpu_info

(riastradh)

kern: Make time_second and time_uptime macros that work atomically.

These use atomic load on platforms with atomic 64-bit load, and
seqlocks on platforms without.

This has the unfortunate side effect of slightly reducing the real
times available on 32-bit platforms, from ending some time in the
year 584942417218 AD, available on 64-bit platforms, to ending some
time in the year 584942417355 AD.  But during that slightly shorter
time, 32-bit platforms can avoid bugs arising from non-atomic access
to time_uptime and time_second.

Note: All platforms still have non-atomic access problems for
bintime, binuptime, nanotime, nanouptime, &c.  This can be addressed
by putting a seqlock around timebasebin and possibly some other
variable -- to be done in a later change.

XXX kernel ABI change -- deleting symbols

(riastradh)