attribute police

(christos)

document cosl with MLINKS and in the man page

(maya)

Expand to cover long double somewhat.

The given data is for double, so use DBL_EPSILON and don't expect better
results.

(maya)

Restore mapping of file id to pid/type/fd.
Use 64bit file id to allow for 32bit fd and 25-26bit pid.

(mlelstv)

Fix typo in 'exceptions'

(kamil)

Fix typo

(kamil)

Introduce check for the support of FPU exceptions

If FPU exceptions are unsupported, skip the SIGFPE crash tests.

Reuse code from tests/kernel/h_segv.c

(kamil)

Adapt to audio2.
- Drop INDEPENDENT property.  Both play and rec share the sample rate.

(isaki)

Adapt to audio2.
- The sample rate seems to be determined by peripherals so
  move uda1341_formats[] structure from ic/uda1341.c to audio_mini2440.c.
- XXX Resetting the device in open() (and close()) might be harmful
  but I don't know details about what does this reset do.

(isaki)

Add missing mutex_exit() in interrupt handler.

(isaki)

Adapt to audio2.
- Starting playback in init_output() no longer works.
  Introduce sc_active flag instead and do it in start_output().

(isaki)

Do not allow 0 as ILL si_code as it's reserved for SI_USER

(kamil)

In ATF t_ptrace_wait* SIGILL tests allow any si_code

These codes are not portable between CPUs and kernels.

Change the checks to return any valid si_code in the defined range of ILL
values.

PR port-sparc/54140 by Andreas Gustafsson

(kamil)

Fix build, install wsmoused

(roy)

KNF. No functional change.

(msaitoh)

No functional change:
- Use __arraycount().
- u_int_{8,16,32}_t -> uint_{8,16,32}_t
- KNF.
- Tabify.
- Remove extra space.

(msaitoh)

Link against expat.

Helps us pass a qt5-qtbase gbm configure check, and avoid PLIST issues.

(maya)

Don't install YP rc.d files with MKYP=no

(roy)

- KNF.
- Use __arraycount.
- Remove extra 'n' from debug printf.

(msaitoh)

KNF. No functional change.

(msaitoh)

Tyop in error message...

(pgoyette)

Better interactive SIGINT handling (when a trap is set), and other
cleanups to the trap code.  No longer silently ignore attempts to
do anything other than set SIGKILL or SIGSTOP to the default ('-")
state.  Don't include those in trap or trap -p output (the former
because they cannot be other than in default state, so simply aren't
included, the latter because it is pointless) but do list them
when requested with trap -p SIG.

Interactive mode SIGINT traps are now run ASAP, rather than after
a command has been entered (so the sequence ^C \n is no longer needed
to generate one).  Further, when trapped, in interactive mode,
while waiting for a user command, a SIGINT acts (aside from the
trap being run) just like when not trapped, aborts the command being
entered (rather than leaving it, which it did when libedit was in use)
prints a new prompt, and starts again (which is what should happen.)

Traps other than SIGINT (which has always been handled special in
interactive mode) are unaffected by this change, as are SIGINT traps
in non-interactive shells.    Or that is the intent anyway.

Fix an in_dotrap ref count bug (was never being decremented... that
was inserted in a place never executed) (relatively harmless) and
add/improve some trap/signal related DEBUG mode tracing.

(kre)

Add gem(4).

(msaitoh)

Expose device type. You can query it with e.g. drvctl -p ld0 disk-info/type.

(mlelstv)

Drop error path from C11 call_once

The original implementation of C11 threads(3) contained check for error
paths, but it was stripped in the calls that are documented to return
no status from an operation. Do the same in call_once(3).

(kamil)

change -h to -n because it is more "portable"

(christos)

Introduce minor changes to the C11 threading library

Switch tss_t type from int to pthread_key_t (no functional change as
pthread_key_t was already typedefed as int).

Noted by <enh from Google>.

Use C11 _Noreturn in thrd_exit(3) instead of NetBSD specific __dead.
The former is documented in the standard as an attribute of thrd_exit(3),
the latter is more portable to pre-C11 compilers, however C11 thread
support library needs C11 compiler for TLS anyway. __dead made a little bit
more point 3 years ago than today as 3 years ago pre-C11 compilers were
more common.

(kamil)

Match the structure order, for better cache utilization.

(maxv)

Provide the hardware error code for NVMM_EXIT_INVALID, useful when
debugging.

(maxv)

Fix compilation failure with gcc-8.
  Equal pointers to 'struct sigaction' should not be passed to sigaction(2).
  So, we pass NULL as an "old sigaction" structure.

(cheusov)

PR/54000: Andreag Gustafsson: Compile the rounding test with
-fround-math since with gcc-7, the default mode ignores fenv settings
(the same effect can be achieved with -O0 :-)

https://gcc.gnu.org/wiki/FloatingPointMath

(christos)

libpthread(3): Import C11 thread support library

(kamil)

Allow os_page sizes greater than the built-in page size. This can happen
for example for COMPAT_NETBSD32 sparc binaries (4K page size because of
MIN_PAGE_SIZE), running on sparc64 (8K pages).

(christos)

RFU

(sevan)

Adapt to audio2.

(isaki)

Adapt zaudio family to audio2.
- wm8731 and wm8750 can share zaudio_formats[].
- Drop INDEPENDENT property.  These share sample rate in play and rec.
- pxa2x0_i2s_setspeed() no longer needs to writeback sample rate.

(isaki)

Bump the libpthread(3) minor number to 4

Added C11 Threading library support.

(kamil)

Adapt to audio2.
- Some models don't have SLINEAR_LE but only SLINEAR_BE is enough.

(isaki)

Accept if bufsize returned from round_buffersize() is a multiple of
blksize.

(isaki)

Adapt to audio2.
- move resetting sc_cmd_index from drain to halt_output since
  drain is obsoleted.

(isaki)

Only warn about being unable to make a smaller stack if asking for debug
messages.

Between ASLR and stack clash random-sized padding, we can't make small
stacks reliably. Not doing so isn't a problem, so don't warn users about
it on every boot.

PR bin/51118

(maya)

Add a complete C11 threads(3) implementation

C11 Thread support library is a portable threading C API between OSs,
similar to std::threads in the C++ world.

The library is implemented as a thin shim over POSIX interfaces.

NetBSD implements the API as a part of the POSIX threading library
(libpthread(3)).

C11 threads(3) are in the process of making them an integral part
of the POSIX standard. The interface has been implemented in major
OSs and used with stopgap libraries for older versions of them.

C11 threading library is already used (with a stopgap implementation)
in the NetBSD distribution in MESA.

Original implementation by myself from 2016.

ATF tests are new and cover almost all interfaces.

Proposed on tech-userlevel@.

(kamil)

SIOCS is 'S'et function and the ioctl argument is ifreq.
SIOCG is 'G'et function and the ioctl armument is ifmediareq.
Before this change, SIOCG modify sc->sc_flowflags unexpectedly.
Don't hook SIOCGIFMEDIA becauise this driver uses MII(4) and ether_ioctl has
the hook(if_ethersubr.c rev. 1.160).
real bug because the modification is overriden in ifmedia_ioctl().
This driver might require some additional fixes for SIOCSIFMTU and other
ioctl()s.

XXX pullup-[78].

(msaitoh)

SIOCS is 'S'et function and the ioctl argument is ifreq.
SIOCG is 'G'et function and the ioctl armument is ifmediareq.
Before this change, SIOCG modify request unexpectedly but it doesn't cause a
real bug because the modification is overriden in ifmedia_ioctl().

(msaitoh)

This driver uses MII(4) and have hook SIOC[GS]IFMEDIA which just pass to
ifmedia_ioctl(), the hook is not required because ether_ioctl has it
(if_ethersubr.c rev. 1.160). This driver might require some additional fixes
for SIOCSIFMTU and other ioctl()s.

(msaitoh)

Fix typo.

(wiz)

This driver uses MII(4) and have hook SIOCGIFMEDIA which just pass to
ifmedia_ioctl(), the hook is not required because ether_ioctl has it
(if_ethersubr.c rev. 1.160). This driver might require some additional fixes
for SIOCSIFMTU and other ioctl()s.

(msaitoh)

This driver uses MII(4) and have hook SIOCGIFMEDIA which just pass to
ifmedia_ioctl(), the hook is not required because ether_ioctl has it
(if_ethersubr.c rev. 1.160). This driver might require some additional fixes
for SIOCSIFMTU and other ioctl()s.

(msaitoh)

Fix building am-utils with MKYP=no

(roy)

Fix building rpcbind with MKYP=no

We add enough of the YP headers to security.c so the security checks
can remain as is.

(roy)

Fix building librpcsvc for MKYP=no

(roy)

Don't install YP headers when MKYP=no

(roy)

KNF. No functional change.

(msaitoh)

ARP: Enable Address Defence again.

Revert the tentative/duplicated check and test for if it's been broadcast
or not. This reverts r1.245.

(roy)

Remove extra test for SIOCSIFMEDIA.

(msaitoh)

Clean up timeout logic. Now the system survives to a ccb timeout; before
the filesystem would stall because one I/O did never completes.

(bouyer)

Remove extra test of SIOCSIFMEDIA.

(msaitoh)

KNF. No functional change.

(msaitoh)

KNF. No functional change.

(msaitoh)

KNF. No functional change.

(msaitoh)

KNF. No functional change.

(msaitoh)

KNF. No functional change.

(msaitoh)

Add/improve debug output for some routines.

(blymn)

In fdt_add_boot_physmem make sure the memory range has pages available
before adding it to the fdt_physmem array.

Fixes a problem that jmcneill@ pointed out to me.

(skrll)

Fix leak of iba.iba_child_devices in fdtbus_attach_i2cbus().  Patch
provided by yarl-baudig %%at%% mailoo.org

(thorpej)

KNF. No functional change.

(msaitoh)

No functional change:
- IFM_AVALID and IFM_ACTIVE are NOT for the media word. Fix comment.
- RFU stands for Reserved for Future Use.

(msaitoh)

Omit files not marked executable from the signature database by default.

Closes PR kern/41669
Reviewed by <agc>

(sevan)

add missing symver-config.h

(mrg)

Add an ifdef _GLIBCXX_COMPAT_ that removes the versioned symbol nodes from
libstdc++ and enable it for now only for the random archives but not for the
shared object (although it does not break anything that I've tested to enable
it for the shared object too). Fixes static linking, which does not handle
multiple versioned symbols properly.

(christos)

Tickets #1245 and #1246

(martin)

Pull up following revision(s) (requested by bouyer in ticket #1246):

sys/arch/arm/arm32/pmap.c: revision 1.373 (via patch)

Fix a deadlock between the pool and pmap codes:

- cpu0 grabs the kernel lock (e.g. from a non-MPSAFE interrupt) and
  calls pool_get().
- cpu1 does a pool_get() on the same pool from MPSAFE code, which needs a
  pool_page_alloc(), which ends up in pmap_extract_coherency().

So cpu0 holds the kernel_lock and wants the pool lock. cpu1 holds the pool
lock and wants the kernel_lock in pmap_extract_coherency().

The pmap code should not rely on kernel_lock. Intead make the
pmap_kernel()->pm_obj_lock a IPL_VM lock and use it as pmap lock
(thus dropping the pmap test pmap_{acquire,release}_pmap_lock()).

This needs to be a IPL_VM because unlike user pmaps, this can be locked
from interrupt context.

Add a IPL_NONE lock for pmap_growkernel(). We can't use
pmap_kernel()->pm_obj_lock here because pmap_grow_map() may sleep.

Make pmap_lock (which may be locked with pm_obj_lock held) a IPL_VM
lock in all case.
reorder a few things to not call pool_get()/pool_put() (which may sleep)
with pm_obj_lock held.

Patch initially posted to port-arm@ on April 19, improved patch (per
suggestions from Nick Hudson and Jason Thorpe) on April 21.

(martin)

Pull up following revision(s) (requested by bouyer in ticket #1245):

sys/arch/arm/arm32/pmap.c: revision 1.372

Don't try to aquire/release the pmap lock when in ddb.

Avoids a deadlock when entering ddb, or on "mach cpu n" ddb command
(the pmap lock may already be held by another CPU, which is halted when
entering ddb).

Posted to port-arm@ on April 19.

(martin)

remove unused label

(christos)

Revert the part of the previous Whitespace commit which clearly was
functional.

This will probably come back with other changes and a proper commit
message.

(skrll)

Don't pass '-' to getopt(3).  My guess is that *very* old getopt
didn't handle "--" terminator, so this code was working around it and
not quite correctly at that.

Fixes weird output from

  $ uniq --long-option
  uniq: uniq: No such file or directory

(for more fun, run that command from /usr/bin as root).

Pointed out by Andreas Krey.

(uwe)

Adapt to audio2.
- XXX Need an accurate list of supported frequencies.

(isaki)

Add -h to the link command so that we replace the target file each time; this
should fix:

*** Failed target:  .BEGIN
*** Failed command: ln -sf /usr/src/sys/external/gpl2/dts/dist/arch/arm/boot/dts dts/arm
*** Error code 1 (ignored)
ln: dts/arm/dts: Permission denied

(christos)

Fix a deadlock between the pool and pmap codes:
- cpu0 grabs the kernel lock (e.g. from a non-MPSAFE interrupt) and
  calls pool_get().
- cpu1 does a pool_get() on the same pool from MPSAFE code, which needs a
  pool_page_alloc(), which ends up in pmap_extract_coherency().

So cpu0 holds the kernel_lock and wants the pool lock. cpu1 holds the pool
lock and wants the kernel_lock in pmap_extract_coherency().

The pmap code should not rely on kernel_lock. Intead make the
pmap_kernel()->pm_obj_lock a IPL_VM lock and use it as pmap lock
(thus dropping the pmap test pmap_{acquire,release}_pmap_lock()).
This needs to be a IPL_VM because unlike user pmaps, this can be locked
from interrupt context.

Add a IPL_NONE lock for pmap_growkernel(). We can't use
pmap_kernel()->pm_obj_lock here because pmap_grow_map() may sleep.

Make pmap_lock (which may be locked with pm_obj_lock held) a IPL_VM
lock in all case.

reorder a few things to not call pool_get()/pool_put() (which may sleep)
with pm_obj_lock held.

Patch initially posted to port-arm@ on April 19, improved patch (per
suggestions from Nick Hudson and Jason Thorpe) on April 21.

(bouyer)

Don't try to aquire/release the pmap lock when in ddb.
Avoids a deadlock when entering ddb, or on "mach cpu n" ddb command
(the pmap lock may already be held by another CPU, which is halted when
entering ddb).
Posted to port-arm@ on April 19.

(bouyer)

Tickets #1242 - #1244

(martin)

Pull up following revision(s) via patch (requested by mrg in ticket #1244):

etc/mtree/NetBSD.dist.base: revision 1.190
etc/mtree/NetBSD.dist.base: revision 1.191
etc/mtree/NetBSD.dist.base: revision 1.192
etc/mtree/NetBSD.dist.base: revision 1.193
etc/mtree/NetBSD.dist.iscsi: revision 1.1
etc/mtree/NetBSD.dist.base: revision 1.186
etc/mtree/NetBSD.dist.base: revision 1.187
etc/mtree/NetBSD.dist.base: revision 1.188
etc/mtree/NetBSD.dist.base: revision 1.189
etc/mtree/Makefile: revision 1.38

- add new entries of gcc=7 and libgomp, libasan, liblsan, libstdc++, libubsan
- add missing header files fpr gcc=7
- add some missing gcc=5 obsolete entries
- create ./usr/include/dev/bluetooth in mtree not make, and move it into the
  base set where all other directories are.

create include/dev/dm here instead of elsewhere, where it may fail
add missing ./usr/include/dev/hdaudio (base).
(perhaps we should stop 'make includes' from creating the target directory
for normal builds as this leads to mtree inconsistencies.)

add missing ./usr/include/dev/hdmicec
add missing ./usr/include/dev/iscsi.
add two more missing subdirs:
  ./usr/include/dev/pud
  ./usr/include/dev/putter

add final missing directory from /usr/include:
  ./usr/include/netmpls

move ./usr/include/dev/iscsi into NetBSD.dist.iscsi that is only
included if MKISCSI != no.

(martin)

Pull up following revision(s) via patch (requested by mrg in ticket #1243):

external/mit/xorg/bin/xinit/Makefile: revision 1.7
external/mit/xorg/lib/libXaw/Makefile: revision 1.12
external/mit/xorg/lib/xkeyboard-config/Makefile: revision 1.13
external/mit/xorg/server/xorg-server.old/hw/xfree86/doc/Makefile: revision 1.2
share/mk/bsd.x11.mk: revision 1.130
external/mit/xorg/bin/sessreg/Makefile: revision 1.7
external/mit/xorg/bin/proxymngr/Makefile: revision 1.4

fix a bunch of man page transforms.

XXX X11MANCPP=yes has only one use left, maybe it can go away!

(martin)

Pull up following revision(s) via patch (requested by mrg in ticket #1243):

external/mit/xorg/bin/xinit/Makefile: revision 1.7
external/mit/xorg/lib/libXaw/Makefile: revision 1.12
external/mit/xorg/lib/xkeyboard-config/Makefile: revision 1.13
external/mit/xorg/server/xorg-server.old/hw/xfree86/doc/Makefile: revision 1.2
share/mk/bsd.x11.mk: revision 1.130
external/mit/xorg/bin/sessreg/Makefile: revision 1.7
external/mit/xorg/bin/proxymngr/Makefile: revision 1.4

fix a bunch of man page transforms.

XXX X11MANCPP=yes has only one use left, maybe it can go away!

(martin)

Pull up following revision(s) via patch (requested by mrg in ticket #1242):

sys/external/bsd/common/include/linux/err.h: revision 1.3
sys/external/bsd/drm2/include/drm/drm_wait_netbsd.h: revision 1.15
sys/external/bsd/common/include/linux/errno.h: revision 1.4
sys/external/bsd/drm2/dist/drm/nouveau/nouveau_drm.c: revision 1.17
sys/external/bsd/drm2/linux/linux_ww_mutex.c: revision 1.6
sys/external/bsd/drm2/dist/drm/nouveau/nouveau_fence.c: revision 1.14

fix various problems i've seen where cv_*wait*() return ERESTART,
which is -3 in netbsd, which we have mapped linux ERESTARTSYS to.

this has a problem because linux code often returns errors and
pointers in the same value, and pointer values between -4095 and
- -1 are considered as error returns, but -3 ends up as 3 and thus
is not considered an error, and mayhem ensues.

with this in place my kabylake system seems actually stable, i
have not triggered any of my prior issues in almost 4 weeks now.

Taylor asked me to write up a description and then wrote most of
the text below for me :-)

In Linux code, we always work with ERESTARTSYS so the code meaning
start over is a positive NetBSD errno safe for PTR_ERR/ERR_PTR.

To achieve this:

1. adapt all cv_waits that return to Linux so they map ERESTART to
  ERESTARTSYS, and

2. adapt all returns to userland so they convert ERESTARTSYS to
  ERESTART.

Leave EINTR and all other error codes alone.

(martin)

KNF. No functional change.

(msaitoh)

KNF. No functional change.

(msaitoh)

Fix compile error.

(msaitoh)

Add awge(4) man page

(maya)

remove unused mii to fix the build

(maya)

Express more confidence in our ability to boot on hardware with ACPI
enabled and don't warn our users that it might not and already suggest
workarounds.

The ability to disable ACPI and SMP is still there, by dropping to
the boot prompt.

(maya)

Adapt am7930 families to audio2.
- Remove {input/output}_conv stuff from am7930_glue.
  Filter pipelines for user encoding are not necessary in audio2
  so the driver only needs to handle its hardware encoding.
- audioamd/vsaudio use an ordinal linear-mulaw conversion filter
  and bba requires special one.  dev/audio/mulaw.c supports this
  variant just for bba.  It might not a good way for bba but it
  keeps all other drivers simple.
- Tested on vsaudio(4) by naru@, bba(4) by tsutsui@ (a few months ago).
  Thanks!

(isaki)

Express more confidence in our ability to boot on hardware with ACPI
enabled and don't warn our users that it might not and already suggest
workarounds.

The ability to disable ACPI and SMP is still there, by dropping to
the boot prompt.

(maya)

Adapt to audio2.
- New audio filter can access softc so it's simple.
- Clear rval of fixphase on halt.

(isaki)

ZOI

(sevan)

Add missing error check.

(msaitoh)

Suggested by Christos:
- Dont' cast malloc()'s return value.
- Use calloc() instead of malloc().

(msaitoh)

regen

(jmcneill)

Add Amlogix GXL internal PHY

(jmcneill)

Add a few more sd drive nodes, with SD2SCSI and similar devices, we can
fit more than four SCSI disks into a mac.

(martin)

This driver uses MII(4) and have hook SIOC[GS]IFMEDIA which just pass to
ifmedia_ioctl(), the hook is not required because ether_ioctl has it
(if_ethersubr.c rev. 1.160). This driver might require some additional fixes
for SIOCSIFMTU and other ioctl()s.

(msaitoh)

Ticket #1241

(martin)

Pull up following revision(s) (requested by knakahara in ticket #1241):

sys/net/if_gif.c: revision 1.146

fix a potential bug of gif(4) check for tunnel duplicate.

This problem has not actualized thanks to check for duplicate
in encap_attach().

(martin)

Pull up following revision(s) (requested by mrg in ticket #1239):

sys/external/bsd/drm2/ttm/ttm_bo_vm.c: revision 1.12

ttm_bo_uvm_fault() is a netbsd API function and thus should return
netbsd errnos.  fix the one "return -ERESTART".

(martin)

This driver does ether_ioctl() on SIOC{ADD,DEL}MULTI, SIOC{G,S}IFMEDIA and
default case in the switch statement. Integrate them to one ether_ioctl().
This driver don't return ENETRESET in ifmedia_ioctl(), so no functional change.

(msaitoh)

This driver does ether_ioctl() on SIOC{ADD,DEL}MULTI, SIOC{G,S}IFMEDIA and
default case in the switch statement. Only the default case didn't check the
return value with ENETRESET. Integrate them to one ether_ioctl() call with
ENETRESET test. This driver might require some additional fixes for SIOCSIFMTU
and other ioctl()s.

(msaitoh)

These drivers do ether_ioctl() on SIOC{ADD,DEL}MULTI, SIOC{G,S}IFMEDIA and
default case in the switch statement. Only the default case didn't check the
return value with ENETRESET. Integrate them to one ether_ioctl() with
ENETRESET test. This change might improve some other ioctl()s which return
ENETRESET by calling if_init().

(msaitoh)

These drivers do ether_ioctl() on SIOC{ADD,DEL}MULTI, SIOC{G,S}IFMEDIA and
default case in the switch statement. Only the default case didn't check the
return value with ENETRESET. Integrate them to one ether_ioctl() with
ENETRESET test. This change might improve SIOCSIFMTU or some other ioctl()s
which return ENETRESET by calling if_init().

(msaitoh)

Tickets #1236 - #1240

(martin)

Pull up following revision(s) (requested by mrg in ticket #1240):

sys/dev/usb/umodem_common.c: revision 1.27
sys/dev/usb/umodemvar.h: revision 1.10
sys/dev/usb/ucom.c: revision 1.122

fix umodem(4) detach:

- ucom(4) needs kpreempt disabled around softint_schedule()
- switch a copied printf() to aprint_error_dev()
- use static normally in umodem_common.c
- remove unused sc_openings in softc, convert sc_dying to real bool
- add sc_refcnt, sc_lock and sc_detach_cv to softc.  usage is:
  - sc_dying is protected by sc_lock
  - sc_detach_cv is matched with sc_lock for cv operations
  - sc_refcnt is increased in open and decreased in close, any time
    it is decreased, it is checked for less than zero, and a broadcast
    performed on sc_detach_cv.  detach waits for sc_refcnt.
- umodem_param() and umodem_set() check for sc_dying

this fixes pullout out an open ucom@umodem.

@skrll.

XXX: pullup

(martin)

Pull up following revision(s) (requested by chs in ticket #1238):

sys/uvm/uvm_pdaemon.c: revision 1.110

Draining pools from the pagedaemon thread can deadlock, because draining
a pool can involve taking a lock which can be held by a thread which is
blocked waiting for memory.  Avoid this by moving the pool-draining work
to a separate worker thread.

(martin)

fix a potential bug of gif(4) check for tunnel duplicate.

This problem has not actualized thanks to check for duplicate
in encap_attach().

(knakahara)

Pull up following revision(s) (requested by chs in ticket #1237):

sys/external/bsd/drm2/ttm/ttm_bus_dma.c: revision 1.8

in ttm_bus_dma_populate(), move the clearing of TTM_PAGE_FLAG_SWAPPED to be
earlier to avoid wrongly asserting if bus_dmamap_load_pglist() fails.

(martin)

Pull up following revision(s) (requested by chs in ticket #1236):

sys/uvm/uvm_fault.c: revision 1.205

If a pager fault method returns ENOMEM but some memory appears to be reclaimable,
wake up the pagedaemon and retry the fault.  This fixes the problems with Xorg
being killed with an "out of swap" message due to a transient memory shortage.

(martin)

On drivers which use MII(4) and have hook SIOC[GS]IFMEDIA which just pass to
ifmedia_ioctl(), the hook is not required because ether_ioctl has it
(if_ethersubr.c rev. 1.160). These drivers don't return ENETRESET in
ifmedia_ioctl(), so no functional change.

(msaitoh)

In drivers which use MII(4) and have hook SIOC[GS]IFMEDIA which just pass to
ifmedia_ioctl(), the hook is not required because ether_ioctl has it
(if_ethersubr.c rev. 1.160). These drivers don't return ENETRESET in
ifmedia_init(), so no functional change.

(msaitoh)

Remove extra SIOCSIFMEDIA check.

(msaitoh)

Don't include mii*.h because this driver doesn't use MII(4).

(msaitoh)

Bump date for previous.

(kre)

PR standards/40554

Update the description of the <& and >& redirection operators
(as indicated would happen in a message appended to the PR a week ago,
which received no opposition - no feedback).

Some rewriting of the section on redirects (including how the word
expansion of the "file" works) to make this simpler & more accurate.

(kre)

Modify driver to use bits(3) to extract capabilities.  Thanks to
Anon Ymous for the fix.

(blymn)

Note efiboot.plist / device tree overlay changes.

(thorpej)

- Add support for a boot configuration file, defaulting to /etc/efiboot.plist.
- Add support for pre-loading EFI environment variables from efiboot.plist.
- Add support for device tree overlays specified in efiboot.plist.

(Man page for efiboot forthcoming.)

(thorpej)

use cpu_name() on a struct cpu_info *.  don't assume it has a ci_dev
member that is valid input to device_xname().

fixes build on arm64.

(mrg)

The third argument of mincore(2) is unsigned only on linux.

(christos)

Fix SOC_MESON8B build

(jmcneill)

in ttm_bus_dma_populate(), move the clearing of TTM_PAGE_FLAG_SWAPPED to be
earlier to avoid wrongly asserting if bus_dmamap_load_pglist() fails.

(chs)

Draining pools from the pagedaemon thread can deadlock, because draining
a pool can involve taking a lock which can be held by a thread which is
blocked waiting for memory.  Avoid this by moving the pool-draining work
to a separate worker thread.

(chs)

If a pager fault method returns ENOMEM but some memory appears to be reclaimable,
wake up the pagedaemon and retry the fault.  This fixes the problems with Xorg
being killed with an "out of swap" message due to a transient memory shortage.

(chs)

Add mesonrng

(jmcneill)

Add support for Meson GXBB and GXL

(jmcneill)

Adapt to audio2.

(isaki)

Add driver for Meson GX EE-domain watchdog timer.

(jmcneill)

Don't rely on the existing state of the GX watchdog control register when
resetting.

(jmcneill)

Fix module clock rate in DDR52 mode and support mmc-ddr-3_3v DT property

(jmcneill)

Adapt to audio2.
- The driver does not support recording so drop CAPTURE property.

(isaki)

Adapt to audio2.
- The driver doesn't seem to support recording.

(isaki)

Provide a separate ap_reset for Meson GX family SoCs

(jmcneill)

wether -> whether

(maya)

Replace build option to enable hex encoded bignum parameters with a sysctl.

(mlelstv)

Add support for Meson GXL internal PHY.

(jmcneill)

Adapt to audio2.

(isaki)

Adapt to audio2.

(isaki)

Adapt to audio2.

(isaki)

Adapt to audio2.
- Remove its own mulaw converter.
- XXX I'm not sure about 3,4ch mode vs 8/14bit mode.

(isaki)

Adapt to audio2.

(isaki)

Adapt to audio2.

(isaki)

Make it compilable (on this branch).
flags here is F(READ|WRITE), not AUOPEN_(READ|WRITE).

(isaki)

Adapt to audio2.
- Drop INDEPENDENT property from es1370 which has only one clock.

(isaki)

Use FOREACH_SAFE when removing devices from our internal list.

(martin)

Adapt cs428x families to audio2.
- recording on cs4280 seems to have its own conversion.
  I will see it later but first make it compilable.

(isaki)

Oops, add accidentally dropped chunks.

(isaki)

Adapt to audio2.
- Use mulaw as default format.  HW supports slinear_be:16 actually
  but it's hard to use due to several hardware restrictions.
- Improve data transfer and interrupt.
Tested by tsutsui@ (a few months ago).  Thank you.

(isaki)

update

(maxv)

Note removal of COMPAT_OSF1.

(maxv)

Rename the PTE bits.

(maxv)

Not yet fully adapted to audio2 but make it compilable.
- Drop FULLDUPLEX property.  It is done in a tricky way.

(isaki)

Adapt to audio2.
- Drop FULLDUPLEX property.  The driver doesn't seem to be written
  as full duplex.

(isaki)