drmkms_sched: Fix link sets in module.

(riastradh)

KNF the KNF

(skrll)

arm32: Refactor splhigh and restore in dosoftints.

No functional change intended.  splhigh always returns ci->ci_cpl,
which should not be changing at this point.  Makes the bracketing by
splhigh/splx clearer.

(riastradh)

aarch64: Refactor splhigh and restore in dosoftints.

No functional change intended.  splhigh always returns ci->ci_cpl,
which should not be changing at this point.  Makes the bracketing by
splhigh/splx clearer.

(riastradh)

Catch up with the per lwp astpending (rather than per cpu) flag in the
preemption code.  NFC as preemption (still) isn't enabled.

(skrll)

if we add a mode for the DFP port, make sure it has a name so we don't crash
later on
now this works again on my 12" powerbook

(macallan)

And, we're not ready for 10.0 yet, so remove a couple premature lines!

Thanks, Rin!

(pgoyette)

remove duplicate entry and blank line that crept in from somewhere.

(pgoyette)

Oopppss, /usr has no `e'  :)

(pgoyette)

Include debug files for new amdgpu modules

(pgoyette)

Enable amdgpu module & firmware on x86

Make it easier to test this one out, add `load amdgpu` before boot netbsd.

XXX there should probably be a drm block in sys/modules/Makefile which
includes aarch64.

(maya)

Ticket #1482

(martin)

regen for ticket #1482

(martin)

Pull up the following revisions, requested by msaitoh in ticket #1482:

sys/dev/pci/pcidevs 1.1422,1.1445-1.1460
via patch
sys/arch/x86/pci/amdsmn.c 1.13

Update pcidevs:
- Add Intel Alder Lake devices and Intel 600 Series PCH devices.
- Add some Intel Xeon Scalable / Skylake-E devices.
- Fix AMD F16_HB from 0x1568 to 0x1538.
- Add some devices for AMD and improve some descriptions to clarify.
- Add VMware AHCI and NVMe.
- Update Intel 700 series Ethernet devices.
- Add some Broadcom devices.
- Add some Broadcom / LSI RAID cards.
- Fix typos and whitespace.

(martin)

Correct stale comment. Thanks rillig@ for pointed out. NFC.

(rin)

s/functin/function/ in copy pasted comment.

(andvar)

s/fucntion/function/ in comments.

(andvar)

Pass MAKEOBJDIR to opt-m-include-dir

Prevent makeing opt-m-include-dir.tmp in src tree.

Reviewed by: rillig

(sjg)

curlwp is only available for ifdef _KERNEL.

(chs)

Fix copy-paste in comment. No binary changes.

(rin)

Fix inverted logic; cr was completely broken by FPU insn with Rc == 1.

Why haven't we noticed this so far...

(rin)

Whitespace

(skrll)

mincore(2) - use .Rv -std

(uwe)

mincore(2): Clarify units of len argument: bytes, not pages.

Minor tidying while here:

- Use .Fa, not .Ar, for function arguments.
- Reduce verbiage of opening sentence.

(riastradh)

t_sig_backtrace: Pacify vfork clobber warning with volatile.

(riastradh)

execinfo(3): Mark sig_backtrace_jump test as xfail.

PR lib/56940

(riastradh)

kern.maxvnodes _can_ be lowered, but not below the number of currently
active vnodes.

(pgoyette)

execinfo: Fix mistake in previous: match `the_loop*' as prefix.

Now that there are two functions we need to catch either one.

(riastradh)

execinfo: Test stack traces through null pointer jump too.

(riastradh)

Some more BITS(3)

(skrll)

G/C the unused bge_rx_mini_chain member from bge_chain_data. It was removed
from the FreeBSD driver in 2003.

(skrll)

Add missing entry for ticket #1476.

(msaitoh)

Fix misleading fsck_ffs messages about wapbl journal replay.

Apparently fsck doesn't have a way to replay it to its internal memory
only. (Someone(TM) should implement this...)

This means that if you use -n, it can't replay the journal. But the
sequence of prints is such that it looks like it did. This is quite
misleading. Add an additional specific warning.

(dholland)

use the pcb of the thread we are tracing rather than always curlwp.

(chs)

sys/lock.h: Need <sys/systm.h> for nullop.

But only do that we're not using SPINLOCK_BACKOFF_HOOK from
<machine/lock.h>.

(riastradh)

aarch64/lock.h: Need <sys/param.h> for _HARDKERNEL.

Add include guard while here.

XXX Why does this aarch64-specific file have #ifdef __aarch64__?

(riastradh)

make: fix out-of-bounds read when parsing an invalid line

Reported by Robert Morris in https://bugs.freebsd.org/265119.

Since 2021-12-14.

(rillig)

amdgpu: Define DC_FPU_START/END on Arm.

(riastradh)

amdgpu: Sprinkle missing COPTS.amdgpu_float options.

For some reason x86 didn't need these but aarch64 does.

(riastradh)

drm: Fix CPPFLAGS for drm scheduler code.

(riastradh)

kern_event.c: Mark KASSERT-only static function as __diagused.

Otherwise clang objects with -Wunneeded-internal-declaration.

(riastradh)

use defparam; noted by tnn@.

(mrg)

Ticket #1481

(martin)

Pull up following revision(s) (requested by nia in ticket #1481):

external/mit/ctwm/libexec/ctwm_app_menu: revision 1.5 (via patch)

ctwm_app_menu: Work around "emacs being installed results in the
generated ctwmrc containing invalid syntax" problem. Emacs installs
a desktop file called "emacsclient" that contains "doublequotes"
in its exec string. Pending a better solution for dumping escaped
shell lines into the config file, hide or sanitize desktop entires
that might mess up the internal ctwmrc quoting rules.

(martin)

Revert our local fix for GCC 10.3 introduced in rev. 1.25:
http://cvsweb.netbsd.org/bsdweb.cgi/src/external/gpl3/gcc/dist/gcc/config/rs6000/rs6000.c#rev1.25

The problem has been fixed differently by upstream:
- master: https://gcc.gnu.org/git/?p=gcc.git;a=commit;h=77eccbf39ed55297802bb66dff5f62507a7239e3
- gcc-10: https://gcc.gnu.org/git/?p=gcc.git;a=commit;h=6142d65f4f528067f31e86bb97c07a661f2d66e7

No regression observed for ATF on oea (sandpoint), booke (evbppc), and
405 (evbppc). The system boots successfully on 403 (evbppc).

(rin)

ctwm_app_menu: Various shell script optimization suggestions by kre@.

(nia)

make MAXLWP a real option that triggers rebuilds properly.

(mrg)

regen

(mrg)

bump the netbsd version GCC to yesterday's date.

(mrg)

whitespace

(skrll)

drm: Split scheduler into separate module.

Used only by amdgpu, and it uses something that is currently broken
from the gcc update on aarch64, so let's make it a little more
optional.

(riastradh)

Add some file system options present in GENERIC kernels.

(simonb)

drm: Go back to using device_unit(pciN) for pci busid.

This isn't arbitrary; it has to match existing logic in userland in
libpciaccess, and until we extend the pci(4) ioctl interface, and
find some way to handle userland compatibility, we can't change it.

(riastradh)

- Handle dtlink in acpi_pack_compat_list().
- Don't pass the _HID value as the device name; always use the ACPI
  node name.

(thorpej)

If there is a compat string list, parenthetically print the first one
in iic_print_direct().  (From thorpej-i2c-spi-conf2 branch.)

(thorpej)

Normalize how motoi2c controllers are attached.

(thorpej)

arm64 needs additional lse.S builds with GCC 10.4.0.

(mrg)

revert the previous; atomicity.cc is also built into here.

(mrg)

atomicity.cc has moved into libsupc++ for everyone.

this was a workaround we applied for sparc in PR#54660, and
now is used for all platforms.

(mrg)

GCC 10.4.0 is now in -current.

(mrg)

regen everything for GCC 10.4.0.

(mrg)

merge GCC 10.4.0.

(mrg)

Split the i2c controller into 2 separate instances.  This aligns with
how the device is represented when using DeviceTree, and also allows us
to de-weirdify the shared motoi2c code.

XXX Because e500 interrupt code doesn't support shared interrupts, we
no longer establish one for the i2c controller, but this isn't a great
loss as motoi2c doesn't support using interrupts right now anyway.

(thorpej)

Stop using deprecated problib functions.

(thorpej)

Revert previous; that was intended to be two separate commits.

(thorpej)

Stop using deprecated proplib functions.

(thorpej)

Stop using deprecated proplib functions.

(thorpej)

Stop using deprecated proplib functions.

(thorpej)

Drop using deprecated proplib functions.

(thorpej)

ctwm_app_menu: Work around "emacs being installed results in the
generated ctwmrc containing invalid syntax" problem. Emacs installs
a desktop file called "emacsclient" that contains "doublequotes"
in its exec string. Pending a better solution for dumping escaped
shell lines into the config file, hide or sanitize desktop entires
that might mess up the internal ctwmrc quoting rules.

(nia)

Add a menu option to set the file extension of the downloaded sets.
amd64 sets are .tar.xz and sysinst fails by default.

(christos)

Build and install pkgconfig files for liblzma and libarchive

(wiz)

fix current libarchive version, and mention 3.6.1 release

(wiz)

fix missing components of c++17.

codecvt.cc was already present and handled for c++98 and c++11,
but cow-string-inst.cc, ostream-inst.cc, and string-inst.cc
were not handled for c++11 and c++17.

move all the handling of this into the Makefiles out of the
mknative-gcc output, and handle the new c++17 files.

fixes a problem reported by dholland.

(mrg)

don't rename codecvt.cc here, do it in a Makefile when consumed.

since the Makefiles already need updates for these, it's easier
if all of the changes are there.

(mrg)

malloc -> kmem.

(skrll)

Add more statistics countes.

- Add many statics counters that the chip has.
- Attach event counters only if available.

(msaitoh)

build glamoregl module with all the glamor objects, not as part of
the base list of libraries.  noted by rjs.

should fix PR#56934.

(mrg)

Add some driver debugging infrastructure

(martin)

s/pedning/pending/

(andvar)

s/Transration/Translation/

(andvar)

Removed unused opt_wapbl.h include.

(simonb)

fix typos in comments and log messages, mainly s/intrrupt/interrupt/.

(andvar)

readlink: sort SEE ALSO

(wiz)

Add an Xr to realpath(1) (which does much the same as readlink -f).
Try not to forget the damn comma!

(kre)

Note realpath(1) update.

(kre)

Add ATF tests for realpath(1)

Note that realpath can act differently for root than for other users
(where an ordinary user will see EACCESS root just barrels right through).

The tests adapt themselves, when run as root, less error cases can be
tested than when run as some other user.

(kre)

Add missing comma.

(wiz)

Make realpath(1) compat with the planned (not yet approved) specification
for POSIX.8 (its next version).  (Should the requirements change, the code
here can be updated).

This adds two new options.  -e and -E
-e just tells realpath to do what it has done since it was imported here.
-E makes realpath more compatible with the coreutils version, and allows
  the final component of the path to not exist (the final component after
  all symlinks have been expanded, not of the arg on the command line - though
  that one not existing is one case of the more general spec.).

POSIX is not going to specify which of those is the default - instead is
planning to require users to always explicitly specify one.

The default (now) here is -E.  This makes us more compat with coreutils.
realpath was added in the first place because it is (apparently) used in
real world scripts - the more we can support, the better.

Note that in all cases where realpath -e succeeds, realpath -E will succeed
as well.  This means that any uses of "realpath file" that have been
working in HEAD will still work.  Some cases that would have failed
will work (by default) now.

(kre)

PR bin/56917

If getfsspecname() fails that will usually mean that a NAME=wedge or
ROOT.x partition is unabailable.  raidframe specified unavailable
partitions as "absent" so in this case, pass "absent" rather than the
unaltered NAME= or ROOT.x string, which the kernel has no clue what
do do with, and doesn't configure the raid at all.

(kre)

drm/ttm: Cast to potentially larger types before << PAGE_SHIFT.

Not sure if this is a problem but it might affect, e.g., i386pae.
Can't hurt, in any case, to cast to what the type will be after the
shift anyway.

(riastradh)

Cleanup the "hup" function - quote properly, and test for the pid
file existing (with something in it) before simply using it, so the
error message can be controlled, rather than just sh's cannot open file msg.

(kre)

Get rid of the (insulting)
We recommend that you create a non-root account...
message from single user mode shell startups.

(kre)

Make this better ...  Allow config file for raidN to be found
in /etc/conf/raid/raidN (as well as in /etc/raidN.conf) (less
clutter in /etc).

Allow config files to exist for raids that are autoconfigured (just
skip anything autoconfigured) - this allows a (simplistic) recovery
mechanism if autoconf fails, and makes it simpler to enable/disable
autoconf at will, while keeping the raid working.

While here, add sh quoting where it is needed, and remove it from
where it is pointless.

(kre)

Rename functions in previous in preparation for future changes.

(thorpej)

Improve error reporting.

(thorpej)

When we init the hardware's rx/tx ring configuration we need to
adjust our internal state, as this implicitly resets the current
descriptor pointer.
Previously "ifconfig eqos0 down; ifconfig eqos0 up" made the interface
non-functional.

(martin)

Fix a couple of nits with savectx():
- Note that this function is only used by dumpsys().
- Don't safe the PS word; there isn't actually a spot for it in the PCB.
- Don't bother returning anything; savectx() is declared void.

(thorpej)

Localize direct refernces to kernel symbols into functions.  NFC.

(thorpej)

Make alpha_wmb() actually a WMB rather than an MB (all uses of
alpha_wmb() have been audited and fixed-up as necessary).

(thorpej)

In the Alpha tc_syncbus(), we need to issue a MB because WMB does not
order reads with respect to writes.  Remove now-redundant tc_wmb()
calls before tc_syncbus().

NFC on MIPS other than removing a redundant wbflush() (tc_wmb() followed
by tc_syncbus()).

(thorpej)

No need to test for __alpha__ || alpha ... just one will suffice.

(thorpej)

mips: Fix cpuids synchronization at boot.

(riastradh)

drm: Use CPPFLAGS.drmkms in all local drm drivers too.

This way we don't pollute the NetBSD kernel namespace with all the
Linux compat shim definitions needed to build drm, except for the
local drm drivers that need the API.

(riastradh)

drm: Keep drm include flags in CPPFLAGS.drmkms after all.

Failures from this are much more obvious to debug than files with
missing #ifdefs, which sometimes lead to build failures but could
also just lead to silent miscompilation.

(riastradh)

Modify comment. Use __BIT() a little. No functional change.

(msaitoh)

use PRIu64 and hopefully fix some clang builds.

(mrg)

drm: Restore pollution of CPPFLAGS with include directories.

Too much trouble to make this fine-grained.

(riastradh)

aarch64: Make cpufunc.h includable without sys/cpu.h first.

(riastradh)

drm: Ifdef out alpha-specific logic in radeon.

Guessing this is handled transparently by bus_space(9) when we ask
for prefetchable linear mappings.  Need an alpha wizard to review!

(riastradh)

drm: Use real pci segment number, not pciN device unit.

(riastradh)

drm/ttm: Handle io_mem_pfn, used by amdgpu.

(riastradh)

amdgpu: Write amdgpu_ttm_io_mem_pfn in terms of bus_space_mmap.

(riastradh)

drm: Undefine `alpha' in the module build too.

(riastradh)

drm: Teach bus_dma_hacks.h about alpha.

Not sure if this is correct!  Someone who knows alpha should vet.

(riastradh)

drm: Undefine `alpha' in CPPFLAGS.  Causes lotsa trouble!

But don't undefine it outside drmkms; `#ifdef alpha' or equivalent is
used elsewhere in-tree.  (Maybe it should be replaced by __alpha__.)

(riastradh)

drm: Make drm_clflush_* noop on alpha.

As far as I know there is no need for this; the CPU has weak ordering
on loads and stores requiring ordering barriers, but I/O devices
participate in the CPU's cache protocol for access to main memory so
there's never any need for explicit cache flushes.

(riastradh)

drm: Narrow scope of drm CPPLAGS just to drm files.

No need for various Linux compat definitions elsewhere in the kernel.

(riastradh)

drm: Propagate CPPFLAGS.drmkms into all the drm drivers.

(riastradh)

drm: Note dependency on sysmon_power.

drm uses this for pswitch events to notify of hotplug.

(riastradh)

drm: Nix drm_agp_borrow.

This horrible kludge dates from before I understood the relation of
genfb_pci and pci drm drivers in the old and new worlds of drm.

The only user of it, in i915, was changed to use agp_i810_borrow
directly in the last drm update, so this hack can die.

(riastradh)

alpha: Fix missing includes in pmap.h.

(riastradh)

linux: Add alpha definitions of mb/wmb/rmb too for fun.

Who wants to plug an amdgpu into their alpha?

(riastradh)

linux: Clarify error message in asm/barrier.h.  Not just for drm.

(riastradh)

linux: Add mips definitions for mb, wmb, rmb.

(riastradh)

linux: Use dsb(sy/st/ld) from arm/cpufunc.h for mb/wmb/rmb.

Works on more Arm CPU types this way.  From skrll.

(riastradh)

wm(4): Use CSR_WRITE_FLUSH, not bus_space_barrier.

The bus space is mapped non-prefetchable and non-cacheable, so there
is no need for bus_space_barrier and it might not even have any
effect -- bus_space_read/write are totally ordered for non-
prefetchable, non-cacheable mappings anyway.  Presumably the intent
here was to wait for acknowledgment of the write from the device --
which bus_space_barrier does not guarantee, and which I expect
requires a read transaction as CSR_WRITE_FLUSH performs.

ok msaitoh@, knakahara@

(riastradh)

wm(4): Use bus_dmamap_sync, not bus_space_barrier, for tx desc.

The newly written tx descriptor needs to be visible to the device
before the TDT register is set.  There's no bus space mappings that
need any barriers, though -- nothing mapped prefetchable or cacheable
here.  So bus_dmamap_sync (via wm_cdtxsync) is necessary, not
bus_space_barrier.  By accident, bus_space_barrier implies the same
barrier instructions as bus_dmamap_sync on some architectures, but it
wouldn't work at all if we were bouncing.

ok msaitoh@, knakahara@

(riastradh)

Fix path in man page to system.ctwmrc.

(nia)

KNF

(skrll)

Don't need sys/malloc.h so remove the #include

(skrll)

Turn a locking botch (shouldn't drop lock on error) into a KASSERT

(skrll)

Fix a problem whereby detaching a device that has open kevent
registrations can result in a UAF: When a device detaches, it
calls seldestroy(), which calls knote_fini(), and when that
returns, the softc that contained the selinfo and klist are freed.
However, any knotes that were registered still linger on with the
kq descriptor they're were associated with, and when the file
descriptors close, those knotes will be f_detach'd, which will
call into the driver instance that no longer exists.

Address this problem by adding a "foplock" mutex to the knote.
This foplock must be held when calling into filter_attach(),
filter_detach(), and filter_event() (XXX not filter_touch();
see code for details).  Now, in klist_fini(), for each knote
that is on the klist that's about to be blown away, acquire
the foplock, replace the knote's filterops with a do-nothing
stub, and release the foplock.

The end result is that:
==> The foplock ensures that calls into filter_*() will get EITHER
    the real backing object's filterops OR the nop stubs.
==> Holing the foplock across the filter_*() calls ensures that
    klist_fini() will not complete until there are no callers inside
    the filterops that are about to be blown away.

(thorpej)

Make some knote implementation details private to kern_event.c.  NFC, and
no ABI change for kevent providers.

(thorpej)

drm: Simplify async framebuffer child task attachment logic.

No need for scheduling the task to fail.  Nix needless error
branches.

(riastradh)

drm: Fix asynchronous attach logic for drm framebuffer devices.

The driver-specific sc_scheduled boolean flag was racy.  It is
necessary to use config_pending_incr/decr to hold up console
selection and userland start (and block detach, not that that works
yet) until the drmfb attach has completed.

(riastradh)

Fix bug revealed by SIGINFO support; Do not treat short read(2)/write(2)
as error (*). This occurs typically when signal is received.

(*) For older version, we already deal with short read(2) from remote
host in sink(). But for other cases, i.e., write(2) to local file in
sink(), read(2)/write(2) in source(), error was raised.

This version of rcp(1) can successfully send/receive files with older
version, even if short read(2)/write(2) occurs by SIGINFO.

Also, when real error occurs, give up immediately instead of continue to
send/receive wrong data.

Clean up the mess a little bit as well...

(rin)

Make it compile w/o warnings for !VCONS_DRAW_INTR

(martin)

Regen for:

Make kqueue event status for vnodes shareable, and for stacked file systems
like nullfs, make the upper vnode share that status with the lower vnode.

And, lo, NetBSD 9.99.99.

Fixes PR kern/56713.

(thorpej)

Make kqueue event status for vnodes shareable, and for stacked file systems
like nullfs, make the upper vnode share that status with the lower vnode.

And, lo, NetBSD 9.99.99.

Fixes PR kern/56713.

(thorpej)

drm: Fix missing barrier definitions on arm.

The definition will work on armv7; don't think we have any drmkms
drivers on armv<7.

(riastradh)

wsdisplay(4): Make most of struct vcons_data private.

More importantly, make sizeof(struct vcons_data) independent of
whether VCONS_DRAW_INTR &c. is defined.

Allocate the private space with kmem rather than in the caller.

This still doesn't have very good separation between interface
parameters and internal state, but it's better than before, and is
necessary to make genfb usable in modules.

In arm_simplefb.c, this removes use of the use_intr member.  That
assignment became redundant with the introduction of vcons_earlyinit,
so there's no need to replace it by anything.

(riastradh)

drm: Comment *mb() vs smp_*mb() and omit default for *mb().

In general membar_*() is _not_ enough for *mb(), because membar_*()
is only for CPU/CPU synchronization on normal memory, and *mb()
requires machine-dependent I/O logic.  Leave it as an `#error' case
if the architecture isn't listed here.  But membar_*() is OK for
smp_*().

Fill in powerpc and sparc while here, as examples of what the
now-removed defaults failed to do.  If you want to use drm on a mips
or alpha system yer gonna hafta fill this in yourself.

(riastradh)

linux/ww_mutex: Make wwm_debug member unconditional.

This way LOCKDEBUG doesn't change any structure sizes.  Shuffle
members around to make better use of padding, on both LP32 and LP64.

(riastradh)

vmwgfx: Depend on drmkms_ttm module.

(riastradh)

radeon: Depend on drmkms_ttm module.

(riastradh)

nouveau: Depend on drmkms_ttm module.

(riastradh)

amdgpu: Depend on drmkms_ttm module.

(riastradh)

drm: Modularize ttm.

(riastradh)

drm: Include GPU scheduler in drmkms.kmod.

Could be split into a separate module, but whatever.

(riastradh)

drm: Sync module/drmkms/Makefile with files.drmkms.

Unconditionally include legacy APIs for viadrmums.

(riastradh)

drm: Dust off module build a little.

(riastradh)

drm: Ifdef out devm_* stuff rather than stub it out.

Stub is not helpful, and this pollutes the symbol namespace.

(riastradh)