cgd(4): Fix criterion for detach when wedgies are held.

The somewhat confusingly named DK_BUSY(dksc, pmask) answers the
following question:

        Suppose I hold either the character or the block device (but
        not both) of all of the partitions in pmask.  Is anyone else
        using the disk, rendering it unsafe to detach?

This is useful for ioctls like CGDIOCCLR and VNDIOCCLR, which must be
issued on open file descriptors for the disk, so the question cannot
simply be answered by testing whether dk_openmask != 0.

Instead, DK_BUSY breaks the question into the following criteria:

        1. Are there any _other_ partitions than those in pmask open
          at all?  If so, it must be someone else, since I only hold
          partitions in pmask -- hence the disk is busy.

        2. Are any of the partitions in pmask open _both_ as a block
          device _and_ as a character device?  If so, it must be
          someone else, since I only hold _either_ the character
          _or_ the block device open but not both -- hence the disk
          is busy.

When config_detach_all runs at shutdown time, it tries to detach
cgd(4), which has DVF_DETACH_SHUTDOWN; this is important so we submit
queued writes to the underlying disk and wait for them to complete
with dk_drain.

If cgd(4) has any dk wedges with file systems mounted still
configured on it, it isn't ready to detach yet.  But asking
DK_BUSY(dksc, 1 << RAW_PART) returns false, because the dk wedges
only hold RAW_PART open as a block device -- so if nobody has
RAW_PART open as a character device, or any other partitions open,
cgd_detach blithely goes on its way to forcibly detach the wedges.

Instead, ask DK_BUSY(dksc, 0), because the caller -- cgd_detach
issued by config_detach_all -- does not, in fact, hold any partitions
open, so it doesn't need to work around them like ioctl(CGDIOCCLR)
does.  Fixes hang in zfs on dk on cgd during shutdown (and probably
also zfs on cgd without any intervening dk but I haven't tested).

(This change might have the side effect that `drvctl -d cgdN' doesn't
work, but I don't care.)

XXX pullup-9
XXX pullup-8 (...-7, -6, -5...)

(riastradh)

make: remove usage message if the debug file cannot be opened

Since a non-writable file is not a syntax error, there is no point in
showing the usage in this situation.  Showing the usage may have been a
copy-and-paste mistake from a few lines below, when this option was
added back in main.c 1.133 from 2006-10-15.

(rillig)

tests/make: cover a debug log file that cannot be opened

(rillig)

make: narrow down the return type of ReadMakefile

No functional change.

(rillig)

make: rename local variables to be simpler

No binary change.

(rillig)

make: clean up 'explode'

Breaking out of the first 'for' loop was unnecessarily complicated.  The
call to strlen was not necessary since f already pointed at the end of
the string.

No functional change.

(rillig)

tests/make: test MAKEFLAGS in POSIX mode

This test covers the function 'explode' in main.c, the case in which the
flags are actually expanded.

(rillig)

make: replace __func__ with actual strings

Make is supposed to be C90-compatible, and __func__ is from C99.

No functional change.

(rillig)

make: clean up comments

(rillig)

make: remove unnecessary words from command line options

Several years ago, the command line options were individual global
variables.  The global variable could therefore not be named 'silent'
since that would have conflicted with local variables of the same name.
After moving the global variable to the namespace 'struct CmdOpts',
there is no conflict anymore.

There doesn't seem to be any risk of naming collisions for the names
'touch' and 'query'.

No functional change.

(rillig)

make: rename eunlink to unlink_file

The name eunlink suggested a relation with the similarly named functions
emalloc or esnprintf, but that was misleading.  Instead, unlink_file
works like unlink, except that it refuses to remove an empty directory.

No functional change.

(rillig)

linux: Fix copypasto in comment.

(riastradh)

i915: Fix timeout return value of i915_request_wait.

For some reason this returns ETIME, not ETIMEDOUT or 0 like most
other wait-with-timeout things in Linux.

(riastradh)

i915: Factor a little out of #ifdef NetBSD.

Doesn't really reduce diff but slightly reduces danger of bad merge.

(riastradh)

i915: Omit diff now that TASK_INTERRUPTIBLE is defined.

(riastradh)

drm: Define TASK_INTERRUPTIBLE so we can reduce diffs a little.

(riastradh)

Revert rev 1.12 of devopen.c.  This had the unintented side effect
of breaking opens on non-root filesystems (eg trying to open/read
"esp:/EFI/NetBSD/boot.cfg" on the EFI system partition).

Ok manu@.  Original problem to be re-addressed.

(simonb)

i915: Disable message citing firmware URL.

Not very helpful for users; helpful only for developers maintaining
the gpufw set, who can find it by searching the source of the
message.  (Also it was missing a newline which is very confusing!)

(riastradh)

linux: Define mb/wmb/rmb as mfence/sfence/lfence on i386 too.

Not the same as membar_*; drm uses mb to order special instructions
like clflush which require mfence for ordering.

(riastradh)

sys: Skip suspendsched on cpu_reboot if we're in ddb.

If we're in ddb, the scheduler and all other CPUs are quiesced
anyway.  But suspendsched will try to take an adaptive lock, which
causes it to crash and re-enter ddb, which isn't very useful for
rebooting.

(riastradh)

i915: Unifdef cnp_enable/disable_backlight.

Not sure why this was ifdef'd out in the first place!  Appears to
have been a mistake in merge.

(riastradh)

drm: Fix locking around accurate vblank counts.

- Make drm_crtc_accurate_vblank_count require the caller to hold the
  event lock, rather than take it internally.

- Fix locking around drm_crtc_accurate_vblank_count and related
  operations in amdgpu and nouveau interrupt handlers.

- Use drm_crtc_vblank_put_locked, not drm_crtc_vblank_put, when we
  already hold the event lock.

PR kern/56557

(riastradh)

Add clang's optnone attribute.

(christos)

namei(9): s/operatoins/operations/

(andvar)

drm: Allow ww_mutex_lock after ww_acquire_done if we already hold it.

In other words, if ww_mutex_lock would return -EALREADY, that's OK and
does not warrant an assertion.

PR kern/56557

(riastradh)

fix various typos, mainly in comments.

(andvar)

mixerctl: add colon after 'Usage', add space after comma

(rillig)

acpi: ohci: Match "generic-ohci" compatible string

(jmcneill)

acpi: Support DT link device ID

The PRP0001 device ID is used to tell the OS to match a driver against
the "compatible" _DSD property. Make acpi_compatible_match aware of this
special device ID and match device compatible entries against the
compatible property when applicable.

(jmcneill)

port-arm/50563: pool allocator corruption due to __MUTEX_PRIVATE

This file has moved on somewhat since the PR was raised, but nevertheless
let's add the CTASSERT "just in case".

(skrll)

Once again things have grown, so the install image allocation size
needs to be bumped.

XXX There's gotta be a better way to maintain this!  :-)

(pgoyette)

Add support for a read-only edid attribute that shows the result of
WSDISPLAYIO_GET_EDID.

Uses code from sys/dev/videomode to parse and print the edid data.

(mlelstv)

lint: don't try to trick mathematics by defining ZERO to a nonzero value

No binary change.

(rillig)

Make re-usuable to userland. The vesagtf standalone build is #if'd out for now.

(mlelstv)

ukbd(4): Wait for callouts and blinking to complete in detach.

This detach routine (and whole driver) has got a lotta other
problems, but let's fix 'em one at a time...

(riastradh)

postfix: enable lint again

Since init.c 1.229 from 2021-12-22, lint correctly parses and interprets
initializers with omitted braces, which are allowed by the C standards
but not recommended.

The warnings about empty declarations at file scope are suppressed
precisely.

Lint2 handles the generated .ln files fine in my local builds, I suspect
some truncated files as the cause for the unspecific "is broken" from
the comment.

(rillig)

Use a white on black terminal regardless of whether we're COLOR or not

(nia)

Run the .profile script in the default xsession to ensure that
the PATH is set properly when starting X from xdm.

(nia)

ehci(4): fix PR kern/56574: uaudio at ehci (not xhci) only outputs
rapid gunfire (patch from nick)

(nia)

malo(4): Fix line breaks in attach output.

(riastradh)

put back the tcpwinsz initialization code.

(christos)

PR/56569: Mike Small: Remove unused code.

(christos)

wsfb: Prefer wide fonts when EDID data is available.

To give us a better chance of picking a readable font, prefer fonts that
will render at least 3mm wide instead of picking the font that will be
closest to that size.

(jmcneill)

drm: Fix missing newline in DRM_WARN (redux).

(riastradh)

Revert "drm: Fix missing newline in DRM_WARN."

Accidentally included more than I intended here.

(riastradh)

drm: Fix missing newline in DRM_WARN.

(riastradh)

drm: Sprinkle some assertions into sg dma logic.

(riastradh)

i915: Use AcpiOsMapMemory, not bus_space_map, for opregion.

Needed because this appears in firmware-type memory mappings, which
are excluded from bus_space_map.

XXX pullup-9 (via manual patch since the code has changed a bit)

(riastradh)

i915: Restore uao_set_pgfl for i915 gem objects, lost in the merge.

Needed to ensure we allocate paddrs that the GPU can cope with.

(riastradh)

Fix copy + pasto, last note was for #1396

(martin)

Ticket #1721

(martin)

Pull up the following (all via patch), requested by msaitoh in ticket #1721:

usr.sbin/cpuctl/arch/i386.c 1.118-1.119, 1.121-1.122
usr.sbin/cpuctl/arch/cpuctl_i386.h 1.6
sys/arch/x86/x86/identcpu_subr.c 1.8-1.9
sys/arch/x86/x86/identcpu.c 1.123
sys/arch/x86/include/cacheinfo.h 1.30
sys/arch/x86/include/cpu.h 1.132

- Fix a bug that some TLB related lines were not printed.
- Fix a bug that STLB is printed as DTLB.
- If a TLB is variable sized, print the max size instead of error message.
- Cosmetic changes to improve readability.

(martin)

Ticket #1391

(martin)

Pull up the following (all via patch), requested by msaitoh in ticket #1396:

usr.sbin/cpuctl/arch/i386.c 1.118-1.119, 1.121-1.122
usr.sbin/cpuctl/arch/cpuctl_i386.h 1.6
sys/arch/x86/x86/identcpu_subr.c 1.8-1.9
sys/arch/x86/x86/identcpu.c 1.123
sys/arch/x86/include/cacheinfo.h 1.30
sys/arch/x86/include/cpu.h 1.132

- Fix a bug that some TLB related lines were not printed.
- Fix a bug that STLB is printed as DTLB.
- If a TLB is variable sized, print the max size instead of error message.
- Cosmetic changes to improve readability.

(martin)

amdgpu: Sprinkle __diagused.

(riastradh)

add some comments

(macallan)

make ALU instructions use the same format as load/store instructions
same binary generated

(macallan)

Enable mailbox API 1.5 support. Tested on ESXi with ixgben 1.10.3.0.

(msaitoh)

Add code to support API version 1.5. No functional change.

  - This change adds almost all code to support API 1.5 except real negotiation
    and upgrade mailbox functions.
  - From ix-3.3.18, ix-3.3.22 and ixv-1.5.24.

(msaitoh)

Whitespace fix. Remove old comment. No functional change.

  - From FreeBSD ixv-1.5.22.

(msaitoh)

Restore some mailbox related functions. No functional change.

  - Revert some part of ixgbe_mbx.c rev. 1.7 and ixgbe_mbx.h rev. 1.11.
    It's also part of FreeBSD 3f66b96d8658f8721e6b1c6f035611bec44675b9.
  - This change reduces diff against FreeBSD ix-3.3.14.

(msaitoh)

Update copyright to 2020.

  FreeBSD: 8455e365f77f5b66ac9521dbcd690f79345ce147

(msaitoh)

Change the first argument of ixgbe_send_vf_msg(). No functional change.

- This is a part of FreeBSD ix-3.2.17.

(msaitoh)

Rename IXGBE_VT_MSGTYPE_{ACK,NACK} to IXGBE_VT_MSGTYPE_{SUCCESS,FAILURE}.

  - Sync with FreeBSD ix-3.3.18.
  - No functional change.

(msaitoh)

Move some definitions. No functional change. Part of FreeBSD ix-3.3.18.

(msaitoh)

Fix missing curlwp_bind()

Fixes kern/56556

(yamaguchi)

add macros to simplify issuing SX instructions, hide the address alignment /
displacement shenanigans SX needs, and make the code more resemble the
inline assembler source it arguably is

(macallan)

acpi: ehci: Don't assume that a transaction translator is available.

Look for companion controllers using the rules in ACPI 6.3 section 9.14.1
"USB 2.0 Host Controllers and _UPC and _PLD", and only assume TT is
present if we can't find a USB 1.1 companion.

(jmcneill)

Add ohci at acpi bindings. No known PNP IDs for this, so match using _CLS
and the PCI class/subclass/interface codes. Maybe this will take off.

(jmcneill)

i915: Fix memory leak in active node allocation under #ifdef NetBSD.

Potential fix for PR kern/56573.

(riastradh)

pool(9): Fix default PR_NOALIGN for large pool caches.

Was broken in recent change to separate some pool cache flags from
pool flags.

Fixes crash in zfs.

(riastradh)

remove extra C in OBJCOPTS (from Anon Ymous)

(christos)

Disable amdgpu as it uses floating point arithmetics and subr_kcov
doesnt handle it.

Ok: riastradh@

(hannken)

Initialize crtc_state to suppress false warning it may be used uninitialized.

Ok: riastradh@

(hannken)

Cast addr to uint64_t before printing.
Ok: riastradh@

(hannken)

Keep constants 32 bit, why does __BIT() return uintmax_t?

(hannken)

Put back the code to add DEBUG=-g in config and to add the debug kernels
in the kernel sets. Having the .gdb kernels in the EXTRA_KERNELS does not
work, since we only walk through the standard kernels in the Makefile. Also
doing DEBUG=-g in config twice does not hurt.

(christos)

tests/make: explain the current behavior of the option '--version'

(rillig)

Use the correct dmatag for all control structures. Also, don't KASSERT that
we're using the low 4GB for them and simply error out.

kern/56570: ehci panic on boot

(skrll)

hyper-v: move idt vector allocating to vmbus_init_interrupts_md()
for refactoring

And, the deallocating is also moved to
vmbus_deinit_interrupts_md().

reviewed by nonaka@n.o.

(yamaguchi)

Move the variable into the section that uses it

(yamaguchi)

delete the extra space

(yamaguchi)

x86: improve error handling related to idt_vec_alloc()

(yamaguchi)

Three fixes

- pass the 64bit DMA tag if the HCCPARAMS says ehci supports it and the
  64bit DMA tag is available/valid.  This should help with the
  "cannot create xfer" error on 64bit systems.

- restrict the control structure memory allocation to the low 4GB
  (CTRLDSSEGMENT was always set to zero anyway)

- if ehci_init fails then tidyup before returning the error.

(skrll)

Whitespace

(skrll)

Reduce code duplication: kmem_cache_create() is now exactly the same as
kmem_cache_create_dtor() except for the dtor argument, so implement
the former in terms of the latter.

(thorpej)

Document PR_PSERIALIZE.  Remove documentation for pool_cache_setpredestruct().

(thorpej)

Do the last change differently:

Instead of having a pre-destruct hook, put knowledge of passive
serialization into the pool allocator directly, enabled by PR_PSERIALIZE
when the pool / pool_cache is initialized.  This will guarantee that
a passive serialization barrier will be performed before the object's
destructor is called, or before the page containing the object is freed
back to the system (in the case of no destructor).  Note that the internal
allocator overhead is different when PR_PSERIALIZE is used (it implies
PR_NOTOUCH, because the objects must remain in a valid state).

In the DRM Linux API shim, this allows us to remove the custom page
allocator for SLAB_TYPESAFE_BY_RCU.

(thorpej)

lint: remove redundant EOF tests in lexer

No functional change.

(rillig)

lint: fix wrong error in strict bool mode in condition with comma

For the result of the comma operator, it doesn't matter whether the
comma itself comes from a system header or not.  Instead, it's the main
operator of the right operand.

Since 2021-11-16.

(rillig)

lint: clean up lex.c

Rename 'struct kwtab' to 'struct keyword' since a single keyword is not
a whole keyword table.

Sync comment for lex_name with reality: sbuf_t no longer contains the
hash value.

Remove redundant tests for EOF, as EOF is neither a space nor a digit
nor an xdigit.

No functional change.

(rillig)

lint: remove spaces around bit-field colon

As seen in /usr/share/misc/style.

No binary change.

(rillig)

lint: rename C89 to C90 in variable names

No binary change.

(rillig)

lint: clean up initialization

No functional change.

(rillig)

tests/lint: improve test for initialization of array of unknown size

Previously, the test didn't show that the size of the resulting object
was updated too early.

(rillig)

lint: use C90 instead of C89 when referring to the C standard

(rillig)

tests/lint: fix space-tab indentation

(rillig)

In zfs_setattr() don't recheck the auth policy for a "nodump" flags
change. zfs_netbsd_setattr() has already checked if this request is
authorised, and our secpolicy_xvattr() doesn't check kauth chflags.

XXX: Fix this propery when we migrate to openzfs.

riastradh@: Seems reasonable.

(simonb)

drm: Omit local diff -- vmem is entirely a NetBSDism.

Use struct vmem rather than vmem_t to obviate need for header file
dependency.

(riastradh)

Fixup NQTXC_CMD_{TCP,UDP} after previous

(skrll)

In a close run thing use #define<tab> everywhere for consistency

(skrll)

Use __BIT() some more. Same code before and after.

(skrll)

add register definitions. the design is BCM2835 I2C look-a-like.

(nisimura)

lint: fix handling of initializations

The implementation from March 2021 added proper support for designators
but didn't model the brace levels correctly.  In particular, it could
not handle additional braces or omitted braces.  In such a case, lint
skipped the remaining initializers from the initialization.  Due to
this, type errors in the remaining initializers went unnoticed.  Another
effect was that arrays of unknown size were wrongly reported as having
size 0.

Both GCC and Clang recommend placing braces around each sub-type that is
initialized, such as a struct, union or array.  Postfix does not follow
these recommendations, therefore lint had to be disabled in
external/ibm-public/postfix/Makefile.inc.  This commit fixes the bugs
mentioned there.

(rillig)

Handle the SIOCGNBRINFO ioctl for compat32.
arp -a works with compat32 now.

Credit to simonb@ for the ndp fix from which this is cribbed.

(roy)

retire Ampro EnCorePP1 board configuration

(nisimura)

tests/lint: extend test for nested initializations

(rillig)

scsi(4): Take kernel lock around entry into autoconf.

This code paths is entered by kthreads marked MP-safe, not just from
autoconf.

I'm not sure this is sufficient -- it's not clear to me whether
anything prevents concurrently scanning the same target.  Someone with
a better understanding of scsi(4) locking will have to audit this.

(For example, maybe it is guaranteed only to happen only either (a)
in autoconf, or (b) in a thread that doesn't start until autoconf is
done.  But I don't know -- and if it is this, it should be asserted
so we can verify it.)

(riastradh)

lint: reword message 187 about too long string literal for initializer

The previous message was imprecise in that it didn't distinguish between
non-terminating and terminating null bytes.

(rillig)

lint: treat incomplete union in the same way as incomplete struct

The newly added tests triggered the assertion in begin_designation since
for incomplete types the initialization is stopped before handling the
first brace.

(rillig)

pursue better register description

(nisimura)

s/filname/filename/

(andvar)

tests/lint: test initialization of incomplete union

The function initialization_lbrace only mentions STRUCT, which looks
suspicious.

(rillig)

lint: rename debug_indent to debug_print_indent

The previous name could be mistaken to mean "increase the indentation of
the debug output".  Instead, the function prints the current indentation.

In externs1.h, the macro definition was a duplicate, the macros were
sorted differently than the functions a few lines above.

No binary change.

(rillig)

- For kmem_cache_create_dtor(), use a pre-destructor to issue the
  synchronize_rcu() if the caller uses SLAB_TYPESAFE_BY_RCU.  A
  special pool allocator is not required in this case.
- Now that SLAB_TYPESAFE_BY_RCU does the right thing, no need to
  call synchronize_rcu() in __i915_request_dtor().

(thorpej)

Rather than calling xc_barrier() in lwp_dtor(), set a pre-destruct hook
on the lwp_cache and invoke the barrier there.

(thorpej)

Add pool_cache_setpredestruct(), which allows a pool cache to specify
a function to be called before the destructor for a batch of one or more
objects is called.  This can be used as a synchronization point by
subsystems that rely on the type-stable nature of pool cache objects or
subsystems that use other forms of passive serialization.

(thorpej)

lint: make function names a bit more concise

No binary change.

(rillig)

tests/lint: remove false assumptions from comments

https://stackoverflow.com/q/65868752

(rillig)

lint: sync comment with recent change of parameter name

(rillig)

Fix build of xlint/line1 with OBJDIR set.

(roy)

lint: invert condition in build_name

No functional change.

(rillig)

lint: rename parameters in check_init_expr

An initialization has a left-hand and a right-hand side, reflect this in
the parameter names to avoid confusion.

No binary change.

(rillig)

Default files to BSD group ownership in line with ffs, after the lead
of FreeBSD <https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=139076>
(patch by hannken@)

(hauke)

files.radeon: add clang -Wno-format for radeon_uvd.c

int vs. short mismatch in UVD firmware version printf.
Also change some makeoptions conditional copypasta i915drmkms -> radeon.

(tnn)

drm: sprinkle some -Wno-unused-function for clang

915_sw_fence.c: for debug_fence_init_onstack
drm_mm.c: for rb_hole_size_to_node
intel_hdmi.c: for intel_hdmi_hdcp2_protocol
intel_guc_submission.c: for rq_prio
nouveau_nvkm_subdev_instmem_base.c: for fake_ioread32_native
nouveau_nvkm_subdev_mmu_vmmnv04.c: for nv04_vmm_pgt_pte

(tnn)

drm_device.h: don't redefine vmem_t

(tnn)

correct register definition error and improve naming

(nisimura)

i915_scheduler.c: node_to_request: duplicate 'const' declaration specifier

(tnn)

i915_gem_shmem.c: shmem_get_pages: fix uninitialize use of "noreclaim"

(tnn)

i915_drm_resume_early: initialize ret

clang -Wsometimes-uninitialized says uninitialized use
on line 2043 when if statement on line 2037 is false

(tnn)

add interrupt logic stuff

(nisimura)

Remove unneeded struct acpi_pci_intr forward declaration.

(skrll)

Remove an empty line

(skrll)

Wrap a long line

(skrll)

Change the usb_mem API to take a bus_dma_tag_t in usb_allocmem instead of
a struct usbd_bus *.

This allows an HCD to use more than one tag.

(skrll)

Pass up bus_dma error from usb_block_allocmem rather than always returning
USBD_NOMEM on error.

(skrll)

Style.

(skrll)

KASSERT(BUS_ADDR_HI32(baddr) == 0) in, and apply some consistent code
style to all of ehci_alloc_{sqh,sqtd,itd,sitd}.

While ehci can address memory >4GB for control structures
EHCI_CTRLDSSEGMENT is always set to zero so ehci assumes it's using the
low 4GB.

(skrll)

Update uvm_pglistalloc_[cs]_ps to return EINVAL if [low, high] doesn't
match any memory.

Useful for bus_dmamem_alloc where a tag might not cover any memory.
This will be used in an update to ehci.

"looks good" from chuq@

(skrll)

Fix a bug where pic_establish_intr would fail to call pic_establish_irq
if a free pic__iplsources slot was found, i.e. an interrupt handler at
the same ipl had been disestablished previously.

(skrll)

G/C pic_iplsource

(skrll)

KNF

(skrll)

improve consistency when attach error cases.

(nisimura)

Fix net.*.rps_hash=toeplitz-othercpus on one CPU systems.

(knakahara)

tpm(4): Fix disabling of rnd source if tpm is deactivated.

Nothing prevents a second worker from being queued when the first one
is about to do rnd_detach_source.  Instead, just set a flag so future
requests don't bother running a new thread; if there's a concurrent
one that's already been scheduled on another CPU, well, too bad, we
get a couple extra log messages but that's fine.

A better way to do this would probably be to detect whether the tpm
is deactivated at attach time, but that requires reading more of the
tpm spec than I care to do when there are alternative ways to
procrastinate like scrubbing the toilet.

(riastradh)

Slight code re-structure and wrap a long line.  Interestingly this gives
the same binary before and after.

(skrll)

drm: add missing KERNEL_LOCK around calls to config_found().

(chs)

Rename:
    MKKDEBUG -> MKDEBUGKERNEL
    MKTOOLSDEBUG -> MKDEBUGTOOLS
while keeping compatibility with the old names. Add missing documentation.
Now all debugging tunables are prefixed with MKDEBUG.

(christos)

Revert previous, not needed MKKDEBUG does it.

(christos)

ieee1394: Need kernel lock around config_found.

(riastradh)

i915: Obviate need for __diagused on variables in GEM_BUG_ON.

(riastradh)

tests/lint: test excess braces around initializers

(rillig)

lint: add grammar rule for the beginning of a designation

This will be necessary to properly implement handling of initializers
and braced initializer-lists.

No functional change for now since the designation is already reset
after each expression and '}'.  To handle initializations properly, the
designation must not be reset after each expression, it must advance to
the next member instead.

(rillig)