When using dm-crypt for full disk encryption, dm-crypt would allocate
an out bio and allocate the same amount of pages as in bio for
encryption.  It currently allocates one page at a time in a loop.  This
is not efficient.  So using mempool page bulk allocator instead of
allocating one page at a time.

The mempool page bulk allocator would improve the IOPS with 1M I/O
by approxiamately 6%.  The test is done on a machine with 80 vCPU and
128GB memory with an encrypted ram device (the impact from storage
hardware could be minimized so that we could benchmark the dm-crypt
layer more accurately).

Before the patch:
Jobs: 1 (f=1): [w(1)][100.0%][w=1301MiB/s][w=1301 IOPS][eta 00m:00s]
crypt: (groupid=0, jobs=1): err= 0: pid=48512: Wed Feb  1 18:11:30 2023
  write: IOPS=1300, BW=1301MiB/s (1364MB/s)(76.2GiB/60001msec); 0 zone resets
    slat (usec): min=724, max=867, avg=765.71, stdev=19.27
    clat (usec): min=4, max=196297, avg=195688.86, stdev=6450.50
     lat (usec): min=801, max=197064, avg=196454.90, stdev=6450.35
    clat percentiles (msec):
     |  1.00th=[  197],  5.00th=[  197], 10.00th=[  197], 20.00th=[  197],
     | 30.00th=[  197], 40.00th=[  197], 50.00th=[  197], 60.00th=[  197],
     | 70.00th=[  197], 80.00th=[  197], 90.00th=[  197], 95.00th=[  197],
     | 99.00th=[  197], 99.50th=[  197], 99.90th=[  197], 99.95th=[  197],
     | 99.99th=[  197]
   bw (  MiB/s): min=  800, max= 1308, per=99.69%, avg=1296.94, stdev=46.02, samples=119
   iops        : min=  800, max= 1308, avg=1296.94, stdev=46.02, samples=119
  lat (usec)   : 10=0.01%, 1000=0.01%
  lat (msec)   : 2=0.01%, 4=0.01%, 10=0.01%, 20=0.02%, 50=0.05%
  lat (msec)   : 100=0.08%, 250=99.83%
  cpu          : usr=3.88%, sys=96.02%, ctx=69, majf=1, minf=9
  IO depths    : 1=0.1%, 2=0.1%, 4=0.1%, 8=0.1%, 16=0.1%, 32=0.1%, >=64=99.9%
     submit    : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
     complete  : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.1%
     issued rwts: total=0,78060,0,0 short=0,0,0,0 dropped=0,0,0,0
     latency   : target=0, window=0, percentile=100.00%, depth=256

Run status group 0 (all jobs):
  WRITE: bw=1301MiB/s (1364MB/s), 1301MiB/s-1301MiB/s (1364MB/s-1364MB/s), io=76.2GiB (81.9GB), run=60001-60001msec

After the patch:
Jobs: 1 (f=1): [w(1)][100.0%][w=1401MiB/s][w=1401 IOPS][eta 00m:00s]
crypt: (groupid=0, jobs=1): err= 0: pid=2171: Wed Feb  1 21:08:16 2023
  write: IOPS=1401, BW=1402MiB/s (1470MB/s)(82.1GiB/60001msec); 0 zone resets
    slat (usec): min=685, max=815, avg=710.77, stdev=13.24
    clat (usec): min=4, max=182206, avg=181658.31, stdev=5810.58
     lat (usec): min=709, max=182913, avg=182369.36, stdev=5810.67
    clat percentiles (msec):
     |  1.00th=[  182],  5.00th=[  182], 10.00th=[  182], 20.00th=[  182],
     | 30.00th=[  182], 40.00th=[  182], 50.00th=[  182], 60.00th=[  182],
     | 70.00th=[  182], 80.00th=[  182], 90.00th=[  182], 95.00th=[  182],
     | 99.00th=[  182], 99.50th=[  182], 99.90th=[  182], 99.95th=[  182],
     | 99.99th=[  182]
   bw (  MiB/s): min=  900, max= 1408, per=99.71%, avg=1397.60, stdev=46.04, samples=119
   iops        : min=  900, max= 1408, avg=1397.60, stdev=46.04, samples=119
  lat (usec)   : 10=0.01%, 750=0.01%
  lat (msec)   : 2=0.01%, 4=0.01%, 10=0.01%, 20=0.02%, 50=0.05%
  lat (msec)   : 100=0.08%, 250=99.83%
  cpu          : usr=3.66%, sys=96.23%, ctx=76, majf=1, minf=9
  IO depths    : 1=0.1%, 2=0.1%, 4=0.1%, 8=0.1%, 16=0.1%, 32=0.1%, >=64=99.9%
     submit    : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
     complete  : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.1%
     issued rwts: total=0,84098,0,0 short=0,0,0,0 dropped=0,0,0,0
     latency   : target=0, window=0, percentile=100.00%, depth=256

Run status group 0 (all jobs):
  WRITE: bw=1402MiB/s (1470MB/s), 1402MiB/s-1402MiB/s (1470MB/s-1470MB/s), io=82.1GiB (88.2GB), run=60001-60001msec

The function tracing also shows the time consumed by page allocations is
reduced significantly.  The test allocated 1M (256 pages) bio in the same
environment.

Before the patch:
It took approximately 600us by excluding the bio_add_page() calls.
2720.630754 |   56)  xfs_io-38859  |   2.571 us    |    mempool_alloc();
2720.630757 |   56)  xfs_io-38859  |   0.937 us    |    bio_add_page();
 2720.630758 |   56)  xfs_io-38859  |   1.772 us    |    mempool_alloc();
 2720.630760 |   56)  xfs_io-38859  |   0.852 us    |    bio_add_page();
….
2720.631559 |   56)  xfs_io-38859  |   2.058 us    |    mempool_alloc();
 2720.631561 |   56)  xfs_io-38859  |   0.717 us    |    bio_add_page();
 2720.631562 |   56)  xfs_io-38859  |   2.014 us    |    mempool_alloc();
 2720.631564 |   56)  xfs_io-38859  |   0.620 us    |    bio_add_page();

After the patch:
It took approxiamately 30us.
11564.266385 |   22) xfs_io-136183  | + 30.551 us   |    __alloc_pages_bulk();

Page allocations overhead is around 6% (600us/9853us) in dm-crypt layer shown by
function trace.  The data also matches the IOPS data shown by fio.

And the benchmark with 4K size I/O doesn't show measurable regression.

Signed-off-by: Yang Shi <shy828301@gmail.com>

With moving crypt_free_buffer_pages() before crypt_alloc_buffer(), we
don't need an extra declaration anymore.

Signed-off-by: Yang Shi <shy828301@gmail.com>

Since v5.13 the page bulk allocator was introduced to allocate order-0
pages in bulk.  There are a few mempool allocator callers which does
order-0 page allocation in a loop, for example, dm-crypt, f2fs compress,
etc.  A mempool page bulk allocator seems useful.  So introduce the
mempool page bulk allocator.

It introduces the below APIs:
  - mempool_init_pages_bulk()
  - mempool_create_pages_bulk()
They initialize the mempool for page bulk allocator.  The pool is filled
by alloc_page() in a loop.

  - mempool_alloc_pages_bulk_array()
  - mempool_alloc_pages_bulk_cb()
They do bulk allocation from mempool.
They do the below conceptually:
  1. Call bulk page allocator
  2. If the allocation is fulfilled then return otherwise try to
     allocate the remaining pages from the mempool
  3. If it is fulfilled then return otherwise retry from #1 with sleepable
     gfp
  4. If it is still failed, sleep for a while to wait for the mempool is
     refilled, then retry from #1
The populated pages will stay on the array until the callers consume them
or free them, or are consumed by the callback immediately.

Since mempool allocator is guaranteed to success in the sleepable context,
so the two APIs return true for success or false for fail.  It is the
caller's responsibility to handle failure case (partial allocation), just
like the page bulk allocator.

The mempool typically is an object agnostic allocator, but bulk allocation
is only supported by pages, so the mempool bulk allocator is for page
allocation only as well.

Signed-off-by: Yang Shi <shy828301@gmail.com>

Extract the common initialization code to __mempool_init() and
__mempool_create().  And extract the common alloc code into an internal
function.  This will make the following patch easier and avoid duplicate
code.

Signed-off-by: Yang Shi <shy828301@gmail.com>

Currently the bulk allocator support to pass pages via list or array,
but neither is suitable for some usecases, for example, dm-crypt, which
doesn't need a list, but array may be too big to fit on stack.  So
adding a new bulk allocator API, which passes in a callback function
that deal with the allocated pages.

The API defined in this patch will be used by the following patches.

Signed-off-by: Yang Shi <shy828301@gmail.com>

…it/rafael/linux-pm

Pull power management fix from Rafael Wysocki:
 "Add a missing NULL pointer check to the cpufreq drver for Qualcomm
  platforms (Manivannan Sadhasivam)"

* tag 'pm-6.2-rc9' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
  cpufreq: qcom-hw: Add missing null pointer check

Pull kvm fixes from Paolo Bonzini:
 "Certain AMD processors are vulnerable to a cross-thread return address
  predictions bug. When running in SMT mode and one of the sibling
  threads transitions out of C0 state, the other thread gets access to
  twice as many entries in the RSB, but unfortunately the predictions of
  the now-halted logical processor are not purged. Therefore, the
  executing processor could speculatively execute from locations that
  the now-halted processor had trained the RSB on.

  The Spectre v2 mitigations cover the Linux kernel, as it fills the RSB
  when context switching to the idle thread. However, KVM allows a VMM
  to prevent exiting guest mode when transitioning out of C0 using the
  KVM_CAP_X86_DISABLE_EXITS capability can be used by a VMM to change
  this behavior. To mitigate the cross-thread return address predictions
  bug, a VMM must not be allowed to override the default behavior to
  intercept C0 transitions.

  These patches introduce a KVM module parameter that, if set, will
  prevent the user from disabling the HLT, MWAIT and CSTATE exits"

* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
  Documentation/hw-vuln: Add documentation for Cross-Thread Return Predictions
  KVM: x86: Mitigate the cross-thread return address predictions bug
  x86/speculation: Identify processors vulnerable to SMT RSB predictions

of_device_get_match_data() may return NULL, so add a check to prevent
potential null pointer dereference.

Issue reported by Qualcomm's internal static analysis tool.

Fixes: 4f79617 ("cpufreq: qcom-hw: Move soc_data to struct qcom_cpufreq")
Signed-off-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>

…rg/pub/scm/linux/kernel/git/akpm/mm

Pull misc fixes from Andrew Morton:
 "Twelve hotfixes, mostly against mm/.

  Five of these fixes are cc:stable"

* tag 'mm-hotfixes-stable-2023-02-13-13-50' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm:
  of: reserved_mem: Have kmemleak ignore dynamically allocated reserved mem
  scripts/gdb: fix 'lx-current' for x86
  lib: parser: optimize match_NUMBER apis to use local array
  mm: shrinkers: fix deadlock in shrinker debugfs
  mm: hwpoison: support recovery from ksm_might_need_to_copy()
  kasan: fix Oops due to missing calls to kasan_arch_is_ready()
  revert "squashfs: harden sanity check in squashfs_read_xattr_id_table"
  fsdax: dax_unshare_iter() should return a valid length
  mm/gup: add folio to list when folio_isolate_lru() succeed
  aio: fix mremap after fork null-deref
  mailmap: add entry for Alexander Mikhalitsyn
  mm: extend max struct page size for kmsan

…cm/linux/kernel/git/pdx86/platform-drivers-x86

Pull x86 platform drivers fix from Hans de Goede:
 "Intel vsec driver Meteor Lake PCI ids addition"

* tag 'platform-drivers-x86-v6.2-5' of git://git.kernel.org/pub/scm/linux/kernel/git/pdx86/platform-drivers-x86:
  platform/x86/intel/vsec: Add support for Meteor Lake

Both Rich Felker and Yoshinori Sato haven't done any work on arch/sh
for a while. As I have been maintaining Debian's sh4 port since 2014,
I am interested to keep the architecture alive.

Signed-off-by: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Acked-by: Yoshinori Sato <ysato@users.sourceforge.jp>
Acked-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

…el/git/trace/linux-trace

Pull tracing fix from Steven Rostedt:
 "Fix showing of TASK_COMM_LEN instead of its value

  The TASK_COMM_LEN was converted from a macro into an enum so that BTF
  would have access to it. But this unfortunately caused TASK_COMM_LEN
  to display in the format fields of trace events, as they are created
  by the TRACE_EVENT() macro and such, macros convert to their values,
  where as enums do not.

  To handle this, instead of using the field itself to be display, save
  the value of the array size as another field in the trace_event_fields
  structure, and use that instead.

  Not only does this fix the issue, but also converts the other trace
  events that have this same problem (but were not breaking tooling).

  With this change, the original work around b3bc854 ("tracing:
  Have TRACE_DEFINE_ENUM affect trace event types as well") could be
  reverted (but that should be done in the merge window)"

* tag 'trace-v6.2-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
  tracing: Fix TASK_COMM_LEN in trace event format file

…nel/git/kdave/linux

Pull btrfs fixes from David Sterba:

 - one more fix for a tree-log 'write time corruption' report, update
   the last dir index directly and don't keep in the log context

 - do VFS-level inode lock around FIEMAP to prevent a deadlock with
   concurrent fsync, the extent-level lock is not sufficient

 - don't cache a single-device filesystem device to avoid cases when a
   loop device is reformatted and the entry gets stale

* tag 'for-6.2-rc7-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:
  btrfs: free device in btrfs_close_devices for a single device filesystem
  btrfs: lock the inode in shared mode before starting fiemap
  btrfs: simplify update of last_dir_index_offset when logging a directory

…git/gregkh/usb

Pull USB fixes from Greg KH:
 "Here are 2 small USB driver fixes that resolve some reported
  regressions and one new device quirk. Specifically these are:

   - new quirk for Alcor Link AK9563 smartcard reader

   - revert of u_ether gadget change in 6.2-rc1 that caused problems

   - typec pin probe fix

  All of these have been in linux-next with no reported problems"

* tag 'usb-6.2-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb:
  usb: core: add quirk for Alcor Link AK9563 smartcard reader
  usb: typec: altmodes/displayport: Fix probe pin assign check
  Revert "usb: gadget: u_ether: Do not make UDC parent of the net device"

…x/kernel/git/efi/efi

Pull EFI fix from Ard Biesheuvel:
 "A fix from Darren to widen the SMBIOS match for detecting Ampere Altra
  machines with problematic firmware. In the mean time, we are working
  on a more precise check, but this is still work in progress"

* tag 'efi-fixes-for-v6.2-4' of git://git.kernel.org/pub/scm/linux/kernel/git/efi/efi:
  arm64: efi: Force the use of SetVirtualAddressMap() on eMAG and Altra Max machines

…l/git/powerpc/linux

Pull powerpc fixes from Michael Ellerman:

 - Fix interrupt exit race with security mitigation switching.

 - Don't select ARCH_WANTS_NO_INSTR until warnings are fixed.

 - Build fix for CONFIG_NUMA=n.

Thanks to Nicholas Piggin, Randy Dunlap, and Sachin Sant.

* tag 'powerpc-6.2-5' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:
  powerpc/64s/interrupt: Fix interrupt exit race with security mitigation switch
  powerpc/kexec_file: fix implicit decl error
  powerpc: Don't select ARCH_WANTS_NO_INSTR

When we upgraded our kernel, we started seeing some page corruption like
the following consistently:

  BUG: Bad page state in process ganesha.nfsd  pfn:1304ca
  page:0000000022261c55 refcount:0 mapcount:-128 mapping:0000000000000000 index:0x0 pfn:0x1304ca
  flags: 0x17ffffc0000000()
  raw: 0017ffffc0000000 ffff8a513ffd4c98 ffffeee24b35ec08 0000000000000000
  raw: 0000000000000000 0000000000000001 00000000ffffff7f 0000000000000000
  page dumped because: nonzero mapcount
  CPU: 0 PID: 15567 Comm: ganesha.nfsd Kdump: loaded Tainted: P    B      O      5.10.158-1.nutanix.20221209.el7.x86_64 #1
  Hardware name: VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform, BIOS 6.00 04/05/2016
  Call Trace:
   dump_stack+0x74/0x96
   bad_page.cold+0x63/0x94
   check_new_page_bad+0x6d/0x80
   rmqueue+0x46e/0x970
   get_page_from_freelist+0xcb/0x3f0
   ? _cond_resched+0x19/0x40
   __alloc_pages_nodemask+0x164/0x300
   alloc_pages_current+0x87/0xf0
   skb_page_frag_refill+0x84/0x110
   ...

Sometimes, it would also show up as corruption in the free list pointer
and cause crashes.

After bisecting the issue, we found the issue started from commit
e320d30 ("mm/page_alloc.c: fix freeing non-compound pages"):

	if (put_page_testzero(page))
		free_the_page(page, order);
	else if (!PageHead(page))
		while (order-- > 0)
			free_the_page(page + (1 << order), order);

So the problem is the check PageHead is racy because at this point we
already dropped our reference to the page.  So even if we came in with
compound page, the page can already be freed and PageHead can return
false and we will end up freeing all the tail pages causing double free.

Fixes: e320d30 ("mm/page_alloc.c: fix freeing non-compound pages")
Link: https://lore.kernel.org/lkml/BYAPR02MB448855960A9656EEA81141FC94D99@BYAPR02MB4488.namprd02.prod.outlook.com/
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: stable@vger.kernel.org
Signed-off-by: Chunwei Chen <david.chen@nutanix.com>
Reviewed-by: Vlastimil Babka <vbabka@suse.cz>
Reviewed-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

After commit 3087c61 ("tools/testing/selftests/bpf: replace open-coded 16 with TASK_COMM_LEN"),
the content of the format file under
/sys/kernel/tracing/events/task/task_newtask was changed from
  field:char comm[16];    offset:12;    size:16;    signed:0;
to
  field:char comm[TASK_COMM_LEN];    offset:12;    size:16;    signed:0;

John reported that this change breaks older versions of perfetto.
Then Mathieu pointed out that this behavioral change was caused by the
use of __stringify(_len), which happens to work on macros, but not on enum
labels. And he also gave the suggestion on how to fix it:
  :One possible solution to make this more robust would be to extend
  :struct trace_event_fields with one more field that indicates the length
  :of an array as an actual integer, without storing it in its stringified
  :form in the type, and do the formatting in f_show where it belongs.

The result as follows after this change,
$ cat /sys/kernel/tracing/events/task/task_newtask/format
        field:char comm[16];    offset:12;      size:16;        signed:0;

Link: https://lore.kernel.org/lkml/Y+QaZtz55LIirsUO@google.com/
Link: https://lore.kernel.org/linux-trace-kernel/20230210155921.4610-1-laoar.shao@gmail.com/
Link: https://lore.kernel.org/linux-trace-kernel/20230212151303.12353-1-laoar.shao@gmail.com

Cc: stable@vger.kernel.org
Cc: Alexei Starovoitov <alexei.starovoitov@gmail.com>
Cc: Kajetan Puchalski <kajetan.puchalski@arm.com>
CC: Qais Yousef <qyousef@layalina.io>
Fixes: 3087c61 ("tools/testing/selftests/bpf: replace open-coded 16 with TASK_COMM_LEN")
Reported-by: John Stultz <jstultz@google.com>
Debugged-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Suggested-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Suggested-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Yafang Shao <laoar.shao@gmail.com>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>

…rnel/git/broonie/spi

Pull spi fixes from Mark Brown:
 "A couple of hopefully final fixes for spi: one driver specific fix for
  an issue with very large transfers and a fix for an issue with the
  locking fixes in spidev merged earlier this release cycle which was
  missed"

* tag 'spi-fix-v6.2-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi:
  spi: spidev: fix a recursive locking error
  spi: dw: Fix wrong FIFO level setting for long xfers

…ux/kernel/git/tip/tip

Pull x86 fixes from Ingo Molnar:
 "Fix a kprobes bug, plus add a new Intel model number to the upstream
  <asm/intel-family.h> header for drivers to use"

* tag 'x86-urgent-2023-02-11' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/cpu: Add Lunar Lake M
  x86/kprobes: Fix 1 byte conditional jump target

…/linux/kernel/git/tip/tip

Pull locking fix from Ingo Molnar:
 "Fix an rtmutex missed-wakeup bug"

* tag 'locking-urgent-2023-02-11' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  rtmutex: Ensure that the top waiter is always woken up

…l/git/cxl/cxl

Pull cxl fixes from Dan Williams:
 "Two fixups for CXL (Compute Express Link) in presence of passthrough
  decoders.

  This primarily helps developers using the QEMU CXL emulation, but with
  the impending arrival of CXL switches these types of topologies will
  be of interest to end users.

   - Fix a crash when shutting down regions in the presence of
     passthrough decoders

   - Fix region creation to understand passthrough decoders instead of
     the narrower definition of passthrough ports"

* tag 'cxl-fixes-6.2' of git://git.kernel.org/pub/scm/linux/kernel/git/cxl/cxl:
  cxl/region: Fix passthrough-decoder detection
  cxl/region: Fix null pointer dereference for resetting decoder

…/kernel/git/nvdimm/nvdimm

Pull libnvdimm fixes from Dan Williams:
 "A fix for an issue that could causes users to inadvertantly reserve
  too much capacity when debugging the KMSAN and persistent memory
  namespace, a lockdep fix, and a kernel-doc build warning:

   - Resolve the conflict between KMSAN and NVDIMM with respect to
     reserving pmem namespace / volume capacity for larger sizeof(struct
     page)

   - Fix a lockdep warning in the the NFIT code

   - Fix a kernel-doc build warning"

* tag 'libnvdimm-fixes-6.2' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm:
  nvdimm: Support sizeof(struct page) > MAX_STRUCT_PAGE_SIZE
  ACPI: NFIT: fix a potential deadlock during NFIT teardown
  dax: super.c: fix kernel-doc bad line warning

…rnel/git/rppt/memblock

Pull memblock revert from Mike Rapoport:
 "Revert 'mm: Always release pages to the buddy allocator in
  memblock_free_late()'

  The pages being freed by memblock_free_late() have already been
  initialized, but if they are in the deferred init range,
  __free_one_page() might access nearby uninitialized pages when trying
  to coalesce buddies, which will cause a crash.

  A proper fix will be more involved so revert this change for the time
  being"

* tag 'fixes-2023-02-11' of git://git.kernel.org/pub/scm/linux/kernel/git/rppt/memblock:
  Revert "mm: Always release pages to the buddy allocator in memblock_free_late()."

…/kernel/git/clk/linux

Pull clk fixes from Stephen Boyd:
 "Two clk driver fixes

   - Use devm_kasprintf() to avoid overflows when forming clk names in
     the Microchip PolarFire driver

   - Fix the pretty broken Ingenic JZ4760 M/N/OD calculation to actually
     work and find proper divisors"

* tag 'clk-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux:
  clk: ingenic: jz4760: Update M/N/OD calculation algorithm
  clk: microchip: mpfs-ccc: Use devm_kasprintf() for allocating formatted strings

…el/git/linusw/linux-pinctrl

Pull pin control fixes from Linus Walleij:
 "Some assorted pin control fixes, the most interesting will be the
  Intel patch fixing a classic problem: laptop touchpad IRQs...

   - Some pin drive register fixes in the Mediatek driver.

   - Return proper error code in the Aspeed driver, and revert and
     ill-advised force-disablement patch that needs to be reworked.

   - Fix AMD driver debug output.

   - Fix potential NULL dereference in the Single driver.

   - Fix a group definition error in the Qualcomm SM8450 LPASS driver.

   - Restore pins used in direct IRQ mode in the Intel driver (This
     fixes some laptop touchpads!)"

* tag 'pinctrl-v6.2-3' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-pinctrl:
  pinctrl: intel: Restore the pins that used to be in Direct IRQ mode
  pinctrl: qcom: sm8450-lpass-lpi: correct swr_rx_data group
  pinctrl: aspeed: Revert "Force to disable the function's signal"
  pinctrl: single: fix potential NULL dereference
  pinctrl: amd: Fix debug output for debounce time
  pinctrl: aspeed: Fix confusing types in return value
  pinctrl: mediatek: Fix the drive register definition of some Pins

…rnel/git/pci/pci

Pull PCI fixes from Bjorn Helgaas:

 - Move to a shared PCI git tree (Bjorn Helgaas)

 - Add Krzysztof Wilczyński as another PCI maintainer (Lorenzo
   Pieralisi)

 - Revert a couple ASPM patches to fix suspend/resume regressions (Bjorn
   Helgaas)

* tag 'pci-v6.2-fixes-2' of git://git.kernel.org/pub/scm/linux/kernel/git/pci/pci:
  Revert "PCI/ASPM: Refactor L1 PM Substates Control Register programming"
  Revert "PCI/ASPM: Save L1 PM Substates Capability for suspend/resume"
  MAINTAINERS: Promote Krzysztof to PCI controller maintainer
  MAINTAINERS: Move to shared PCI tree

This reverts commit 5e85eba.

Thomas Witt reported that 5e85eba ("PCI/ASPM: Refactor L1 PM Substates
Control Register programming") broke suspend/resume on a Tuxedo
Infinitybook S 14 v5, which seems to use a Clevo L140CU Mainboard.

The main symptom is:

  iwlwifi 0000:02:00.0: Unable to change power state from D3hot to D0, device inaccessible
  nvme 0000:03:00.0: Unable to change power state from D3hot to D0, device inaccessible

and the machine is only partially usable after resume.  It can't run dmesg
and can't do a clean reboot.  This happens on every suspend/resume cycle.

Revert 5e85eba until we can figure out the root cause.

Fixes: 5e85eba ("PCI/ASPM: Refactor L1 PM Substates Control Register programming")
Link: https://bugzilla.kernel.org/show_bug.cgi?id=216877
Reported-by: Thomas Witt <kernel@witt.link>
Tested-by: Thomas Witt <kernel@witt.link>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: stable@vger.kernel.org	# v6.1+
Cc: Vidya Sagar <vidyas@nvidia.com>

This reverts commit 4ff116d.

Tasev Nikola and Mark Enriquez reported that resume from suspend was broken
in v6.1-rc1.  Tasev bisected to a47126e ("PCI/PTM: Cache PTM
Capability offset"), but we can't figure out how that could be related.

Mark saw the same symptoms and bisected to 4ff116d ("PCI/ASPM: Save L1
PM Substates Capability for suspend/resume"), which does have a connection:
it restores L1 Substates configuration while ASPM L1 may be enabled:

  pci_restore_state
    pci_restore_aspm_l1ss_state
      aspm_program_l1ss
        pci_write_config_dword(PCI_L1SS_CTL1, ctl1)         # L1SS restore
    pci_restore_pcie_state
      pcie_capability_write_word(PCI_EXP_LNKCTL, cap[i++])  # L1 restore

which is a problem because PCIe r6.0, sec 5.5.4, requires that:

  If setting either or both of the enable bits for ASPM L1 PM
  Substates, both ports must be configured as described in this
  section while ASPM L1 is disabled.

Separately, Thomas Witt reported that 5e85eba ("PCI/ASPM: Refactor L1
PM Substates Control Register programming") broke suspend/resume, and it
depends on 4ff116d.

Revert 4ff116d ("PCI/ASPM: Save L1 PM Substates Capability for
suspend/resume") to fix the resume issue and enable revert of 5e85eba
to fix the issue Thomas reported.

Note that reverting 4ff116d means L1 Substates config may be lost on
suspend/resume.  As far as we know the system will use more power but will
still *work* correctly.

Fixes: 4ff116d ("PCI/ASPM: Save L1 PM Substates Capability for suspend/resume")
Link: https://bugzilla.kernel.org/show_bug.cgi?id=216782
Link: https://bugzilla.kernel.org/show_bug.cgi?id=216877
Reported-by: Tasev Nikola <tasev.stefanoska@skynet.be>
Reported-by: Mark Enriquez <enriquezmark36@gmail.com>
Reported-by: Thomas Witt <kernel@witt.link>
Tested-by: Mark Enriquez <enriquezmark36@gmail.com>
Tested-by: Thomas Witt <kernel@witt.link>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: stable@vger.kernel.org	# v6.1+
Cc: Vidya Sagar <vidyas@nvidia.com>

…nel/git/soc/soc

Pull ARM SoC fixes from Arnd Bergmann:
 "All the changes this time are minor devicetree corrections, the
  majority being for 64-bit Rockchip SoC support. These are a couple of
  corrections for properties that are in violation of the binding, some
  that put the machine into safer operating points for the eMMC and
  thermal settings, and missing properties that prevented rk356x PCIe
  and ethernet from working correctly.

  The changes for amlogic and mediatek address incorrect properties that
  were preventing the display support on MT8195 and the MMC support on
  various Meson SoCs from working correctly.

  The stihxxx-b2120 change fixes the GPIO polarity for the DVB tuner to
  allow this to be used correctly after a futre driver change, though it
  has no effect on older kernels"

* tag 'soc-fixes-6.2-4' of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc:
  arm64: dts: meson-gx: Make mmc host controller interrupts level-sensitive
  arm64: dts: meson-g12-common: Make mmc host controller interrupts level-sensitive
  arm64: dts: meson-axg: Make mmc host controller interrupts level-sensitive
  ARM: dts: stihxxx-b2120: fix polarity of reset line of tsin0 port
  arm64: dts: mediatek: mt8195: Fix vdosys* compatible strings
  arm64: dts: rockchip: align rk3399 DMC OPP table with bindings
  arm64: dts: rockchip: set sdmmc0 speed to sd-uhs-sdr50 on rock-3a
  arm64: dts: rockchip: fix probe of analog sound card on rock-3a
  arm64: dts: rockchip: add missing #interrupt-cells to rk356x pcie2x1
  arm64: dts: rockchip: fix input enable pinconf on rk3399
  ARM: dts: rockchip: add power-domains property to dp node on rk3288
  arm64: dts: rockchip: add io domain setting to rk3566-box-demo
  arm64: dts: rockchip: remove unsupported property from sdmmc2 for rock-3a
  arm64: dts: rockchip: drop unused LED mode property from rk3328-roc-cc
  arm64: dts: rockchip: reduce thermal limits on rk3399-pinephone-pro
  arm64: dts: rockchip: use correct reset names for rk3399 crypto nodes

…inux/kernel/git/riscv/linux

Pull RISC-V fixes from Palmer Dabbelt:
 "This is a little bigger that I'd hope for this late in the cycle, but
  they're all pretty concrete fixes and the only one that's bigger than
  a few lines is pmdp_collapse_flush() (which is almost all
  boilerplate/comment). It's also all bug fixes for issues that have
  been around for a while.

  So I think it's not all that scary, just bad timing.

   - avoid partial TLB fences for huge pages, which are disallowed by
     the ISA

   - avoid missing a frame when dumping stacks

   - avoid misaligned accesses (and possibly overflows) in kprobes

   - fix a race condition in tracking page dirtiness"

* tag 'riscv-for-linus-6.2-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux:
  riscv: Fixup race condition on PG_dcache_clean in flush_icache_pte
  riscv: kprobe: Fixup misaligned load text
  riscv: stacktrace: Fix missing the first frame
  riscv: mm: Implement pmdp_collapse_flush for THP

Pull ceph fix from Ilya Dryomov:
 "A fix for a pretty embarrassing omission in the session flush handler
  from Xiubo, marked for stable"

* tag 'ceph-for-6.2-rc8' of https://github.com/ceph/ceph-client:
  ceph: flush cap releases when the session is flushed

Pull block fix from Jens Axboe:
 "A single fix for a smatch regression introduced in this merge window"

* tag 'block-6.2-2023-02-10' of git://git.kernel.dk/linux:
  nvme-auth: mark nvme_auth_wq static

…l/git/tiwai/sound

Pull sound fixes from Takashi Iwai:
 "Hopefully the last one for 6.2, a collection of the fixes that have
  been gathered since the last pull.

  All changes are small and trivial device-specific fixes"

* tag 'sound-6.2-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
  ALSA: hda/realtek: Add Positivo N14KP6-TG
  ASoC: topology: Return -ENOMEM on memory allocation failure
  ALSA: emux: Avoid potential array out-of-bound in snd_emux_xg_control()
  ASoC: fsl_sai: fix getting version from VERID
  ALSA: hda/realtek: fix mute/micmute LEDs don't work for a HP platform.
  ALSA: hda/realtek: Add quirk for ASUS UM3402 using CS35L41
  ASoC: codecs: es8326: Fix DTS properties reading
  ASoC: tas5805m: add missing page switch.
  ASoC: tas5805m: rework to avoid scheduling while atomic.
  ALSA: hda/realtek: Enable mute/micmute LEDs on HP Elitebook, 645 G9
  ASoC: SOF: amd: Fix for handling spurious interrupts from DSP
  ALSA: hda/realtek: Fix the speaker output on Samsung Galaxy Book2 Pro 360
  ALSA: pci: lx6464es: fix a debug loop
  ASoC: rt715-sdca: fix clock stop prepare timeout issue