The unhold_lkb() function decrements the lock's kref, and
asserts that the ref count was not the final one.  Use the
kref_put release function (which should not be called) to
call the assert, rather than doing the assert based on the
kref_put return value.  Using kill_lkb() as the release
function doesn't make sense if we only want to assert.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>

This patch will disable use of deprecated timeout features if
CONFIG_DLM_DEPRECATED_API is not set.  The deprecated features
will be removed in upcoming kernel release v6.2.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>

This patch adds a CONFIG_DLM_DEPRECATED_API Kconfig option
that must be enabled to use two timeout-related features
that we intend to remove in kernel v6.2.  Warnings are
printed if either is enabled and used.  Neither has ever
been used as far as we know.

. The DLM_LSFL_TIMEWARN lockspace creation flag will be
  removed, along with the associated configfs entry for
  setting the timeout.  Setting the flag and configfs file
  would cause dlm to track how long locks were waiting
  for reply messages.  After a timeout, a kernel message
  would be logged, and a netlink message would be sent
  to userspace.  Recently, midcomms messages have been
  added that produce much better logging about actual
  problems with messages.  No use has ever been found
  for the netlink messages.

. The userspace libdlm API has allowed the DLM_LKF_TIMEOUT
  flag with a timeout value to be set in lock requests.
  The lock request would be cancelled after the timeout.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>

Remove the unused timeout parameter from dlm_user_adopt_orphan().
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>

This patch removes warning messages that could be logged when
remote requests had been waiting on a reply message for some timeout
period (which could be set through configfs, but was rarely enabled.)
The improved midcomms layer now carefully tracks all messages and
replies, and logs much more useful messages if there is an actual
problem.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>

This patch fixes some grammar output in lowcomms implementation by
removing the "successful" word which should be "successfully" but it
can never be unsuccessfully so we remove it.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>

This patch adds comments about the difference between the lower 2 bytes
of lkb flags and the 2 upper bytes of the lkb IFL flags. In short the
upper 2 bytes will be handled as internal flags whereas the lower 2
bytes are part of the DLM protocol and are used to exchange messages.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>

This patch cleans up the handling of recovery results by moving it from
ls_recover() to the caller do_ls_recovery(). This makes the error handling
clearer.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>

Make dlm_new_lockspace() wait until a full recovery completes
sucessfully or fails. Previously, dlm_new_lockspace() returned
to the caller after dlm_recover_members() finished, which is
only partially through recovery.  The result of the previous
behavior is that the new lockspace would not be usable for some
time (especially with overlapping recoveries), and some errors
in the later part of recovery could not be returned to the caller.

Kernel callers gfs2 and cluster-md have their own wait handling to
wait for recovery to complete after calling dlm_new_lockspace().
This continues to work, but will be unnecessary.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>

A lockspace can be "stopped" multiple times consecutively before
being "started" (when recoveries overlap.)  In this case, the
lsop_recover_prep callback only needs to be called once when the
lockspace is first stopped, and not repeatedly for each stop.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>

Make clear that a particular recovery iteration must not be aborted
before membership changes are applied to the members list (ls_nodes)
and midcomms layer.  Interrupting recovery before this can result
in missing node-specific changes in midcomms or through lsops.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>

This patch adds the resource name to dlm tracepoints.  The name
usually comes through the lkb_resource, but in some cases a resource
may not yet be associated with an lkb, in which case the name and
namelen parameters are used.

It should be okay to access the lkb_resource and the res_name field at
the time when the tracepoint is invoked. The resource is assigned to a
lkb and it's reference is being held during the tracepoint call. During
this time the resource cannot be freed. Also a lkb will never switch
its assigned resource. The name of a dlm_rsb is assigned at creation
time and should never be changed during runtime as well.

The TP_printk() call uses always a hexadecimal string array
representation for the resource name (which is not necessarily ascii.)
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>

This patch removes a dereference of lksb of lkb when calling ast
tracepoint. First it reduces additional overhead, even if traces
are not active. Second we can deference it in TP_fast_assign from
the existing lkb parameter.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>

This patch moves the trace calls for ast and bast to before the
ast and bast callback functions are called rather than after.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>

This patch changes the handling of a plock operation that was interrupted
while waiting for a user space reply from dlm_controld.  (This is not
the lock blocking state, i.e. locks_lock_file_wait().)

Currently, when an op is interrupted while waiting on user space, the
op is removed.  When the user space result later arrives, a kernel
message is loggged: "dev_write no op...".  This can be seen from a test
such as "stress-ng --fcntl 100" and interrupting it with ctrl-c.

Now, leave the op in place when interrupted and remove it when the
result arrives (the result will be ignored.)  With this change, the
logged message is not expected to appear, and would indicate a bug.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>

This patch refactors do_unlock_close() by using only struct dlm_plock_info
as a parameter.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>

This patch reverses the commit bcfad426 ("dlm: improve plock logging
if interrupted") by moving it to debug level and notifying the user an op
was removed.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>

This patch adds the pid information which requested the lock operation
to the debug log output.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>

This patch will use the list helper list_first_entry() instead of using
list_entry() to get the first element of a list.
Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>

Pull x86 fixes from Thomas Gleixner:

 - Make RESERVE_BRK() work again with older binutils. The recent
   'simplification' broke that.

 - Make early #VE handling increment RIP when successful.

 - Make the #VE code consistent vs. the RIP adjustments and add
   comments.

 - Handle load_unaligned_zeropad() across page boundaries correctly in
   #VE when the second page is shared.

* tag 'x86-urgent-2022-06-19' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/tdx: Handle load_unaligned_zeropad() page-cross to a shared page
  x86/tdx: Clarify RIP adjustments in #VE handler
  x86/tdx: Fix early #VE handling
  x86/mm: Fix RESERVE_BRK() for older binutils

Pull build tooling updates from Thomas Gleixner:

 - Remove obsolete CONFIG_X86_SMAP reference from objtool

 - Fix overlapping text section failures in faddr2line for real

 - Remove OBJECT_FILES_NON_STANDARD usage from x86 ftrace and replace it
   with finegrained annotations so objtool can validate that code
   correctly.

* tag 'objtool-urgent-2022-06-19' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/ftrace: Remove OBJECT_FILES_NON_STANDARD usage
  faddr2line: Fix overlapping text section failures, the sequel
  objtool: Fix obsolete reference to CONFIG_X86_SMAP

Pull scheduler fix from Thomas Gleixner:
 "A single scheduler fix plugging a race between sched_setscheduler()
  and balance_push().

  sched_setscheduler() spliced the balance callbacks accross a lock
  break which makes it possible for an interleaving schedule() to
  observe an empty list"

* tag 'sched-urgent-2022-06-19' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  sched: Fix balance_push() vs __sched_setscheduler()

Pull lockdep fix from Thomas Gleixner:
 "A RT fix for lockdep.

  lockdep invokes prandom_u32() to create cookies. This worked until
  prandom_u32() was switched to the real random generator, which takes a
  spinlock for extraction, which does not work on RT when invoked from
  atomic contexts.

  lockdep has no requirement for real random numbers and it turns out
  sched_clock() is good enough to create the cookie. That works
  everywhere and is faster"

* tag 'locking-urgent-2022-06-19' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  locking/lockdep: Use sched_clock() for random numbers

Pull irq fixes from Thomas Gleixner:
 "A set of interrupt subsystem updates:

  Core:

   - Ensure runtime power management for chained interrupts

  Drivers:

   - A collection of OF node refcount fixes

   - Unbreak MIPS uniprocessor builds

   - Fix xilinx interrupt controller Kconfig dependencies

   - Add a missing compatible string to the Uniphier driver"

* tag 'irq-urgent-2022-06-19' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  irqchip/loongson-liointc: Use architecture register to get coreid
  irqchip/uniphier-aidet: Add compatible string for NX1 SoC
  dt-bindings: interrupt-controller/uniphier-aidet: Add bindings for NX1 SoC
  irqchip/realtek-rtl: Fix refcount leak in map_interrupts
  irqchip/gic-v3: Fix refcount leak in gic_populate_ppi_partitions
  irqchip/gic-v3: Fix error handling in gic_populate_ppi_partitions
  irqchip/apple-aic: Fix refcount leak in aic_of_ic_init
  irqchip/apple-aic: Fix refcount leak in build_fiq_affinity
  irqchip/gic/realview: Fix refcount leak in realview_gic_of_init
  irqchip/xilinx: Remove microblaze+zynq dependency
  genirq: PM: Use runtime PM for chained interrupts

Merge tag 'char-misc-5.19-rc3-take2' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc

Pull char/misc driver fixes for real from Greg KH:
 "Let's tag the proper branch this time...

  Here are some small char/misc driver fixes for 5.19-rc3 that resolve
  some reported issues.

  They include:

   - mei driver fixes

   - comedi driver fix

   - rtsx build warning fix

   - fsl-mc-bus driver fix

  All of these have been in linux-next for a while with no reported
  issues"

This is what the merge in commit f0ec9c65 _should_ have merged, but
Greg fat-fingered the pull request and I got some small changes from
linux-next instead there. Credit to Nathan Chancellor for eagle-eyes.

Link: https://lore.kernel.org/all/Yqywy+Md2AfGDu8v@dev-arch.thelio-3990X/

* tag 'char-misc-5.19-rc3-take2' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc:
  bus: fsl-mc-bus: fix KASAN use-after-free in fsl_mc_bus_remove()
  mei: me: add raptor lake point S DID
  mei: hbm: drop capability response on early shutdown
  mei: me: set internal pg flag to off on hardware reset
  misc: rtsx: Fix clang -Wsometimes-uninitialized in rts5261_init_from_hw()
  comedi: vmk80xx: fix expression for tx buffer size

Pull i2c fixes from Wolfram Sang:
 "MAINTAINERS rectifications and a few minor driver fixes"

* tag 'i2c-for-5.19-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux:
  i2c: mediatek: Fix an error handling path in mtk_i2c_probe()
  i2c: designware: Use standard optional ref clock implementation
  MAINTAINERS: core DT include belongs to core
  MAINTAINERS: add include/dt-bindings/i2c to I2C SUBSYSTEM HOST DRIVERS
  i2c: npcm7xx: Add check for platform_driver_register
  MAINTAINERS: Update Synopsys DesignWare I2C to Supported

Pull xfs fixes from Darrick Wong:
 "There's not a whole lot this time around (I'm still on vacation) but
  here are some important fixes for new features merged in -rc1:

   - Fix a bug where inode flag changes would accidentally drop nrext64

   - Fix a race condition when toggling LARP mode"

* tag 'xfs-5.19-fixes-1' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:
  xfs: preserve DIFLAG2_NREXT64 when setting other inode attributes
  xfs: fix variable state usage
  xfs: fix TOCTOU race involving the new logged xattrs control knob

Pull ext4 fixes from Ted Ts'o:
 "Fix a variety of bugs, many of which were found by folks using fuzzing
  or error injection.

  Also fix up how test_dummy_encryption mount option is handled for the
  new mount API.

  Finally, fix/cleanup a number of comments and ext4 Documentation
  files"

* tag 'ext4_for_linus_stable' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4:
  ext4: fix a doubled word "need" in a comment
  ext4: add reserved GDT blocks check
  ext4: make variable "count" signed
  ext4: correct the judgment of BUG in ext4_mb_normalize_request
  ext4: fix bug_on ext4_mb_use_inode_pa
  ext4: fix up test_dummy_encryption handling for new mount API
  ext4: use kmemdup() to replace kmalloc + memcpy
  ext4: fix super block checksum incorrect after mount
  ext4: improve write performance with disabled delalloc
  ext4: fix warning when submitting superblock in ext4_commit_super()
  ext4, doc: remove unnecessary escaping
  ext4: fix incorrect comment in ext4_bio_write_page()
  fs: fix jbd2_journal_try_to_free_buffers() kernel-doc comment

Pull cifs client fixes from Steve French:
 "Two cifs debugging improvements - one found to deal with debugging a
  multichannel problem and one for a recent fallocate issue

  This does include the two larger multichannel reconnect (dynamically
  adjusting interfaces on reconnect) patches, because we recently found
  an additional problem with multichannel to one server type that I want
  to include at the same time"

* tag '5.19-rc2-smb3-client-fixes' of git://git.samba.org/sfrench/cifs-2.6:
  cifs: when a channel is not found for server, log its connection id
  smb3: add trace point for SMB2_set_eof

Signed-off-by: Xiang wangx <wangxiang@cdjrlc.com>
Link: https://lore.kernel.org/r/20220605091503.12513-1-wangxiang@cdjrlc.comSigned-off-by: Theodore Ts'o <tytso@mit.edu>

We capture a NULL pointer issue when resizing a corrupt ext4 image which
is freshly clear resize_inode feature (not run e2fsck). It could be
simply reproduced by following steps. The problem is because of the
resize_inode feature was cleared, and it will convert the filesystem to
meta_bg mode in ext4_resize_fs(), but the es->s_reserved_gdt_blocks was
not reduced to zero, so could we mistakenly call reserve_backup_gdb()
and passing an uninitialized resize_inode to it when adding new group
descriptors.

 mkfs.ext4 /dev/sda 3G
 tune2fs -O ^resize_inode /dev/sda #forget to run requested e2fsck
 mount /dev/sda /mnt
 resize2fs /dev/sda 8G

 ========
 BUG: kernel NULL pointer dereference, address: 0000000000000028
 CPU: 19 PID: 3243 Comm: resize2fs Not tainted 5.18.0-rc7-00001-gfde086c5ebfd #748
 ...
 RIP: 0010:ext4_flex_group_add+0xe08/0x2570
 ...
 Call Trace:
  <TASK>
  ext4_resize_fs+0xbec/0x1660
  __ext4_ioctl+0x1749/0x24e0
  ext4_ioctl+0x12/0x20
  __x64_sys_ioctl+0xa6/0x110
  do_syscall_64+0x3b/0x90
  entry_SYSCALL_64_after_hwframe+0x44/0xae
 RIP: 0033:0x7f2dd739617b
 ========

The fix is simple, add a check in ext4_resize_begin() to make sure that
the es->s_reserved_gdt_blocks is zero when the resize_inode feature is
disabled.

Cc: stable@kernel.org
Signed-off-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Ritesh Harjani <ritesh.list@gmail.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20220601092717.763694-1-yi.zhang@huawei.comSigned-off-by: Theodore Ts'o <tytso@mit.edu>

Since dx_make_map() may return -EFSCORRUPTED now, so change "count" to
be a signed integer so we can correctly check for an error code returned
by dx_make_map().

Fixes: 46c116b9 ("ext4: verify dir block before splitting it")
Cc: stable@kernel.org
Signed-off-by: Ding Xiang <dingxiang@cmss.chinamobile.com>
Link: https://lore.kernel.org/r/20220530100047.537598-1-dingxiang@cmss.chinamobile.comSigned-off-by: Theodore Ts'o <tytso@mit.edu>

ext4_mb_normalize_request() can move logical start of allocated blocks
to reduce fragmentation and better utilize preallocation. However logical
block requested as a start of allocation (ac->ac_o_ex.fe_logical) should
always be covered by allocated blocks so we should check that by
modifying and to or in the assertion.
Signed-off-by: Baokun Li <libaokun1@huawei.com>
Reviewed-by: Ritesh Harjani <ritesh.list@gmail.com>
Link: https://lore.kernel.org/r/20220528110017.354175-3-libaokun1@huawei.comSigned-off-by: Theodore Ts'o <tytso@mit.edu>

Hulk Robot reported a BUG_ON:
==================================================================
kernel BUG at fs/ext4/mballoc.c:3211!
[...]
RIP: 0010:ext4_mb_mark_diskspace_used.cold+0x85/0x136f
[...]
Call Trace:
 ext4_mb_new_blocks+0x9df/0x5d30
 ext4_ext_map_blocks+0x1803/0x4d80
 ext4_map_blocks+0x3a4/0x1a10
 ext4_writepages+0x126d/0x2c30
 do_writepages+0x7f/0x1b0
 __filemap_fdatawrite_range+0x285/0x3b0
 file_write_and_wait_range+0xb1/0x140
 ext4_sync_file+0x1aa/0xca0
 vfs_fsync_range+0xfb/0x260
 do_fsync+0x48/0xa0
[...]
==================================================================

Above issue may happen as follows:
-------------------------------------
do_fsync
 vfs_fsync_range
  ext4_sync_file
   file_write_and_wait_range
    __filemap_fdatawrite_range
     do_writepages
      ext4_writepages
       mpage_map_and_submit_extent
        mpage_map_one_extent
         ext4_map_blocks
          ext4_mb_new_blocks
           ext4_mb_normalize_request
            >>> start + size <= ac->ac_o_ex.fe_logical
           ext4_mb_regular_allocator
            ext4_mb_simple_scan_group
             ext4_mb_use_best_found
              ext4_mb_new_preallocation
               ext4_mb_new_inode_pa
                ext4_mb_use_inode_pa
                 >>> set ac->ac_b_ex.fe_len <= 0
           ext4_mb_mark_diskspace_used
            >>> BUG_ON(ac->ac_b_ex.fe_len <= 0);

we can easily reproduce this problem with the following commands:
	`fallocate -l100M disk`
	`mkfs.ext4 -b 1024 -g 256 disk`
	`mount disk /mnt`
	`fsstress -d /mnt -l 0 -n 1000 -p 1`

The size must be smaller than or equal to EXT4_BLOCKS_PER_GROUP.
Therefore, "start + size <= ac->ac_o_ex.fe_logical" may occur
when the size is truncated. So start should be the start position of
the group where ac_o_ex.fe_logical is located after alignment.
In addition, when the value of fe_logical or EXT4_BLOCKS_PER_GROUP
is very large, the value calculated by start_off is more accurate.

Cc: stable@kernel.org
Fixes: cd648b8a ("ext4: trim allocation requests to group size")
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Baokun Li <libaokun1@huawei.com>
Reviewed-by: Ritesh Harjani <ritesh.list@gmail.com>
Link: https://lore.kernel.org/r/20220528110017.354175-2-libaokun1@huawei.comSigned-off-by: Theodore Ts'o <tytso@mit.edu>

Since ext4 was converted to the new mount API, the test_dummy_encryption
mount option isn't being handled entirely correctly, because the needed
fscrypt_set_test_dummy_encryption() helper function combines
parsing/checking/applying into one function.  That doesn't work well
with the new mount API, which split these into separate steps.

This was sort of okay anyway, due to the parsing logic that was copied
from fscrypt_set_test_dummy_encryption() into ext4_parse_param(),
combined with an additional check in ext4_check_test_dummy_encryption().
However, these overlooked the case of changing the value of
test_dummy_encryption on remount, which isn't allowed but ext4 wasn't
detecting until ext4_apply_options() when it's too late to fail.
Another bug is that if test_dummy_encryption was specified multiple
times with an argument, memory was leaked.

Fix this up properly by using the new helper functions that allow
splitting up the parse/check/apply steps for test_dummy_encryption.

Fixes: cebe85d5 ("ext4: switch to the new mount api")
Signed-off-by: Eric Biggers <ebiggers@google.com>
Link: https://lore.kernel.org/r/20220526040412.173025-1-ebiggers@kernel.orgSigned-off-by: Theodore Ts'o <tytso@mit.edu>

Replace kmalloc + memcpy with kmemdup()
Signed-off-by: Shuqi Zhang <zhangshuqi3@huawei.com>
Reviewed-by: Ritesh Harjani <ritesh.list@gmail.com>
Link: https://lore.kernel.org/r/20220525030120.803330-1-zhangshuqi3@huawei.comSigned-off-by: Theodore Ts'o <tytso@mit.edu>

We got issue as follows:
[home]# mount  /dev/sda  test
EXT4-fs (sda): warning: mounting fs with errors, running e2fsck is recommended
[home]# dmesg
EXT4-fs (sda): warning: mounting fs with errors, running e2fsck is recommended
EXT4-fs (sda): Errors on filesystem, clearing orphan list.
EXT4-fs (sda): recovery complete
EXT4-fs (sda): mounted filesystem with ordered data mode. Quota mode: none.
[home]# debugfs /dev/sda
debugfs 1.46.5 (30-Dec-2021)
Checksum errors in superblock!  Retrying...

Reason is ext4_orphan_cleanup will reset ‘s_last_orphan’ but not update
super block checksum.

To solve above issue, defer update super block checksum after
ext4_orphan_cleanup.
Signed-off-by: Ye Bin <yebin10@huawei.com>
Cc: stable@kernel.org
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Ritesh Harjani <ritesh.list@gmail.com>
Link: https://lore.kernel.org/r/20220525012904.1604737-1-yebin10@huawei.comSigned-off-by: Theodore Ts'o <tytso@mit.edu>

cifs_ses_get_chan_index gets the index for a given server pointer.
When a match is not found, we warn about a possible bug.
However, printing details about the non-matching server could be
more useful to debug here.
Signed-off-by: Shyam Prasad N <sprasad@microsoft.com>
Signed-off-by: Steve French <stfrench@microsoft.com>

load_unaligned_zeropad() can lead to unwanted loads across page boundaries.
The unwanted loads are typically harmless. But, they might be made to
totally unrelated or even unmapped memory. load_unaligned_zeropad()
relies on exception fixup (#PF, #GP and now #VE) to recover from these
unwanted loads.

In TDX guests, the second page can be shared page and a VMM may configure
it to trigger #VE.

The kernel assumes that #VE on a shared page is an MMIO access and tries to
decode instruction to handle it. In case of load_unaligned_zeropad() it
may result in confusion as it is not MMIO access.

Fix it by detecting split page MMIO accesses and failing them.
load_unaligned_zeropad() will recover using exception fixups.

The issue was discovered by analysis and reproduced artificially. It was
not triggered during testing.

[ dhansen: fix up changelogs and comments for grammar and clarity,
	   plus incorporate Kirill's off-by-one fix]
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Link: https://lkml.kernel.org/r/20220614120135.14812-4-kirill.shutemov@linux.intel.com