mm/damon/core: move damon_set_targets() into dbgfs

damon_set_targets() function is defined in the core for general use cases,
but called from only dbgfs.  Also, because the function is for general use
cases, dbgfs does additional handling of pid type target id case.  To make
the situation simpler, this commit moves the function into dbgfs and makes
it to do the pid type case handling on its own.

Link: https://lkml.kernel.org/r/20211230100723.2238-4-sj@kernel.org
Signed-off-by: SeongJae Park <sj@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Docs/admin-guide/mm/damon/usage: update for changed initail_regions file input

A previous commit made init_regions debugfs file to use target index
instead of target id for specifying the target of the init regions.  This
commit updates the usage document to reflect the change.

Link: https://lkml.kernel.org/r/20211230100723.2238-3-sj@kernel.org
Signed-off-by: SeongJae Park <sj@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/damon/dbgfs/init_regions: use target index instead of target id

Patch series "Remove the type-unclear target id concept".

DAMON asks each monitoring target ('struct damon_target') to have one
'unsigned long' integer called 'id', which should be unique among the
targets of same monitoring context.  Meaning of it is, however, totally up
to the monitoring primitives that registered to the monitoring context.
For example, the virtual address spaces monitoring primitives treats the
id as a 'struct pid' pointer.

This makes the code flexible but ugly, not well-documented, and
type-unsafe[1].  Also, identification of each target can be done via its
index.  For the reason, this patchset removes the concept and uses clear
type definition.

[1] https://lore.kernel.org/linux-mm/20211013154535.4aaeaaf9d0182922e405dd1e@linux-foundation.org/

This patch (of 4):

Target id is a 'unsigned long' data, which can be interpreted differently
by each monitoring primitives.  For example, it means 'struct pid *' for
the virtual address spaces monitoring, while it means nothing but an
integer to be displayed to debugfs interface users for the physical
address space monitoring.  It's flexible but makes code ugly and
type-unsafe[1].

To be prepared for eventual removal of the concept, this commit removes a
use case of the concept in 'init_regions' debugfs file handling.  In
detail, this commit replaces use of the id with the index of each target
in the context's targets list.

[1] https://lore.kernel.org/linux-mm/20211013154535.4aaeaaf9d0182922e405dd1e@linux-foundation.org/

Link: https://lkml.kernel.org/r/20211230100723.2238-1-sj@kernel.org
Link: https://lkml.kernel.org/r/20211230100723.2238-2-sj@kernel.org
Signed-off-by: SeongJae Park <sj@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/hmm.c: remove unneeded local variable ret

The local variable ret is always 0. Remove it to make code more tight.

Link: https://lkml.kernel.org/r/20220125124833.39718-1-linmiaohe@huawei.com
Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
Reviewed-by: Muchun Song <songmuchun@bytedance.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

kfence: allow use of a deferrable timer

Allow the use of a deferrable timer, which does not force CPU wake-ups
when the system is idle.  A consequence is that the sample interval
becomes very unpredictable, to the point that it is not guaranteed that
the KFENCE KUnit test still passes.

Nevertheless, on power-constrained systems this may be preferable, so
let's give the user the option should they accept the above trade-off.

Link: https://lkml.kernel.org/r/20220308141415.3168078-1-elver@google.com
Signed-off-by: Marco Elver <elver@google.com>
Reviewed-by: Alexander Potapenko <glider@google.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

kfence: test: try to avoid test_gfpzero trigger rcu_stall

When CONFIG_KFENCE_NUM_OBJECTS is set to a big number, kfence
kunit-test-case test_gfpzero will eat up nearly all the CPU's resources
and rcu_stall is reported as the following log which is cut from a
physical server.

  rcu: INFO: rcu_sched self-detected stall on CPU
  rcu:  68-....: (14422 ticks this GP) idle=6ce/1/0x4000000000000002
  softirq=592/592 fqs=7500 (t=15004 jiffies g=10677 q=20019)
  Task dump for CPU 68:
  task:kunit_try_catch state:R  running task
  stack:    0 pid: 9728 ppid:     2 flags:0x0000020a
  Call trace:
   dump_backtrace+0x0/0x1e4
   show_stack+0x20/0x2c
   sched_show_task+0x148/0x170
   ...
   rcu_sched_clock_irq+0x70/0x180
   update_process_times+0x68/0xb0
   tick_sched_handle+0x38/0x74
   ...
   gic_handle_irq+0x78/0x2c0
   el1_irq+0xb8/0x140
   kfree+0xd8/0x53c
   test_alloc+0x264/0x310 [kfence_test]
   test_gfpzero+0xf4/0x840 [kfence_test]
   kunit_try_run_case+0x48/0x20c
   kunit_generic_run_threadfn_adapter+0x28/0x34
   kthread+0x108/0x13c
   ret_from_fork+0x10/0x18

To avoid rcu_stall and unacceptable latency, a schedule point is
added to test_gfpzero.

Link: https://lkml.kernel.org/r/20220309083753.1561921-4-liupeng256@huawei.com
Signed-off-by: Peng Liu <liupeng256@huawei.com>
Reviewed-by: Marco Elver <elver@google.com>
Tested-by: Brendan Higgins <brendanhiggins@google.com>
Cc: Alexander Potapenko <glider@google.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Wang Kefeng <wangkefeng.wang@huawei.com>
Cc: Daniel Latypov <dlatypov@google.com>
Cc: David Gow <davidgow@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

kunit: make kunit_test_timeout compatible with comment

In function kunit_test_timeout, it is declared "300 * MSEC_PER_SEC"
represent 5min.  However, it is wrong when dealing with arm64 whose
default HZ = 250, or some other situations.  Use msecs_to_jiffies to fix
this, and kunit_test_timeout will work as desired.

Link: https://lkml.kernel.org/r/20220309083753.1561921-3-liupeng256@huawei.com
Fixes: 5f3e06208920 ("kunit: test: add support for test abort")
Signed-off-by: Peng Liu <liupeng256@huawei.com>
Reviewed-by: Marco Elver <elver@google.com>
Reviewed-by: Daniel Latypov <dlatypov@google.com>
Reviewed-by: Brendan Higgins <brendanhiggins@google.com>
Tested-by: Brendan Higgins <brendanhiggins@google.com>
Cc: Alexander Potapenko <glider@google.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Wang Kefeng <wangkefeng.wang@huawei.com>
Cc: David Gow <davidgow@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

kunit: fix UAF when run kfence test case test_gfpzero

Patch series "kunit: fix a UAF bug and do some optimization", v2.

This series is to fix UAF (use after free) when running kfence test case
test_gfpzero, which is time costly.  This UAF bug can be easily triggered
by setting CONFIG_KFENCE_NUM_OBJECTS = 65535.  Furthermore, some
optimization for kunit tests has been done.

This patch (of 3):

Kunit will create a new thread to run an actual test case, and the main
process will wait for the completion of the actual test thread until
overtime.  The variable "struct kunit test" has local property in function
kunit_try_catch_run, and will be used in the test case thread.  Task
kunit_try_catch_run will free "struct kunit test" when kunit runs
overtime, but the actual test case is still run and an UAF bug will be
triggered.

The above problem has been both observed in a physical machine and qemu
platform when running kfence kunit tests.  The problem can be triggered
when setting CONFIG_KFENCE_NUM_OBJECTS = 65535.  Under this setting, the
test case test_gfpzero will cost hours and kunit will run to overtime.
The follows show the panic log.

  BUG: unable to handle page fault for address: ffffffff82d882e9

  Call Trace:
   kunit_log_append+0x58/0xd0
   ...
   test_alloc.constprop.0.cold+0x6b/0x8a [kfence_test]
   test_gfpzero.cold+0x61/0x8ab [kfence_test]
   kunit_try_run_case+0x4c/0x70
   kunit_generic_run_threadfn_adapter+0x11/0x20
   kthread+0x166/0x190
   ret_from_fork+0x22/0x30
  Kernel panic - not syncing: Fatal exception
  Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS
  Ubuntu-1.8.2-1ubuntu1 04/01/2014

To solve this problem, the test case thread should be stopped when the
kunit frame runs overtime.  The stop signal will send in function
kunit_try_catch_run, and test_gfpzero will handle it.

Link: https://lkml.kernel.org/r/20220309083753.1561921-1-liupeng256@huawei.com
Link: https://lkml.kernel.org/r/20220309083753.1561921-2-liupeng256@huawei.com
Signed-off-by: Peng Liu <liupeng256@huawei.com>
Reviewed-by: Marco Elver <elver@google.com>
Reviewed-by: Brendan Higgins <brendanhiggins@google.com>
Tested-by: Brendan Higgins <brendanhiggins@google.com>
Cc: Alexander Potapenko <glider@google.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Wang Kefeng <wangkefeng.wang@huawei.com>
Cc: Daniel Latypov <dlatypov@google.com>
Cc: David Gow <davidgow@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

kfence: alloc kfence_pool after system startup

Allow enabling KFENCE after system startup by allocating its pool via the
page allocator. This provides the flexibility to enable KFENCE even if it
wasn't enabled at boot time.

Link: https://lkml.kernel.org/r/20220307074516.6920-3-dtcccc@linux.alibaba.com
Signed-off-by: Tianchen Ding <dtcccc@linux.alibaba.com>
Reviewed-by: Marco Elver <elver@google.com>
Tested-by: Peng Liu <liupeng256@huawei.com>
Cc: Alexander Potapenko <glider@google.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

kfence: allow re-enabling KFENCE after system startup

Patch series "provide the flexibility to enable KFENCE", v3.

If CONFIG_CONTIG_ALLOC is not supported, we fallback to try
alloc_pages_exact().  Allocating pages in this way has limits about
MAX_ORDER (default 11).  So we will not support allocating kfence pool
after system startup with a large KFENCE_NUM_OBJECTS.

When handling failures in kfence_init_pool_late(), we pair
free_pages_exact() to alloc_pages_exact() for compatibility consideration,
though it actually does the same as free_contig_range().

This patch (of 2):

If once KFENCE is disabled by:
echo 0 > /sys/module/kfence/parameters/sample_interval
KFENCE could never be re-enabled until next rebooting.

Allow re-enabling it by writing a positive num to sample_interval.

Link: https://lkml.kernel.org/r/20220307074516.6920-1-dtcccc@linux.alibaba.com
Link: https://lkml.kernel.org/r/20220307074516.6920-2-dtcccc@linux.alibaba.com
Signed-off-by: Tianchen Ding <dtcccc@linux.alibaba.com>
Reviewed-by: Marco Elver <elver@google.com>
Cc: Alexander Potapenko <glider@google.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/kfence: remove unnecessary CONFIG_KFENCE option

In mm/Makefile has:

  obj-$(CONFIG_KFENCE) += kfence/

So that we don't need 'obj-$(CONFIG_KFENCE) :=' in mm/kfence/Makefile,
delete it from mm/kfence/Makefile.

Link: https://lkml.kernel.org/r/20220221065525.21344-1-tangmeng@uniontech.com
Signed-off-by: tangmeng <tangmeng@uniontech.com>
Reviewed-by: Marco Elver <elver@google.com>
Cc: Alexander Potapenko <glider@google.com>
Cc: Dmitriy Vyukov <dvyukov@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/page_table_check.c: use strtobool for param parsing

Use strtobool rather than open coding "on" and "off" parsing.

Link: https://lkml.kernel.org/r/20220227181038.126926-1-linux@treblig.org
Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/highmem: remove unnecessary done label

Remove unnecessary done label to simplify the code.

Link: https://lkml.kernel.org/r/20220126092542.64659-1-linmiaohe@huawei.com
Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
Reviewed-by: Muchun Song <songmuchun@bytedance.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Acked-by: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

highmem: document kunmap_local()

Some users of kmap() add an offset to the kmap() address to be used
during the mapping.

When converting to kmap_local_page() the base address does not need to
be stored because any address within the page can be used in
kunmap_local().  However, this was not clear from the documentation and
cause some questions.[1]

Document that any address in the page can be used in kunmap_local() to
clarify this for future users.

[1] https://lore.kernel.org/lkml/20211213154543.GM3538886@iweiny-DESK2.sc.intel.com/

[ira.weiny@intel.com: updates per Christoph]
Link: https://lkml.kernel.org/r/20220124182138.816693-1-ira.weiny@intel.com
Link: https://lkml.kernel.org/r/20220124013045.806718-1-ira.weiny@intel.com
Signed-off-by: Ira Weiny <ira.weiny@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/early_ioremap: declare early_memremap_pgprot_adjust()

The mm/ directory can almost fully be built with W=1, which would help
in local development.  One remaining issue is missing prototype for
early_memremap_pgprot_adjust().

Thus add a declaration for this function.  Use mm/internal.h instead of
asm/early_ioremap.h to avoid missing type definitions and unnecessary
exposure.

Link: https://lkml.kernel.org/r/20220314165724.16071-2-vbabka@suse.cz
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: David Hildenbrand <david@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/usercopy: return 1 from hardened_usercopy __setup() handler

__setup() handlers should return 1 if the command line option is handled
and 0 if not (or maybe never return 0; it just pollutes init's
environment).  This prevents:

  Unknown kernel command line parameters \
  "BOOT_IMAGE=/boot/bzImage-517rc5 hardened_usercopy=off", will be \
  passed to user space.

  Run /sbin/init as init process
   with arguments:
     /sbin/init
   with environment:
     HOME=/
     TERM=linux
     BOOT_IMAGE=/boot/bzImage-517rc5
     hardened_usercopy=off
or
     hardened_usercopy=on
but when "hardened_usercopy=foo" is used, there is no Unknown kernel
command line parameter.

Return 1 to indicate that the boot option has been handled.
Print a warning if strtobool() returns an error on the option string,
but do not mark this as in unknown command line option and do not cause
init's environment to be polluted with this string.

Link: https://lkml.kernel.org/r/20220222034249.14795-1-rdunlap@infradead.org
Link: lore.kernel.org/r/64644a2f-4a20-bab3-1e15-3b2cdd0defe3@omprussia.ru
Fixes: b5cb15d9372ab ("usercopy: Allow boot cmdline disabling of hardening")
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Reported-by: Igor Zhbanov <i.zhbanov@omprussia.ru>
Acked-by: Chris von Recklinghausen <crecklin@redhat.com>
Cc: Kees Cook <keescook@chromium.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm: uninline copy_overflow()

While building a small config with CONFIG_CC_OPTIMISE_FOR_SIZE, I ended
up with more than 50 times the following function in vmlinux because GCC
doesn't honor the 'inline' keyword:

c00243bc <copy_overflow>:
c00243bc: 94 21 ff f0  stwu    r1,-16(r1)
c00243c0: 7c 85 23 78  mr      r5,r4
c00243c4: 7c 64 1b 78  mr      r4,r3
c00243c8: 3c 60 c0 62  lis     r3,-16286
c00243cc: 7c 08 02 a6  mflr    r0
c00243d0: 38 63 5e e5  addi    r3,r3,24293
c00243d4: 90 01 00 14  stw     r0,20(r1)
c00243d8: 4b ff 82 45  bl      c001c61c <__warn_printk>
c00243dc: 0f e0 00 00  twui    r0,0
c00243e0: 80 01 00 14  lwz     r0,20(r1)
c00243e4: 38 21 00 10  addi    r1,r1,16
c00243e8: 7c 08 03 a6  mtlr    r0
c00243ec: 4e 80 00 20  blr

With -Winline, GCC tells:

/include/linux/thread_info.h:212:20: warning: inlining failed in call to 'copy_overflow': call is unlikely and code size would grow [-Winline]

copy_overflow() is a non conditional warning called by check_copy_size()
on an error path.

check_copy_size() have to remain inlined in order to benefit from
constant folding, but copy_overflow() is not worth inlining.

Uninline the warning when CONFIG_BUG is selected.

When CONFIG_BUG is not selected, WARN() does nothing so skip it.

This reduces the size of vmlinux by almost 4kbytes.

Link: https://lkml.kernel.org/r/e1723b9cfa924bcefcd41f69d0025b38e4c9364e.1644819985.git.christophe.leroy@csgroup.eu
Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu>
Cc: David Laight <David.Laight@ACULAB.COM>
Cc: Anshuman Khandual <anshuman.khandual@arm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm: remove usercopy_warn()

Users of usercopy_warn() were removed by commit 53944f171a89 ("mm:
remove HARDENED_USERCOPY_FALLBACK")

Remove it.

Link: https://lkml.kernel.org/r/5f26643fc70b05f8455b60b99c30c17d635fa640.1644231910.git.christophe.leroy@csgroup.eu
Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu>
Reviewed-by: Miaohe Lin <linmiaohe@huawei.com>
Reviewed-by: Stephen Kitt <steve@sk2.org>
Reviewed-by: Muchun Song <songmuchun@bytedance.com>
Cc: Kees Cook <keescook@chromium.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/zswap.c: allow handling just same-value filled pages

Zswap has an ability to efficiently store same-value filled pages, which
can be turned on and off using the "same_filled_pages_enabled"
parameter.

However, there is currently no way to enable just this (lightweight)
functionality, while not making use of the whole compressed page storage
machinery.

Add a "non_same_filled_pages_enabled" parameter which allows disabling
handling of pages that aren't same-value filled.  This way zswap can be
run in such lightweight same-value filled pages only mode.

Link: https://lkml.kernel.org/r/7dbafa963e8bab43608189abbe2067f4b9287831.1641247624.git.maciej.szmigiero@oracle.com
Signed-off-by: Maciej S. Szmigiero <maciej.szmigiero@oracle.com>
Cc: Seth Jennings <sjenning@redhat.com>
Cc: Dan Streetman <ddstreet@ieee.org>
Cc: Vitaly Wool <vitaly.wool@konsulko.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/thp: ClearPageDoubleMap in first page_add_file_rmap()

PageDoubleMap is maintained differently for anon and for shmem+file: the
shmem+file one was never cleared, because a safe place to do so could
not be found; so it would blight future use of the cached hugepage until
evicted.

See https://lore.kernel.org/lkml/1571938066-29031-1-git-send-email-yang.shi@linux.alibaba.com/

But page_add_file_rmap() does provide a safe place to do so (though later
than one might wish): allowing testing to return to an initial state
without a damaging drop_caches.

Link: https://lkml.kernel.org/r/61c5cf99-a962-9a25-597a-53ab1bd8fbc0@google.com
Fixes: 9a73f61bdb8a ("thp, mlock: do not mlock PTE-mapped file huge pages")
Signed-off-by: Hugh Dickins <hughd@google.com>
Reviewed-by: Yang Shi <shy828301@gmail.com>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm: only re-generate demotion targets when a numa node changes its N_CPU state

Abhishek reported that after patch [1], hotplug operations are taking
roughly double the expected time.  [2]

The reason behind is that the CPU callbacks that
migrate_on_reclaim_init() sets always call set_migration_target_nodes()
whenever a CPU is brought up/down.

But we only care about numa nodes going from having cpus to become
cpuless, and vice versa, as that influences the demotion_target order.

We do already have two CPU callbacks (vmstat_cpu_online() and
vmstat_cpu_dead()) that check exactly that, so get rid of the CPU
callbacks in migrate_on_reclaim_init() and only call
set_migration_target_nodes() from vmstat_cpu_{dead,online}() whenever a
numa node change its N_CPU state.

[1] https://lore.kernel.org/linux-mm/20210721063926.3024591-2-ying.huang@intel.com/
[2] https://lore.kernel.org/linux-mm/eb438ddd-2919-73d4-bd9f-b7eecdd9577a@linux.vnet.ibm.com/

[osalvador@suse.de: add feedback from Huang Ying]
Link: https://lkml.kernel.org/r/20220314150945.12694-1-osalvador@suse.de
Link: https://lkml.kernel.org/r/20220310120749.23077-1-osalvador@suse.de
Fixes: 884a6e5d1f93b ("mm/migrate: update node demotion order on hotplug events")
Signed-off-by: Oscar Salvador <osalvador@suse.de>
Reviewed-by: Baolin Wang <baolin.wang@linux.alibaba.com>
Tested-by: Baolin Wang <baolin.wang@linux.alibaba.com>
Reported-by: Abhishek Goel <huntbag@linux.vnet.ibm.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: "Huang, Ying" <ying.huang@intel.com>
Cc: Abhishek Goel <huntbag@linux.vnet.ibm.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

drivers/base/memory: clarify adding and removing of memory blocks

Let's make it clearer at which places we actually add and remove memory
blocks -- streamlining the terminology -- and highlight which memory block
start out online and which start out as offline.

 * rename add_memory_block -> add_boot_memory_block
 * rename init_memory_block -> add_memory_block
 * rename unregister_memory -> remove_memory_block
 * rename register_memory -> __add_memory_block
 * add add_hotplug_memory_block
 * mark add_boot_memory_block with __init (suggested by Oscar)

__add_memory_block() is  a pure helper for add_memory_block(), remove
the somewhat obvious comment.

Link: https://lkml.kernel.org/r/20220221154531.11382-1-david@redhat.com
Signed-off-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Oscar Salvador <osalvador@suse.de>
Cc: "Rafael J. Wysocki" <rafael@kernel.org>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Michal Hocko <mhocko@suse.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

drivers/base/memory: determine and store zone for single-zone memory blocks

test_pages_in_a_zone() is just another nasty PFN walker that can easily
stumble over ZONE_DEVICE memory ranges falling into the same memory block
as ordinary system RAM: the memmap of parts of these ranges might possibly
be uninitialized.  In fact, we observed (on an older kernel) with UBSAN:

  UBSAN: Undefined behaviour in ./include/linux/mm.h:1133:50
  index 7 is out of range for type 'zone [5]'
  CPU: 121 PID: 35603 Comm: read_all Kdump: loaded Tainted: [...]
  Hardware name: Dell Inc. PowerEdge R7425/08V001, BIOS 1.12.2 11/15/2019
  Call Trace:
   dump_stack+0x9a/0xf0
   ubsan_epilogue+0x9/0x7a
   __ubsan_handle_out_of_bounds+0x13a/0x181
   test_pages_in_a_zone+0x3c4/0x500
   show_valid_zones+0x1fa/0x380
   dev_attr_show+0x43/0xb0
   sysfs_kf_seq_show+0x1c5/0x440
   seq_read+0x49d/0x1190
   vfs_read+0xff/0x300
   ksys_read+0xb8/0x170
   do_syscall_64+0xa5/0x4b0
   entry_SYSCALL_64_after_hwframe+0x6a/0xdf
  RIP: 0033:0x7f01f4439b52

We seem to stumble over a memmap that contains a garbage zone id.  While
we could try inserting pfn_to_online_page() calls, it will just make
memory offlining slower, because we use test_pages_in_a_zone() to make
sure we're offlining pages that all belong to the same zone.

Let's just get rid of this PFN walker and determine the single zone of a
memory block -- if any -- for early memory blocks during boot.  For memory
onlining, we know the single zone already.  Let's avoid any additional
memmap scanning and just rely on the zone information available during
boot.

For memory hot(un)plug, we only really care about memory blocks that:
* span a single zone (and, thereby, a single node)
* are completely System RAM (IOW, no holes, no ZONE_DEVICE)
If one of these conditions is not met, we reject memory offlining.
Hotplugged memory blocks (starting out offline), always meet both
conditions.

There are three scenarios to handle:

(1) Memory hot(un)plug

A memory block with zone == NULL cannot be offlined, corresponding to
our previous test_pages_in_a_zone() check.

After successful memory onlining/offlining, we simply set the zone
accordingly.
* Memory onlining: set the zone we just used for onlining
* Memory offlining: set zone = NULL

So a hotplugged memory block starts with zone = NULL. Once memory
onlining is done, we set the proper zone.

(2) Boot memory with !CONFIG_NUMA

We know that there is just a single pgdat, so we simply scan all zones
of that pgdat for an intersection with our memory block PFN range when
adding the memory block. If more than one zone intersects (e.g., DMA and
DMA32 on x86 for the first memory block) we set zone = NULL and
consequently mimic what test_pages_in_a_zone() used to do.

(3) Boot memory with CONFIG_NUMA

At the point in time we create the memory block devices during boot, we
don't know yet which nodes *actually* span a memory block. While we could
scan all zones of all nodes for intersections, overlapping nodes complicate
the situation and scanning all nodes is possibly expensive. But that
problem has already been solved by the code that sets the node of a memory
block and creates the link in the sysfs --
do_register_memory_block_under_node().

So, we hook into the code that sets the node id for a memory block. If
we already have a different node id set for the memory block, we know
that multiple nodes *actually* have PFNs falling into our memory block:
we set zone = NULL and consequently mimic what test_pages_in_a_zone() used
to do. If there is no node id set, we do the same as (2) for the given
node.

Note that the call order in driver_init() is:
-> memory_dev_init(): create memory block devices
-> node_dev_init(): link memory block devices to the node and set the
    node id

So in summary, we detect if there is a single zone responsible for this
memory block and we consequently store the zone in that case in the
memory block, updating it during memory onlining/offlining.

Link: https://lkml.kernel.org/r/20220210184359.235565-3-david@redhat.com
Signed-off-by: David Hildenbrand <david@redhat.com>
Reported-by: Rafael Parra <rparrazo@redhat.com>
Reviewed-by: Oscar Salvador <osalvador@suse.de>
Cc: "Rafael J. Wysocki" <rafael@kernel.org>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Rafael Parra <rparrazo@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

drivers/base/node: rename link_mem_sections() to register_memory_block_under_node()

Patch series "drivers/base/memory: determine and store zone for single-zone memory blocks", v2.

I remember talking to Michal in the past about removing
test_pages_in_a_zone(), which we use for:
* verifying that a memory block we intend to offline is really only managed
  by a single zone. We don't support offlining of memory blocks that are
  managed by multiple zones (e.g., multiple nodes, DMA and DMA32)
* exposing that zone to user space via
  /sys/devices/system/memory/memory*/valid_zones

Now that I identified some more cases where test_pages_in_a_zone() might
go wrong, and we received an UBSAN report (see patch #3), let's get rid of
this PFN walker.

So instead of detecting the zone at runtime with test_pages_in_a_zone() by
scanning the memmap, let's determine and remember for each memory block if
it's managed by a single zone.  The stored zone can then be used for the
above two cases, avoiding a manual lookup using test_pages_in_a_zone().

This avoids eventually stumbling over uninitialized memmaps in corner
cases, especially when ZONE_DEVICE ranges partly fall into memory block
(that are responsible for managing System RAM).

Handling memory onlining is easy, because we online to exactly one zone.
Handling boot memory is more tricky, because we want to avoid scanning all
zones of all nodes to detect possible zones that overlap with the physical
memory region of interest.  Fortunately, we already have code that
determines the applicable nodes for a memory block, to create sysfs links
-- we'll hook into that.

Patch #1 is a simple cleanup I had laying around for a longer time.
Patch #2 contains the main logic to remove test_pages_in_a_zone() and
further details.

[1] https://lkml.kernel.org/r/20220128144540.153902-1-david@redhat.com
[2] https://lkml.kernel.org/r/20220203105212.30385-1-david@redhat.com

This patch (of 2):

Let's adjust the stale terminology, making it match
unregister_memory_block_under_nodes() and
do_register_memory_block_under_node().  We're dealing with memory block
devices, which span 1..X memory sections.

Link: https://lkml.kernel.org/r/20220210184359.235565-1-david@redhat.com
Link: https://lkml.kernel.org/r/20220210184359.235565-2-david@redhat.com
Signed-off-by: David Hildenbrand <david@redhat.com>
Acked-by: Oscar Salvador <osalvador@suse.de>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Michal Hocko <mhocko@suse.com>
Cc: "Rafael J. Wysocki" <rafael@kernel.org>
Cc: Rafael Parra <rparrazo@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/memory_hotplug: fix misplaced comment in offline_pages

It's misplaced since commit 7960509329c2 ("mm, memory_hotplug: print
reason for the offlining failure").  Move it to the right place.

Link: https://lkml.kernel.org/r/20220207133643.23427-5-linmiaohe@huawei.com
Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Oscar Salvador <osalvador@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/memory_hotplug: clean up try_offline_node

We can use helper macro node_spanned_pages to check whether node spans
pages.  And we can change the parameter of check_cpu_on_node to nid as
that's what it really cares.  Thus we can further get rid of the local
variable pgdat and improve the readability a bit.

Link: https://lkml.kernel.org/r/20220207133643.23427-4-linmiaohe@huawei.com
Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Oscar Salvador <osalvador@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/memory_hotplug: avoid calling zone_intersects() for ZONE_NORMAL

If zid reaches ZONE_NORMAL, the caller will always get the NORMAL zone no
matter what zone_intersects() returns.  So we can save some possible cpu
cycles by avoid calling zone_intersects() for ZONE_NORMAL.

Link: https://lkml.kernel.org/r/20220207133643.23427-3-linmiaohe@huawei.com
Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Oscar Salvador <osalvador@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/memory_hotplug: remove obsolete comment of __add_pages

Patch series "A few cleanup patches around memory_hotplug".

This series contains a few patches to fix obsolete and misplaced comments,
clean up the try_offline_node function and so on.

This patch (of 4):

Since commit f1dd2cd13c4b ("mm, memory_hotplug: do not associate hotadded
memory to zones until online"), there is no need to pass in the zone.

[akpm@linux-foundation.org: remove the comment altogether, per David]

Link: https://lkml.kernel.org/r/20220207133643.23427-1-linmiaohe@huawei.com
Link: https://lkml.kernel.org/r/20220207133643.23427-2-linmiaohe@huawei.com
Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
Cc: David Hildenbrand <david@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

drivers/base/node: consolidate node device subsystem initialization in node_dev_init()

...  and call node_dev_init() after memory_dev_init() from driver_init(),
so before any of the existing arch/subsys calls.  All online nodes should
be known at that point: early during boot, arch code determines node and
zone ranges and sets the relevant nodes online; usually this happens in
setup_arch().

This is in line with memory_dev_init(), which initializes the memory
device subsystem and creates all memory block devices.

Similar to memory_dev_init(), panic() if anything goes wrong, we don't
want to continue with such basic initialization errors.

The important part is that node_dev_init() gets called after
memory_dev_init() and after cpu_dev_init(), but before any of the relevant
archs call register_cpu() to register the new cpu device under the node
device.  The latter should be the case for the current users of
topology_init().

Link: https://lkml.kernel.org/r/20220203105212.30385-1-david@redhat.com
Signed-off-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Oscar Salvador <osalvador@suse.de>
Tested-by: Anatoly Pugachev <matorola@gmail.com> (sparc64)
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Oscar Salvador <osalvador@suse.de>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Will Deacon <will@kernel.org>
Cc: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Paul Walmsley <paul.walmsley@sifive.com>
Cc: Palmer Dabbelt <palmer@dabbelt.com>
Cc: Albert Ou <aou@eecs.berkeley.edu>
Cc: Heiko Carstens <hca@linux.ibm.com>
Cc: Vasily Gorbik <gor@linux.ibm.com>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Cc: Rich Felker <dalias@libc.org>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: "Rafael J. Wysocki" <rafael@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

drivers/base/memory: add memory block to memory group after registration succeeded

If register_memory() fails, we freed the memory block but already added
the memory block to the group list, not good.  Let's defer adding the
block to the memory group to after registering the memory block device.

We do handle it properly during unregister_memory(), but that's not
called when the registration fails.

Link: https://lkml.kernel.org/r/20220128144540.153902-1-david@redhat.com
Fixes: 028fc57a1c36 ("drivers/base/memory: introduce "memory groups" to logically group memory blocks")
Signed-off-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Oscar Salvador <osalvador@suse.de>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: "Rafael J. Wysocki" <rafael@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

memcg: do not tweak node in alloc_mem_cgroup_per_node_info

alloc_mem_cgroup_per_node_info is allocated for each possible node and
this used to be a problem because !node_online nodes didn't have
appropriate data structure allocated.  This has changed by "mm: handle
uninitialized numa nodes gracefully" so we can drop the special casing
here.

Link: https://lkml.kernel.org/r/20220127085305.20890-7-mhocko@kernel.org
Signed-off-by: Wei Yang <richard.weiyang@gmail.com>
Signed-off-by: Michal Hocko <mhocko@suse.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Alexey Makhalov <amakhalov@vmware.com>
Cc: Dennis Zhou <dennis@kernel.org>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Christoph Lameter <cl@linux.com>
Cc: Nico Pache <npache@redhat.com>
Cc: Wei Yang <richard.weiyang@gmail.com>
Cc: Mike Rapoport <rppt@linux.ibm.com>
Cc: Oscar Salvador <osalvador@suse.de>
Cc: Rafael Aquini <raquini@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm: make free_area_init_node aware of memory less nodes

free_area_init_node is also called from memory less node initialization
path (free_area_init_memoryless_node).  It doesn't really make much sense
to display the physical memory range for those nodes: Initmem setup node
XX [mem 0x0000000000000000-0x0000000000000000]

Instead be explicit that the node is memoryless: Initmem setup node XX as
memoryless

Link: https://lkml.kernel.org/r/20220127085305.20890-6-mhocko@kernel.org
Signed-off-by: Michal Hocko <mhocko@suse.com>
Acked-by: Rafael Aquini <raquini@redhat.com>
Acked-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Mike Rapoport <rppt@linux.ibm.com>
Reviewed-by: Oscar Salvador <osalvador@suse.de>
Cc: Alexey Makhalov <amakhalov@vmware.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Dennis Zhou <dennis@kernel.org>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Nico Pache <npache@redhat.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Wei Yang <richard.weiyang@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm, memory_hotplug: reorganize new pgdat initialization

When a !node_online node is brought up it needs a hotplug specific
initialization because the node could be either uninitialized yet or it
could have been recycled after previous hotremove.  hotadd_init_pgdat is
responsible for that.

Internal pgdat state is initialized at two places currently
- hotadd_init_pgdat
- free_area_init_core_hotplug

There is no real clear cut what should go where but this patch's chosen to
move the whole internal state initialization into
free_area_init_core_hotplug.  hotadd_init_pgdat is still responsible to
pull all the parts together - most notably to initialize zonelists because
those depend on the overall topology.

This patch doesn't introduce any functional change.

Link: https://lkml.kernel.org/r/20220127085305.20890-5-mhocko@kernel.org
Signed-off-by: Michal Hocko <mhocko@suse.com>
Acked-by: Rafael Aquini <raquini@redhat.com>
Acked-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Oscar Salvador <osalvador@suse.de>
Cc: Alexey Makhalov <amakhalov@vmware.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Dennis Zhou <dennis@kernel.org>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Mike Rapoport <rppt@linux.ibm.com>
Cc: Nico Pache <npache@redhat.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Wei Yang <richard.weiyang@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm, memory_hotplug: drop arch_free_nodedata

Prior to "mm: handle uninitialized numa nodes gracefully" memory hotplug
used to allocate pgdat when memory has been added to a node
(hotadd_init_pgdat) arch_free_nodedata has been only used in the failure
path because once the pgdat is exported (to be visible by NODA_DATA(nid))
it cannot really be freed because there is no synchronization available
for that.

pgdat is allocated for each possible nodes now so the memory hotplug
doesn't need to do the ever use arch_free_nodedata so drop it.

This patch doesn't introduce any functional change.

Link: https://lkml.kernel.org/r/20220127085305.20890-4-mhocko@kernel.org
Signed-off-by: Michal Hocko <mhocko@suse.com>
Acked-by: Rafael Aquini <raquini@redhat.com>
Acked-by: David Hildenbrand <david@redhat.com>
Acked-by: Mike Rapoport <rppt@linux.ibm.com>
Reviewed-by: Oscar Salvador <osalvador@suse.de>
Cc: Alexey Makhalov <amakhalov@vmware.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Dennis Zhou <dennis@kernel.org>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Nico Pache <npache@redhat.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Wei Yang <richard.weiyang@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm: handle uninitialized numa nodes gracefully

We have had several reports [1][2][3] that page allocator blows up when an
allocation from a possible node is requested.  The underlying reason is
that NODE_DATA for the specific node is not allocated.

NUMA specific initialization is arch specific and it can vary a lot.  E.g.
x86 tries to initialize all nodes that have some cpu affinity (see
init_cpu_to_node) but this can be insufficient because the node might be
cpuless for example.

One way to address this problem would be to check for !node_online nodes
when trying to get a zonelist and silently fall back to another node.
That is unfortunately adding a branch into allocator hot path and it
doesn't handle any other potential NODE_DATA users.

This patch takes a different approach (following a lead of [3]) and it pre
allocates pgdat for all possible nodes in an arch indipendent code -
free_area_init.  All uninitialized nodes are treated as memoryless nodes.
node_state of the node is not changed because that would lead to other
side effects - e.g.  sysfs representation of such a node and from past
discussions [4] it is known that some tools might have problems digesting
that.

Newly allocated pgdat only gets a minimal initialization and the rest of
the work is expected to be done by the memory hotplug - hotadd_new_pgdat
(renamed to hotadd_init_pgdat).

generic_alloc_nodedata is changed to use the memblock allocator because
neither page nor slab allocators are available at the stage when all
pgdats are allocated.  Hotplug doesn't allocate pgdat anymore so we can
use the early boot allocator.  The only arch specific implementation is
ia64 and that is changed to use the early allocator as well.

[1] http://lkml.kernel.org/r/20211101201312.11589-1-amakhalov@vmware.com
[2] http://lkml.kernel.org/r/20211207224013.880775-1-npache@redhat.com
[3] http://lkml.kernel.org/r/20190114082416.30939-1-mhocko@kernel.org
[4] http://lkml.kernel.org/r/20200428093836.27190-1-srikar@linux.vnet.ibm.com

[akpm@linux-foundation.org: replace comment, per Mike]

Link: https://lkml.kernel.org/r/Yfe7RBeLCijnWBON@dhcp22.suse.cz
Reported-by: Alexey Makhalov <amakhalov@vmware.com>
Tested-by: Alexey Makhalov <amakhalov@vmware.com>
Reported-by: Nico Pache <npache@redhat.com>
Acked-by: Rafael Aquini <raquini@redhat.com>
Tested-by: Rafael Aquini <raquini@redhat.com>
Acked-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Oscar Salvador <osalvador@suse.de>
Acked-by: Mike Rapoport <rppt@linux.ibm.com>
Signed-off-by: Michal Hocko <mhocko@suse.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Dennis Zhou <dennis@kernel.org>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Wei Yang <richard.weiyang@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm, memory_hotplug: make arch_alloc_nodedata independent on CONFIG_MEMORY_HOTPLUG

Patch series "mm, memory_hotplug: handle unitialized numa node gracefully".

The core of the fix is patch 2 which also links existing bug reports.  The
high level goal is to have all possible numa nodes have their pgdat
allocated and initialized so

for_each_possible_node(nid)
NODE_DATA(nid)

will never return garbage.  This has proven to be problem in several
places when an offline numa node is used for an allocation just to realize
that node_data and therefore allocation fallback zonelists are not
initialized and such an allocation request blows up.

There were attempts to address that by checking node_online in several
places including the page allocator.  This patchset approaches the problem
from a different perspective and instead of special casing, which just
adds a runtime overhead, it allocates pglist_data for each possible node.
This can add some memory overhead for platforms with high number of
possible nodes if they do not contain any memory.  This should be a rather
rare configuration though.

How to test this? David has provided and excellent howto:
http://lkml.kernel.org/r/6e5ebc19-890c-b6dd-1924-9f25c441010d@redhat.com

Patches 1 and 3-6 are mostly cleanups.  The patchset has been reviewed by
Rafael (thanks!) and the core fix tested by Rafael and Alexey (thanks to
both).  David has tested as per instructions above and hasn't found any
fallouts in the memory hotplug scenarios.

This patch (of 6):

This is a preparatory patch and it doesn't introduce any functional
change.  It merely pulls out arch_alloc_nodedata (and co) outside of
CONFIG_MEMORY_HOTPLUG because the following patch will need to call this
from the generic MM code.

Link: https://lkml.kernel.org/r/20220127085305.20890-1-mhocko@kernel.org
Link: https://lkml.kernel.org/r/20220127085305.20890-2-mhocko@kernel.org
Signed-off-by: Michal Hocko <mhocko@suse.com>
Acked-by: Rafael Aquini <raquini@redhat.com>
Acked-by: David Hildenbrand <david@redhat.com>
Acked-by: Mike Rapoport <rppt@linux.ibm.com>
Reviewed-by: Oscar Salvador <osalvador@suse.de>
Reviewed-by: Wei Yang <richard.weiyang@gmail.com>
Cc: Alexey Makhalov <amakhalov@vmware.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Dennis Zhou <dennis@kernel.org>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Nico Pache <npache@redhat.com>
Cc: Tejun Heo <tj@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm: madvise: skip unmapped vma holes passed to process_madvise

The process_madvise() system call is expected to skip holes in vma passed
through 'struct iovec' vector list.  But do_madvise, which
process_madvise() calls for each vma, returns ENOMEM in case of unmapped
holes, despite the VMA is processed.

Thus process_madvise() should treat ENOMEM as expected and consider the
VMA passed to as processed and continue processing other vma's in the
vector list.  Returning -ENOMEM to user, despite the VMA is processed,
will be unable to figure out where to start the next madvise.

Link: https://lkml.kernel.org/r/4f091776142f2ebf7b94018146de72318474e686.1647008754.git.quic_charante@quicinc.com
Fixes: ecb8ac8b1f14("mm/madvise: introduce process_madvise() syscall: an external memory hinting API")
Signed-off-by: Charan Teja Kalla <quic_charante@quicinc.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Nadav Amit <nadav.amit@gmail.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm: madvise: return correct bytes advised with process_madvise

Patch series "mm: madvise: return correct bytes processed with
process_madvise", v2.  With the process_madvise(), always choose to return
non zero processed bytes over an error.  This can help the user to know on
which VMA, passed in the 'struct iovec' vector list, is failed to advise
thus can take the decission of retrying/skipping on that VMA.

This patch (of 2):

The process_madvise() system call returns error even after processing some
VMA's passed in the 'struct iovec' vector list which leaves the user
confused to know where to restart the advise next.  It is also against
this syscall man page[1] documentation where it mentions that "return
value may be less than the total number of requested bytes, if an error
occurred after some iovec elements were already processed.".

Consider a user passed 10 VMA's in the 'struct iovec' vector list of which
9 are processed but one.  Then it just returns the error caused on that
failed VMA despite the first 9 VMA's processed, leaving the user confused
about on which VMA it is failed.  Returning the number of bytes processed
here can help the user to know which VMA it is failed on and thus can
retry/skip the advise on that VMA.

[1]https://man7.org/linux/man-pages/man2/process_madvise.2.html.

Link: https://lkml.kernel.org/r/cover.1647008754.git.quic_charante@quicinc.com
Link: https://lkml.kernel.org/r/125b61a0edcee5c2db8658aed9d06a43a19ccafc.1647008754.git.quic_charante@quicinc.com
Fixes: ecb8ac8b1f14("mm/madvise: introduce process_madvise() syscall: an external memory hinting API")
Signed-off-by: Charan Teja Kalla <quic_charante@quicinc.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: David Rientjes <rientjes@google.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Nadav Amit <nadav.amit@gmail.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/madvise: use vma_lookup() instead of find_vma()

Using vma_lookup() verifies the start address is contained in the found
vma.  This results in easier to read the code.

Link: https://lkml.kernel.org/r/20220311082731.63513-1-linmiaohe@huawei.com
Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/hwpoison: check the subpage, not the head page

Hardware poison is tracked on a per-page basis, not on the head page.

Link: https://lkml.kernel.org/r/20220130013042.1906881-1-willy@infradead.org
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Acked-by: Naoya Horiguchi <naoya.horiguchi@nec.com>
Reviewed-by: Yang Shi <shy828301@gmail.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Mike Kravetz <mike.kravetz@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/ksm: use helper macro __ATTR_RW

Use helper macro __ATTR_RW to define KSM_ATTR to make code more clear.
Minor readability improvement.

Link: https://lkml.kernel.org/r/20220221115809.26381-1-linmiaohe@huawei.com
Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/vmstat: add event for ksm swapping in copy

When faults in from swap what used to be a KSM page and that page had been
swapped in before, system has to make a copy, and leaves remerging the
pages to a later pass of ksmd.

That is not good for performace, we'd better to reduce this kind of copy.
There are some ways to reduce it, for example lessen swappiness or
madvise(, , MADV_MERGEABLE) range.  So add this event to support doing
this tuning.  Just like this patch: "mm, THP, swap: add THP swapping out
fallback counting".

Link: https://lkml.kernel.org/r/20220113023839.758845-1-yang.yang29@zte.com.cn
Signed-off-by: Yang Yang <yang.yang29@zte.com.cn>
Reviewed-by: Ran Xiaokai <ran.xiaokai@zte.com.cn>
Cc: Hugh Dickins <hughd@google.com>
Cc: Yang Shi <yang.shi@linux.alibaba.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Saravanan D <saravanand@fb.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm: page_io: fix psi memory pressure error on cold swapins

Once upon a time, all swapins counted toward memory pressure[1].  Then
Joonsoo introduced workingset detection for anonymous pages and we gained
the ability to distinguish hot from cold swapins[2][3].  But we failed to
update swap_readpage() accordingly, and now we account partial memory
pressure in the swapin path of cold memory.

Not for all situations - which adds more inconsistency: paths using the
conventional submit_bio() and lock_page() route will not see much pressure
- unless storage itself is heavily congested and the bio submissions
stall.  ZRAM and ZSWAP do most of the work directly from swap_readpage()
and will see all swapins reflected as pressure.

IOW, a workload doing cold swapins could see little to no pressure
reported with on-disk swap, but potentially high pressure with a zram or
zswap backend.  That confuses any psi-based health monitoring, load
shedding, proactive reclaim, or userspace OOM killing schemes that might
be in place for the workload.

Restore consistency by making all swapin stall accounting conditional on
the page actually being part of the workingset.

[1] commit 937790699be9 ("mm/page_io.c: annotate refault stalls from swap_readpage")
[2] commit aae466b0052e ("mm/swap: implement workingset detection for anonymous LRU")
[3] commit cad8320b4b39 ("mm/swap: don't SetPageWorkingset unconditionally during swapin")

Link: https://lkml.kernel.org/r/20220214214921.419687-1-hannes@cmpxchg.org
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Reported-by: CGEL <cgel.zte@gmail.com>
Acked-by: Minchan Kim <minchan@kernel.org>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Yu Zhao <yuzhao@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

memory tiering: skip to scan fast memory

If the NUMA balancing isn't used to optimize the page placement among
sockets but only among memory types, the hot pages in the fast memory
node couldn't be migrated (promoted) to anywhere.  So it's unnecessary
to scan the pages in the fast memory node via changing their PTE/PMD
mapping to be PROT_NONE.  So that the page faults could be avoided too.

In the test, if only the memory tiering NUMA balancing mode is enabled,
the number of the NUMA balancing hint faults for the DRAM node is
reduced to almost 0 with the patch.  While the benchmark score doesn't
change visibly.

Link: https://lkml.kernel.org/r/20220221084529.1052339-4-ying.huang@intel.com
Signed-off-by: "Huang, Ying" <ying.huang@intel.com>
Suggested-by: Dave Hansen <dave.hansen@linux.intel.com>
Tested-by: Baolin Wang <baolin.wang@linux.alibaba.com>
Reviewed-by: Baolin Wang <baolin.wang@linux.alibaba.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Reviewed-by: Oscar Salvador <osalvador@suse.de>
Reviewed-by: Yang Shi <shy828301@gmail.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Rik van Riel <riel@surriel.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Zi Yan <ziy@nvidia.com>
Cc: Wei Xu <weixugc@google.com>
Cc: Shakeel Butt <shakeelb@google.com>
Cc: zhongjiang-ali <zhongjiang-ali@linux.alibaba.com>
Cc: Feng Tang <feng.tang@intel.com>
Cc: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

NUMA balancing: optimize page placement for memory tiering system

With the advent of various new memory types, some machines will have
multiple types of memory, e.g.  DRAM and PMEM (persistent memory).  The
memory subsystem of these machines can be called memory tiering system,
because the performance of the different types of memory are usually
different.

In such system, because of the memory accessing pattern changing etc,
some pages in the slow memory may become hot globally.  So in this
patch, the NUMA balancing mechanism is enhanced to optimize the page
placement among the different memory types according to hot/cold
dynamically.

In a typical memory tiering system, there are CPUs, fast memory and slow
memory in each physical NUMA node.  The CPUs and the fast memory will be
put in one logical node (called fast memory node), while the slow memory
will be put in another (faked) logical node (called slow memory node).
That is, the fast memory is regarded as local while the slow memory is
regarded as remote.  So it's possible for the recently accessed pages in
the slow memory node to be promoted to the fast memory node via the
existing NUMA balancing mechanism.

The original NUMA balancing mechanism will stop to migrate pages if the
free memory of the target node becomes below the high watermark.  This
is a reasonable policy if there's only one memory type.  But this makes
the original NUMA balancing mechanism almost do not work to optimize
page placement among different memory types.  Details are as follows.

It's the common cases that the working-set size of the workload is
larger than the size of the fast memory nodes.  Otherwise, it's
unnecessary to use the slow memory at all.  So, there are almost always
no enough free pages in the fast memory nodes, so that the globally hot
pages in the slow memory node cannot be promoted to the fast memory
node.  To solve the issue, we have 2 choices as follows,

a. Ignore the free pages watermark checking when promoting hot pages
   from the slow memory node to the fast memory node.  This will
   create some memory pressure in the fast memory node, thus trigger
   the memory reclaiming.  So that, the cold pages in the fast memory
   node will be demoted to the slow memory node.

b. Define a new watermark called wmark_promo which is higher than
   wmark_high, and have kswapd reclaiming pages until free pages reach
   such watermark.  The scenario is as follows: when we want to promote
   hot-pages from a slow memory to a fast memory, but fast memory's free
   pages would go lower than high watermark with such promotion, we wake
   up kswapd with wmark_promo watermark in order to demote cold pages and
   free us up some space.  So, next time we want to promote hot-pages we
   might have a chance of doing so.

The choice "a" may create high memory pressure in the fast memory node.
If the memory pressure of the workload is high, the memory pressure
may become so high that the memory allocation latency of the workload
is influenced, e.g.  the direct reclaiming may be triggered.

The choice "b" works much better at this aspect.  If the memory
pressure of the workload is high, the hot pages promotion will stop
earlier because its allocation watermark is higher than that of the
normal memory allocation.  So in this patch, choice "b" is implemented.
A new zone watermark (WMARK_PROMO) is added.  Which is larger than the
high watermark and can be controlled via watermark_scale_factor.

In addition to the original page placement optimization among sockets,
the NUMA balancing mechanism is extended to be used to optimize page
placement according to hot/cold among different memory types.  So the
sysctl user space interface (numa_balancing) is extended in a backward
compatible way as follow, so that the users can enable/disable these
functionality individually.

The sysctl is converted from a Boolean value to a bits field.  The
definition of the flags is,

- 0: NUMA_BALANCING_DISABLED
- 1: NUMA_BALANCING_NORMAL
- 2: NUMA_BALANCING_MEMORY_TIERING

We have tested the patch with the pmbench memory accessing benchmark
with the 80:20 read/write ratio and the Gauss access address
distribution on a 2 socket Intel server with Optane DC Persistent
Memory Model.  The test results shows that the pmbench score can
improve up to 95.9%.

Thanks Andrew Morton to help fix the document format error.

Link: https://lkml.kernel.org/r/20220221084529.1052339-3-ying.huang@intel.com
Signed-off-by: "Huang, Ying" <ying.huang@intel.com>
Tested-by: Baolin Wang <baolin.wang@linux.alibaba.com>
Reviewed-by: Baolin Wang <baolin.wang@linux.alibaba.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Reviewed-by: Oscar Salvador <osalvador@suse.de>
Reviewed-by: Yang Shi <shy828301@gmail.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Rik van Riel <riel@surriel.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Zi Yan <ziy@nvidia.com>
Cc: Wei Xu <weixugc@google.com>
Cc: Shakeel Butt <shakeelb@google.com>
Cc: zhongjiang-ali <zhongjiang-ali@linux.alibaba.com>
Cc: Randy Dunlap <rdunlap@infradead.org>
Cc: Feng Tang <feng.tang@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

NUMA Balancing: add page promotion counter

Patch series "NUMA balancing: optimize memory placement for memory tiering system", v13

With the advent of various new memory types, some machines will have
multiple types of memory, e.g.  DRAM and PMEM (persistent memory).  The
memory subsystem of these machines can be called memory tiering system,
because the performance of the different types of memory are different.

After commit c221c0b0308f ("device-dax: "Hotplug" persistent memory for
use like normal RAM"), the PMEM could be used as the cost-effective
volatile memory in separate NUMA nodes.  In a typical memory tiering
system, there are CPUs, DRAM and PMEM in each physical NUMA node.  The
CPUs and the DRAM will be put in one logical node, while the PMEM will
be put in another (faked) logical node.

To optimize the system overall performance, the hot pages should be
placed in DRAM node.  To do that, we need to identify the hot pages in
the PMEM node and migrate them to DRAM node via NUMA migration.

In the original NUMA balancing, there are already a set of existing
mechanisms to identify the pages recently accessed by the CPUs in a node
and migrate the pages to the node.  So we can reuse these mechanisms to
build the mechanisms to optimize the page placement in the memory
tiering system.  This is implemented in this patchset.

At the other hand, the cold pages should be placed in PMEM node.  So, we
also need to identify the cold pages in the DRAM node and migrate them
to PMEM node.

In commit 26aa2d199d6f ("mm/migrate: demote pages during reclaim"), a
mechanism to demote the cold DRAM pages to PMEM node under memory
pressure is implemented.  Based on that, the cold DRAM pages can be
demoted to PMEM node proactively to free some memory space on DRAM node
to accommodate the promoted hot PMEM pages.  This is implemented in this
patchset too.

We have tested the solution with the pmbench memory accessing benchmark
with the 80:20 read/write ratio and the Gauss access address
distribution on a 2 socket Intel server with Optane DC Persistent Memory
Model.  The test results shows that the pmbench score can improve up to
95.9%.

This patch (of 3):

In a system with multiple memory types, e.g.  DRAM and PMEM, the CPU
and DRAM in one socket will be put in one NUMA node as before, while
the PMEM will be put in another NUMA node as described in the
description of the commit c221c0b0308f ("device-dax: "Hotplug"
persistent memory for use like normal RAM").  So, the NUMA balancing
mechanism will identify all PMEM accesses as remote access and try to
promote the PMEM pages to DRAM.

To distinguish the number of the inter-type promoted pages from that of
the inter-socket migrated pages.  A new vmstat count is added.  The
counter is per-node (count in the target node).  So this can be used to
identify promotion imbalance among the NUMA nodes.

Link: https://lkml.kernel.org/r/20220301085329.3210428-1-ying.huang@intel.com
Link: https://lkml.kernel.org/r/20220221084529.1052339-1-ying.huang@intel.com
Link: https://lkml.kernel.org/r/20220221084529.1052339-2-ying.huang@intel.com
Signed-off-by: "Huang, Ying" <ying.huang@intel.com>
Reviewed-by: Yang Shi <shy828301@gmail.com>
Tested-by: Baolin Wang <baolin.wang@linux.alibaba.com>
Reviewed-by: Baolin Wang <baolin.wang@linux.alibaba.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Reviewed-by: Oscar Salvador <osalvador@suse.de>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Rik van Riel <riel@surriel.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Zi Yan <ziy@nvidia.com>
Cc: Wei Xu <weixugc@google.com>
Cc: Shakeel Butt <shakeelb@google.com>
Cc: zhongjiang-ali <zhongjiang-ali@linux.alibaba.com>
Cc: Feng Tang <feng.tang@intel.com>
Cc: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

powerpc/fadump: opt out from freeing pages on cma activation failure

With commit a4e92ce8e4c8 ("powerpc/fadump: Reservationless firmware
assisted dump"), Linux kernel's Contiguous Memory Allocator (CMA) based
reservation was introduced in fadump.  That change was aimed at using CMA
to let applications utilize the memory reserved for fadump while blocking
it from being used for kernel pages.  The assumption was, even if CMA
activation fails for whatever reason, the memory still remains reserved to
avoid it from being used for kernel pages.  But commit 072355c1cf2d
("mm/cma: expose all pages to the buddy if activation of an area fails")
breaks this assumption as it started exposing all pages to buddy allocator
on CMA activation failure.  It led to warning messages like below while
running crash-utility on vmcore of a kernel having above two commits:

  crash: seek error: kernel virtual address: <from reserved region>

To fix this problem, opt out from exposing pages to buddy allocator on CMA
activation failure for fadump reserved memory.

Link: https://lkml.kernel.org/r/20220117075246.36072-3-hbathini@linux.ibm.com
Signed-off-by: Hari Bathini <hbathini@linux.ibm.com>
Acked-by: David Hildenbrand <david@redhat.com>
Acked-by: Michael Ellerman <mpe@ellerman.id.au>
Cc: Mahesh Salgaonkar <mahesh@linux.ibm.com>
Cc: Mike Kravetz <mike.kravetz@oracle.com>
Cc: Oscar Salvador <osalvador@suse.de>
Cc: Sourabh Jain <sourabhjain@linux.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/cma: provide option to opt out from exposing pages on activation failure

Patch series "powerpc/fadump: handle CMA activation failure appropriately", v3.

Commit 072355c1cf2d ("mm/cma: expose all pages to the buddy if
activation of an area fails") started exposing all pages to buddy
allocator on CMA activation failure.  But there can be CMA users that
want to handle the reserved memory differently on CMA allocation
failure.

Provide an option to opt out from exposing pages to buddy for such
cases.

Link: https://lkml.kernel.org/r/20220117075246.36072-1-hbathini@linux.ibm.com
Link: https://lkml.kernel.org/r/20220117075246.36072-2-hbathini@linux.ibm.com
Signed-off-by: Hari Bathini <hbathini@linux.ibm.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Cc: Oscar Salvador <osalvador@suse.de>
Cc: Mike Kravetz <mike.kravetz@oracle.com>
Cc: Mahesh Salgaonkar <mahesh@linux.ibm.com>
Cc: Sourabh Jain <sourabhjain@linux.ibm.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/thp: refix __split_huge_pmd_locked() for migration PMD

Migration entries do not contribute to a page's reference count: move
__split_huge_pmd_locked()'s page_ref_add() into pmd_migration's else
block (along with the page_count() check - a page is quite likely to
have reference count frozen to 0 when a migration entry is found).

This will fix a very rare anonymous memory leak, after a
split_huge_pmd() raced with an anon split_huge_page() or an anon THP
migrate_pages(): since the wrongly raised refcount stopped the page
(perhaps small, perhaps huge, depending on when the race hit) from ever
being freed.

At first I thought there were worse risks, from prematurely unfreezing a
frozen page: but now think that would only affect page cache pages,
which do not come this way (except for anonymous pages in swap cache,
perhaps).

Link: https://lkml.kernel.org/r/84792468-f512-e48f-378c-e34c3641e97@google.com
Fixes: ec0abae6dcdf ("mm/thp: fix __split_huge_pmd_locked() for migration PMD")
Signed-off-by: Hugh Dickins <hughd@google.com>
Reviewed-by: Yang Shi <shy828301@gmail.com>
Cc: Ralph Campbell <rcampbell@nvidia.com>
Cc: Zi Yan <ziy@nvidia.com>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/migrate: fix race between lock page and clear PG_Isolated

When memory is tight, system may start to compact memory for large
continuous memory demands.  If one process tries to lock a memory page
that is being locked and isolated for compaction, it may wait a long time
or even forever.  This is because compaction will perform non-atomic
PG_Isolated clear while holding page lock, this may overwrite PG_waiters
set by the process that can't obtain the page lock and add itself to the
waiting queue to wait for the lock to be unlocked.

  CPU1                            CPU2
  lock_page(page); (successful)
                                  lock_page(); (failed)
  __ClearPageIsolated(page);      SetPageWaiters(page) (may be overwritten)
  unlock_page(page);

The solution is to not perform non-atomic operation on page flags while
holding page lock.

Link: https://lkml.kernel.org/r/20220315030515.20263-1-andrew.yang@mediatek.com
Signed-off-by: andrew.yang <andrew.yang@mediatek.com>
Cc: Matthias Brugger <matthias.bgg@gmail.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: "Vlastimil Babka" <vbabka@suse.cz>
Cc: David Howells <dhowells@redhat.com>
Cc: "William Kucharski" <william.kucharski@oracle.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Yang Shi <shy828301@gmail.com>
Cc: Marc Zyngier <maz@kernel.org>
Cc: Nicholas Tang <nicholas.tang@mediatek.com>
Cc: Kuan-Ying Lee <Kuan-Ying.Lee@mediatek.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm,migrate: fix establishing demotion target

In commit ac16ec835314 ("mm: migrate: support multiple target nodes
demotion"), after the first demotion target node is found, we will
continue to check the next candidate obtained via find_next_best_node().
This is to find all demotion target nodes with same NUMA distance.  But
one side effect of find_next_best_node() is that the candidate node
returned will be set in "used" parameter, even if the candidate node isn't
passed in the following NUMA distance checking, the candidate node will
not be used as demotion target node for the following nodes.  For example,
for system as follows,

node distances:
node   0   1   2   3
  0:  10  21  17  28
  1:  21  10  28  17
  2:  17  28  10  28
  3:  28  17  28  10

when we establish demotion target node for node 0, in the first round node
2 is added to the demotion target node set.  Then in the second round,
node 3 is checked and failed because distance(0, 3) > distance(0, 2).  But
node 3 is set in "used" nodemask too.  When we establish demotion target
node for node 1, there is no available node.  This is wrong, node 3 should
be set as the demotion target of node 1.

To fix this, if the candidate node is failed to pass the distance
checking, it will be cleared in "used" nodemask.  So that it can be used
for the following node.

The bug can be reproduced and fixed with this patch on a 2 socket server
machine with DRAM and PMEM.

Link: https://lkml.kernel.org/r/20220128055940.1792614-1-ying.huang@intel.com
Fixes: ac16ec835314 ("mm: migrate: support multiple target nodes demotion")
Signed-off-by: "Huang, Ying" <ying.huang@intel.com>
Reviewed-by: Baolin Wang <baolin.wang@linux.alibaba.com>
Cc: Baolin Wang <baolin.wang@linux.alibaba.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Zi Yan <ziy@nvidia.com>
Cc: Oscar Salvador <osalvador@suse.de>
Cc: Yang Shi <shy828301@gmail.com>
Cc: zhongjiang-ali <zhongjiang-ali@linux.alibaba.com>
Cc: Xunlei Pang <xlpang@linux.alibaba.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/oom_kill: remove unneeded is_memcg_oom check

oom_cpuset_eligible() is always called when !is_memcg_oom().  Remove this
unnecessary check.

Link: https://lkml.kernel.org/r/20220224115933.20154-1-linmiaohe@huawei.com
Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
Acked-by: David Rientjes <rientjes@google.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mempolicy: mbind_range() set_policy() after vma_merge()

v2.6.34 commit 9d8cebd4bcd7 ("mm: fix mbind vma merge problem") introduced
vma_merge() to mbind_range(); but unlike madvise, mlock and mprotect, it
put a "continue" to next vma where its precedents go to update flags on
current vma before advancing: that left vma with the wrong setting in the
infamous vma_merge() case 8.

v3.10 commit 1444f92c8498 ("mm: merging memory blocks resets mempolicy")
tried to fix that in vma_adjust(), without fully understanding the issue.

v3.11 commit 3964acd0dbec ("mm: mempolicy: fix mbind_range() &&
vma_adjust() interaction") reverted that, and went about the fix in the
right way, but chose to optimize out an unnecessary mpol_dup() with a
prior mpol_equal() test.  But on tmpfs, that also pessimized out the vital
call to its ->set_policy(), leaving the new mbind unenforced.

The user visible effect was that the pages got allocated on the local
node (happened to be 0), after the mbind() caller had specifically
asked for them to be allocated on node 1.  There was not any page
migration involved in the case reported: the pages simply got allocated
on the wrong node.

Just delete that optimization now (though it could be made conditional on
vma not having a set_policy).  Also remove the "next" variable: it turned
out to be blameless, but also pointless.

Link: https://lkml.kernel.org/r/319e4db9-64ae-4bca-92f0-ade85d342ff@google.com
Fixes: 3964acd0dbec ("mm: mempolicy: fix mbind_range() && vma_adjust() interaction")
Signed-off-by: Hugh Dickins <hughd@google.com>
Acked-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: Liam R. Howlett <Liam.Howlett@oracle.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm: compaction: cleanup the compaction trace events

As Steven suggested [1], we should access the pointers from the trace
event to avoid dereferencing them to the tracepoint function when the
tracepoint is disabled.

[1] https://lkml.org/lkml/2021/11/3/409

Link: https://lkml.kernel.org/r/4cd393b4d57f8f01ed72c001509b28e3a3b1a8c1.1646985115.git.baolin.wang@linux.alibaba.com
Signed-off-by: Baolin Wang <baolin.wang@linux.alibaba.com>
Cc: Steven Rostedt (Google) <rostedt@goodmis.org>
Cc: Ingo Molnar <mingo@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm: vmscan: fix documentation for page_check_references()

Commit b518154e59aa ("mm/vmscan: protect the workingset on anonymous
LRU") requires to look twice for both mapped anon/file pages are used
more than once to take the decission of reclaim or activation.  Correct
the documentation accordingly.

Link: https://lkml.kernel.org/r/1646925640-21324-1-git-send-email-quic_charante@quicinc.com
Signed-off-by: Charan Teja Kalla <quic_charante@quicinc.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm: workingset: replace IRQ-off check with a lockdep assert.

Commit 68d48e6a2df57 ("mm: workingset: add vmstat counter for shadow
nodes") introduced an IRQ-off check to ensure that a lock is held which
also disabled interrupts.  This does not work the same way on PREEMPT_RT
because none of the locks, that are held, disable interrupts.

Replace this check with a lockdep assert which ensures that the lock is
held.

Link: https://lkml.kernel.org/r/20220301122143.1521823-3-bigeasy@linutronix.de
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Tejun Heo <tj@kernel.org>
Cc: Zefan Li <lizefan.x@bytedance.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm: lru_cache_disable: replace work queue synchronization with synchronize_rcu

On systems that run FIFO:1 applications that busy loop, any SCHED_OTHER
task that attempts to execute on such a CPU (such as work threads) will
not be scheduled, which leads to system hangs.

Commit d479960e44f27e0e5 ("mm: disable LRU pagevec during the migration
temporarily") relies on queueing work items on all online CPUs to ensure
visibility of lru_disable_count.

To fix this, replace the usage of work items with synchronize_rcu,
which provides the same guarantees.

Readers of lru_disable_count are protected by either disabling
preemption or rcu_read_lock:

  preempt_disable, local_irq_disable  [bh_lru_lock()]
  rcu_read_lock                       [rt_spin_lock CONFIG_PREEMPT_RT]
  preempt_disable                     [local_lock !CONFIG_PREEMPT_RT]

Since v5.1 kernel, synchronize_rcu() is guaranteed to wait on
preempt_disable() regions of code.  So any CPU which sees
lru_disable_count = 0 will have exited the critical section when
synchronize_rcu() returns.

Link: https://lkml.kernel.org/r/Yin7hDxdt0s/x+fp@fuller.cnet
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Reviewed-by: Nicolas Saenz Julienne <nsaenzju@redhat.com>
Acked-by: Minchan Kim <minchan@kernel.org>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Cc: Paul E. McKenney <paulmck@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/list_lru: optimize memcg_reparent_list_lru_node()

Since commit 2c80cd57c743 ("mm/list_lru.c: fix list_lru_count_node() to
be race free"), we are tracking the total number of lru entries in a
list_lru_node in its nr_items field.

In the case of memcg_reparent_list_lru_node(), there is nothing to be
done if nr_items is 0.  We don't even need to take the nlru->lock as no
new lru entry could be added by a racing list_lru_add() to the draining
src_idx memcg at this point.

On systems that serve a lot of containers, it is possible that there can
be thousands of list_lru's present due to the fact that each container
may mount its own container specific filesystems.  As a typical
container uses only a few cpus, it is likely that only the list_lru_node
that contains those cpus will be utilized while the rests may be empty.
In other words, there can be a lot of list_lru_node with 0 nr_items.

By skipping a lock/unlock operation and loading a cacheline from
memcg_lrus, a sizeable number of cpu cycles can be saved.  That can be
substantial if we are talking about thousands of list_lru_node's with 0
nr_items.

Link: https://lkml.kernel.org/r/20220309144000.1470138-1-longman@redhat.com
Signed-off-by: Waiman Long <longman@redhat.com>
Reviewed-by: Roman Gushchin <roman.gushchin@linux.dev>
Cc: Muchun Song <songmuchun@bytedance.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Shakeel Butt <shakeelb@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm: __isolate_lru_page_prepare() in isolate_migratepages_block()

__isolate_lru_page_prepare() conflates two unrelated functions, with the
flags to one disjoint from the flags to the other; and hides some of the
important checks outside of isolate_migratepages_block(), where the
sequence is better to be visible.  It comes from the days of lumpy
reclaim, before compaction, when the combination made more sense.

Move what's needed by mm/compaction.c isolate_migratepages_block() inline
there, and what's needed by mm/vmscan.c isolate_lru_pages() inline there.

Shorten "isolate_mode" to "mode", so the sequence of conditions is easier
to read.  Declare a "mapping" variable, to save one call to page_mapping()
(but not another: calling again after page is locked is necessary).
Simplify isolate_lru_pages() with a "move_to" list pointer.

Link: https://lkml.kernel.org/r/879d62a8-91cc-d3c6-fb3b-69768236df68@google.com
Signed-off-by: Hugh Dickins <hughd@google.com>
Acked-by: David Rientjes <rientjes@google.com>
Reviewed-by: Alex Shi <alexs@kernel.org>
Cc: Alexander Duyck <alexander.duyck@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/fs: delete PF_SWAPWRITE

PF_SWAPWRITE has been redundant since v3.2 commit ee72886d8ed5 ("mm:
vmscan: do not writeback filesystem pages in direct reclaim").

Coincidentally, NeilBrown's current patch "remove inode_congested()"
deletes may_write_to_inode(), which appeared to be the one function which
took notice of PF_SWAPWRITE.  But if you study the old logic, and the
conditions under which may_write_to_inode() was called, you discover that
flag and function have been pointless for a decade.

Link: https://lkml.kernel.org/r/75e80e7-742d-e3bd-531-614db8961e4@google.com
Signed-off-by: Hugh Dickins <hughd@google.com>
Cc: NeilBrown <neilb@suse.de>
Cc: Jan Kara <jack@suse.de>
Cc: "Darrick J. Wong" <djwong@kernel.org>
Cc: Dave Chinner <david@fromorbit.com>
Cc: Matthew Wilcox <willy@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

userfaultfd/selftests: fix uninitialized_var.cocci warning

Fix following coccicheck warning:
tools/testing/selftests/vm/userfaultfd.c:556:23-24:
WARNING this kind of initialization is deprecated

`unsigned long page_nr = *(&page_nr)` has the same form of
uninitialized_var() macro. I remove the redundant assignement. It has
been tested with gcc (Debian 8.3.0-6) 8.3.0.

The patch which removed uninitialized_var() is:
https://lore.kernel.org/all/20121028102007.GA7547@gmail.com/ And there is
very few "/* GCC */" comments in the Linux kernel code now.

Link: https://lkml.kernel.org/r/20220304082333.9252-1-guozhengkui@vivo.com
Signed-off-by: Guo Zhengkui <guozhengkui@vivo.com>
Cc: Shuah Khan <shuah@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

userfaultfd: provide unmasked address on page-fault

Userfaultfd is supposed to provide the full address (i.e., unmasked) of
the faulting access back to userspace.  However, that is not the case for
quite some time.

Even running "userfaultfd_demo" from the userfaultfd man page provides the
wrong output (and contradicts the man page).  Notice that
"UFFD_EVENT_PAGEFAULT event" shows the masked address (7fc5e30b3000) and
not the first read address (0x7fc5e30b300f).

Address returned by mmap() = 0x7fc5e30b3000

fault_handler_thread():
    poll() returns: nready = 1; POLLIN = 1; POLLERR = 0
    UFFD_EVENT_PAGEFAULT event: flags = 0; address = 7fc5e30b3000
(uffdio_copy.copy returned 4096)
Read address 0x7fc5e30b300f in main(): A
Read address 0x7fc5e30b340f in main(): A
Read address 0x7fc5e30b380f in main(): A
Read address 0x7fc5e30b3c0f in main(): A

The exact address is useful for various reasons and specifically for
prefetching decisions.  If it is known that the memory is populated by
certain objects whose size is not page-aligned, then based on the faulting
address, the uffd-monitor can decide whether to prefetch and prefault the
adjacent page.

This bug has been for quite some time in the kernel: since commit
1a29d85eb0f1 ("mm: use vmf->address instead of of vmf->virtual_address")
vmf->virtual_address"), which dates back to 2016.  A concern has been
raised that existing userspace application might rely on the old/wrong
behavior in which the address is masked.  Therefore, it was suggested to
provide the masked address unless the user explicitly asks for the exact
address.

Add a new userfaultfd feature UFFD_FEATURE_EXACT_ADDRESS to direct
userfaultfd to provide the exact address.  Add a new "real_address" field
to vmf to hold the unmasked address.  Provide the address to userspace
accordingly.

Initialize real_address in various code-paths to be consistent with
address, even when it is not used, to be on the safe side.

[namit@vmware.com: initialize real_address on all code paths, per Jan]
Link: https://lkml.kernel.org/r/20220226022655.350562-1-namit@vmware.com
[akpm@linux-foundation.org: fix typo in comment, per Jan]

Link: https://lkml.kernel.org/r/20220218041003.3508-1-namit@vmware.com
Signed-off-by: Nadav Amit <namit@vmware.com>
Acked-by: Peter Xu <peterx@redhat.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Acked-by: Mike Rapoport <rppt@linux.ibm.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm: remove unneeded local variable follflags

We can pass FOLL_GET | FOLL_DUMP to follow_page directly to simplify the
code a bit in add_page_for_migration and split_huge_pages_pid.

Link: https://lkml.kernel.org/r/20220311072002.35575-1-linmiaohe@huawei.com
Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
Reviewed-by: Anshuman Khandual <anshuman.khandual@arm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/hugetlb.c: export PageHeadHuge()

Export PageHeadHuge() - it's used by folio_test_hugetlb() and thence by
such as folio_file_page() and folio_contains().  Matthew suggested I use
the first of those instead of doing the same calculation manually - but I
can't call it from a module.

Kirill suggested rearranging things to put it in a header, but that
introduces header dependencies because of where constants are defined.

[akpm@linux-foundation.org: s/EXPORT_SYMBOL/EXPORT_SYMBOL_GPL/, per Christoph]

Link: https://lkml.kernel.org/r/2494562.1646054576@warthog.procyon.org.uk
Link: https://lore.kernel.org/r/163707085314.3221130.14783857863702203440.stgit@warthog.procyon.org.uk/
Signed-off-by: David Howells <dhowells@redhat.com>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: Kirill A. Shutemov <kirill@shutemov.name>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Mike Kravetz <mike.kravetz@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/hugetlb: use helper macro __ATTR_RW

Use helper macro __ATTR_RW to define HSTATE_ATTR to make code more clear.
Minor readability improvement.

Link: https://lkml.kernel.org/r/20220222112731.33479-1-linmiaohe@huawei.com
Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
Reviewed-by: Mike Kravetz <mike.kravetz@oracle.com>
Reviewed-by: Muchun Song <songmuchun@bytedance.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

hugetlb: clean up potential spectre issue warnings

Recently introduced code allows numa nodes to be specified on the kernel
command line for hugetlb allocations or CMA reservations.  The node
values are user specified and used as indicies into arrays.  This
generated the following smatch warnings:

  mm/hugetlb.c:4170 hugepages_setup() warn: potential spectre issue 'default_hugepages_in_node' [w]
  mm/hugetlb.c:4172 hugepages_setup() warn: potential spectre issue 'parsed_hstate->max_huge_pages_node' [w]
  mm/hugetlb.c:6898 cmdline_parse_hugetlb_cma() warn: potential spectre issue 'hugetlb_cma_size_in_node' [w] (local cap)

Clean up by using array_index_nospec to sanitize array indicies.

The routine cmdline_parse_hugetlb_cma has the same overflow/truncation
issue addressed in [1].  That is also fixed with this change.

[1] https://lore.kernel.org/linux-mm/20220209134018.8242-1-liuyuntao10@huawei.com/

As Michal pointed out, this is unlikely to be exploitable because it is
__init code.  But the patch suppresses the warnings.

[mike.kravetz@oracle.com: v2]
Link: https://lkml.kernel.org/r/20220218212946.35441-1-mike.kravetz@oracle.com
Link: https://lkml.kernel.org/r/20220217234218.192885-1-mike.kravetz@oracle.com
Signed-off-by: Mike Kravetz <mike.kravetz@oracle.com>
Cc: Baolin Wang <baolin.wang@linux.alibaba.com>
Cc: Zhenguo Yao <yaozhenguo1@gmail.com>
Cc: Liu Yuntao <liuyuntao10@huawei.com>
Cc: Dan Carpenter <dan.carpenter@oracle.com>
Cc: Michal Hocko <mhocko@suse.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/hugetlb: generalize ARCH_WANT_GENERAL_HUGETLB

ARCH_WANT_GENERAL_HUGETLB config has duplicate definitions on platforms
that subscribe it.  Instead make it a generic config option which can be
selected on applicable platforms when required.

Link: https://lkml.kernel.org/r/1643718465-4324-1-git-send-email-anshuman.khandual@arm.com
Signed-off-by: Anshuman Khandual <anshuman.khandual@arm.com>
Cc: Russell King <linux@armlinux.org.uk>
Cc: Paul Walmsley <paul.walmsley@sifive.com>
Cc: Palmer Dabbelt <palmer@dabbelt.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Mike Kravetz <mike.kravetz@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm: sparsemem: move vmemmap related to HugeTLB to CONFIG_HUGETLB_PAGE_FREE_VMEMMAP

The vmemmap_remap_free/alloc are relevant to HugeTLB, so move those
functiongs to the scope of CONFIG_HUGETLB_PAGE_FREE_VMEMMAP.

Link: https://lkml.kernel.org/r/20211101031651.75851-6-songmuchun@bytedance.com
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
Reviewed-by: Barry Song <song.bao.hua@hisilicon.com>
Cc: Bodeddula Balasubramaniam <bodeddub@amazon.com>
Cc: Chen Huang <chenhuang5@huawei.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Fam Zheng <fam.zheng@bytedance.com>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Mike Kravetz <mike.kravetz@oracle.com>
Cc: Oscar Salvador <osalvador@suse.de>
Cc: Qi Zheng <zhengqi.arch@bytedance.com>
Cc: Xiongchun Duan <duanxiongchun@bytedance.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

selftests: vm: add a hugetlb test case

Since the head vmemmap page frame associated with each HugeTLB page is
reused, we should hide the PG_head flag of tail struct page from the
user.  Add a tese case to check whether it is work properly.  The test
steps are as follows.

  1) alloc 2MB hugeTLB
  2) get each page frame
  3) apply those APIs in each page frame
  4) Those APIs work completely the same as before.

Reading the flags of a page by /proc/kpageflags is done in
stable_page_flags(), which has invoked PageHead(), PageTail(),
PageCompound() and compound_head().

If those APIs work properly, the head page must have 15 and 17 bits set.
And tail pages must have 16 and 17 bits set but 15 bit unset.  Those
flags are checked in check_page_flags().

Link: https://lkml.kernel.org/r/20211101031651.75851-5-songmuchun@bytedance.com
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
Reviewed-by: Barry Song <song.bao.hua@hisilicon.com>
Cc: Bodeddula Balasubramaniam <bodeddub@amazon.com>
Cc: Chen Huang <chenhuang5@huawei.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Fam Zheng <fam.zheng@bytedance.com>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Mike Kravetz <mike.kravetz@oracle.com>
Cc: Oscar Salvador <osalvador@suse.de>
Cc: Qi Zheng <zhengqi.arch@bytedance.com>
Cc: Xiongchun Duan <duanxiongchun@bytedance.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm: sparsemem: use page table lock to protect kernel pmd operations

The init_mm.page_table_lock is used to protect kernel page tables, we
can use it to serialize splitting vmemmap PMD mappings instead of mmap
write lock, which can increase the concurrency of vmemmap_remap_free().

Actually, It increase the concurrency between allocations of HugeTLB
pages.  But it is not the only benefit.  There are a lot of users of
mmap read lock of init_mm.  The mmap write lock is holding through
vmemmap_remap_free(), removing mmap write lock usage to make it does not
affect other users of mmap read lock.  It is not making anything worse
and always a win to move.

Now the kernel page table walker does not hold the page_table_lock when
walking pmd entries.  There may be consistency issue of a pmd entry,
because pmd entry might change from a huge pmd entry to a PTE page
table.  There is only one user of kernel page table walker, namely
ptdump.  The ptdump already considers the consistency, which use a local
variable to cache the value of pmd entry.  But we also need to update
->action to ACTION_CONTINUE to make sure the walker does not walk every
pte entry again when concurrent thread has split the huge pmd.

Link: https://lkml.kernel.org/r/20211101031651.75851-4-songmuchun@bytedance.com
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
Cc: Barry Song <song.bao.hua@hisilicon.com>
Cc: Bodeddula Balasubramaniam <bodeddub@amazon.com>
Cc: Chen Huang <chenhuang5@huawei.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Fam Zheng <fam.zheng@bytedance.com>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Mike Kravetz <mike.kravetz@oracle.com>
Cc: Oscar Salvador <osalvador@suse.de>
Cc: Qi Zheng <zhengqi.arch@bytedance.com>
Cc: Xiongchun Duan <duanxiongchun@bytedance.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm: hugetlb: replace hugetlb_free_vmemmap_enabled with a static_key

The page_fixed_fake_head() is used throughout memory management and the
conditional check requires checking a global variable, although the
overhead of this check may be small, it increases when the memory cache
comes under pressure.  Also, the global variable will not be modified
after system boot, so it is very appropriate to use static key machanism.

Link: https://lkml.kernel.org/r/20211101031651.75851-3-songmuchun@bytedance.com
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
Reviewed-by: Barry Song <song.bao.hua@hisilicon.com>
Cc: Bodeddula Balasubramaniam <bodeddub@amazon.com>
Cc: Chen Huang <chenhuang5@huawei.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Fam Zheng <fam.zheng@bytedance.com>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Mike Kravetz <mike.kravetz@oracle.com>
Cc: Oscar Salvador <osalvador@suse.de>
Cc: Qi Zheng <zhengqi.arch@bytedance.com>
Cc: Xiongchun Duan <duanxiongchun@bytedance.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm: hugetlb: free the 2nd vmemmap page associated with each HugeTLB page

Patch series "Free the 2nd vmemmap page associated with each HugeTLB
page", v7.

This series can minimize the overhead of struct page for 2MB HugeTLB
pages significantly.  It further reduces the overhead of struct page by
12.5% for a 2MB HugeTLB compared to the previous approach, which means
2GB per 1TB HugeTLB.  It is a nice gain.  Comments and reviews are
welcome.  Thanks.

The main implementation and details can refer to the commit log of patch
1.  In this series, I have changed the following four helpers, the
following table shows the impact of the overhead of those helpers.

+------------------+-----------------------+
|       APIs       | head page | tail page |
+------------------+-----------+-----------+
|    PageHead()    |     Y     |     N     |
+------------------+-----------+-----------+
|    PageTail()    |     Y     |     N     |
+------------------+-----------+-----------+
|  PageCompound()  |     N     |     N     |
+------------------+-----------+-----------+
|  compound_head() |     Y     |     N     |
+------------------+-----------+-----------+

Y: Overhead is increased.
N: Overhead is _NOT_ increased.

It shows that the overhead of those helpers on a tail page don't change
between "hugetlb_free_vmemmap=on" and "hugetlb_free_vmemmap=off".  But the
overhead on a head page will be increased when "hugetlb_free_vmemmap=on"
(except PageCompound()).  So I believe that Matthew Wilcox's folio series
will help with this.

The users of PageHead() and PageTail() are much less than compound_head()
and most users of PageTail() are VM_BUG_ON(), so I have done some tests
about the overhead of compound_head() on head pages.

I have tested the overhead of calling compound_head() on a head page,
which is 2.11ns (Measure the call time of 10 million times
compound_head(), and then average).

For a head page whose address is not aligned with PAGE_SIZE or a
non-compound page, the overhead of compound_head() is 2.54ns which is
increased by 20%.  For a head page whose address is aligned with
PAGE_SIZE, the overhead of compound_head() is 2.97ns which is increased by
40%.  Most pages are the former.  I do not think the overhead is
significant since the overhead of compound_head() itself is low.

This patch (of 5):

This patch minimizes the overhead of struct page for 2MB HugeTLB pages
significantly.  It further reduces the overhead of struct page by 12.5%
for a 2MB HugeTLB compared to the previous approach, which means 2GB per
1TB HugeTLB (2MB type).

After the feature of "Free sonme vmemmap pages of HugeTLB page" is
enabled, the mapping of the vmemmap addresses associated with a 2MB
HugeTLB page becomes the figure below.

     HugeTLB                    struct pages(8 pages)         page frame(8 pages)
 +-----------+ ---virt_to_page---> +-----------+   mapping to   +-----------+---> PG_head
 |           |                     |     0     | -------------> |     0     |
 |           |                     +-----------+                +-----------+
 |           |                     |     1     | -------------> |     1     |
 |           |                     +-----------+                +-----------+
 |           |                     |     2     | ----------------^ ^ ^ ^ ^ ^
 |           |                     +-----------+                   | | | | |
 |           |                     |     3     | ------------------+ | | | |
 |           |                     +-----------+                     | | | |
 |           |                     |     4     | --------------------+ | | |
 |    2MB    |                     +-----------+                       | | |
 |           |                     |     5     | ----------------------+ | |
 |           |                     +-----------+                         | |
 |           |                     |     6     | ------------------------+ |
 |           |                     +-----------+                           |
 |           |                     |     7     | --------------------------+
 |           |                     +-----------+
 |           |
 |           |
 |           |
 +-----------+

As we can see, the 2nd vmemmap page frame (indexed by 1) is reused and
remaped. However, the 2nd vmemmap page frame is also can be freed to
the buddy allocator, then we can change the mapping from the figure
above to the figure below.

    HugeTLB                    struct pages(8 pages)         page frame(8 pages)
 +-----------+ ---virt_to_page---> +-----------+   mapping to   +-----------+---> PG_head
 |           |                     |     0     | -------------> |     0     |
 |           |                     +-----------+                +-----------+
 |           |                     |     1     | ---------------^ ^ ^ ^ ^ ^ ^
 |           |                     +-----------+                  | | | | | |
 |           |                     |     2     | -----------------+ | | | | |
 |           |                     +-----------+                    | | | | |
 |           |                     |     3     | -------------------+ | | | |
 |           |                     +-----------+                      | | | |
 |           |                     |     4     | ---------------------+ | | |
 |    2MB    |                     +-----------+                        | | |
 |           |                     |     5     | -----------------------+ | |
 |           |                     +-----------+                          | |
 |           |                     |     6     | -------------------------+ |
 |           |                     +-----------+                            |
 |           |                     |     7     | ---------------------------+
 |           |                     +-----------+
 |           |
 |           |
 |           |
 +-----------+

After we do this, all tail vmemmap pages (1-7) are mapped to the head
vmemmap page frame (0).  In other words, there are more than one page
struct with PG_head associated with each HugeTLB page.  We __know__ that
there is only one head page struct, the tail page structs with PG_head are
fake head page structs.  We need an approach to distinguish between those
two different types of page structs so that compound_head(), PageHead()
and PageTail() can work properly if the parameter is the tail page struct
but with PG_head.

The following code snippet describes how to distinguish between real and
fake head page struct.

if (test_bit(PG_head, &page->flags)) {
unsigned long head = READ_ONCE(page[1].compound_head);

if (head & 1) {
if (head == (unsigned long)page + 1)
==> head page struct
else
==> tail page struct
} else
==> head page struct
}

We can safely access the field of the @page[1] with PG_head because the
@page is a compound page composed with at least two contiguous pages.

[songmuchun@bytedance.com: restore lost comment changes]

Link: https://lkml.kernel.org/r/20211101031651.75851-1-songmuchun@bytedance.com
Link: https://lkml.kernel.org/r/20211101031651.75851-2-songmuchun@bytedance.com
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
Reviewed-by: Barry Song <song.bao.hua@hisilicon.com>
Cc: Mike Kravetz <mike.kravetz@oracle.com>
Cc: Oscar Salvador <osalvador@suse.de>
Cc: Michal Hocko <mhocko@suse.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Chen Huang <chenhuang5@huawei.com>
Cc: Bodeddula Balasubramaniam <bodeddub@amazon.com>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Xiongchun Duan <duanxiongchun@bytedance.com>
Cc: Fam Zheng <fam.zheng@bytedance.com>
Cc: Qi Zheng <zhengqi.arch@bytedance.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/mlock: fix potential imbalanced rlimit ucounts adjustment

user_shm_lock forgets to set allowed to 0 when get_ucounts fails.  So
the later user_shm_unlock might do the extra dec_rlimit_ucounts.  Fix
this by resetting allowed to 0.

Link: https://lkml.kernel.org/r/20220310132417.41189-1-linmiaohe@huawei.com
Fixes: d7c9e99aee48 ("Reimplement RLIMIT_MEMLOCK on top of ucounts")
Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Acked-by: Hugh Dickins <hughd@google.com>
Cc: Herbert van den Bergh <herbert.van.den.bergh@oracle.com>
Cc: Chris Mason <chris.mason@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm, fault-injection: declare should_fail_alloc_page()

The mm/ directory can almost fully be built with W=1, which would help
in local development.  One remaining issue is missing prototype for
should_fail_alloc_page().  Thus add it next to the should_failslab()
prototype.

Note the previous attempt by commit f7173090033c ("mm/page_alloc: make
should_fail_alloc_page() static") had to be reverted by commit
54aa386661fe as it caused an unresolved symbol error with
CONFIG_DEBUG_INFO_BTF=y

Link: https://lkml.kernel.org/r/20220314165724.16071-1-vbabka@suse.cz
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: David Hildenbrand <david@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/memory-failure.c: make non-LRU movable pages unhandlable

We can not really handle non-LRU movable pages in memory failure.
Typically they are balloon, zsmalloc, etc.

Assuming we run into a base (4K) non-LRU movable page, we could reach as
far as identify_page_state(), it should not fall into any category
except me_unknown.

For the non-LRU compound movable pages, they could be taken for
transhuge pages but it's unexpected to split non-LRU movable pages using
split_huge_page_to_list in memory_failure.  So we could just simply make
non-LRU movable pages unhandlable to avoid these possible nasty cases.

Link: https://lkml.kernel.org/r/20220312074613.4798-4-linmiaohe@huawei.com
Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
Suggested-by: Yang Shi <shy828301@gmail.com>
Reviewed-by: Yang Shi <shy828301@gmail.com>
Acked-by: Naoya Horiguchi <naoya.horiguchi@nec.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Mike Kravetz <mike.kravetz@oracle.com>
Cc: Tony Luck <tony.luck@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/memory-failure.c: avoid calling invalidate_inode_page() with unexpected pages

Since commit 042c4f32323b ("mm/truncate: Inline invalidate_complete_page()
into its one caller"), invalidate_inode_page() can invalidate the pages
in the swap cache because the check of page->mapping != mapping is
removed.  But invalidate_inode_page() is not expected to deal with the
pages in swap cache.  Also non-lru movable page can reach here too.
They're not page cache pages.  Skip these pages by checking
PageSwapCache and PageLRU.

Link: https://lkml.kernel.org/r/20220312074613.4798-3-linmiaohe@huawei.com
Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Mike Kravetz <mike.kravetz@oracle.com>
Cc: Naoya Horiguchi <naoya.horiguchi@nec.com>
Cc: Tony Luck <tony.luck@intel.com>
Cc: Yang Shi <shy828301@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/memory-failure.c: fix race with changing page compound again

Patch series "A few fixup patches for memory failure", v2.

This series contains a few patches to fix the race with changing page
compound page, make non-LRU movable pages unhandlable and so on.  More
details can be found in the respective changelogs.

There is a race window where we got the compound_head, the hugetlb page
could be freed to buddy, or even changed to another compound page just
before we try to get hwpoison page.  Think about the below race window:

  CPU 1   CPU 2
  memory_failure_hugetlb
  struct page *head = compound_head(p);
  hugetlb page might be freed to
  buddy, or even changed to another
  compound page.

  get_hwpoison_page -- page is not what we want now...

If this race happens, just bail out.  Also MF_MSG_DIFFERENT_PAGE_SIZE is
introduced to record this event.

[akpm@linux-foundation.org: s@/**@/*@, per Naoya Horiguchi]

Link: https://lkml.kernel.org/r/20220312074613.4798-1-linmiaohe@huawei.com
Link: https://lkml.kernel.org/r/20220312074613.4798-2-linmiaohe@huawei.com
Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
Acked-by: Naoya Horiguchi <naoya.horiguchi@nec.com>
Cc: Tony Luck <tony.luck@intel.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Mike Kravetz <mike.kravetz@oracle.com>
Cc: Yang Shi <shy828301@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/hwpoison: add in-use hugepage hwpoison filter judgement

After successfully obtaining the reference count of the huge page, it is
still necessary to call hwpoison_filter() to make a filter judgement,
otherwise the filter hugepage will be unmaped and the related process
may be killed.

Link: https://lkml.kernel.org/r/20220223082254.2769757-1-luofei@unicloud.com
Signed-off-by: luofei <luofei@unicloud.com>
Reviewed-by: Miaohe Lin <linmiaohe@huawei.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Naoya Horiguchi <naoya.horiguchi@nec.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Tony Luck <tony.luck@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/hwpoison: avoid the impact of hwpoison_filter() return value on mce handler

When the hwpoison page meets the filter conditions, it should not be
regarded as successful memory_failure() processing for mce handler, but
should return a distinct value, otherwise mce handler regards the error
page has been identified and isolated, which may lead to calling
set_mce_nospec() to change page attribute, etc.

Here memory_failure() return -EOPNOTSUPP to indicate that the error
event is filtered, mce handler should not take any action for this
situation and hwpoison injector should treat as correct.

Link: https://lkml.kernel.org/r/20220223082135.2769649-1-luofei@unicloud.com
Signed-off-by: luofei <luofei@unicloud.com>
Acked-by: Borislav Petkov <bp@suse.de>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Miaohe Lin <linmiaohe@huawei.com>
Cc: Naoya Horiguchi <naoya.horiguchi@nec.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Tony Luck <tony.luck@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/hwpoison-inject: support injecting hwpoison to free page

memory_failure() can handle free buddy page.  Support injecting hwpoison
to free page by adding is_free_buddy_page check when hwpoison filter is
disabled.

[akpm@linux-foundation.org: export is_free_buddy_page() to modules]

Link: https://lkml.kernel.org/r/20220218092052.3853-1-linmiaohe@huawei.com
Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
Cc: Naoya Horiguchi <naoya.horiguchi@nec.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/memory-failure.c: remove unnecessary PageTransTail check

When we reach here, we're guaranteed to have non-compound page as thp is
already splited.  Remove this unnecessary PageTransTail check.

Link: https://lkml.kernel.org/r/20220218090118.1105-9-linmiaohe@huawei.com
Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
Acked-by: Naoya Horiguchi <naoya.horiguchi@nec.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/memory-failure.c: remove obsolete comment in __soft_offline_page

Since commit add05cecef80 ("mm: soft-offline: don't free target page in
successful page migration"), set_migratetype_isolate logic is removed.
Remove this obsolete comment.

Link: https://lkml.kernel.org/r/20220218090118.1105-8-linmiaohe@huawei.com
Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/memory-failure.c: rework the try_to_unmap logic in hwpoison_user_mappings()

Only for hugetlb pages in shared mappings, try_to_unmap should take
semaphore in write mode here.  Rework the code to make it clear.

Link: https://lkml.kernel.org/r/20220218090118.1105-7-linmiaohe@huawei.com
Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
Acked-by: Naoya Horiguchi <naoya.horiguchi@nec.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/memory-failure.c: remove PageSlab check in hwpoison_filter_dev

Since commit 03e5ac2fc3bf ("mm: fix crash when using XFS on loopback"),
page_mapping() can handle the Slab pages.  So remove this unnecessary
PageSlab check and obsolete comment.

Link: https://lkml.kernel.org/r/20220218090118.1105-6-linmiaohe@huawei.com
Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
Acked-by: Naoya Horiguchi <naoya.horiguchi@nec.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/memory-failure.c: fix race with changing page more robustly

We're only intended to deal with the non-Compound page after we split
thp in memory_failure.  However, the page could have changed compound
pages due to race window.  If this happens, we could retry once to
hopefully handle the page next round.  Also remove unneeded orig_head.
It's always equal to the hpage.  So we can use hpage directly and remove
this redundant one.

Link: https://lkml.kernel.org/r/20220218090118.1105-5-linmiaohe@huawei.com
Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
Acked-by: Naoya Horiguchi <naoya.horiguchi@nec.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/memory-failure.c: rework the signaling logic in kill_proc

BUS_MCEERR_AR code is only sent when MF_ACTION_REQUIRED is set and the
target is current.  Rework the code to make this clear.

Link: https://lkml.kernel.org/r/20220218090118.1105-4-linmiaohe@huawei.com
Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
Acked-by: Naoya Horiguchi <naoya.horiguchi@nec.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/memory-failure.c: catch unexpected -EFAULT from vma_address()

It's unexpected to walk the page table when vma_address() return
-EFAULT.  But dev_pagemap_mapping_shift() is called only when vma
associated to the error page is found already in
collect_procs_{file,anon}, so vma_address() should not return -EFAULT
except with some bug, as Naoya pointed out.  We can use VM_BUG_ON_VMA()
to catch this bug here.

Link: https://lkml.kernel.org/r/20220218090118.1105-3-linmiaohe@huawei.com
Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
Acked-by: Naoya Horiguchi <naoya.horiguchi@nec.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/memory-failure.c: minor clean up for memory_failure_dev_pagemap

Patch series "A few cleanup and fixup patches for memory failure", v3.

This series contains a few patches to simplify the code logic, remove
unneeded variable and remove obsolete comment.  Also we fix race
changing page more robustly in memory_failure.  More details can be
found in the respective changelogs.

This patch (of 8):

The flags always has MF_ACTION_REQUIRED and MF_MUST_KILL set.  So we do
not need to check these flags again.

Link: https://lkml.kernel.org/r/20220218090118.1105-1-linmiaohe@huawei.com
Link: https://lkml.kernel.org/r/20220218090118.1105-2-linmiaohe@huawei.com
Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
Acked-by: Naoya Horiguchi <naoya.horiguchi@nec.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm: invalidate hwpoison page cache page in fault path

Sometimes the page offlining code can leave behind a hwpoisoned clean
page cache page.  This can lead to programs being killed over and over
and over again as they fault in the hwpoisoned page, get killed, and
then get re-spawned by whatever wanted to run them.

This is particularly embarrassing when the page was offlined due to
having too many corrected memory errors.  Now we are killing tasks due
to them trying to access memory that probably isn't even corrupted.

This problem can be avoided by invalidating the page from the page fault
handler, which already has a branch for dealing with these kinds of
pages.  With this patch we simply pretend the page fault was successful
if the page was invalidated, return to userspace, incur another page
fault, read in the file from disk (to a new memory page), and then
everything works again.

Link: https://lkml.kernel.org/r/20220212213740.423efcea@imladris.surriel.com
Signed-off-by: Rik van Riel <riel@surriel.com>
Reviewed-by: Miaohe Lin <linmiaohe@huawei.com>
Acked-by: Naoya Horiguchi <naoya.horiguchi@nec.com>
Reviewed-by: Oscar Salvador <osalvador@suse.de>
Cc: John Hubbard <jhubbard@nvidia.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/hwpoison: fix error page recovered but reported "not recovered"

When an uncorrected memory error is consumed there is a race between the
CMCI from the memory controller reporting an uncorrected error with a
UCNA signature, and the core reporting and SRAR signature machine check
when the data is about to be consumed.

If the CMCI wins that race, the page is marked poisoned when
uc_decode_notifier() calls memory_failure() and the machine check
processing code finds the page already poisoned.  It calls
kill_accessing_process() to make sure a SIGBUS is sent.  But returns the
wrong error code.

Console log looks like this:

  mce: Uncorrected hardware memory error in user-access at 3710b3400
  Memory failure: 0x3710b3: recovery action for dirty LRU page: Recovered
  Memory failure: 0x3710b3: already hardware poisoned
  Memory failure: 0x3710b3: Sending SIGBUS to einj_mem_uc:361438 due to hardware memory corruption
  mce: Memory error not recovered

kill_accessing_process() is supposed to return -EHWPOISON to notify that
SIGBUS is already set to the process and kill_me_maybe() doesn't have to
send it again.  But current code simply fails to do this, so fix it to
make sure to work as intended.  This change avoids the noise message
"Memory error not recovered" and skips duplicate SIGBUSs.

[tony.luck@intel.com: reword some parts of commit message]

Link: https://lkml.kernel.org/r/20220113231117.1021405-1-naoya.horiguchi@linux.dev
Fixes: a3f5d80ea401 ("mm,hwpoison: send SIGBUS with error virutal address")
Signed-off-by: Naoya Horiguchi <naoya.horiguchi@nec.com>
Reported-by: Youquan Song <youquan.song@intel.com>
Cc: Tony Luck <tony.luck@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/memory-failure.c: remove obsolete comment

With the introduction of mf_mutex, most of memory error handling process
is mutually exclusive, so the in-line comment about subtlety about
double-checking PageHWPoison is no more correct.  So remove it.

Link: https://lkml.kernel.org/r/20220125025601.3054511-1-naoya.horiguchi@linux.dev
Signed-off-by: Naoya Horiguchi <naoya.horiguchi@nec.com>
Suggested-by: Mike Kravetz <mike.kravetz@oracle.com>
Reviewed-by: Miaohe Lin <linmiaohe@huawei.com>
Reviewed-by: Anshuman Khandual <anshuman.khandual@arm.com>
Reviewed-by: Oscar Salvador <osalvador@suse.de>
Reviewed-by: Yang Shi <shy828301@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/page_alloc: check high-order pages for corruption during PCP operations

Eric Dumazet pointed out that commit 44042b449872 ("mm/page_alloc: allow
high-order pages to be stored on the per-cpu lists") only checks the
head page during PCP refill and allocation operations.  This was an
oversight and all pages should be checked.  This will incur a small
performance penalty but it's necessary for correctness.

Link: https://lkml.kernel.org/r/20220310092456.GJ15701@techsingularity.net
Fixes: 44042b449872 ("mm/page_alloc: allow high-order pages to be stored on the per-cpu lists")
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Reported-by: Eric Dumazet <edumazet@google.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Shakeel Butt <shakeelb@google.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Acked-by: David Rientjes <rientjes@google.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Wei Xu <weixugc@google.com>
Cc: Greg Thelen <gthelen@google.com>
Cc: Hugh Dickins <hughd@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/page_alloc: call check_new_pages() while zone spinlock is not held

For high order pages not using pcp, rmqueue() is currently calling the
costly check_new_pages() while zone spinlock is held, and hard irqs
masked.

This is not needed, we can release the spinlock sooner to reduce zone
spinlock contention.

Note that after this patch, we call __mod_zone_freepage_state() before
deciding to leak the page because it is in bad state.

Link: https://lkml.kernel.org/r/20220304170215.1868106-1-eric.dumazet@gmail.com
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Shakeel Butt <shakeelb@google.com>
Acked-by: David Rientjes <rientjes@google.com>
Acked-by: Mel Gorman <mgorman@techsingularity.net>
Reviewed-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Wei Xu <weixugc@google.com>
Cc: Greg Thelen <gthelen@google.com>
Cc: Hugh Dickins <hughd@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm: count time in drain_all_pages during direct reclaim as memory pressure

When page allocation in direct reclaim path fails, the system will make
one attempt to shrink per-cpu page lists and free pages from high alloc
reserves.  Draining per-cpu pages into buddy allocator can be a very
slow operation because it's done using workqueues and the task in direct
reclaim waits for all of them to finish before proceeding.  Currently
this time is not accounted as psi memory stall.

While testing mobile devices under extreme memory pressure, when
allocations are failing during direct reclaim, we notices that psi
events which would be expected in such conditions were not triggered.
After profiling these cases it was determined that the reason for
missing psi events was that a big chunk of time spent in direct reclaim
is not accounted as memory stall, therefore psi would not reach the
levels at which an event is generated.  Further investigation revealed
that the bulk of that unaccounted time was spent inside drain_all_pages
call.

A typical captured case when drain_all_pages path gets activated:

__alloc_pages_slowpath  took 44.644.613ns
    __perform_reclaim   took    751.668ns (1.7%)
    drain_all_pages     took 43.887.167ns (98.3%)

PSI in this case records the time spent in __perform_reclaim but ignores
drain_all_pages, IOW it misses 98.3% of the time spent in
__alloc_pages_slowpath.

Annotate __alloc_pages_direct_reclaim in its entirety so that delays
from handling page allocation failure in the direct reclaim path are
accounted as memory stall.

Link: https://lkml.kernel.org/r/20220223194812.1299646-1-surenb@google.com
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Reported-by: Tim Murray <timmurray@google.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Michal Hocko <mhocko@suse.com>
Reviewed-by: Shakeel Butt <shakeelb@google.com>
Cc: Petr Mladek <pmladek@suse.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Roman Gushchin <roman.gushchin@linux.dev>
Cc: Minchan Kim <minchan@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

arch/x86/mm/numa: Do not initialize nodes twice

On x86, prior to ("mm: handle uninitialized numa nodes gracecully"), NUMA
nodes could be allocated at three different places.

 - numa_register_memblks
 - init_cpu_to_node
 - init_gi_nodes

All these calls happen at setup_arch, and have the following order:

setup_arch
  ...
  x86_numa_init
   numa_init
    numa_register_memblks
  ...
  init_cpu_to_node
   init_memory_less_node
    alloc_node_data
    free_area_init_memoryless_node
  init_gi_nodes
   init_memory_less_node
    alloc_node_data
    free_area_init_memoryless_node

numa_register_memblks() is only interested in those nodes which have
memory, so it skips over any memoryless node it founds.  Later on, when
we have read ACPI's SRAT table, we call init_cpu_to_node() and
init_gi_nodes(), which initialize any memoryless node we might have that
have either CPU or Initiator affinity, meaning we allocate pg_data_t
struct for them and we mark them as ONLINE.

So far so good, but the thing is that after ("mm: handle uninitialized
numa nodes gracefully"), we allocate all possible NUMA nodes in
free_area_init(), meaning we have a picture like the following:

setup_arch
  x86_numa_init
   numa_init
    numa_register_memblks  <-- allocate non-memoryless node
  x86_init.paging.pagetable_init
   ...
    free_area_init
     free_area_init_memoryless <-- allocate memoryless node
  init_cpu_to_node
   alloc_node_data             <-- allocate memoryless node with CPU
   free_area_init_memoryless_node
  init_gi_nodes
   alloc_node_data             <-- allocate memoryless node with Initiator
   free_area_init_memoryless_node

free_area_init() already allocates all possible NUMA nodes, but
init_cpu_to_node() and init_gi_nodes() are clueless about that, so they
go ahead and allocate a new pg_data_t struct without checking anything,
meaning we end up allocating twice.

It should be mad clear that this only happens in the case where
memoryless NUMA node happens to have a CPU/Initiator affinity.

So get rid of init_memory_less_node() and just set the node online.

Note that setting the node online is needed, otherwise we choke down the
chain when bringup_nonboot_cpus() ends up calling
__try_online_node()->register_one_node()->...  and we blow up in
bus_add_device().  As can be seen here:

  BUG: kernel NULL pointer dereference, address: 0000000000000060
  #PF: supervisor read access in kernel mode
  #PF: error_code(0x0000) - not-present page
  PGD 0 P4D 0
  Oops: 0000 [#1] PREEMPT SMP DEBUG_PAGEALLOC PTI
  CPU: 0 PID: 1 Comm: swapper/0 Not tainted 5.17.0-rc4-1-default+ #45
  Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.0.0-prebuilt.qemu-project.org 04/4
  RIP: 0010:bus_add_device+0x5a/0x140
  Code: 8b 74 24 20 48 89 df e8 84 96 ff ff 85 c0 89 c5 75 38 48 8b 53 50 48 85 d2 0f 84 bb 00 004
  RSP: 0000:ffffc9000022bd10 EFLAGS: 00010246
  RAX: 0000000000000000 RBX: ffff888100987400 RCX: ffff8881003e4e19
  RDX: ffff8881009a5e00 RSI: ffff888100987400 RDI: ffff888100987400
  RBP: 0000000000000000 R08: ffff8881003e4e18 R09: ffff8881003e4c98
  R10: 0000000000000000 R11: ffff888100402bc0 R12: ffffffff822ceba0
  R13: 0000000000000000 R14: ffff888100987400 R15: 0000000000000000
  FS:  0000000000000000(0000) GS:ffff88853fc00000(0000) knlGS:0000000000000000
  CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
  CR2: 0000000000000060 CR3: 000000000200a001 CR4: 00000000001706b0
  Call Trace:
   device_add+0x4c0/0x910
   __register_one_node+0x97/0x2d0
   __try_online_node+0x85/0xc0
   try_online_node+0x25/0x40
   cpu_up+0x4f/0x100
   bringup_nonboot_cpus+0x4f/0x60
   smp_init+0x26/0x79
   kernel_init_freeable+0x130/0x2f1
   kernel_init+0x17/0x150
   ret_from_fork+0x22/0x30

The reason is simple, by the time bringup_nonboot_cpus() gets called, we
did not register the node_subsys bus yet, so we crash when
bus_add_device() tries to dereference bus()->p.

The following shows the order of the calls:

kernel_init_freeable
 smp_init
  bringup_nonboot_cpus
   ...
     bus_add_device()      <- we did not register node_subsys yet
 do_basic_setup
  do_initcalls
   postcore_initcall(register_node_type);
    register_node_type
     subsys_system_register
      subsys_register
       bus_register         <- register node_subsys bus

Why setting the node online saves us then? Well, simply because
__try_online_node() backs off when the node is online, meaning we do not
end up calling register_one_node() in the first place.

This is subtle, broken and deserves a deep analysis and thought about
how to put this into shape, but for now let us have this easy fix for
the leaking memory issue.

[osalvador@suse.de: add comments]
Link: https://lkml.kernel.org/r/20220221142649.3457-1-osalvador@suse.de
Link: https://lkml.kernel.org/r/20220218224302.5282-2-osalvador@suse.de
Fixes: da4490c958ad ("mm: handle uninitialized numa nodes gracefully")
Signed-off-by: Oscar Salvador <osalvador@suse.de>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Rafael Aquini <raquini@redhat.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Wei Yang <richard.weiyang@gmail.com>
Cc: Dennis Zhou <dennis@kernel.org>
Cc: Alexey Makhalov <amakhalov@vmware.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/page_alloc: do not prefetch buddies during bulk free

free_pcppages_bulk() has taken two passes through the pcp lists since
commit 0a5f4e5b4562 ("mm/free_pcppages_bulk: do not hold lock when
picking pages to free") due to deferring the cost of selecting PCP lists
until the zone lock is held.

As the list processing now takes place under the zone lock, it's less
clear that this will always benefit for two reasons.

1. There is a guaranteed cost to calculating the buddy which definitely
   has to be calculated again. However, as the zone lock is held and
   there is no deferring of buddy merging, there is no guarantee that the
   prefetch will have completed when the second buddy calculation takes
   place and buddies are being merged.  With or without the prefetch, there
   may be further stalls depending on how many pages get merged. In other
   words, a stall due to merging is inevitable and at best only one stall
   might be avoided at the cost of calculating the buddy location twice.

2. As the zone lock is held, prefetch_nr makes less sense as once
   prefetch_nr expires, the cache lines of interest have already been
   merged.

The main concern is that there is a definite cost to calculating the
buddy location early for the prefetch and it is a "maybe win" depending
on whether the CPU prefetch logic and memory is fast enough.  Remove the
prefetch logic on the basis that reduced instructions in a path is
always a saving where as the prefetch might save one memory stall
depending on the CPU and memory.

In most cases, this has marginal benefit as the calculations are a small
part of the overall freeing of pages.  However, it was detectable on at
least one machine.

                              5.17.0-rc3             5.17.0-rc3
                    mm-highpcplimit-v2r1     mm-noprefetch-v1r1
Min       elapsed      630.00 (   0.00%)      610.00 (   3.17%)
Amean     elapsed      639.00 (   0.00%)      623.00 *   2.50%*
Max       elapsed      660.00 (   0.00%)      660.00 (   0.00%)

Link: https://lkml.kernel.org/r/20220221094119.15282-2-mgorman@techsingularity.net
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Suggested-by: Aaron Lu <aaron.lu@intel.com>
Reviewed-by: Vlastimil Babka <vbabka@suse.cz>
Reviewed-by: Aaron Lu <aaron.lu@intel.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/page_alloc: limit number of high-order pages on PCP during bulk free

When a PCP is mostly used for frees then high-order pages can exist on
PCP lists for some time.  This is problematic when the allocation
pattern is all allocations from one CPU and all frees from another
resulting in colder pages being used.  When bulk freeing pages, limit
the number of high-order pages that are stored on the PCP lists.

Netperf running on localhost exhibits this pattern and while it does not
matter for some machines, it does matter for others with smaller caches
where cache misses cause problems due to reduced page reuse.  Pages
freed directly to the buddy list may be reused quickly while still cache
hot where as storing on the PCP lists may be cold by the time
free_pcppages_bulk() is called.

Using perf kmem:mm_page_alloc, the 5 most used page frames were

5.17-rc3
  13041 pfn=0x111a30
  13081 pfn=0x5814d0
  13097 pfn=0x108258
  13121 pfn=0x689598
  13128 pfn=0x5814d8

5.17-revert-highpcp
 192009 pfn=0x54c140
 195426 pfn=0x1081d0
 200908 pfn=0x61c808
 243515 pfn=0xa9dc20
 402523 pfn=0x222bb8

5.17-full-series
 142693 pfn=0x346208
 162227 pfn=0x13bf08
 166413 pfn=0x2711e0
 166950 pfn=0x2702f8

The spread is wider as there is still time before pages freed to one PCP
get released with a tradeoff between fast reuse and reduced zone lock
acquisition.

On the machine used to gather the traces, the headline performance was
equivalent.

netperf-tcp
                            5.17.0-rc3             5.17.0-rc3             5.17.0-rc3
                               vanilla  mm-reverthighpcp-v1r1     mm-highpcplimit-v2
Hmean     64         839.93 (   0.00%)      840.77 (   0.10%)      841.02 (   0.13%)
Hmean     128       1614.22 (   0.00%)     1622.07 *   0.49%*     1636.41 *   1.37%*
Hmean     256       2952.00 (   0.00%)     2953.19 (   0.04%)     2977.76 *   0.87%*
Hmean     1024     10291.67 (   0.00%)    10239.17 (  -0.51%)    10434.41 *   1.39%*
Hmean     2048     17335.08 (   0.00%)    17399.97 (   0.37%)    17134.81 *  -1.16%*
Hmean     3312     22628.15 (   0.00%)    22471.97 (  -0.69%)    22422.78 (  -0.91%)
Hmean     4096     25009.50 (   0.00%)    24752.83 *  -1.03%*    24740.41 (  -1.08%)
Hmean     8192     32745.01 (   0.00%)    31682.63 *  -3.24%*    32153.50 *  -1.81%*
Hmean     16384    39759.59 (   0.00%)    36805.78 *  -7.43%*    38948.13 *  -2.04%*

On a 1-socket skylake machine with a small CPU cache that suffers more if
cache misses are too high

netperf-tcp
                            5.17.0-rc3             5.17.0-rc3             5.17.0-rc3
                               vanilla    mm-reverthighpcp-v1     mm-highpcplimit-v2
Hmean     64         938.95 (   0.00%)      941.50 *   0.27%*      943.61 *   0.50%*
Hmean     128       1843.10 (   0.00%)     1857.58 *   0.79%*     1861.09 *   0.98%*
Hmean     256       3573.07 (   0.00%)     3667.45 *   2.64%*     3674.91 *   2.85%*
Hmean     1024     13206.52 (   0.00%)    13487.80 *   2.13%*    13393.21 *   1.41%*
Hmean     2048     22870.23 (   0.00%)    23337.96 *   2.05%*    23188.41 *   1.39%*
Hmean     3312     31001.99 (   0.00%)    32206.50 *   3.89%*    31863.62 *   2.78%*
Hmean     4096     35364.59 (   0.00%)    36490.96 *   3.19%*    36112.54 *   2.11%*
Hmean     8192     48497.71 (   0.00%)    49954.05 *   3.00%*    49588.26 *   2.25%*
Hmean     16384    58410.86 (   0.00%)    60839.80 *   4.16%*    62282.96 *   6.63%*

Note that this was a machine that did not benefit from caching high-order
pages and performance is almost restored with the series applied.  It's
not fully restored as cache misses are still higher.  This is a trade-off
between optimising for a workload that does all allocs on one CPU and
frees on another or more general workloads that need high-order pages for
SLUB and benefit from avoiding zone->lock for every SLUB refill/drain.

Link: https://lkml.kernel.org/r/20220217002227.5739-7-mgorman@techsingularity.net
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Reviewed-by: Vlastimil Babka <vbabka@suse.cz>
Tested-by: Aaron Lu <aaron.lu@intel.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Jesper Dangaard Brouer <brouer@redhat.com>
Cc: Michal Hocko <mhocko@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/page_alloc: free pages in a single pass during bulk free

free_pcppages_bulk() has taken two passes through the pcp lists since
commit 0a5f4e5b4562 ("mm/free_pcppages_bulk: do not hold lock when
picking pages to free") due to deferring the cost of selecting PCP lists
until the zone lock is held.  Now that list selection is simplier, the
main cost during selection is bulkfree_pcp_prepare() which in the normal
case is a simple check and prefetching.  As the list manipulations have
cost in itself, go back to freeing pages in a single pass.

The series up to this point was evaulated using a trunc microbenchmark
that is truncating sparse files stored in page cache (mmtests config
config-io-trunc).  Sparse files were used to limit filesystem
interaction.  The results versus a revert of storing high-order pages in
the PCP lists is

1-socket Skylake
                               5.17.0-rc3             5.17.0-rc3             5.17.0-rc3
                                  vanilla      mm-reverthighpcp-v1     mm-highpcpopt-v2
 Min       elapsed      540.00 (   0.00%)      530.00 (   1.85%)      530.00 (   1.85%)
 Amean     elapsed      543.00 (   0.00%)      530.00 *   2.39%*      530.00 *   2.39%*
 Stddev    elapsed        4.83 (   0.00%)        0.00 ( 100.00%)        0.00 ( 100.00%)
 CoeffVar  elapsed        0.89 (   0.00%)        0.00 ( 100.00%)        0.00 ( 100.00%)
 Max       elapsed      550.00 (   0.00%)      530.00 (   3.64%)      530.00 (   3.64%)
 BAmean-50 elapsed      540.00 (   0.00%)      530.00 (   1.85%)      530.00 (   1.85%)
 BAmean-95 elapsed      542.22 (   0.00%)      530.00 (   2.25%)      530.00 (   2.25%)
 BAmean-99 elapsed      542.22 (   0.00%)      530.00 (   2.25%)      530.00 (   2.25%)

2-socket CascadeLake
                               5.17.0-rc3             5.17.0-rc3             5.17.0-rc3
                                  vanilla    mm-reverthighpcp-v1       mm-highpcpopt-v2
 Min       elapsed      510.00 (   0.00%)      500.00 (   1.96%)      500.00 (   1.96%)
 Amean     elapsed      529.00 (   0.00%)      521.00 (   1.51%)      510.00 *   3.59%*
 Stddev    elapsed       16.63 (   0.00%)       12.87 (  22.64%)       11.55 (  30.58%)
 CoeffVar  elapsed        3.14 (   0.00%)        2.47 (  21.46%)        2.26 (  27.99%)
 Max       elapsed      550.00 (   0.00%)      540.00 (   1.82%)      530.00 (   3.64%)
 BAmean-50 elapsed      516.00 (   0.00%)      512.00 (   0.78%)      500.00 (   3.10%)
 BAmean-95 elapsed      526.67 (   0.00%)      518.89 (   1.48%)      507.78 (   3.59%)
 BAmean-99 elapsed      526.67 (   0.00%)      518.89 (   1.48%)      507.78 (   3.59%)

The original motivation for multi-passes was will-it-scale page_fault1
using $nr_cpu processes.

2-socket CascadeLake (40 cores, 80 CPUs HT enabled)
                                                     5.17.0-rc3                 5.17.0-rc3
                                                        vanilla           mm-highpcpopt-v2
 Hmean     page_fault1-processes-2        2694662.26 (   0.00%)      2695780.35 (   0.04%)
 Hmean     page_fault1-processes-5        6425819.34 (   0.00%)      6435544.57 *   0.15%*
 Hmean     page_fault1-processes-8        9642169.10 (   0.00%)      9658962.39 (   0.17%)
 Hmean     page_fault1-processes-12      12167502.10 (   0.00%)     12190163.79 (   0.19%)
 Hmean     page_fault1-processes-21      15636859.03 (   0.00%)     15612447.26 (  -0.16%)
 Hmean     page_fault1-processes-30      25157348.61 (   0.00%)     25169456.65 (   0.05%)
 Hmean     page_fault1-processes-48      27694013.85 (   0.00%)     27671111.46 (  -0.08%)
 Hmean     page_fault1-processes-79      25928742.64 (   0.00%)     25934202.02 (   0.02%) <--
 Hmean     page_fault1-processes-110     25730869.75 (   0.00%)     25671880.65 *  -0.23%*
 Hmean     page_fault1-processes-141     25626992.42 (   0.00%)     25629551.61 (   0.01%)
 Hmean     page_fault1-processes-172     25611651.35 (   0.00%)     25614927.99 (   0.01%)
 Hmean     page_fault1-processes-203     25577298.75 (   0.00%)     25583445.59 (   0.02%)
 Hmean     page_fault1-processes-234     25580686.07 (   0.00%)     25608240.71 (   0.11%)
 Hmean     page_fault1-processes-265     25570215.47 (   0.00%)     25568647.58 (  -0.01%)
 Hmean     page_fault1-processes-296     25549488.62 (   0.00%)     25543935.00 (  -0.02%)
 Hmean     page_fault1-processes-320     25555149.05 (   0.00%)     25575696.74 (   0.08%)

The differences are mostly within the noise and the difference close to
$nr_cpus is negligible.

Link: https://lkml.kernel.org/r/20220217002227.5739-6-mgorman@techsingularity.net
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Reviewed-by: Vlastimil Babka <vbabka@suse.cz>
Tested-by: Aaron Lu <aaron.lu@intel.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Jesper Dangaard Brouer <brouer@redhat.com>
Cc: Michal Hocko <mhocko@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/page_alloc: drain the requested list first during bulk free

Prior to the series, pindex 0 (order-0 MIGRATE_UNMOVABLE) was always
skipped first and the precise reason is forgotten.  A potential reason
may have been to artificially preserve MIGRATE_UNMOVABLE but there is no
reason why that would be optimal as it depends on the workload.  The
more likely reason is that it was less complicated to do a pre-increment
instead of a post-increment in terms of overall code flow.  As
free_pcppages_bulk() now typically receives the pindex of the PCP list
that exceeded high, always start draining that list.

Link: https://lkml.kernel.org/r/20220217002227.5739-5-mgorman@techsingularity.net
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Reviewed-by: Vlastimil Babka <vbabka@suse.cz>
Tested-by: Aaron Lu <aaron.lu@intel.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Jesper Dangaard Brouer <brouer@redhat.com>
Cc: Michal Hocko <mhocko@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/page_alloc: simplify how many pages are selected per pcp list during bulk free

free_pcppages_bulk() selects pages to free by round-robining between
lists.  Originally this was to evenly shrink pages by migratetype but
uneven freeing is inevitable due to high pages.  Simplify list selection
by starting with a list that definitely has pages on it in
free_unref_page_commit() and for drain, it does not matter where
draining starts as all pages are removed.

Link: https://lkml.kernel.org/r/20220217002227.5739-4-mgorman@techsingularity.net
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Reviewed-by: Vlastimil Babka <vbabka@suse.cz>
Tested-by: Aaron Lu <aaron.lu@intel.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Jesper Dangaard Brouer <brouer@redhat.com>
Cc: Michal Hocko <mhocko@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>