From Roman ("percpu: partial chunk depopulation"):
In our [Facebook] production experience the percpu memory allocator is
sometimes struggling with returning the memory to the system. A typical
example is a creation of several thousands memory cgroups (each has
several chunks of the percpu data used for vmstats, vmevents,
ref counters etc). Deletion and complete releasing of these cgroups
doesn't always lead to a shrinkage of the percpu memory, so that
sometimes there are several GB's of memory wasted.
The underlying problem is the fragmentation: to release an underlying
chunk all percpu allocations should be released first. The percpu
allocator tends to top up chunks to improve the utilization. It means
new small-ish allocations (e.g. percpu ref counters) are placed onto
almost filled old-ish chunks, effectively pinning them in memory.
This patchset solves this problem by implementing a partial depopulation
of percpu chunks: chunks with many empty pages are being asynchronously
depopulated and the pages are returned to the system.
To illustrate the problem the following script can be used:
--
cd /sys/fs/cgroup
mkdir percpu_test
echo "+memory" > percpu_test/cgroup.subtree_control
cat /proc/meminfo | grep Percpu
for i in `seq 1 1000`; do
mkdir percpu_test/cg_"${i}"
for j in `seq 1 10`; do
mkdir percpu_test/cg_"${i}"_"${j}"
done
done
cat /proc/meminfo | grep Percpu
for i in `seq 1 1000`; do
for j in `seq 1 10`; do
rmdir percpu_test/cg_"${i}"_"${j}"
done
done
sleep 10
cat /proc/meminfo | grep Percpu
for i in `seq 1 1000`; do
rmdir percpu_test/cg_"${i}"
done
rmdir percpu_test
--
It creates 11000 memory cgroups and removes every 10 out of 11.
It prints the initial size of the percpu memory, the size after
creating all cgroups and the size after deleting most of them.
Results:
vanilla:
./percpu_test.sh
Percpu: 7488 kB
Percpu: 481152 kB
Percpu: 481152 kB
with this patchset applied:
./percpu_test.sh
Percpu: 7488 kB
Percpu: 481408 kB
Percpu: 135552 kB
The total size of the percpu memory was reduced by more than 3.5 times.
This patch:
This patch implements partial depopulation of percpu chunks.
As of now, a chunk can be depopulated only as a part of the final
destruction, if there are no more outstanding allocations. However
to minimize a memory waste it might be useful to depopulate a
partially filed chunk, if a small number of outstanding allocations
prevents the chunk from being fully reclaimed.
This patch implements the following depopulation process: it scans
over the chunk pages, looks for a range of empty and populated pages
and performs the depopulation. To avoid races with new allocations,
the chunk is previously isolated. After the depopulation the chunk is
sidelined to a special list or freed. New allocations prefer using
active chunks to sidelined chunks. If a sidelined chunk is used, it is
reintegrated to the active lists.
The depopulation is scheduled on the free path if the chunk is all of
the following:
1) has more than 1/4 of total pages free and populated
2) the system has enough free percpu pages aside of this chunk
3) isn't the reserved chunk
4) isn't the first chunk
If it's already depopulated but got free populated pages, it's a good
target too. The chunk is moved to a special slot,
pcpu_to_depopulate_slot, chunk->isolated is set, and the balance work
item is scheduled. On isolation, these pages are removed from the
pcpu_nr_empty_pop_pages. It is constantly replaced to the
to_depopulate_slot when it meets these qualifications.
pcpu_reclaim_populated() iterates over the to_depopulate_slot until it
becomes empty. The depopulation is performed in the reverse direction to
keep populated pages close to the beginning. Depopulated chunks are
sidelined to preferentially avoid them for new allocations. When no
active chunk can suffice a new allocation, sidelined chunks are first
checked before creating a new chunk.
Signed-off-by: Roman Gushchin <guro@fb.com>
Co-developed-by: Dennis Zhou <dennis@kernel.org>
Signed-off-by: Dennis Zhou <dennis@kernel.org>
Tested-by: Pratik Sampat <psampat@linux.ibm.com>
Signed-off-by: Dennis Zhou <dennis@kernel.org>
This prepares for adding a to_depopulate list and sidelined list after
the free slot in the set of lists in pcpu_slot.
Signed-off-by: Dennis Zhou <dennis@kernel.org>
Acked-by: Roman Gushchin <guro@fb.com>
Signed-off-by: Dennis Zhou <dennis@kernel.org>
Factor out the pcpu_check_block_hint() helper, which will be useful
in the future. The new function checks if the allocation can likely
fit within the contig hint.
Signed-off-by: Roman Gushchin <guro@fb.com>
Signed-off-by: Dennis Zhou <dennis@kernel.org>
Signed-off-by: Dennis Zhou <dennis@kernel.org>
Since kernel v5.1, fanotify_init(2) supports the flag FAN_REPORT_FID
for identifying objects using file handle and fsid in events.
fanotify_mark(2) fails with -ENODEV when trying to set a mark on
filesystems that report null f_fsid in stasfs(2).
Use the digest of uuid as f_fsid for tmpfs to uniquely identify tmpfs
objects as best as possible and allow setting an fanotify mark that
reports events with file handles on tmpfs.
Link: https://lore.kernel.org/r/20210322173944.449469-3-amir73il@gmail.com
Acked-by: Hugh Dickins <hughd@google.com>
Reviewed-by: Christian Brauner <christian.brauner@ubuntu.com>
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
READ_ONCE() cannot be used for reading PTEs. Use ptep_get() instead, to
avoid the following errors:
CC mm/ptdump.o
In file included from <command-line>:
mm/ptdump.c: In function 'ptdump_pte_entry':
include/linux/compiler_types.h:320:38: error: call to '__compiletime_assert_207' declared with attribute error: Unsupported access size for {READ,WRITE}_ONCE().
320 | _compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)
| ^
include/linux/compiler_types.h:301:4: note: in definition of macro '__compiletime_assert'
301 | prefix ## suffix(); \
| ^~~~~~
include/linux/compiler_types.h:320:2: note: in expansion of macro '_compiletime_assert'
320 | _compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)
| ^~~~~~~~~~~~~~~~~~~
include/asm-generic/rwonce.h:36:2: note: in expansion of macro 'compiletime_assert'
36 | compiletime_assert(__native_word(t) || sizeof(t) == sizeof(long long), \
| ^~~~~~~~~~~~~~~~~~
include/asm-generic/rwonce.h:49:2: note: in expansion of macro 'compiletime_assert_rwonce_type'
49 | compiletime_assert_rwonce_type(x); \
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
mm/ptdump.c:114:14: note: in expansion of macro 'READ_ONCE'
114 | pte_t val = READ_ONCE(*pte);
| ^~~~~~~~~
make[2]: *** [mm/ptdump.o] Error 1
See commit 481e980a7c ("mm: Allow arches to provide ptep_get()") and
commit c0e1c8c22b ("powerpc/8xx: Provide ptep_get() with 16k pages")
for details.
Link: https://lkml.kernel.org/r/912b349e2bcaa88939904815ca0af945740c6bd4.1618478922.git.christophe.leroy@csgroup.eu
Fixes: 30d621f672 ("mm: add generic ptdump")
Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu>
Cc: Steven Price <steven.price@arm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Mapping dirty helpers have, so far, been only used on X86, but a port of
vmwgfx to ARM64 exposed a problem which results in a compilation error
on ARM64 systems:
mm/mapping_dirty_helpers.c: In function `wp_clean_pud_entry':
mm/mapping_dirty_helpers.c:172:32: error: implicit declaration of function `pud_dirty'; did you mean `pmd_dirty'? [-Werror=implicit-function-declaration]
This is due to the fact that mapping_dirty_helpers code assumes that
pud_dirty is always defined, which is not the case for architectures
that don't define CONFIG_HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD.
ARM64 arch is a little inconsistent when it comes to PUD hugepage
helpers, e.g. it defines pud_young but not pud_dirty but regardless of
that the core kernel code shouldn't assume that any of the PUD hugepage
helpers are available unless CONFIG_HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD
is defined. This prevents compilation errors whenever one of the
drivers is ported to new architectures.
Link: https://lkml.kernel.org/r/20210409165151.694574-1-zackr@vmware.com
Signed-off-by: Zack Rusin <zackr@vmware.com>
Reviewed-by: Thomas Hellstrm (Intel) <thomas_os@shipmail.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Fix stray kernel-doc warnings in mm/ due to mis-typed or missing function
names.
Quietens these kernel-doc warnings:
mm/mmu_gather.c:264: warning: expecting prototype for tlb_gather_mmu(). Prototype was for __tlb_gather_mmu() instead
mm/oom_kill.c:180: warning: expecting prototype for Check whether unreclaimable slab amount is greater than(). Prototype was for should_dump_unreclaim_slab() instead
mm/shuffle.c:155: warning: expecting prototype for shuffle_free_memory(). Prototype was for __shuffle_free_memory() instead
Link: https://lkml.kernel.org/r/20210411210642.11362-1-rdunlap@infradead.org
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Reviewed-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
__pcpu_balance_workfn() became fairly big and hard to follow, but in
fact it consists of two fully independent parts, responsible for
the destruction of excessive free chunks and population of necessarily
amount of free pages.
In order to simplify the code and prepare for adding of a new
functionality, split it in two functions:
1) pcpu_balance_free,
2) pcpu_balance_populated.
Move the taking/releasing of the pcpu_alloc_mutex to an upper level
to keep the current synchronization in place.
Signed-off-by: Roman Gushchin <guro@fb.com>
Reviewed-by: Dennis Zhou <dennis@kernel.org>
Signed-off-by: Dennis Zhou <dennis@kernel.org>
Since the commit 3e54097beb ("percpu: manage chunks based on
contig_bits instead of free_bytes") chunks are sorted based on the
size of the biggest continuous free area instead of the total number
of free bytes. Update the corresponding comment to reflect this.
Signed-off-by: Roman Gushchin <guro@fb.com>
Signed-off-by: Dennis Zhou <dennis@kernel.org>
Pull RCU changes from Paul E. McKenney:
- Bitmap support for "N" as alias for last bit
- kvfree_rcu updates
- mm_dump_obj() updates. (One of these is to mm, but was suggested by Andrew Morton.)
- RCU callback offloading update
- Polling RCU grace-period interfaces
- Realtime-related RCU updates
- Tasks-RCU updates
- Torture-test updates
- Torture-test scripting updates
- Miscellaneous fixes
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Architectures supported by KASAN_HW_TAGS can provide a sync or async mode
of execution. On an MTE enabled arm64 hw for example this can be identified
with the synchronous or asynchronous tagging mode of execution.
In synchronous mode, an exception is triggered if a tag check fault occurs.
In asynchronous mode, if a tag check fault occurs, the TFSR_EL1 register is
updated asynchronously. The kernel checks the corresponding bits
periodically.
KASAN requires a specific kernel command line parameter to make use of this
hw features.
Add KASAN HW execution mode kernel command line parameter.
Note: This patch adds the kasan.mode kernel parameter and the
sync/async kernel command line options to enable the described features.
[ Add a new var instead of exposing kasan_arg_mode to be consistent with
flags for other command line arguments. ]
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Andrey Ryabinin <aryabinin@virtuozzo.com>
Cc: Alexander Potapenko <glider@google.com>
Cc: Andrey Konovalov <andreyknvl@google.com>
Reviewed-by: Andrey Konovalov <andreyknvl@google.com>
Acked-by: Catalin Marinas <catalin.marinas@arm.com>
Acked-by: Andrey Konovalov <andreyknvl@google.com>
Tested-by: Andrey Konovalov <andreyknvl@google.com>
Signed-off-by: Vincenzo Frascino <vincenzo.frascino@arm.com>
Signed-off-by: Andrey Konovalov <andreyknvl@google.com>
Link: https://lore.kernel.org/r/20210315132019.33202-3-vincenzo.frascino@arm.com
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Pull percpu fix from Dennis Zhou:
"This contains a fix for sporadically failing atomic percpu
allocations.
I only caught it recently while I was reviewing a new series [1] and
simultaneously saw reports by btrfs in xfstests [2] and [3].
In v5.9, memcg accounting was extended to percpu done by adding a
second type of chunk. I missed an interaction with the free page float
count used to ensure we can support atomic allocations. If one type of
chunk has no free pages, but the other has enough to satisfy the free
page float requirement, we will not repopulate the free pages for the
former type of chunk. This led to the sporadically failing atomic
allocations"
Link: https://lore.kernel.org/linux-mm/20210324190626.564297-1-guro@fb.com/ [1]
Link: https://lore.kernel.org/linux-mm/20210401185158.3275.409509F4@e16-tech.com/ [2]
Link: https://lore.kernel.org/linux-mm/CAL3q7H5RNBjCi708GH7jnczAOe0BLnacT9C+OBgA-Dx9jhB6SQ@mail.gmail.com/ [3]
* 'for-5.12-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/dennis/percpu:
percpu: make pcpu_nr_empty_pop_pages per chunk type
When we do coredump for user process signal, this may be an SIGBUS signal
with BUS_MCEERR_AR or BUS_MCEERR_AO code, which means this signal is
resulted from ECC memory fail like SRAR or SRAO, we expect the memory
recovery work is finished correctly, then the get_dump_page() will not
return the error page as its process pte is set invalid by
memory_failure().
But memory_failure() may fail, and the process's related pte may not be
correctly set invalid, for current code, we will return the poison page,
get it dumped, and then lead to system panic as its in kernel code.
So check the poison status in get_dump_page(), and if TRUE, return NULL.
There maybe other scenario that is also better to check the posion status
and not to panic, so make a wrapper for this check, Thanks to David's
suggestion(<david@redhat.com>).
[akpm@linux-foundation.org: s/0/false/]
[yaoaili@kingsoft.com: is_page_poisoned() arg cannot be null, per Matthew]
Link: https://lkml.kernel.org/r/20210322115233.05e4e82a@alex-virtual-machine
Link: https://lkml.kernel.org/r/20210319104437.6f30e80d@alex-virtual-machine
Signed-off-by: Aili Yao <yaoaili@kingsoft.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Naoya Horiguchi <naoya.horiguchi@nec.com>
Cc: Oscar Salvador <osalvador@suse.de>
Cc: Mike Kravetz <mike.kravetz@oracle.com>
Cc: Aili Yao <yaoaili@kingsoft.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The state of CONFIG_INIT_ON_ALLOC_DEFAULT_ON (and ...ON_FREE...) did not
change the assembly ordering of the static branches: they were always out
of line. Use the new jump_label macros to check the CONFIG settings to
default to the "expected" state, which slightly optimizes the resulting
assembly code.
Signed-off-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Alexander Potapenko <glider@google.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Link: https://lore.kernel.org/r/20210401232347.2791257-3-keescook@chromium.org
There are code paths that rely on zero_pfn to be fully initialized
before core_initcall. For example, wq_sysfs_init() is a core_initcall
function that eventually results in a call to kernel_execve, which
causes a page fault with a subsequent mmput. If zero_pfn is not
initialized by then it may not get cleaned up properly and result in an
error:
BUG: Bad rss-counter state mm:(ptrval) type:MM_ANONPAGES val:1
Here is an analysis of the race as seen on a MIPS device. On this
particular MT7621 device (Ubiquiti ER-X), zero_pfn is PFN 0 until
initialized, at which point it becomes PFN 5120:
1. wq_sysfs_init calls into kobject_uevent_env at core_initcall:
kobject_uevent_env+0x7e4/0x7ec
kset_register+0x68/0x88
bus_register+0xdc/0x34c
subsys_virtual_register+0x34/0x78
wq_sysfs_init+0x1c/0x4c
do_one_initcall+0x50/0x1a8
kernel_init_freeable+0x230/0x2c8
kernel_init+0x10/0x100
ret_from_kernel_thread+0x14/0x1c
2. kobject_uevent_env() calls call_usermodehelper_exec() which executes
kernel_execve asynchronously.
3. Memory allocations in kernel_execve cause a page fault, bumping the
MM reference counter:
add_mm_counter_fast+0xb4/0xc0
handle_mm_fault+0x6e4/0xea0
__get_user_pages.part.78+0x190/0x37c
__get_user_pages_remote+0x128/0x360
get_arg_page+0x34/0xa0
copy_string_kernel+0x194/0x2a4
kernel_execve+0x11c/0x298
call_usermodehelper_exec_async+0x114/0x194
4. In case zero_pfn has not been initialized yet, zap_pte_range does
not decrement the MM_ANONPAGES RSS counter and the BUG message is
triggered shortly afterwards when __mmdrop checks the ref counters:
__mmdrop+0x98/0x1d0
free_bprm+0x44/0x118
kernel_execve+0x160/0x1d8
call_usermodehelper_exec_async+0x114/0x194
ret_from_kernel_thread+0x14/0x1c
To avoid races such as described above, initialize init_zero_pfn at
early_initcall level. Depending on the architecture, ZERO_PAGE is
either constant or gets initialized even earlier, at paging_init, so
there is no issue with initializing zero_pfn earlier.
Link: https://lkml.kernel.org/r/CALCv0x2YqOXEAy2Q=hafjhHCtTHVodChv1qpM=niAXOpqEbt7w@mail.gmail.com
Signed-off-by: Ilya Lipnitskiy <ilya.lipnitskiy@gmail.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: stable@vger.kernel.org
Tested-by: 周琰杰 (Zhou Yanjie) <zhouyanjie@wanyeetech.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Merge misc fixes from Andrew Morton:
"14 patches.
Subsystems affected by this patch series: mm (hugetlb, kasan, gup,
selftests, z3fold, kfence, memblock, and highmem), squashfs, ia64,
gcov, and mailmap"
* emailed patches from Andrew Morton <akpm@linux-foundation.org>:
mailmap: update Andrey Konovalov's email address
mm/highmem: fix CONFIG_DEBUG_KMAP_LOCAL_FORCE_MAP
mm: memblock: fix section mismatch warning again
kfence: make compatible with kmemleak
gcov: fix clang-11+ support
ia64: fix format strings for err_inject
ia64: mca: allocate early mca with GFP_ATOMIC
squashfs: fix xattr id and id lookup sanity checks
squashfs: fix inode lookup sanity checks
z3fold: prevent reclaim/free race for headless pages
selftests/vm: fix out-of-tree build
mm/mmu_notifiers: ensure range_end() is paired with range_start()
kasan: fix per-page tags for non-page_alloc pages
hugetlb_cgroup: fix imbalanced css_get and css_put pair for shared mappings
Because memblock allocations are registered with kmemleak, the KFENCE
pool was seen by kmemleak as one large object. Later allocations
through kfence_alloc() that were registered with kmemleak via
slab_post_alloc_hook() would then overlap and trigger a warning.
Therefore, once the pool is initialized, we can remove (free) it from
kmemleak again, since it should be treated as allocator-internal and be
seen as "free memory".
The second problem is that kmemleak is passed the rounded size, and not
the originally requested size, which is also the size of KFENCE objects.
To avoid kmemleak scanning past the end of an object and trigger a
KFENCE out-of-bounds error, fix the size if it is a KFENCE object.
For simplicity, to avoid a call to kfence_ksize() in
slab_post_alloc_hook() (and avoid new IS_ENABLED(CONFIG_DEBUG_KMEMLEAK)
guard), just call kfence_ksize() in mm/kmemleak.c:create_object().
Link: https://lkml.kernel.org/r/20210317084740.3099921-1-elver@google.com
Signed-off-by: Marco Elver <elver@google.com>
Reported-by: Luis Henriques <lhenriques@suse.de>
Reviewed-by: Catalin Marinas <catalin.marinas@arm.com>
Tested-by: Luis Henriques <lhenriques@suse.de>
Cc: Alexander Potapenko <glider@google.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Andrey Konovalov <andreyknvl@google.com>
Cc: Jann Horn <jannh@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
If one or more notifiers fails .invalidate_range_start(), invoke
.invalidate_range_end() for "all" notifiers. If there are multiple
notifiers, those that did not fail are expecting _start() and _end() to
be paired, e.g. KVM's mmu_notifier_count would become imbalanced.
Disallow notifiers that can fail _start() from implementing _end() so
that it's unnecessary to either track which notifiers rejected _start(),
or had already succeeded prior to a failed _start().
Note, the existing behavior of calling _start() on all notifiers even
after a previous notifier failed _start() was an unintented "feature".
Make it canon now that the behavior is depended on for correctness.
As of today, the bug is likely benign:
1. The only caller of the non-blocking notifier is OOM kill.
2. The only notifiers that can fail _start() are the i915 and Nouveau
drivers.
3. The only notifiers that utilize _end() are the SGI UV GRU driver
and KVM.
4. The GRU driver will never coincide with the i195/Nouveau drivers.
5. An imbalanced kvm->mmu_notifier_count only causes soft lockup in the
_guest_, and the guest is already doomed due to being an OOM victim.
Fix the bug now to play nice with future usage, e.g. KVM has a
potential use case for blocking memslot updates in KVM while an
invalidation is in-progress, and failure to unblock would result in said
updates being blocked indefinitely and hanging.
Found by inspection. Verified by adding a second notifier in KVM that
periodically returns -EAGAIN on non-blockable ranges, triggering OOM,
and observing that KVM exits with an elevated notifier count.
Link: https://lkml.kernel.org/r/20210311180057.1582638-1-seanjc@google.com
Fixes: 93065ac753 ("mm, oom: distinguish blockable mode for mmu notifiers")
Signed-off-by: Sean Christopherson <seanjc@google.com>
Suggested-by: Jason Gunthorpe <jgg@ziepe.ca>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Ben Gardon <bgardon@google.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: "Jérôme Glisse" <jglisse@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Dimitri Sivanich <dimitri.sivanich@hpe.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The current implementation of hugetlb_cgroup for shared mappings could
have different behavior. Consider the following two scenarios:
1.Assume initial css reference count of hugetlb_cgroup is 1:
1.1 Call hugetlb_reserve_pages with from = 1, to = 2. So css reference
count is 2 associated with 1 file_region.
1.2 Call hugetlb_reserve_pages with from = 2, to = 3. So css reference
count is 3 associated with 2 file_region.
1.3 coalesce_file_region will coalesce these two file_regions into
one. So css reference count is 3 associated with 1 file_region
now.
2.Assume initial css reference count of hugetlb_cgroup is 1 again:
2.1 Call hugetlb_reserve_pages with from = 1, to = 3. So css reference
count is 2 associated with 1 file_region.
Therefore, we might have one file_region while holding one or more css
reference counts. This inconsistency could lead to imbalanced css_get()
and css_put() pair. If we do css_put one by one (i.g. hole punch case),
scenario 2 would put one more css reference. If we do css_put all
together (i.g. truncate case), scenario 1 will leak one css reference.
The imbalanced css_get() and css_put() pair would result in a non-zero
reference when we try to destroy the hugetlb cgroup. The hugetlb cgroup
directory is removed __but__ associated resource is not freed. This
might result in OOM or can not create a new hugetlb cgroup in a busy
workload ultimately.
In order to fix this, we have to make sure that one file_region must
hold exactly one css reference. So in coalesce_file_region case, we
should release one css reference before coalescence. Also only put css
reference when the entire file_region is removed.
The last thing to note is that the caller of region_add() will only hold
one reference to h_cg->css for the whole contiguous reservation region.
But this area might be scattered when there are already some
file_regions reside in it. As a result, many file_regions may share only
one h_cg->css reference. In order to ensure that one file_region must
hold exactly one css reference, we should do css_get() for each
file_region and release the reference held by caller when they are done.
[linmiaohe@huawei.com: fix imbalanced css_get and css_put pair for shared mappings]
Link: https://lkml.kernel.org/r/20210316023002.53921-1-linmiaohe@huawei.com
Link: https://lkml.kernel.org/r/20210301120540.37076-1-linmiaohe@huawei.com
Fixes: 075a61d07a ("hugetlb_cgroup: add accounting for shared mappings")
Reported-by: kernel test robot <lkp@intel.com> (auto build test ERROR)
Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
Reviewed-by: Mike Kravetz <mike.kravetz@oracle.com>
Cc: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Cc: Wanpeng Li <liwp.linux@gmail.com>
Cc: Mina Almasry <almasrymina@google.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Merge misc fixes from Andrew Morton:
"28 patches.
Subsystems affected by this series: mm (memblock, pagealloc, hugetlb,
highmem, kfence, oom-kill, madvise, kasan, userfaultfd, memcg, and
zram), core-kernel, kconfig, fork, binfmt, MAINTAINERS, kbuild, and
ia64"
* emailed patches from Andrew Morton <akpm@linux-foundation.org>: (28 commits)
zram: fix broken page writeback
zram: fix return value on writeback_store
mm/memcg: set memcg when splitting page
mm/memcg: rename mem_cgroup_split_huge_fixup to split_page_memcg and add nr_pages argument
ia64: fix ptrace(PTRACE_SYSCALL_INFO_EXIT) sign
ia64: fix ia64_syscall_get_set_arguments() for break-based syscalls
mm/userfaultfd: fix memory corruption due to writeprotect
kasan: fix KASAN_STACK dependency for HW_TAGS
kasan, mm: fix crash with HW_TAGS and DEBUG_PAGEALLOC
mm/madvise: replace ptrace attach requirement for process_madvise
include/linux/sched/mm.h: use rcu_dereference in in_vfork()
kfence: fix reports if constant function prefixes exist
kfence, slab: fix cache_alloc_debugcheck_after() for bulk allocations
kfence: fix printk format for ptrdiff_t
linux/compiler-clang.h: define HAVE_BUILTIN_BSWAP*
MAINTAINERS: exclude uapi directories in API/ABI section
binfmt_misc: fix possible deadlock in bm_register_write
mm/highmem.c: fix zero_user_segments() with start > end
hugetlb: do early cow when page pinned on src mm
mm: use is_cow_mapping() across tree where proper
...
Userfaultfd self-test fails occasionally, indicating a memory corruption.
Analyzing this problem indicates that there is a real bug since mmap_lock
is only taken for read in mwriteprotect_range() and defers flushes, and
since there is insufficient consideration of concurrent deferred TLB
flushes in wp_page_copy(). Although the PTE is flushed from the TLBs in
wp_page_copy(), this flush takes place after the copy has already been
performed, and therefore changes of the page are possible between the time
of the copy and the time in which the PTE is flushed.
To make matters worse, memory-unprotection using userfaultfd also poses a
problem. Although memory unprotection is logically a promotion of PTE
permissions, and therefore should not require a TLB flush, the current
userrfaultfd code might actually cause a demotion of the architectural PTE
permission: when userfaultfd_writeprotect() unprotects memory region, it
unintentionally *clears* the RW-bit if it was already set. Note that this
unprotecting a PTE that is not write-protected is a valid use-case: the
userfaultfd monitor might ask to unprotect a region that holds both
write-protected and write-unprotected PTEs.
The scenario that happens in selftests/vm/userfaultfd is as follows:
cpu0 cpu1 cpu2
---- ---- ----
[ Writable PTE
cached in TLB ]
userfaultfd_writeprotect()
[ write-*unprotect* ]
mwriteprotect_range()
mmap_read_lock()
change_protection()
change_protection_range()
...
change_pte_range()
[ *clear* “write”-bit ]
[ defer TLB flushes ]
[ page-fault ]
...
wp_page_copy()
cow_user_page()
[ copy page ]
[ write to old
page ]
...
set_pte_at_notify()
A similar scenario can happen:
cpu0 cpu1 cpu2 cpu3
---- ---- ---- ----
[ Writable PTE
cached in TLB ]
userfaultfd_writeprotect()
[ write-protect ]
[ deferred TLB flush ]
userfaultfd_writeprotect()
[ write-unprotect ]
[ deferred TLB flush]
[ page-fault ]
wp_page_copy()
cow_user_page()
[ copy page ]
... [ write to page ]
set_pte_at_notify()
This race exists since commit 292924b260 ("userfaultfd: wp: apply
_PAGE_UFFD_WP bit"). Yet, as Yu Zhao pointed, these races became apparent
since commit 09854ba94c ("mm: do_wp_page() simplification") which made
wp_page_copy() more likely to take place, specifically if page_count(page)
> 1.
To resolve the aforementioned races, check whether there are pending
flushes on uffd-write-protected VMAs, and if there are, perform a flush
before doing the COW.
Further optimizations will follow to avoid during uffd-write-unprotect
unnecassary PTE write-protection and TLB flushes.
Link: https://lkml.kernel.org/r/20210304095423.3825684-1-namit@vmware.com
Fixes: 09854ba94c ("mm: do_wp_page() simplification")
Signed-off-by: Nadav Amit <namit@vmware.com>
Suggested-by: Yu Zhao <yuzhao@google.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Tested-by: Peter Xu <peterx@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Pavel Emelyanov <xemul@openvz.org>
Cc: Mike Kravetz <mike.kravetz@oracle.com>
Cc: Mike Rapoport <rppt@linux.vnet.ibm.com>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Will Deacon <will@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: <stable@vger.kernel.org> [5.9+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
process_madvise currently requires ptrace attach capability.
PTRACE_MODE_ATTACH gives one process complete control over another
process. It effectively removes the security boundary between the two
processes (in one direction). Granting ptrace attach capability even to a
system process is considered dangerous since it creates an attack surface.
This severely limits the usage of this API.
The operations process_madvise can perform do not affect the correctness
of the operation of the target process; they only affect where the data is
physically located (and therefore, how fast it can be accessed). What we
want is the ability for one process to influence another process in order
to optimize performance across the entire system while leaving the
security boundary intact.
Replace PTRACE_MODE_ATTACH with a combination of PTRACE_MODE_READ and
CAP_SYS_NICE. PTRACE_MODE_READ to prevent leaking ASLR metadata and
CAP_SYS_NICE for influencing process performance.
Link: https://lkml.kernel.org/r/20210303185807.2160264-1-surenb@google.com
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Reviewed-by: Kees Cook <keescook@chromium.org>
Acked-by: Minchan Kim <minchan@kernel.org>
Acked-by: David Rientjes <rientjes@google.com>
Cc: Jann Horn <jannh@google.com>
Cc: Jeff Vander Stoep <jeffv@google.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Shakeel Butt <shakeelb@google.com>
Cc: Tim Murray <timmurray@google.com>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: James Morris <jmorris@namei.org>
Cc: <stable@vger.kernel.org> [5.10+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
zero_user_segments() is used from __block_write_begin_int(), for example
like the following
zero_user_segments(page, 4096, 1024, 512, 918)
But new the zero_user_segments() implementation for for HIGHMEM +
TRANSPARENT_HUGEPAGE doesn't handle "start > end" case correctly, and hits
BUG_ON(). (we can fix __block_write_begin_int() instead though, it is the
old and multiple usage)
Also it calls kmap_atomic() unnecessarily while start == end == 0.
Link: https://lkml.kernel.org/r/87v9ab60r4.fsf@mail.parknet.co.jp
Fixes: 0060ef3b4e ("mm: support THPs in zero_user_segments")
Signed-off-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
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>
