linux-toradex.git/mm/hugetlb.c, branch v5.4-rc1

hugetlbfs: don't retry when pool page allocations start to fail

2019-09-24T22:54:10+00:00

When allocating hugetlbfs pool pages via /proc/sys/vm/nr_hugepages, the
pages will be interleaved between all nodes of the system.  If nodes are
not equal, it is quite possible for one node to fill up before the others.
When this happens, the code still attempts to allocate pages from the
full node.  This results in calls to direct reclaim and compaction which
slow things down considerably.

When allocating pool pages, note the state of the previous allocation for
each node.  If previous allocation failed, do not use the aggressive retry
algorithm on successive attempts.  The allocation will still succeed if
there is memory available, but it will not try as hard to free up memory.

Link: http://lkml.kernel.org/r/20190806014744.15446-5-mike.kravetz@oracle.com
Signed-off-by: Mike Kravetz 
Acked-by: Vlastimil Babka 
Cc: Hillf Danton 
Cc: Johannes Weiner 
Cc: Mel Gorman 
Cc: Michal Hocko 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

hugetlbfs: fix hugetlb page migration/fault race causing SIGBUS

2019-08-13T23:06:53+00:00

Li Wang discovered that LTP/move_page12 V2 sometimes triggers SIGBUS in
the kernel-v5.2.3 testing.  This is caused by a race between hugetlb
page migration and page fault.

If a hugetlb page can not be allocated to satisfy a page fault, the task
is sent SIGBUS.  This is normal hugetlbfs behavior.  A hugetlb fault
mutex exists to prevent two tasks from trying to instantiate the same
page.  This protects against the situation where there is only one
hugetlb page, and both tasks would try to allocate.  Without the mutex,
one would fail and SIGBUS even though the other fault would be
successful.

There is a similar race between hugetlb page migration and fault.
Migration code will allocate a page for the target of the migration.  It
will then unmap the original page from all page tables.  It does this
unmap by first clearing the pte and then writing a migration entry.  The
page table lock is held for the duration of this clear and write
operation.  However, the beginnings of the hugetlb page fault code
optimistically checks the pte without taking the page table lock.  If
clear (as it can be during the migration unmap operation), a hugetlb
page allocation is attempted to satisfy the fault.  Note that the page
which will eventually satisfy this fault was already allocated by the
migration code.  However, the allocation within the fault path could
fail which would result in the task incorrectly being sent SIGBUS.

Ideally, we could take the hugetlb fault mutex in the migration code
when modifying the page tables.  However, locks must be taken in the
order of hugetlb fault mutex, page lock, page table lock.  This would
require significant rework of the migration code.  Instead, the issue is
addressed in the hugetlb fault code.  After failing to allocate a huge
page, take the page table lock and check for huge_pte_none before
returning an error.  This is the same check that must be made further in
the code even if page allocation is successful.

Link: http://lkml.kernel.org/r/20190808000533.7701-1-mike.kravetz@oracle.com
Fixes: 290408d4a250 ("hugetlb: hugepage migration core")
Signed-off-by: Mike Kravetz 
Reported-by: Li Wang 
Tested-by: Li Wang 
Reviewed-by: Naoya Horiguchi 
Acked-by: Michal Hocko 
Cc: Cyril Hrubis 
Cc: Xishi Qiu 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

mm: hugetlb: soft-offline: dissolve_free_huge_page() return zero on !PageHuge

2019-06-29T08:43:45+00:00

madvise(MADV_SOFT_OFFLINE) often returns -EBUSY when calling soft offline
for hugepages with overcommitting enabled.  That was caused by the
suboptimal code in current soft-offline code.  See the following part:

    ret = migrate_pages(&pagelist, new_page, NULL, MPOL_MF_MOVE_ALL,
                            MIGRATE_SYNC, MR_MEMORY_FAILURE);
    if (ret) {
            ...
    } else {
            /*
             * We set PG_hwpoison only when the migration source hugepage
             * was successfully dissolved, because otherwise hwpoisoned
             * hugepage remains on free hugepage list, then userspace will
             * find it as SIGBUS by allocation failure. That's not expected
             * in soft-offlining.
             */
            ret = dissolve_free_huge_page(page);
            if (!ret) {
                    if (set_hwpoison_free_buddy_page(page))
                            num_poisoned_pages_inc();
            }
    }
    return ret;

Here dissolve_free_huge_page() returns -EBUSY if the migration source page
was freed into buddy in migrate_pages(), but even in that case we actually
has a chance that set_hwpoison_free_buddy_page() succeeds.  So that means
current code gives up offlining too early now.

dissolve_free_huge_page() checks that a given hugepage is suitable for
dissolving, where we should return success for !PageHuge() case because
the given hugepage is considered as already dissolved.

This change also affects other callers of dissolve_free_huge_page(), which
are cleaned up together.

[n-horiguchi@ah.jp.nec.com: v3]
  Link: http://lkml.kernel.org/r/1560761476-4651-3-git-send-email-n-horiguchi@ah.jp.nec.comLink: http://lkml.kernel.org/r/1560154686-18497-3-git-send-email-n-horiguchi@ah.jp.nec.com
Fixes: 6bc9b56433b76 ("mm: fix race on soft-offlining")
Signed-off-by: Naoya Horiguchi 
Reported-by: Chen, Jerry T 
Tested-by: Chen, Jerry T 
Reviewed-by: Mike Kravetz 
Reviewed-by: Oscar Salvador 
Cc: Michal Hocko 
Cc: Xishi Qiu 
Cc: "Chen, Jerry T" 
Cc: "Zhuo, Qiuxu" 
Cc: 	[4.19+]
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

treewide: Add SPDX license identifier for missed files

2019-05-21T08:50:45+00:00

Add SPDX license identifiers to all files which:

 - Have no license information of any form

 - Have EXPORT_.*_SYMBOL_GPL inside which was used in the
   initial scan/conversion to ignore the file

These files fall under the project license, GPL v2 only. The resulting SPDX
license identifier is:

  GPL-2.0-only

Signed-off-by: Thomas Gleixner 
Signed-off-by: Greg Kroah-Hartman

hugetlbfs: always use address space in inode for resv_map pointer

2019-05-14T16:47:50+00:00

Continuing discussion about 58b6e5e8f1ad ("hugetlbfs: fix memory leak for
resv_map") brought up the issue that inode->i_mapping may not point to the
address space embedded within the inode at inode eviction time.  The
hugetlbfs truncate routine handles this by explicitly using inode->i_data.
However, code cleaning up the resv_map will still use the address space
pointed to by inode->i_mapping.  Luckily, private_data is NULL for address
spaces in all such cases today but, there is no guarantee this will
continue.

Change all hugetlbfs code getting a resv_map pointer to explicitly get it
from the address space embedded within the inode.  In addition, add more
comments in the code to indicate why this is being done.

Link: http://lkml.kernel.org/r/20190419204435.16984-1-mike.kravetz@oracle.com
Signed-off-by: Mike Kravetz 
Reported-by: Yufen Yu 
Cc: Michal Hocko 
Cc: Naoya Horiguchi 
Cc: "Kirill A . Shutemov" 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

mm/mmu_notifier: use correct mmu_notifier events for each invalidation

2019-05-14T16:47:49+00:00

This updates each existing invalidation to use the correct mmu notifier
event that represent what is happening to the CPU page table.  See the
patch which introduced the events to see the rational behind this.

Link: http://lkml.kernel.org/r/20190326164747.24405-7-jglisse@redhat.com
Signed-off-by: Jérôme Glisse 
Reviewed-by: Ralph Campbell 
Reviewed-by: Ira Weiny 
Cc: Christian König 
Cc: Joonas Lahtinen 
Cc: Jani Nikula 
Cc: Rodrigo Vivi 
Cc: Jan Kara 
Cc: Andrea Arcangeli 
Cc: Peter Xu 
Cc: Felix Kuehling 
Cc: Jason Gunthorpe 
Cc: Ross Zwisler 
Cc: Dan Williams 
Cc: Paolo Bonzini 
Cc: Radim Krcmar 
Cc: Michal Hocko 
Cc: Christian Koenig 
Cc: John Hubbard 
Cc: Arnd Bergmann 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

mm/mmu_notifier: contextual information for event triggering invalidation

2019-05-14T16:47:49+00:00

CPU page table update can happens for many reasons, not only as a result
of a syscall (munmap(), mprotect(), mremap(), madvise(), ...) but also as
a result of kernel activities (memory compression, reclaim, migration,
...).

Users of mmu notifier API track changes to the CPU page table and take
specific action for them.  While current API only provide range of virtual
address affected by the change, not why the changes is happening.

This patchset do the initial mechanical convertion of all the places that
calls mmu_notifier_range_init to also provide the default MMU_NOTIFY_UNMAP
event as well as the vma if it is know (most invalidation happens against
a given vma).  Passing down the vma allows the users of mmu notifier to
inspect the new vma page protection.

The MMU_NOTIFY_UNMAP is always the safe default as users of mmu notifier
should assume that every for the range is going away when that event
happens.  A latter patch do convert mm call path to use a more appropriate
events for each call.

This is done as 2 patches so that no call site is forgotten especialy
as it uses this following coccinelle patch:

%<----------------------------------------------------------------------
@@
identifier I1, I2, I3, I4;
@@
static inline void mmu_notifier_range_init(struct mmu_notifier_range *I1,
+enum mmu_notifier_event event,
+unsigned flags,
+struct vm_area_struct *vma,
struct mm_struct *I2, unsigned long I3, unsigned long I4) { ... }

@@
@@
-#define mmu_notifier_range_init(range, mm, start, end)
+#define mmu_notifier_range_init(range, event, flags, vma, mm, start, end)

@@
expression E1, E3, E4;
identifier I1;
@@
<...
mmu_notifier_range_init(E1,
+MMU_NOTIFY_UNMAP, 0, I1,
I1->vm_mm, E3, E4)
...>

@@
expression E1, E2, E3, E4;
identifier FN, VMA;
@@
FN(..., struct vm_area_struct *VMA, ...) {
<...
mmu_notifier_range_init(E1,
+MMU_NOTIFY_UNMAP, 0, VMA,
E2, E3, E4)
...> }

@@
expression E1, E2, E3, E4;
identifier FN, VMA;
@@
FN(...) {
struct vm_area_struct *VMA;
<...
mmu_notifier_range_init(E1,
+MMU_NOTIFY_UNMAP, 0, VMA,
E2, E3, E4)
...> }

@@
expression E1, E2, E3, E4;
identifier FN;
@@
FN(...) {
<...
mmu_notifier_range_init(E1,
+MMU_NOTIFY_UNMAP, 0, NULL,
E2, E3, E4)
...> }
---------------------------------------------------------------------->%

Applied with:
spatch --all-includes --sp-file mmu-notifier.spatch fs/proc/task_mmu.c --in-place
spatch --sp-file mmu-notifier.spatch --dir kernel/events/ --in-place
spatch --sp-file mmu-notifier.spatch --dir mm --in-place

Link: http://lkml.kernel.org/r/20190326164747.24405-6-jglisse@redhat.com
Signed-off-by: Jérôme Glisse 
Reviewed-by: Ralph Campbell 
Reviewed-by: Ira Weiny 
Cc: Christian König 
Cc: Joonas Lahtinen 
Cc: Jani Nikula 
Cc: Rodrigo Vivi 
Cc: Jan Kara 
Cc: Andrea Arcangeli 
Cc: Peter Xu 
Cc: Felix Kuehling 
Cc: Jason Gunthorpe 
Cc: Ross Zwisler 
Cc: Dan Williams 
Cc: Paolo Bonzini 
Cc: Radim Krcmar 
Cc: Michal Hocko 
Cc: Christian Koenig 
Cc: John Hubbard 
Cc: Arnd Bergmann 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

hugetlb: use same fault hash key for shared and private mappings

2019-05-14T16:47:48+00:00

hugetlb uses a fault mutex hash table to prevent page faults of the
same pages concurrently.  The key for shared and private mappings is
different.  Shared keys off address_space and file index.  Private keys
off mm and virtual address.  Consider a private mappings of a populated
hugetlbfs file.  A fault will map the page from the file and if needed
do a COW to map a writable page.

Hugetlbfs hole punch uses the fault mutex to prevent mappings of file
pages.  It uses the address_space file index key.  However, private
mappings will use a different key and could race with this code to map
the file page.  This causes problems (BUG) for the page cache remove
code as it expects the page to be unmapped.  A sample stack is:

page dumped because: VM_BUG_ON_PAGE(page_mapped(page))
kernel BUG at mm/filemap.c:169!
...
RIP: 0010:unaccount_page_cache_page+0x1b8/0x200
...
Call Trace:
__delete_from_page_cache+0x39/0x220
delete_from_page_cache+0x45/0x70
remove_inode_hugepages+0x13c/0x380
? __add_to_page_cache_locked+0x162/0x380
hugetlbfs_fallocate+0x403/0x540
? _cond_resched+0x15/0x30
? __inode_security_revalidate+0x5d/0x70
? selinux_file_permission+0x100/0x130
vfs_fallocate+0x13f/0x270
ksys_fallocate+0x3c/0x80
__x64_sys_fallocate+0x1a/0x20
do_syscall_64+0x5b/0x180
entry_SYSCALL_64_after_hwframe+0x44/0xa9

There seems to be another potential COW issue/race with this approach
of different private and shared keys as noted in commit 8382d914ebf7
("mm, hugetlb: improve page-fault scalability").

Since every hugetlb mapping (even anon and private) is actually a file
mapping, just use the address_space index key for all mappings.  This
results in potentially more hash collisions.  However, this should not
be the common case.

Link: http://lkml.kernel.org/r/20190328234704.27083-3-mike.kravetz@oracle.com
Link: http://lkml.kernel.org/r/20190412165235.t4sscoujczfhuiyt@linux-r8p5
Fixes: b5cec28d36f5 ("hugetlbfs: truncate_hugepages() takes a range of pages")
Signed-off-by: Mike Kravetz 
Reviewed-by: Naoya Horiguchi 
Reviewed-by: Davidlohr Bueso 
Cc: Joonsoo Kim 
Cc: "Kirill A . Shutemov" 
Cc: Michal Hocko 
Cc: 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

hugetlbfs: on restore reserve error path retain subpool reservation

2019-05-14T16:47:48+00:00

When a huge page is allocated, PagePrivate() is set if the allocation
consumed a reservation.  When freeing a huge page, PagePrivate is checked.
If set, it indicates the reservation should be restored.  PagePrivate
being set at free huge page time mostly happens on error paths.

When huge page reservations are created, a check is made to determine if
the mapping is associated with an explicitly mounted filesystem.  If so,
pages are also reserved within the filesystem.  The default action when
freeing a huge page is to decrement the usage count in any associated
explicitly mounted filesystem.  However, if the reservation is to be
restored the reservation/use count within the filesystem should not be
decrementd.  Otherwise, a subsequent page allocation and free for the same
mapping location will cause the file filesystem usage to go 'negative'.

Filesystem                         Size  Used Avail Use% Mounted on
nodev                              4.0G -4.0M  4.1G    - /opt/hugepool

To fix, when freeing a huge page do not adjust filesystem usage if
PagePrivate() is set to indicate the reservation should be restored.

I did not cc stable as the problem has been around since reserves were
added to hugetlbfs and nobody has noticed.

Link: http://lkml.kernel.org/r/20190328234704.27083-2-mike.kravetz@oracle.com
Signed-off-by: Mike Kravetz 
Reviewed-by: Naoya Horiguchi 
Cc: Davidlohr Bueso 
Cc: Joonsoo Kim 
Cc: Michal Hocko 
Cc: "Kirill A . Shutemov" 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

mm/hugetlb: get rid of NODEMASK_ALLOC

2019-05-14T16:47:48+00:00

NODEMASK_ALLOC is used to allocate a nodemask bitmap, and it does it by
first determining whether it should be allocated on the stack or
dynamically, depending on NODES_SHIFT.  Right now, it goes the dynamic
path whenever the nodemask_t is above 32 bytes.

Although we could bump it to a reasonable value, the largest a nodemask_t
can get is 128 bytes, so since __nr_hugepages_store_common is called from
a rather short stack we can just get rid of the NODEMASK_ALLOC call here.

This reduces some code churn and complexity.

Link: http://lkml.kernel.org/r/20190402133415.21983-1-osalvador@suse.de
Signed-off-by: Oscar Salvador 
Reviewed-by: Mike Kravetz 
Reviewed-by: Naoya Horiguchi 
Cc: Alex Ghiti 
Cc: David Rientjes 
Cc: Jing Xiangfeng 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds