| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
MMU notifiers can sleep, but in page_mkclean_one() we call
mmu_notifier_invalidate_page() under page table lock.
Let's instead use mmu_notifier_invalidate_range() outside
page_vma_mapped_walk() loop.
[[email protected]: try_to_unmap_one() do not call mmu_notifier under ptl]
Link: http://lkml.kernel.org/r/[email protected]
Link: http://lkml.kernel.org/r/[email protected]
Fixes: c7ab0d2fdc84 ("mm: convert try_to_unmap_one() to use page_vma_mapped_walk()")
Signed-off-by: Kirill A. Shutemov <[email protected]>
Signed-off-by: Jérôme Glisse <[email protected]>
Reported-by: axie <[email protected]>
Cc: Alex Deucher <[email protected]>
Cc: "Writer, Tim" <[email protected]>
Cc: <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
We saw many list corruption warnings on shmem shrinklist:
WARNING: CPU: 18 PID: 177 at lib/list_debug.c:59 __list_del_entry+0x9e/0xc0
list_del corruption. prev->next should be ffff9ae5694b82d8, but was ffff9ae5699ba960
Modules linked in: intel_rapl sb_edac edac_core x86_pkg_temp_thermal coretemp iTCO_wdt iTCO_vendor_support crct10dif_pclmul crc32_pclmul ghash_clmulni_intel raid0 dcdbas shpchp wmi hed i2c_i801 ioatdma lpc_ich i2c_smbus acpi_cpufreq tcp_diag inet_diag sch_fq_codel ipmi_si ipmi_devintf ipmi_msghandler igb ptp crc32c_intel pps_core i2c_algo_bit i2c_core dca ipv6 crc_ccitt
CPU: 18 PID: 177 Comm: kswapd1 Not tainted 4.9.34-t3.el7.twitter.x86_64 #1
Hardware name: Dell Inc. PowerEdge C6220/0W6W6G, BIOS 2.2.3 11/07/2013
Call Trace:
dump_stack+0x4d/0x66
__warn+0xcb/0xf0
warn_slowpath_fmt+0x4f/0x60
__list_del_entry+0x9e/0xc0
shmem_unused_huge_shrink+0xfa/0x2e0
shmem_unused_huge_scan+0x20/0x30
super_cache_scan+0x193/0x1a0
shrink_slab.part.41+0x1e3/0x3f0
shrink_slab+0x29/0x30
shrink_node+0xf9/0x2f0
kswapd+0x2d8/0x6c0
kthread+0xd7/0xf0
ret_from_fork+0x22/0x30
WARNING: CPU: 23 PID: 639 at lib/list_debug.c:33 __list_add+0x89/0xb0
list_add corruption. prev->next should be next (ffff9ae5699ba960), but was ffff9ae5694b82d8. (prev=ffff9ae5694b82d8).
Modules linked in: intel_rapl sb_edac edac_core x86_pkg_temp_thermal coretemp iTCO_wdt iTCO_vendor_support crct10dif_pclmul crc32_pclmul ghash_clmulni_intel raid0 dcdbas shpchp wmi hed i2c_i801 ioatdma lpc_ich i2c_smbus acpi_cpufreq tcp_diag inet_diag sch_fq_codel ipmi_si ipmi_devintf ipmi_msghandler igb ptp crc32c_intel pps_core i2c_algo_bit i2c_core dca ipv6 crc_ccitt
CPU: 23 PID: 639 Comm: systemd-udevd Tainted: G W 4.9.34-t3.el7.twitter.x86_64 #1
Hardware name: Dell Inc. PowerEdge C6220/0W6W6G, BIOS 2.2.3 11/07/2013
Call Trace:
dump_stack+0x4d/0x66
__warn+0xcb/0xf0
warn_slowpath_fmt+0x4f/0x60
__list_add+0x89/0xb0
shmem_setattr+0x204/0x230
notify_change+0x2ef/0x440
do_truncate+0x5d/0x90
path_openat+0x331/0x1190
do_filp_open+0x7e/0xe0
do_sys_open+0x123/0x200
SyS_open+0x1e/0x20
do_syscall_64+0x61/0x170
entry_SYSCALL64_slow_path+0x25/0x25
The problem is that shmem_unused_huge_shrink() moves entries from the
global sbinfo->shrinklist to its local lists and then releases the
spinlock. However, a parallel shmem_setattr() could access one of these
entries directly and add it back to the global shrinklist if it is
removed, with the spinlock held.
The logic itself looks solid since an entry could be either in a local
list or the global list, otherwise it is removed from one of them by
list_del_init(). So probably the race condition is that, one CPU is in
the middle of INIT_LIST_HEAD() but the other CPU calls list_empty()
which returns true too early then the following list_add_tail() sees a
corrupted entry.
list_empty_careful() is designed to fix this situation.
[[email protected]: add comments]
Link: http://lkml.kernel.org/r/[email protected]
Fixes: 779750d20b93 ("shmem: split huge pages beyond i_size under memory pressure")
Signed-off-by: Cong Wang <[email protected]>
Acked-by: Linus Torvalds <[email protected]>
Acked-by: Kirill A. Shutemov <[email protected]>
Cc: Hugh Dickins <[email protected]>
Cc: <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Revert commit bb01b64cfab7 ("mm/balloon_compaction.c: enqueue zero page
to balloon device")'
Zeroing ballon pages is rather time consuming, especially when a lot of
pages are in flight. E.g. 7GB worth of ballooned memory takes 2.8s with
__GFP_ZERO while it takes ~491ms without it.
The original commit argued that zeroing will help ksmd to merge these
pages on the host but this argument is assuming that the host actually
marks balloon pages for ksm which is not universally true. So we pay
performance penalty for something that even might not be used in the end
which is wrong. The host can zero out pages on its own when there is a
need.
[[email protected]: new changelog text]
Link: http://lkml.kernel.org/r/[email protected]
Fixes: bb01b64cfab7 ("mm/balloon_compaction.c: enqueue zero page to balloon device")
Signed-off-by: Wei Wang <[email protected]>
Acked-by: Michael S. Tsirkin <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Cc: zhenwei.pi <[email protected]>
Cc: David Hildenbrand <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Nadav reported KSM can corrupt the user data by the TLB batching
race[1]. That means data user written can be lost.
Quote from Nadav Amit:
"For this race we need 4 CPUs:
CPU0: Caches a writable and dirty PTE entry, and uses the stale value
for write later.
CPU1: Runs madvise_free on the range that includes the PTE. It would
clear the dirty-bit. It batches TLB flushes.
CPU2: Writes 4 to /proc/PID/clear_refs , clearing the PTEs soft-dirty.
We care about the fact that it clears the PTE write-bit, and of
course, batches TLB flushes.
CPU3: Runs KSM. Our purpose is to pass the following test in
write_protect_page():
if (pte_write(*pvmw.pte) || pte_dirty(*pvmw.pte) ||
(pte_protnone(*pvmw.pte) && pte_savedwrite(*pvmw.pte)))
Since it will avoid TLB flush. And we want to do it while the PTE is
stale. Later, and before replacing the page, we would be able to
change the page.
Note that all the operations the CPU1-3 perform canhappen in parallel
since they only acquire mmap_sem for read.
We start with two identical pages. Everything below regards the same
page/PTE.
CPU0 CPU1 CPU2 CPU3
---- ---- ---- ----
Write the same
value on page
[cache PTE as
dirty in TLB]
MADV_FREE
pte_mkclean()
4 > clear_refs
pte_wrprotect()
write_protect_page()
[ success, no flush ]
pages_indentical()
[ ok ]
Write to page
different value
[Ok, using stale
PTE]
replace_page()
Later, CPU1, CPU2 and CPU3 would flush the TLB, but that is too late.
CPU0 already wrote on the page, but KSM ignored this write, and it got
lost"
In above scenario, MADV_FREE is fixed by changing TLB batching API
including [set|clear]_tlb_flush_pending. Remained thing is soft-dirty
part.
This patch changes soft-dirty uses TLB batching API instead of
flush_tlb_mm and KSM checks pending TLB flush by using
mm_tlb_flush_pending so that it will flush TLB to avoid data lost if
there are other parallel threads pending TLB flush.
[1] http://lkml.kernel.org/r/[email protected]
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Minchan Kim <[email protected]>
Signed-off-by: Nadav Amit <[email protected]>
Reported-by: Nadav Amit <[email protected]>
Tested-by: Nadav Amit <[email protected]>
Reviewed-by: Andrea Arcangeli <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: Hugh Dickins <[email protected]>
Cc: "David S. Miller" <[email protected]>
Cc: Andy Lutomirski <[email protected]>
Cc: Heiko Carstens <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jeff Dike <[email protected]>
Cc: Martin Schwidefsky <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: Nadav Amit <[email protected]>
Cc: Rik van Riel <[email protected]>
Cc: Russell King <[email protected]>
Cc: Sergey Senozhatsky <[email protected]>
Cc: Tony Luck <[email protected]>
Cc: Yoshinori Sato <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Nadav reported parallel MADV_DONTNEED on same range has a stale TLB
problem and Mel fixed it[1] and found same problem on MADV_FREE[2].
Quote from Mel Gorman:
"The race in question is CPU 0 running madv_free and updating some PTEs
while CPU 1 is also running madv_free and looking at the same PTEs.
CPU 1 may have writable TLB entries for a page but fail the pte_dirty
check (because CPU 0 has updated it already) and potentially fail to
flush.
Hence, when madv_free on CPU 1 returns, there are still potentially
writable TLB entries and the underlying PTE is still present so that a
subsequent write does not necessarily propagate the dirty bit to the
underlying PTE any more. Reclaim at some unknown time at the future
may then see that the PTE is still clean and discard the page even
though a write has happened in the meantime. I think this is possible
but I could have missed some protection in madv_free that prevents it
happening."
This patch aims for solving both problems all at once and is ready for
other problem with KSM, MADV_FREE and soft-dirty story[3].
TLB batch API(tlb_[gather|finish]_mmu] uses [inc|dec]_tlb_flush_pending
and mmu_tlb_flush_pending so that when tlb_finish_mmu is called, we can
catch there are parallel threads going on. In that case, forcefully,
flush TLB to prevent for user to access memory via stale TLB entry
although it fail to gather page table entry.
I confirmed this patch works with [4] test program Nadav gave so this
patch supersedes "mm: Always flush VMA ranges affected by zap_page_range
v2" in current mmotm.
NOTE:
This patch modifies arch-specific TLB gathering interface(x86, ia64,
s390, sh, um). It seems most of architecture are straightforward but
s390 need to be careful because tlb_flush_mmu works only if
mm->context.flush_mm is set to non-zero which happens only a pte entry
really is cleared by ptep_get_and_clear and friends. However, this
problem never changes the pte entries but need to flush to prevent
memory access from stale tlb.
[1] http://lkml.kernel.org/r/[email protected]
[2] http://lkml.kernel.org/r/[email protected]
[3] http://lkml.kernel.org/r/[email protected]
[4] https://patchwork.kernel.org/patch/9861621/
[[email protected]: decrease tlb flush pending count in tlb_finish_mmu]
Link: http://lkml.kernel.org/r/20170808080821.GA31730@bbox
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Minchan Kim <[email protected]>
Signed-off-by: Nadav Amit <[email protected]>
Reported-by: Nadav Amit <[email protected]>
Reported-by: Mel Gorman <[email protected]>
Acked-by: Mel Gorman <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Russell King <[email protected]>
Cc: Tony Luck <[email protected]>
Cc: Martin Schwidefsky <[email protected]>
Cc: "David S. Miller" <[email protected]>
Cc: Heiko Carstens <[email protected]>
Cc: Yoshinori Sato <[email protected]>
Cc: Jeff Dike <[email protected]>
Cc: Andrea Arcangeli <[email protected]>
Cc: Andy Lutomirski <[email protected]>
Cc: Hugh Dickins <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: Nadav Amit <[email protected]>
Cc: Rik van Riel <[email protected]>
Cc: Sergey Senozhatsky <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Currently, tlb_flush_pending is used only for CONFIG_[NUMA_BALANCING|
COMPACTION] but upcoming patches to solve subtle TLB flush batching
problem will use it regardless of compaction/NUMA so this patch doesn't
remove the dependency.
[[email protected]: remove more ifdefs from world's ugliest printk statement]
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Minchan Kim <[email protected]>
Signed-off-by: Nadav Amit <[email protected]>
Acked-by: Mel Gorman <[email protected]>
Cc: "David S. Miller" <[email protected]>
Cc: Andrea Arcangeli <[email protected]>
Cc: Andy Lutomirski <[email protected]>
Cc: Heiko Carstens <[email protected]>
Cc: Hugh Dickins <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jeff Dike <[email protected]>
Cc: Martin Schwidefsky <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: Nadav Amit <[email protected]>
Cc: Rik van Riel <[email protected]>
Cc: Russell King <[email protected]>
Cc: Sergey Senozhatsky <[email protected]>
Cc: Tony Luck <[email protected]>
Cc: Yoshinori Sato <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
This patch is a preparatory patch for solving race problems caused by
TLB batch. For that, we will increase/decrease TLB flush pending count
of mm_struct whenever tlb_[gather|finish]_mmu is called.
Before making it simple, this patch separates architecture specific part
and rename it to arch_tlb_[gather|finish]_mmu and generic part just
calls it.
It shouldn't change any behavior.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Minchan Kim <[email protected]>
Signed-off-by: Nadav Amit <[email protected]>
Acked-by: Mel Gorman <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Russell King <[email protected]>
Cc: Tony Luck <[email protected]>
Cc: Martin Schwidefsky <[email protected]>
Cc: "David S. Miller" <[email protected]>
Cc: Heiko Carstens <[email protected]>
Cc: Yoshinori Sato <[email protected]>
Cc: Jeff Dike <[email protected]>
Cc: Andrea Arcangeli <[email protected]>
Cc: Andy Lutomirski <[email protected]>
Cc: Hugh Dickins <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: Nadav Amit <[email protected]>
Cc: Rik van Riel <[email protected]>
Cc: Sergey Senozhatsky <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
While deferring TLB flushes is a good practice, the reverted patch
caused pending TLB flushes to be checked while the page-table lock is
not taken. As a result, in architectures with weak memory model (PPC),
Linux may miss a memory-barrier, miss the fact TLB flushes are pending,
and cause (in theory) a memory corruption.
Since the alternative of using smp_mb__after_unlock_lock() was
considered a bit open-coded, and the performance impact is expected to
be small, the previous patch is reverted.
This reverts b0943d61b8fa ("mm: numa: defer TLB flush for THP migration
as long as possible").
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Nadav Amit <[email protected]>
Suggested-by: Mel Gorman <[email protected]>
Acked-by: Mel Gorman <[email protected]>
Acked-by: Rik van Riel <[email protected]>
Cc: Minchan Kim <[email protected]>
Cc: Sergey Senozhatsky <[email protected]>
Cc: Andy Lutomirski <[email protected]>
Cc: "David S. Miller" <[email protected]>
Cc: Andrea Arcangeli <[email protected]>
Cc: Heiko Carstens <[email protected]>
Cc: Hugh Dickins <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jeff Dike <[email protected]>
Cc: Martin Schwidefsky <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: Nadav Amit <[email protected]>
Cc: Russell King <[email protected]>
Cc: Tony Luck <[email protected]>
Cc: Yoshinori Sato <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Patch series "fixes of TLB batching races", v6.
It turns out that Linux TLB batching mechanism suffers from various
races. Races that are caused due to batching during reclamation were
recently handled by Mel and this patch-set deals with others. The more
fundamental issue is that concurrent updates of the page-tables allow
for TLB flushes to be batched on one core, while another core changes
the page-tables. This other core may assume a PTE change does not
require a flush based on the updated PTE value, while it is unaware that
TLB flushes are still pending.
This behavior affects KSM (which may result in memory corruption) and
MADV_FREE and MADV_DONTNEED (which may result in incorrect behavior). A
proof-of-concept can easily produce the wrong behavior of MADV_DONTNEED.
Memory corruption in KSM is harder to produce in practice, but was
observed by hacking the kernel and adding a delay before flushing and
replacing the KSM page.
Finally, there is also one memory barrier missing, which may affect
architectures with weak memory model.
This patch (of 7):
Setting and clearing mm->tlb_flush_pending can be performed by multiple
threads, since mmap_sem may only be acquired for read in
task_numa_work(). If this happens, tlb_flush_pending might be cleared
while one of the threads still changes PTEs and batches TLB flushes.
This can lead to the same race between migration and
change_protection_range() that led to the introduction of
tlb_flush_pending. The result of this race was data corruption, which
means that this patch also addresses a theoretically possible data
corruption.
An actual data corruption was not observed, yet the race was was
confirmed by adding assertion to check tlb_flush_pending is not set by
two threads, adding artificial latency in change_protection_range() and
using sysctl to reduce kernel.numa_balancing_scan_delay_ms.
Link: http://lkml.kernel.org/r/[email protected]
Fixes: 20841405940e ("mm: fix TLB flush race between migration, and
change_protection_range")
Signed-off-by: Nadav Amit <[email protected]>
Acked-by: Mel Gorman <[email protected]>
Acked-by: Rik van Riel <[email protected]>
Acked-by: Minchan Kim <[email protected]>
Cc: Andy Lutomirski <[email protected]>
Cc: Hugh Dickins <[email protected]>
Cc: "David S. Miller" <[email protected]>
Cc: Andrea Arcangeli <[email protected]>
Cc: Heiko Carstens <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jeff Dike <[email protected]>
Cc: Martin Schwidefsky <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: Russell King <[email protected]>
Cc: Sergey Senozhatsky <[email protected]>
Cc: Tony Luck <[email protected]>
Cc: Yoshinori Sato <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
huge_add_to_page_cache->add_to_page_cache implicitly unlocks the page
before returning in case of errors.
The error returned was -EEXIST by running UFFDIO_COPY on a non-hole
offset of a VM_SHARED hugetlbfs mapping. It was an userland bug that
triggered it and the kernel must cope with it returning -EEXIST from
ioctl(UFFDIO_COPY) as expected.
page dumped because: VM_BUG_ON_PAGE(!PageLocked(page))
kernel BUG at mm/filemap.c:964!
invalid opcode: 0000 [#1] SMP
CPU: 1 PID: 22582 Comm: qemu-system-x86 Not tainted 4.11.11-300.fc26.x86_64 #1
RIP: unlock_page+0x4a/0x50
Call Trace:
hugetlb_mcopy_atomic_pte+0xc0/0x320
mcopy_atomic+0x96f/0xbe0
userfaultfd_ioctl+0x218/0xe90
do_vfs_ioctl+0xa5/0x600
SyS_ioctl+0x79/0x90
entry_SYSCALL_64_fastpath+0x1a/0xa9
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Andrea Arcangeli <[email protected]>
Tested-by: Maxime Coquelin <[email protected]>
Reviewed-by: Mike Kravetz <[email protected]>
Cc: "Dr. David Alan Gilbert" <[email protected]>
Cc: Mike Rapoport <[email protected]>
Cc: Alexey Perevalov <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
The RDMA subsystem can generate several thousand of these messages per
second eventually leading to a kernel crash. Ratelimit these messages
to prevent this crash.
Doug said:
"I've been carrying a version of this for several kernel versions. I
don't remember when they started, but we have one (and only one) class
of machines: Dell PE R730xd, that generate these errors. When it
happens, without a rate limit, we get rcu timeouts and kernel oopses.
With the rate limit, we just get a lot of annoying kernel messages but
the machine continues on, recovers, and eventually the memory
operations all succeed"
And:
"> Well... why are all these EBUSY's occurring? It sounds inefficient
> (at least) but if it is expected, normal and unavoidable then
> perhaps we should just remove that message altogether?
I don't have an answer to that question. To be honest, I haven't
looked real hard. We never had this at all, then it started out of the
blue, but only on our Dell 730xd machines (and it hits all of them),
but no other classes or brands of machines. And we have our 730xd
machines loaded up with different brands and models of cards (for
instance one dedicated to mlx4 hardware, one for qib, one for mlx5, an
ocrdma/cxgb4 combo, etc), so the fact that it hit all of the machines
meant it wasn't tied to any particular brand/model of RDMA hardware.
To me, it always smelled of a hardware oddity specific to maybe the
CPUs or mainboard chipsets in these machines, so given that I'm not an
mm expert anyway, I never chased it down.
A few other relevant details: it showed up somewhere around 4.8/4.9 or
thereabouts. It never happened before, but the prinkt has been there
since the 3.18 days, so possibly the test to trigger this message was
changed, or something else in the allocator changed such that the
situation started happening on these machines?
And, like I said, it is specific to our 730xd machines (but they are
all identical, so that could mean it's something like their specific
ram configuration is causing the allocator to hit this on these
machine but not on other machines in the cluster, I don't want to say
it's necessarily the model of chipset or CPU, there are other bits of
identicalness between these machines)"
Link: http://lkml.kernel.org/r/499c0f6cc10d6eb829a67f2a4d75b4228a9b356e.1501695897.git.jtoppins@redhat.com
Signed-off-by: Jonathan Toppins <[email protected]>
Reviewed-by: Doug Ledford <[email protected]>
Tested-by: Doug Ledford <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Vlastimil Babka <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: Hillf Danton <[email protected]>
Cc: <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
As Tetsuo points out:
"Commit 385386cff4c6 ("mm: vmstat: move slab statistics from zone to
node counters") broke "Slab:" field of /proc/meminfo . It shows nearly
0kB"
In addition to /proc/meminfo, this problem also affects the slab
counters OOM/allocation failure info dumps, can cause early -ENOMEM from
overcommit protection, and miscalculate image size requirements during
suspend-to-disk.
This is because the patch in question switched the slab counters from
the zone level to the node level, but forgot to update the global
accessor functions to read the aggregate node data instead of the
aggregate zone data.
Use global_node_page_state() to access the global slab counters.
Fixes: 385386cff4c6 ("mm: vmstat: move slab statistics from zone to node counters")
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Reported-by: Tetsuo Handa <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Cc: Josef Bacik <[email protected]>
Cc: Vladimir Davydov <[email protected]>
Cc: Stefan Agner <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Andre Wild reported the following warning:
WARNING: CPU: 2 PID: 1205 at kernel/cpu.c:240 lockdep_assert_cpus_held+0x4c/0x60
Modules linked in:
CPU: 2 PID: 1205 Comm: bash Not tainted 4.13.0-rc2-00022-gfd2b2c57ec20 #10
Hardware name: IBM 2964 N96 702 (z/VM 6.4.0)
task: 00000000701d8100 task.stack: 0000000073594000
Krnl PSW : 0704f00180000000 0000000000145e24 (lockdep_assert_cpus_held+0x4c/0x60)
...
Call Trace:
lockdep_assert_cpus_held+0x42/0x60)
stop_machine_cpuslocked+0x62/0xf0
build_all_zonelists+0x92/0x150
numa_zonelist_order_handler+0x102/0x150
proc_sys_call_handler.isra.12+0xda/0x118
proc_sys_write+0x34/0x48
__vfs_write+0x3c/0x178
vfs_write+0xbc/0x1a0
SyS_write+0x66/0xc0
system_call+0xc4/0x2b0
locks held by bash/1205:
#0: (sb_writers#4){.+.+.+}, at: vfs_write+0xa6/0x1a0
#1: (zl_order_mutex){+.+...}, at: numa_zonelist_order_handler+0x44/0x150
#2: (zonelists_mutex){+.+...}, at: numa_zonelist_order_handler+0xf4/0x150
Last Breaking-Event-Address:
lockdep_assert_cpus_held+0x48/0x60
This can be easily triggered with e.g.
echo n > /proc/sys/vm/numa_zonelist_order
In commit 3f906ba23689a ("mm/memory-hotplug: switch locking to a percpu
rwsem") memory hotplug locking was changed to fix a potential deadlock.
This also switched the stop_machine() invocation within
build_all_zonelists() to stop_machine_cpuslocked() which now expects
that online cpus are locked when being called.
This assumption is not true if build_all_zonelists() is being called
from numa_zonelist_order_handler().
In order to fix this simply add a mem_hotplug_begin()/mem_hotplug_done()
pair to numa_zonelist_order_handler().
Link: http://lkml.kernel.org/r/[email protected]
Fixes: 3f906ba23689a ("mm/memory-hotplug: switch locking to a percpu rwsem")
Signed-off-by: Heiko Carstens <[email protected]>
Reported-by: Andre Wild <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Cc: KAMEZAWA Hiroyuki <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Cc: Vlastimil Babka <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
When a thread is OOM-killed during swap_readpage() operation, an oops
occurs because end_swap_bio_read() is calling wake_up_process() based on
an assumption that the thread which called swap_readpage() is still
alive.
Out of memory: Kill process 525 (polkitd) score 0 or sacrifice child
Killed process 525 (polkitd) total-vm:528128kB, anon-rss:0kB, file-rss:4kB, shmem-rss:0kB
oom_reaper: reaped process 525 (polkitd), now anon-rss:0kB, file-rss:0kB, shmem-rss:0kB
general protection fault: 0000 [#1] SMP DEBUG_PAGEALLOC
Modules linked in: nf_conntrack_netbios_ns nf_conntrack_broadcast ip6t_rpfilter ipt_REJECT nf_reject_ipv4 ip6t_REJECT nf_reject_ipv6 xt_conntrack ip_set nfnetlink ebtable_nat ebtable_broute bridge stp llc ip6table_nat nf_conntrack_ipv6 nf_defrag_ipv6 nf_nat_ipv6 ip6table_mangle ip6table_raw iptable_nat nf_conntrack_ipv4 nf_defrag_ipv4 nf_nat_ipv4 nf_nat nf_conntrack iptable_mangle iptable_raw ebtable_filter ebtables ip6table_filter ip6_tables iptable_filter coretemp ppdev pcspkr vmw_balloon sg shpchp vmw_vmci parport_pc parport i2c_piix4 ip_tables xfs libcrc32c sd_mod sr_mod cdrom ata_generic pata_acpi vmwgfx ahci libahci drm_kms_helper ata_piix syscopyarea sysfillrect sysimgblt fb_sys_fops mptspi scsi_transport_spi ttm e1000 mptscsih drm mptbase i2c_core libata serio_raw
CPU: 0 PID: 0 Comm: swapper/0 Not tainted 4.13.0-rc2-next-20170725 #129
Hardware name: VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform, BIOS 6.00 07/31/2013
task: ffffffffb7c16500 task.stack: ffffffffb7c00000
RIP: 0010:__lock_acquire+0x151/0x12f0
Call Trace:
<IRQ>
lock_acquire+0x59/0x80
_raw_spin_lock_irqsave+0x3b/0x4f
try_to_wake_up+0x3b/0x410
wake_up_process+0x10/0x20
end_swap_bio_read+0x6f/0xf0
bio_endio+0x92/0xb0
blk_update_request+0x88/0x270
scsi_end_request+0x32/0x1c0
scsi_io_completion+0x209/0x680
scsi_finish_command+0xd4/0x120
scsi_softirq_done+0x120/0x140
__blk_mq_complete_request_remote+0xe/0x10
flush_smp_call_function_queue+0x51/0x120
generic_smp_call_function_single_interrupt+0xe/0x20
smp_trace_call_function_single_interrupt+0x22/0x30
smp_call_function_single_interrupt+0x9/0x10
call_function_single_interrupt+0xa7/0xb0
</IRQ>
RIP: 0010:native_safe_halt+0x6/0x10
default_idle+0xe/0x20
arch_cpu_idle+0xa/0x10
default_idle_call+0x1e/0x30
do_idle+0x187/0x200
cpu_startup_entry+0x6e/0x70
rest_init+0xd0/0xe0
start_kernel+0x456/0x477
x86_64_start_reservations+0x24/0x26
x86_64_start_kernel+0xf7/0x11a
secondary_startup_64+0xa5/0xa5
Code: c3 49 81 3f 20 9e 0b b8 41 bc 00 00 00 00 44 0f 45 e2 83 fe 01 0f 87 62 ff ff ff 89 f0 49 8b 44 c7 08 48 85 c0 0f 84 52 ff ff ff <f0> ff 80 98 01 00 00 8b 3d 5a 49 c4 01 45 8b b3 18 0c 00 00 85
RIP: __lock_acquire+0x151/0x12f0 RSP: ffffa01f39e03c50
---[ end trace 6c441db499169b1e ]---
Kernel panic - not syncing: Fatal exception in interrupt
Kernel Offset: 0x36000000 from 0xffffffff81000000 (relocation range: 0xffffffff80000000-0xffffffffbfffffff)
---[ end Kernel panic - not syncing: Fatal exception in interrupt
Fix it by holding a reference to the thread.
[[email protected]: add comment]
Fixes: 23955622ff8d231b ("swap: add block io poll in swapin path")
Signed-off-by: Tetsuo Handa <[email protected]>
Reviewed-by: Shaohua Li <[email protected]>
Cc: Tim Chen <[email protected]>
Cc: Huang Ying <[email protected]>
Cc: Jens Axboe <[email protected]>
Cc: Hugh Dickins <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Mike reported kernel goes oops with ltp:zram03 testcase.
zram: Added device: zram0
zram0: detected capacity change from 0 to 107374182400
BUG: unable to handle kernel paging request at 0000306d61727a77
IP: zs_map_object+0xb9/0x260
PGD 0
P4D 0
Oops: 0000 [#1] SMP
Dumping ftrace buffer:
(ftrace buffer empty)
Modules linked in: zram(E) xfs(E) libcrc32c(E) btrfs(E) xor(E) raid6_pq(E) loop(E) ebtable_filter(E) ebtables(E) ip6table_filter(E) ip6_tables(E) iptable_filter(E) ip_tables(E) x_tables(E) af_packet(E) br_netfilter(E) bridge(E) stp(E) llc(E) iscsi_ibft(E) iscsi_boot_sysfs(E) nls_iso8859_1(E) nls_cp437(E) vfat(E) fat(E) intel_powerclamp(E) coretemp(E) cdc_ether(E) kvm_intel(E) usbnet(E) mii(E) kvm(E) irqbypass(E) crct10dif_pclmul(E) crc32_pclmul(E) crc32c_intel(E) iTCO_wdt(E) ghash_clmulni_intel(E) bnx2(E) iTCO_vendor_support(E) pcbc(E) ioatdma(E) ipmi_ssif(E) aesni_intel(E) i5500_temp(E) i2c_i801(E) aes_x86_64(E) lpc_ich(E) shpchp(E) mfd_core(E) crypto_simd(E) i7core_edac(E) dca(E) glue_helper(E) cryptd(E) ipmi_si(E) button(E) acpi_cpufreq(E) ipmi_devintf(E) pcspkr(E) ipmi_msghandler(E)
nfsd(E) auth_rpcgss(E) nfs_acl(E) lockd(E) grace(E) sunrpc(E) ext4(E) crc16(E) mbcache(E) jbd2(E) sd_mod(E) ata_generic(E) i2c_algo_bit(E) ata_piix(E) drm_kms_helper(E) ahci(E) syscopyarea(E) sysfillrect(E) libahci(E) sysimgblt(E) fb_sys_fops(E) uhci_hcd(E) ehci_pci(E) ttm(E) ehci_hcd(E) libata(E) drm(E) megaraid_sas(E) usbcore(E) sg(E) dm_multipath(E) dm_mod(E) scsi_dh_rdac(E) scsi_dh_emc(E) scsi_dh_alua(E) scsi_mod(E) efivarfs(E) autofs4(E) [last unloaded: zram]
CPU: 6 PID: 12356 Comm: swapon Tainted: G E 4.13.0.g87b2c3f-default #194
Hardware name: IBM System x3550 M3 -[7944K3G]-/69Y5698 , BIOS -[D6E150AUS-1.10]- 12/15/2010
task: ffff880158d2c4c0 task.stack: ffffc90001680000
RIP: 0010:zs_map_object+0xb9/0x260
Call Trace:
zram_bvec_rw.isra.26+0xe8/0x780 [zram]
zram_rw_page+0x6e/0xa0 [zram]
bdev_read_page+0x81/0xb0
do_mpage_readpage+0x51a/0x710
mpage_readpages+0x122/0x1a0
blkdev_readpages+0x1d/0x20
__do_page_cache_readahead+0x1b2/0x270
ondemand_readahead+0x180/0x2c0
page_cache_sync_readahead+0x31/0x50
generic_file_read_iter+0x7e7/0xaf0
blkdev_read_iter+0x37/0x40
__vfs_read+0xce/0x140
vfs_read+0x9e/0x150
SyS_read+0x46/0xa0
entry_SYSCALL_64_fastpath+0x1a/0xa5
Code: 81 e6 00 c0 3f 00 81 fe 00 00 16 00 0f 85 9f 01 00 00 0f b7 13 65 ff 05 5e 07 dc 7e 66 c1 ea 02 81 e2 ff 01 00 00 49 8b 54 d4 08 <8b> 4a 48 41 0f af ce 81 e1 ff 0f 00 00 41 89 c9 48 c7 c3 a0 70
RIP: zs_map_object+0xb9/0x260 RSP: ffffc90001683988
CR2: 0000306d61727a77
He bisected the problem is [1].
After commit cf8e0fedf078 ("mm/zsmalloc: simplify zs_max_alloc_size
handling"), zram doesn't use double pointer for pool->size_class any
more in zs_create_pool so counter function zs_destroy_pool don't need to
free it, either.
Otherwise, it does kfree wrong address and then, kernel goes Oops.
Link: http://lkml.kernel.org/r/20170725062650.GA12134@bbox
Fixes: cf8e0fedf078 ("mm/zsmalloc: simplify zs_max_alloc_size handling")
Signed-off-by: Minchan Kim <[email protected]>
Reported-by: Mike Galbraith <[email protected]>
Tested-by: Mike Galbraith <[email protected]>
Reviewed-by: Sergey Senozhatsky <[email protected]>
Cc: Jerome Marchand <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
gcc-7 produces this warning:
mm/kasan/report.c: In function 'kasan_report':
mm/kasan/report.c:351:3: error: 'info.first_bad_addr' may be used uninitialized in this function [-Werror=maybe-uninitialized]
print_shadow_for_address(info->first_bad_addr);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
mm/kasan/report.c:360:27: note: 'info.first_bad_addr' was declared here
The code seems fine as we only print info.first_bad_addr when there is a
shadow, and we always initialize it in that case, but this is relatively
hard for gcc to figure out after the latest rework.
Adding an intialization to the most likely value together with the other
struct members shuts up that warning.
Fixes: b235b9808664 ("kasan: unify report headers")
Link: https://patchwork.kernel.org/patch/9641417/
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Arnd Bergmann <[email protected]>
Suggested-by: Alexander Potapenko <[email protected]>
Suggested-by: Andrey Ryabinin <[email protected]>
Acked-by: Andrey Ryabinin <[email protected]>
Cc: Dmitry Vyukov <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
When mremap is called with MREMAP_FIXED it unmaps memory at the
destination address without notifying userfaultfd monitor.
If the destination were registered with userfaultfd, the monitor has no
way to distinguish between the old and new ranges and to properly relate
the page faults that would occur in the destination region.
Fixes: 897ab3e0c49e ("userfaultfd: non-cooperative: add event for memory unmaps")
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Mike Rapoport <[email protected]>
Acked-by: Pavel Emelyanov <[email protected]>
Cc: Andrea Arcangeli <[email protected]>
Cc: <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
leaving stale TLB entries
Nadav Amit identified a theoritical race between page reclaim and
mprotect due to TLB flushes being batched outside of the PTL being held.
He described the race as follows:
CPU0 CPU1
---- ----
user accesses memory using RW PTE
[PTE now cached in TLB]
try_to_unmap_one()
==> ptep_get_and_clear()
==> set_tlb_ubc_flush_pending()
mprotect(addr, PROT_READ)
==> change_pte_range()
==> [ PTE non-present - no flush ]
user writes using cached RW PTE
...
try_to_unmap_flush()
The same type of race exists for reads when protecting for PROT_NONE and
also exists for operations that can leave an old TLB entry behind such
as munmap, mremap and madvise.
For some operations like mprotect, it's not necessarily a data integrity
issue but it is a correctness issue as there is a window where an
mprotect that limits access still allows access. For munmap, it's
potentially a data integrity issue although the race is massive as an
munmap, mmap and return to userspace must all complete between the
window when reclaim drops the PTL and flushes the TLB. However, it's
theoritically possible so handle this issue by flushing the mm if
reclaim is potentially currently batching TLB flushes.
Other instances where a flush is required for a present pte should be ok
as either the page lock is held preventing parallel reclaim or a page
reference count is elevated preventing a parallel free leading to
corruption. In the case of page_mkclean there isn't an obvious path
that userspace could take advantage of without using the operations that
are guarded by this patch. Other users such as gup as a race with
reclaim looks just at PTEs. huge page variants should be ok as they
don't race with reclaim. mincore only looks at PTEs. userfault also
should be ok as if a parallel reclaim takes place, it will either fault
the page back in or read some of the data before the flush occurs
triggering a fault.
Note that a variant of this patch was acked by Andy Lutomirski but this
was for the x86 parts on top of his PCID work which didn't make the 4.13
merge window as expected. His ack is dropped from this version and
there will be a follow-on patch on top of PCID that will include his
ack.
[[email protected]: tweak comments]
[[email protected]: fix spello]
Link: http://lkml.kernel.org/r/[email protected]
Reported-by: Nadav Amit <[email protected]>
Signed-off-by: Mel Gorman <[email protected]>
Cc: Andy Lutomirski <[email protected]>
Cc: <[email protected]> [v4.4+]
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Commit 9a291a7c9428 ("mm/hugetlb: report -EHWPOISON not -EFAULT when
FOLL_HWPOISON is specified") causes __get_user_pages to ignore certain
errors from follow_hugetlb_page. After such error, __get_user_pages
subsequently calls faultin_page on the same VMA and start address that
follow_hugetlb_page failed on instead of returning the error immediately
as it should.
In follow_hugetlb_page, when hugetlb_fault returns a value covered under
VM_FAULT_ERROR, follow_hugetlb_page returns it without setting nr_pages
to 0 as __get_user_pages expects in this case, which causes the
following to happen in __get_user_pages: the "while (nr_pages)" check
succeeds, we skip the "if (!vma..." check because we got a VMA the last
time around, we find no page with follow_page_mask, and we call
faultin_page, which calls hugetlb_fault for the second time.
This issue also slightly changes how __get_user_pages works. Before, it
only returned error if it had made no progress (i = 0). But now,
follow_hugetlb_page can clobber "i" with an error code since its new
return path doesn't check for progress. So if "i" is nonzero before a
failing call to follow_hugetlb_page, that indication of progress is lost
and __get_user_pages can return error even if some pages were
successfully pinned.
To fix this, change follow_hugetlb_page so that it updates nr_pages,
allowing __get_user_pages to fail immediately and restoring the "error
only if no progress" behavior to __get_user_pages.
Tested that __get_user_pages returns when expected on error from
hugetlb_fault in follow_hugetlb_page.
Fixes: 9a291a7c9428 ("mm/hugetlb: report -EHWPOISON not -EFAULT when FOLL_HWPOISON is specified")
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Daniel Jordan <[email protected]>
Acked-by: Punit Agrawal <[email protected]>
Cc: Andrea Arcangeli <[email protected]>
Cc: "Aneesh Kumar K.V" <[email protected]>
Cc: Gerald Schaefer <[email protected]>
Cc: James Morse <[email protected]>
Cc: "Kirill A. Shutemov" <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Mike Kravetz <[email protected]>
Cc: Naoya Horiguchi <[email protected]>
Cc: zhong jiang <[email protected]>
Cc: <[email protected]> [4.12.x]
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs
Pull ->s_options removal from Al Viro:
"Preparations for fsmount/fsopen stuff (coming next cycle). Everything
gets moved to explicit ->show_options(), killing ->s_options off +
some cosmetic bits around fs/namespace.c and friends. Basically, the
stuff needed to work with fsmount series with minimum of conflicts
with other work.
It's not strictly required for this merge window, but it would reduce
the PITA during the coming cycle, so it would be nice to have those
bits and pieces out of the way"
* 'work.mount' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
isofs: Fix isofs_show_options()
VFS: Kill off s_options and helpers
orangefs: Implement show_options
9p: Implement show_options
isofs: Implement show_options
afs: Implement show_options
affs: Implement show_options
befs: Implement show_options
spufs: Implement show_options
bpf: Implement show_options
ramfs: Implement show_options
pstore: Implement show_options
omfs: Implement show_options
hugetlbfs: Implement show_options
VFS: Don't use save/replace_mount_options if not using generic_show_options
VFS: Provide empty name qstr
VFS: Make get_filesystem() return the affected filesystem
VFS: Clean up whitespace in fs/namespace.c and fs/super.c
Provide a function to create a NUL-terminated string from unterminated data
|
|
Jörn Engel noticed that the expand_upwards() function might not return
-ENOMEM in case the requested address is (unsigned long)-PAGE_SIZE and
if the architecture didn't defined TASK_SIZE as multiple of PAGE_SIZE.
Affected architectures are arm, frv, m68k, blackfin, h8300 and xtensa
which all define TASK_SIZE as 0xffffffff, but since none of those have
an upwards-growing stack we currently have no actual issue.
Nevertheless let's fix this just in case any of the architectures with
an upward-growing stack (currently parisc, metag and partly ia64) define
TASK_SIZE similar.
Link: http://lkml.kernel.org/r/[email protected]
Fixes: bd726c90b6b8 ("Allow stack to grow up to address space limit")
Signed-off-by: Helge Deller <[email protected]>
Reported-by: Jörn Engel <[email protected]>
Cc: Hugh Dickins <[email protected]>
Cc: Oleg Nesterov <[email protected]>
Cc: <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Currently the writeback statistics code uses a percpu counters to hold
various statistics. Furthermore we have 2 families of functions - those
which disable local irq and those which doesn't and whose names begin
with double underscore. However, they both end up calling
__add_wb_stats which in turn calls percpu_counter_add_batch which is
already irq-safe.
Exploiting this fact allows to eliminated the __wb_* functions since
they don't add any further protection than we already have.
Furthermore, refactor the wb_* function to call __add_wb_stat directly
without the irq-disabling dance. This will likely result in better
runtime of code which deals with modifying the stat counters.
While at it also document why percpu_counter_add_batch is in fact
preempt and irq-safe since at least 3 people got confused.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Nikolay Borisov <[email protected]>
Acked-by: Tejun Heo <[email protected]>
Reviewed-by: Jan Kara <[email protected]>
Cc: Josef Bacik <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: Jeff Layton <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Page migration (for memory hotplug, soft_offline_page or mbind) needs to
allocate a new memory. This can trigger an oom killer if the target
memory is depleated. Although quite unlikely, still possible,
especially for the memory hotplug (offlining of memoery).
Up to now we didn't really have reasonable means to back off.
__GFP_NORETRY can fail just too easily and __GFP_THISNODE sticks to a
single node and that is not suitable for all callers.
But now that we have __GFP_RETRY_MAYFAIL we should use it. It is
preferable to fail the migration than disrupt the system by killing some
processes.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Michal Hocko <[email protected]>
Acked-by: Vlastimil Babka <[email protected]>
Cc: Alex Belits <[email protected]>
Cc: Chris Wilson <[email protected]>
Cc: Christoph Hellwig <[email protected]>
Cc: Darrick J. Wong <[email protected]>
Cc: David Daney <[email protected]>
Cc: Johannes Weiner <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: NeilBrown <[email protected]>
Cc: Ralf Baechle <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Now that __GFP_RETRY_MAYFAIL has a reasonable semantic regardless of the
request size we can drop the hackish implementation for !costly orders.
__GFP_RETRY_MAYFAIL retries as long as the reclaim makes a forward
progress and backs of when we are out of memory for the requested size.
Therefore we do not need to enforce__GFP_NORETRY for !costly orders just
to silent the oom killer anymore.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Michal Hocko <[email protected]>
Acked-by: Vlastimil Babka <[email protected]>
Cc: Alex Belits <[email protected]>
Cc: Chris Wilson <[email protected]>
Cc: Christoph Hellwig <[email protected]>
Cc: Darrick J. Wong <[email protected]>
Cc: David Daney <[email protected]>
Cc: Johannes Weiner <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: NeilBrown <[email protected]>
Cc: Ralf Baechle <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
semantic
__GFP_REPEAT was designed to allow retry-but-eventually-fail semantic to
the page allocator. This has been true but only for allocations
requests larger than PAGE_ALLOC_COSTLY_ORDER. It has been always
ignored for smaller sizes. This is a bit unfortunate because there is
no way to express the same semantic for those requests and they are
considered too important to fail so they might end up looping in the
page allocator for ever, similarly to GFP_NOFAIL requests.
Now that the whole tree has been cleaned up and accidental or misled
usage of __GFP_REPEAT flag has been removed for !costly requests we can
give the original flag a better name and more importantly a more useful
semantic. Let's rename it to __GFP_RETRY_MAYFAIL which tells the user
that the allocator would try really hard but there is no promise of a
success. This will work independent of the order and overrides the
default allocator behavior. Page allocator users have several levels of
guarantee vs. cost options (take GFP_KERNEL as an example)
- GFP_KERNEL & ~__GFP_RECLAIM - optimistic allocation without _any_
attempt to free memory at all. The most light weight mode which even
doesn't kick the background reclaim. Should be used carefully because
it might deplete the memory and the next user might hit the more
aggressive reclaim
- GFP_KERNEL & ~__GFP_DIRECT_RECLAIM (or GFP_NOWAIT)- optimistic
allocation without any attempt to free memory from the current
context but can wake kswapd to reclaim memory if the zone is below
the low watermark. Can be used from either atomic contexts or when
the request is a performance optimization and there is another
fallback for a slow path.
- (GFP_KERNEL|__GFP_HIGH) & ~__GFP_DIRECT_RECLAIM (aka GFP_ATOMIC) -
non sleeping allocation with an expensive fallback so it can access
some portion of memory reserves. Usually used from interrupt/bh
context with an expensive slow path fallback.
- GFP_KERNEL - both background and direct reclaim are allowed and the
_default_ page allocator behavior is used. That means that !costly
allocation requests are basically nofail but there is no guarantee of
that behavior so failures have to be checked properly by callers
(e.g. OOM killer victim is allowed to fail currently).
- GFP_KERNEL | __GFP_NORETRY - overrides the default allocator behavior
and all allocation requests fail early rather than cause disruptive
reclaim (one round of reclaim in this implementation). The OOM killer
is not invoked.
- GFP_KERNEL | __GFP_RETRY_MAYFAIL - overrides the default allocator
behavior and all allocation requests try really hard. The request
will fail if the reclaim cannot make any progress. The OOM killer
won't be triggered.
- GFP_KERNEL | __GFP_NOFAIL - overrides the default allocator behavior
and all allocation requests will loop endlessly until they succeed.
This might be really dangerous especially for larger orders.
Existing users of __GFP_REPEAT are changed to __GFP_RETRY_MAYFAIL
because they already had their semantic. No new users are added.
__alloc_pages_slowpath is changed to bail out for __GFP_RETRY_MAYFAIL if
there is no progress and we have already passed the OOM point.
This means that all the reclaim opportunities have been exhausted except
the most disruptive one (the OOM killer) and a user defined fallback
behavior is more sensible than keep retrying in the page allocator.
[[email protected]: fix arch/sparc/kernel/mdesc.c]
[[email protected]: semantic fix]
Link: http://lkml.kernel.org/r/[email protected]
[[email protected]: address other thing spotted by Vlastimil]
Link: http://lkml.kernel.org/r/[email protected]
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Michal Hocko <[email protected]>
Acked-by: Vlastimil Babka <[email protected]>
Cc: Alex Belits <[email protected]>
Cc: Chris Wilson <[email protected]>
Cc: Christoph Hellwig <[email protected]>
Cc: Darrick J. Wong <[email protected]>
Cc: David Daney <[email protected]>
Cc: Johannes Weiner <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: NeilBrown <[email protected]>
Cc: Ralf Baechle <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
With gcc 4.1.2:
mm/memory.o: In function `create_huge_pmd':
memory.c:(.text+0x93e): undefined reference to `do_huge_pmd_anonymous_page'
Interestingly, create_huge_pmd() is emitted in the assembler output, but
never called.
Converting transparent_hugepage_enabled() from a macro to a static
inline function reduced the ability of the compiler to remove unused
code.
Fix this by marking create_huge_pmd() inline.
Fixes: 16981d763501c0e0 ("mm: improve readability of transparent_hugepage_enabled()")
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Geert Uytterhoeven <[email protected]>
Acked-by: Arnd Bergmann <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
The helper function get_wild_bug_type() does not need to be in global
scope, so make it static.
Cleans up sparse warning:
"symbol 'get_wild_bug_type' was not declared. Should it be static?"
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Colin Ian King <[email protected]>
Acked-by: Dmitry Vyukov <[email protected]>
Cc: Andrey Ryabinin <[email protected]>
Cc: Alexander Potapenko <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
They return positive value, that is, true, if non-zero value is found.
Rename them to reduce confusion.
Link: http://lkml.kernel.org/r/20170516012350.GA16015@js1304-desktop
Signed-off-by: Joonsoo Kim <[email protected]>
Cc: Andrey Ryabinin <[email protected]>
Cc: Alexander Potapenko <[email protected]>
Cc: Dmitry Vyukov <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
KASAN doesn't happen work with memory hotplug because hotplugged memory
doesn't have any shadow memory. So any access to hotplugged memory
would cause a crash on shadow check.
Use memory hotplug notifier to allocate and map shadow memory when the
hotplugged memory is going online and free shadow after the memory
offlined.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Andrey Ryabinin <[email protected]>
Cc: "H. Peter Anvin" <[email protected]>
Cc: Alexander Potapenko <[email protected]>
Cc: Catalin Marinas <[email protected]>
Cc: Dmitry Vyukov <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Cc: Will Deacon <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
For some unaligned memory accesses we have to check additional byte of
the shadow memory. Currently we load that byte speculatively to have
only single load + branch on the optimistic fast path.
However, this approach has some downsides:
- It's unaligned access, so this prevents porting KASAN on
architectures which doesn't support unaligned accesses.
- We have to map additional shadow page to prevent crash if speculative
load happens near the end of the mapped memory. This would
significantly complicate upcoming memory hotplug support.
I wasn't able to notice any performance degradation with this patch. So
these speculative loads is just a pain with no gain, let's remove them.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Andrey Ryabinin <[email protected]>
Acked-by: Dmitry Vyukov <[email protected]>
Cc: Alexander Potapenko <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Catalin Marinas <[email protected]>
Cc: Will Deacon <[email protected]>
Cc: "H. Peter Anvin" <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Cc: Ingo Molnar <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
There is missing optimization in zero_p4d_populate() that can save some
memory when mapping zero shadow. Implement it like as others.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Joonsoo Kim <[email protected]>
Acked-by: Andrey Ryabinin <[email protected]>
Cc: "Kirill A . Shutemov" <[email protected]>
Cc: Alexander Potapenko <[email protected]>
Cc: Dmitry Vyukov <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Commit 40f9fb8cffc6 ("mm/zsmalloc: support allocating obj with size of
ZS_MAX_ALLOC_SIZE") fixes a size calculation error that prevented
zsmalloc to allocate an object of the maximal size (ZS_MAX_ALLOC_SIZE).
I think however the fix is unneededly complicated.
This patch replaces the dynamic calculation of zs_size_classes at init
time by a compile time calculation that uses the DIV_ROUND_UP() macro
already used in get_size_class_index().
[[email protected]: use min_t]
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Jerome Marchand <[email protected]>
Acked-by: Minchan Kim <[email protected]>
Cc: Sergey Senozhatsky <[email protected]>
Cc: Mahendran Ganesh <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Andrey reported a potential deadlock with the memory hotplug lock and
the cpu hotplug lock.
The reason is that memory hotplug takes the memory hotplug lock and then
calls stop_machine() which calls get_online_cpus(). That's the reverse
lock order to get_online_cpus(); get_online_mems(); in mm/slub_common.c
The problem has been there forever. The reason why this was never
reported is that the cpu hotplug locking had this homebrewn recursive
reader writer semaphore construct which due to the recursion evaded the
full lock dep coverage. The memory hotplug code copied that construct
verbatim and therefor has similar issues.
Three steps to fix this:
1) Convert the memory hotplug locking to a per cpu rwsem so the
potential issues get reported proper by lockdep.
2) Lock the online cpus in mem_hotplug_begin() before taking the memory
hotplug rwsem and use stop_machine_cpuslocked() in the page_alloc
code to avoid recursive locking.
3) The cpu hotpluck locking in #2 causes a recursive locking of the cpu
hotplug lock via __offline_pages() -> lru_add_drain_all(). Solve this
by invoking lru_add_drain_all_cpuslocked() instead.
Link: http://lkml.kernel.org/r/[email protected]
Reported-by: Andrey Ryabinin <[email protected]>
Signed-off-by: Thomas Gleixner <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Acked-by: Vlastimil Babka <[email protected]>
Cc: Vladimir Davydov <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Davidlohr Bueso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
The rework of the cpu hotplug locking unearthed potential deadlocks with
the memory hotplug locking code.
The solution for these is to rework the memory hotplug locking code as
well and take the cpu hotplug lock before the memory hotplug lock in
mem_hotplug_begin(), but this will cause a recursive locking of the cpu
hotplug lock when the memory hotplug code calls lru_add_drain_all().
Split out the inner workings of lru_add_drain_all() into
lru_add_drain_all_cpuslocked() so this function can be invoked from the
memory hotplug code with the cpu hotplug lock held.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Thomas Gleixner <[email protected]>
Reported-by: Andrey Ryabinin <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Acked-by: Vlastimil Babka <[email protected]>
Cc: Vladimir Davydov <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Davidlohr Bueso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Use rlimit() helper instead of manually writing whole chain from current
task to rlim_cur.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Krzysztof Opasiak <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
list_lru_count_node() iterates over all memcgs to get the total number of
entries on the node but it can race with memcg_drain_all_list_lrus(),
which migrates the entries from a dead cgroup to another. This can return
incorrect number of entries from list_lru_count_node().
Fix this by keeping track of entries per node and simply return it in
list_lru_count_node().
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Sahitya Tummala <[email protected]>
Acked-by: Vladimir Davydov <[email protected]>
Cc: Jan Kara <[email protected]>
Cc: Alexander Polakov <[email protected]>
Cc: Al Viro <[email protected]>
Cc: <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
expand_stack(vma) fails if address < stack_guard_gap even if there is no
vma->vm_prev. I don't think this makes sense, and we didn't do this
before the recent commit 1be7107fbe18 ("mm: larger stack guard gap,
between vmas").
We do not need a gap in this case, any address is fine as long as
security_mmap_addr() doesn't object.
This also simplifies the code, we know that address >= prev->vm_end and
thus underflow is not possible.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Oleg Nesterov <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Cc: Hugh Dickins <[email protected]>
Cc: Larry Woodman <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Commit 1be7107fbe18 ("mm: larger stack guard gap, between vmas") has
introduced a regression in some rust and Java environments which are
trying to implement their own stack guard page. They are punching a new
MAP_FIXED mapping inside the existing stack Vma.
This will confuse expand_{downwards,upwards} into thinking that the
stack expansion would in fact get us too close to an existing non-stack
vma which is a correct behavior wrt safety. It is a real regression on
the other hand.
Let's work around the problem by considering PROT_NONE mapping as a part
of the stack. This is a gros hack but overflowing to such a mapping
would trap anyway an we only can hope that usespace knows what it is
doing and handle it propely.
Fixes: 1be7107fbe18 ("mm: larger stack guard gap, between vmas")
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Michal Hocko <[email protected]>
Debugged-by: Vlastimil Babka <[email protected]>
Cc: Ben Hutchings <[email protected]>
Cc: Willy Tarreau <[email protected]>
Cc: Oleg Nesterov <[email protected]>
Cc: Rik van Riel <[email protected]>
Cc: Hugh Dickins <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
presently pages in the balloon device have random value, and these pages
will be scanned by ksmd on the host. They usually cannot be merged.
Enqueue zero pages will resolve this problem.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: zhenwei.pi <[email protected]>
Cc: Gioh Kim <[email protected]>
Cc: Vlastimil Babka <[email protected]>
Cc: Minchan Kim <[email protected]>
Cc: Konstantin Khlebnikov <[email protected]>
Cc: Rafael Aquini <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
The align_offset parameter is used by bitmap_find_next_zero_area_off()
to represent the offset of map's base from the previous alignment
boundary; the function ensures that the returned index, plus the
align_offset, honors the specified align_mask.
The logic introduced by commit b5be83e308f7 ("mm: cma: align to physical
address, not CMA region position") has the cma driver calculate the
offset to the *next* alignment boundary. In most cases, the base
alignment is greater than that specified when making allocations,
resulting in a zero offset whether we align up or down. In the example
given with the commit, the base alignment (8MB) was half the requested
alignment (16MB) so the math also happened to work since the offset is
8MB in both directions. However, when requesting allocations with an
alignment greater than twice that of the base, the returned index would
not be correctly aligned.
Also, the align_order arguments of cma_bitmap_aligned_mask() and
cma_bitmap_aligned_offset() should not be negative so the argument type
was made unsigned.
Fixes: b5be83e308f7 ("mm: cma: align to physical address, not CMA region position")
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Angus Clark <[email protected]>
Signed-off-by: Doug Berger <[email protected]>
Acked-by: Gregory Fong <[email protected]>
Cc: Doug Berger <[email protected]>
Cc: Angus Clark <[email protected]>
Cc: Laura Abbott <[email protected]>
Cc: Vlastimil Babka <[email protected]>
Cc: Greg Kroah-Hartman <[email protected]>
Cc: Lucas Stach <[email protected]>
Cc: Catalin Marinas <[email protected]>
Cc: Shiraz Hashim <[email protected]>
Cc: Jaewon Kim <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
__remove_zone() sets up up zone_type, but never uses it for anything.
This does not cause a warning, due to the (necessary) use of
-Wno-unused-but-set-variable. However, it's noise, so just delete it.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: John Hubbard <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
get_cpu_var() disables preemption and returns the per-CPU version of the
variable. Disabling preemption is useful to ensure atomic access to the
variable within the critical section.
In this case however, after the per-CPU version of the variable is
obtained the ->free_lock is acquired. For that reason it seems the raw
accessor could be used. It only seems that ->slots_ret should be
retested (because with disabled preemption this variable can not be set
to NULL otherwise).
This popped up during PREEMPT-RT testing because it tries to take
spinlocks in a preempt disabled section. In RT, spinlocks can sleep.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Cc: Tim Chen <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Cc: Ying Huang <[email protected]>
Cc: Hugh Dickins <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Since current_order starts as MAX_ORDER-1 and is then only decremented,
the second half of the loop condition seems superfluous. However, if
order is 0, we may decrement current_order past 0, making it UINT_MAX.
This is obviously too subtle ([1], [2]).
Since we need to add some comment anyway, change the two variables to
signed, making the counting-down for loop look more familiar, and
apparently also making gcc generate slightly smaller code.
[1] https://lkml.org/lkml/2016/6/20/493
[2] https://lkml.org/lkml/2017/6/19/345
[[email protected]: fix up reject fixupping]
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Rasmus Villemoes <[email protected]>
Reported-by: Hao Lee <[email protected]>
Acked-by: Wei Yang <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Acked-by: Vlastimil Babka <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
pagetypeinfo_showmixedcount_print is found to take a lot of time to
complete and it does this holding the zone lock and disabling
interrupts. In some cases it is found to take more than a second (On a
2.4GHz,8Gb RAM,arm64 cpu).
Avoid taking the zone lock similar to what is done by read_page_owner,
which means possibility of inaccurate results.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Vinayak Menon <[email protected]>
Acked-by: Vlastimil Babka <[email protected]>
Cc: Joonsoo Kim <[email protected]>
Cc: zhongjiang <[email protected]>
Cc: Sergey Senozhatsky <[email protected]>
Cc: Sudip Mukherjee <[email protected]>
Cc: Johannes Weiner <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Sebastian Andrzej Siewior <[email protected]>
Cc: David Rientjes <[email protected]>
Cc: Minchan Kim <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
new_page is yet another duplication of the migration callback which has
to handle hugetlb migration specially. We can safely use the generic
new_page_nodemask for the same purpose.
Please note that gigantic hugetlb pages do not need any special handling
because alloc_huge_page_nodemask will make sure to check pages in all
per node pools. The reason this was done previously was that
alloc_huge_page_node treated NO_NUMA_NODE and a specific node
differently and so alloc_huge_page_node(nid) would check on this
specific node.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Michal Hocko <[email protected]>
Acked-by: Vlastimil Babka <[email protected]>
Reported-by: Vlastimil Babka <[email protected]>
Reviewed-by: Mike Kravetz <[email protected]>
Tested-by: Mike Kravetz <[email protected]>
Cc: Naoya Horiguchi <[email protected]>
Cc: Mel Gorman <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
alloc_huge_page_nodemask tries to allocate from any numa node in the
allowed node mask starting from lower numa nodes. This might lead to
filling up those low NUMA nodes while others are not used. We can
reduce this risk by introducing a concept of the preferred node similar
to what we have in the regular page allocator. We will start allocating
from the preferred nid and then iterate over all allowed nodes in the
zonelist order until we try them all.
This is mimicing the page allocator logic except it operates on per-node
mempools. dequeue_huge_page_vma already does this so distill the
zonelist logic into a more generic dequeue_huge_page_nodemask and use it
in alloc_huge_page_nodemask.
This will allow us to use proper per numa distance fallback also for
alloc_huge_page_node which can use alloc_huge_page_nodemask now and we
can get rid of alloc_huge_page_node helper which doesn't have any user
anymore.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Michal Hocko <[email protected]>
Acked-by: Vlastimil Babka <[email protected]>
Reviewed-by: Mike Kravetz <[email protected]>
Tested-by: Mike Kravetz <[email protected]>
Cc: Naoya Horiguchi <[email protected]>
Cc: Mel Gorman <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Patch series "mm, hugetlb: allow proper node fallback dequeue".
While working on a hugetlb migration issue addressed in a separate
patchset[1] I have noticed that the hugetlb allocations from the
preallocated pool are quite subotimal.
[1] //lkml.kernel.org/r/[email protected]
There is no fallback mechanism implemented and no notion of preferred
node. I have tried to work around it but Vlastimil was right to push
back for a more robust solution. It seems that such a solution is to
reuse zonelist approach we use for the page alloctor.
This series has 3 patches. The first one tries to make hugetlb
allocation layers more clear. The second one implements the zonelist
hugetlb pool allocation and introduces a preferred node semantic which
is used by the migration callbacks. The last patch is a clean up.
This patch (of 3):
Hugetlb allocation path for fresh huge pages is unnecessarily complex
and it mixes different interfaces between layers.
__alloc_buddy_huge_page is the central place to perform a new
allocation. It checks for the hugetlb overcommit and then relies on
__hugetlb_alloc_buddy_huge_page to invoke the page allocator. This is
all good except that __alloc_buddy_huge_page pushes vma and address down
the callchain and so __hugetlb_alloc_buddy_huge_page has to deal with
two different allocation modes - one for memory policy and other node
specific (or to make it more obscure node non-specific) requests.
This just screams for a reorganization.
This patch pulls out all the vma specific handling up to
__alloc_buddy_huge_page_with_mpol where it belongs.
__alloc_buddy_huge_page will get nodemask argument and
__hugetlb_alloc_buddy_huge_page will become a trivial wrapper over the
page allocator.
In short:
__alloc_buddy_huge_page_with_mpol - memory policy handling
__alloc_buddy_huge_page - overcommit handling and accounting
__hugetlb_alloc_buddy_huge_page - page allocator layer
Also note that __hugetlb_alloc_buddy_huge_page and its cpuset retry loop
is not really needed because the page allocator already handles the
cpusets update.
Finally __hugetlb_alloc_buddy_huge_page had a special case for node
specific allocations (when no policy is applied and there is a node
given). This has relied on __GFP_THISNODE to not fallback to a different
node. alloc_huge_page_node is the only caller which relies on this
behavior so move the __GFP_THISNODE there.
Not only does this remove quite some code it also should make those
layers easier to follow and clear wrt responsibilities.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Michal Hocko <[email protected]>
Acked-by: Vlastimil Babka <[email protected]>
Reviewed-by: Mike Kravetz <[email protected]>
Tested-by: Mike Kravetz <[email protected]>
Cc: Naoya Horiguchi <[email protected]>
Cc: Mel Gorman <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
During the debugging of the problem described in
https://lkml.org/lkml/2017/5/17/542 and fixed by Tetsuo Handa in
https://lkml.org/lkml/2017/5/19/383 , I've found that the existing debug
output is not really useful to understand issues related to the oom
reaper.
So, I assume, that adding some tracepoints might help with debugging of
similar issues.
Trace the following events:
1) a process is marked as an oom victim,
2) a process is added to the oom reaper list,
3) the oom reaper starts reaping process's mm,
4) the oom reaper finished reaping,
5) the oom reaper skips reaping.
How it works in practice? Below is an example which show how the problem
mentioned above can be found: one process is added twice to the
oom_reaper list:
$ cd /sys/kernel/debug/tracing
$ echo "oom:mark_victim" > set_event
$ echo "oom:wake_reaper" >> set_event
$ echo "oom:skip_task_reaping" >> set_event
$ echo "oom:start_task_reaping" >> set_event
$ echo "oom:finish_task_reaping" >> set_event
$ cat trace_pipe
allocate-502 [001] .... 91.836405: mark_victim: pid=502
allocate-502 [001] .N.. 91.837356: wake_reaper: pid=502
allocate-502 [000] .N.. 91.871149: wake_reaper: pid=502
oom_reaper-23 [000] .... 91.871177: start_task_reaping: pid=502
oom_reaper-23 [000] .N.. 91.879511: finish_task_reaping: pid=502
oom_reaper-23 [000] .... 91.879580: skip_task_reaping: pid=502
Link: http://lkml.kernel.org/r/20170530185231.GA13412@castle
Signed-off-by: Roman Gushchin <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Cc: Tetsuo Handa <[email protected]>
Cc: Johannes Weiner <[email protected]>
Cc: Vladimir Davydov <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
MADV_FREE is identical to MADV_DONTNEED from the point of view of uffd
monitor. The monitor has to stop handling #PF events in the range being
freed. We are reusing userfaultfd_remove callback along with the logic
required to re-get and re-validate the VMA which may change or disappear
because userfaultfd_remove releases mmap_sem.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Mike Rapoport <[email protected]>
Cc: Andrea Arcangeli <[email protected]>
Cc: Hillf Danton <[email protected]>
Cc: Pavel Emelyanov <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
The condition checking for THP straddling end of invalidated range is
wrong - it checks 'index' against 'end' but 'index' has been already
advanced to point to the end of THP and thus the condition can never be
true. As a result THP straddling 'end' has been fully invalidated.
Given the nature of invalidate_mapping_pages(), this could be only
performance issue. In fact, we are lucky the condition is wrong because
if it was ever true, we'd leave locked page behind.
Fix the condition checking for THP straddling 'end' and also properly
unlock the page. Also update the comment before the condition to
explain why we decide not to invalidate the page as it was not clear to
me and I had to ask Kirill.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Jan Kara <[email protected]>
Acked-by: Kirill A. Shutemov <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|