page_mapcount_is_zero() calculates accurately how many mappings a hugepage
has in order to check against 0 only. This is a waste of cpu time. We
can do this via page_not_mapped() to save some possible atomic_read
cycles. Remove the function page_mapcount_is_zero() as it's not used
anymore and move page_not_mapped() above try_to_unmap() to avoid
identifier undeclared compilation error.
Link: https://lkml.kernel.org/r/20210130084904.35307-1-linmiaohe@huawei.com
Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Patch series "mm/memory_hotplug: Pre-validate the address range with platform", v5.
This series adds a mechanism allowing platforms to weigh in and
prevalidate incoming address range before proceeding further with the
memory hotplug. This helps prevent potential platform errors for the
given address range, down the hotplug call chain, which inevitably fails
the hotplug itself.
This mechanism was suggested by David Hildenbrand during another
discussion with respect to a memory hotplug fix on arm64 platform.
https://lore.kernel.org/linux-arm-kernel/1600332402-30123-1-git-send-email-anshuman.khandual@arm.com/
This mechanism focuses on the addressibility aspect and not [sub] section
alignment aspect. Hence check_hotplug_memory_range() and check_pfn_span()
have been left unchanged.
This patch (of 4):
This introduces mhp_range_allowed() which can be called in various memory
hotplug paths to prevalidate the address range which is being added, with
the platform. Then mhp_range_allowed() calls mhp_get_pluggable_range()
which provides applicable address range depending on whether linear
mapping is required or not. For ranges that require linear mapping, it
calls a new arch callback arch_get_mappable_range() which the platform can
override. So the new callback, in turn provides the platform an
opportunity to configure acceptable memory hotplug address ranges in case
there are constraints.
This mechanism will help prevent platform specific errors deep down during
hotplug calls. This drops now redundant
check_hotplug_memory_addressable() check in __add_pages() but instead adds
a VM_BUG_ON() check which would ensure that the range has been validated
with mhp_range_allowed() earlier in the call chain. Besides
mhp_get_pluggable_range() also can be used by potential memory hotplug
callers to avail the allowed physical range which would go through on a
given platform.
This does not really add any new range check in generic memory hotplug but
instead compensates for lost checks in arch_add_memory() where applicable
and check_hotplug_memory_addressable(), with unified mhp_range_allowed().
[akpm@linux-foundation.org: make pagemap_range() return -EINVAL when mhp_range_allowed() fails]
Link: https://lkml.kernel.org/r/1612149902-7867-1-git-send-email-anshuman.khandual@arm.com
Link: https://lkml.kernel.org/r/1612149902-7867-2-git-send-email-anshuman.khandual@arm.com
Signed-off-by: Anshuman Khandual <anshuman.khandual@arm.com>
Suggested-by: David Hildenbrand <david@redhat.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Oscar Salvador <osalvador@suse.de>
Cc: Heiko Carstens <hca@linux.ibm.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Vasily Gorbik <gor@linux.ibm.com> # s390
Cc: Will Deacon <will@kernel.org>
Cc: Ard Biesheuvel <ardb@kernel.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Jason Wang <jasowang@redhat.com>
Cc: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Pankaj Gupta <pankaj.gupta@cloud.ionos.com>
Cc: Pankaj Gupta <pankaj.gupta.linux@gmail.com>
Cc: teawater <teawaterz@linux.alibaba.com>
Cc: Wei Yang <richard.weiyang@linux.alibaba.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
While pfn_to_online_page() is able to determine pfn_valid() at subsection
granularity it is not able to reliably determine if a given pfn is also
online if the section is mixes ZONE_{NORMAL,MOVABLE} with ZONE_DEVICE.
This means that pfn_to_online_page() may return invalid @page objects.
For example with a memory map like:
100000000-1fbffffff : System RAM
142000000-143002e16 : Kernel code
143200000-143713fff : Kernel rodata
143800000-143b15b7f : Kernel data
144227000-144ffffff : Kernel bss
1fc000000-2fbffffff : Persistent Memory (legacy)
1fc000000-2fbffffff : namespace0.0
This command:
echo 0x1fc000000 > /sys/devices/system/memory/soft_offline_page
...succeeds when it should fail. When it succeeds it touches an
uninitialized page and may crash or cause other damage (see
dissolve_free_huge_page()).
While the memory map above is contrived via the memmap=ss!nn kernel
command line option, the collision happens in practice on shipping
platforms. The memory controller resources that decode spans of physical
address space are a limited resource. One technique platform-firmware
uses to conserve those resources is to share a decoder across 2 devices to
keep the address range contiguous. Unfortunately the unit of operation of
a decoder is 64MiB while the Linux section size is 128MiB. This results
in situations where, without subsection hotplug memory mappings with
different lifetimes collide into one object that can only express one
lifetime.
Update move_pfn_range_to_zone() to flag (SECTION_TAINT_ZONE_DEVICE) a
section that mixes ZONE_DEVICE pfns with other online pfns. With
SECTION_TAINT_ZONE_DEVICE to delineate, pfn_to_online_page() can fall back
to a slow-path check for ZONE_DEVICE pfns in an online section. In the
fast path online_section() for a full ZONE_DEVICE section returns false.
Because the collision case is rare, and for simplicity, the
SECTION_TAINT_ZONE_DEVICE flag is never cleared once set.
[dan.j.williams@intel.com: fix CONFIG_ZONE_DEVICE=n build]
Link: https://lkml.kernel.org/r/CAPcyv4iX+7LAgAeSqx7Zw-Zd=ZV9gBv8Bo7oTbwCOOqJoZ3+Yg@mail.gmail.com
Link: https://lkml.kernel.org/r/161058500675.1840162.7887862152161279354.stgit@dwillia2-desk3.amr.corp.intel.com
Fixes: ba72b4c8cf ("mm/sparsemem: support sub-section hotplug")
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Reported-by: Michal Hocko <mhocko@suse.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Reported-by: David Hildenbrand <david@redhat.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Oscar Salvador <osalvador@suse.de>
Cc: Naoya Horiguchi <naoya.horiguchi@nec.com>
Cc: Qian Cai <cai@lca.pw>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Patch series "mm: Fix pfn_to_online_page() with respect to ZONE_DEVICE", v4.
A pfn-walker that uses pfn_to_online_page() may inadvertently translate a
pfn as online and in the page allocator, when it is offline managed by a
ZONE_DEVICE mapping (details in Patch 3: ("mm: Teach pfn_to_online_page()
about ZONE_DEVICE section collisions")).
The 2 proposals under consideration are teach pfn_to_online_page() to be
precise in the presence of mixed-zone sections, or teach the memory-add
code to drop the System RAM associated with ZONE_DEVICE collisions. In
order to not regress memory capacity by a few 10s to 100s of MiB the
approach taken in this set is to add precision to pfn_to_online_page().
In the course of validating pfn_to_online_page() a couple other fixes
fell out:
1/ soft_offline_page() fails to drop the reference taken in the
madvise(..., MADV_SOFT_OFFLINE) case.
2/ memory_failure() uses get_dev_pagemap() to lookup ZONE_DEVICE pages,
however that mapping may contain data pages and metadata raw pfns.
Introduce pgmap_pfn_valid() to delineate the 2 types and fail the
handling of raw metadata pfns.
This patch (of 4);
pfn_to_online_page() is already too large to be a macro or an inline
function. In anticipation of further logic changes / growth, move it out
of line.
No functional change, just code movement.
Link: https://lkml.kernel.org/r/161058499000.1840162.702316708443239771.stgit@dwillia2-desk3.amr.corp.intel.com
Link: https://lkml.kernel.org/r/161058499608.1840162.10165648147615238793.stgit@dwillia2-desk3.amr.corp.intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Reported-by: Michal Hocko <mhocko@kernel.org>
Acked-by: Michal Hocko <mhocko@suse.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Oscar Salvador <osalvador@suse.de>
Cc: Naoya Horiguchi <naoya.horiguchi@nec.com>
Cc: Qian Cai <cai@lca.pw>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Byte-accounted items are used for slab object accounting at the cgroup
level, because the objects in a slab page can belong to different cgroups.
At the global level these items always change in multiples of whole slab
pages. The vmstat code exploits this and stores these items as pages
internally, which allows for more compact per-cpu data.
This optimization isn't self-evident from the asserts and the division in
the stat update functions. Provide the reader with some context.
Link: https://lkml.kernel.org/r/20210202184411.118614-1-hannes@cmpxchg.org
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
On NOHZ, the periodic vmstat flushers on each CPU can go to sleep and
won't wake up until stat changes are detected in the per-cpu deltas of the
zone vmstat counters.
In commit 75ef718405 ("mm, vmstat: add infrastructure for per-node
vmstats") per-node counters were introduced, and subsequently most stats
were moved from the zone to the node level. However, the node counters
weren't added to the NOHZ wakeup detection.
In theory this can cause per-cpu errors to remain in the user-reported
stats indefinitely. In practice this only affects a handful of sub
counters (file_mapped, dirty and writeback e.g.) because other page state
changes at the node level likely involve a change at the zone level as
well (alloc and free, lru ops). Also, nobody has complained.
Fix it up for completeness: wake up vmstat refreshing on node changes.
Also remove the BUILD_BUG_ONs that assert counter size; we haven't relied
on it since we added sizeof() to the range calculation in commit
13c9aaf7fa ("mm/vmstat.c: fix NUMA statistics updates").
Link: https://lkml.kernel.org/r/20210202184342.118513-1-hannes@cmpxchg.org
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Let's count the number of CMA pages per zone and print them in
/proc/zoneinfo.
Having access to the total number of CMA pages per zone is helpful for
debugging purposes to know where exactly the CMA pages ended up, and to
figure out how many pages of a zone might behave differently, even after
some of these pages might already have been allocated.
As one example, CMA pages part of a kernel zone cannot be used for
ordinary kernel allocations but instead behave more like ZONE_MOVABLE.
For now, we are only able to get the global nr+free cma pages from
/proc/meminfo and the free cma pages per zone from /proc/zoneinfo.
Example after this patch when booting a 6 GiB QEMU VM with
"hugetlb_cma=2G":
# cat /proc/zoneinfo | grep cma
cma 0
nr_free_cma 0
cma 0
nr_free_cma 0
cma 524288
nr_free_cma 493016
cma 0
cma 0
# cat /proc/meminfo | grep Cma
CmaTotal: 2097152 kB
CmaFree: 1972064 kB
Note: We print even without CONFIG_CMA, just like "nr_free_cma"; this way,
one can be sure when spotting "cma 0", that there are definetly no
CMA pages located in a zone.
[david@redhat.com: v2]
Link: https://lkml.kernel.org/r/20210128164533.18566-1-david@redhat.com
[david@redhat.com: v3]
Link: https://lkml.kernel.org/r/20210129113451.22085-1-david@redhat.com
Link: https://lkml.kernel.org/r/20210127101813.6370-3-david@redhat.com
Signed-off-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Oscar Salvador <osalvador@suse.de>
Acked-by: David Rientjes <rientjes@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "Peter Zijlstra (Intel)" <peterz@infradead.org>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Wei Yang <richard.weiyang@linux.alibaba.com>
Cc: Zi Yan <ziy@nvidia.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Right now, if activation fails, we might already have exposed some pages
to the buddy for CMA use (although they will never get actually used by
CMA), and some pages won't be exposed to the buddy at all.
Let's check for "single zone" early and on error, don't expose any pages
for CMA use - instead, expose them to the buddy available for any use.
Simply call free_reserved_page() on every single page - easier than going
via free_reserved_area(), converting back and forth between pfns and virt
addresses.
In addition, make sure to fixup totalcma_pages properly.
Example: 6 GiB QEMU VM with "... hugetlb_cma=2G movablecore=20% ...":
[ 0.006891] hugetlb_cma: reserve 2048 MiB, up to 2048 MiB per node
[ 0.006893] cma: Reserved 2048 MiB at 0x0000000100000000
[ 0.006893] hugetlb_cma: reserved 2048 MiB on node 0
...
[ 0.175433] cma: CMA area hugetlb0 could not be activated
Before this patch:
# cat /proc/meminfo
MemTotal: 5867348 kB
MemFree: 5692808 kB
MemAvailable: 5542516 kB
...
CmaTotal: 2097152 kB
CmaFree: 1884160 kB
After this patch:
# cat /proc/meminfo
MemTotal: 6077308 kB
MemFree: 5904208 kB
MemAvailable: 5747968 kB
...
CmaTotal: 0 kB
CmaFree: 0 kB
Note: cma_init_reserved_mem() makes sure that we always cover full
pageblocks / MAX_ORDER - 1 pages.
Link: https://lkml.kernel.org/r/20210127101813.6370-2-david@redhat.com
Signed-off-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Zi Yan <ziy@nvidia.com>
Reviewed-by: Oscar Salvador <osalvador@suse.de>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "Peter Zijlstra (Intel)" <peterz@infradead.org>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Wei Yang <richard.weiyang@linux.alibaba.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Currently cma areas without a fixed base are allocated close to the end of
the node. This placement is sub-optimal because of compaction: it brings
pages into the cma area. In particular, it can bring in hot executable
pages, even if there is a plenty of free memory on the machine. This
results in cma allocation failures.
Instead let's place cma areas close to the beginning of a node. In this
case the compaction will help to free cma areas, resulting in better cma
allocation success rates.
If there is enough memory let's try to allocate bottom-up starting with
4GB to exclude any possible interference with DMA32. On smaller machines
or in a case of a failure, stick with the old behavior.
16GB vm, 2GB cma area:
With this patch:
[ 0.000000] Command line: root=/dev/vda3 rootflags=subvol=/root systemd.unified_cgroup_hierarchy=1 enforcing=0 console=ttyS0,115200 hugetlb_cma=2G
[ 0.002928] hugetlb_cma: reserve 2048 MiB, up to 2048 MiB per node
[ 0.002930] cma: Reserved 2048 MiB at 0x0000000100000000
[ 0.002931] hugetlb_cma: reserved 2048 MiB on node 0
Without this patch:
[ 0.000000] Command line: root=/dev/vda3 rootflags=subvol=/root systemd.unified_cgroup_hierarchy=1 enforcing=0 console=ttyS0,115200 hugetlb_cma=2G
[ 0.002930] hugetlb_cma: reserve 2048 MiB, up to 2048 MiB per node
[ 0.002933] cma: Reserved 2048 MiB at 0x00000003c0000000
[ 0.002934] hugetlb_cma: reserved 2048 MiB on node 0
v2:
- switched to memblock_set_bottom_up(true), by Mike
- start with 4GB, by Mike
[guro@fb.com: whitespace fix, per Mike]
Link: https://lkml.kernel.org/r/20201221170551.GB3428478@carbon.DHCP.thefacebook.com
[guro@fb.com: fix 32-bit warnings]
Link: https://lkml.kernel.org/r/20201223163537.GA4011967@carbon.DHCP.thefacebook.com
[guro@fb.com: fix 32-bit systems]
[akpm@linux-foundation.org: build fix]
Link: https://lkml.kernel.org/r/20201217201214.3414100-1-guro@fb.com
Signed-off-by: Roman Gushchin <guro@fb.com>
Reviewed-by: Mike Rapoport <rppt@linux.ibm.com>
Cc: Wonhyuk Yang <vvghjk1234@gmail.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Rik van Riel <riel@surriel.com>
Cc: Michal Hocko <mhocko@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Patch series "mm,thp,shm: limit shmem THP alloc gfp_mask", v6.
The allocation flags of anonymous transparent huge pages can be controlled
through the files in /sys/kernel/mm/transparent_hugepage/defrag, which can
help the system from getting bogged down in the page reclaim and
compaction code when many THPs are getting allocated simultaneously.
However, the gfp_mask for shmem THP allocations were not limited by those
configuration settings, and some workloads ended up with all CPUs stuck on
the LRU lock in the page reclaim code, trying to allocate dozens of THPs
simultaneously.
This patch applies the same configurated limitation of THPs to shmem
hugepage allocations, to prevent that from happening.
This way a THP defrag setting of "never" or "defer+madvise" will result in
quick allocation failures without direct reclaim when no 2MB free pages
are available.
With this patch applied, THP allocations for tmpfs will be a little more
aggressive than today for files mmapped with MADV_HUGEPAGE, and a little
less aggressive for files that are not mmapped or mapped without that
flag.
This patch (of 4):
The allocation flags of anonymous transparent huge pages can be controlled
through the files in /sys/kernel/mm/transparent_hugepage/defrag, which can
help the system from getting bogged down in the page reclaim and
compaction code when many THPs are getting allocated simultaneously.
However, the gfp_mask for shmem THP allocations were not limited by those
configuration settings, and some workloads ended up with all CPUs stuck on
the LRU lock in the page reclaim code, trying to allocate dozens of THPs
simultaneously.
This patch applies the same configurated limitation of THPs to shmem
hugepage allocations, to prevent that from happening.
Controlling the gfp_mask of THP allocations through the knobs in sysfs
allows users to determine the balance between how aggressively the system
tries to allocate THPs at fault time, and how much the application may end
up stalling attempting those allocations.
This way a THP defrag setting of "never" or "defer+madvise" will result in
quick allocation failures without direct reclaim when no 2MB free pages
are available.
With this patch applied, THP allocations for tmpfs will be a little more
aggressive than today for files mmapped with MADV_HUGEPAGE, and a little
less aggressive for files that are not mmapped or mapped without that
flag.
Link: https://lkml.kernel.org/r/20201124194925.623931-1-riel@surriel.com
Link: https://lkml.kernel.org/r/20201124194925.623931-2-riel@surriel.com
Signed-off-by: Rik van Riel <riel@surriel.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Xu Yu <xuyu@linux.alibaba.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: Hugh Dickins <hughd@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Patch series "Overhaul multi-page lookups for THP", v4.
This THP prep patchset changes several page cache iteration APIs to only
return head pages.
- It's only possible to tag head pages in the page cache, so only
return head pages, not all their subpages.
- Factor a lot of common code out of the various batch lookup routines
- Add mapping_seek_hole_data()
- Unify find_get_entries() and pagevec_lookup_entries()
- Make find_get_entries only return head pages, like find_get_entry().
These are only loosely connected, but they seem to make sense together as
a series.
This patch (of 14):
Pagecache tags are used for dirty page writeback. Since dirtiness is
tracked on a per-THP basis, we only want to return the head page rather
than each subpage of a tagged page. All the filesystems which use huge
pages today are in-memory, so there are no tagged huge pages today.
Link: https://lkml.kernel.org/r/20201112212641.27837-2-willy@infradead.org
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: William Kucharski <william.kucharski@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Cc: Hugh Dickins <hughd@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Yang Shi <yang.shi@linux.alibaba.com>
Cc: Dave Chinner <dchinner@redhat.com>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Now, NUMA balancing can only optimize the page placement among the NUMA
nodes if the default memory policy is used. Because the memory policy
specified explicitly should take precedence. But this seems too strict in
some situations. For example, on a system with 4 NUMA nodes, if the
memory of an application is bound to the node 0 and 1, NUMA balancing can
potentially migrate the pages between the node 0 and 1 to reduce
cross-node accessing without breaking the explicit memory binding policy.
So in this patch, we add MPOL_F_NUMA_BALANCING mode flag to
set_mempolicy() when mode is MPOL_BIND. With the flag specified, NUMA
balancing will be enabled within the thread to optimize the page placement
within the constrains of the specified memory binding policy. With the
newly added flag, the NUMA balancing control mechanism becomes,
- sysctl knob numa_balancing can enable/disable the NUMA balancing
globally.
- even if sysctl numa_balancing is enabled, the NUMA balancing will be
disabled for the memory areas or applications with the explicit
memory policy by default.
- MPOL_F_NUMA_BALANCING can be used to enable the NUMA balancing for
the applications when specifying the explicit memory policy
(MPOL_BIND).
Various page placement optimization based on the NUMA balancing can be
done with these flags. As the first step, in this patch, if the memory of
the application is bound to multiple nodes (MPOL_BIND), and in the hint
page fault handler the accessing node are in the policy nodemask, the page
will be tried to be migrated to the accessing node to reduce the
cross-node accessing.
If the newly added MPOL_F_NUMA_BALANCING flag is specified by an
application on an old kernel version without its support, set_mempolicy()
will return -1 and errno will be set to EINVAL. The application can use
this behavior to run on both old and new kernel versions.
And if the MPOL_F_NUMA_BALANCING flag is specified for the mode other than
MPOL_BIND, set_mempolicy() will return -1 and errno will be set to EINVAL
as before. Because we don't support optimization based on the NUMA
balancing for these modes.
In the previous version of the patch, we tried to reuse MPOL_MF_LAZY for
mbind(). But that flag is tied to MPOL_MF_MOVE.*, so it seems not a good
API/ABI for the purpose of the patch.
And because it's not clear whether it's necessary to enable NUMA balancing
for a specific memory area inside an application, so we only add the flag
at the thread level (set_mempolicy()) instead of the memory area level
(mbind()). We can do that when it become necessary.
To test the patch, we run a test case as follows on a 4-node machine with
192 GB memory (48 GB per node).
1. Change pmbench memory accessing benchmark to call set_mempolicy()
to bind its memory to node 1 and 3 and enable NUMA balancing. Some
related code snippets are as follows,
#include <numaif.h>
#include <numa.h>
struct bitmask *bmp;
int ret;
bmp = numa_parse_nodestring("1,3");
ret = set_mempolicy(MPOL_BIND | MPOL_F_NUMA_BALANCING,
bmp->maskp, bmp->size + 1);
/* If MPOL_F_NUMA_BALANCING isn't supported, fall back to MPOL_BIND */
if (ret < 0 && errno == EINVAL)
ret = set_mempolicy(MPOL_BIND, bmp->maskp, bmp->size + 1);
if (ret < 0) {
perror("Failed to call set_mempolicy");
exit(-1);
}
2. Run a memory eater on node 3 to use 40 GB memory before running pmbench.
3. Run pmbench with 64 processes, the working-set size of each process
is 640 MB, so the total working-set size is 64 * 640 MB = 40 GB. The
CPU and the memory (as in step 1.) of all pmbench processes is bound
to node 1 and 3. So, after CPU usage is balanced, some pmbench
processes run on the CPUs of the node 3 will access the memory of
the node 1.
4. After the pmbench processes run for 100 seconds, kill the memory
eater. Now it's possible for some pmbench processes to migrate
their pages from node 1 to node 3 to reduce cross-node accessing.
Test results show that, with the patch, the pages can be migrated from
node 1 to node 3 after killing the memory eater, and the pmbench score
can increase about 17.5%.
Link: https://lkml.kernel.org/r/20210120061235.148637-2-ying.huang@intel.com
Signed-off-by: "Huang, Ying" <ying.huang@intel.com>
Acked-by: Mel Gorman <mgorman@suse.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Rik van Riel <riel@surriel.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: "Matthew Wilcox (Oracle)" <willy@infradead.org>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Compaction always operates on pages from a single given zone when
isolating both pages to migrate and freepages. Pageblock boundaries are
intersected with zone boundaries to be safe in case zone starts or ends in
the middle of pageblock. The use of pageblock_pfn_to_page() protects
against non-contiguous pageblocks.
The functions fast_isolate_freepages() and fast_isolate_around() don't
currently protect the fast freepage isolation thoroughly enough against
these corner cases, and can result in freepage isolation operate outside
of zone boundaries:
- in fast_isolate_freepages() if we get a pfn from the first pageblock
of a zone that starts in the middle of that pageblock, 'highest' can
be a pfn outside of the zone.
If we fail to isolate anything in this function, we may then call
fast_isolate_around() on a pfn outside of the zone and there
effectively do a set_pageblock_skip(page_to_pfn(highest)) which may
currently hit a VM_BUG_ON() in some configurations
- fast_isolate_around() checks only the zone end boundary and not
beginning, nor that the pageblock is contiguous (with
pageblock_pfn_to_page()) so it's possible that we end up calling
isolate_freepages_block() on a range of pfn's from two different
zones and end up e.g. isolating freepages under the wrong zone's
lock.
This patch should fix the above issues.
Link: https://lkml.kernel.org/r/20210217173300.6394-1-vbabka@suse.cz
Fixes: 5a811889de ("mm, compaction: use free lists to quickly locate a migration target")
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Acked-by: David Rientjes <rientjes@google.com>
Acked-by: Mel Gorman <mgorman@techsingularity.net>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
In the fast_find_migrateblock(), it iterates ocer the freelist to find the
proper pageblock. But there are some misbehaviors.
First, if the page we found is equal to cc->migrate_pfn, it is considered
that we didn't find a suitable pageblock. Secondly, if the loop was
terminated because order is less than PAGE_ALLOC_COSTLY_ORDER, it could be
considered that we found a suitable one. Thirdly, if the skip bit is set
on the page block and we goto continue, it doesn't check nr_scanned.
Fourthly, if the page block's skip bit is set, it checks that page block
is the last of list, which is unnecessary.
Link: https://lkml.kernel.org/r/20210128130411.6125-1-vvghjk1234@gmail.com
Fixes: 70b44595ea ("mm, compaction: use free lists to quickly locate a migration source")
Signed-off-by: Wonhyuk Yang <vvghjk1234@gmail.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
should_proactive_compact_node() returns true when sum of the weighted
fragmentation score of all the zones in the node is greater than the
wmark_high of compaction, which then triggers the proactive compaction
that operates on the individual zones of the node. But proactive
compaction runs on the zone only when its weighted fragmentation score
is greater than wmark_low(=wmark_high - 10).
This means that the sum of the weighted fragmentation scores of all the
zones can exceed the wmark_high but individual weighted fragmentation zone
scores can still be less than wmark_low which makes the unnecessary
trigger of the proactive compaction only to return doing nothing.
Issue with the return of proactive compaction with out even trying is its
deferral. It is simply deferred for 1 << COMPACT_MAX_DEFER_SHIFT if the
scores across the proactive compaction is same, thinking that compaction
didn't make any progress but in reality it didn't even try. With the
delay between successive retries for proactive compaction is 500msec, it
can result into the deferral for ~30sec with out even trying the proactive
compaction.
Test scenario is that: compaction_proactiveness=50 thus the wmark_low = 50
and wmark_high = 60. System have 2 zones(Normal and Movable) with sizes
5GB and 6GB respectively. After opening some apps on the android, the
weighted fragmentation scores of these zones are 47 and 49 respectively.
Since the sum of these fragmentation scores are above the wmark_high which
triggers the proactive compaction and there since the individual zones
weighted fragmentation scores are below wmark_low, it returns without
trying the proactive compaction. As a result the weighted fragmentation
scores of the zones are still 47 and 49 which makes the existing logic to
defer the compaction thinking that noprogress is made across the
compaction.
Fix this by checking just zone fragmentation score, not the weighted, in
__compact_finished() and use the zones weighted fragmentation score in
fragmentation_score_node(). In the test case above, If the weighted
average of is above wmark_high, then individual score (not adjusted) of
atleast one zone has to be above wmark_high. Thus it avoids the
unnecessary trigger and deferrals of the proactive compaction.
Link: https://lkml.kernel.org/r/1610989938-31374-1-git-send-email-charante@codeaurora.org
Signed-off-by: Charan Teja Reddy <charante@codeaurora.org>
Suggested-by: Vlastimil Babka <vbabka@suse.cz>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Reviewed-by: Khalid Aziz <khalid.aziz@oracle.com>
Acked-by: David Rientjes <rientjes@google.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Nitin Gupta <ngupta@nitingupta.dev>
Cc: Vinayak Menon <vinmenon@codeaurora.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
