| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
and propagate through down the call stack.
Link: http://lkml.kernel.org/r/20181124091411.GC10969@avx2
Signed-off-by: Alexey Dobriyan <[email protected]>
Reviewed-by: Andrew Morton <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Those strings are immutable as well.
Link: http://lkml.kernel.org/r/20181124090508.GB10877@avx2
Signed-off-by: Alexey Dobriyan <[email protected]>
Reviewed-by: Andrew Morton <[email protected]>
Reviewed-by: David Hildenbrand <[email protected]>
Acked-by: Vlastimil Babka <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Those strings are immutable in fact.
Link: http://lkml.kernel.org/r/20181124090327.GA10877@avx2
Signed-off-by: Alexey Dobriyan <[email protected]>
Reviewed-by: Andrew Morton <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
An external fragmentation event was previously described as
When the page allocator fragments memory, it records the event using
the mm_page_alloc_extfrag event. If the fallback_order is smaller
than a pageblock order (order-9 on 64-bit x86) then it's considered
an event that will cause external fragmentation issues in the future.
The kernel reduces the probability of such events by increasing the
watermark sizes by calling set_recommended_min_free_kbytes early in the
lifetime of the system. This works reasonably well in general but if
there are enough sparsely populated pageblocks then the problem can still
occur as enough memory is free overall and kswapd stays asleep.
This patch introduces a watermark_boost_factor sysctl that allows a zone
watermark to be temporarily boosted when an external fragmentation causing
events occurs. The boosting will stall allocations that would decrease
free memory below the boosted low watermark and kswapd is woken if the
calling context allows to reclaim an amount of memory relative to the size
of the high watermark and the watermark_boost_factor until the boost is
cleared. When kswapd finishes, it wakes kcompactd at the pageblock order
to clean some of the pageblocks that may have been affected by the
fragmentation event. kswapd avoids any writeback, slab shrinkage and swap
from reclaim context during this operation to avoid excessive system
disruption in the name of fragmentation avoidance. Care is taken so that
kswapd will do normal reclaim work if the system is really low on memory.
This was evaluated using the same workloads as "mm, page_alloc: Spread
allocations across zones before introducing fragmentation".
1-socket Skylake machine
config-global-dhp__workload_thpfioscale XFS (no special madvise)
4 fio threads, 1 THP allocating thread
--------------------------------------
4.20-rc3 extfrag events < order 9: 804694
4.20-rc3+patch: 408912 (49% reduction)
4.20-rc3+patch1-4: 18421 (98% reduction)
4.20.0-rc3 4.20.0-rc3
lowzone-v5r8 boost-v5r8
Amean fault-base-1 653.58 ( 0.00%) 652.71 ( 0.13%)
Amean fault-huge-1 0.00 ( 0.00%) 178.93 * -99.00%*
4.20.0-rc3 4.20.0-rc3
lowzone-v5r8 boost-v5r8
Percentage huge-1 0.00 ( 0.00%) 5.12 ( 100.00%)
Note that external fragmentation causing events are massively reduced by
this path whether in comparison to the previous kernel or the vanilla
kernel. The fault latency for huge pages appears to be increased but that
is only because THP allocations were successful with the patch applied.
1-socket Skylake machine
global-dhp__workload_thpfioscale-madvhugepage-xfs (MADV_HUGEPAGE)
-----------------------------------------------------------------
4.20-rc3 extfrag events < order 9: 291392
4.20-rc3+patch: 191187 (34% reduction)
4.20-rc3+patch1-4: 13464 (95% reduction)
thpfioscale Fault Latencies
4.20.0-rc3 4.20.0-rc3
lowzone-v5r8 boost-v5r8
Min fault-base-1 912.00 ( 0.00%) 905.00 ( 0.77%)
Min fault-huge-1 127.00 ( 0.00%) 135.00 ( -6.30%)
Amean fault-base-1 1467.55 ( 0.00%) 1481.67 ( -0.96%)
Amean fault-huge-1 1127.11 ( 0.00%) 1063.88 * 5.61%*
4.20.0-rc3 4.20.0-rc3
lowzone-v5r8 boost-v5r8
Percentage huge-1 77.64 ( 0.00%) 83.46 ( 7.49%)
As before, massive reduction in external fragmentation events, some jitter
on latencies and an increase in THP allocation success rates.
2-socket Haswell machine
config-global-dhp__workload_thpfioscale XFS (no special madvise)
4 fio threads, 5 THP allocating threads
----------------------------------------------------------------
4.20-rc3 extfrag events < order 9: 215698
4.20-rc3+patch: 200210 (7% reduction)
4.20-rc3+patch1-4: 14263 (93% reduction)
4.20.0-rc3 4.20.0-rc3
lowzone-v5r8 boost-v5r8
Amean fault-base-5 1346.45 ( 0.00%) 1306.87 ( 2.94%)
Amean fault-huge-5 3418.60 ( 0.00%) 1348.94 ( 60.54%)
4.20.0-rc3 4.20.0-rc3
lowzone-v5r8 boost-v5r8
Percentage huge-5 0.78 ( 0.00%) 7.91 ( 910.64%)
There is a 93% reduction in fragmentation causing events, there is a big
reduction in the huge page fault latency and allocation success rate is
higher.
2-socket Haswell machine
global-dhp__workload_thpfioscale-madvhugepage-xfs (MADV_HUGEPAGE)
-----------------------------------------------------------------
4.20-rc3 extfrag events < order 9: 166352
4.20-rc3+patch: 147463 (11% reduction)
4.20-rc3+patch1-4: 11095 (93% reduction)
thpfioscale Fault Latencies
4.20.0-rc3 4.20.0-rc3
lowzone-v5r8 boost-v5r8
Amean fault-base-5 6217.43 ( 0.00%) 7419.67 * -19.34%*
Amean fault-huge-5 3163.33 ( 0.00%) 3263.80 ( -3.18%)
4.20.0-rc3 4.20.0-rc3
lowzone-v5r8 boost-v5r8
Percentage huge-5 95.14 ( 0.00%) 87.98 ( -7.53%)
There is a large reduction in fragmentation events with some jitter around
the latencies and success rates. As before, the high THP allocation
success rate does mean the system is under a lot of pressure. However, as
the fragmentation events are reduced, it would be expected that the
long-term allocation success rate would be higher.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Mel Gorman <[email protected]>
Acked-by: Vlastimil Babka <[email protected]>
Cc: Andrea Arcangeli <[email protected]>
Cc: David Rientjes <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Zi Yan <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
This is a preparation patch that copies the GFP flag __GFP_KSWAPD_RECLAIM
into alloc_flags. This is a preparation patch only that avoids having to
pass gfp_mask through a long callchain in a future patch.
Note that the setting in the fast path happens in alloc_flags_nofragment()
and it may be claimed that this has nothing to do with ALLOC_NO_FRAGMENT.
That's true in this patch but is not true later so it's done now for
easier review to show where the flag needs to be recorded.
No functional change.
[[email protected]: ALLOC_KSWAPD flag needs to be applied in the !CONFIG_ZONE_DMA32 case]
Link: http://lkml.kernel.org/r/[email protected]
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Mel Gorman <[email protected]>
Reviewed-by: Andrew Morton <[email protected]>
Cc: Andrea Arcangeli <[email protected]>
Cc: David Rientjes <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Vlastimil Babka <[email protected]>
Cc: Zi Yan <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
This is a preparation patch only, no functional change.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Mel Gorman <[email protected]>
Acked-by: Vlastimil Babka <[email protected]>
Cc: Andrea Arcangeli <[email protected]>
Cc: David Rientjes <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Zi Yan <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Patch series "Fragmentation avoidance improvements", v5.
It has been noted before that fragmentation avoidance (aka
anti-fragmentation) is not perfect. Given sufficient time or an adverse
workload, memory gets fragmented and the long-term success of high-order
allocations degrades. This series defines an adverse workload, a definition
of external fragmentation events (including serious) ones and a series
that reduces the level of those fragmentation events.
The details of the workload and the consequences are described in more
detail in the changelogs. However, from patch 1, this is a high-level
summary of the adverse workload. The exact details are found in the
mmtests implementation.
The broad details of the workload are as follows;
1. Create an XFS filesystem (not specified in the configuration but done
as part of the testing for this patch)
2. Start 4 fio threads that write a number of 64K files inefficiently.
Inefficiently means that files are created on first access and not
created in advance (fio parameterr create_on_open=1) and fallocate
is not used (fallocate=none). With multiple IO issuers this creates
a mix of slab and page cache allocations over time. The total size
of the files is 150% physical memory so that the slabs and page cache
pages get mixed
3. Warm up a number of fio read-only threads accessing the same files
created in step 2. This part runs for the same length of time it
took to create the files. It'll fault back in old data and further
interleave slab and page cache allocations. As it's now low on
memory due to step 2, fragmentation occurs as pageblocks get
stolen.
4. While step 3 is still running, start a process that tries to allocate
75% of memory as huge pages with a number of threads. The number of
threads is based on a (NR_CPUS_SOCKET - NR_FIO_THREADS)/4 to avoid THP
threads contending with fio, any other threads or forcing cross-NUMA
scheduling. Note that the test has not been used on a machine with less
than 8 cores. The benchmark records whether huge pages were allocated
and what the fault latency was in microseconds
5. Measure the number of events potentially causing external fragmentation,
the fault latency and the huge page allocation success rate.
6. Cleanup
Overall the series reduces external fragmentation causing events by over 94%
on 1 and 2 socket machines, which in turn impacts high-order allocation
success rates over the long term. There are differences in latencies and
high-order allocation success rates. Latencies are a mixed bag as they
are vulnerable to exact system state and whether allocations succeeded
so they are treated as a secondary metric.
Patch 1 uses lower zones if they are populated and have free memory
instead of fragmenting a higher zone. It's special cased to
handle a Normal->DMA32 fallback with the reasons explained
in the changelog.
Patch 2-4 boosts watermarks temporarily when an external fragmentation
event occurs. kswapd wakes to reclaim a small amount of old memory
and then wakes kcompactd on completion to recover the system
slightly. This introduces some overhead in the slowpath. The level
of boosting can be tuned or disabled depending on the tolerance
for fragmentation vs allocation latency.
Patch 5 stalls some movable allocation requests to let kswapd from patch 4
make some progress. The duration of the stalls is very low but it
is possible to tune the system to avoid fragmentation events if
larger stalls can be tolerated.
The bulk of the improvement in fragmentation avoidance is from patches
1-4 but patch 5 can deal with a rare corner case and provides the option
of tuning a system for THP allocation success rates in exchange for
some stalls to control fragmentation.
This patch (of 5):
The page allocator zone lists are iterated based on the watermarks of each
zone which does not take anti-fragmentation into account. On x86, node 0
may have multiple zones while other nodes have one zone. A consequence is
that tasks running on node 0 may fragment ZONE_NORMAL even though
ZONE_DMA32 has plenty of free memory. This patch special cases the
allocator fast path such that it'll try an allocation from a lower local
zone before fragmenting a higher zone. In this case, stealing of
pageblocks or orders larger than a pageblock are still allowed in the fast
path as they are uninteresting from a fragmentation point of view.
This was evaluated using a benchmark designed to fragment memory before
attempting THP allocations. It's implemented in mmtests as the following
configurations
configs/config-global-dhp__workload_thpfioscale
configs/config-global-dhp__workload_thpfioscale-defrag
configs/config-global-dhp__workload_thpfioscale-madvhugepage
e.g. from mmtests
./run-mmtests.sh --run-monitor --config configs/config-global-dhp__workload_thpfioscale test-run-1
The broad details of the workload are as follows;
1. Create an XFS filesystem (not specified in the configuration but done
as part of the testing for this patch).
2. Start 4 fio threads that write a number of 64K files inefficiently.
Inefficiently means that files are created on first access and not
created in advance (fio parameter create_on_open=1) and fallocate
is not used (fallocate=none). With multiple IO issuers this creates
a mix of slab and page cache allocations over time. The total size
of the files is 150% physical memory so that the slabs and page cache
pages get mixed.
3. Warm up a number of fio read-only processes accessing the same files
created in step 2. This part runs for the same length of time it
took to create the files. It'll refault old data and further
interleave slab and page cache allocations. As it's now low on
memory due to step 2, fragmentation occurs as pageblocks get
stolen.
4. While step 3 is still running, start a process that tries to allocate
75% of memory as huge pages with a number of threads. The number of
threads is based on a (NR_CPUS_SOCKET - NR_FIO_THREADS)/4 to avoid THP
threads contending with fio, any other threads or forcing cross-NUMA
scheduling. Note that the test has not been used on a machine with less
than 8 cores. The benchmark records whether huge pages were allocated
and what the fault latency was in microseconds.
5. Measure the number of events potentially causing external fragmentation,
the fault latency and the huge page allocation success rate.
6. Cleanup the test files.
Note that due to the use of IO and page cache that this benchmark is not
suitable for running on large machines where the time to fragment memory
may be excessive. Also note that while this is one mix that generates
fragmentation that it's not the only mix that generates fragmentation.
Differences in workload that are more slab-intensive or whether SLUB is
used with high-order pages may yield different results.
When the page allocator fragments memory, it records the event using the
mm_page_alloc_extfrag ftrace event. If the fallback_order is smaller than
a pageblock order (order-9 on 64-bit x86) then it's considered to be an
"external fragmentation event" that may cause issues in the future.
Hence, the primary metric here is the number of external fragmentation
events that occur with order < 9. The secondary metric is allocation
latency and huge page allocation success rates but note that differences
in latencies and what the success rate also can affect the number of
external fragmentation event which is why it's a secondary metric.
1-socket Skylake machine
config-global-dhp__workload_thpfioscale XFS (no special madvise)
4 fio threads, 1 THP allocating thread
--------------------------------------
4.20-rc3 extfrag events < order 9: 804694
4.20-rc3+patch: 408912 (49% reduction)
thpfioscale Fault Latencies
4.20.0-rc3 4.20.0-rc3
vanilla lowzone-v5r8
Amean fault-base-1 662.92 ( 0.00%) 653.58 * 1.41%*
Amean fault-huge-1 0.00 ( 0.00%) 0.00 ( 0.00%)
4.20.0-rc3 4.20.0-rc3
vanilla lowzone-v5r8
Percentage huge-1 0.00 ( 0.00%) 0.00 ( 0.00%)
Fault latencies are slightly reduced while allocation success rates remain
at zero as this configuration does not make any special effort to allocate
THP and fio is heavily active at the time and either filling memory or
keeping pages resident. However, a 49% reduction of serious fragmentation
events reduces the changes of external fragmentation being a problem in
the future.
Vlastimil asked during review for a breakdown of the allocation types
that are falling back.
vanilla
3816 MIGRATE_UNMOVABLE
800845 MIGRATE_MOVABLE
33 MIGRATE_UNRECLAIMABLE
patch
735 MIGRATE_UNMOVABLE
408135 MIGRATE_MOVABLE
42 MIGRATE_UNRECLAIMABLE
The majority of the fallbacks are due to movable allocations and this is
consistent for the workload throughout the series so will not be presented
again as the primary source of fallbacks are movable allocations.
Movable fallbacks are sometimes considered "ok" to fallback because they
can be migrated. The problem is that they can fill an
unmovable/reclaimable pageblock causing those allocations to fallback
later and polluting pageblocks with pages that cannot move. If there is a
movable fallback, it is pretty much guaranteed to affect an
unmovable/reclaimable pageblock and while it might not be enough to
actually cause a unmovable/reclaimable fallback in the future, we cannot
know that in advance so the patch takes the only option available to it.
Hence, it's important to control them. This point is also consistent
throughout the series and will not be repeated.
1-socket Skylake machine
global-dhp__workload_thpfioscale-madvhugepage-xfs (MADV_HUGEPAGE)
-----------------------------------------------------------------
4.20-rc3 extfrag events < order 9: 291392
4.20-rc3+patch: 191187 (34% reduction)
thpfioscale Fault Latencies
4.20.0-rc3 4.20.0-rc3
vanilla lowzone-v5r8
Amean fault-base-1 1495.14 ( 0.00%) 1467.55 ( 1.85%)
Amean fault-huge-1 1098.48 ( 0.00%) 1127.11 ( -2.61%)
thpfioscale Percentage Faults Huge
4.20.0-rc3 4.20.0-rc3
vanilla lowzone-v5r8
Percentage huge-1 78.57 ( 0.00%) 77.64 ( -1.18%)
Fragmentation events were reduced quite a bit although this is known
to be a little variable. The latencies and allocation success rates
are similar but they were already quite high.
2-socket Haswell machine
config-global-dhp__workload_thpfioscale XFS (no special madvise)
4 fio threads, 5 THP allocating threads
----------------------------------------------------------------
4.20-rc3 extfrag events < order 9: 215698
4.20-rc3+patch: 200210 (7% reduction)
thpfioscale Fault Latencies
4.20.0-rc3 4.20.0-rc3
vanilla lowzone-v5r8
Amean fault-base-5 1350.05 ( 0.00%) 1346.45 ( 0.27%)
Amean fault-huge-5 4181.01 ( 0.00%) 3418.60 ( 18.24%)
4.20.0-rc3 4.20.0-rc3
vanilla lowzone-v5r8
Percentage huge-5 1.15 ( 0.00%) 0.78 ( -31.88%)
The reduction of external fragmentation events is slight and this is
partially due to the removal of __GFP_THISNODE in commit ac5b2c18911f
("mm: thp: relax __GFP_THISNODE for MADV_HUGEPAGE mappings") as THP
allocations can now spill over to remote nodes instead of fragmenting
local memory.
2-socket Haswell machine
global-dhp__workload_thpfioscale-madvhugepage-xfs (MADV_HUGEPAGE)
-----------------------------------------------------------------
4.20-rc3 extfrag events < order 9: 166352
4.20-rc3+patch: 147463 (11% reduction)
thpfioscale Fault Latencies
4.20.0-rc3 4.20.0-rc3
vanilla lowzone-v5r8
Amean fault-base-5 6138.97 ( 0.00%) 6217.43 ( -1.28%)
Amean fault-huge-5 2294.28 ( 0.00%) 3163.33 * -37.88%*
thpfioscale Percentage Faults Huge
4.20.0-rc3 4.20.0-rc3
vanilla lowzone-v5r8
Percentage huge-5 96.82 ( 0.00%) 95.14 ( -1.74%)
There was a slight reduction in external fragmentation events although the
latencies were higher. The allocation success rate is high enough that
the system is struggling and there is quite a lot of parallel reclaim and
compaction activity. There is also a certain degree of luck on whether
processes start on node 0 or not for this patch but the relevance is
reduced later in the series.
Overall, the patch reduces the number of external fragmentation causing
events so the success of THP over long periods of time would be improved
for this adverse workload.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Mel Gorman <[email protected]>
Acked-by: Vlastimil Babka <[email protected]>
Cc: David Rientjes <[email protected]>
Cc: Andrea Arcangeli <[email protected]>
Cc: Zi Yan <[email protected]>
Cc: Michal Hocko <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Userspace should always be in charge of how to online memory and if memory
should be onlined automatically in the kernel. Let's drop the parameter
to overwrite this - XEN passes memhp_auto_online, just like add_memory(),
so we can directly use that instead internally.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: David Hildenbrand <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Reviewed-by: Oscar Salvador <[email protected]>
Acked-by: Juergen Gross <[email protected]>
Cc: Boris Ostrovsky <[email protected]>
Cc: Stefano Stabellini <[email protected]>
Cc: Dan Williams <[email protected]>
Cc: Pavel Tatashin <[email protected]>
Cc: David Hildenbrand <[email protected]>
Cc: Joonsoo Kim <[email protected]>
Cc: Arun KS <[email protected]>
Cc: Mathieu Malaterre <[email protected]>
Cc: Stephen Rothwell <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
In cb5e39b8038b ("drivers: base: refactor add_memory_section() to
add_memory_block()"), add_memory_block() is introduced, which is only
invoked in memory_dev_init().
When combining these two loops in memory_dev_init() and
add_memory_block(), they looks like this:
for (i = 0; i < NR_MEM_SECTIONS; i += sections_per_block)
for (j = i;
(j < i + sections_per_block) && j < NR_MEM_SECTIONS;
j++)
Since it is sure the (i < NR_MEM_SECTIONS) and j sits in its own memory
block, the check of (j < NR_MEM_SECTIONS) is not necessary.
This patch just removes this check.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Wei Yang <[email protected]>
Reviewed-by: Andrew Morton <[email protected]>
Cc: Greg Kroah-Hartman <[email protected]>
Cc: "Rafael J. Wysocki" <[email protected]>
Cc: Seth Jennings <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
__memblock_free_early() is only used by the convenience wrappers, so
essentially we wrap a call to memblock_free() twice. Replace calls of
__memblock_free_early() with calls to memblock_free() and drop the former.
Link: http://lkml.kernel.org/r/20181125102940.GE28634@rapoport-lnx
Signed-off-by: Mike Rapoport <[email protected]>
Reviewed-by: Andrew Morton <[email protected]>
Cc: Wentao Wang <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Wentao Wang <[email protected]>
Reviewed-by: Andrew Morton <[email protected]>
Cc: Mike Rapoport <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
There are multiple places of freeing a page, they all do the same things
so a common function can be used to reduce code duplicate.
It also avoids bug fixed in one function but left in another.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Aaron Lu <[email protected]>
Acked-by: Vlastimil Babka <[email protected]>
Cc: Alexander Duyck <[email protected]>
Cc: Ilias Apalodimas <[email protected]>
Cc: Jesper Dangaard Brouer <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: Pankaj gupta <[email protected]>
Cc: Pawel Staszewski <[email protected]>
Cc: Tariq Toukan <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
page_frag_free() calls __free_pages_ok() to free the page back to Buddy.
This is OK for high order page, but for order-0 pages, it misses the
optimization opportunity of using Per-Cpu-Pages and can cause zone lock
contention when called frequently.
Pawel Staszewski recently shared his result of 'how Linux kernel handles
normal traffic'[1] and from perf data, Jesper Dangaard Brouer found the
lock contention comes from page allocator:
mlx5e_poll_tx_cq
|
--16.34%--napi_consume_skb
|
|--12.65%--__free_pages_ok
| |
| --11.86%--free_one_page
| |
| |--10.10%--queued_spin_lock_slowpath
| |
| --0.65%--_raw_spin_lock
|
|--1.55%--page_frag_free
|
--1.44%--skb_release_data
Jesper explained how it happened: mlx5 driver RX-page recycle mechanism is
not effective in this workload and pages have to go through the page
allocator. The lock contention happens during mlx5 DMA TX completion
cycle. And the page allocator cannot keep up at these speeds.[2]
I thought that __free_pages_ok() are mostly freeing high order pages and
thought this is an lock contention for high order pages but Jesper
explained in detail that __free_pages_ok() here are actually freeing
order-0 pages because mlx5 is using order-0 pages to satisfy its page pool
allocation request.[3]
The free path as pointed out by Jesper is:
skb_free_head()
-> skb_free_frag()
-> page_frag_free()
And the pages being freed on this path are order-0 pages.
Fix this by doing similar things as in __page_frag_cache_drain() - send
the being freed page to PCP if it's an order-0 page, or directly to Buddy
if it is a high order page.
With this change, Paweł hasn't noticed lock contention yet in his
workload and Jesper has noticed a 7% performance improvement using a micro
benchmark and lock contention is gone. Ilias' test on a 'low' speed 1Gbit
interface on an cortex-a53 shows ~11% performance boost testing with
64byte packets and __free_pages_ok() disappeared from perf top.
[1]: https://www.spinics.net/lists/netdev/msg531362.html
[2]: https://www.spinics.net/lists/netdev/msg531421.html
[3]: https://www.spinics.net/lists/netdev/msg531556.html
[[email protected]: add comment]
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Aaron Lu <[email protected]>
Reported-by: Pawel Staszewski <[email protected]>
Analysed-by: Jesper Dangaard Brouer <[email protected]>
Acked-by: Vlastimil Babka <[email protected]>
Acked-by: Mel Gorman <[email protected]>
Acked-by: Jesper Dangaard Brouer <[email protected]>
Acked-by: Ilias Apalodimas <[email protected]>
Tested-by: Ilias Apalodimas <[email protected]>
Acked-by: Alexander Duyck <[email protected]>
Acked-by: Tariq Toukan <[email protected]>
Acked-by: Pankaj gupta <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
At Maintainer Summit, Greg brought up a topic I proposed around
EXPORT_SYMBOL_GPL usage. The motivation was considerations for when
EXPORT_SYMBOL_GPL is warranted and the criteria for taking the exceptional
step of reclassifying an existing export. Specifically, I wanted to make
the case that although the line is fuzzy and hard to specify in abstract
terms, it is nonetheless clear that devm_memremap_pages() and HMM
(Heterogeneous Memory Management) have crossed it. The
devm_memremap_pages() facility should have been EXPORT_SYMBOL_GPL from the
beginning, and HMM as a derivative of that functionality should have
naturally picked up that designation as well.
Contrary to typical rules, the HMM infrastructure was merged upstream with
zero in-tree consumers. There was a promise at the time that those users
would be merged "soon", but it has been over a year with no drivers
arriving. While the Nouveau driver is about to belatedly make good on
that promise it is clear that HMM was targeted first and foremost at an
out-of-tree consumer.
HMM is derived from devm_memremap_pages(), a facility Christoph and I
spearheaded to support persistent memory. It combines a device lifetime
model with a dynamically created 'struct page' / memmap array for any
physical address range. It enables coordination and control of the many
code paths in the kernel built to interact with memory via 'struct page'
objects. With HMM the integration goes even deeper by allowing device
drivers to hook and manipulate page fault and page free events.
One interpretation of when EXPORT_SYMBOL is suitable is when it is
exporting stable and generic leaf functionality. The
devm_memremap_pages() facility continues to see expanding use cases,
peer-to-peer DMA being the most recent, with no clear end date when it
will stop attracting reworks and semantic changes. It is not suitable to
export devm_memremap_pages() as a stable 3rd party driver API due to the
fact that it is still changing and manipulates core behavior. Moreover,
it is not in the best interest of the long term development of the core
memory management subsystem to permit any external driver to effectively
define its own system-wide memory management policies with no
encouragement to engage with upstream.
I am also concerned that HMM was designed in a way to minimize further
engagement with the core-MM. That, with these hooks in place,
device-drivers are free to implement their own policies without much
consideration for whether and how the core-MM could grow to meet that
need. Going forward not only should HMM be EXPORT_SYMBOL_GPL, but the
core-MM should be allowed the opportunity and stimulus to change and
address these new use cases as first class functionality.
Original changelog:
hmm_devmem_add(), and hmm_devmem_add_resource() duplicated
devm_memremap_pages() and are now simple now wrappers around the core
facility to inject a dev_pagemap instance into the global pgmap_radix and
hook page-idle events. The devm_memremap_pages() interface is base
infrastructure for HMM. HMM has more and deeper ties into the kernel
memory management implementation than base ZONE_DEVICE which is itself a
EXPORT_SYMBOL_GPL facility.
Originally, the HMM page structure creation routines copied the
devm_memremap_pages() code and reused ZONE_DEVICE. A cleanup to unify the
implementations was discussed during the initial review:
http://lkml.iu.edu/hypermail/linux/kernel/1701.2/00812.html Recent work to
extend devm_memremap_pages() for the peer-to-peer-DMA facility enabled
this cleanup to move forward.
In addition to the integration with devm_memremap_pages() HMM depends on
other GPL-only symbols:
mmu_notifier_unregister_no_release
percpu_ref
region_intersects
__class_create
It goes further to consume / indirectly expose functionality that is not
exported to any other driver:
alloc_pages_vma
walk_page_range
HMM is derived from devm_memremap_pages(), and extends deep core-kernel
fundamentals. Similar to devm_memremap_pages(), mark its entry points
EXPORT_SYMBOL_GPL().
[[email protected]: PCI/P2PDMA: match interface changes to devm_memremap_pages()]
Link: http://lkml.kernel.org/r/[email protected]
Link: http://lkml.kernel.org/r/154275560565.76910.15919297436557795278.stgit@dwillia2-desk3.amr.corp.intel.com
Signed-off-by: Dan Williams <[email protected]>
Signed-off-by: Logan Gunthorpe <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Cc: Logan Gunthorpe <[email protected]>
Cc: "Jérôme Glisse" <[email protected]>
Cc: Balbir Singh <[email protected]>,
Cc: Michal Hocko <[email protected]>
Cc: Benjamin Herrenschmidt <[email protected]>
Cc: <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Commit e8d513483300 ("memremap: change devm_memremap_pages interface to
use struct dev_pagemap") refactored devm_memremap_pages() to allow a
dev_pagemap instance to be supplied. Passing in a dev_pagemap interface
simplifies the design of pgmap type drivers in that they can rely on
container_of() to lookup any private data associated with the given
dev_pagemap instance.
In addition to the cleanups this also gives hmm users multi-order-radix
improvements that arrived with commit ab1b597ee0e4 "mm,
devm_memremap_pages: use multi-order radix for ZONE_DEVICE lookups"
As part of the conversion to the devm_memremap_pages() method of
handling the percpu_ref relative to when pages are put, the percpu_ref
completion needs to move to hmm_devmem_ref_exit(). See 71389703839e
("mm, zone_device: Replace {get, put}_zone_device_page...") for details.
Link: http://lkml.kernel.org/r/154275560053.76910.10870962637383152392.stgit@dwillia2-desk3.amr.corp.intel.com
Signed-off-by: Dan Williams <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Reviewed-by: Jérôme Glisse <[email protected]>
Acked-by: Balbir Singh <[email protected]>
Cc: Logan Gunthorpe <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
devm semantics arrange for resources to be torn down when
device-driver-probe fails or when device-driver-release completes.
Similar to devm_memremap_pages() there is no need to support an explicit
remove operation when the users properly adhere to devm semantics.
Note that devm_kzalloc() automatically handles allocating node-local
memory.
Link: http://lkml.kernel.org/r/154275559545.76910.9186690723515469051.stgit@dwillia2-desk3.amr.corp.intel.com
Signed-off-by: Dan Williams <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Reviewed-by: Jérôme Glisse <[email protected]>
Cc: "Jérôme Glisse" <[email protected]>
Cc: Logan Gunthorpe <[email protected]>
Cc: Balbir Singh <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
In preparation for consolidating all ZONE_DEVICE enabling via
devm_memremap_pages(), teach it how to handle the constraints of
MEMORY_DEVICE_PRIVATE ranges.
[[email protected]: call move_pfn_range_to_zone for MEMORY_DEVICE_PRIVATE]
Link: http://lkml.kernel.org/r/154275559036.76910.12434636179931292607.stgit@dwillia2-desk3.amr.corp.intel.com
Signed-off-by: Dan Williams <[email protected]>
Reviewed-by: Jérôme Glisse <[email protected]>
Acked-by: Christoph Hellwig <[email protected]>
Reported-by: Logan Gunthorpe <[email protected]>
Reviewed-by: Logan Gunthorpe <[email protected]>
Cc: Balbir Singh <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
The last step before devm_memremap_pages() returns success is to allocate
a release action, devm_memremap_pages_release(), to tear the entire setup
down. However, the result from devm_add_action() is not checked.
Checking the error from devm_add_action() is not enough. The api
currently relies on the fact that the percpu_ref it is using is killed by
the time the devm_memremap_pages_release() is run. Rather than continue
this awkward situation, offload the responsibility of killing the
percpu_ref to devm_memremap_pages_release() directly. This allows
devm_memremap_pages() to do the right thing relative to init failures and
shutdown.
Without this change we could fail to register the teardown of
devm_memremap_pages(). The likelihood of hitting this failure is tiny as
small memory allocations almost always succeed. However, the impact of
the failure is large given any future reconfiguration, or disable/enable,
of an nvdimm namespace will fail forever as subsequent calls to
devm_memremap_pages() will fail to setup the pgmap_radix since there will
be stale entries for the physical address range.
An argument could be made to require that the ->kill() operation be set in
the @pgmap arg rather than passed in separately. However, it helps code
readability, tracking the lifetime of a given instance, to be able to grep
the kill routine directly at the devm_memremap_pages() call site.
Link: http://lkml.kernel.org/r/154275558526.76910.7535251937849268605.stgit@dwillia2-desk3.amr.corp.intel.com
Signed-off-by: Dan Williams <[email protected]>
Fixes: e8d513483300 ("memremap: change devm_memremap_pages interface...")
Reviewed-by: "Jérôme Glisse" <[email protected]>
Reported-by: Logan Gunthorpe <[email protected]>
Reviewed-by: Logan Gunthorpe <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Cc: Balbir Singh <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Given the fact that devm_memremap_pages() requires a percpu_ref that is
torn down by devm_memremap_pages_release() the current support for mapping
RAM is broken.
Support for remapping "System RAM" has been broken since the beginning and
there is no existing user of this this code path, so just kill the support
and make it an explicit error.
This cleanup also simplifies a follow-on patch to fix the error path when
setting a devm release action for devm_memremap_pages_release() fails.
Link: http://lkml.kernel.org/r/154275557997.76910.14689813630968180480.stgit@dwillia2-desk3.amr.corp.intel.com
Signed-off-by: Dan Williams <[email protected]>
Reviewed-by: "Jérôme Glisse" <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Reviewed-by: Logan Gunthorpe <[email protected]>
Cc: Balbir Singh <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
devm_memremap_pages() is a facility that can create struct page entries
for any arbitrary range and give drivers the ability to subvert core
aspects of page management.
Specifically the facility is tightly integrated with the kernel's memory
hotplug functionality. It injects an altmap argument deep into the
architecture specific vmemmap implementation to allow allocating from
specific reserved pages, and it has Linux specific assumptions about page
structure reference counting relative to get_user_pages() and
get_user_pages_fast(). It was an oversight and a mistake that this was
not marked EXPORT_SYMBOL_GPL from the outset.
Again, devm_memremap_pagex() exposes and relies upon core kernel internal
assumptions and will continue to evolve along with 'struct page', memory
hotplug, and support for new memory types / topologies. Only an in-kernel
GPL-only driver is expected to keep up with this ongoing evolution. This
interface, and functionality derived from this interface, is not suitable
for kernel-external drivers.
Link: http://lkml.kernel.org/r/154275557457.76910.16923571232582744134.stgit@dwillia2-desk3.amr.corp.intel.com
Signed-off-by: Dan Williams <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Cc: "Jérôme Glisse" <[email protected]>
Cc: Balbir Singh <[email protected]>
Cc: Logan Gunthorpe <[email protected]>
Cc: <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
fallbacks
In the enum migratetype definition, MIGRATE_MOVABLE is before
MIGRATE_RECLAIMABLE. Change the order of them to match the enumeration's
order.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Huang Shijie <[email protected]>
Reviewed-by: Andrew Morton <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Reference counters should use refcount_t rather than atomic_t, since the
refcount_t implementation can prevent overflows, reducing the
exploitability of reference leak bugs. userfaultfd_ctx::refcount is a
reference counter with the usual semantics, so convert it to refcount_t.
Note: I replaced the BUG() on incrementing a 0 refcount with just
refcount_inc(), since part of the semantics of refcount_t is that that
incrementing a 0 refcount is not allowed; with CONFIG_REFCOUNT_FULL,
refcount_inc() already checks for it and warns.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Eric Biggers <[email protected]>
Reviewed-by: Andrew Morton <[email protected]>
Cc: Andrea Arcangeli <[email protected]>
Reviewed-by: Mike Rapoport <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Since a2468cc9bfdf ("swap: choose swap device according to numa node"),
avail_lists field of swap_info_struct is changed to an array with
MAX_NUMNODES elements. This made swap_info_struct size increased to 40KiB
and needs an order-4 page to hold it.
This is not optimal in that:
1 Most systems have way less than MAX_NUMNODES(1024) nodes so it
is a waste of memory;
2 It could cause swapon failure if the swap device is swapped on
after system has been running for a while, due to no order-4
page is available as pointed out by Vasily Averin.
Solve the above two issues by using nr_node_ids(which is the actual
possible node number the running system has) for avail_lists instead of
MAX_NUMNODES.
nr_node_ids is unknown at compile time so can't be directly used when
declaring this array. What I did here is to declare avail_lists as zero
element array and allocate space for it when allocating space for
swap_info_struct. The reason why keep using array but not pointer is
plist_for_each_entry needs the field to be part of the struct, so pointer
will not work.
This patch is on top of Vasily Averin's fix commit. I think the use of
kvzalloc for swap_info_struct is still needed in case nr_node_ids is
really big on some systems.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Aaron Lu <[email protected]>
Reviewed-by: Andrew Morton <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Cc: Vasily Averin <[email protected]>
Cc: Huang Ying <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Commit fa5e084e43eb ("vmscan: do not unconditionally treat zones that
fail zone_reclaim() as full") changed the return value of
node_reclaim(). The original return value 0 means NODE_RECLAIM_SOME
after this commit.
While the return value of node_reclaim() when CONFIG_NUMA is n is not
changed. This will leads to call zone_watermark_ok() again.
This patch fixes the return value by adjusting to NODE_RECLAIM_NOSCAN.
Since node_reclaim() is only called in page_alloc.c, move it to
mm/internal.h.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Wei Yang <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Reviewed-by: Matthew Wilcox <[email protected]>
Cc: Mel Gorman <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Now that totalram_pages and managed_pages are atomic varibles, no need of
managed_page_count spinlock. The lock had really a weak consistency
guarantee. It hasn't been used for anything but the update but no reader
actually cares about all the values being updated to be in sync.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Arun KS <[email protected]>
Reviewed-by: Konstantin Khlebnikov <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Acked-by: Vlastimil Babka <[email protected]>
Cc: David Hildenbrand <[email protected]>
Reviewed-by: Pavel Tatashin <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
totalram_pages and totalhigh_pages are made static inline function.
Main motivation was that managed_page_count_lock handling was complicating
things. It was discussed in length here,
https://lore.kernel.org/patchwork/patch/995739/#1181785 So it seemes
better to remove the lock and convert variables to atomic, with preventing
poteintial store-to-read tearing as a bonus.
[[email protected]: coding style fixes]
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Arun KS <[email protected]>
Suggested-by: Michal Hocko <[email protected]>
Suggested-by: Vlastimil Babka <[email protected]>
Reviewed-by: Konstantin Khlebnikov <[email protected]>
Reviewed-by: Pavel Tatashin <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Acked-by: Vlastimil Babka <[email protected]>
Cc: David Hildenbrand <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
totalram_pages, zone->managed_pages and totalhigh_pages updates are
protected by managed_page_count_lock, but readers never care about it.
Convert these variables to atomic to avoid readers potentially seeing a
store tear.
This patch converts zone->managed_pages. Subsequent patches will convert
totalram_panges, totalhigh_pages and eventually managed_page_count_lock
will be removed.
Main motivation was that managed_page_count_lock handling was complicating
things. It was discussed in length here,
https://lore.kernel.org/patchwork/patch/995739/#1181785 So it seemes
better to remove the lock and convert variables to atomic, with preventing
poteintial store-to-read tearing as a bonus.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Arun KS <[email protected]>
Suggested-by: Michal Hocko <[email protected]>
Suggested-by: Vlastimil Babka <[email protected]>
Reviewed-by: Konstantin Khlebnikov <[email protected]>
Reviewed-by: David Hildenbrand <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Acked-by: Vlastimil Babka <[email protected]>
Reviewed-by: Pavel Tatashin <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Patch series "mm: convert totalram_pages, totalhigh_pages and managed
pages to atomic", v5.
This series converts totalram_pages, totalhigh_pages and
zone->managed_pages to atomic variables.
totalram_pages, zone->managed_pages and totalhigh_pages updates are
protected by managed_page_count_lock, but readers never care about it.
Convert these variables to atomic to avoid readers potentially seeing a
store tear.
Main motivation was that managed_page_count_lock handling was complicating
things. It was discussed in length here,
https://lore.kernel.org/patchwork/patch/995739/#1181785 It seemes better
to remove the lock and convert variables to atomic. With the change,
preventing poteintial store-to-read tearing comes as a bonus.
This patch (of 4):
This is in preparation to a later patch which converts totalram_pages and
zone->managed_pages to atomic variables. Please note that re-reading the
value might lead to a different value and as such it could lead to
unexpected behavior. There are no known bugs as a result of the current
code but it is better to prevent from them in principle.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Arun KS <[email protected]>
Reviewed-by: Konstantin Khlebnikov <[email protected]>
Reviewed-by: David Hildenbrand <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Acked-by: Vlastimil Babka <[email protected]>
Reviewed-by: Pavel Tatashin <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
per_cpu_pageset is cleared by memset, it is not necessary to reset it
again.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Wei Yang <[email protected]>
Reviewed-by: Andrew Morton <[email protected]>
Cc: Michal Hocko <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Per-cpu numa_node provides a default node for each possible cpu. The
association gets initialized during the boot when the architecture
specific code explores cpu->NUMA affinity. When the whole NUMA node is
removed though we are clearing this association
try_offline_node
check_and_unmap_cpu_on_node
unmap_cpu_on_node
numa_clear_node
numa_set_node(cpu, NUMA_NO_NODE)
This means that whoever calls cpu_to_node for a cpu associated with such a
node will get NUMA_NO_NODE. This is problematic for two reasons. First
it is fragile because __alloc_pages_node would simply blow up on an
out-of-bound access. We have encountered this when loading kvm module
BUG: unable to handle kernel paging request at 00000000000021c0
IP: __alloc_pages_nodemask+0x93/0xb70
PGD 800000ffe853e067 PUD 7336bbc067 PMD 0
Oops: 0000 [#1] SMP
[...]
CPU: 88 PID: 1223749 Comm: modprobe Tainted: G W 4.4.156-94.64-default #1
RIP: __alloc_pages_nodemask+0x93/0xb70
RSP: 0018:ffff887354493b40 EFLAGS: 00010202
RAX: 00000000000021c0 RBX: 0000000000000000 RCX: 0000000000000000
RDX: 0000000000000000 RSI: 0000000000000002 RDI: 00000000014000c0
RBP: 00000000014000c0 R08: ffffffffffffffff R09: 0000000000000000
R10: ffff88fffc89e790 R11: 0000000000014000 R12: 0000000000000101
R13: ffffffffa0772cd4 R14: ffffffffa0769ac0 R15: 0000000000000000
FS: 00007fdf2f2f1700(0000) GS:ffff88fffc880000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00000000000021c0 CR3: 00000077205ee000 CR4: 0000000000360670
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
alloc_vmcs_cpu+0x3d/0x90 [kvm_intel]
hardware_setup+0x781/0x849 [kvm_intel]
kvm_arch_hardware_setup+0x28/0x190 [kvm]
kvm_init+0x7c/0x2d0 [kvm]
vmx_init+0x1e/0x32c [kvm_intel]
do_one_initcall+0xca/0x1f0
do_init_module+0x5a/0x1d7
load_module+0x1393/0x1c90
SYSC_finit_module+0x70/0xa0
entry_SYSCALL_64_fastpath+0x1e/0xb7
DWARF2 unwinder stuck at entry_SYSCALL_64_fastpath+0x1e/0xb7
on an older kernel but the code is basically the same in the current Linus
tree as well. alloc_vmcs_cpu could use alloc_pages_nodemask which would
recognize NUMA_NO_NODE and use alloc_pages_node which would translate it
to numa_mem_id but that is wrong as well because it would use a cpu
affinity of the local CPU which might be quite far from the original node.
It is also reasonable to expect that cpu_to_node will provide a sane
value and there might be many more callers like that.
The second problem is that __register_one_node relies on cpu_to_node to
properly associate cpus back to the node when it is onlined. We do not
want to lose that link as there is no arch independent way to get it from
the early boot time AFAICS.
Drop the whole check_and_unmap_cpu_on_node machinery and keep the
association to fix both issues. The NODE_DATA(nid) is not deallocated so
it will stay in place and if anybody wants to allocate from that node then
a fallback node will be used.
Thanks to Vlastimil Babka for his live system debugging skills that helped
debugging the issue.
Link: http://lkml.kernel.org/r/[email protected]
Fixes: e13fe8695c57 ("cpu-hotplug,memory-hotplug: clear cpu_to_node() when offlining the node")
Signed-off-by: Michal Hocko <[email protected]>
Debugged-by: Vlastimil Babka <[email protected]>
Reported-by: Miroslav Benes <[email protected]>
Acked-by: Anshuman Khandual <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
We should get rid of this function. It no longer serves its purpose.
This is a historical artifact from 2005 where do_brk was called outside of
the core mm. We do have a proper abstraction in vm_brk_flags and that one
does the locking properly so there is no need to use this function.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Yangtao Li <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Acked-by: David Rientjes <[email protected]>
Cc: Dan Williams <[email protected]>
Cc: Dominik Brodowski <[email protected]>
Cc: Dave Hansen <[email protected]>
Cc: David Woodhouse <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Replace jhash2 with xxhash.
Perf numbers:
Intel(R) Xeon(R) CPU E5-2420 v2 @ 2.20GHz
ksm: crc32c hash() 12081 MB/s
ksm: xxh64 hash() 8770 MB/s
ksm: xxh32 hash() 4529 MB/s
ksm: jhash2 hash() 1569 MB/s
Sioh Lee did some testing:
crc32c_intel: 1084.10ns
crc32c (no hardware acceleration): 7012.51ns
xxhash32: 2227.75ns
xxhash64: 1413.16ns
jhash2: 5128.30ns
As jhash2 always will be slower (for data size like PAGE_SIZE). Don't use
it in ksm at all.
Use only xxhash for now, because for using crc32c, cryptoapi must be
initialized first - that requires some tricky solution to work well in all
situations.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Timofey Titovets <[email protected]>
Signed-off-by: leesioh <[email protected]>
Reviewed-by: Pavel Tatashin <[email protected]>
Reviewed-by: Mike Rapoport <[email protected]>
Reviewed-by: Andrew Morton <[email protected]>
Cc: Andrea Arcangeli <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Patch series "Currently used jhash are slow enough and replace it allow as
to make KSM", v8.
Apeed (in kernel):
ksm: crc32c hash() 12081 MB/s
ksm: xxh64 hash() 8770 MB/s
ksm: xxh32 hash() 4529 MB/s
ksm: jhash2 hash() 1569 MB/s
Sioh Lee's testing (copy from other mail):
Test platform: openstack cloud platform (NEWTON version)
Experiment node: openstack based cloud compute node (CPU: xeon E5-2620 v3, memory 64gb)
VM: (2 VCPU, RAM 4GB, DISK 20GB) * 4
Linux kernel: 4.14 (latest version)
KSM setup - sleep_millisecs: 200ms, pages_to_scan: 200
Experiment process:
Firstly, we turn off KSM and launch 4 VMs. Then we turn on the KSM and
measure the checksum computation time until full_scans become two.
The experimental results (the experimental value is the average of the measured values)
crc32c_intel: 1084.10ns
crc32c (no hardware acceleration): 7012.51ns
xxhash32: 2227.75ns
xxhash64: 1413.16ns
jhash2: 5128.30ns
In summary, the result shows that crc32c_intel has advantages over all of
the hash function used in the experiment. (decreased by 84.54% compared
to crc32c, 78.86% compared to jhash2, 51.33% xxhash32, 23.28% compared to
xxhash64) the results are similar to those of Timofey.
But, use only xxhash for now, because for using crc32c, cryptoapi must be
initialized first - that require some tricky solution to work good in all
situations.
So:
- First patch implement compile time pickup of fastest implementation of
xxhash for target platform.
- The second patch replaces jhash2 with xxhash
This patch (of 2):
xxh32() - fast on both 32/64-bit platforms
xxh64() - fast only on 64-bit platform
Create xxhash() which will pick up the fastest version at compile time.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Timofey Titovets <[email protected]>
Reviewed-by: Pavel Tatashin <[email protected]>
Reviewed-by: Mike Rapoport <[email protected]>
Reviewed-by: Andrew Morton <[email protected]>
Cc: Andrea Arcangeli <[email protected]>
Cc: leesioh <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Heiko has complained that his log is swamped by warnings from
has_unmovable_pages
[ 20.536664] page dumped because: has_unmovable_pages
[ 20.536792] page:000003d081ff4080 count:1 mapcount:0 mapping:000000008ff88600 index:0x0 compound_mapcount: 0
[ 20.536794] flags: 0x3fffe0000010200(slab|head)
[ 20.536795] raw: 03fffe0000010200 0000000000000100 0000000000000200 000000008ff88600
[ 20.536796] raw: 0000000000000000 0020004100000000 ffffffff00000001 0000000000000000
[ 20.536797] page dumped because: has_unmovable_pages
[ 20.536814] page:000003d0823b0000 count:1 mapcount:0 mapping:0000000000000000 index:0x0
[ 20.536815] flags: 0x7fffe0000000000()
[ 20.536817] raw: 07fffe0000000000 0000000000000100 0000000000000200 0000000000000000
[ 20.536818] raw: 0000000000000000 0000000000000000 ffffffff00000001 0000000000000000
which are not triggered by the memory hotplug but rather CMA allocator.
The original idea behind dumping the page state for all call paths was
that these messages will be helpful debugging failures. From the above it
seems that this is not the case for the CMA path because we are lacking
much more context. E.g the second reported page might be a CMA allocated
page. It is still interesting to see a slab page in the CMA area but it
is hard to tell whether this is bug from the above output alone.
Address this issue by dumping the page state only on request. Both
start_isolate_page_range and has_unmovable_pages already have an argument
to ignore hwpoison pages so make this argument more generic and turn it
into flags and allow callers to combine non-default modes into a mask.
While we are at it, has_unmovable_pages call from
is_pageblock_removable_nolock (sysfs removable file) is questionable to
report the failure so drop it from there as well.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Michal Hocko <[email protected]>
Reported-by: Heiko Carstens <[email protected]>
Reviewed-by: Oscar Salvador <[email protected]>
Cc: Anshuman Khandual <[email protected]>
Cc: Stephen Rothwell <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
There is only very limited information printed when the memory offlining
fails:
[ 1984.506184] rac1 kernel: memory offlining [mem 0x82600000000-0x8267fffffff] failed due to signal backoff
This tells us that the failure is triggered by the userspace intervention
but it doesn't tell us much more about the underlying reason. It might be
that the page migration failes repeatedly and the userspace timeout
expires and send a signal or it might be some of the earlier steps
(isolation, memory notifier) takes too long.
If the migration failes then it would be really helpful to see which page
that and its state. The same applies to the isolation phase. If we fail
to isolate a page from the allocator then knowing the state of the page
would be helpful as well.
Dump the page state that fails to get isolated or migrated. This will
tell us more about the failure and what to focus on during debugging.
[[email protected]: add missing printk arg]
[[email protected]: tweak dump_page() `reason' text]
Link: http://lkml.kernel.org/r/[email protected]
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Michal Hocko <[email protected]>
Reviewed-by: Andrew Morton <[email protected]>
Reviewed-by: Oscar Salvador <[email protected]>
Reviewed-by: Anshuman Khandual <[email protected]>
Cc: Baoquan He <[email protected]>
Cc: Oscar Salvador <[email protected]>
Cc: William Kucharski <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
The memory offlining failure reporting is inconsistent and insufficient.
Some error paths simply do not report the failure to the log at all. When
we do report there are no details about the reason of the failure and
there are several of them which makes memory offlining failures hard to
debug.
Make sure that the
memory offlining [mem %#010llx-%#010llx] failed
message is printed for all failures and also provide a short textual
reason for the failure e.g.
[ 1984.506184] rac1 kernel: memory offlining [mem 0x82600000000-0x8267fffffff] failed due to signal backoff
this tells us that the offlining has failed because of a signal pending
aka user intervention.
[[email protected]: tweak messages a bit]
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Michal Hocko <[email protected]>
Reviewed-by: Andrew Morton <[email protected]>
Reviewed-by: Oscar Salvador <[email protected]>
Reviewed-by: Anshuman Khandual <[email protected]>
Cc: Baoquan He <[email protected]>
Cc: Oscar Salvador <[email protected]>
Cc: William Kucharski <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
This function is never called from a context which would provide
misaligned pfn range so drop the pointless check.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Michal Hocko <[email protected]>
Reviewed-by: Andrew Morton <[email protected]>
Reviewed-by: Oscar Salvador <[email protected]>
Reviewed-by: Anshuman Khandual <[email protected]>
Cc: Baoquan He <[email protected]>
Cc: Oscar Salvador <[email protected]>
Cc: William Kucharski <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
__dump_page messages use KERN_EMERG resp. KERN_ALERT loglevel (this is
the case since 2004). Most callers of this function are really detecting
a critical page state and BUG right after. On the other hand the function
is called also from contexts which just want to inform about the page
state and those would rather not disrupt logs that much (e.g. some
systems route these messages to the normal console).
Reduce the loglevel to KERN_WARNING to make dump_page easier to reuse for
other contexts while those messages will still make it to the kernel log
in most setups. Even if the loglevel setup filters warnings away those
paths that are really critical already print the more targeted error or
panic and that should make it to the kernel log.
[[email protected]: fix __dump_page()]
Link: http://lkml.kernel.org/r/[email protected]
[[email protected]: s/KERN_WARN/KERN_WARNING/, per Michal]
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Michal Hocko <[email protected]>
Reviewed-by: Andrew Morton <[email protected]>
Reviewed-by: Anshuman Khandual <[email protected]>
Cc: Baoquan He <[email protected]>
Cc: Oscar Salvador <[email protected]>
Cc: Oscar Salvador <[email protected]>
Cc: William Kucharski <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
I have been promissing to improve memory offlining failures debugging for
quite some time. As things stand now we get only very limited information
in the kernel log when the offlining fails. It is usually only
[ 1984.506184] rac1 kernel: memory offlining [mem 0x82600000000-0x8267fffffff] failed
with no further details. We do not know what exactly fails and for what
reason. Whenever I was forced to debug such a failure I've always had to
do a debugging patch to tell me more. We can enable some tracepoints but
it would be much better to get a better picture without using them.
This patch series does 2 things. The first one is to make dump_page more
usable by printing more information about the mapping patch 1. Then it
reduces the log level from emerg to warning so that this function is
usable from less critical context patch 2. Then I have added more
detailed information about the offlining failure patch 4 and finally add
dump_page to isolation and offlining migration paths. Patch 3 is a
trivial cleanup.
This patch (of 6):
__dump_page prints the mapping pointer but that is quite unhelpful for
many reports because the pointer itself only helps to distinguish anon/ksm
mappings from other ones (because of lowest bits set). Sometimes it would
be much more helpful to know what kind of mapping that is actually and if
we know this is a file mapping then also try to resolve the dentry name.
[[email protected]: fix a width vs precision bug in printk]
Link: http://lkml.kernel.org/r/[email protected]
[[email protected]: use %dp to print dentry]
Link: http://lkml.kernel.org/r/[email protected]
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Michal Hocko <[email protected]>
Reviewed-by: Andrew Morton <[email protected]>
Reviewed-by: Anshuman Khandual <[email protected]>
Reviewed-by: William Kucharski <[email protected]>
Cc: Oscar Salvador <[email protected]>
Cc: Baoquan He <[email protected]>
Cc: Oscar Salvador <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
It's a trivial simplification for get_next_ra_size() and clear enough for
humans to understand.
It also fixes potential overflow if ra->size(< ra_pages) is too large.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Gao Xiang <[email protected]>
Reviewed-by: Fengguang Wu <[email protected]>
Reviewed-by: Matthew Wilcox <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
This happened while running in qemu-system-aarch64, the AMBA PL011 UART
driver when enabling CONFIG_DEBUG_TEST_DRIVER_REMOVE.
arch_initcall(pl011_init) came before subsys_initcall(default_bdi_init),
devtmpfs' handle_remove() crashes because the reference count is a NULL
pointer only because wb->bdi hasn't been initialized yet.
Rework so that wb_put have an extra check if wb->bdi before decrement
wb->refcnt and also add a WARN_ON_ONCE to get a warning if it happens
again in other drivers.
Link: http://lkml.kernel.org/r/[email protected]
Fixes: 52ebea749aae ("writeback: make backing_dev_info host cgroup-specific bdi_writebacks")
Signed-off-by: Arnd Bergmann <[email protected]>
Signed-off-by: Anders Roxell <[email protected]>
Co-developed-by: Arnd Bergmann <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Contrary to its name, mmu_notifier_synchronize() does not synchronize the
notifier's SRCU instance, but rather waits for RCU callbacks to finish.
i.e. it invokes rcu_barrier(). The RCU documentation is quite clear on
this matter, explicitly calling out that rcu_barrier() does not imply
synchronize_rcu().
As there are no callers of mmu_notifier_synchronize() and it's unclear
whether any user of mmu_notifier_call_srcu() will ever want to barrier on
their callbacks, simply remove the function.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Sean Christopherson <[email protected]>
Reviewed-by: Andrew Morton <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Jérôme Glisse <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
In hot remove, we try to clear poisoned pages, but a small optimization to
check if num_poisoned_pages is 0 helps remove the iteration through
nr_pages.
[[email protected]: tweak comment text]
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Balbir Singh <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Acked-by: Naoya Horiguchi <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
The (root-only) page owner read might allocate a large size of memory with
a large read count. Allocation fails can easily occur when doing high
order allocations.
Clamp buffer size to PAGE_SIZE to avoid arbitrary size allocation
and avoid allocation fails due to high order allocation.
[[email protected]: use min_t()]
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Miles Chen <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Cc: Joe Perches <[email protected]>
Cc: Matthew Wilcox <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Multiple people have reported the following sparse warning:
./include/linux/slab.h:332:43: warning: dubious: x & !y
The minimal fix would be to change the logical & to boolean &&, which
emits the same code, but Andrew has suggested that the branch-avoiding
tricks are maybe not worthwile. David Laight provided a nice comparison
of disassembly of multiple variants, which shows that the current version
produces a 4 deep dependency chain, and fixing the sparse warning by
changing logical and to multiplication emits an IMUL, making it even more
expensive.
The code as rewritten by this patch yielded the best disassembly, with a
single predictable branch for the most common case, and a ternary operator
for the rest, which gcc seems to compile without a branch or cmov by
itself.
The result should be more readable, without a sparse warning and probably
also faster for the common case.
Link: http://lkml.kernel.org/r/[email protected]
Fixes: 1291523f2c1d ("mm, slab/slub: introduce kmalloc-reclaimable caches")
Reviewed-by: Andrew Morton <[email protected]>
Signed-off-by: Vlastimil Babka <[email protected]>
Reported-by: Bart Van Assche <[email protected]>
Reported-by: Darryl T. Agostinelli <[email protected]>
Reported-by: Masahiro Yamada <[email protected]>
Suggested-by: Andrew Morton <[email protected]>
Suggested-by: David Laight <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
If __cmpxchg_double_slab() fails and (l != m), current code records
transition states of slub action.
Update the action after __cmpxchg_double_slab() success to record the
final state.
[[email protected]: more whitespace cleanup]
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Wei Yang <[email protected]>
Reviewed-by: Andrew Morton <[email protected]>
Cc: Christoph Lameter <[email protected]>
Cc: Pekka Enberg <[email protected]>
Cc: David Rientjes <[email protected]>
Cc: Joonsoo Kim <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
node_match() is a static function and is only invoked in slub.c.
In all three places, `page' is ensured to be valid.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Wei Yang <[email protected]>
Acked-by: Christoph Lameter <[email protected]>
Cc: Pekka Enberg <[email protected]>
Cc: David Rientjes <[email protected]>
Cc: Joonsoo Kim <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
cpu_slab is a per cpu variable which is allocated in all or none. If a
cpu_slab failed to be allocated, the slub is not usable.
We could use cpu_slab without validation in __flush_cpu_slab().
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Wei Yang <[email protected]>
Reviewed-by: Andrew Morton <[email protected]>
Cc: Christoph Lameter <[email protected]>
Cc: Pekka Enberg <[email protected]>
Cc: David Rientjes <[email protected]>
Cc: Joonsoo Kim <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
WARN_ON() already contains an unlikely(), so it's not necessary to use
unlikely.
Also change WARN_ON() back to WARN_ON_ONCE() to avoid potentially
spamming dmesg with user-triggerable large allocations.
[[email protected]: s/WARN_ON/WARN_ON_ONCE/, per Vlastimil]
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Yangtao Li <[email protected]>
Acked-by: Vlastimil Babka <[email protected]>
Reviewed-by: Andrew Morton <[email protected]>
Acked-by: David Rientjes <[email protected]>
Cc: Christoph Lameter <[email protected]>
Cc: Pekka Enberg <[email protected]>
Cc: Joonsoo Kim <[email protected]>
Cc: Dmitry Vyukov <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
For sync io read in ocfs2_read_blocks_sync(), first clear bh uptodate flag
and submit the io, second wait io done, last check whether bh uptodate, if
not return io error.
If two sync io for the same bh were issued, it could be the first io done
and set uptodate flag, but just before check that flag, the second io came
in and cleared uptodate, then ocfs2_read_blocks_sync() for the first io
will return IO error.
Indeed it's not necessary to clear uptodate flag, as the io end handler
end_buffer_read_sync() will set or clear it based on io succeed or failed.
The following message was found from a nfs server but the underlying
storage returned no error.
[4106438.567376] (nfsd,7146,3):ocfs2_get_suballoc_slot_bit:2780 ERROR: read block 1238823695 failed -5
[4106438.567569] (nfsd,7146,3):ocfs2_get_suballoc_slot_bit:2812 ERROR: status = -5
[4106438.567611] (nfsd,7146,3):ocfs2_test_inode_bit:2894 ERROR: get alloc slot and bit failed -5
[4106438.567643] (nfsd,7146,3):ocfs2_test_inode_bit:2932 ERROR: status = -5
[4106438.567675] (nfsd,7146,3):ocfs2_get_dentry:94 ERROR: test inode bit failed -5
Same issue in non sync read ocfs2_read_blocks(), fixed it as well.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Junxiao Bi <[email protected]>
Reviewed-by: Changwei Ge <[email protected]>
Reviewed-by: Yiwen Jiang <[email protected]>
Cc: Joel Becker <[email protected]>
Cc: Joseph Qi <[email protected]>
Cc: Jun Piao <[email protected]>
Cc: Mark Fasheh <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|