- 06 Nov, 2021 40 commits
-
-
Mel Gorman authored
do_writepages throttles on congestion if the writepages() fails due to a lack of memory but congestion_wait() is partially broken as the congestion state is not updated for all BDIs. This patch stalls waiting for a number of pages to complete writeback that located on the local node. The main weakness is that there is no correlation between the location of the inode's pages and locality but that is still better than congestion_wait. Link: https://lkml.kernel.org/r/20211022144651.19914-5-mgorman@techsingularity.netSigned-off-by:
Mel Gorman <mgorman@techsingularity.net> Acked-by:
Vlastimil Babka <vbabka@suse.cz> Cc: Andreas Dilger <adilger.kernel@dilger.ca> Cc: "Darrick J . Wong" <djwong@kernel.org> Cc: Dave Chinner <david@fromorbit.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Jonathan Corbet <corbet@lwn.net> Cc: Matthew Wilcox <willy@infradead.org> Cc: Michal Hocko <mhocko@suse.com> Cc: NeilBrown <neilb@suse.de> Cc: Rik van Riel <riel@surriel.com> Cc: "Theodore Ts'o" <tytso@mit.edu> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org>
-
Mel Gorman authored
Memcg reclaim throttles on congestion if no reclaim progress is made. This makes little sense, it might be due to writeback or a host of other factors. For !memcg reclaim, it's messy. Direct reclaim primarily is throttled in the page allocator if it is failing to make progress. Kswapd throttles if too many pages are under writeback and marked for immediate reclaim. This patch explicitly throttles if reclaim is failing to make progress. [vbabka@suse.cz: Remove redundant code] Link: https://lkml.kernel.org/r/20211022144651.19914-4-mgorman@techsingularity.netSigned-off-by:
-
Mel Gorman authored
Page reclaim throttles on congestion if too many parallel reclaim instances have isolated too many pages. This makes no sense, excessive parallelisation has nothing to do with writeback or congestion. This patch creates an additional workqueue to sleep on when too many pages are isolated. The throttled tasks are woken when the number of isolated pages is reduced or a timeout occurs. There may be some false positive wakeups for GFP_NOIO/GFP_NOFS callers but the tasks will throttle again if necessary. [shy828301@gmail.com: Wake up from compaction context] [vbabka@suse.cz: Account number of throttled tasks only for writeback] Link: https://lkml.kernel.org/r/20211022144651.19914-3-mgorman@techsingularity.netSigned-off-by:
-
Mel Gorman authored
Patch series "Remove dependency on congestion_wait in mm/", v5. This series that removes all calls to congestion_wait in mm/ and deletes wait_iff_congested. It's not a clever implementation but congestion_wait has been broken for a long time [1]. Even if congestion throttling worked, it was never a great idea. While excessive dirty/writeback pages at the tail of the LRU is one possibility that reclaim may be slow, there is also the problem of too many pages being isolated and reclaim failing for other reasons (elevated references, too many pages isolated, excessive LRU contention etc). This series replaces the "congestion" throttling with 3 different types. - If there are too many dirty/writeback pages, sleep until a timeout or enough pages get cleaned - If too many pages are isolated, sleep until enough isolated pages are either reclaimed or put back on the LRU - If no progress is being made, direct reclaim tasks sleep until another task makes progress with acceptable efficiency. This was initially tested with a mix of workloads that used to trigger corner cases that no longer work. A new test case was created called "stutterp" (pagereclaim-stutterp-noreaders in mmtests) using a freshly created XFS filesystem. Note that it may be necessary to increase the timeout of ssh if executing remotely as ssh itself can get throttled and the connection may timeout. stutterp varies the number of "worker" processes from 4 up to NR_CPUS*4 to check the impact as the number of direct reclaimers increase. It has four types of worker. - One "anon latency" worker creates small mappings with mmap() and times how long it takes to fault the mapping reading it 4K at a time - X file writers which is fio randomly writing X files where the total size of the files add up to the allowed dirty_ratio. fio is allowed to run for a warmup period to allow some file-backed pages to accumulate. The duration of the warmup is based on the best-case linear write speed of the storage. - Y file readers which is fio randomly reading small files - Z anon memory hogs which continually map (100-dirty_ratio)% of memory - Total estimated WSS = (100+dirty_ration) percentage of memory X+Y+Z+1 == NR_WORKERS varying from 4 up to NR_CPUS*4 The intent is to maximise the total WSS with a mix of file and anon memory where some anonymous memory must be swapped and there is a high likelihood of dirty/writeback pages reaching the end of the LRU. The test can be configured to have no background readers to stress dirty/writeback pages. The results below are based on having zero readers. The short summary of the results is that the series works and stalls until some event occurs but the timeouts may need adjustment. The test results are not broken down by patch as the series should be treated as one block that replaces a broken throttling mechanism with a working one. Finally, three machines were tested but I'm reporting the worst set of results. The other two machines had much better latencies for example. First the results of the "anon latency" latency stutterp 5.15.0-rc1 5.15.0-rc1 vanilla mm-reclaimcongest-v5r4 Amean mmap-4 31.4003 ( 0.00%) 2661.0198 (-8374.52%) Amean mmap-7 38.1641 ( 0.00%) 149.2891 (-291.18%) Amean mmap-12 60.0981 ( 0.00%) 187.8105 (-212.51%) Amean mmap-21 161.2699 ( 0.00%) 213.9107 ( -32.64%) Amean mmap-30 174.5589 ( 0.00%) 377.7548 (-116.41%) Amean mmap-48 8106.8160 ( 0.00%) 1070.5616 ( 86.79%) Stddev mmap-4 41.3455 ( 0.00%) 27573.9676 (-66591.66%) Stddev mmap-7 53.5556 ( 0.00%) 4608.5860 (-8505.23%) Stddev mmap-12 171.3897 ( 0.00%) 5559.4542 (-3143.75%) Stddev mmap-21 1506.6752 ( 0.00%) 5746.2507 (-281.39%) Stddev mmap-30 557.5806 ( 0.00%) 7678.1624 (-1277.05%) Stddev mmap-48 61681.5718 ( 0.00%) 14507.2830 ( 76.48%) Max-90 mmap-4 31.4243 ( 0.00%) 83.1457 (-164.59%) Max-90 mmap-7 41.0410 ( 0.00%) 41.0720 ( -0.08%) Max-90 mmap-12 66.5255 ( 0.00%) 53.9073 ( 18.97%) Max-90 mmap-21 146.7479 ( 0.00%) 105.9540 ( 27.80%) Max-90 mmap-30 193.9513 ( 0.00%) 64.3067 ( 66.84%) Max-90 mmap-48 277.9137 ( 0.00%) 591.0594 (-112.68%) Max mmap-4 1913.8009 ( 0.00%) 299623.9695 (-15555.96%) Max mmap-7 2423.9665 ( 0.00%) 204453.1708 (-8334.65%) Max mmap-12 6845.6573 ( 0.00%) 221090.3366 (-3129.64%) Max mmap-21 56278.6508 ( 0.00%) 213877.3496 (-280.03%) Max mmap-30 19716.2990 ( 0.00%) 216287.6229 (-997.00%) Max mmap-48 477923.9400 ( 0.00%) 245414.8238 ( 48.65%) For most thread counts, the time to mmap() is unfortunately increased. In earlier versions of the series, this was lower but a large number of throttling events were reaching their timeout increasing the amount of inefficient scanning of the LRU. There is no prioritisation of reclaim tasks making progress based on each tasks rate of page allocation versus progress of reclaim. The variance is also impacted for high worker counts but in all cases, the differences in latency are not statistically significant due to very large maximum outliers. Max-90 shows that 90% of the stalls are comparable but the Max results show the massive outliers which are increased to to stalling. It is expected that this will be very machine dependant. Due to the test design, reclaim is difficult so allocations stall and there are variances depending on whether THPs can be allocated or not. The amount of memory will affect exactly how bad the corner cases are and how often they trigger. The warmup period calculation is not ideal as it's based on linear writes where as fio is randomly writing multiple files from multiple tasks so the start state of the test is variable. For example, these are the latencies on a single-socket machine that had more memory Amean mmap-4 42.2287 ( 0.00%) 49.6838 * -17.65%* Amean mmap-7 216.4326 ( 0.00%) 47.4451 * 78.08%* Amean mmap-12 2412.0588 ( 0.00%) 51.7497 ( 97.85%) Amean mmap-21 5546.2548 ( 0.00%) 51.8862 ( 99.06%) Amean mmap-30 1085.3121 ( 0.00%) 72.1004 ( 93.36%) The overall system CPU usage and elapsed time is as follows 5.15.0-rc3 5.15.0-rc3 vanilla mm-reclaimcongest-v5r4 Duration User 6989.03 983.42 Duration System 7308.12 799.68 Duration Elapsed 2277.67 2092.98 The patches reduce system CPU usage by 89% as the vanilla kernel is rarely stalling. The high-level /proc/vmstats show 5.15.0-rc1 5.15.0-rc1 vanilla mm-reclaimcongest-v5r2 Ops Direct pages scanned 1056608451.00 503594991.00 Ops Kswapd pages scanned 109795048.00 147289810.00 Ops Kswapd pages reclaimed 63269243.00 31036005.00 Ops Direct pages reclaimed 10803973.00 6328887.00 Ops Kswapd efficiency % 57.62 21.07 Ops Kswapd velocity 48204.98 57572.86 Ops Direct efficiency % 1.02 1.26 Ops Direct velocity 463898.83 196845.97 Kswapd scanned less pages but the detailed pattern is different. The vanilla kernel scans slowly over time where as the patches exhibits burst patterns of scan activity. Direct reclaim scanning is reduced by 52% due to stalling. The pattern for stealing pages is also slightly different. Both kernels exhibit spikes but the vanilla kernel when reclaiming shows pages being reclaimed over a period of time where as the patches tend to reclaim in spikes. The difference is that vanilla is not throttling and instead scanning constantly finding some pages over time where as the patched kernel throttles and reclaims in spikes. Ops Percentage direct scans 90.59 77.37 For direct reclaim, vanilla scanned 90.59% of pages where as with the patches, 77.37% were direct reclaim due to throttling Ops Page writes by reclaim 2613590.00 1687131.00 Page writes from reclaim context are reduced. Ops Page writes anon 2932752.00 1917048.00 And there is less swapping. Ops Page reclaim immediate 996248528.00 107664764.00 The number of pages encountered at the tail of the LRU tagged for immediate reclaim but still dirty/writeback is reduced by 89%. Ops Slabs scanned 164284.00 153608.00 Slab scan activity is similar. ftrace was used to gather stall activity Vanilla ------- 1 writeback_wait_iff_congested: usec_timeout=100000 usec_delayed=16000 2 writeback_wait_iff_congested: usec_timeout=100000 usec_delayed=12000 8 writeback_wait_iff_congested: usec_timeout=100000 usec_delayed=8000 29 writeback_wait_iff_congested: usec_timeout=100000 usec_delayed=4000 82394 writeback_wait_iff_congested: usec_timeout=100000 usec_delayed=0 The fast majority of wait_iff_congested calls do not stall at all. What is likely happening is that cond_resched() reschedules the task for a short period when the BDI is not registering congestion (which it never will in this test setup). 1 writeback_congestion_wait: usec_timeout=100000 usec_delayed=120000 2 writeback_congestion_wait: usec_timeout=100000 usec_delayed=132000 4 writeback_congestion_wait: usec_timeout=100000 usec_delayed=112000 380 writeback_congestion_wait: usec_timeout=100000 usec_delayed=108000 778 writeback_congestion_wait: usec_timeout=100000 usec_delayed=104000 congestion_wait if called always exceeds the timeout as there is no trigger to wake it up. Bottom line: Vanilla will throttle but it's not effective. Patch series ------------ Kswapd throttle activity was always due to scanning pages tagged for immediate reclaim at the tail of the LRU 1 usec_timeout=100000 usect_delayed=72000 reason=VMSCAN_THROTTLE_WRITEBACK 4 usec_timeout=100000 usect_delayed=20000 reason=VMSCAN_THROTTLE_WRITEBACK 5 usec_timeout=100000 usect_delayed=12000 reason=VMSCAN_THROTTLE_WRITEBACK 6 usec_timeout=100000 usect_delayed=16000 reason=VMSCAN_THROTTLE_WRITEBACK 11 usec_timeout=100000 usect_delayed=100000 reason=VMSCAN_THROTTLE_WRITEBACK 11 usec_timeout=100000 usect_delayed=8000 reason=VMSCAN_THROTTLE_WRITEBACK 94 usec_timeout=100000 usect_delayed=0 reason=VMSCAN_THROTTLE_WRITEBACK 112 usec_timeout=100000 usect_delayed=4000 reason=VMSCAN_THROTTLE_WRITEBACK The majority of events did not stall or stalled for a short period. Roughly 16% of stalls reached the timeout before expiry. For direct reclaim, the number of times stalled for each reason were 6624 reason=VMSCAN_THROTTLE_ISOLATED 93246 reason=VMSCAN_THROTTLE_NOPROGRESS 96934 reason=VMSCAN_THROTTLE_WRITEBACK The most common reason to stall was due to excessive pages tagged for immediate reclaim at the tail of the LRU followed by a failure to make forward. A relatively small number were due to too many pages isolated from the LRU by parallel threads For VMSCAN_THROTTLE_ISOLATED, the breakdown of delays was 9 usec_timeout=20000 usect_delayed=4000 reason=VMSCAN_THROTTLE_ISOLATED 12 usec_timeout=20000 usect_delayed=16000 reason=VMSCAN_THROTTLE_ISOLATED 83 usec_timeout=20000 usect_delayed=20000 reason=VMSCAN_THROTTLE_ISOLATED 6520 usec_timeout=20000 usect_delayed=0 reason=VMSCAN_THROTTLE_ISOLATED Most did not stall at all. A small number reached the timeout. For VMSCAN_THROTTLE_NOPROGRESS, the breakdown of stalls were all over the map 1 usec_timeout=500000 usect_delayed=324000 reason=VMSCAN_THROTTLE_NOPROGRESS 1 usec_timeout=500000 usect_delayed=332000 reason=VMSCAN_THROTTLE_NOPROGRESS 1 usec_timeout=500000 usect_delayed=348000 reason=VMSCAN_THROTTLE_NOPROGRESS 1 usec_timeout=500000 usect_delayed=360000 reason=VMSCAN_THROTTLE_NOPROGRESS 2 usec_timeout=500000 usect_delayed=228000 reason=VMSCAN_THROTTLE_NOPROGRESS 2 usec_timeout=500000 usect_delayed=260000 reason=VMSCAN_THROTTLE_NOPROGRESS 2 usec_timeout=500000 usect_delayed=340000 reason=VMSCAN_THROTTLE_NOPROGRESS 2 usec_timeout=500000 usect_delayed=364000 reason=VMSCAN_THROTTLE_NOPROGRESS 2 usec_timeout=500000 usect_delayed=372000 reason=VMSCAN_THROTTLE_NOPROGRESS 2 usec_timeout=500000 usect_delayed=428000 reason=VMSCAN_THROTTLE_NOPROGRESS 2 usec_timeout=500000 usect_delayed=460000 reason=VMSCAN_THROTTLE_NOPROGRESS 2 usec_timeout=500000 usect_delayed=464000 reason=VMSCAN_THROTTLE_NOPROGRESS 3 usec_timeout=500000 usect_delayed=244000 reason=VMSCAN_THROTTLE_NOPROGRESS 3 usec_timeout=500000 usect_delayed=252000 reason=VMSCAN_THROTTLE_NOPROGRESS 3 usec_timeout=500000 usect_delayed=272000 reason=VMSCAN_THROTTLE_NOPROGRESS 4 usec_timeout=500000 usect_delayed=188000 reason=VMSCAN_THROTTLE_NOPROGRESS 4 usec_timeout=500000 usect_delayed=268000 reason=VMSCAN_THROTTLE_NOPROGRESS 4 usec_timeout=500000 usect_delayed=328000 reason=VMSCAN_THROTTLE_NOPROGRESS 4 usec_timeout=500000 usect_delayed=380000 reason=VMSCAN_THROTTLE_NOPROGRESS 4 usec_timeout=500000 usect_delayed=392000 reason=VMSCAN_THROTTLE_NOPROGRESS 4 usec_timeout=500000 usect_delayed=432000 reason=VMSCAN_THROTTLE_NOPROGRESS 5 usec_timeout=500000 usect_delayed=204000 reason=VMSCAN_THROTTLE_NOPROGRESS 5 usec_timeout=500000 usect_delayed=220000 reason=VMSCAN_THROTTLE_NOPROGRESS 5 usec_timeout=500000 usect_delayed=412000 reason=VMSCAN_THROTTLE_NOPROGRESS 5 usec_timeout=500000 usect_delayed=436000 reason=VMSCAN_THROTTLE_NOPROGRESS 6 usec_timeout=500000 usect_delayed=488000 reason=VMSCAN_THROTTLE_NOPROGRESS 7 usec_timeout=500000 usect_delayed=212000 reason=VMSCAN_THROTTLE_NOPROGRESS 7 usec_timeout=500000 usect_delayed=300000 reason=VMSCAN_THROTTLE_NOPROGRESS 7 usec_timeout=500000 usect_delayed=316000 reason=VMSCAN_THROTTLE_NOPROGRESS 7 usec_timeout=500000 usect_delayed=472000 reason=VMSCAN_THROTTLE_NOPROGRESS 8 usec_timeout=500000 usect_delayed=248000 reason=VMSCAN_THROTTLE_NOPROGRESS 8 usec_timeout=500000 usect_delayed=356000 reason=VMSCAN_THROTTLE_NOPROGRESS 8 usec_timeout=500000 usect_delayed=456000 reason=VMSCAN_THROTTLE_NOPROGRESS 9 usec_timeout=500000 usect_delayed=124000 reason=VMSCAN_THROTTLE_NOPROGRESS 9 usec_timeout=500000 usect_delayed=376000 reason=VMSCAN_THROTTLE_NOPROGRESS 9 usec_timeout=500000 usect_delayed=484000 reason=VMSCAN_THROTTLE_NOPROGRESS 10 usec_timeout=500000 usect_delayed=172000 reason=VMSCAN_THROTTLE_NOPROGRESS 10 usec_timeout=500000 usect_delayed=420000 reason=VMSCAN_THROTTLE_NOPROGRESS 10 usec_timeout=500000 usect_delayed=452000 reason=VMSCAN_THROTTLE_NOPROGRESS 11 usec_timeout=500000 usect_delayed=256000 reason=VMSCAN_THROTTLE_NOPROGRESS 12 usec_timeout=500000 usect_delayed=112000 reason=VMSCAN_THROTTLE_NOPROGRESS 12 usec_timeout=500000 usect_delayed=116000 reason=VMSCAN_THROTTLE_NOPROGRESS 12 usec_timeout=500000 usect_delayed=144000 reason=VMSCAN_THROTTLE_NOPROGRESS 12 usec_timeout=500000 usect_delayed=152000 reason=VMSCAN_THROTTLE_NOPROGRESS 12 usec_timeout=500000 usect_delayed=264000 reason=VMSCAN_THROTTLE_NOPROGRESS 12 usec_timeout=500000 usect_delayed=384000 reason=VMSCAN_THROTTLE_NOPROGRESS 12 usec_timeout=500000 usect_delayed=424000 reason=VMSCAN_THROTTLE_NOPROGRESS 12 usec_timeout=500000 usect_delayed=492000 reason=VMSCAN_THROTTLE_NOPROGRESS 13 usec_timeout=500000 usect_delayed=184000 reason=VMSCAN_THROTTLE_NOPROGRESS 13 usec_timeout=500000 usect_delayed=444000 reason=VMSCAN_THROTTLE_NOPROGRESS 14 usec_timeout=500000 usect_delayed=308000 reason=VMSCAN_THROTTLE_NOPROGRESS 14 usec_timeout=500000 usect_delayed=440000 reason=VMSCAN_THROTTLE_NOPROGRESS 14 usec_timeout=500000 usect_delayed=476000 reason=VMSCAN_THROTTLE_NOPROGRESS 16 usec_timeout=500000 usect_delayed=140000 reason=VMSCAN_THROTTLE_NOPROGRESS 17 usec_timeout=500000 usect_delayed=232000 reason=VMSCAN_THROTTLE_NOPROGRESS 17 usec_timeout=500000 usect_delayed=240000 reason=VMSCAN_THROTTLE_NOPROGRESS 17 usec_timeout=500000 usect_delayed=280000 reason=VMSCAN_THROTTLE_NOPROGRESS 18 usec_timeout=500000 usect_delayed=404000 reason=VMSCAN_THROTTLE_NOPROGRESS 20 usec_timeout=500000 usect_delayed=148000 reason=VMSCAN_THROTTLE_NOPROGRESS 20 usec_timeout=500000 usect_delayed=216000 reason=VMSCAN_THROTTLE_NOPROGRESS 20 usec_timeout=500000 usect_delayed=468000 reason=VMSCAN_THROTTLE_NOPROGRESS 21 usec_timeout=500000 usect_delayed=448000 reason=VMSCAN_THROTTLE_NOPROGRESS 23 usec_timeout=500000 usect_delayed=168000 reason=VMSCAN_THROTTLE_NOPROGRESS 23 usec_timeout=500000 usect_delayed=296000 reason=VMSCAN_THROTTLE_NOPROGRESS 25 usec_timeout=500000 usect_delayed=132000 reason=VMSCAN_THROTTLE_NOPROGRESS 25 usec_timeout=500000 usect_delayed=352000 reason=VMSCAN_THROTTLE_NOPROGRESS 26 usec_timeout=500000 usect_delayed=180000 reason=VMSCAN_THROTTLE_NOPROGRESS 27 usec_timeout=500000 usect_delayed=284000 reason=VMSCAN_THROTTLE_NOPROGRESS 28 usec_timeout=500000 usect_delayed=164000 reason=VMSCAN_THROTTLE_NOPROGRESS 29 usec_timeout=500000 usect_delayed=136000 reason=VMSCAN_THROTTLE_NOPROGRESS 30 usec_timeout=500000 usect_delayed=200000 reason=VMSCAN_THROTTLE_NOPROGRESS 30 usec_timeout=500000 usect_delayed=400000 reason=VMSCAN_THROTTLE_NOPROGRESS 31 usec_timeout=500000 usect_delayed=196000 reason=VMSCAN_THROTTLE_NOPROGRESS 32 usec_timeout=500000 usect_delayed=156000 reason=VMSCAN_THROTTLE_NOPROGRESS 33 usec_timeout=500000 usect_delayed=224000 reason=VMSCAN_THROTTLE_NOPROGRESS 35 usec_timeout=500000 usect_delayed=128000 reason=VMSCAN_THROTTLE_NOPROGRESS 35 usec_timeout=500000 usect_delayed=176000 reason=VMSCAN_THROTTLE_NOPROGRESS 36 usec_timeout=500000 usect_delayed=368000 reason=VMSCAN_THROTTLE_NOPROGRESS 36 usec_timeout=500000 usect_delayed=496000 reason=VMSCAN_THROTTLE_NOPROGRESS 37 usec_timeout=500000 usect_delayed=312000 reason=VMSCAN_THROTTLE_NOPROGRESS 38 usec_timeout=500000 usect_delayed=304000 reason=VMSCAN_THROTTLE_NOPROGRESS 40 usec_timeout=500000 usect_delayed=288000 reason=VMSCAN_THROTTLE_NOPROGRESS 43 usec_timeout=500000 usect_delayed=408000 reason=VMSCAN_THROTTLE_NOPROGRESS 55 usec_timeout=500000 usect_delayed=416000 reason=VMSCAN_THROTTLE_NOPROGRESS 56 usec_timeout=500000 usect_delayed=76000 reason=VMSCAN_THROTTLE_NOPROGRESS 58 usec_timeout=500000 usect_delayed=120000 reason=VMSCAN_THROTTLE_NOPROGRESS 59 usec_timeout=500000 usect_delayed=208000 reason=VMSCAN_THROTTLE_NOPROGRESS 61 usec_timeout=500000 usect_delayed=68000 reason=VMSCAN_THROTTLE_NOPROGRESS 71 usec_timeout=500000 usect_delayed=192000 reason=VMSCAN_THROTTLE_NOPROGRESS 71 usec_timeout=500000 usect_delayed=480000 reason=VMSCAN_THROTTLE_NOPROGRESS 79 usec_timeout=500000 usect_delayed=60000 reason=VMSCAN_THROTTLE_NOPROGRESS 82 usec_timeout=500000 usect_delayed=320000 reason=VMSCAN_THROTTLE_NOPROGRESS 82 usec_timeout=500000 usect_delayed=92000 reason=VMSCAN_THROTTLE_NOPROGRESS 85 usec_timeout=500000 usect_delayed=64000 reason=VMSCAN_THROTTLE_NOPROGRESS 85 usec_timeout=500000 usect_delayed=80000 reason=VMSCAN_THROTTLE_NOPROGRESS 88 usec_timeout=500000 usect_delayed=84000 reason=VMSCAN_THROTTLE_NOPROGRESS 90 usec_timeout=500000 usect_delayed=160000 reason=VMSCAN_THROTTLE_NOPROGRESS 90 usec_timeout=500000 usect_delayed=292000 reason=VMSCAN_THROTTLE_NOPROGRESS 94 usec_timeout=500000 usect_delayed=56000 reason=VMSCAN_THROTTLE_NOPROGRESS 118 usec_timeout=500000 usect_delayed=88000 reason=VMSCAN_THROTTLE_NOPROGRESS 119 usec_timeout=500000 usect_delayed=72000 reason=VMSCAN_THROTTLE_NOPROGRESS 126 usec_timeout=500000 usect_delayed=108000 reason=VMSCAN_THROTTLE_NOPROGRESS 146 usec_timeout=500000 usect_delayed=52000 reason=VMSCAN_THROTTLE_NOPROGRESS 148 usec_timeout=500000 usect_delayed=36000 reason=VMSCAN_THROTTLE_NOPROGRESS 148 usec_timeout=500000 usect_delayed=48000 reason=VMSCAN_THROTTLE_NOPROGRESS 159 usec_timeout=500000 usect_delayed=28000 reason=VMSCAN_THROTTLE_NOPROGRESS 178 usec_timeout=500000 usect_delayed=44000 reason=VMSCAN_THROTTLE_NOPROGRESS 183 usec_timeout=500000 usect_delayed=40000 reason=VMSCAN_THROTTLE_NOPROGRESS 237 usec_timeout=500000 usect_delayed=100000 reason=VMSCAN_THROTTLE_NOPROGRESS 266 usec_timeout=500000 usect_delayed=32000 reason=VMSCAN_THROTTLE_NOPROGRESS 313 usec_timeout=500000 usect_delayed=24000 reason=VMSCAN_THROTTLE_NOPROGRESS 347 usec_timeout=500000 usect_delayed=96000 reason=VMSCAN_THROTTLE_NOPROGRESS 470 usec_timeout=500000 usect_delayed=20000 reason=VMSCAN_THROTTLE_NOPROGRESS 559 usec_timeout=500000 usect_delayed=16000 reason=VMSCAN_THROTTLE_NOPROGRESS 964 usec_timeout=500000 usect_delayed=12000 reason=VMSCAN_THROTTLE_NOPROGRESS 2001 usec_timeout=500000 usect_delayed=104000 reason=VMSCAN_THROTTLE_NOPROGRESS 2447 usec_timeout=500000 usect_delayed=8000 reason=VMSCAN_THROTTLE_NOPROGRESS 7888 usec_timeout=500000 usect_delayed=4000 reason=VMSCAN_THROTTLE_NOPROGRESS 22727 usec_timeout=500000 usect_delayed=0 reason=VMSCAN_THROTTLE_NOPROGRESS 51305 usec_timeout=500000 usect_delayed=500000 reason=VMSCAN_THROTTLE_NOPROGRESS The full timeout is often hit but a large number also do not stall at all. The remainder slept a little allowing other reclaim tasks to make progress. While this timeout could be further increased, it could also negatively impact worst-case behaviour when there is no prioritisation of what task should make progress. For VMSCAN_THROTTLE_WRITEBACK, the breakdown was 1 usec_timeout=100000 usect_delayed=44000 reason=VMSCAN_THROTTLE_WRITEBACK 2 usec_timeout=100000 usect_delayed=76000 reason=VMSCAN_THROTTLE_WRITEBACK 3 usec_timeout=100000 usect_delayed=80000 reason=VMSCAN_THROTTLE_WRITEBACK 5 usec_timeout=100000 usect_delayed=48000 reason=VMSCAN_THROTTLE_WRITEBACK 5 usec_timeout=100000 usect_delayed=84000 reason=VMSCAN_THROTTLE_WRITEBACK 6 usec_timeout=100000 usect_delayed=72000 reason=VMSCAN_THROTTLE_WRITEBACK 7 usec_timeout=100000 usect_delayed=88000 reason=VMSCAN_THROTTLE_WRITEBACK 11 usec_timeout=100000 usect_delayed=56000 reason=VMSCAN_THROTTLE_WRITEBACK 12 usec_timeout=100000 usect_delayed=64000 reason=VMSCAN_THROTTLE_WRITEBACK 16 usec_timeout=100000 usect_delayed=92000 reason=VMSCAN_THROTTLE_WRITEBACK 24 usec_timeout=100000 usect_delayed=68000 reason=VMSCAN_THROTTLE_WRITEBACK 28 usec_timeout=100000 usect_delayed=32000 reason=VMSCAN_THROTTLE_WRITEBACK 30 usec_timeout=100000 usect_delayed=60000 reason=VMSCAN_THROTTLE_WRITEBACK 30 usec_timeout=100000 usect_delayed=96000 reason=VMSCAN_THROTTLE_WRITEBACK 32 usec_timeout=100000 usect_delayed=52000 reason=VMSCAN_THROTTLE_WRITEBACK 42 usec_timeout=100000 usect_delayed=40000 reason=VMSCAN_THROTTLE_WRITEBACK 77 usec_timeout=100000 usect_delayed=28000 reason=VMSCAN_THROTTLE_WRITEBACK 99 usec_timeout=100000 usect_delayed=36000 reason=VMSCAN_THROTTLE_WRITEBACK 137 usec_timeout=100000 usect_delayed=24000 reason=VMSCAN_THROTTLE_WRITEBACK 190 usec_timeout=100000 usect_delayed=20000 reason=VMSCAN_THROTTLE_WRITEBACK 339 usec_timeout=100000 usect_delayed=16000 reason=VMSCAN_THROTTLE_WRITEBACK 518 usec_timeout=100000 usect_delayed=12000 reason=VMSCAN_THROTTLE_WRITEBACK 852 usec_timeout=100000 usect_delayed=8000 reason=VMSCAN_THROTTLE_WRITEBACK 3359 usec_timeout=100000 usect_delayed=4000 reason=VMSCAN_THROTTLE_WRITEBACK 7147 usec_timeout=100000 usect_delayed=0 reason=VMSCAN_THROTTLE_WRITEBACK 83962 usec_timeout=100000 usect_delayed=100000 reason=VMSCAN_THROTTLE_WRITEBACK The majority hit the timeout in direct reclaim context although a sizable number did not stall at all. This is very different to kswapd where only a tiny percentage of stalls due to writeback reached the timeout. Bottom line, the throttling appears to work and the wakeup events may limit worst case stalls. There might be some grounds for adjusting timeouts but it's likely futile as the worst-case scenarios depend on the workload, memory size and the speed of the storage. A better approach to improve the series further would be to prioritise tasks based on their rate of allocation with the caveat that it may be very expensive to track. This patch (of 5): Page reclaim throttles on wait_iff_congested under the following conditions: - kswapd is encountering pages under writeback and marked for immediate reclaim implying that pages are cycling through the LRU faster than pages can be cleaned. - Direct reclaim will stall if all dirty pages are backed by congested inodes. wait_iff_congested is almost completely broken with few exceptions. This patch adds a new node-based workqueue and tracks the number of throttled tasks and pages written back since throttling started. If enough pages belonging to the node are written back then the throttled tasks will wake early. If not, the throttled tasks sleeps until the timeout expires. [neilb@suse.de: Uninterruptible sleep and simpler wakeups] [hdanton@sina.com: Avoid race when reclaim starts] [vbabka@suse.cz: vmstat irq-safe api, clarifications] Link: https://lore.kernel.org/linux-mm/45d8b7a6-8548-65f5-cccf-9f451d4ae3d4@kernel.dk/ [1] Link: https://lkml.kernel.org/r/20211022144651.19914-1-mgorman@techsingularity.net Link: https://lkml.kernel.org/r/20211022144651.19914-2-mgorman@techsingularity.netSigned-off-by: -
Kai Song authored
We fix the following warning when building kernel with W=1: mm/vmscan.c:1362:6: warning: variable 'err' set but not used [-Wunused-but-set-variable] Link: https://lkml.kernel.org/r/20210924181218.21165-1-songkai01@inspur.comSigned-off-by:
Kai Song <songkai01@inspur.com> Reviewed-by:
Yang Shi <shy828301@gmail.com> Signed-off-by:
-
Miaohe Lin authored
Isolating a free page in an isolated pageblock is expected to always work as watermarks don't apply here. But if __isolate_free_page() failed, due to condition changes, the page will be left on the free list. And the page will be put back to free list again via __putback_isolated_page(). This may trigger VM_BUG_ON_PAGE() on page->flags checking in __free_one_page() if PageReported is set. Or we will corrupt the free list because list_add() will be called for pages already on another list. Add a VM_WARN_ON() to complain about this change. Link: https://lkml.kernel.org/r/20210914114508.23725-1-linmiaohe@huawei.com Fixes: 3c605096 ("mm/page_alloc: restrict max order of merging on isolated pageblock") Signed-off-by:
Miaohe Lin <linmiaohe@huawei.com> Reviewed-by:
David Hildenbrand <david@redhat.com> Acked-by:
-
Miaohe Lin authored
In start_isolate_page_range() undo path, pfn_to_online_page() just checks the first pfn in a pageblock while __first_valid_page() will traverse the pageblock until the first online pfn is found. So we may miss the call to unset_migratetype_isolate() in undo path and pages will remain isolated unexpectedly. Fix this by calling undo_isolate_page_range() and this will also help to simplify the code further. Note we shouldn't ever trigger it because MAX_ORDER-1 aligned pfn ranges shouldn't contain memory holes now. Link: https://lkml.kernel.org/r/20210914114348.15569-1-linmiaohe@huawei.com Fixes: 2ce13640 ("mm: __first_valid_page skip over offline pages") Signed-off-by:
-
Axel Rasmussen authored
Today, we assert that the ioctls the kernel reports as supported for a registration match a precomputed list. We decide which ioctls are supported by examining the memory type. Then, in several locations we "fix up" this list by adding or removing things this initial decision got wrong. What ioctls the kernel reports is actually a function of several things: - The memory type - Kernel feature support (e.g., no writeprotect on aarch64) - The registration type (e.g., CONTINUE only supported for MINOR mode) So, we can't fully compute this at the start, in set_test_type. It varies per test, depending on what registration mode(s) those tests use. Instead, introduce a new function which computes the correct list. This centralizes the add/remove of ioctls depending on these function inputs in one place, so we don't have to repeat ourselves in various tests. Not only is the resulting code a bit shorter, but it fixes a real bug in the existing code: previously, we would incorrectly require the writeprotect ioctl to be present on aarch64, where it isn't actually supported. Link: https://lkml.kernel.org/r/20210930212309.4001967-4-axelrasmussen@google.comSigned-off-by:
Axel Rasmussen <axelrasmussen@google.com> Reviewed-by:
Peter Xu <peterx@redhat.com> Cc: Shuah Khan <shuah@kernel.org> Signed-off-by:
-
Axel Rasmussen authored
Before any tests are run, in set_test_type, we decide what feature(s) we are going to be testing, based upon our command line arguments. However, the supported features are not just a function of the memory type being used, so this is broken. For instance, consider writeprotect support. It is "normally" supported for anonymous memory, but furthermore it requires that the kernel has CONFIG_HAVE_ARCH_USERFAULTFD_WP. So, it is *not* supported at all on aarch64, for example. So, this fixes this by querying the kernel for the set of features it supports in set_test_type, by opening a userfaultfd and issuing a UFFDIO_API ioctl. Based upon the reported features, we toggle what tests are enabled. Link: https://lkml.kernel.org/r/20210930212309.4001967-3-axelrasmussen@google.comSigned-off-by:
-
Axel Rasmussen authored
Patch series "Small userfaultfd selftest fixups", v2. This patch (of 3): Two arguments for doing this: First, and maybe most importantly, the resulting code is significantly shorter / simpler. Then, we avoid using GNU libc extensions. Why does this matter? It makes testing userfaultfd with the selftest easier e.g. on distros which use something other than glibc (e.g., Alpine, which uses musl); basically, it makes the test more portable. Link: https://lkml.kernel.org/r/20210930212309.4001967-2-axelrasmussen@google.comSigned-off-by:
-
Mike Kravetz authored
In commit 7118fc29 ("hugetlb: address ref count racing in prep_compound_gigantic_page"), page_ref_freeze is used to atomically zero the ref count of tail pages iff they are 1. The unconditional call to set_page_count(0) was left in the code. This call is after page_ref_freeze so it is really a noop. Remove redundant and unnecessary set_page_count call. Link: https://lkml.kernel.org/r/20211026220635.35187-1-mike.kravetz@oracle.com Fixes: 7118fc29 ("hugetlb: address ref count racing in prep_compound_gigantic_page") Signed-off-by:
Mike Kravetz <mike.kravetz@oracle.com> Suggested-by:
Pasha Tatashin <pasha.tatashin@soleen.com> Reviewed-by:
Matthew Wilcox (Oracle) <willy@infradead.org> Reviewed-by:
Oscar Salvador <osalvador@suse.de> Reviewed-by:
Muchun Song <songmuchun@bytedance.com> Signed-off-by:
-
Baolin Wang authored
When calling hugetlb_resv_map_add(), we've guaranteed that the parameter 'to' is always larger than 'from', so it never returns a negative value from hugetlb_resv_map_add(). Thus remove the redundant VM_BUG_ON(). Link: https://lkml.kernel.org/r/2b565552f3d06753da1e8dda439c0d96d6d9a5a3.1634797639.git.baolin.wang@linux.alibaba.comSigned-off-by:
Baolin Wang <baolin.wang@linux.alibaba.com> Reviewed-by:
-
Baolin Wang authored
The callers of has_same_uncharge_info() has accessed the original file_region and new file_region, and they are impossible to be NULL now. So we can remove the file_region validation in has_same_uncharge_info() to simplify the code. Link: https://lkml.kernel.org/r/97fc68d3f8d34f63c204645e10d7a718997e50b7.1634797639.git.baolin.wang@linux.alibaba.comSigned-off-by:
-
Baolin Wang authored
After commit 8382d914 ("mm, hugetlb: improve page-fault scalability"), the hugetlb_instantiation_mutex lock had been replaced by hugetlb_fault_mutex_table to serializes faults on the same logical page. Thus update the obsolete hugetlb_instantiation_mutex related comments. Link: https://lkml.kernel.org/r/4b3febeae37455ff7b74aa0aad16cc6909cf0926.1634797639.git.baolin.wang@linux.alibaba.comSigned-off-by:
-
Baolin Wang authored
Patch series "Some cleanups and improvements for hugetlb". This patchset does some cleanups and improvements for hugetlb and hugetlb_cgroup. This patch (of 4): Since commit 726b7bbe ("hugetlb_cgroup: fix illegal access to memory"), the hugetlb_cgroup_from_counter() macro is not used any more, remove it. Link: https://lkml.kernel.org/r/cover.1634797639.git.baolin.wang@linux.alibaba.com Link: https://lkml.kernel.org/r/f03b29b801fa9942466ab15334ec09988e124ae6.1634797639.git.baolin.wang@linux.alibaba.comSigned-off-by:
-
Ran Jianping authored
Remove duplicate includes 'unistd.h' included in '/tools/testing/selftests/vm/hugepage-mremap.c' is duplicated.It is also included on 23 line. Link: https://lkml.kernel.org/r/20211018102336.869726-1-ran.jianping@zte.com.cnSigned-off-by:
Ran Jianping <ran.jianping@zte.com.cn> Reported-by:
Zeal Robot <zealci@zte.com.cn> Cc: Shuah Khan <shuah@kernel.org> Signed-off-by:
-
Baolin Wang authored
Now the size of CMA area for gigantic hugepages runtime allocation is balanced for all online nodes, but we also want to specify the size of CMA per-node, or only one node in some cases, which are similar with patch [1]. For example, on some multi-nodes systems, each node's memory can be different, allocating the same size of CMA for each node is not suitable for the low-memory nodes. Meanwhile some workloads like DPDK mentioned by Zhenguo in patch [1] only need hugepages in one node. On the other hand, we have some machines with multiple types of memory, like DRAM and PMEM (persistent memory). On this system, we may want to specify all the hugepages only on DRAM node, or specify the proportion of DRAM node and PMEM node, to tuning the performance of the workloads. Thus this patch adds node format for 'hugetlb_cma' parameter to support specifying the size of CMA per-node. An example is as follows: hugetlb_cma=0:5G,2:5G which means allocating 5G size of CMA area on node 0 and node 2 respectively. And the users should use the node specific sysfs file to allocate the gigantic hugepages if specified the CMA size on that node. Link: https://lkml.kernel.org/r/20211005054729.86457-1-yaozhenguo1@gmail.com [1] Link: https://lkml.kernel.org/r/bb790775ca60bb8f4b26956bb3f6988f74e075c7.1634261144.git.baolin.wang@linux.alibaba.comSigned-off-by:
-
Mina Almasry authored
[almasrymina@google.com: v8] Link: https://lkml.kernel.org/r/20211014200542.4126947-2-almasrymina@google.com [wanjiabing@vivo.com: remove duplicated include in hugepage-mremap] Link: https://lkml.kernel.org/r/20211021122944.8857-1-wanjiabing@vivo.com Link: https://lkml.kernel.org/r/20211013195825.3058275-2-almasrymina@google.comSigned-off-by:
Mina Almasry <almasrymina@google.com> Signed-off-by:
Wan Jiabing <wanjiabing@vivo.com> Acked-by:
-
Mina Almasry authored
Support mremap() for hugepage backed vma segment by simply repositioning page table entries. The page table entries are repositioned to the new virtual address on mremap(). Hugetlb mremap() support is of course generic; my motivating use case is a library (hugepage_text), which reloads the ELF text of executables in hugepages. This significantly increases the execution performance of said executables. Restrict the mremap operation on hugepages to up to the size of the original mapping as the underlying hugetlb reservation is not yet capable of handling remapping to a larger size. During the mremap() operation we detect pmd_share'd mappings and we unshare those during the mremap(). On access and fault the sharing is established again. Link: https://lkml.kernel.org/r/20211013195825.3058275-1-almasrymina@google.comSigned-off-by:
-
Liangcai Fan authored
When initializing transparent huge pages, min_free_kbytes would be calculated according to what khugepaged expected. So when transparent huge pages get disabled, min_free_kbytes should be recalculated instead of the higher value set by khugepaged. Link: https://lkml.kernel.org/r/1633937809-16558-1-git-send-email-liangcaifan19@gmail.comSigned-off-by:
Liangcai Fan <liangcaifan19@gmail.com> Signed-off-by:
Chunyan Zhang <zhang.lyra@gmail.com> Cc: Mike Kravetz <mike.kravetz@oracle.com> Signed-off-by:
-
Mike Kravetz authored
Demote page functionality will split a huge page into a number of huge pages of a smaller size. For example, on x86 a 1GB huge page can be demoted into 512 2M huge pages. Demotion is done 'in place' by simply splitting the huge page. Added '*_for_demote' wrappers for remove_hugetlb_page, destroy_compound_hugetlb_page and prep_compound_gigantic_page for use by demote code. [mike.kravetz@oracle.com: v4] Link: https://lkml.kernel.org/r/6ca29b8e-527c-d6ec-900e-e6a43e4f8b73@oracle.com Link: https://lkml.kernel.org/r/20211007181918.136982-6-mike.kravetz@oracle.comSigned-off-by:
-
Mike Kravetz authored
The routines remove_hugetlb_page and destroy_compound_gigantic_page will remove a gigantic page and make the set of base pages ready to be returned to a lower level allocator. In the process of doing this, they make all base pages reference counted. The routine prep_compound_gigantic_page creates a gigantic page from a set of base pages. It assumes that all these base pages are reference counted. During demotion, a gigantic page will be split into huge pages of a smaller size. This logically involves use of the routines, remove_hugetlb_page, and destroy_compound_gigantic_page followed by prep_compound*_page for each smaller huge page. When pages are reference counted (ref count >= 0), additional speculative ref counts could be taken as described in previous commits [1] and [2]. This could result in errors while demoting a huge page. Quite a bit of code would need to be created to handle all possible issues. Instead of dealing with the possibility of speculative ref counts, avoid the possibility by keeping ref counts at zero during the demote process. Add a boolean 'demote' to the routines remove_hugetlb_page, destroy_compound_gigantic_page and prep_compound_gigantic_page. If the boolean is set, the remove and destroy routines will not reference count pages and the prep routine will not expect reference counted pages. '*_for_demote' wrappers of the routines will be added in a subsequent patch where this functionality is used. [1] https://lore.kernel.org/linux-mm/20210622021423.154662-3-mike.kravetz@oracle.com/ [2] https://lore.kernel.org/linux-mm/20210809184832.18342-3-mike.kravetz@oracle.com/ Link: https://lkml.kernel.org/r/20211007181918.136982-5-mike.kravetz@oracle.comSigned-off-by:
-
Mike Kravetz authored
When huge page demotion is fully implemented, gigantic pages can be demoted to a smaller huge page size. For example, on x86 a 1G page can be demoted to 512 2M pages. However, gigantic pages can potentially be allocated from CMA. If a gigantic page which was allocated from CMA is demoted, the corresponding demoted pages needs to be returned to CMA. Use the new interface cma_pages_valid() to determine if a non-gigantic hugetlb page should be freed to CMA. Also, clear mapping field of these pages as expected by cma_release. This also requires a change to CMA region creation for gigantic pages. CMA uses a per-region bit map to track allocations. When setting up the region, you specify how many pages each bit represents. Currently, only gigantic pages are allocated/freed from CMA so the region is set up such that one bit represents a gigantic page size allocation. With demote, a gigantic page (allocation) could be split into smaller size pages. And, these smaller size pages will be freed to CMA. So, since the per-region bit map needs to be set up to represent the smallest allocation/free size, it now needs to be set to the smallest huge page size which can be freed to CMA. Unfortunately, we set up the CMA region for huge pages before we set up huge pages sizes (hstates). So, technically we do not know the smallest huge page size as this can change via command line options and architecture specific code. Therefore, at region setup time we use HUGETLB_PAGE_ORDER as the smallest possible huge page size that can be given back to CMA. It is possible that this value is sub-optimal for some architectures/config options. If needed, this can be addressed in follow on work. Link: https://lkml.kernel.org/r/20211007181918.136982-4-mike.kravetz@oracle.comSigned-off-by:
-
Mike Kravetz authored
Add new interface cma_pages_valid() which indicates if the specified pages are part of a CMA region. This interface will be used in a subsequent patch by hugetlb code. In order to keep the same amount of DEBUG information, a pr_debug() call was added to cma_pages_valid(). In the case where the page passed to cma_release is not in cma region, the debug message will be printed from cma_pages_valid as opposed to cma_release. Link: https://lkml.kernel.org/r/20211007181918.136982-3-mike.kravetz@oracle.comSigned-off-by:
-
Mike Kravetz authored
Patch series "hugetlb: add demote/split page functionality", v4. The concurrent use of multiple hugetlb page sizes on a single system is becoming more common. One of the reasons is better TLB support for gigantic page sizes on x86 hardware. In addition, hugetlb pages are being used to back VMs in hosting environments. When using hugetlb pages to back VMs, it is often desirable to preallocate hugetlb pools. This avoids the delay and uncertainty of allocating hugetlb pages at VM startup. In addition, preallocating huge pages minimizes the issue of memory fragmentation that increases the longer the system is up and running. In such environments, a combination of larger and smaller hugetlb pages are preallocated in anticipation of backing VMs of various sizes. Over time, the preallocated pool of smaller hugetlb pages may become depleted while larger hugetlb pages still remain. In such situations, it is desirable to convert larger hugetlb pages to smaller hugetlb pages. Converting larger to smaller hugetlb pages can be accomplished today by first freeing the larger page to the buddy allocator and then allocating the smaller pages. For example, to convert 50 GB pages on x86: gb_pages=`cat .../hugepages-1048576kB/nr_hugepages` m2_pages=`cat .../hugepages-2048kB/nr_hugepages` echo $(($gb_pages - 50)) > .../hugepages-1048576kB/nr_hugepages echo $(($m2_pages + 25600)) > .../hugepages-2048kB/nr_hugepages On an idle system this operation is fairly reliable and results are as expected. The number of 2MB pages is increased as expected and the time of the operation is a second or two. However, when there is activity on the system the following issues arise: 1) This process can take quite some time, especially if allocation of the smaller pages is not immediate and requires migration/compaction. 2) There is no guarantee that the total size of smaller pages allocated will match the size of the larger page which was freed. This is because the area freed by the larger page could quickly be fragmented. In a test environment with a load that continually fills the page cache with clean pages, results such as the following can be observed: Unexpected number of 2MB pages allocated: Expected 25600, have 19944 real 0m42.092s user 0m0.008s sys 0m41.467s To address these issues, introduce the concept of hugetlb page demotion. Demotion provides a means of 'in place' splitting of a hugetlb page to pages of a smaller size. This avoids freeing pages to buddy and then trying to allocate from buddy. Page demotion is controlled via sysfs files that reside in the per-hugetlb page size and per node directories. - demote_size Target page size for demotion, a smaller huge page size. File can be written to chose a smaller huge page size if multiple are available. - demote Writable number of hugetlb pages to be demoted To demote 50 GB huge pages, one would: cat .../hugepages-1048576kB/free_hugepages /* optional, verify free pages */ cat .../hugepages-1048576kB/demote_size /* optional, verify target size */ echo 50 > .../hugepages-1048576kB/demote Only hugetlb pages which are free at the time of the request can be demoted. Demotion does not add to the complexity of surplus pages and honors reserved huge pages. Therefore, when a value is written to the sysfs demote file, that value is only the maximum number of pages which will be demoted. It is possible fewer will actually be demoted. The recently introduced per-hstate mutex is used to synchronize demote operations with other operations that modify hugetlb pools. Real world use cases -------------------- The above scenario describes a real world use case where hugetlb pages are used to back VMs on x86. Both issues of long allocation times and not necessarily getting the expected number of smaller huge pages after a free and allocate cycle have been experienced. The occurrence of these issues is dependent on other activity within the host and can not be predicted. This patch (of 5): Two new sysfs files are added to demote hugtlb pages. These files are both per-hugetlb page size and per node. Files are: demote_size - The size in Kb that pages are demoted to. (read-write) demote - The number of huge pages to demote. (write-only) By default, demote_size is the next smallest huge page size. Valid huge page sizes less than huge page size may be written to this file. When huge pages are demoted, they are demoted to this size. Writing a value to demote will result in an attempt to demote that number of hugetlb pages to an appropriate number of demote_size pages. NOTE: Demote interfaces are only provided for huge page sizes if there is a smaller target demote huge page size. For example, on x86 1GB huge pages will have demote interfaces. 2MB huge pages will not have demote interfaces. This patch does not provide full demote functionality. It only provides the sysfs interfaces. It also provides documentation for the new interfaces. [mike.kravetz@oracle.com: n_mask initialization does not need to be protected by the mutex] Link: https://lkml.kernel.org/r/0530e4ef-2492-5186-f919-5db68edea654@oracle.com Link: https://lkml.kernel.org/r/20211007181918.136982-2-mike.kravetz@oracle.comSigned-off-by: -
Peter Xu authored
Remove __unmap_hugepage_range() from the header file, because it is only used in hugetlb.c. Link: https://lkml.kernel.org/r/20210917165108.9341-1-peterx@redhat.comSigned-off-by:
John Hubbard <jhubbard@nvidia.com> Reviewed-by:
-
Yang Shi authored
Currently hwpoison doesn't handle non-anonymous THP, but since v4.8 THP support for tmpfs and read-only file cache has been added. They could be offlined by split THP, just like anonymous THP. Link: https://lkml.kernel.org/r/20211020210755.23964-7-shy828301@gmail.comSigned-off-by:
Naoya Horiguchi <naoya.horiguchi@nec.com> Cc: Hugh Dickins <hughd@google.com> Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Cc: Matthew Wilcox <willy@infradead.org> Cc: Oscar Salvador <osalvador@suse.de> Cc: Peter Xu <peterx@redhat.com> Signed-off-by:
-
Yang Shi authored
The current behavior of memory failure is to truncate the page cache regardless of dirty or clean. If the page is dirty the later access will get the obsolete data from disk without any notification to the users. This may cause silent data loss. It is even worse for shmem since shmem is in-memory filesystem, truncating page cache means discarding data blocks. The later read would return all zero. The right approach is to keep the corrupted page in page cache, any later access would return error for syscalls or SIGBUS for page fault, until the file is truncated, hole punched or removed. The regular storage backed filesystems would be more complicated so this patch is focused on shmem. This also unblock the support for soft offlining shmem THP. [arnd@arndb.de: fix uninitialized variable use in me_pagecache_clean()] Link: https://lkml.kernel.org/r/20211022064748.4173718-1-arnd@kernel.org Link: https://lkml.kernel.org/r/20211020210755.23964-6-shy828301@gmail.comSigned-off-by:
Arnd Bergmann <arnd@arndb.de> Cc: Hugh Dickins <hughd@google.com> Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Cc: Matthew Wilcox <willy@infradead.org> Cc: Naoya Horiguchi <naoya.horiguchi@nec.com> Cc: Oscar Salvador <osalvador@suse.de> Cc: Peter Xu <peterx@redhat.com> Signed-off-by:
-
Yang Shi authored
Memory failure will report failure if the page still has extra pinned refcount other than from hwpoison after the handler is done. Actually the check is not necessary for all handlers, so move the check into specific handlers. This would make the following keeping shmem page in page cache patch easier. There may be expected extra pin for some cases, for example, when the page is dirty and in swapcache. Link: https://lkml.kernel.org/r/20211020210755.23964-5-shy828301@gmail.comSigned-off-by:
-
Yang Shi authored
Patch series "Solve silent data loss caused by poisoned page cache (shmem/tmpfs)", v5. When discussing the patch that splits page cache THP in order to offline the poisoned page, Noaya mentioned there is a bigger problem [1] that prevents this from working since the page cache page will be truncated if uncorrectable errors happen. By looking this deeper it turns out this approach (truncating poisoned page) may incur silent data loss for all non-readonly filesystems if the page is dirty. It may be worse for in-memory filesystem, e.g. shmem/tmpfs since the data blocks are actually gone. To solve this problem we could keep the poisoned dirty page in page cache then notify the users on any later access, e.g. page fault, read/write, etc. The clean page could be truncated as is since they can be reread from disk later on. The consequence is the filesystems may find poisoned page and manipulate it as healthy page since all the filesystems actually don't check if the page is poisoned or not in all the relevant paths except page fault. In general, we need make the filesystems be aware of poisoned page before we could keep the poisoned page in page cache in order to solve the data loss problem. To make filesystems be aware of poisoned page we should consider: - The page should be not written back: clearing dirty flag could prevent from writeback. - The page should not be dropped (it shows as a clean page) by drop caches or other callers: the refcount pin from hwpoison could prevent from invalidating (called by cache drop, inode cache shrinking, etc), but it doesn't avoid invalidation in DIO path. - The page should be able to get truncated/hole punched/unlinked: it works as it is. - Notify users when the page is accessed, e.g. read/write, page fault and other paths (compression, encryption, etc). The scope of the last one is huge since almost all filesystems need do it once a page is returned from page cache lookup. There are a couple of options to do it: 1. Check hwpoison flag for every path, the most straightforward way. 2. Return NULL for poisoned page from page cache lookup, the most callsites check if NULL is returned, this should have least work I think. But the error handling in filesystems just return -ENOMEM, the error code will incur confusion to the users obviously. 3. To improve #2, we could return error pointer, e.g. ERR_PTR(-EIO), but this will involve significant amount of code change as well since all the paths need check if the pointer is ERR or not just like option #1. I did prototypes for both #1 and #3, but it seems #3 may require more changes than #1. For #3 ERR_PTR will be returned so all the callers need to check the return value otherwise invalid pointer may be dereferenced, but not all callers really care about the content of the page, for example, partial truncate which just sets the truncated range in one page to 0. So for such paths it needs additional modification if ERR_PTR is returned. And if the callers have their own way to handle the problematic pages we need to add a new FGP flag to tell FGP functions to return the pointer to the page. It may happen very rarely, but once it happens the consequence (data corruption) could be very bad and it is very hard to debug. It seems this problem had been slightly discussed before, but seems no action was taken at that time. [2] As the aforementioned investigation, it needs huge amount of work to solve the potential data loss for all filesystems. But it is much easier for in-memory filesystems and such filesystems actually suffer more than others since even the data blocks are gone due to truncating. So this patchset starts from shmem/tmpfs by taking option #1. TODO: * The unpoison has been broken since commit 0ed950d1 ("mm,hwpoison: make get_hwpoison_page() call get_any_page()"), and this patch series make refcount check for unpoisoning shmem page fail. * Expand to other filesystems. But I haven't heard feedback from filesystem developers yet. Patch breakdown: Patch #1: cleanup, depended by patch #2 Patch #2: fix THP with hwpoisoned subpage(s) PMD map bug Patch #3: coding style cleanup Patch #4: refactor and preparation. Patch #5: keep the poisoned page in page cache and handle such case for all the paths. Patch #6: the previous patches unblock page cache THP split, so this patch add page cache THP split support. This patch (of 4): A minor cleanup to the indent. Link: https://lkml.kernel.org/r/20211020210755.23964-1-shy828301@gmail.com Link: https://lkml.kernel.org/r/20211020210755.23964-4-shy828301@gmail.comSigned-off-by: -
Rikard Falkeborn authored
The only usage of hwp_walk_ops is to pass its address to walk_page_range() which takes a pointer to const mm_walk_ops as argument. Make it const to allow the compiler to put it in read-only memory. Link: https://lkml.kernel.org/r/20211014075042.17174-3-rikard.falkeborn@gmail.comSigned-off-by:
Rikard Falkeborn <rikard.falkeborn@gmail.com> Acked-by:
Anshuman Khandual <anshuman.khandual@arm.com> Signed-off-by:
-
Marco Elver authored
PagePoisoned() accesses page->flags which can be updated concurrently: | BUG: KCSAN: data-race in next_uptodate_page / unlock_page | | write (marked) to 0xffffea00050f37c0 of 8 bytes by task 1872 on cpu 1: | instrument_atomic_write include/linux/instrumented.h:87 [inline] | clear_bit_unlock_is_negative_byte include/asm-generic/bitops/instrumented-lock.h:74 [inline] | unlock_page+0x102/0x1b0 mm/filemap.c:1465 | filemap_map_pages+0x6c6/0x890 mm/filemap.c:3057 | ... | read to 0xffffea00050f37c0 of 8 bytes by task 1873 on cpu 0: | PagePoisoned include/linux/page-flags.h:204 [inline] | PageReadahead include/linux/page-flags.h:382 [inline] | next_uptodate_page+0x456/0x830 mm/filemap.c:2975 | ... | CPU: 0 PID: 1873 Comm: systemd-udevd Not tainted 5.11.0-rc4-00001-gf9ce0be7 #1 To avoid the compiler tearing or otherwise optimizing the access, use READ_ONCE() to access flags. Link: https://lore.kernel.org/all/20210826144157.GA26950@xsang-OptiPlex-9020/ Link: https://lkml.kernel.org/r/20210913113542.2658064-1-elver@google.comReported-by:
kernel test robot <oliver.sang@intel.com> Signed-off-by:
Marco Elver <elver@google.com> Acked-by:
Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by:
Will Deacon <will@kernel.org> Cc: Marco Elver <elver@google.com> Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com> Signed-off-by:
-
Wang ShaoBo authored
This patch uses clamp() to simplify code in init_per_zone_wmark_min(). Link: https://lkml.kernel.org/r/20211021034830.1049150-1-bobo.shaobowang@huawei.comSigned-off-by:
Wang ShaoBo <bobo.shaobowang@huawei.com> Reviewed-by:
-
Sebastian Andrzej Siewior authored
drain_local_pages_wq() disables preemption to avoid CPU migration during CPU hotplug and can't use cpus_read_lock(). Using migrate_disable() works here, too. The scheduler won't take the CPU offline until the task left the migrate-disable section. The problem with disabled preemption here is that drain_local_pages() acquires locks which are turned into sleeping locks on PREEMPT_RT and can't be acquired with disabled preemption. Use migrate_disable() in drain_local_pages_wq(). Link: https://lkml.kernel.org/r/20211015210933.viw6rjvo64qtqxn4@linutronix.deSigned-off-by:
Sebastian Andrzej Siewior <bigeasy@linutronix.de> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Peter Zijlstra <peterz@infradead.org> Signed-off-by:
-
Christophe Leroy authored
The generic version of arch_is_kernel_initmem_freed() now does the same as s390 version. Remove the s390 version. Link: https://lkml.kernel.org/r/b6feb5dfe611a322de482762fc2df3a9eece70c7.1633001016.git.christophe.leroy@csgroup.euSigned-off-by:
Christophe Leroy <christophe.leroy@csgroup.eu> Acked-by:
Heiko Carstens <hca@linux.ibm.com> Cc: Gerald Schaefer <gerald.schaefer@linux.ibm.com> Cc: Kefeng Wang <wangkefeng.wang@huawei.com> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Paul Mackerras <paulus@ozlabs.org> Signed-off-by:
-
Christophe Leroy authored
The generic version of arch_is_kernel_initmem_freed() now does the same as powerpc version. Remove the powerpc version. Link: https://lkml.kernel.org/r/c53764eb45d41491e2b21da2e7812239897dbebb.1633001016.git.christophe.leroy@csgroup.euSigned-off-by:
-
Christophe Leroy authored
Commit 7a5da02d ("locking/lockdep: check for freed initmem in static_obj()") added arch_is_kernel_initmem_freed() which is supposed to report whether an object is part of already freed init memory. For the time being, the generic version of arch_is_kernel_initmem_freed() always reports 'false', allthough free_initmem() is generically called on all architectures. Therefore, change the generic version of arch_is_kernel_initmem_freed() to check whether free_initmem() has been called. If so, then check if a given address falls into init memory. To ease the use of system_state, move it out of line into its only caller which is lockdep.c Link: https://lkml.kernel.org/r/1d40783e676e07858be97d881f449ee7ea8adfb1.1633001016.git.christophe.leroy@csgroup.euSigned-off-by:
-
Christophe Leroy authored
core_kernel_text() considers that until system_state in at least SYSTEM_RUNNING, init memory is valid. But init memory is freed a few lines before setting SYSTEM_RUNNING, so we have a small period of time when core_kernel_text() is wrong. Create an intermediate system state called SYSTEM_FREEING_INIT that is set before starting freeing init memory, and use it in core_kernel_text() to report init memory invalid earlier. Link: https://lkml.kernel.org/r/9ecfdee7dd4d741d172cb93ff1d87f1c58127c9a.1633001016.git.christophe.leroy@csgroup.euSigned-off-by:
-
Liangcai Fan authored
min/low/high_wmark_pages(z) is defined as (z->_watermark[WMARK_MIN/LOW/HIGH] + z->watermark_boost) If kswapd is frequently woken up due to the increase of min/low/high_wmark_pages, printing watermark_boost can quickly locate whether watermark_boost or _watermark[WMARK_MIN/LOW/HIGH] caused min/low/high_wmark_pages to increase. Link: https://lkml.kernel.org/r/1632472566-12246-1-git-send-email-liangcaifan19@gmail.comSigned-off-by:
-
Feng Tang authored
There was a report that starting an Ubuntu in docker while using cpuset to bind it to movable nodes (a node only has movable zone, like a node for hotplug or a Persistent Memory node in normal usage) will fail due to memory allocation failure, and then OOM is involved and many other innocent processes got killed. It can be reproduced with command: $ docker run -it --rm --cpuset-mems 4 ubuntu:latest bash -c "grep Mems_allowed /proc/self/status" (where node 4 is a movable node) runc:[2:INIT] invoked oom-killer: gfp_mask=0x500cc2(GFP_HIGHUSER|__GFP_ACCOUNT), order=0, oom_score_adj=0 CPU: 8 PID: 8291 Comm: runc:[2:INIT] Tainted: G W I E 5.8.2-0.g71b519a-default #1 openSUSE Tumbleweed (unreleased) Hardware name: Dell Inc. PowerEdge R640/0PHYDR, BIOS 2.6.4 04/09/2020 Call Trace: dump_stack+0x6b/0x88 dump_header+0x4a/0x1e2 oom_kill_process.cold+0xb/0x10 out_of_memory.part.0+0xaf/0x230 out_of_memory+0x3d/0x80 __alloc_pages_slowpath.constprop.0+0x954/0xa20 __alloc_pages_nodemask+0x2d3/0x300 pipe_write+0x322/0x590 new_sync_write+0x196/0x1b0 vfs_write+0x1c3/0x1f0 ksys_write+0xa7/0xe0 do_syscall_64+0x52/0xd0 entry_SYSCALL_64_after_hwframe+0x44/0xa9 Mem-Info: active_anon:392832 inactive_anon:182 isolated_anon:0 active_file:68130 inactive_file:151527 isolated_file:0 unevictable:2701 dirty:0 writeback:7 slab_reclaimable:51418 slab_unreclaimable:116300 mapped:45825 shmem:735 pagetables:2540 bounce:0 free:159849484 free_pcp:73 free_cma:0 Node 4 active_anon:1448kB inactive_anon:0kB active_file:0kB inactive_file:0kB unevictable:0kB isolated(anon):0kB isolated(file):0kB mapped:0kB dirty:0kB writeback:0kB shmem:0kB shmem_thp: 0kB shmem_pmdmapped: 0kB anon_thp: 0kB writeback_tmp:0kB all_unreclaimable? no Node 4 Movable free:130021408kB min:9140kB low:139160kB high:269180kB reserved_highatomic:0KB active_anon:1448kB inactive_anon:0kB active_file:0kB inactive_file:0kB unevictable:0kB writepending:0kB present:130023424kB managed:130023424kB mlocked:0kB kernel_stack:0kB pagetables:0kB bounce:0kB free_pcp:292kB local_pcp:84kB free_cma:0kB lowmem_reserve[]: 0 0 0 0 0 Node 4 Movable: 1*4kB (M) 0*8kB 0*16kB 1*32kB (M) 0*64kB 0*128kB 1*256kB (M) 1*512kB (M) 1*1024kB (M) 0*2048kB 31743*4096kB (M) = 130021156kB oom-kill:constraint=CONSTRAINT_CPUSET,nodemask=(null),cpuset=docker-9976a269caec812c134fa317f27487ee36e1129beba7278a463dd53e5fb9997b.scope,mems_allowed=4,global_oom,task_memcg=/system.slice/containerd.service,task=containerd,pid=4100,uid=0 Out of memory: Killed process 4100 (containerd) total-vm:4077036kB, anon-rss:51184kB, file-rss:26016kB, shmem-rss:0kB, UID:0 pgtables:676kB oom_score_adj:0 oom_reaper: reaped process 8248 (docker), now anon-rss:0kB, file-rss:0kB, shmem-rss:0kB oom_reaper: reaped process 2054 (node_exporter), now anon-rss:0kB, file-rss:0kB, shmem-rss:0kB oom_reaper: reaped process 1452 (systemd-journal), now anon-rss:0kB, file-rss:8564kB, shmem-rss:4kB oom_reaper: reaped process 2146 (munin-node), now anon-rss:0kB, file-rss:0kB, shmem-rss:0kB oom_reaper: reaped process 8291 (runc:[2:INIT]), now anon-rss:0kB, file-rss:0kB, shmem-rss:0kB The reason is that in this case, the target cpuset nodes only have movable zone, while the creation of an OS in docker sometimes needs to allocate memory in non-movable zones (dma/dma32/normal) like GFP_HIGHUSER, and the cpuset limit forbids the allocation, then out-of-memory killing is involved even when normal nodes and movable nodes both have many free memory. The OOM killer cannot help to resolve the situation as there is no usable memory for the request in the cpuset scope. The only reasonable measure to take is to fail the allocation right away and have the caller to deal with it. So add a check for cases like this in the slowpath of allocation, and bail out early returning NULL for the allocation. As page allocation is one of the hottest path in kernel, this check will hurt all users with sane cpuset configuration, add a static branch check and detect the abnormal config in cpuset memory binding setup so that the extra check cost in page allocation is not paid by everyone. [thanks to Micho Hocko and David Rientjes for suggesting not handling it inside OOM code, adding cpuset check, refining comments] Link: https://lkml.kernel.org/r/1632481657-68112-1-git-send-email-feng.tang@intel.comSigned-off-by:Feng Tang <feng.tang@intel.com> Suggested-by:
Michal Hocko <mhocko@suse.com> Acked-by:
-
