kernel/git/stable/linux.git/mm/migrate.c, branch linux-4.1.y

mm: Export migrate_page_move_mapping and migrate_page_copy

2016-07-11T00:20:01+00:00

[ Upstream commit 1118dce773d84f39ebd51a9fe7261f9169cb056e ] Export these symbols such that UBIFS can implement ->migratepage. Cc: stable@vger.kernel.org Signed-off-by: Richard Weinberger Acked-by: Christoph Hellwig Signed-off-by: Sasha Levin

mm: check __PG_HWPOISON separately from PAGE_FLAGS_CHECK_AT_*

2016-05-17T17:42:51+00:00

[ Upstream commit f4c18e6f7b5bbb5b528b3334115806b0d76f50f9 ] The race condition addressed in commit add05cecef80 ("mm: soft-offline: don't free target page in successful page migration") was not closed completely, because that can happen not only for soft-offline, but also for hard-offline. Consider that a slab page is about to be freed into buddy pool, and then an uncorrected memory error hits the page just after entering __free_one_page(), then VM_BUG_ON_PAGE(page->flags & PAGE_FLAGS_CHECK_AT_PREP) is triggered, despite the fact that it's not necessary because the data on the affected page is not consumed. To solve it, this patch drops __PG_HWPOISON from page flag checks at allocation/free time. I think it's justified because __PG_HWPOISON flags is defined to prevent the page from being reused, and setting it outside the page's alloc-free cycle is a designed behavior (not a bug.) For recent months, I was annoyed about BUG_ON when soft-offlined page remains on lru cache list for a while, which is avoided by calling put_page() instead of putback_lru_page() in page migration's success path. This means that this patch reverts a major change from commit add05cecef80 about the new refcounting rule of soft-offlined pages, so "reuse window" revives. This will be closed by a subsequent patch. Signed-off-by: Naoya Horiguchi Cc: Andi Kleen Cc: Dean Nelson Cc: Tony Luck Cc: "Kirill A. Shutemov" Cc: Hugh Dickins Cc: David Rientjes Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds Signed-off-by: Sasha Levin

mm: soft-offline: don't free target page in successful page migration

2016-05-17T17:42:50+00:00

[ Upstream commit add05cecef803f3372c5fc1d2a964171872daf9f ] Stress testing showed that soft offline events for a process iterating "mmap-pagefault-munmap" loop can trigger VM_BUG_ON(PAGE_FLAGS_CHECK_AT_PREP) in __free_one_page(): Soft offlining page 0x70fe1 at 0x70100008d000 Soft offlining page 0x705fb at 0x70300008d000 page:ffffea0001c3f840 count:0 mapcount:0 mapping: (null) index:0x2 flags: 0x1fffff80800000(hwpoison) page dumped because: VM_BUG_ON_PAGE(page->flags & ((1 << 25) - 1)) ------------[ cut here ]------------ kernel BUG at /src/linux-dev/mm/page_alloc.c:585! invalid opcode: 0000 [#1] SMP DEBUG_PAGEALLOC Modules linked in: cfg80211 rfkill crc32c_intel microcode ppdev parport_pc pcspkr serio_raw virtio_balloon parport i2c_piix4 virtio_blk virtio_net ata_generic pata_acpi floppy CPU: 3 PID: 1779 Comm: test_base_madv_ Not tainted 4.0.0-v4.0-150511-1451-00009-g82360a3730e6 #139 RIP: free_pcppages_bulk+0x52a/0x6f0 Call Trace: drain_pages_zone+0x3d/0x50 drain_local_pages+0x1d/0x30 on_each_cpu_mask+0x46/0x80 drain_all_pages+0x14b/0x1e0 soft_offline_page+0x432/0x6e0 SyS_madvise+0x73c/0x780 system_call_fastpath+0x12/0x17 Code: ff 89 45 b4 48 8b 45 c0 48 83 b8 a8 00 00 00 00 0f 85 e3 fb ff ff 0f 1f 00 0f 0b 48 8b 7d 90 48 c7 c6 e8 95 a6 81 e8 e6 32 02 00 <0f> 0b 8b 45 cc 49 89 47 30 41 8b 47 18 83 f8 ff 0f 85 10 ff ff RIP [] free_pcppages_bulk+0x52a/0x6f0 RSP ---[ end trace 53926436e76d1f35 ]--- When soft offline successfully migrates page, the source page is supposed to be freed. But there is a race condition where a source page looks isolated (i.e. the refcount is 0 and the PageHWPoison is set) but somewhat linked to pcplist. Then another soft offline event calls drain_all_pages() and tries to free such hwpoisoned page, which is forbidden. This odd page state seems to happen due to the race between put_page() in putback_lru_page() and __pagevec_lru_add_fn(). But I don't want to play with tweaking drain code as done in commit 9ab3b598d2df "mm: hwpoison: drop lru_add_drain_all() in __soft_offline_page()", or to change page freeing code for this soft offline's purpose. Instead, let's think about the difference between hard offline and soft offline. There is an interesting difference in how to isolate the in-use page between these, that is, hard offline marks PageHWPoison of the target page at first, and doesn't free it by keeping its refcount 1. OTOH, soft offline tries to free the target page then marks PageHWPoison. This difference might be the source of complexity and result in bugs like the above. So making soft offline isolate with keeping refcount can be a solution for this problem. We can pass to page migration code the "reason" which shows the caller, so let's use this more to avoid calling putback_lru_page() when called from soft offline, which effectively does the isolation for soft offline. With this change, target pages of soft offline never be reused without changing migratetype, so this patch also removes the related code. Signed-off-by: Naoya Horiguchi Cc: Andi Kleen Cc: Tony Luck Cc: "Kirill A. Shutemov" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds Signed-off-by: Sasha Levin

mm: numa: quickly fail allocations for NUMA balancing on full nodes

2016-03-09T18:15:12+00:00

[ Upstream commit 8479eba7781fa9ffb28268840de6facfc12c35a7 ] Commit 4167e9b2cf10 ("mm: remove GFP_THISNODE") removed the GFP_THISNODE flag combination due to confusing semantics. It noted that alloc_misplaced_dst_page() was one such user after changes made by commit e97ca8e5b864 ("mm: fix GFP_THISNODE callers and clarify"). Unfortunately when GFP_THISNODE was removed, users of alloc_misplaced_dst_page() started waking kswapd and entering direct reclaim because the wrong GFP flags are cleared. The consequence is that workloads that used to fit into memory now get reclaimed which is addressed by this patch. The problem can be demonstrated with "mutilate" that exercises memcached which is software dedicated to memory object caching. The configuration uses 80% of memory and is run 3 times for varying numbers of clients. The results on a 4-socket NUMA box are mutilate 4.4.0 4.4.0 vanilla numaswap-v1 Hmean 1 8394.71 ( 0.00%) 8395.32 ( 0.01%) Hmean 4 30024.62 ( 0.00%) 34513.54 ( 14.95%) Hmean 7 32821.08 ( 0.00%) 70542.96 (114.93%) Hmean 12 55229.67 ( 0.00%) 93866.34 ( 69.96%) Hmean 21 39438.96 ( 0.00%) 85749.21 (117.42%) Hmean 30 37796.10 ( 0.00%) 50231.49 ( 32.90%) Hmean 47 18070.91 ( 0.00%) 38530.13 (113.22%) The metric is queries/second with the more the better. The results are way outside of the noise and the reason for the improvement is obvious from some of the vmstats 4.4.0 4.4.0 vanillanumaswap-v1r1 Minor Faults 1929399272 2146148218 Major Faults 19746529 3567 Swap Ins 57307366 9913 Swap Outs 50623229 17094 Allocation stalls 35909 443 DMA allocs 0 0 DMA32 allocs 72976349 170567396 Normal allocs 5306640898 5310651252 Movable allocs 0 0 Direct pages scanned 404130893 799577 Kswapd pages scanned 160230174 0 Kswapd pages reclaimed 55928786 0 Direct pages reclaimed 1843936 41921 Page writes file 2391 0 Page writes anon 50623229 17094 The vanilla kernel is swapping like crazy with large amounts of direct reclaim and kswapd activity. The figures are aggregate but it's known that the bad activity is throughout the entire test. Note that simple streaming anon/file memory consumers also see this problem but it's not as obvious. In those cases, kswapd is awake when it should not be. As there are at least two reclaim-related bugs out there, it's worth spelling out the user-visible impact. This patch only addresses bugs related to excessive reclaim on NUMA hardware when the working set is larger than a NUMA node. There is a bug related to high kswapd CPU usage but the reports are against laptops and other UMA hardware and is not addressed by this patch. Signed-off-by: Mel Gorman Cc: Vlastimil Babka Cc: Johannes Weiner Cc: David Rientjes Cc: [4.1+] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds Signed-off-by: Sasha Levin

mm: migrate: hugetlb: putback destination hugepage to active list

2015-10-22T21:43:15+00:00

commit 3aaa76e125c1dd58c9b599baa8c6021896874c12 upstream. Since commit bcc54222309c ("mm: hugetlb: introduce page_huge_active") each hugetlb page maintains its active flag to avoid a race condition betwe= en multiple calls of isolate_huge_page(), but current kernel doesn't set the f= lag on a hugepage allocated by migration because the proper putback routine isn= 't called. This means that users could still encounter the race referred to by bcc54222309c in this special case, so this patch fixes it. Fixes: bcc54222309c ("mm: hugetlb: introduce page_huge_active") Signed-off-by: Naoya Horiguchi Cc: Michal Hocko Cc: Andi Kleen Cc: Hugh Dickins Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds Signed-off-by: Greg Kroah-Hartman

mm/migrate: check-before-clear PageSwapCache

2015-04-15T23:35:17+00:00

With the page flag sanitization patchset, an invalid usage of ClearPageSwapCache() is detected in migration_page_copy(). migrate_page_copy() is shared by both normal and hugepage (both thp and hugetlb) code path, so let's check PageSwapCache() and clear it if it's set to avoid misuse of the invalid clear operation. Signed-off-by: Naoya Horiguchi Acked-by: Kirill A. Shutemov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

mm: numa: remove migrate_ratelimited

2015-04-14T23:49:05+00:00

This code is dead since commit 9e645ab6d089 ("sched/numa: Continue PTE scanning even if migrate rate limited") so remove it. Signed-off-by: Mel Gorman Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

mm/migrate: mark unmap_and_move() "noinline" to avoid ICE in gcc 4.7.3

2015-04-14T23:48:59+00:00

With gcc version 4.7.3 (Ubuntu/Linaro 4.7.3-12ubuntu1) : mm/migrate.c: In function `migrate_pages': mm/migrate.c:1148:1: internal compiler error: in push_minipool_fix, at config/arm/arm.c:13500 Please submit a full bug report, with preprocessed source if appropriate. See for instructions. Preprocessed source stored into /tmp/ccPoM1tr.out file, please attach this to your bugreport. make[1]: *** [mm/migrate.o] Error 1 make: *** [mm/migrate.o] Error 2 Mark unmap_and_move() (which is used in a single place only) "noinline" to work around this compiler bug. [akpm@linux-foundation.org: make it conditional on gcc-4.7.3 and arm] [khilman@kernel.org: fine-tune compiler versions] [akpm@linux-foundation.org: fix comment] Signed-off-by: Geert Uytterhoeven Reported-by: Kevin Hilman Cc: Marc Zyngier Tested-by: Kevin Hilman Tested-by: Lina Iyer Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

mm: convert p[te|md]_mknonnuma and remaining page table manipulations

2015-02-13T02:54:08+00:00

With PROT_NONE, the traditional page table manipulation functions are sufficient. [andre.przywara@arm.com: fix compiler warning in pmdp_invalidate()] [akpm@linux-foundation.org: fix build with STRICT_MM_TYPECHECKS] Signed-off-by: Mel Gorman Acked-by: Linus Torvalds Acked-by: Aneesh Kumar Tested-by: Sasha Levin Cc: Benjamin Herrenschmidt Cc: Dave Jones Cc: Hugh Dickins Cc: Ingo Molnar Cc: Kirill Shutemov Cc: Paul Mackerras Cc: Rik van Riel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

mm: numa: do not dereference pmd outside of the lock during NUMA hinting fault

2015-02-13T02:54:08+00:00

Automatic NUMA balancing depends on being able to protect PTEs to trap a fault and gather reference locality information. Very broadly speaking it would mark PTEs as not present and use another bit to distinguish between NUMA hinting faults and other types of faults. It was universally loved by everybody and caused no problems whatsoever. That last sentence might be a lie. This series is very heavily based on patches from Linus and Aneesh to replace the existing PTE/PMD NUMA helper functions with normal change protections. I did alter and add parts of it but I consider them relatively minor contributions. At their suggestion, acked-bys are in there but I've no problem converting them to Signed-off-by if requested. AFAIK, this has received no testing on ppc64 and I'm depending on Aneesh for that. I tested trinity under kvm-tool and passed and ran a few other basic tests. At the time of writing, only the short-lived tests have completed but testing of V2 indicated that long-term testing had no surprises. In most cases I'm leaving out detail as it's not that interesting. specjbb single JVM: There was negligible performance difference in the benchmark itself for short runs. However, system activity is higher and interrupts are much higher over time -- possibly TLB flushes. Migrations are also higher. Overall, this is more overhead but considering the problems faced with the old approach I think we just have to suck it up and find another way of reducing the overhead. specjbb multi JVM: Negligible performance difference to the actual benchmark but like the single JVM case, the system overhead is noticeably higher. Again, interrupts are a major factor. autonumabench: This was all over the place and about all that can be reasonably concluded is that it's different but not necessarily better or worse. autonumabench 3.18.0-rc5 3.18.0-rc5 mmotm-20141119 protnone-v3r3 User NUMA01 32380.24 ( 0.00%) 21642.92 ( 33.16%) User NUMA01_THEADLOCAL 22481.02 ( 0.00%) 22283.22 ( 0.88%) User NUMA02 3137.00 ( 0.00%) 3116.54 ( 0.65%) User NUMA02_SMT 1614.03 ( 0.00%) 1543.53 ( 4.37%) System NUMA01 322.97 ( 0.00%) 1465.89 (-353.88%) System NUMA01_THEADLOCAL 91.87 ( 0.00%) 49.32 ( 46.32%) System NUMA02 37.83 ( 0.00%) 14.61 ( 61.38%) System NUMA02_SMT 7.36 ( 0.00%) 7.45 ( -1.22%) Elapsed NUMA01 716.63 ( 0.00%) 599.29 ( 16.37%) Elapsed NUMA01_THEADLOCAL 553.98 ( 0.00%) 539.94 ( 2.53%) Elapsed NUMA02 83.85 ( 0.00%) 83.04 ( 0.97%) Elapsed NUMA02_SMT 86.57 ( 0.00%) 79.15 ( 8.57%) CPU NUMA01 4563.00 ( 0.00%) 3855.00 ( 15.52%) CPU NUMA01_THEADLOCAL 4074.00 ( 0.00%) 4136.00 ( -1.52%) CPU NUMA02 3785.00 ( 0.00%) 3770.00 ( 0.40%) CPU NUMA02_SMT 1872.00 ( 0.00%) 1959.00 ( -4.65%) System CPU usage of NUMA01 is worse but it's an adverse workload on this machine so I'm reluctant to conclude that it's a problem that matters. On the other workloads that are sensible on this machine, system CPU usage is great. Overall time to complete the benchmark is comparable 3.18.0-rc5 3.18.0-rc5 mmotm-20141119protnone-v3r3 User 59612.50 48586.44 System 460.22 1537.45 Elapsed 1442.20 1304.29 NUMA alloc hit 5075182 5743353 NUMA alloc miss 0 0 NUMA interleave hit 0 0 NUMA alloc local 5075174 5743339 NUMA base PTE updates 637061448 443106883 NUMA huge PMD updates 1243434 864747 NUMA page range updates 1273699656 885857347 NUMA hint faults 1658116 1214277 NUMA hint local faults 959487 754113 NUMA hint local percent 57 62 NUMA pages migrated 5467056 61676398 The NUMA pages migrated look terrible but when I looked at a graph of the activity over time I see that the massive spike in migration activity was during NUMA01. This correlates with high system CPU usage and could be simply down to bad luck but any modifications that affect that workload would be related to scan rates and migrations, not the protection mechanism. For all other workloads, migration activity was comparable. Overall, headline performance figures are comparable but the overhead is higher, mostly in interrupts. To some extent, higher overhead from this approach was anticipated but not to this degree. It's going to be necessary to reduce this again with a separate series in the future. It's still worth going ahead with this series though as it's likely to avoid constant headaches with Xen and is probably easier to maintain. This patch (of 10): A transhuge NUMA hinting fault may find the page is migrating and should wait until migration completes. The check is race-prone because the pmd is deferenced outside of the page lock and while the race is tiny, it'll be larger if the PMD is cleared while marking PMDs for hinting fault. This patch closes the race. Signed-off-by: Mel Gorman Cc: Aneesh Kumar K.V Cc: Benjamin Herrenschmidt Cc: Dave Jones Cc: Hugh Dickins Cc: Ingo Molnar Cc: Kirill Shutemov Cc: Linus Torvalds Cc: Paul Mackerras Cc: Rik van Riel Cc: Sasha Levin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds