- 04 Jun, 2020 40 commits
-
-
Johannes Weiner authored
With the advent of fast random IO devices (SSDs, PMEM) and in-memory swap devices such as zswap, it's possible for swap to be much faster than filesystems, and for swapping to be preferable over thrashing filesystem caches. Allow setting swappiness - which defines the rough relative IO cost of cache misses between page cache and swap-backed pages - to reflect such situations by making the swap-preferred range configurable. Signed-off-by:
Johannes Weiner <hannes@cmpxchg.org> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Michal Hocko <mhocko@suse.com> Cc: Minchan Kim <minchan@kernel.org> Cc: Rik van Riel <riel@surriel.com> Link: http://lkml.kernel.org/r/20200520232525.798933-4-hannes@cmpxchg.orgSigned-off-by:
Linus Torvalds <torvalds@linux-foundation.org>
-
Johannes Weiner authored
Having statistics on pages scanned and pages reclaimed for both anon and file pages makes it easier to evaluate changes to LRU balancing. While at it, clean up the stat-keeping mess for isolation, putback, reclaim stats etc. a bit: first the physical LRU operation (isolation and putback), followed by vmstats, reclaim_stats, and then vm events. Signed-off-by:
-
Johannes Weiner authored
The reclaim code that balances between swapping and cache reclaim tries to predict likely reuse based on in-memory reference patterns alone. This works in many cases, but when it fails it cannot detect when the cache is thrashing pathologically, or when we're in the middle of a swap storm. The high seek cost of rotational drives under which the algorithm evolved also meant that mistakes could quickly result in lockups from too aggressive swapping (which is predominantly random IO). As a result, the balancing code has been tuned over time to a point where it mostly goes for page cache and defers swapping until the VM is under significant memory pressure. The resulting strategy doesn't make optimal caching decisions - where optimal is the least amount of IO required to execute the workload. The proliferation of fast random IO devices such as SSDs, in-memory compression such as zswap, and persistent memory technologies on the horizon, has made this undesirable behavior very noticable: Even in the presence of large amounts of cold anonymous memory and a capable swap device, the VM refuses to even seriously scan these pages, and can leave the page cache thrashing needlessly. This series sets out to address this. Since commit ("a528910e mm: thrash detection-based file cache sizing") we have exact tracking of refault IO - the ultimate cost of reclaiming the wrong pages. This allows us to use an IO cost based balancing model that is more aggressive about scanning anonymous memory when the cache is thrashing, while being able to avoid unnecessary swap storms. These patches base the LRU balance on the rate of refaults on each list, times the relative IO cost between swap device and filesystem (swappiness), in order to optimize reclaim for least IO cost incurred. History I floated these changes in 2016. At the time they were incomplete and full of workarounds due to a lack of infrastructure in the reclaim code: We didn't have PageWorkingset, we didn't have hierarchical cgroup statistics, and problems with the cgroup swap controller. As swapping wasn't too high a priority then, the patches stalled out. With all dependencies in place now, here we are again with much cleaner, feature-complete patches. I kept the acks for patches that stayed materially the same :-) Below is a series of test results that demonstrate certain problematic behavior of the current code, as well as showcase the new code's more predictable and appropriate balancing decisions. Test #1: No convergence This test shows an edge case where the VM currently doesn't converge at all on a new file workingset with a stale anon/tmpfs set. The test sets up a cold anon set the size of 3/4 RAM, then tries to establish a new file set half the size of RAM (flat access pattern). The vanilla kernel refuses to even scan anon pages and never converges. The file set is perpetually served from the filesystem. The first test kernel is with the series up to the workingset patch applied. This allows thrashing page cache to challenge the anonymous workingset. The VM then scans the lists based on the current scanned/rotated balancing algorithm. It converges on a stable state where all cold anon pages are pushed out and the fileset is served entirely from cache: noconverge/5.7-rc5-mm noconverge/5.7-rc5-mm-workingset Scanned 417719308.00 ( +0.00%) 64091155.00 ( -84.66%) Reclaimed 417711094.00 ( +0.00%) 61640308.00 ( -85.24%) Reclaim efficiency % 100.00 ( +0.00%) 96.18 ( -3.78%) Scanned file 417719308.00 ( +0.00%) 59211118.00 ( -85.83%) Scanned anon 0.00 ( +0.00%) 4880037.00 ( ) Swapouts 0.00 ( +0.00%) 2439957.00 ( ) Swapins 0.00 ( +0.00%) 257.00 ( ) Refaults 415246605.00 ( +0.00%) 59183722.00 ( -85.75%) Restore refaults 0.00 ( +0.00%) 54988252.00 ( ) The second test kernel is with the full patch series applied, which replaces the scanned/rotated ratios with refault/swapin rate-based balancing. It evicts the cold anon pages more aggressively in the presence of a thrashing cache and the absence of swapins, and so converges with about 60% of the IO and reclaim activity: noconverge/5.7-rc5-mm-workingset noconverge/5.7-rc5-mm-lrubalance Scanned 64091155.00 ( +0.00%) 37579741.00 ( -41.37%) Reclaimed 61640308.00 ( +0.00%) 35129293.00 ( -43.01%) Reclaim efficiency % 96.18 ( +0.00%) 93.48 ( -2.78%) Scanned file 59211118.00 ( +0.00%) 32708385.00 ( -44.76%) Scanned anon 4880037.00 ( +0.00%) 4871356.00 ( -0.18%) Swapouts 2439957.00 ( +0.00%) 2435565.00 ( -0.18%) Swapins 257.00 ( +0.00%) 262.00 ( +1.94%) Refaults 59183722.00 ( +0.00%) 32675667.00 ( -44.79%) Restore refaults 54988252.00 ( +0.00%) 28480430.00 ( -48.21%) We're triggering this case in host sideloading scenarios: When a host's primary workload is not saturating the machine (primary load is usually driven by user activity), we can optimistically sideload a batch job; if user activity picks up and the primary workload needs the whole host during this time, we freeze the sideload and rely on it getting pushed to swap. Frequently that swapping doesn't happen and the completely inactive sideload simply stays resident while the expanding primary worklad is struggling to gain ground. Test #2: Kernel build This test is a a kernel build that is slightly memory-restricted (make -j4 inside a 400M cgroup). Despite the very aggressive swapping of cold anon pages in test #1, this test shows that the new kernel carefully balances swap against cache refaults when both the file and the cache set are pressured. It shows the patched kernel to be slightly better at finding the coldest memory from the combined anon and file set to evict under pressure. The result is lower aggregate reclaim and paging activity: z 5.7-rc5-mm 5.7-rc5-mm-lrubalance Real time 210.60 ( +0.00%) 210.97 ( +0.18%) User time 745.42 ( +0.00%) 746.48 ( +0.14%) System time 69.78 ( +0.00%) 69.79 ( +0.02%) Scanned file 354682.00 ( +0.00%) 293661.00 ( -17.20%) Scanned anon 465381.00 ( +0.00%) 378144.00 ( -18.75%) Swapouts 185920.00 ( +0.00%) 147801.00 ( -20.50%) Swapins 34583.00 ( +0.00%) 32491.00 ( -6.05%) Refaults 212664.00 ( +0.00%) 172409.00 ( -18.93%) Restore refaults 48861.00 ( +0.00%) 80091.00 ( +63.91%) Total paging IO 433167.00 ( +0.00%) 352701.00 ( -18.58%) Test #3: Overload This next test is not about performance, but rather about the predictability of the algorithm. The current balancing behavior doesn't always lead to comprehensible results, which makes performance analysis and parameter tuning (swappiness e.g.) very difficult. The test shows the balancing behavior under equivalent anon and file input. Anon and file sets are created of equal size (3/4 RAM), have the same access patterns (a hot-cold gradient), and synchronized access rates. Swappiness is raised from the default of 60 to 100 to indicate equal IO cost between swap and cache. With the vanilla balancing code, anon scans make up around 9% of the total pages scanned, or a ~1:10 ratio. This is a surprisingly skewed ratio, and it's an outcome that is hard to explain given the input parameters to the VM. The new balancing model targets a 1:2 balance: All else being equal, reclaiming a file page costs one page IO - the refault; reclaiming an anon page costs two IOs - the swapout and the swapin. In the test we observe a ~1:3 balance. The scanned and paging IO numbers indicate that the anon LRU algorithm we have in place right now does a slightly worse job at picking the coldest pages compared to the file algorithm. There is ongoing work to improve this, like Joonsoo's anon workingset patches; however, it's difficult to compare the two aging strategies when the balancing between them is behaving unintuitively. The slightly less efficient anon reclaim results in a deviation from the optimal 1:2 scan ratio we would like to see here - however, 1:3 is much closer to what we'd want to see in this test than the vanilla kernel's aging of 10+ cache pages for every anonymous one: overload-100/5.7-rc5-mm-workingset overload-100/5.7-rc5-mm-lrubalance-realfile Scanned 533633725.00 ( +0.00%) 595687785.00 ( +11.63%) Reclaimed 494325440.00 ( +0.00%) 518154380.00 ( +4.82%) Reclaim efficiency % 92.63 ( +0.00%) 86.98 ( -6.03%) Scanned file 484532894.00 ( +0.00%) 456937722.00 ( -5.70%) Scanned anon 49100831.00 ( +0.00%) 138750063.00 ( +182.58%) Swapouts 8096423.00 ( +0.00%) 48982142.00 ( +504.98%) Swapins 10027384.00 ( +0.00%) 62325044.00 ( +521.55%) Refaults 479819973.00 ( +0.00%) 451309483.00 ( -5.94%) Restore refaults 426422087.00 ( +0.00%) 399914067.00 ( -6.22%) Total paging IO 497943780.00 ( +0.00%) 562616669.00 ( +12.99%) Test #4: Parallel IO It's important to note that these patches only affect the situation where the kernel has to reclaim workingset memory, which is usually a transitionary period. The vast majority of page reclaim occuring in a system is from trimming the ever-expanding page cache. These patches don't affect cache trimming behavior. We never swap as long as we only have use-once cache moving through the file LRU, we only consider swapping when the cache is actively thrashing. The following test demonstrates this. It has an anon workingset that takes up half of RAM and then writes a file that is twice the size of RAM out to disk. As the cache is funneled through the inactive file list, no anon pages are scanned (aside from apparently some background noise of 10 pages): 5.7-rc5-mm 5.7-rc5-mm-lrubalance Scanned 10714722.00 ( +0.00%) 10723445.00 ( +0.08%) Reclaimed 10703596.00 ( +0.00%) 10712166.00 ( +0.08%) Reclaim efficiency % 99.90 ( +0.00%) 99.89 ( -0.00%) Scanned file 10714722.00 ( +0.00%) 10723435.00 ( +0.08%) Scanned anon 0.00 ( +0.00%) 10.00 ( ) Swapouts 0.00 ( +0.00%) 7.00 ( ) Swapins 0.00 ( +0.00%) 0.00 ( +0.00%) Refaults 92.00 ( +0.00%) 41.00 ( -54.84%) Restore refaults 0.00 ( +0.00%) 0.00 ( +0.00%) Total paging IO 92.00 ( +0.00%) 48.00 ( -47.31%) This patch (of 14): Currently, THP are counted as single pages until they are split right before being swapped out. However, at that point the VM is already in the middle of reclaim, and adjusting the LRU balance then is useless. Always account THP by the number of basepages, and remove the fixup from the splitting path. Signed-off-by:Shakeel Butt <shakeelb@google.com> Signed-off-by:
Rik van Riel <riel@surriel.com> Reviewed-by:
Michal Hocko <mhocko@suse.com> Acked-by:
Minchan Kim <minchan@kernel.org> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Link: http://lkml.kernel.org/r/20200520232525.798933-1-hannes@cmpxchg.org Link: http://lkml.kernel.org/r/20200520232525.798933-2-hannes@cmpxchg.orgSigned-off-by:
-
Johannes Weiner authored
The previous patches have simplified the access rules around page->mem_cgroup somewhat: 1. We never change page->mem_cgroup while the page is isolated by somebody else. This was by far the biggest exception to our rules and it didn't stop at lock_page() or lock_page_memcg(). 2. We charge pages before they get put into page tables now, so the somewhat fishy rule about "can be in page table as long as it's still locked" is now gone and boiled down to having an exclusive reference to the page. Document the new rules. Any of the following will stabilize the page->mem_cgroup association: - the page lock - LRU isolation - lock_page_memcg() - exclusive access to the page Signed-off-by:
Alex Shi <alex.shi@linux.alibaba.com> Reviewed-by:
Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Hugh Dickins <hughd@google.com> Cc: "Kirill A. Shutemov" <kirill@shutemov.name> Cc: Michal Hocko <mhocko@suse.com> Cc: Roman Gushchin <guro@fb.com> Cc: Shakeel Butt <shakeelb@google.com> Cc: Balbir Singh <bsingharora@gmail.com> Link: http://lkml.kernel.org/r/20200508183105.225460-20-hannes@cmpxchg.orgSigned-off-by:
-
Johannes Weiner authored
Swapin faults were the last event to charge pages after they had already been put on the LRU list. Now that we charge directly on swapin, the lrucare portion of the charge code is unused. Signed-off-by:
-
Alex Shi authored
Signed-off-by:
-
Johannes Weiner authored
Right now, users that are otherwise memory controlled can easily escape their containment and allocate significant amounts of memory that they're not being charged for. That's because swap readahead pages are not being charged until somebody actually faults them into their page table. This can be exploited with MADV_WILLNEED, which triggers arbitrary readahead allocations without charging the pages. There are additional problems with the delayed charging of swap pages: 1. To implement refault/workingset detection for anonymous pages, we need to have a target LRU available at swapin time, but the LRU is not determinable until the page has been charged. 2. To implement per-cgroup LRU locking, we need page->mem_cgroup to be stable when the page is isolated from the LRU; otherwise, the locks change under us. But swapcache gets charged after it's already on the LRU, and even if we cannot isolate it ourselves (since charging is not exactly optional). The previous patch ensured we always maintain cgroup ownership records for swap pages. This patch moves the swapcache charging point from the fault handler to swapin time to fix all of the above problems. v2: simplify swapin error checking (Joonsoo) [hughd@google.com: fix livelock in __read_swap_cache_async()] Link: http://lkml.kernel.org/r/alpine.LSU.2.11.2005212246080.8458@eggly.anvilsSigned-off-by:
Hugh Dickins <hughd@google.com> Signed-off-by:
-
Johannes Weiner authored
Without swap page tracking, users that are otherwise memory controlled can easily escape their containment and allocate significant amounts of memory that they're not being charged for. That's because swap does readahead, but without the cgroup records of who owned the page at swapout, readahead pages don't get charged until somebody actually faults them into their page table and we can identify an owner task. This can be maliciously exploited with MADV_WILLNEED, which triggers arbitrary readahead allocations without charging the pages. Make swap swap page tracking an integral part of memcg and remove the Kconfig options. In the first place, it was only made configurable to allow users to save some memory. But the overhead of tracking cgroup ownership per swap page is minimal - 2 byte per page, or 512k per 1G of swap, or 0.04%. Saving that at the expense of broken containment semantics is not something we should present as a coequal option. The swapaccount=0 boot option will continue to exist, and it will eliminate the page_counter overhead and hide the swap control files, but it won't disable swap slot ownership tracking. This patch makes sure we always have the cgroup records at swapin time; the next patch will fix the actual bug by charging readahead swap pages at swapin time rather than at fault time. v2: fix double swap charge bug in cgroup1/cgroup2 code gating [hannes@cmpxchg.org: fix crash with cgroup_disable=memory] Link: http://lkml.kernel.org/r/20200521215855.GB815153@cmpxchg.orgSigned-off-by:
Michal Hocko <mhocko@kernel.org> Signed-off-by:
-
Johannes Weiner authored
A few cleanups to streamline the swap controller setup: - Replace the do_swap_account flag with cgroup_memory_noswap. This brings it in line with other functionality that is usually available unless explicitly opted out of - nosocket, nokmem. - Remove the really_do_swap_account flag that stores the boot option and is later used to switch the do_swap_account. It's not clear why this indirection is/was necessary. Use do_swap_account directly. - Minor coding style polishing Signed-off-by:
-
Johannes Weiner authored
There are no more users. RIP in peace. [arnd@arndb.de: fix an unused-function warning] Link: http://lkml.kernel.org/r/20200528095640.151454-1-arnd@arndb.deSigned-off-by:
Arnd Bergmann <arnd@arndb.de> Signed-off-by:
-
Johannes Weiner authored
With the page->mapping requirement gone from memcg, we can charge anon and file-thp pages in one single step, right after they're allocated. This removes two out of three API calls - especially the tricky commit step that needed to happen at just the right time between when the page is "set up" and when it's "published" - somewhat vague and fluid concepts that varied by page type. All we need is a freshly allocated page and a memcg context to charge. v2: prevent double charges on pre-allocated hugepages in khugepaged [hannes@cmpxchg.org: Fix crash - *hpage could be ERR_PTR instead of NULL] Link: http://lkml.kernel.org/r/20200512215813.GA487759@cmpxchg.orgSigned-off-by:
-
Johannes Weiner authored
With rmap memcg locking already in place for NR_ANON_MAPPED, it's just a small step to remove the MEMCG_RSS_HUGE wart and switch memcg to the native NR_ANON_THPS accounting sites. [hannes@cmpxchg.org: fixes] Link: http://lkml.kernel.org/r/20200512121750.GA397968@cmpxchg.orgSigned-off-by:
Naresh Kamboju <naresh.kamboju@linaro.org> Reviewed-by:
-
Johannes Weiner authored
Memcg maintains a private MEMCG_RSS counter. This divergence from the generic VM accounting means unnecessary code overhead, and creates a dependency for memcg that page->mapping is set up at the time of charging, so that page types can be told apart. Convert the generic accounting sites to mod_lruvec_page_state and friends to maintain the per-cgroup vmstat counter of NR_ANON_MAPPED. We use lock_page_memcg() to stabilize page->mem_cgroup during rmap changes, the same way we do for NR_FILE_MAPPED. With the previous patch removing MEMCG_CACHE and the private NR_SHMEM counter, this patch finally eliminates the need to have page->mapping set up at charge time. However, we need to have page->mem_cgroup set up by the time rmap runs and does the accounting, so switch the commit and the rmap callbacks around. v2: fix temporary accounting bug by switching rmap<->commit (Joonsoo) Signed-off-by:
-
Johannes Weiner authored
Memcg maintains private MEMCG_CACHE and NR_SHMEM counters. This divergence from the generic VM accounting means unnecessary code overhead, and creates a dependency for memcg that page->mapping is set up at the time of charging, so that page types can be told apart. Convert the generic accounting sites to mod_lruvec_page_state and friends to maintain the per-cgroup vmstat counters of NR_FILE_PAGES and NR_SHMEM. The page is already locked in these places, so page->mem_cgroup is stable; we only need minimal tweaks of two mem_cgroup_migrate() calls to ensure it's set up in time. Then replace MEMCG_CACHE with NR_FILE_PAGES and delete the private NR_SHMEM accounting sites. Signed-off-by:
-
Johannes Weiner authored
Anonymous compound pages can be mapped by ptes, which means that if we want to track NR_MAPPED_ANON, NR_ANON_THPS on a per-cgroup basis, we have to be prepared to see tail pages in our accounting functions. Make mod_lruvec_page_state() and lock_page_memcg() deal with tail pages correctly, namely by redirecting to the head page which has the page->mem_cgroup set up. Signed-off-by:
-
Johannes Weiner authored
When memcg uses the generic vmstat counters, it doesn't need to do anything at charging and uncharging time. It does, however, need to migrate counts when pages move to a different cgroup in move_account. Prepare the move_account function for the arrival of NR_FILE_PAGES, NR_ANON_MAPPED, NR_ANON_THPS etc. by having a branch for files and a branch for anon, which can then divided into sub-branches. Signed-off-by:
-
Johannes Weiner authored
The uncharge batching code adds up the anon, file, kmem counts to determine the total number of pages to uncharge and references to drop. But the next patches will remove the anon and file counters. Maintain an aggregate nr_pages in the uncharge_gather struct. Signed-off-by:
-
Johannes Weiner authored
The try/commit/cancel protocol that memcg uses dates back to when pages used to be uncharged upon removal from the page cache, and thus couldn't be committed before the insertion had succeeded. Nowadays, pages are uncharged when they are physically freed; it doesn't matter whether the insertion was successful or not. For the page cache, the transaction dance has become unnecessary. Introduce a mem_cgroup_charge() function that simply charges a newly allocated page to a cgroup and sets up page->mem_cgroup in one single step. If the insertion fails, the caller doesn't have to do anything but free/put the page. Then switch the page cache over to this new API. Subsequent patches will also convert anon pages, but it needs a bit more prep work. Right now, memcg depends on page->mapping being already set up at the time of charging, so that it can maintain its own MEMCG_CACHE and MEMCG_RSS counters. For anon, page->mapping is set under the same pte lock under which the page is publishd, so a single charge point that can block doesn't work there just yet. The following prep patches will replace the private memcg counters with the generic vmstat counters, thus removing the page->mapping dependency, then complete the transition to the new single-point charge API and delete the old transactional scheme. v2: leave shmem swapcache when charging fails to avoid double IO (Joonsoo) v3: rebase on preceeding shmem simplification patch Signed-off-by:
-
Johannes Weiner authored
The cgroup swaprate throttling is about matching new anon allocations to the rate of available IO when that is being throttled. It's the io controller hooking into the VM, rather than a memory controller thing. Rename mem_cgroup_throttle_swaprate() to cgroup_throttle_swaprate(), and drop the @memcg argument which is only used to check whether the preceding page charge has succeeded and the fault is proceeding. We could decouple the call from mem_cgroup_try_charge() here as well, but that would cause unnecessary churn: the following patches convert all callsites to a new charge API and we'll decouple as we go along. Signed-off-by:
-
Johannes Weiner authored
Commit 215c02bc ("tmpfs: fix shmem_getpage_gfp() VM_BUG_ON") recognized that hole punching can race with swapin and removed the BUG_ON() for a truncated entry from the swapin path. The patch also added a swapcache deletion to optimize this rare case: Since swapin has the page locked, and free_swap_and_cache() merely trylocks, this situation can leave the page stranded in swapcache. Usually, page reclaim picks up stale swapcache pages, and the race can happen at any other time when the page is locked. (The same happens for non-shmem swapin racing with page table zapping.) The thinking here was: we already observed the race and we have the page locked, we may as well do the cleanup instead of waiting for reclaim. However, this optimization complicates the next patch which moves the cgroup charging code around. As this is just a minor speedup for a race condition that is so rare that it required a fuzzer to trigger the original BUG_ON(), it's no longer worth the complications. Suggested-by:
-
Johannes Weiner authored
The memcg charging API carries a boolean @compound parameter that tells whether the page we're dealing with is a hugepage. mem_cgroup_commit_charge() has another boolean @lrucare that indicates whether the page needs LRU locking or not while charging. The majority of callsites know those parameters at compile time, which results in a lot of naked "false, false" argument lists. This makes for cryptic code and is a breeding ground for subtle mistakes. Thankfully, the huge page state can be inferred from the page itself and doesn't need to be passed along. This is safe because charging completes before the page is published and somebody may split it. Simplify the callsites by removing @compound, and let memcg infer the state by using hpage_nr_pages() unconditionally. That function does PageTransHuge() to identify huge pages, which also helpfully asserts that nobody passes in tail pages by accident. The following patches will introduce a new charging API, best not to carry over unnecessary weight. Signed-off-by:
-
Johannes Weiner authored
The move_lock is a per-memcg lock, but the VM accounting code that needs to acquire it comes from the page and follows page->mem_cgroup under RCU protection. That means that the page becomes unlocked not when we drop the move_lock, but when we update page->mem_cgroup. And that assignment doesn't imply any memory ordering. If that pointer write gets reordered against the reads of the page state - page_mapped, PageDirty etc. the state may change while we rely on it being stable and we can end up corrupting the counters. Place an SMP memory barrier to make sure we're done with all page state by the time the new page->mem_cgroup becomes visible. Also replace the open-coded move_lock with a lock_page_memcg() to make it more obvious what we're serializing against. Signed-off-by:
-
Johannes Weiner authored
Patch series "mm: memcontrol: charge swapin pages on instantiation", v2. This patch series reworks memcg to charge swapin pages directly at swapin time, rather than at fault time, which may be much later, or not happen at all. Changes in version 2: - prevent double charges on pre-allocated hugepages in khugepaged - leave shmem swapcache when charging fails to avoid double IO (Joonsoo) - fix temporary accounting bug by switching rmap<->commit (Joonsoo) - fix double swap charge bug in cgroup1/cgroup2 code gating - simplify swapin error checking (Joonsoo) - mm: memcontrol: document the new swap control behavior (Alex) - review tags The delayed swapin charging scheme we have right now causes problems: - Alex's per-cgroup lru_lock patches rely on pages that have been isolated from the LRU to have a stable page->mem_cgroup; otherwise the lock may change underneath him. Swapcache pages are charged only after they are added to the LRU, and charging doesn't follow the LRU isolation protocol. - Joonsoo's anon workingset patches need a suitable LRU at the time the page enters the swap cache and displaces the non-resident info. But the correct LRU is only available after charging. - It's a containment hole / DoS vector. Users can trigger arbitrarily large swap readahead using MADV_WILLNEED. The memory is never charged unless somebody actually touches it. - It complicates the page->mem_cgroup stabilization rules In order to charge pages directly at swapin time, the memcg code base needs to be prepared, and several overdue cleanups become a necessity: To charge pages at swapin time, we need to always have cgroup ownership tracking of swap records. We also cannot rely on page->mapping to tell apart page types at charge time, because that's only set up during a page fault. To eliminate the page->mapping dependency, memcg needs to ditch its private page type counters (MEMCG_CACHE, MEMCG_RSS, NR_SHMEM) in favor of the generic vmstat counters and accounting sites, such as NR_FILE_PAGES, NR_ANON_MAPPED etc. To switch to generic vmstat counters, the charge sequence must be adjusted such that page->mem_cgroup is set up by the time these counters are modified. The series is structured as follows: 1. Bug fixes 2. Decoupling charging from rmap 3. Swap controller integration into memcg 4. Direct swapin charging This patch (of 19): When replacing one page with another one in the cache, we have to decrease the file count of the old page's NUMA node and increase the one of the new NUMA node, otherwise the old node leaks the count and the new node eventually underflows its counter. Fixes: 74d60958 ("page cache: Add and replace pages using the XArray") Signed-off-by:
Balbir Singh <bsingharora@gmail.com> Cc: Hugh Dickins <hughd@google.com> Cc: Michal Hocko <mhocko@suse.com> Cc: "Kirill A. Shutemov" <kirill@shutemov.name> Cc: Roman Gushchin <guro@fb.com> Link: http://lkml.kernel.org/r/20200508183105.225460-1-hannes@cmpxchg.org Link: http://lkml.kernel.org/r/20200508183105.225460-2-hannes@cmpxchg.orgSigned-off-by:
-
Qiwu Chen authored
try_to_compact_zone() has been replaced by try_to_compact_pages(), which is necessary to be updated in the comment of should_continue_reclaim(). Signed-off-by:
Qiwu Chen <chenqiwu@xiaomi.com> Signed-off-by:
-
Maninder Singh authored
commit 3c710c1a ("mm, vmscan extract shrink_page_list reclaim counters into a struct") changed data type for the function, so changing return type for funciton and its caller. Signed-off-by:
Vaneet Narang <v.narang@samsung.com> Signed-off-by:
Maninder Singh <maninder1.s@samsung.com> Signed-off-by:
-
Jaewon Kim authored
Fix an nr_isolate_* mismatch problem between cma and dirty lazyfree pages. If try_to_unmap_one is used for reclaim and it detects a dirty lazyfree page, then the lazyfree page is changed to a normal anon page having SwapBacked by commit 802a3a92 ("mm: reclaim MADV_FREE pages"). Even with the change, reclaim context correctly counts isolated files because it uses is_file_lru to distinguish file. And the change to anon is not happened if try_to_unmap_one is used for migration. So migration context like compaction also correctly counts isolated files even though it uses page_is_file_lru insted of is_file_lru. Recently page_is_file_cache was renamed to page_is_file_lru by commit 9de4f22a ("mm: code cleanup for MADV_FREE"). But the nr_isolate_* mismatch problem happens on cma alloc. There is reclaim_clean_pages_from_list which is being used only by cma. It was introduced by commit 02c6de8d ("mm: cma: discard clean pages during contiguous allocation instead of migration") to reclaim clean file pages without migration. The cma alloc uses both reclaim_clean_pages_from_list and migrate_pages, and it uses page_is_file_lru to count isolated files. If there are dirty lazyfree pages allocated from cma memory region, the pages are counted as isolated file at the beginging but are counted as isolated anon after finished. Mem-Info: Node 0 active_anon:3045904kB inactive_anon:611448kB active_file:14892kB inactive_file:205636kB unevictable:10416kB isolated(anon):0kB isolated(file):37664kB mapped:630216kB dirty:384kB writeback:0kB shmem:42576kB writeback_tmp:0kB unstable:0kB all_unreclaimable? no Like log above, there were too much isolated files, 37664kB, which triggers too_many_isolated in reclaim even when there is no actually isolated file in system wide. It could be reproducible by running two programs, writing on MADV_FREE page and doing cma alloc, respectively. Although isolated anon is 0, I found that the internal value of isolated anon was the negative value of isolated file. Fix this by compensating the isolated count for both LRU lists. Count non-discarded lazyfree pages in shrink_page_list, then compensate the counted number in reclaim_clean_pages_from_list. Reported-by:
Yong-Taek Lee <ytk.lee@samsung.com> Suggested-by:
Jaewon Kim <jaewon31.kim@samsung.com> Signed-off-by:
-
Wei Yang authored
We already defined the helper update_lru_size(). Let's use this to reduce code duplication. Signed-off-by:
Wei Yang <richard.weiyang@gmail.com> Signed-off-by:
Baoquan He <bhe@redhat.com> Reviewed-by:
-
Matthew Wilcox (Oracle) authored
None of the three callers of get_compound_page_dtor() want to know the value; they just want to call the function. Replace it with destroy_compound_page() which calls the dtor for them. Signed-off-by:
Matthew Wilcox (Oracle) <willy@infradead.org> Signed-off-by:
Anshuman Khandual <anshuman.khandual@arm.com> Reviewed-by:
David Hildenbrand <david@redhat.com> Acked-by:
Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Link: http://lkml.kernel.org/r/20200517105051.9352-1-willy@infradead.orgSigned-off-by:
-
Anshuman Khandual authored
There are multiple similar definitions for arch_clear_hugepage_flags() on various platforms. Lets just add it's generic fallback definition for platforms that do not override. This help reduce code duplication. Signed-off-by:
Mike Kravetz <mike.kravetz@oracle.com> Cc: Russell King <linux@armlinux.org.uk> Cc: Catalin Marinas <catalin.marinas@arm.com> Cc: Will Deacon <will@kernel.org> Cc: Tony Luck <tony.luck@intel.com> Cc: Fenghua Yu <fenghua.yu@intel.com> Cc: Thomas Bogendoerfer <tsbogend@alpha.franken.de> Cc: "James E.J. Bottomley" <James.Bottomley@HansenPartnership.com> Cc: Helge Deller <deller@gmx.de> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Cc: Paul Walmsley <paul.walmsley@sifive.com> Cc: Palmer Dabbelt <palmer@dabbelt.com> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Vasily Gorbik <gor@linux.ibm.com> Cc: Christian Borntraeger <borntraeger@de.ibm.com> Cc: Yoshinori Sato <ysato@users.sourceforge.jp> Cc: Rich Felker <dalias@libc.org> Cc: "David S. Miller" <davem@davemloft.net> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@redhat.com> Cc: Borislav Petkov <bp@alien8.de> Cc: "H. Peter Anvin" <hpa@zytor.com> Link: http://lkml.kernel.org/r/1588907271-11920-4-git-send-email-anshuman.khandual@arm.comSigned-off-by:
-
Anshuman Khandual authored
There are multiple similar definitions for is_hugepage_only_range() on various platforms. Lets just add it's generic fallback definition for platforms that do not override. This help reduce code duplication. Signed-off-by:
-
Anshuman Khandual authored
Patch series "mm/hugetlb: Add some new generic fallbacks", v3. This series adds the following new generic fallbacks. Before that it drops __HAVE_ARCH_HUGE_PTEP_GET from arm64 platform. 1. is_hugepage_only_range() 2. arch_clear_hugepage_flags() After this arm (32 bit) remains the sole platform defining it's own huge_ptep_get() via __HAVE_ARCH_HUGE_PTEP_GET. This patch (of 3): Platform specific huge_ptep_get() is required only when fetching the huge PTE involves more than just dereferencing the page table pointer. This is not the case on arm64 platform. Hence huge_ptep_pte() can be dropped along with it's __HAVE_ARCH_HUGE_PTEP_GET subscription. Before that, it updates the generic huge_ptep_get() with READ_ONCE() which will prevent known page table issues with THP on arm64. Signed-off-by:
-
Li Xinhai authored
When huge_pte_offset() is called, the parameter sz can only be PUD_SIZE or PMD_SIZE. If sz is PUD_SIZE and code can reach pud, then *pud must be none, or normal hugetlb entry, or non-present (migration or hwpoisoned) hugetlb entry, and we can directly return pud. When sz is PMD_SIZE, pud must be none or present, and if code can reach pmd, we can directly return pmd. So after this patch the code is simplified by first check on the parameter sz, and avoid unnecessary checks in current code. Same semantics of existing code is maintained. More details about relevant commits: commit 9b19df29 ("mm/hugetlb.c: make huge_pte_offset() consistent and document behaviour") changed the code path for pud and pmd handling, see comments about why this patch intends to change it. ... pud = pud_offset(p4d, addr); if (sz != PUD_SIZE && pud_none(*pud)) // [1] return NULL; /* hugepage or swap? */ if (pud_huge(*pud) || !pud_present(*pud)) // [2] return (pte_t *)pud; pmd = pmd_offset(pud, addr); if (sz != PMD_SIZE && pmd_none(*pmd)) // [3] return NULL; /* hugepage or swap? */ if (pmd_huge(*pmd) || !pmd_present(*pmd)) // [4] return (pte_t *)pmd; return NULL; // [5] ... [1]: this is necessary, return NULL for sz == PMD_SIZE; [2]: if sz == PUD_SIZE, all valid values of pud entry will cause return; [3]: dead code, sz != PMD_SIZE never true; [4]: all valid values of pmd entry will cause return; [5]: dead code, because of check in [4]. Now, this patch combines [1] and [2] for pud, and combines [3], [4] and [5] for pmd, so avoid unnecessary checks. I don't try to catch any invalid values in page table entry, as that will be checked by caller and avoid extra branch in this function. Also no assert on sz must equal PUD_SIZE or PMD_SIZE, since this function only call for hugetlb mapping. For commit 3c1d7e6c ("mm/hugetlb: fix a addressing exception caused by huge_pte_offset"), since we don't read the entry more than once now, variable pud_entry and pmd_entry are not needed. Signed-off-by:
Li Xinhai <lixinhai.lxh@gmail.com> Signed-off-by:
-
Mike Kravetz authored
Previously, a check for hugepages_supported was added before processing hugetlb command line parameters. On some architectures such as powerpc, hugepages_supported() is not set to true until after command line processing. Therefore, no hugetlb command line parameters would be accepted. Remove the additional checks for hugepages_supported. In hugetlb_init, print a warning if !hugepages_supported and command line parameters were specified. Reported-by:
Sandipan Das <sandipan.osd@gmail.com> Signed-off-by:
-
Mike Kravetz authored
With all hugetlb page processing done in a single file clean up code. - Make code match desired semantics - Update documentation with semantics - Make all warnings and errors messages start with 'HugeTLB:'. - Consistently name command line parsing routines. - Warn if !hugepages_supported() and command line parameters have been specified. - Add comments to code - Describe some of the subtle interactions - Describe semantics of command line arguments This patch also fixes issues with implicitly setting the number of gigantic huge pages to preallocate. Previously on X86 command line, hugepages=2 default_hugepagesz=1G would result in zero 1G pages being preallocated and, # grep HugePages_Total /proc/meminfo HugePages_Total: 0 # sysctl -a | grep nr_hugepages vm.nr_hugepages = 2 vm.nr_hugepages_mempolicy = 2 # cat /proc/sys/vm/nr_hugepages 2 After this patch 2 gigantic pages will be preallocated and all the proc, sysfs, sysctl and meminfo files will accurately reflect this. To address the issue with gigantic pages, a small change in behavior was made to command line processing. Previously the command line, hugepages=128 default_hugepagesz=2M hugepagesz=2M hugepages=256 would result in the allocation of 256 2M huge pages. The value 128 would be ignored without any warning. After this patch, 128 2M pages will be allocated and a warning message will be displayed indicating the value of 256 is ignored. This change in behavior is required because allocation of implicitly specified gigantic pages must be done when the default_hugepagesz= is encountered for gigantic pages. Previously the code waited until later in the boot process (hugetlb_init), to allocate pages of default size. However the bootmem allocator required for gigantic allocations is not available at this time. Signed-off-by:Sandipan Das <sandipan@linux.ibm.com> Acked-by: Gerald Schaefer <gerald.schaefer@de.ibm.com> [s390] Acked-by:
Will Deacon <will@kernel.org> Cc: Albert Ou <aou@eecs.berkeley.edu> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Catalin Marinas <catalin.marinas@arm.com> Cc: Christian Borntraeger <borntraeger@de.ibm.com> Cc: Christophe Leroy <christophe.leroy@c-s.fr> Cc: Dave Hansen <dave.hansen@linux.intel.com> Cc: David S. Miller <davem@davemloft.net> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jonathan Corbet <corbet@lwn.net> Cc: Longpeng <longpeng2@huawei.com> Cc: Mina Almasry <almasrymina@google.com> Cc: Nitesh Narayan Lal <nitesh@redhat.com> Cc: Palmer Dabbelt <palmer@dabbelt.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Paul Walmsley <paul.walmsley@sifive.com> Cc: Peter Xu <peterx@redhat.com> Cc: Randy Dunlap <rdunlap@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Vasily Gorbik <gor@linux.ibm.com> Cc: Anders Roxell <anders.roxell@linaro.org> Cc: "Aneesh Kumar K.V" <aneesh.kumar@linux.ibm.com> Cc: Qian Cai <cai@lca.pw> Cc: Stephen Rothwell <sfr@canb.auug.org.au> Link: http://lkml.kernel.org/r/20200417185049.275845-5-mike.kravetz@oracle.comSigned-off-by:
-
Mike Kravetz authored
hugetlb_add_hstate() prints a warning if the hstate already exists. This was originally done as part of kernel command line parsing. If 'hugepagesz=' was specified more than once, the warning pr_warn("hugepagesz= specified twice, ignoring\n"); would be printed. Some architectures want to enable all huge page sizes. They would call hugetlb_add_hstate for all supported sizes. However, this was done after command line processing and as a result hstates could have already been created for some sizes. To make sure no warning were printed, there would often be code like: if (!size_to_hstate(size) hugetlb_add_hstate(ilog2(size) - PAGE_SHIFT) The only time we want to print the warning is as the result of command line processing. So, remove the warning from hugetlb_add_hstate and add it to the single arch independent routine processing "hugepagesz=". After this, calls to size_to_hstate() in arch specific code can be removed and hugetlb_add_hstate can be called without worrying about warning messages. [mike.kravetz@oracle.com: fix hugetlb initialization] Link: http://lkml.kernel.org/r/4c36c6ce-3774-78fa-abc4-b7346bf24348@oracle.com Link: http://lkml.kernel.org/r/20200428205614.246260-5-mike.kravetz@oracle.comSigned-off-by:Anders Roxell <anders.roxell@linaro.org> Acked-by:
Mina Almasry <almasrymina@google.com> Acked-by: Gerald Schaefer <gerald.schaefer@de.ibm.com> [s390] Acked-by:
-
Mike Kravetz authored
Now that architectures provide arch_hugetlb_valid_size(), parsing of "hugepagesz=" can be done in architecture independent code. Create a single routine to handle hugepagesz= parsing and remove all arch specific routines. We can also remove the interface hugetlb_bad_size() as this is no longer used outside arch independent code. This also provides consistent behavior of hugetlbfs command line options. The hugepagesz= option should only be specified once for a specific size, but some architectures allow multiple instances. This appears to be more of an oversight when code was added by some architectures to set up ALL huge pages sizes. Signed-off-by:
Peter Xu <peterx@redhat.com> Acked-by:
-
Mike Kravetz authored
Patch series "Clean up hugetlb boot command line processing", v4. Longpeng(Mike) reported a weird message from hugetlb command line processing and proposed a solution [1]. While the proposed patch does address the specific issue, there are other related issues in command line processing. As hugetlbfs evolved, updates to command line processing have been made to meet immediate needs and not necessarily in a coordinated manner. The result is that some processing is done in arch specific code, some is done in arch independent code and coordination is problematic. Semantics can vary between architectures. The patch series does the following: - Define arch specific arch_hugetlb_valid_size routine used to validate passed huge page sizes. - Move hugepagesz= command line parsing out of arch specific code and into an arch independent routine. - Clean up command line processing to follow desired semantics and document those semantics. [1] https://lore.kernel.org/linux-mm/20200305033014.1152-1-longpeng2@huawei.com This patch (of 3): The architecture independent routine hugetlb_default_setup sets up the default huge pages size. It has no way to verify if the passed value is valid, so it accepts it and attempts to validate at a later time. This requires undocumented cooperation between the arch specific and arch independent code. For architectures that support more than one huge page size, provide a routine arch_hugetlb_valid_size to validate a huge page size. hugetlb_default_setup can use this to validate passed values. arch_hugetlb_valid_size will also be used in a subsequent patch to move processing of the "hugepagesz=" in arch specific code to a common routine in arch independent code. Signed-off-by:
-
Kirill A. Shutemov authored
'max_ptes_shared' specifies how many pages can be shared across multiple processes. Exceeding the number would block the collapse:: /sys/kernel/mm/transparent_hugepage/khugepaged/max_ptes_shared A higher value may increase memory footprint for some workloads. By default, at least half of pages has to be not shared. [colin.king@canonical.com: fix several spelling mistakes] Link: http://lkml.kernel.org/r/20200420084241.65433-1-colin.king@canonical.comSigned-off-by:
Colin Ian King <colin.king@canonical.com> Signed-off-by:
Zi Yan <ziy@nvidia.com> Reviewed-by:
William Kucharski <william.kucharski@oracle.com> Reviewed-by:
Yang Shi <yang.shi@linux.alibaba.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: John Hubbard <jhubbard@nvidia.com> Cc: Mike Kravetz <mike.kravetz@oracle.com> Cc: Ralph Campbell <rcampbell@nvidia.com> Link: http://lkml.kernel.org/r/20200416160026.16538-9-kirill.shutemov@linux.intel.comSigned-off-by:
-
Kirill A. Shutemov authored
Currently we have different copy-on-write semantics for anon- and file-THP. For anon-THP we try to allocate huge page on the write fault, but on file-THP we split PMD and allocate 4k page. Arguably, file-THP semantics is more desirable: we don't necessary want to unshare full PMD range from the parent on the first access. This is the primary reason THP is unusable for some workloads, like Redis. The original THP refcounting didn't allow to have PTE-mapped compound pages, so we had no options, but to allocate huge page on CoW (with fallback to 512 4k pages). The current refcounting doesn't have such limitations and we can cut a lot of complex code out of fault path. khugepaged is now able to recover THP from such ranges if the configuration allows. Signed-off-by:
-
Kirill A. Shutemov authored
We can collapse PTE-mapped compound pages. We only need to avoid handling them more than once: lock/unlock page only once if it's present in the PMD range multiple times as it handled on compound level. The same goes for LRU isolation and putback. Signed-off-by:
-
