- 05 Mar, 2024 8 commits
-
-
Vishal Moola (Oracle) authored
Patch series "Handle hugetlb faults under the VMA lock", v2. It is generally safe to handle hugetlb faults under the VMA lock. The only time this is unsafe is when no anon_vma has been allocated to this vma yet, so we can use vmf_anon_prepare() instead of anon_vma_prepare() to bailout if necessary. This should only happen for the first hugetlb page in the vma. Additionally, this patchset begins to use struct vm_fault within hugetlb_fault(). This works towards cleaning up hugetlb code, and should significantly reduce the number of arguments passed to functions. The last patch in this series may cause ltp hugemmap10 to "fail". This is because vmf_anon_prepare() may bailout with no anon_vma under the VMA lock after allocating a folio for the hugepage. In free_huge_folio(), this folio is completely freed on bailout iff there is a surplus of hugetlb pages. This will remove a folio off the freelist and decrement the number of hugepages while ltp expects these counters to remain unchanged on failure. The rest of the ltp testcases pass. This patch (of 2): In order to handle hugetlb faults under the VMA lock, hugetlb can use vmf_anon_prepare() to ensure we can safely prepare an anon_vma. Change it to be a non-static function so it can be used within hugetlb as well. Link: https://lkml.kernel.org/r/20240221234732.187629-6-vishal.moola@gmail.com Link: https://lkml.kernel.org/r/20240221234732.187629-2-vishal.moola@gmail.comSigned-off-by:
Vishal Moola (Oracle) <vishal.moola@gmail.com> Reviewed-by:
Matthew Wilcox (Oracle) <willy@infradead.org> Cc: Muchun Song <muchun.song@linux.dev> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org>
-
Hao Ge authored
Make check_new_page() return bool like check_new_pages() Link: https://lkml.kernel.org/r/20240222091932.54799-1-gehao@kylinos.cnSigned-off-by:
Hao Ge <gehao@kylinos.cn> Reviewed-by:
-
Yosry Ahmed authored
The first argument of switch_mm_irqs_off() is unused by the x86 implementation. Make sure that x86 code never passes a non-NULL value to make this clear. Update the only non violating caller, switch_mm(). Link: https://lkml.kernel.org/r/20240222190911.1903054-2-yosryahmed@google.comSigned-off-by:
Yosry Ahmed <yosryahmed@google.com> Suggested-by:
Dave Hansen <dave.hansen@intel.com> Acked-by:
-
Yosry Ahmed authored
Commit accf6b23d1e5a ("x86/mm: clarify "prev" usage in switch_mm_irqs_off()") attempted to clarify x86's usage of the arguments passed by generic code, specifically the "prev" argument the is unused by x86. However, it could have done a better job with the comment above switch_mm_irqs_off(). Rewrite this comment according to Dave Hansen's suggestion. Link: https://lkml.kernel.org/r/20240222190911.1903054-1-yosryahmed@google.com Fixes: 3cfd6625 ("x86/mm: clarify "prev" usage in switch_mm_irqs_off()") Signed-off-by: -
Matthew Cassell authored
Commit 44b414c8 ("mm/util.c: add warning if __vm_enough_memory fails") adds debug information which gives the process id and executable name should __vm_enough_memory() fail. Adding the number of pages to the failure message would benefit application developers and system administrators in debugging overambitious memory requests by providing a point of reference to the amount of memory causing __vm_enough_memory() to fail. 1. Set appropriate kernel tunable to reach code path for failure message: # echo 2 > /proc/sys/vm/overcommit_memory 2. Test program to generate failure - requests 1 gibibyte per iteration: #include <stdlib.h> #include <stdio.h> int main(int argc, char **argv) { for(;;) { if(malloc(1<<30) == NULL) break; printf("allocated 1 GiB\n"); } return 0; } 3. Output: Before: __vm_enough_memory: pid: 1218, comm: a.out, not enough memory for the allocation After: __vm_enough_memory: pid: 1137, comm: a.out, bytes: 1073741824, not enough memory for the allocation Link: https://lkml.kernel.org/r/20240222194617.1255-1-mcassell411@gmail.comSigned-off-by:
Matthew Cassell <mcassell411@gmail.com> Cc: David Hildenbrand <david@redhat.com> Signed-off-by:
-
Byungchul Park authored
Memoryless nodes do not have any memory to migrate to, so, as an optimization, stop trying it. Link: https://lkml.kernel.org/r/20240219041920.1183-1-byungchul@sk.com Link: https://lkml.kernel.org/r/20240216111502.79759-1-byungchul@sk.com Fixes: c574bbe9 ("NUMA balancing: optimize page placement for memory tiering system") Signed-off-by:
Byungchul Park <byungchul@sk.com> Reviewed-by:
Oscar Salvador <osalvador@suse.de> Reviewed-by:
"Huang, Ying" <ying.huang@intel.com> Reviewed-by:
Phil Auld <pauld@redhat.com> Reviewed-by:
Davidlohr Bueso <dave@stgolabs.net> Acked-by:
David Hildenbrand <david@redhat.com> Cc: Benjamin Segall <bsegall@google.com> Cc: Daniel Bristot de Oliveira <bristot@redhat.com> Cc: Dietmar Eggemann <dietmar.eggemann@arm.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Juri Lelli <juri.lelli@redhat.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Valentin Schneider <vschneid@redhat.com> Cc: Vincent Guittot <vincent.guittot@linaro.org> Signed-off-by:
-
Chengming Zhou authored
All zswap entries will take a reference of zswap_pool when zswap_store(), and drop it when free. Change it to use the percpu_ref is better for scalability performance. Although percpu_ref use a bit more memory which should be ok for our use case, since we almost have only one zswap_pool to be using. The performance gain is for zswap_store/load hotpath. Testing kernel build (32 threads) in tmpfs with memory.max=2GB. (zswap shrinker and writeback enabled with one 50GB swapfile, on a 128 CPUs x86-64 machine, below is the average of 5 runs) mm-unstable zswap-global-lru real 63.20 63.12 user 1061.75 1062.95 sys 268.74 264.44 [chengming.zhou@linux.dev: fix zswap_pools_lock usages after changing to percpu_ref] Link: https://lkml.kernel.org/r/20240228154954.3028626-1-chengming.zhou@linux.dev Link: https://lkml.kernel.org/r/20240210-zswap-global-lru-v3-2-200495333595@bytedance.comSigned-off-by:Chengming Zhou <zhouchengming@bytedance.com> Reviewed-by:
Nhat Pham <nphamcs@gmail.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Yosry Ahmed <yosryahmed@google.com> Cc: Chengming Zhou <chengming.zhou@linux.dev> Signed-off-by:
-
Chengming Zhou authored
Patch series "mm/zswap: optimize for dynamic zswap_pools", v3. Dynamic pool creation has been supported for a long time, which maybe not used so much in practice. But with the per-memcg lru merged, the current structure of zswap_pool's lru and shrinker become less optimal. In the current structure, each zswap_pool has its own lru, shrinker and shrink_work, but only the latest zswap_pool will be the current used. 1. When memory has pressure, all shrinkers of zswap_pools will try to shrink its lru list, there is no order between them. 2. When zswap limit hit, only the last zswap_pool's shrink_work will try to shrink its own lru, which is inefficient. A more natural way is to have a global zswap lru shared between all zswap_pools, and so is the shrinker. The code becomes much simpler too. Another optimization is changing zswap_pool kref to percpu_ref, which will be taken reference by every zswap entry. So the scalability is better. Testing kernel build (32 threads) in tmpfs with memory.max=2GB. (zswap shrinker and writeback enabled with one 50GB swapfile, on a 128 CPUs x86-64 machine, below is the average of 5 runs) mm-unstable zswap-global-lru real 63.20 63.12 user 1061.75 1062.95 sys 268.74 264.44 This patch (of 3): Dynamic zswap_pool creation may create/reuse to have multiple zswap_pools in a list, only the first will be current used. Each zswap_pool has its own lru and shrinker, which is not necessary and has its problem: 1. When memory has pressure, all shrinker of zswap_pools will try to shrink its own lru, there is no order between them. 2. When zswap limit hit, only the last zswap_pool's shrink_work will try to shrink its lru list. The rationale here was to try and empty the old pool first so that we can completely drop it. However, since we only support exclusive loads now, the LRU ordering should be entirely decided by the order of stores, so the oldest entries on the LRU will naturally be from the oldest pool. Anyway, having a global lru and shrinker shared by all zswap_pools is better and efficient. Link: https://lkml.kernel.org/r/20240210-zswap-global-lru-v3-0-200495333595@bytedance.com Link: https://lkml.kernel.org/r/20240210-zswap-global-lru-v3-1-200495333595@bytedance.comSigned-off-by:
-
- 24 Feb, 2024 32 commits
-
-
Matthew Wilcox (Oracle) authored
Use the new writeback_iter() directly instead of indirecting through a callback. [hch@lst.de: ported to the while based iter style] Link: https://lkml.kernel.org/r/20240215063649.2164017-15-hch@lst.deSigned-off-by:
Christoph Hellwig <hch@lst.de> Reviewed-by:
Brian Foster <bfoster@redhat.com> Reviewed-by:
Jan Kara <jack@suse.cz> Cc: Christian Brauner <brauner@kernel.org> Cc: Dave Chinner <dchinner@redhat.com> Cc: David Howells <dhowells@redhat.com> Signed-off-by:
-
Christoph Hellwig authored
Refactor the code left in write_cache_pages into an iterator that the file system can call to get the next folio for a writeback operation: struct folio *folio = NULL; while ((folio = writeback_iter(mapping, wbc, folio, &error))) { error = <do per-folio writeback>; } The twist here is that the error value is passed by reference, so that the iterator can restore it when breaking out of the loop. Handling of the magic AOP_WRITEPAGE_ACTIVATE value stays outside the iterator and needs is just kept in the write_cache_pages legacy wrapper. in preparation for eventually killing it off. Heavily based on a for_each* based iterator from Matthew Wilcox. Link: https://lkml.kernel.org/r/20240215063649.2164017-14-hch@lst.deSigned-off-by: -
Matthew Wilcox (Oracle) authored
Move the loop for should-we-write-this-folio to writeback_get_folio. [hch@lst.de: fold loop into existing helper instead of a separate one per Jan] Link: https://lkml.kernel.org/r/20240215063649.2164017-13-hch@lst.deSigned-off-by:
Dave Chinner <dchinner@redhat.com> Cc: Christian Brauner <brauner@kernel.org> Cc: David Howells <dhowells@redhat.com> Signed-off-by:
-
Matthew Wilcox (Oracle) authored
Instead of keeping our own local iterator variable, use the one just added to folio_batch. Link: https://lkml.kernel.org/r/20240215063649.2164017-12-hch@lst.deSigned-off-by:
-
Matthew Wilcox (Oracle) authored
Add a loop counter inside the folio_batch to let us iterate from 0-nr instead of decrementing nr and treating the batch as a stack. It would generate some very weird and suboptimal I/O patterns for page writeback to iterate over the batch as a stack. Link: https://lkml.kernel.org/r/20240215063649.2164017-11-hch@lst.deSigned-off-by:
-
Matthew Wilcox (Oracle) authored
Collapse the two nested loops into one. This is needed as a step towards turning this into an iterator. Note that this drops the "index <= end" check in the previous outer loop and just relies on filemap_get_folios_tag() to return 0 entries when index > end. This actually has a subtle implication when end == -1 because then the returned index will be -1 as well and thus if there is page present on index -1, we could be looping indefinitely. But as the comment in filemap_get_folios_tag documents this as already broken anyway we should not worry about it here either. The fix for that would probably a change to the filemap_get_folios_tag() calling convention. [hch@lst.de: update the commit log per Jan] Link: https://lkml.kernel.org/r/20240215063649.2164017-10-hch@lst.deSigned-off-by:
-
Matthew Wilcox (Oracle) authored
This simple helper will be the basis of the writeback iterator. To make this work, we need to remember the current index and end positions in writeback_control. [hch@lst.de: heavily rebased, add helpers to get the tag and end index, don't keep the end index in struct writeback_control] Link: https://lkml.kernel.org/r/20240215063649.2164017-9-hch@lst.deSigned-off-by:
-
Matthew Wilcox (Oracle) authored
Reduce write_cache_pages() by about 30 lines; much of it is commentary, but it all bundles nicely into an obvious function. [hch@lst.de: rename should_writeback_folio to folio_prepare_writeback per Jan] Link: https://lkml.kernel.org/r/20240215063649.2164017-8-hch@lst.deSigned-off-by:
-
Christoph Hellwig authored
Rework the way we deal with the cleanup after the writepage call. First handle the magic AOP_WRITEPAGE_ACTIVATE separately from real error returns to get it out of the way of the actual error handling path. The split the handling on intgrity vs non-integrity branches first, and return early using a goto for the non-ingegrity early loop condition to remove the need for the done and done_index local variables, and for assigning the error to ret when we can just return error directly. Link: https://lkml.kernel.org/r/20240215063649.2164017-7-hch@lst.deSigned-off-by:
-
Christoph Hellwig authored
mapping->writeback_index is only [1] used as the starting point for range_cyclic writeback, so there is no point in updating it for other types of writeback. [1] except for btrfs_defrag_file which does really odd things with mapping->writeback_index. But btrfs doesn't use write_cache_pages at all, so this isn't relevant here. Link: https://lkml.kernel.org/r/20240215063649.2164017-6-hch@lst.deSigned-off-by:
-
Christoph Hellwig authored
When exiting write_cache_pages early due to a non-integrity write failure, wbc->nr_to_write currently doesn't account for the folio we just failed to write. This doesn't matter because the callers always ingore the value on a failure, but moving the update to common code will allow to simplify the code, so do it. Link: https://lkml.kernel.org/r/20240215063649.2164017-5-hch@lst.deSigned-off-by:
-
Christoph Hellwig authored
When write_cache_pages finishes writing out a folio, it fails to update done_index to account for the number of pages in the folio just written. That means when range_cyclic writeback is restarted, it will be restarted at this folio instead of after it as it should. Fix that by updating done_index before breaking out of the loop. Link: https://lkml.kernel.org/r/20240215063649.2164017-4-hch@lst.deSigned-off-by:
-
Matthew Wilcox (Oracle) authored
[hch@lst.de: split from a larger patch] Link: https://lkml.kernel.org/r/20240215063649.2164017-3-hch@lst.deSigned-off-by:
-
Christoph Hellwig authored
Patch series "convert write_cache_pages() to an iterator", v8. This is an evolution of the series Matthew Wilcox originally sent in June 2023, which has changed quite a bit since and now has a while based iterator. This patch (of 14): mapping_set_error should only be called on 0 returns (which it ignores) or a negative error code. writepage_cb ends up being able to call writepage_cb on the magic AOP_WRITEPAGE_ACTIVATE return value from ->writepage which means success but the caller needs to unlock the page. Ignore that and just call mapping_set_error on negative errors. (no fixes tag as this goes back more than 20 years over various renames and refactors so I've given up chasing down the original introduction) Link: https://lkml.kernel.org/r/20240215063649.2164017-1-hch@lst.de Link: https://lkml.kernel.org/r/20240215063649.2164017-2-hch@lst.deSigned-off-by:
-
Hao Ge authored
bad_range() can return bool, so let us change it. Link: https://lkml.kernel.org/r/20240221073227.276234-1-gehao@kylinos.cnSigned-off-by:
-
Barry Song authored
The purpose is stopping splitting large folios whose mapcount are 2 or above. Folios whose estimated_shares = 0 should be still perfect and even better candidates than estimated_shares = 1. Consider a pte-mapped large folio with 16 subpages, if we unmap 1-15, the current code will split folios and reclaim them while madvise goes on this folio; but if we unmap subpage 0, we will keep this folio and break. This is weird. For pmd-mapped large folios, we can still use "= 1" as the condition as anyway we have the entire map for it. So this patch doesn't change the condition for pmd-mapped large folios. This also explains why we had been using "= 1" for both pmd-mapped and pte-mapped large folios before commit 07e8c82b ("madvise: convert madvise_cold_or_pageout_pte_range() to use folios"), because in the past, we used the mapcount of the specific subpage, since the subpage had pte present, its mapcount wouldn't be 0. The problem can be quite easily reproduced by writing a small program, unmapping the first subpage of a pte-mapped large folio vs. unmapping anyone other than the first subpage. Link: https://lkml.kernel.org/r/20240221085036.105621-1-21cnbao@gmail.com Fixes: 2f406263 ("madvise:madvise_cold_or_pageout_pte_range(): don't use mapcount() against large folio for sharing check") Signed-off-by:
Barry Song <v-songbaohua@oppo.com> Reviewed-by:
-
Barry Song authored
The code is quite hard to read, we are still writing swap_map after errors happen. Though the written value is as before, has_cache = count & SWAP_HAS_CACHE; count &= ~SWAP_HAS_CACHE; [snipped] WRITE_ONCE(p->swap_map[offset], count | has_cache); It would be better to entirely drop the WRITE_ONCE for both performance and readability. [akpm@linux-foundation.org: avoid using goto] Link: https://lkml.kernel.org/r/20240221091028.123122-1-21cnbao@gmail.comSigned-off-by:
-
Jan Kara authored
Report quota options among the set of mount options. This allows proper user visibility into whether quotas are enabled or not. Link: https://lkml.kernel.org/r/20240129120131.21145-1-jack@suse.cz Fixes: e09764cf ("shmem: quota support") Signed-off-by:
Carlos Maiolino <cmaiolino@redhat.com> Acked-by:
Hugh Dickins <hughd@google.com> Signed-off-by:
-
Zi Yan authored
During migration in a memory compaction, free pages are placed in an array of page lists based on their order. But the desired free page order (i.e., the order of a source page) might not be always present, thus leading to migration failures and premature compaction termination. Split a high order free pages when source migration page has a lower order to increase migration successful rate. Note: merging free pages when a migration fails and a lower order free page is returned via compaction_free() is possible, but there is too much work. Since the free pages are not buddy pages, it is hard to identify these free pages using existing PFN-based page merging algorithm. Link: https://lkml.kernel.org/r/20240220183220.1451315-5-zi.yan@sent.comSigned-off-by:
Zi Yan <ziy@nvidia.com> Reviewed-by:
Baolin Wang <baolin.wang@linux.alibaba.com> Reviewed-by:
Vlastimil Babka <vbabka@suse.cz> Tested-by:
Yu Zhao <yuzhao@google.com> Cc: Adam Manzanares <a.manzanares@samsung.com> Cc: David Hildenbrand <david@redhat.com> Cc: Huang Ying <ying.huang@intel.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Kemeng Shi <shikemeng@huaweicloud.com> Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Cc: Luis Chamberlain <mcgrof@kernel.org> Cc: Matthew Wilcox (Oracle) <willy@infradead.org> Cc: Mel Gorman <mgorman@techsingularity.net> Cc: Ryan Roberts <ryan.roberts@arm.com> Cc: Vishal Moola (Oracle) <vishal.moola@gmail.com> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: Yin Fengwei <fengwei.yin@intel.com> Signed-off-by:
-
Zi Yan authored
Before last commit, memory compaction only migrates order-0 folios and skips >0 order folios. Last commit splits all >0 order folios during compaction. This commit migrates >0 order folios during compaction by keeping isolated free pages at their original size without splitting them into order-0 pages and using them directly during migration process. What is different from the prior implementation: 1. All isolated free pages are kept in a NR_PAGE_ORDERS array of page lists, where each page list stores free pages in the same order. 2. All free pages are not post_alloc_hook() processed nor buddy pages, although their orders are stored in first page's private like buddy pages. 3. During migration, in new page allocation time (i.e., in compaction_alloc()), free pages are then processed by post_alloc_hook(). When migration fails and a new page is returned (i.e., in compaction_free()), free pages are restored by reversing the post_alloc_hook() operations using newly added free_pages_prepare_fpi_none(). Step 3 is done for a latter optimization that splitting and/or merging free pages during compaction becomes easier. Note: without splitting free pages, compaction can end prematurely due to migration will return -ENOMEM even if there is free pages. This happens when no order-0 free page exist and compaction_alloc() return NULL. Link: https://lkml.kernel.org/r/20240220183220.1451315-4-zi.yan@sent.comSigned-off-by:
-
Zi Yan authored
migrate_pages() supports >0 order folio migration and during compaction, even if compaction_alloc() cannot provide >0 order free pages, migrate_pages() can split the source page and try to migrate the base pages from the split. It can be a baseline and start point for adding support for compacting >0 order folios. Link: https://lkml.kernel.org/r/20240220183220.1451315-3-zi.yan@sent.comSigned-off-by:
-
Zi Yan authored
Patch series "Enable >0 order folio memory compaction", v7. This patchset enables >0 order folio memory compaction, which is one of the prerequisitions for large folio support[1]. I am aware of that split free pages is necessary for folio migration in compaction, since if >0 order free pages are never split and no order-0 free page is scanned, compaction will end prematurely due to migration returns -ENOMEM. Free page split becomes a must instead of an optimization. lkp ncompare results (on a 8-CPU (Intel Xeon E5-2650 v4 @2.20GHz) 16G VM) for default LRU (-no-mglru) and CONFIG_LRU_GEN are shown at the bottom, copied from V3[4]. In sum, most of vm-scalability applications do not see performance change, and the others see ~4% to ~26% performance boost under default LRU and ~2% to ~6% performance boost under CONFIG_LRU_GEN. Overview === To support >0 order folio compaction, the patchset changes how free pages used for migration are kept during compaction. Free pages used to be split into order-0 pages that are post allocation processed (i.e., PageBuddy flag cleared, page order stored in page->private is zeroed, and page reference is set to 1). Now all free pages are kept in a NR_PAGE_ORDER array of page lists based on their order without post allocation process. When migrate_pages() asks for a new page, one of the free pages, based on the requested page order, is then processed and given out. And THP <2MB would need this feature. [1] https://lore.kernel.org/linux-mm/f8d47176-03a8-99bf-a813-b5942830fd73@arm.com/ [2] https://lore.kernel.org/linux-mm/20231113170157.280181-1-zi.yan@sent.com/ [3] https://lore.kernel.org/linux-mm/20240123034636.1095672-1-zi.yan@sent.com/ [4] https://lore.kernel.org/linux-mm/20240202161554.565023-1-zi.yan@sent.com/ [5] https://lore.kernel.org/linux-mm/20240212163510.859822-1-zi.yan@sent.com/ [6] https://lore.kernel.org/linux-mm/20240214220420.1229173-1-zi.yan@sent.com/ [7] https://lore.kernel.org/linux-mm/20240216170432.1268753-1-zi.yan@sent.com/ This patch (of 4): Commit 0a54864f ("kasan: remove PG_skip_kasan_poison flag") removes the use of fpi_flags in should_skip_kasan_poison() and fpi_flags is only passed to should_skip_kasan_poison() in free_pages_prepare(). Remove the unused parameter. Link: https://lkml.kernel.org/r/20240220183220.1451315-1-zi.yan@sent.com Link: https://lkml.kernel.org/r/20240220183220.1451315-2-zi.yan@sent.comSigned-off-by:
-
Chengming Zhou authored
I have been actively contributing to zswap and reviewing zswap patches for a while, and I am already getting CC'd on most of them. So add myself as a reviewer, will continue to work on it and help with the review process. Link: https://lkml.kernel.org/r/20240220073851.865113-1-chengming.zhou@linux.devSigned-off-by:
-
Chengming Zhou authored
Actually we seldom use the class_idx returned from get_zspage_mapping(), only the zspage->fullness is useful, just use zspage->fullness to remove this helper. Note zspage->fullness is not stable outside pool->lock, remove redundant "VM_BUG_ON(fullness != ZS_INUSE_RATIO_0)" in async_free_zspage() since we already have the same VM_BUG_ON() in __free_zspage(), which is safe to access zspage->fullness with pool->lock held. Link: https://lkml.kernel.org/r/20240220-b4-zsmalloc-cleanup-v1-3-5c5ee4ccdd87@bytedance.comSigned-off-by:
Sergey Senozhatsky <senozhatsky@chromium.org> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Minchan Kim <minchan@kernel.org> Cc: Nhat Pham <nphamcs@gmail.com> Cc: Yosry Ahmed <yosryahmed@google.com> Signed-off-by:
-
Chengming Zhou authored
We must remove_zspage() from its current fullness list, then use insert_zspage() to update its fullness and insert to new fullness list. Obviously, remove_zspage() doesn't need the fullness parameter. Link: https://lkml.kernel.org/r/20240220-b4-zsmalloc-cleanup-v1-2-5c5ee4ccdd87@bytedance.comSigned-off-by:
-
Chengming Zhou authored
Patch series "mm/zsmalloc: some cleanup for get/set_zspage_mapping()". The discussion[1] with Sergey shows there are some cleanup works to do in get/set_zspage_mapping(): - the fullness returned from get_zspage_mapping() is not stable outside pool->lock, this usage pattern is confusing, but should be ok in this free_zspage path. - we seldom use the class_idx returned from get_zspage_mapping(), only free_zspage path use to get its class. - set_zspage_mapping() always set the zspage->class, but it's never changed after zspage allocated. [1] https://lore.kernel.org/all/a6c22e30-cf10-4122-91bc-ceb9fb57a5d6@bytedance.com/ This patch (of 3): We only need to update zspage->fullness when insert_zspage(), since zspage->class is never changed after allocated. Link: https://lkml.kernel.org/r/20240220-b4-zsmalloc-cleanup-v1-0-5c5ee4ccdd87@bytedance.com Link: https://lkml.kernel.org/r/20240220-b4-zsmalloc-cleanup-v1-1-5c5ee4ccdd87@bytedance.comSigned-off-by:
-
Kefeng Wang authored
No need to continue try compact memory if pending fatal signal, allow loop termination earlier in compact_nodes(). The existing fatal_signal_pending() check does make compact_zone() break out of the while loop, but it still enters the next zone/next nid, and some unnecessary functions(eg, lru_add_drain) are called. There was no observable benefit from the new test, it is just found from code inspection when refactoring compact_node(). Link: https://lkml.kernel.org/r/20240208022508.1771534-1-wangkefeng.wang@huawei.comSigned-off-by:
Kefeng Wang <wangkefeng.wang@huawei.com> Cc: David Hildenbrand <david@redhat.com> Signed-off-by:
-
Barry Song authored
We used to rely on the returned -ENOSPC of zpool_malloc() to increase reject_compress_poor. But the code wouldn't get to there after commit 744e1885 ("crypto: scomp - fix req->dst buffer overflow") as the new code will goto out immediately after the special compression case happens. So there might be no longer a chance to execute zpool_malloc now. We are incorrectly increasing zswap_reject_compress_fail instead. Thus, we need to fix the counters handling right after compressions return ENOSPC. This patch also centralizes the counters handling for all of compress_poor, compress_fail and alloc_fail. Link: https://lkml.kernel.org/r/20240219211935.72394-1-21cnbao@gmail.com Fixes: 744e1885 ("crypto: scomp - fix req->dst buffer overflow") Signed-off-by:
-
Zhongkun He authored
The comment is confusing that Pool lock should be held as this function accesses first_num above the __encode_handle() because first_num is the element of z3fold_header which is protected by z3fold_header->page_lock. I found the same comment for encode_handle() in zbud.c by accident ,Pool lock should be held as this function accesses first|last_chunks, which is the element of zbud_header and it does not have any lock, so pool lock should be held. Z3fold is based on zbud, maybe the comment come from zbud, but it was wrong, so fix it. Link: https://lkml.kernel.org/r/20240219024453.2240147-1-hezhongkun.hzk@bytedance.comSigned-off-by:
Zhongkun He <hezhongkun.hzk@bytedance.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Miaohe Lin <linmiaohe@huawei.com> Cc: Vitaly Wool <vitaly.wool@konsulko.com> Signed-off-by:
-
Chengming Zhou authored
The zspage->isolated is not used anywhere, we don't need to maintain it, which needs to hold the heavy pool lock to update it, so just remove it. Link: https://lkml.kernel.org/r/20240219-b4-szmalloc-migrate-v1-3-34cd49c6545b@bytedance.comSigned-off-by:
-
Chengming Zhou authored
The migrate write lock is to protect the race between zspage migration and zspage objects' map users. We only need to lock out the map users of src zspage, not dst zspage, which is safe to map by users concurrently, since we only need to do obj_malloc() from dst zspage. So we can remove the migrate_write_lock_nested() use case. As we are here, cleanup the __zs_compact() by moving putback_zspage() outside of migrate_write_unlock since we hold pool lock, no malloc or free users can come in. Link: https://lkml.kernel.org/r/20240219-b4-szmalloc-migrate-v1-2-34cd49c6545b@bytedance.comSigned-off-by:
-
Chengming Zhou authored
Patch series "mm/zsmalloc: fix and optimize objects/page migration". This series is to fix and optimize the zsmalloc objects/page migration. This patch (of 3): migrate_write_lock() is a empty function when !CONFIG_COMPACTION, in which case zs_compact() can be triggered from shrinker reclaim context. (Maybe it's better to rename it to zs_shrink()?) And zspage map object users rely on this migrate_read_lock() so object won't be migrated elsewhere. Fix it by always implementing the migrate_write_lock() related functions. Link: https://lkml.kernel.org/r/20240219-b4-szmalloc-migrate-v1-0-34cd49c6545b@bytedance.com Link: https://lkml.kernel.org/r/20240219-b4-szmalloc-migrate-v1-1-34cd49c6545b@bytedance.comSigned-off-by:
-
