- 14 Oct, 2020 40 commits
-
-
Matthew Wilcox (Oracle) authored
Patch series "Fix PageDoubleMap". This is a purely theoretical problem for now as none of the filesystems which use PG_private_2 (ie PG_fscache) are being converted at this time, but it's confusing to leave it like this. This patch (of 2): PG_private_2 is defined as being PF_ANY (applicable to tail pages as well as regular & head pages). That means that the first tail page of a double-map page will appear to have Private2 set. Use the Workingset bit instead which is defined as PF_HEAD so any attempt to access the Workingset bit on a tail page will redirect to the head page's Workingset bit. Signed-off-by:
Matthew Wilcox (Oracle) <willy@infradead.org> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Reviewed-by:
Zi Yan <ziy@nvidia.com> Link: https://lkml.kernel.org/r/20200629151933.15671-1-willy@infradead.org Link: https://lkml.kernel.org/r/20200629151933.15671-2-willy@infradead.orgSigned-off-by:
Linus Torvalds <torvalds@linux-foundation.org>
-
Chinwen Chang authored
smaps_rollup will try to grab mmap_lock and go through the whole vma list until it finishes the iterating. When encountering large processes, the mmap_lock will be held for a longer time, which may block other write requests like mmap and munmap from progressing smoothly. There are upcoming mmap_lock optimizations like range-based locks, but the lock applied to smaps_rollup would be the coarse type, which doesn't avoid the occurrence of unpleasant contention. To solve aforementioned issue, we add a check which detects whether anyone wants to grab mmap_lock for write attempts. Signed-off-by:
Chinwen Chang <chinwen.chang@mediatek.com> Signed-off-by:
-
Chinwen Chang authored
Extend smap_gather_stats to support indicated beginning address at which it should start gathering. To achieve the goal, we add a new parameter @start assigned by the caller and try to refactor it for simplicity. If @start is 0, it will use the range of @vma for gathering. Signed-off-by:
Steven Price <steven.price@arm.com> Cc: Michel Lespinasse <walken@google.com> Cc: Alexey Dobriyan <adobriyan@gmail.com> Cc: Daniel Jordan <daniel.m.jordan@oracle.com> Cc: Daniel Kiss <daniel.kiss@arm.com> Cc: Davidlohr Bueso <dbueso@suse.de> Cc: Huang Ying <ying.huang@intel.com> Cc: Jason Gunthorpe <jgg@ziepe.ca> Cc: Jimmy Assarsson <jimmyassarsson@gmail.com> Cc: Laurent Dufour <ldufour@linux.ibm.com> Cc: "Matthew Wilcox (Oracle)" <willy@infradead.org> Cc: Matthias Brugger <matthias.bgg@gmail.com> Cc: Song Liu <songliubraving@fb.com> Cc: Vlastimil Babka <vbabka@suse.cz> Link: http://lkml.kernel.org/r/1597715898-3854-3-git-send-email-chinwen.chang@mediatek.comSigned-off-by:
-
Chinwen Chang authored
Patch series "Try to release mmap_lock temporarily in smaps_rollup", v4. Recently, we have observed some janky issues caused by unpleasantly long contention on mmap_lock which is held by smaps_rollup when probing large processes. To address the problem, we let smaps_rollup detect if anyone wants to acquire mmap_lock for write attempts. If yes, just release the lock temporarily to ease the contention. smaps_rollup is a procfs interface which allows users to summarize the process's memory usage without the overhead of seq_* calls. Android uses it to sample the memory usage of various processes to balance its memory pool sizes. If no one wants to take the lock for write requests, smaps_rollup with this patch will behave like the original one. Although there are on-going mmap_lock optimizations like range-based locks, the lock applied to smaps_rollup would be the coarse one, which is hard to avoid the occurrence of aforementioned issues. So the detection and temporary release for write attempts on mmap_lock in smaps_rollup is still necessary. This patch (of 3): Add new API to query if someone wants to acquire mmap_lock for write attempts. Using this instead of rwsem_is_contended makes it more tolerant of future changes to the lock type. Signed-off-by:
Michel Lespinasse <walken@google.com> Cc: Alexey Dobriyan <adobriyan@gmail.com> Cc: Daniel Jordan <daniel.m.jordan@oracle.com> Cc: Daniel Kiss <daniel.kiss@arm.com> Cc: Davidlohr Bueso <dbueso@suse.de> Cc: Huang Ying <ying.huang@intel.com> Cc: Jason Gunthorpe <jgg@ziepe.ca> Cc: Jimmy Assarsson <jimmyassarsson@gmail.com> Cc: Laurent Dufour <ldufour@linux.ibm.com> Cc: "Matthew Wilcox (Oracle)" <willy@infradead.org> Cc: Matthias Brugger <matthias.bgg@gmail.com> Cc: Song Liu <songliubraving@fb.com> Cc: Vlastimil Babka <vbabka@suse.cz> Link: http://lkml.kernel.org/r/1597715898-3854-1-git-send-email-chinwen.chang@mediatek.com Link: http://lkml.kernel.org/r/1597715898-3854-2-git-send-email-chinwen.chang@mediatek.comSigned-off-by:
-
Wei Yang authored
These two functions share the same logic except ignore a different vma. Let's reuse the code. Signed-off-by:
Wei Yang <richard.weiyang@linux.alibaba.com> Signed-off-by:
-
Wei Yang authored
__vma_unlink_common() and __vma_unlink() are counterparts. Since there is no function named __vma_unlink(), let's rename __vma_unlink_common() to __vma_unlink() to make the code more self-explanatory and easy for audience to understand. Otherwise we may expect there are several variants of vma_unlink() and __vma_unlink_common() is used by them. Signed-off-by:
-
Yanfei Xu authored
The code has declared a vma_struct named vma which is assigned a value of vmf->vma. Thus, use variable vma directly here. Signed-off-by:
Yanfei Xu <yanfei.xu@windriver.com> Signed-off-by:
-
Yanfei Xu authored
It's "pte_alloc_one", not "pte_alloc_pne". Let's fix that. Signed-off-by:
David Hildenbrand <david@redhat.com> Link: http://lkml.kernel.org/r/20200818104339.5310-1-yanfei.xu@windriver.comSigned-off-by:
-
Matthew Wilcox authored
We account the PTE level of the page tables to the process in order to make smarter OOM decisions and help diagnose why memory is fragmented. For these same reasons, we should account pages allocated for PMDs. With larger process address spaces and ASLR, the number of PMDs in use is higher than it used to be so the inaccuracy is starting to matter. [rppt@linux.ibm.com: arm: __pmd_free_tlb(): call page table destructor] Link: https://lkml.kernel.org/r/20200825111303.GB69694@linux.ibm.comSigned-off-by:
Mike Rapoport <rppt@linux.ibm.com> Cc: Abdul Haleem <abdhalee@linux.vnet.ibm.com> Cc: Andy Lutomirski <luto@kernel.org> Cc: Arnd Bergmann <arnd@arndb.de> Cc: Christophe Leroy <christophe.leroy@csgroup.eu> Cc: Joerg Roedel <joro@8bytes.org> Cc: Max Filippov <jcmvbkbc@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Satheesh Rajendran <sathnaga@linux.vnet.ibm.com> Cc: Stafford Horne <shorne@gmail.com> Cc: Naresh Kamboju <naresh.kamboju@linaro.org> Cc: Anders Roxell <anders.roxell@linaro.org> Link: http://lkml.kernel.org/r/20200627184642.GF25039@casper.infradead.orgSigned-off-by:
-
John Hubbard authored
Avoid accidental wrong builds, due to built-in rules working just a little bit too well--but not quite as well as required for our situation here. In other words, "make userfaultfd" (for example) is supposed to fail to build at all, because this Makefile only supports either "make" (all), or "make /full/path". However, the built-in rules, if not suppressed, will pick up CFLAGS and the initial LDLIBS (but not the target-specific LDLIBS, because those are only set for the full path target!). This causes it to get pretty far into building things despite using incorrect values such as an *occasionally* incomplete LDLIBS value. Signed-off-by:
John Hubbard <jhubbard@nvidia.com> Signed-off-by:
-
John Hubbard authored
Patch series "selftests/vm: fix some minor aggravating factors in the Makefile". This fixes a couple of minor aggravating factors that I ran across while trying to do some changes in selftests/vm. These are simple things, but like most things with GNU Make, it's rarely obvious what's wrong until you understand *the entire Makefile and all of its includes*. So while there is, of course, joy in learning those details, I thought I'd fix these little things, so as to allow others to skip out on the Joy if they so choose. :) First of all, if you have an item (let's choose userfaultfd for an example) that fails to build, you might do this: $ make -j32 # ...you observe a failed item in the threaded output # OK, let's get a closer look $ make # ...but now the build quietly "succeeds". That's what Patch 0001 fixes. Second, if you instead attempt this approach for your closer look (a casual mistake, as it's not supported): $ make userfaultfd # ...userfaultfd fails to link, due to incomplete LDLIBS That's what Patch 0002 fixes. This patch (of 2): If one or more of these selftest fail to build, then after the first failure, subsequent invocations of "make" will make it appear that there are no build failures, after all. That's because the failed build products remain, with up-to-date timestamps, thus tricking Make (and you!) into believing that there's nothing else to build. Fix this by telling Make to delete targets that didn't completely succeed. Signed-off-by: -
Ralph Campbell authored
The code in mc_handle_swap_pte() checks for non_swap_entry() and returns NULL before checking is_device_private_entry() so device private pages are never handled. Fix this by checking for non_swap_entry() after handling device private swap PTEs. I assume the memory cgroup accounting would be off somehow when moving a process to another memory cgroup. Currently, the device private page is charged like a normal anonymous page when allocated and is uncharged when the page is freed so I think that path is OK. Signed-off-by:
Ralph Campbell <rcampbell@nvidia.com> Signed-off-by:
Johannes Weiner <hannes@cmpxchg.org> Cc: Michal Hocko <mhocko@kernel.org> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Cc: Jerome Glisse <jglisse@redhat.com> Cc: Balbir Singh <bsingharora@gmail.com> Cc: Ira Weiny <ira.weiny@intel.com> Link: https://lkml.kernel.org/r/20201009215952.2726-1-rcampbell@nvidia.com xFixes: c733a828 ("mm/memcontrol: support MEMORY_DEVICE_PRIVATE") Signed-off-by:
-
Bharata B Rao authored
Object cgroup charging is done for all the objects during allocation, but during freeing, uncharging ends up happening for only one object in the case of bulk allocation/freeing. Fix this by having a separate call to uncharge all the objects from kmem_cache_free_bulk() and by modifying memcg_slab_free_hook() to take care of bulk uncharging. Fixes: 964d4bd3 ("mm: memcg/slab: save obj_cgroup for non-root slab objects" Signed-off-by:
Bharata B Rao <bharata@linux.ibm.com> Signed-off-by:
Roman Gushchin <guro@fb.com> Cc: Christoph Lameter <cl@linux.com> Cc: David Rientjes <rientjes@google.com> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: Shakeel Butt <shakeelb@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Michal Hocko <mhocko@kernel.org> Cc: Tejun Heo <tj@kernel.org> Cc: <stable@vger.kernel.org> Link: https://lkml.kernel.org/r/20201009060423.390479-1-bharata@linux.ibm.comSigned-off-by:
-
Miaohe Lin authored
Since commit 79dfdacc ("memcg: make oom_lock 0 and 1 based rather than counter"), the mem_cgroup_unmark_under_oom() is added and the comment of the mem_cgroup_oom_unlock() is moved here. But this comment make no sense here because mem_cgroup_oom_lock() does not operate on under_oom field. So we reword the comment as this would be helpful. [Thanks Michal Hocko for rewording this comment.] Signed-off-by:
Miaohe Lin <linmiaohe@huawei.com> Signed-off-by:
Michal Hocko <mhocko@suse.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Link: https://lkml.kernel.org/r/20200930095336.21323-1-linmiaohe@huawei.comSigned-off-by:
-
Miaohe Lin authored
Since commit bbec2e15 ("mm: rename page_counter's count/limit into usage/max"), page_counter_limit() is renamed to page_counter_set_max(). So replace page_counter_limit with page_counter_set_max in comment. Signed-off-by:
-
Muchun Song authored
In the cgroup v1, we have a numa_stat interface. This is useful for providing visibility into the numa locality information within an memcg since the pages are allowed to be allocated from any physical node. One of the use cases is evaluating application performance by combining this information with the application's CPU allocation. But the cgroup v2 does not. So this patch adds the missing information. Suggested-by:
Shakeel Butt <shakeelb@google.com> Signed-off-by:
Muchun Song <songmuchun@bytedance.com> Signed-off-by:
-
Waiman Long authored
The swap page counter is v2 only while memsw is v1 only. As v1 and v2 controllers cannot be active at the same time, there is no point to keep both swap and memsw page counters in mem_cgroup. The previous patch has made sure that memsw page counter is updated and accessed only when in v1 code paths. So it is now safe to alias the v1 memsw page counter to v2 swap page counter. This saves 14 long's in the size of mem_cgroup. This is a saving of 112 bytes for 64-bit archs. While at it, also document which page counters are used in v1 and/or v2. Signed-off-by:
Waiman Long <longman@redhat.com> Signed-off-by:
-
Waiman Long authored
mem_cgroup_get_max() used to get memory+swap max from both the v1 memsw and v2 memory+swap page counters & return the maximum of these 2 values. This is redundant and it is more efficient to just get either the v1 or the v2 values depending on which one is currently in use. [longman@redhat.com: v4] Link: https://lkml.kernel.org/r/20200914150928.7841-1-longman@redhat.comSigned-off-by:
-
Waiman Long authored
Patch series "mm/memcg: Miscellaneous cleanups and streamlining", v2. This patch (of 3): Since commit 0a31bc97 ("mm: memcontrol: rewrite uncharge API") and commit 00501b53 ("mm: memcontrol: rewrite charge API") in v3.17, the enum charge_type was no longer used anywhere. However, the enum itself was not removed at that time. Remove the obsolete enum charge_type now. Signed-off-by:
Chris Down <chris@chrisdown.name> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Cc: Tejun Heo <tj@kernel.org> Cc: Roman Gushchin <guro@fb.com> Cc: Yafang Shao <laoar.shao@gmail.com> Link: https://lkml.kernel.org/r/20200914024452.19167-1-longman@redhat.com Link: https://lkml.kernel.org/r/20200914024452.19167-2-longman@redhat.comSigned-off-by:
-
Miaohe Lin authored
Since commit bbec2e15 ("mm: rename page_counter's count/limit into usage/max"), the arg @reclaim has no priority field anymore. Signed-off-by:
-
Roman Gushchin authored
mem_cgroup_from_obj() checks the lowest bit of the page->mem_cgroup pointer to determine if the page has an attached obj_cgroup vector instead of a regular memcg pointer. If it's not set, it simple returns the page->mem_cgroup value as a struct mem_cgroup pointer. The commit 10befea9 ("mm: memcg/slab: use a single set of kmem_caches for all allocations") changed the moment when this bit is set: if previously it was set on the allocation of the slab page, now it can be set well after, when the first accounted object is allocated on this page. It opened a race: if page->mem_cgroup is set concurrently after the first page_has_obj_cgroups(page) check, a pointer to the obj_cgroups array can be returned as a memory cgroup pointer. A simple check for page->mem_cgroup pointer for NULL before the page_has_obj_cgroups() check fixes the race. Indeed, if the pointer is not NULL, it's either a simple mem_cgroup pointer or a pointer to obj_cgroup vector. The pointer can be asynchronously changed from NULL to (obj_cgroup_vec | 0x1UL), but can't be changed from a valid memcg pointer to objcg vector or back. If the object passed to mem_cgroup_from_obj() is a slab object and page->mem_cgroup is NULL, it means that the object is not accounted, so the function must return NULL. I've discovered the race looking at the code, so far I haven't seen it in the wild. Fixes: 10befea9 ("mm: memcg/slab: use a single set of kmem_caches for all allocations") Signed-off-by:
-
Gustavo A. R. Silva authored
Use the preferred form for passing the size of a structure type. The alternative form where the structure type is spelled out hurts readability and introduces an opportunity for a bug when the object type is changed but the corresponding object identifier to which the sizeof operator is applied is not. Signed-off-by:
Gustavo A. R. Silva <gustavoars@kernel.org> Signed-off-by:
-
Gustavo A. R. Silva authored
Make use of the flex_array_size() helper to calculate the size of a flexible array member within an enclosing structure. This helper offers defense-in-depth against potential integer overflows, while at the same time makes it explicitly clear that we are dealing with a flexible array member. Also, remove unnecessary braces. Signed-off-by:
-
Ira Weiny authored
While reviewing Protection Key Supervisor support it was pointed out that using a counter to track static branch enable was an anti-pattern which was better solved using the provided static_branch_{inc,dec} functions.[1] Fix up devmap_managed_key to work the same way. Also this should be safer because there is a very small (very unlikely) race when multiple callers try to enable at the same time. [1] https://lore.kernel.org/lkml/20200714194031.GI5523@worktop.programming.kicks-ass.net/Signed-off-by:Ira Weiny <ira.weiny@intel.com> Signed-off-by:
William Kucharski <william.kucharski@oracle.com> Cc: Dan Williams <dan.j.williams@intel.com> Cc: Vishal Verma <vishal.l.verma@intel.com> Link: https://lkml.kernel.org/r/20200810235319.2796597-1-ira.weiny@intel.comSigned-off-by:
-
Miaohe Lin authored
If we failed to drain inode, we would forget to free the swap address space allocated by init_swap_address_space() above. Fixes: dc617f29 ("vfs: don't allow writes to swap files") Signed-off-by:
Darrick J. Wong <darrick.wong@oracle.com> Link: https://lkml.kernel.org/r/20200930101803.53884-1-linmiaohe@huawei.comSigned-off-by:
-
Miaohe Lin authored
It's unnecessary to goto the out label while out label is just below. Signed-off-by:
-
Miaohe Lin authored
Since commit 9c4e6b1a ("mm, mlock, vmscan: no more skipping pagevecs"), unevictable pages do not goes directly back onto zone's unevictable list. Signed-off-by:
-
Miaohe Lin authored
The out label is only used in one place and return ret directly without something like resource cleanup or lock release and so on. So we should remove this jump label and do some cleanup. Signed-off-by:
-
Miaohe Lin authored
enable_swap_slots_cache() always return zero and its return value is just ignored by the caller. So make enable_swap_slots_cache() void. Signed-off-by:
-
Miaohe Lin authored
Since commit 07d80269 ("mm: devmap: refactor 1-based refcounting for ZONE_DEVICE pages"), we have renamed the func put_devmap_managed_page() to page_is_devmap_managed(). Signed-off-by:
-
Yu Zhao authored
To activate a page, mark_page_accessed() always holds a reference on it. It either gets a new reference when adding a page to lru_pvecs.activate_page or reuses an existing one it previously got when it added a page to lru_pvecs.lru_add. So it doesn't call SetPageActive() on a page that doesn't have any reference left. Therefore, the race is impossible these days (I didn't brother to dig into its history). For other paths, namely reclaim and migration, a reference count is always held while calling SetPageActive() on a page. SetPageSlabPfmemalloc() also uses SetPageActive(), but it's irrelevant to LRU pages. Signed-off-by:
Yu Zhao <yuzhao@google.com> Signed-off-by:
Yang Shi <shy828301@gmail.com> Cc: Alexander Duyck <alexander.h.duyck@linux.intel.com> Cc: David Hildenbrand <david@redhat.com> Cc: Huang Ying <ying.huang@intel.com> Cc: Hugh Dickins <hughd@google.com> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Mel Gorman <mgorman@techsingularity.net> Cc: Michal Hocko <mhocko@suse.com> Cc: Nicholas Piggin <npiggin@gmail.com> Cc: Qian Cai <cai@lca.pw> Link: http://lkml.kernel.org/r/20200818184704.3625199-2-yuzhao@google.comSigned-off-by:
-
Yu Zhao authored
We don't initially add anon pages to active lruvec after commit b518154e ("mm/vmscan: protect the workingset on anonymous LRU"). Remove activate_page() from unuse_pte(), which seems to be missed by the commit. And make the function static while we are at it. Before the commit, we called lru_cache_add_active_or_unevictable() to add new ksm pages to active lruvec. Therefore, activate_page() wasn't necessary for them in the first place. Signed-off-by:
-
Gao Xiang authored
SWP_FS is used to make swap_{read,write}page() go through the filesystem, and it's only used for swap files over NFS for now. Otherwise it will directly submit IO to blockdev according to swapfile extents reported by filesystems in advance. As Matthew pointed out [1], SWP_FS naming is somewhat confusing, so let's rename to SWP_FS_OPS. [1] https://lore.kernel.org/r/20200820113448.GM17456@casper.infradead.orgSuggested-by:Gao Xiang <hsiangkao@redhat.com> Signed-off-by:
-
John Hubbard authored
As suggested by Dan Carpenter, fortify unpin_user_pages() just a bit, against a typical caller mistake: check if the npages arg is really a -ERRNO value, which would blow up the unpinning loop: WARN and return. If this new WARN_ON() fires, then the system *might* be leaking pages (by leaving them pinned), but probably not. More likely, gup/pup returned a hard -ERRNO error to the caller, who erroneously passed it here. Signed-off-by:
Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by:
-
Barry Song authored
gup prohibits users from calling get_user_pages() with FOLL_PIN. But it allows users to call get_user_pages() with FOLL_LONGTERM only. It seems insensible. Since FOLL_LONGTERM is a stricter case of FOLL_PIN, we should prohibit users from calling get_user_pages() with FOLL_LONGTERM while not with FOLL_PIN. mm/gup_benchmark.c used to be the only user who did this improperly. But it has been fixed by moving to use pin_user_pages(). [akpm@linux-foundation.org: fix CONFIG_MMU=n build] Link: https://lkml.kernel.org/r/CA+G9fYuNS3k0DVT62twfV746pfNhCSrk5sVMcOcQ1PGGnEseyw@mail.gmail.comSigned-off-by:
Barry Song <song.bao.hua@hisilicon.com> Signed-off-by:
-
Barry Song authored
According to Documentation/core-api/pin_user_pages.rst, FOLL_PIN is a prerequisite to FOLL_LONGTERM. Another way of saying that is, FOLL_LONGTERM is a specific case, more restrictive case of FOLL_PIN. Almost all kernel modules are using pin_user_pages() with FOLL_LONGTERM, mm/gup_benchmark.c seems to the only exception in which FOLL_PIN is not a prerequisite to FOLL_LONGTERM. Signed-off-by:
-
Barry Song authored
In the beginning, mm/gup_benchmark.c supported get_user_pages_fast() only, but right now, it supports the benchmarking of a couple of get_user_pages() related calls like: * get_user_pages_fast() * get_user_pages() * pin_user_pages_fast() * pin_user_pages() The documentation is confusing and needs update. Signed-off-by:
-
Yafang Shao authored
Our users reported that there're some random latency spikes when their RT process is running. Finally we found that latency spike is caused by FADV_DONTNEED. Which may call lru_add_drain_all() to drain LRU cache on remote CPUs, and then waits the per-cpu work to complete. The wait time is uncertain, which may be tens millisecond. That behavior is unreasonable, because this process is bound to a specific CPU and the file is only accessed by itself, IOW, there should be no pagecache pages on a per-cpu pagevec of a remote CPU. That unreasonable behavior is partially caused by the wrong comparation of the number of invalidated pages and the number of the target. For example, if (count < (end_index - start_index + 1)) The count above is how many pages were invalidated in the local CPU, and (end_index - start_index + 1) is how many pages should be invalidated. The usage of (end_index - start_index + 1) is incorrect, because they are virtual addresses, which may not mapped to pages. Besides that, there may be holes between start and end. So we'd better check whether there are still pages on per-cpu pagevec after drain the local cpu, and then decide whether or not to call lru_add_drain_all(). After I applied it with a hotfix to our production environment, most of the lru_add_drain_all() can be avoided. Suggested-by:Mel Gorman <mgorman@suse.de> Signed-off-by:
Yafang Shao <laoar.shao@gmail.com> Signed-off-by:
-
Matthew Wilcox (Oracle) authored
We dereference page->mapping and page->index directly after calling find_subpage() and these fields are not valid for tail pages. While commit 4101196b ("mm: page cache: store only head pages in i_pages") introduced the call to find_subpage(), the problem existed prior to this; I'm going to suggest all the way back to when THPs first existed. The user-visible effects of this are almost negligible. To hit it, you have to mmap a tmpfs file at an unaligned address and then it's only a disabled optimisation causing page faults to happen more frequently than they otherwise would. Fix this by keeping both head and page pointers and checking the appropriate one. We could use page_mapping() and page_to_index(), but that's higher overhead. Signed-off-by:
Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Cc: William Kucharski <william.kucharski@oracle.com> Link: https://lkml.kernel.org/r/20200911012532.24761-1-willy@infradead.orgSigned-off-by:
-
Matthew Wilcox (Oracle) authored
Add a new FGP_HEAD flag which avoids calling find_subpage() and add a convenience wrapper for it. Signed-off-by:
-
