- 12 Dec, 2023 19 commits
-
-
Nhat Pham authored
Patch series "workload-specific and memory pressure-driven zswap writeback", v8. There are currently several issues with zswap writeback: 1. There is only a single global LRU for zswap, making it impossible to perform worload-specific shrinking - an memcg under memory pressure cannot determine which pages in the pool it owns, and often ends up writing pages from other memcgs. This issue has been previously observed in practice and mitigated by simply disabling memcg-initiated shrinking: https://lore.kernel.org/all/20230530232435.3097106-1-nphamcs@gmail.com/T/#u But this solution leaves a lot to be desired, as we still do not have an avenue for an memcg to free up its own memory locked up in the zswap pool. 2. We only shrink the zswap pool when the user-defined limit is hit. This means that if we set the limit too high, cold data that are unlikely to be used again will reside in the pool, wasting precious memory. It is hard to predict how much zswap space will be needed ahead of time, as this depends on the workload (specifically, on factors such as memory access patterns and compressibility of the memory pages). This patch series solves these issues by separating the global zswap LRU into per-memcg and per-NUMA LRUs, and performs workload-specific (i.e memcg- and NUMA-aware) zswap writeback under memory pressure. The new shrinker does not have any parameter that must be tuned by the user, and can be opted in or out on a per-memcg basis. As a proof of concept, we ran the following synthetic benchmark: build the linux kernel in a memory-limited cgroup, and allocate some cold data in tmpfs to see if the shrinker could write them out and improved the overall performance. Depending on the amount of cold data generated, we observe from 14% to 35% reduction in kernel CPU time used in the kernel builds. This patch (of 6): The interface of list_lru is based on the assumption that the list node and the data it represents belong to the same allocated on the correct node/memcg. While this assumption is valid for existing slab objects LRU such as dentries and inodes, it is undocumented, and rather inflexible for certain potential list_lru users (such as the upcoming zswap shrinker and the THP shrinker). It has caused us a lot of issues during our development. This patch changes list_lru interface so that the caller must explicitly specify numa node and memcg when adding and removing objects. The old list_lru_add() and list_lru_del() are renamed to list_lru_add_obj() and list_lru_del_obj(), respectively. It also extends the list_lru API with a new function, list_lru_putback, which undoes a previous list_lru_isolate call. Unlike list_lru_add, it does not increment the LRU node count (as list_lru_isolate does not decrement the node count). list_lru_putback also allows for explicit memcg and NUMA node selection. Link: https://lkml.kernel.org/r/20231130194023.4102148-1-nphamcs@gmail.com Link: https://lkml.kernel.org/r/20231130194023.4102148-2-nphamcs@gmail.comSigned-off-by:
Nhat Pham <nphamcs@gmail.com> Suggested-by:
Johannes Weiner <hannes@cmpxchg.org> Acked-by:
Bagas Sanjaya <bagasdotme@gmail.com> Cc: Chris Li <chrisl@kernel.org> Cc: Dan Streetman <ddstreet@ieee.org> Cc: Domenico Cerasuolo <cerasuolodomenico@gmail.com> Cc: Michal Hocko <mhocko@kernel.org> Cc: Muchun Song <muchun.song@linux.dev> Cc: Roman Gushchin <roman.gushchin@linux.dev> Cc: Seth Jennings <sjenning@redhat.com> Cc: Shakeel Butt <shakeelb@google.com> Cc: Shuah Khan <shuah@kernel.org> Cc: Vitaly Wool <vitaly.wool@konsulko.com> Cc: Yosry Ahmed <yosryahmed@google.com> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org>
-
Peng Zhang authored
Now it seems that the incoming 'end' is already pointing to the last item, so we can simplify this function, considering only whether the last slot is being used. This has passed the maple tree test suite. Link: https://lkml.kernel.org/r/20231120070937.35481-6-zhangpeng.00@bytedance.comSigned-off-by:
Peng Zhang <zhangpeng.00@bytedance.com> Reviewed-by:
Liam R. Howlett <Liam.Howlett@oracle.com> Cc: Dan Carpenter <dan.carpenter@linaro.org> Signed-off-by:
-
Peng Zhang authored
There are two identical checks, delete one of them. Link: https://lkml.kernel.org/r/20231120070937.35481-5-zhangpeng.00@bytedance.comSigned-off-by:
-
Peng Zhang authored
The parameter maple_type is not used, so remove it. Link: https://lkml.kernel.org/r/20231120070937.35481-4-zhangpeng.00@bytedance.comSigned-off-by:
-
Peng Zhang authored
When the child node is the first child of its parent node, mas->min does not need to be updated. This can reduce the number of ascending times in some cases. Link: https://lkml.kernel.org/r/20231120070937.35481-3-zhangpeng.00@bytedance.comSigned-off-by:
-
Peng Zhang authored
Patch series "Some cleanups of maple tree", v2. These are some small cleanups of maple tree. This patch (of 5): Put the check for gap before its reference to avoid Smatch static check warnings. This is not a bug, it's just a validation program. Even with this change, Smatch may still generate warnings because MT_BUG_ON() doesn't necessarily stop the program. It may require fixing Smatch itself to avoid these warnings. Link: https://lkml.kernel.org/r/20231120070937.35481-1-zhangpeng.00@bytedance.com Link: https://lkml.kernel.org/r/20231120070937.35481-2-zhangpeng.00@bytedance.comSigned-off-by:
Dan Carpenter <dan.carpenter@linaro.org> Closes: http://lists.infradead.org/pipermail/maple-tree/2023-November/003046.htmlReviewed-by:
-
Jiapeng Chong authored
The function are defined in the maple_tree.c file, but not called elsewhere, so delete the unused function. lib/maple_tree.c:689:29: warning: unused function 'mas_pivot'. Link: https://lkml.kernel.org/r/20231027084944.24888-1-jiapeng.chong@linux.alibaba.comSigned-off-by:
Jiapeng Chong <jiapeng.chong@linux.alibaba.com> Reported-by:
Abaci Robot <abaci@linux.alibaba.com> Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=7064Acked-by:
David Hildenbrand <david@redhat.com> Reviewed-by:
-
Liam R. Howlett authored
mtree_range_walk() needed to be updated to avoid checking if there was a pivot value. On closer examination, the code could avoid setting min or max in certain scenarios. The commit removes the extra check for pivot[offset] before setting max and only sets max when necessary. It also only sets min if it is necessary by checking offset 0 prior to the loop (as it has always done). The commit also drops a dead node check since the end of the node will return the array size when the last slot is occupied (by a potential reuse in a dead node). The data will be discarded later if the node is marked dead. Benchmarking these changes results in an increase in performance of 5.45% using the BENCH_WALK in the maple tree test code. Link: https://lkml.kernel.org/r/20231101171629.3612299-13-Liam.Howlett@oracle.comSigned-off-by:
-
Liam R. Howlett authored
Since the pivot being set is now reliable, the optimized loop no longer needs to find the node end. The redundant check for a dead node can also be avoided as there is no danger of using the wrong pivot since the results will be thrown out in the case of a dead node by the later check. This patch also adds a benchmark test for the function to the maple tree test framework. The benchmark shows an average increase performance of 5.98% over 3 runs with this commit. Link: https://lkml.kernel.org/r/20231101171629.3612299-12-Liam.Howlett@oracle.comSigned-off-by:
-
Liam R. Howlett authored
ma_wr_state was previously tracking the end of the node for writing. Since the implementation of the ma_state end tracking, this is duplicated work. This patch removes the maple write state tracking of the end of the node and uses the maple state end instead. Link: https://lkml.kernel.org/r/20231101171629.3612299-11-Liam.Howlett@oracle.comSigned-off-by:
-
Liam R. Howlett authored
Now that the status of the maple state is outside of the node, the mas_searchable() function can be dropped for easier open-coding of what is going on. Link: https://lkml.kernel.org/r/20231101171629.3612299-10-Liam.Howlett@oracle.comSigned-off-by:
-
Liam R. Howlett authored
The maple tree node is overloaded to keep status as well as the active node. This, unfortunately, results in a re-walk on underflow or overflow. Since the maple state has room, the status can be placed in its own enum in the structure. Once an underflow/overflow is detected, certain modes can restore the status to active and others may need to re-walk just that one node to see the entry. The status being an enum has the benefit of detecting unhandled status in switch statements. [Liam.Howlett@oracle.com: fix comments about MAS_*] Link: https://lkml.kernel.org/r/20231106154124.614247-1-Liam.Howlett@oracle.com [Liam.Howlett@oracle.com: update forking to separate maple state and node] Link: https://lkml.kernel.org/r/20231106154551.615042-1-Liam.Howlett@oracle.com [Liam.Howlett@oracle.com: fix mas_prev() state separation code] Link: https://lkml.kernel.org/r/20231207193319.4025462-1-Liam.Howlett@oracle.com Link: https://lkml.kernel.org/r/20231101171629.3612299-9-Liam.Howlett@oracle.comSigned-off-by:
-
Liam R. Howlett authored
There are a few functions which were inlined but are somewhat too large to inline, so remove the inline key word. There are also several very small functions which are used in critical code sections which gcc was not inlining, so make this more strict and use __always_line for these functions. Link: https://lkml.kernel.org/r/20231101171629.3612299-8-Liam.Howlett@oracle.comSigned-off-by:
-
Liam R. Howlett authored
The node end is set during the walk, so use the resulting end instead of re-fetching it. Link: https://lkml.kernel.org/r/20231101171629.3612299-7-Liam.Howlett@oracle.comSigned-off-by:
-
Liam R. Howlett authored
When looking for the next entry, don't recalculate the node end as it is now tracked in the maple state. Link: https://lkml.kernel.org/r/20231101171629.3612299-6-Liam.Howlett@oracle.comSigned-off-by:
-
Liam R. Howlett authored
Analysis of the mas_for_each() iteration showed that there is a significant time spent finding the end of a node. This time can be greatly reduced if the end of the node is cached in the maple state. Care must be taken to update & invalidate as necessary. Link: https://lkml.kernel.org/r/20231101171629.3612299-5-Liam.Howlett@oracle.comSigned-off-by:
-
Liam R. Howlett authored
__mas_set_range() was created to shortcut resetting the maple state and a debug check was added to the caller (the vma iterator) to ensure the internal maple state remains safe to use. Move the debug check from the vma iterator into the maple tree itself so other users do not incorrectly use the advanced maple state modification. Fallout from this change include a large amount of debug setup needed to be moved to earlier in the header, and the maple_tree.h radix-tree test code needed to move the inclusion of the header to after the atomic define. None of those changes have functional changes. Link: https://lkml.kernel.org/r/20231101171629.3612299-4-Liam.Howlett@oracle.comSigned-off-by:
-
Liam R. Howlett authored
mas_erase() may not deal correctly with all maple states. Make the function more robust by ensuring the state is in one of the two acceptable states. Link: https://lkml.kernel.org/r/20231101171629.3612299-3-Liam.Howlett@oracle.comSigned-off-by:
-
Liam R. Howlett authored
Patch series "maple_tree: iterator state changes". These patches have some general cleanup and a change to separate the maple state status tracking from the maple state node. The maple state status change allows for walks to continue from previous places when the status needs to be recorded to make logical sense for the next call to the maple state. For instance, it allows for prev/next to function in a way that better resembles the linked list. It also allows switch statements to be used to detect missed states during compile, and the addition of fast-path "active" state is cleaner as an enum. While making the status change, perf showed some very small (one line) functions that were not inlined even with the inline key word. Making these small functions __always_inline is less expensive according to perf. As part of that change, some inlines have been dropped from larger functions. Perf also showed that the commonly used mas_for_each() iterator was spending a lot of time finding the end of the node. This series introduces caching of the end of the node in the maple state (and updating it during writes). This caching along with the inline changes yielded at 23.25% improvement on the BENCH_MAS_FOR_EACH maple tree test framework benchmark. I've also included a change to mtree_range_walk and mtree_lookup_walk to take advantage of Peng's change [1] to the initial pivot setup. mmtests did not produce any significant gains. [1] https://lore.kernel.org/all/20230711035444.526-1-zhangpeng.00@bytedance.com/T/#u This patch (of 12): Removing the default types from the switch statements will cause compile warnings on missing cases. Link: https://lkml.kernel.org/r/20231101171629.3612299-2-Liam.Howlett@oracle.comSigned-off-by:
-
- 11 Dec, 2023 21 commits
-
-
Juntong Deng authored
Record and report more information to help us find the cause of the bug and to help us correlate the error with other system events. This patch adds recording and showing CPU number and timestamp at allocation and free (controlled by CONFIG_KASAN_EXTRA_INFO). The timestamps in the report use the same format and source as printk. Error occurrence timestamp is already implicit in the printk log, and CPU number is already shown by dump_stack_lvl, so there is no need to add it. In order to record CPU number and timestamp at allocation and free, corresponding members need to be added to the relevant data structures, which will lead to increased memory consumption. In Generic KASAN, members are added to struct kasan_track. Since in most cases, alloc meta is stored in the redzone and free meta is stored in the object or the redzone, memory consumption will not increase much. In SW_TAGS KASAN and HW_TAGS KASAN, members are added to struct kasan_stack_ring_entry. Memory consumption increases as the size of struct kasan_stack_ring_entry increases (this part of the memory is allocated by memblock), but since this is configurable, it is up to the user to choose. Link: https://lkml.kernel.org/r/VI1P193MB0752BD991325D10E4AB1913599BDA@VI1P193MB0752.EURP193.PROD.OUTLOOK.COMSigned-off-by:
Juntong Deng <juntong.deng@outlook.com> Cc: Alexander Potapenko <glider@google.com> Cc: Andrey Konovalov <andreyknvl@gmail.com> Cc: Andrey Ryabinin <ryabinin.a.a@gmail.com> Cc: Dmitry Vyukov <dvyukov@google.com> Cc: Vincenzo Frascino <vincenzo.frascino@arm.com> Signed-off-by:
-
Sergey Senozhatsky authored
Use kmap_local_page() instead of kmap_atomic() which has been deprecated. Link: https://lkml.kernel.org/r/20231128083845.848008-1-senozhatsky@chromium.orgSigned-off-by:
Sergey Senozhatsky <senozhatsky@chromium.org> Acked-by:
Minchan Kim <minchan@kernel.org> Signed-off-by:
-
Dmitry Rokosov authored
To maintain the correct state, it is important to ensure that events for the memory cgroup v2 are aligned with the sample cgroup codes. Link: https://lkml.kernel.org/r/20231123071945.25811-4-ddrokosov@salutedevices.comSigned-off-by:
Dmitry Rokosov <ddrokosov@salutedevices.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Michal Hocko <mhocko@kernel.org> Cc: Muchun Song <muchun.song@linux.dev> Cc: Roman Gushchin <roman.gushchin@linux.dev> Cc: Shakeel Butt <shakeelb@google.com> Signed-off-by:
-
Dmitry Rokosov authored
This is a simple listener for memory events that handles counter changes in runtime. It can be set up for a specific memory cgroup v2. The output example: ===== $ /tmp/memcg_event_listener test Initialized MEMCG events with counters: MEMCG events: low: 0 high: 0 max: 0 oom: 0 oom_kill: 0 oom_group_kill: 0 Started monitoring memory events from '/sys/fs/cgroup/test/memory.events'... Received event in /sys/fs/cgroup/test/memory.events: *** 1 MEMCG oom_kill event, change counter 0 => 1 Received event in /sys/fs/cgroup/test/memory.events: *** 1 MEMCG oom_kill event, change counter 1 => 2 Received event in /sys/fs/cgroup/test/memory.events: *** 1 MEMCG oom_kill event, change counter 2 => 3 Received event in /sys/fs/cgroup/test/memory.events: *** 1 MEMCG oom_kill event, change counter 3 => 4 Received event in /sys/fs/cgroup/test/memory.events: *** 2 MEMCG max events, change counter 0 => 2 Received event in /sys/fs/cgroup/test/memory.events: *** 8 MEMCG max events, change counter 2 => 10 *** 1 MEMCG oom event, change counter 0 => 1 Received event in /sys/fs/cgroup/test/memory.events: *** 1 MEMCG oom_kill event, change counter 4 => 5 ^CExiting memcg event listener... ===== Link: https://lkml.kernel.org/r/20231123071945.25811-3-ddrokosov@salutedevices.comSigned-off-by:
-
Dmitry Rokosov authored
Patch series "samples: introduce cgroup events listeners", v3. To begin with, this patch series relocates the cgroup example code to the samples/cgroup directory, which is the appropriate location for such code snippets. Furthermore, a new memcg events listener is introduced. This listener is a simple yet effective tool for monitoring memory events and managing counter changes during runtime. Additionally, as per Andrew Morton's suggestion, a helpful reminder comment is included in the memcontrol implementation. This comment serves to ensure that the samples code is updated whenever new events are added. This patch (of 3): Move the cgroup_event_listener for cgroup v1 to the samples directory. This suggestion was proposed by Andrew Morton during the discussion [1]. Link: https://lore.kernel.org/all/20231106140934.3f5d4960141562fe8da53906@linux-foundation.org/ [1] Link: https://lkml.kernel.org/r/20231123071945.25811-1-ddrokosov@salutedevices.com Link: https://lkml.kernel.org/r/20231123071945.25811-2-ddrokosov@salutedevices.comSigned-off-by:
-
Muchun Song authored
There are still some places where it does not be converted to folio, this patch convert all of them to folio. And this patch also does some trival cleanup to fix the code style problems. Link: https://lkml.kernel.org/r/20231127084645.27017-5-songmuchun@bytedance.comSigned-off-by:
Muchun Song <songmuchun@bytedance.com> Reviewed-by:
Mike Kravetz <mike.kravetz@oracle.com> Cc: Kefeng Wang <wangkefeng.wang@huawei.com> Signed-off-by:
-
Muchun Song authored
To check a page whether it is self-hosted needs to traverse the page table (e.g. pmd_off_k()), however, we already have done this in the next calling of vmemmap_remap_range(). Moving PageVmemmapSelfHosted() check to vmemmap_pmd_entry() could simplify the code a bit. Link: https://lkml.kernel.org/r/20231127084645.27017-4-songmuchun@bytedance.comSigned-off-by:
-
Muchun Song authored
It is unnecessary to implement a series of dedicated page table walking helpers since there is already a general one walk_page_range_novma(). So use it to simplify the code. Link: https://lkml.kernel.org/r/20231127084645.27017-3-songmuchun@bytedance.comSigned-off-by:
-
Muchun Song authored
The 8782fb61 ("mm: pagewalk: Fix race between unmap and page walker") introduces an assertion to walk_page_range_novma() to make all the users of page table walker is safe. However, the race only exists for walking the user page tables. And it is ridiculous to hold a particular user mmap write lock against the changes of the kernel page tables. So only assert at least mmap read lock when walking the kernel page tables. And some users matching this case could downgrade to a mmap read lock to relief the contention of mmap lock of init_mm, it will be nicer in hugetlb (only holding mmap read lock) in the next patch. Link: https://lkml.kernel.org/r/20231127084645.27017-2-songmuchun@bytedance.comSigned-off-by:
-
Fabio M. De Francesco authored
kmap_atomic() has been deprecated in favor of kmap_local_page(). Therefore, replace kmap_atomic() with kmap_local_page() in swapfile.c. kmap_atomic() is implemented like a kmap_local_page() which also disables page-faults and preemption (the latter only in !PREEMPT_RT kernels). The kernel virtual addresses returned by these two API are only valid in the context of the callers (i.e., they cannot be handed to other threads). With kmap_local_page() the mappings are per thread and CPU local like in kmap_atomic(); however, they can handle page-faults and can be called from any context (including interrupts). The tasks that call kmap_local_page() can be preempted and, when they are scheduled to run again, the kernel virtual addresses are restored and are still valid. In mm/swapfile.c, the blocks of code between the mappings and un-mappings do not depend on the above-mentioned side effects of kmap_atomic(), so that the mere replacements of the old API with the new one is all that is required (i.e., there is no need to explicitly call pagefault_disable() and/or preempt_disable()). Link: https://lkml.kernel.org/r/20231127155452.586387-1-fabio.maria.de.francesco@linux.intel.comSigned-off-by:
Fabio M. De Francesco <fabio.maria.de.francesco@linux.intel.com> Cc: Ira Weiny <ira.weiny@intel.com> Signed-off-by:
-
Fabio M. De Francesco authored
kmap_atomic() has been deprecated in favor of kmap_local_page(). Therefore, replace kmap_atomic() with kmap_local_page() in zswap.c. kmap_atomic() is implemented like a kmap_local_page() which also disables page-faults and preemption (the latter only in !PREEMPT_RT kernels). The kernel virtual addresses returned by these two API are only valid in the context of the callers (i.e., they cannot be handed to other threads). With kmap_local_page() the mappings are per thread and CPU local like in kmap_atomic(); however, they can handle page-faults and can be called from any context (including interrupts). The tasks that call kmap_local_page() can be preempted and, when they are scheduled to run again, the kernel virtual addresses are restored and are still valid. In mm/zswap.c, the blocks of code between the mappings and un-mappings do not depend on the above-mentioned side effects of kmap_atomic(), so that the mere replacements of the old API with the new one is all that is required (i.e., there is no need to explicitly call pagefault_disable() and/or preempt_disable()). Link: https://lkml.kernel.org/r/20231127160058.586446-1-fabio.maria.de.francesco@linux.intel.comSigned-off-by:
-
Yong Wang authored
When the system is under oom, it prints out the RSS information of each process. However, we don't know the size of rss_anon, rss_file, and rss_shmem. To distinguish the memory occupied by anonymous or file mappings or shmem, could help us identify the root cause of the oom. So this patch adds RSS details, which refers to the /proc/<pid>/status[1]. It can help us know more about process memory usage. Example of oom including the new rss_* fields: [ 1630.902466] Tasks state (memory values in pages): [ 1630.902870] [ pid ] uid tgid total_vm rss rss_anon rss_file rss_shmem pgtables_bytes swapents oom_score_adj name [ 1630.903619] [ 149] 0 149 486 288 0 288 0 36864 0 0 ash [ 1630.904210] [ 156] 0 156 153531 153345 153345 0 0 1269760 0 0 mm_test [1] commit 8cee852e ("mm, procfs: breakdown RSS for anon, shmem and file in /proc/pid/status"). Link: https://lkml.kernel.org/r/202311231840181856667@zte.com.cnSigned-off-by:
Yong Wang <wang.yong12@zte.com.cn> Reviewed-by:
Yang Yang <yang.yang29@zte.com.cn> Cc: Hugh Dickins <hughd@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Xuexin Jiang <jiang.xuexin@zte.com.cn> Cc: Michal Hocko <mhocko@suse.com> Signed-off-by:
-
Andy Shevchenko authored
kernel-doc is not happy about documentation in list_lru.h: list_lru.h:90: warning: Function parameter or member 'lru' not described in 'list_lru_add' list_lru.h:90: warning: Excess function parameter 'list_lru' description in 'list_lru_add' list_lru.h:90: warning: No description found for return value of 'list_lru_add' list_lru.h:103: warning: Function parameter or member 'lru' not described in 'list_lru_del' list_lru.h:103: warning: Excess function parameter 'list_lru' description in 'list_lru_del' list_lru.h:103: warning: No description found for return value of 'list_lru_del' list_lru.h:116: warning: No description found for return value of 'list_lru_count_one' list_lru.h:168: warning: No description found for return value of 'list_lru_walk_one' list_lru.h:185: warning: No description found for return value of 'list_lru_walk_one_irq' Fix the documentation accordingly. While at it, fix the references to the parameters in functions inside the long descriptions, on which the above script is not complaining (yet?). Link: https://lkml.kernel.org/r/20231123172320.2434780-1-andriy.shevchenko@linux.intel.comSigned-off-by:
Andy Shevchenko <andriy.shevchenko@linux.intel.com> Cc: Rasmus Villemoes <linux@rasmusvillemoes.dk> Signed-off-by:
-
Peter Xu authored
This is a bug found not by any report but only by code observations. When GUP sees a devpmd/devpud and if page==NULL is returned, it means a fault is probably required. Here falling through when page==NULL can cause unexpected behavior. Fix both cases by catching the page==NULL cases with no_page_table(). Link: https://lkml.kernel.org/r/20231123180222.1048297-1-peterx@redhat.com Fixes: 3565fce3 ("mm, x86: get_user_pages() for dax mappings") Fixes: 080dbb61 ("mm/follow_page_mask: split follow_page_mask to smaller functions.") Signed-off-by:
Peter Xu <peterx@redhat.com> Acked-by:
-
Charan Teja Kalla authored
__alloc_pages_direct_reclaim() is called from slowpath allocation where high atomic reserves can be unreserved after there is a progress in reclaim and yet no suitable page is found. Later should_reclaim_retry() gets called from slow path allocation to decide if the reclaim needs to be retried before OOM kill path is taken. should_reclaim_retry() checks the available(reclaimable + free pages) memory against the min wmark levels of a zone and returns: a) true, if it is above the min wmark so that slow path allocation will do the reclaim retries. b) false, thus slowpath allocation takes oom kill path. should_reclaim_retry() can also unreserves the high atomic reserves **but only after all the reclaim retries are exhausted.** In a case where there are almost none reclaimable memory and free pages contains mostly the high atomic reserves but allocation context can't use these high atomic reserves, makes the available memory below min wmark levels hence false is returned from should_reclaim_retry() leading the allocation request to take OOM kill path. This can turn into a early oom kill if high atomic reserves are holding lot of free memory and unreserving of them is not attempted. (early)OOM is encountered on a VM with the below state: [ 295.998653] Normal free:7728kB boost:0kB min:804kB low:1004kB high:1204kB reserved_highatomic:8192KB active_anon:4kB inactive_anon:0kB active_file:24kB inactive_file:24kB unevictable:1220kB writepending:0kB present:70732kB managed:49224kB mlocked:0kB bounce:0kB free_pcp:688kB local_pcp:492kB free_cma:0kB [ 295.998656] lowmem_reserve[]: 0 32 [ 295.998659] Normal: 508*4kB (UMEH) 241*8kB (UMEH) 143*16kB (UMEH) 33*32kB (UH) 7*64kB (UH) 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 7752kB Per above log, the free memory of ~7MB exist in the high atomic reserves is not freed up before falling back to oom kill path. Fix it by trying to unreserve the high atomic reserves in should_reclaim_retry() before __alloc_pages_direct_reclaim() can fallback to oom kill path. Link: https://lkml.kernel.org/r/1700823445-27531-1-git-send-email-quic_charante@quicinc.com Fixes: 0aaa29a5 ("mm, page_alloc: reserve pageblocks for high-order atomic allocations on demand") Signed-off-by:
Charan Teja Kalla <quic_charante@quicinc.com> Reported-by:
Chris Goldsworthy <quic_cgoldswo@quicinc.com> Suggested-by:
Michal Hocko <mhocko@suse.com> Acked-by:
David Rientjes <rientjes@google.com> Cc: Chris Goldsworthy <quic_cgoldswo@quicinc.com> Cc: David Hildenbrand <david@redhat.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Mel Gorman <mgorman@techsingularity.net> Cc: Pavankumar Kondeti <quic_pkondeti@quicinc.com> Cc: Vlastimil Babka <vbabka@suse.cz> Signed-off-by:
-
Charan Teja Kalla authored
Highatomic reserves are set to roughly 1% of zone for maximum and a pageblock size for minimum. Encountered a system with the below configuration: Normal free:7728kB boost:0kB min:804kB low:1004kB high:1204kB reserved_highatomic:8192KB managed:49224kB On such systems, even a single pageblock makes highatomic reserves are set to ~8% of the zone memory. This high value can easily exert pressure on the zone. Per discussion with Michal and Mel, it is not much useful to reserve the memory for highatomic allocations on such small systems[1]. Since the minimum size for high atomic reserves is always going to be a pageblock size and if 1% of zone managed pages is going to be below pageblock size, don't reserve memory for high atomic allocations. Thanks Michal for this suggestion[2]. Since no memory is being reserved for high atomic allocations and if respective allocation failures are seen, this patch can be reverted. [1] https://lore.kernel.org/linux-mm/20231117161956.d3yjdxhhm4rhl7h2@techsingularity.net/ [2] https://lore.kernel.org/linux-mm/ZVYRJMUitykepLRy@tiehlicka/ Link: https://lkml.kernel.org/r/c3a2a48e2cfe08176a80eaf01c110deb9e918055.1700821416.git.quic_charante@quicinc.comSigned-off-by:
-
Charan Teja Kalla authored
Patch series "mm: page_alloc: fixes for high atomic reserve caluculations", v3. The state of the system where the issue exposed shown in oom kill logs: [ 295.998653] Normal free:7728kB boost:0kB min:804kB low:1004kB high:1204kB reserved_highatomic:8192KB active_anon:4kB inactive_anon:0kB active_file:24kB inactive_file:24kB unevictable:1220kB writepending:0kB present:70732kB managed:49224kB mlocked:0kB bounce:0kB free_pcp:688kBlocal_pcp:492kB free_cma:0kB [ 295.998656] lowmem_reserve[]: 0 32 [ 295.998659] Normal: 508*4kB (UMEH) 241*8kB (UMEH) 143*16kB (UMEH) 33*32kB (UH) 7*64kB (UH) 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 7752kB From the above, it is seen that ~16MB of memory reserved for high atomic reserves against the expectation of 1% reserves which is fixed in the 1st patch. Don't reserve the high atomic page blocks if 1% of zone memory size is below a pageblock size. This patch (of 2): reserve_highatomic_pageblock() aims to reserve the 1% of the managed pages of a zone, which is used for the high order atomic allocations. It uses the below calculation to reserve: static void reserve_highatomic_pageblock(struct page *page, ....) { ....... max_managed = (zone_managed_pages(zone) / 100) + pageblock_nr_pages; if (zone->nr_reserved_highatomic >= max_managed) goto out; zone->nr_reserved_highatomic += pageblock_nr_pages; set_pageblock_migratetype(page, MIGRATE_HIGHATOMIC); move_freepages_block(zone, page, MIGRATE_HIGHATOMIC, NULL); out: .... } Since we are always appending the 1% of zone managed pages count to pageblock_nr_pages, the minimum it is turning into 2 pageblocks as the nr_reserved_highatomic is incremented/decremented in pageblock sizes. Encountered a system(actually a VM running on the Linux kernel) with the below zone configuration: Normal free:7728kB boost:0kB min:804kB low:1004kB high:1204kB reserved_highatomic:8192KB managed:49224kB The existing calculations making it to reserve the 8MB(with pageblock size of 4MB) i.e. 16% of the zone managed memory. Reserving such high amount of memory can easily exert memory pressure in the system thus may lead into unnecessary reclaims till unreserving of high atomic reserves. Since high atomic reserves are managed in pageblock size granules, as MIGRATE_HIGHATOMIC is set for such pageblock, fix the calculations for high atomic reserves as, minimum is pageblock size , maximum is approximately 1% of the zone managed pages. Link: https://lkml.kernel.org/r/cover.1700821416.git.quic_charante@quicinc.com Link: https://lkml.kernel.org/r/1660034138397b82a0a8b6ae51cbe96bd583d89e.1700821416.git.quic_charante@quicinc.comSigned-off-by:Mel Gorman <mgorman@techsingularity.net> Acked-by:
-
Peter Xu authored
Hugetlb vmemmap default option (HUGETLB_PAGE_OPTIMIZE_VMEMMAP_DEFAULT_ON) is a sub-option to hugetlbfs, but it shows in the same level as hugetlbfs itself, under "Pesudo filesystems". Make the vmemmap option a sub-option to hugetlbfs, by changing hugetlbfs into a menuconfig. When moving it, fix a typo 'v' spot by Randy. Link: https://lkml.kernel.org/r/20231124151902.1075697-1-peterx@redhat.comSigned-off-by:
-
Serge Semin authored
Based on the init_unavailable_range() method and it's callee semantics no multi-line info messages are intended to be printed to the console. Thus append the '\n' symbol to the respective info string. Link: https://lkml.kernel.org/r/20231122182419.30633-7-fancer.lancer@gmail.comSigned-off-by:
Serge Semin <fancer.lancer@gmail.com> Reviewed-by:
Mike Rapoport (IBM) <rppt@kernel.org> Signed-off-by:
-
Serge Semin authored
Besides of the already described reasons the pages backended memory holes might be persistent due to having memory mapped IO spaces behind those ranges in the framework of flatmem kernel config. Add such note to the init_unavailable_range() method kdoc in order to point out to one more reason of having the function executed for such regions. [fancer.lancer@gmail.com: update per Mike] Link: https://lkml.kernel.org/r/20231202111855.18392-1-fancer.lancer@gmail.com Link: https://lkml.kernel.org/r/20231122182419.30633-6-fancer.lancer@gmail.comSigned-off-by:
-
Alexander Gordeev authored
Rename ptdesc _refcount field to __page_refcount similar to the other unused page fields. Link: https://lkml.kernel.org/r/982bdc652ba79a606c3d01c905766e7e076b3315.1700594815.git.agordeev@linux.ibm.comSigned-off-by:
Alexander Gordeev <agordeev@linux.ibm.com> Suggested-by:
Vishal Moola <vishal.moola@gmail.com> Cc: Gerald Schaefer <gerald.schaefer@linux.ibm.com> Cc: Heiko Carstens <hca@linux.ibm.com> Signed-off-by:
-
