- 29 Dec, 2023 4 commits
-
-
Matthew Wilcox (Oracle) authored
While sector_t is always defined as a u64 today, that hasn't always been the case and it might not always be the same size as loff_t in the future. Link: https://lkml.kernel.org/r/20231109210608.2252323-5-willy@infradead.orgSigned-off-by:
Matthew Wilcox (Oracle) <willy@infradead.org> Cc: Hannes Reinecke <hare@suse.de> Cc: Luis Chamberlain <mcgrof@kernel.org> Cc: Pankaj Raghav <p.raghav@samsung.com> Cc: Ryusuke Konishi <konishi.ryusuke@gmail.com> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org>
-
Matthew Wilcox (Oracle) authored
We must not shift by a negative number so work in terms of a byte offset to avoid the awkward shift left-or-right-depending-on-sign option. This means we need to use check_mul_overflow() to ensure that a large block number does not result in a wrap. Link: https://lkml.kernel.org/r/20231109210608.2252323-4-willy@infradead.orgSigned-off-by:
Nathan Chancellor <nathan@kernel.org> Cc: Hannes Reinecke <hare@suse.de> Cc: Luis Chamberlain <mcgrof@kernel.org> Cc: Pankaj Raghav <p.raghav@samsung.com> Cc: Ryusuke Konishi <konishi.ryusuke@gmail.com> [nathan@kernel.org: add cast in grow_buffers() to avoid a multiplication libcall] Link: https://lkml.kernel.org/r/20231128-avoid-muloti4-grow_buffers-v1-1-bc3d0f0ec483@kernel.orgSigned-off-by:
-
Matthew Wilcox (Oracle) authored
The calculation of block from index doesn't work for devices with a block size larger than PAGE_SIZE as we end up shifting by a negative number. Instead, calculate the number of the first block from the folio's position in the block device. We no longer need to pass sizebits to grow_dev_folio(). Link: https://lkml.kernel.org/r/20231109210608.2252323-3-willy@infradead.orgSigned-off-by:
Pankaj Raghav <p.raghav@samsung.com> Cc: Hannes Reinecke <hare@suse.de> Cc: Luis Chamberlain <mcgrof@kernel.org> Cc: Ryusuke Konishi <konishi.ryusuke@gmail.com> Signed-off-by:
-
Matthew Wilcox (Oracle) authored
Patch series "More buffer_head cleanups", v2. The first patch is a left-over from last cycle. The rest fix "obvious" block size > PAGE_SIZE problems. I haven't tested with a large block size setup (but I have done an ext4 xfstests run). This patch (of 7): Rename grow_dev_page() to grow_dev_folio() and make it return a bool. Document what that bool means; it's more subtle than it first appears. Also rename the 'failed' label to 'unlock' beacuse it's not exactly 'failed'. It just hasn't succeeded. Link: https://lkml.kernel.org/r/20231109210608.2252323-2-willy@infradead.orgSigned-off-by:
-
- 20 Dec, 2023 36 commits
-
-
Borislav Petkov (AMD) authored
Use the same splat markers as panic does for easier matching by external tools scanning kernel dmesg for splats. Link: https://lkml.kernel.org/r/20231218135339.23209-1-bp@alien8.deSigned-off-by:
Borislav Petkov (AMD) <bp@alien8.de> Cc: Andrey Ryabinin <ryabinin.a.a@gmail.com> Signed-off-by:
-
Yajun Deng authored
There is redundant code in __free_pages_ok(). Use free_one_page() simplify it. Link: https://lkml.kernel.org/r/20231216030503.2126130-1-yajun.deng@linux.devSigned-off-by:
Yajun Deng <yajun.deng@linux.dev> Reviewed-by:
-
Peng Zhang authored
The last range stored in maple tree is typically quite large. By checking if it exceeds the sum of the remaining ranges in that node, it is possible to avoid checking all other gaps. Running the maple tree test suite in user mode almost always results in a near 100% hit rate for this optimization. Link: https://lkml.kernel.org/r/20231215074632.82045-1-zhangpeng.00@bytedance.comSigned-off-by:
Peng Zhang <zhangpeng.00@bytedance.com> Reviewed-by:
Liam R. Howlett <Liam.Howlett@oracle.com> Signed-off-by:
-
Fabio M. De Francesco authored
kmap() has been deprecated in favor of kmap_local_page(). Therefore, replace kmap() with kmap_local_page() in mm/memory.c. There are two main problems with kmap(): (1) It comes with an overhead as the mapping space is restricted and protected by a global lock for synchronization and (2) it also requires global TLB invalidation when the kmap's pool wraps and it might block when the mapping space is fully utilized until a slot becomes available. With kmap_local_page() the mappings are per thread, CPU local, can take page-faults, and can be called from any context (including interrupts). It is faster than kmap() in kernels with HIGHMEM enabled. The tasks can be preempted and, when they are scheduled to run again, the kernel virtual addresses are restored and still valid. Obviously, thread locality implies that the kernel virtual addresses returned by kmap_local_page() are only valid in the context of the callers (i.e., they cannot be handed to other threads). The use of kmap_local_page() in mm/memory.c does not break the above-mentioned assumption, so it is allowed and preferred. Link: https://lkml.kernel.org/r/20231215084417.2002370-1-fabio.maria.de.francesco@linux.intel.com Link: https://lkml.kernel.org/r/20231214081039.1919328-1-fabio.maria.de.francesco@linux.intel.comSigned-off-by:
Fabio M. De Francesco <fabio.maria.de.francesco@linux.intel.com> Reviewed-by:
Ira Weiny <ira.weiny@intel.com> Signed-off-by:
-
Randy Dunlap authored
Correct typos/spellos and punctutation. Link: https://lkml.kernel.org/r/20231213043316.10128-1-rdunlap@infradead.orgSigned-off-by:
Randy Dunlap <rdunlap@infradead.org> Signed-off-by:
-
SeongJae Park authored
There are eight command inputs for 'state' DAMON sysfs file, and those are verbosely explained in multiple paragraphs. It is not easy to find explanation of specific command, and getting whole picture of supported commands. Replace the paragraphs with a list. Link: https://lkml.kernel.org/r/20231213190338.54146-7-sj@kernel.orgSigned-off-by:
SeongJae Park <sj@kernel.org> Cc: Jonathan Corbet <corbet@lwn.net> Signed-off-by:
-
SeongJae Park authored
'Sysfs Files Hierarchy' section of DAMON usage document shows whole picture of the interface. Then sections for detailed explanation of the files follow. Due to the amount of the files, navigating between the whole picture and the section for specific files sometimes require no subtle amount of scrolling. Add links from the whole picture to the dedicated sections for making the navigation easier. Link: https://lkml.kernel.org/r/20231213190338.54146-6-sj@kernel.orgSigned-off-by:
-
SeongJae Park authored
The label for context DAMON sysfs directory section is having name sysfs_contexts. The name would be better to be used for the contexts directory. Rename it to represent a single context. Link: https://lkml.kernel.org/r/20231213190338.54146-5-sj@kernel.orgSigned-off-by:
-
SeongJae Park authored
The execution model and data structures section at the end of the design document is briefly explaining how DAMON works overall. Knowing that first may help better drawing the overall picture. It may also help better understanding following detailed sections. Move it to the beginning of the document. Link: https://lkml.kernel.org/r/20231213190338.54146-4-sj@kernel.orgSigned-off-by:
-
SeongJae Park authored
Commit 35f5d941 ("mm/damon: implement a function for max nr_accesses safe calculation") has fixed an overflow bug that could cause divide-by-zero. Add a kunit test for the bug to ensure similar bugs are not introduced again. Link: https://lkml.kernel.org/r/20231213190338.54146-3-sj@kernel.orgSigned-off-by:
-
SeongJae Park authored
Patch series "mm/damon: misc updates for 6.8". Update comments, tests, and documents for DAMON. This patch (of 6): SeongJae is using his kernel.org account for DAMON development. Update the old email addresses on the comments of DAMON source files. Link: https://lkml.kernel.org/r/20231213190338.54146-1-sj@kernel.org Link: https://lkml.kernel.org/r/20231213190338.54146-2-sj@kernel.orgSigned-off-by:
-
Kevin Hao authored
A freezable kernel thread can enter frozen state during freezing by either calling try_to_freeze() or using wait_event_freezable() and its variants. However, there is no need to use both methods simultaneously. Link: https://lkml.kernel.org/r/20231213090906.1070985-1-haokexin@gmail.comSigned-off-by:
Kevin Hao <haokexin@gmail.com> Acked-by:
David Hildenbrand <david@redhat.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: "Rafael J. Wysocki" <rafael@kernel.org> Cc: Pavel Machek <pavel@ucw.cz> Signed-off-by:
-
SeongJae Park authored
Add a test for reproducing the update_schemes_tried_{regions,bytes} command-causing indefinite hang bug that fixed by commit 7d6fa31a ("mm/damon/sysfs-schemes: add timeout for update_schemes_tried_regions"), to avoid mistakenly re-introducing the bug. Refer to the fix commit for more details of the bug. Link: https://lkml.kernel.org/r/20231212194810.54457-6-sj@kernel.orgSigned-off-by: -
SeongJae Park authored
Add a selftest for verifying the accuracy of DAMON's access monitoring functionality. The test starts a program of artificial access pattern, monitor the access pattern using DAMON, and check if DAMON finds expected amount of hot data region (working set size) with only acceptable error rate. Note that the acceptable error rate is set with only naive assumptions and small number of tests. Hence failures of the test may not always mean DAMON is broken. Rather than that, those could be a signal to better understand the real accuracy level of DAMON in wider environments. Based on further finding, we could optimize DAMON or adjust the expectation of the test. Link: https://lkml.kernel.org/r/20231212194810.54457-5-sj@kernel.orgSigned-off-by:
-
SeongJae Park authored
Implement update_schemes_tried_bytes command of DAMON sysfs interface in _damon_sysfs.py. It is not only making the update, but also read the updated value from the sysfs interface and store it in the Kdamond python objects so that the user of the module can easily get the value. Link: https://lkml.kernel.org/r/20231212194810.54457-4-sj@kernel.orgSigned-off-by:
-
SeongJae Park authored
Extend the tests-writing-purpose DAMON sysfs control module to support the kdamonds start functionality. Link: https://lkml.kernel.org/r/20231212194810.54457-3-sj@kernel.orgSigned-off-by:
-
SeongJae Park authored
Patch series "selftests/damon: add Python-written DAMON functionality tests", v2. DAMON exports most of its functionality via its sysfs interface. Hence most DAMON functionality tests could be implemented using the interface. However, because the interfaces require simple but multiple operations for many controls, writing all such tests from the scratch could be repetitive and time consuming. Implement a minimum DAMON sysfs control module, and a couple of DAMON functionality tests using the control module. The first test is for ensuring minimum accuracy of data access monitoring, and the second test is for finding if a previously found and fixed bug is introduced again. Note that the DAMON sysfs control module is only for avoiding duplicating code in tests. For convenient and general control of DAMON, users should use DAMON user-space tools that developed for the purpose, such as damo[1]. [1] https://github.com/damonitor/damo Patches Sequence ---------------- This patchset is constructed with five patches. The first three patches implement a Python-written test implementation-purpose DAMON sysfs control module. The implementation is incrementally done in the sequence of the basic data structure (first patch) first, kdamonds start command (second patch) next, and finally DAMOS tried bytes update command (third patch). Then two patches for implementing selftests using the module follows. The fourth patch implements a basic functionality test of DAMON for working set estimation accuracy. Finally, the fifth patch implements a corner case test for a previously found bug. This patch (of 5): Implement a python module for DAMON sysfs controls. The module is aimed to be useful for writing DAMON functionality tests in future. Nonetheless, this module is only representing a subset of DAMON sysfs files. Following commits will implement more DAMON sysfs controls. Link: https://lkml.kernel.org/r/20231212194810.54457-1-sj@kernel.org Link: https://lkml.kernel.org/r/20231212194810.54457-2-sj@kernel.orgSigned-off-by:
-
Randy Dunlap authored
Fix typos/grammar and spellos in documentation. Link: https://lkml.kernel.org/r/20231210063839.29967-1-rdunlap@infradead.orgSigned-off-by:
-
Jiapeng Chong authored
Avoid pointer type value compared with 0 to make code clear. ./tools/testing/radix-tree/maple.c:34142:15-16: WARNING comparing pointer to 0. Link: https://lkml.kernel.org/r/20231208020450.7003-1-jiapeng.chong@linux.alibaba.comReported-by:
Abaci Robot <abaci@linux.alibaba.com> Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=7696Signed-off-by:
Jiapeng Chong <jiapeng.chong@linux.alibaba.com> Cc: Liam R. Howlett <Liam.Howlett@oracle.com> Cc: Matthew Wilcox (Oracle) <willy@infradead.org> Signed-off-by:
-
Ryan Roberts authored
Add tests similar to the existing PMD-sized THP tests, but which operate on memory backed by (PTE-mapped) multi-size THP. This reuses all the existing infrastructure. If the test suite detects that multi-size THP is not supported by the kernel, the new tests are skipped. Link: https://lkml.kernel.org/r/20231207161211.2374093-11-ryan.roberts@arm.comSigned-off-by:
Ryan Roberts <ryan.roberts@arm.com> Reviewed-by:
Kefeng Wang <wangkefeng.wang@huawei.com> Tested-by:
John Hubbard <jhubbard@nvidia.com> Cc: Alistair Popple <apopple@nvidia.com> Cc: Anshuman Khandual <anshuman.khandual@arm.com> Cc: Barry Song <v-songbaohua@oppo.com> Cc: Catalin Marinas <catalin.marinas@arm.com> Cc: David Rientjes <rientjes@google.com> Cc: "Huang, Ying" <ying.huang@intel.com> Cc: Hugh Dickins <hughd@google.com> Cc: Itaru Kitayama <itaru.kitayama@gmail.com> Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Cc: Luis Chamberlain <mcgrof@kernel.org> Cc: Matthew Wilcox (Oracle) <willy@infradead.org> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: Yang Shi <shy828301@gmail.com> Cc: Yin Fengwei <fengwei.yin@intel.com> Cc: Yu Zhao <yuzhao@google.com> Cc: Zi Yan <ziy@nvidia.com> Signed-off-by:
-
Ryan Roberts authored
do_run_with_thp() prepares (PMD-sized) THP memory into different states before running tests. With the introduction of multi-size THP, we would like to reuse this logic to also test those smaller THP sizes. So let's add a thpsize parameter which tells the function what size THP it should operate on. A separate commit will utilize this change to add new tests for multi-size THP, where available. Link: https://lkml.kernel.org/r/20231207161211.2374093-10-ryan.roberts@arm.comSigned-off-by:
-
Ryan Roberts authored
The `collapse_max_ptes_none` test was previously failing when a THP size less than PMD-size had enabled="always". The root cause is because the test faults in 1 page less than the threshold it set for collapsing. But when THP is enabled always, we "over allocate" and therefore the threshold is passed, and collapse unexpectedly succeeds. Solve this by enlightening khugepaged selftest. Add a command line option to pass in the desired THP size that should be used for all anonymous allocations. The harness will then explicitly configure a THP size as requested and modify the `collapse_max_ptes_none` test so that it faults in the threshold minus the number of pages in the configured THP size. If no command line option is provided, default to order 0, as per previous behaviour. I chose to use an order in the command line interface, since this makes the interface agnostic of base page size, making it easier to invoke from run_vmtests.sh. Link: https://lkml.kernel.org/r/20231207161211.2374093-9-ryan.roberts@arm.comSigned-off-by:
-
Ryan Roberts authored
Save and restore the new per-size hugepage enabled setting, if available on the running kernel. Since the number of per-size directories is not fixed, solve this as simply as possible by catering for a maximum number in the thp_settings struct (20). Each array index is the order. The value of THP_NEVER is changed to 0 so that all of these new settings default to THP_NEVER and the user only needs to fill in the ones they want to enable. Link: https://lkml.kernel.org/r/20231207161211.2374093-8-ryan.roberts@arm.comSigned-off-by:
-
Ryan Roberts authored
The khugepaged test has a useful framework for save/restore/pop/push of all thp settings via the sysfs interface. This will be useful to explicitly control multi-size THP settings in other tests, so let's move it out of khugepaged and into its own thp_settings.[c|h] utility. Link: https://lkml.kernel.org/r/20231207161211.2374093-7-ryan.roberts@arm.comSigned-off-by:
Alistair Popple <apopple@nvidia.com> Acked-by:
-
Ryan Roberts authored
Previously, the saved thp settings would be restored upon a signal or at the natural end of the test suite. But there are some tests that directly call exit() upon failure. In this case, the thp settings were not being restored, which could then influence other tests. Fix this by installing an atexit() handler to do the actual restore. The signal handler can now just call exit() and the atexit handler is invoked. Link: https://lkml.kernel.org/r/20231207161211.2374093-6-ryan.roberts@arm.comSigned-off-by:
-
Ryan Roberts authored
Introduce the logic to allow THP to be configured (through the new sysfs interface we just added) to allocate large folios to back anonymous memory, which are larger than the base page size but smaller than PMD-size. We call this new THP extension "multi-size THP" (mTHP). mTHP continues to be PTE-mapped, but in many cases can still provide similar benefits to traditional PMD-sized THP: Page faults are significantly reduced (by a factor of e.g. 4, 8, 16, etc. depending on the configured order), but latency spikes are much less prominent because the size of each page isn't as huge as the PMD-sized variant and there is less memory to clear in each page fault. The number of per-page operations (e.g. ref counting, rmap management, lru list management) are also significantly reduced since those ops now become per-folio. Some architectures also employ TLB compression mechanisms to squeeze more entries in when a set of PTEs are virtually and physically contiguous and approporiately aligned. In this case, TLB misses will occur less often. The new behaviour is disabled by default, but can be enabled at runtime by writing to /sys/kernel/mm/transparent_hugepage/hugepage-XXkb/enabled (see documentation in previous commit). The long term aim is to change the default to include suitable lower orders, but there are some risks around internal fragmentation that need to be better understood first. [ryan.roberts@arm.com: resolve some multi-size THP review nits] Link: https://lkml.kernel.org/r/20231214160251.3574571-1-ryan.roberts@arm.com Link: https://lkml.kernel.org/r/20231207161211.2374093-5-ryan.roberts@arm.comSigned-off-by:
-
Ryan Roberts authored
In preparation for adding support for anonymous multi-size THP, introduce new sysfs structure that will be used to control the new behaviours. A new directory is added under transparent_hugepage for each supported THP size, and contains an `enabled` file, which can be set to "inherit" (to inherit the global setting), "always", "madvise" or "never". For now, the kernel still only supports PMD-sized anonymous THP, so only 1 directory is populated. The first half of the change converts transhuge_vma_suitable() and hugepage_vma_check() so that they take a bitfield of orders for which the user wants to determine support, and the functions filter out all the orders that can't be supported, given the current sysfs configuration and the VMA dimensions. The resulting functions are renamed to thp_vma_suitable_orders() and thp_vma_allowable_orders() respectively. Convenience functions that take a single, unencoded order and return a boolean are also defined as thp_vma_suitable_order() and thp_vma_allowable_order(). The second half of the change implements the new sysfs interface. It has been done so that each supported THP size has a `struct thpsize`, which describes the relevant metadata and is itself a kobject. This is pretty minimal for now, but should make it easy to add new per-thpsize files to the interface if needed in future (e.g. per-size defrag). Rather than keep the `enabled` state directly in the struct thpsize, I've elected to directly encode it into huge_anon_orders_[always|madvise|inherit] bitfields since this reduces the amount of work required in thp_vma_allowable_orders() which is called for every page fault. See Documentation/admin-guide/mm/transhuge.rst, as modified by this commit, for details of how the new sysfs interface works. [ryan.roberts@arm.com: fix build warning when CONFIG_SYSFS is disabled] Link: https://lkml.kernel.org/r/20231211125320.3997543-1-ryan.roberts@arm.com Link: https://lkml.kernel.org/r/20231207161211.2374093-4-ryan.roberts@arm.comSigned-off-by:
Barry Song <v-songbaohua@oppo.com> Tested-by:
-
Ryan Roberts authored
In preparation for supporting anonymous multi-size THP, improve folio_add_new_anon_rmap() to allow a non-pmd-mappable, large folio to be passed to it. In this case, all contained pages are accounted using the order-0 folio (or base page) scheme. Link: https://lkml.kernel.org/r/20231207161211.2374093-3-ryan.roberts@arm.comSigned-off-by:
Yu Zhao <yuzhao@google.com> Reviewed-by:
Yin Fengwei <fengwei.yin@intel.com> Reviewed-by:
-
Ryan Roberts authored
Patch series "Multi-size THP for anonymous memory", v9. A series to implement multi-size THP (mTHP) for anonymous memory (previously called "small-sized THP" and "large anonymous folios"). The objective of this is to improve performance by allocating larger chunks of memory during anonymous page faults: 1) Since SW (the kernel) is dealing with larger chunks of memory than base pages, there are efficiency savings to be had; fewer page faults, batched PTE and RMAP manipulation, reduced lru list, etc. In short, we reduce kernel overhead. This should benefit all architectures. 2) Since we are now mapping physically contiguous chunks of memory, we can take advantage of HW TLB compression techniques. A reduction in TLB pressure speeds up kernel and user space. arm64 systems have 2 mechanisms to coalesce TLB entries; "the contiguous bit" (architectural) and HPA (uarch). This version incorporates David's feedback on the core patches (#3, #4) and adds some RB and TB tags (see change log for details). By default, the existing behaviour (and performance) is maintained. The user must explicitly enable multi-size THP to see the performance benefit. This is done via a new sysfs interface (as recommended by David Hildenbrand - thanks to David for the suggestion)! This interface is inspired by the existing per-hugepage-size sysfs interface used by hugetlb, provides full backwards compatibility with the existing PMD-size THP interface, and provides a base for future extensibility. See [9] for detailed discussion of the interface. This series is based on mm-unstable (715b67adf4c8). Prerequisites ============= I'm removing this section on the basis that I don't believe what we were previously calling prerequisites are really prerequisites anymore. We originally defined them when mTHP was a compile-time feature. There is now a runtime control to opt-in to mTHP; when disabled, correctness and performance are as before. When enabled, the code is still correct/robust, but in the absence of the one remaining item (compaction) there may be a performance impact in some corners. See the old list in the v8 cover letter at [8]. And a longer explanation of my thinking here [10]. SUMMARY: I don't think we should hold this series up, waiting for the items on the prerequisites list. I believe this series should be ready now so hopefully can be added to mm-unstable for some testing, then fingers crossed for v6.8. Testing ======= The series includes patches for mm selftests to enlighten the cow and khugepaged tests to explicitly test with multi-size THP, in the same way that PMD-sized THP is tested. The new tests all pass, and no regressions are observed in the mm selftest suite. I've also run my usual kernel compilation and java script benchmarks without any issues. Refer to my performance numbers posted with v6 [6]. (These are for multi-size THP only - they do not include the arm64 contpte follow-on series). John Hubbard at Nvidia has indicated dramatic 10x performance improvements for some workloads at [11]. (Observed using v6 of this series as well as the arm64 contpte series). Kefeng Wang at Huawei has also indicated he sees improvements at [12] although there are some latency regressions also. I've also checked that there is no regression in the write fault path when mTHP is disabled using a microbenchmark. I ran it for a baseline kernel, as well as v8 and v9. I repeated on Ampere Altra (bare metal) and Apple M2 (VM): | | m2 vm | altra | |--------------|---------------------|---------------------| | kernel | mean | std_rel | mean | std_rel | |--------------|----------|----------|----------|----------| | baseline | 0.000% | 0.341% | 0.000% | 3.581% | | anonfolio-v8 | 0.005% | 0.272% | 5.068% | 1.128% | | anonfolio-v9 | -0.013% | 0.442% | 0.107% | 1.788% | There is no measurable difference on M2, but altra has a slow down in v8 which is fixed in v9 by moving the THP order check to be inline within thp_vma_allowable_orders(), as suggested by David. This patch (of 10): In preparation for the introduction of anonymous multi-size THP, we would like to be able to split them when they have unmapped subpages, in order to free those unused pages under memory pressure. So remove the artificial requirement that the large folio needed to be at least PMD-sized. Link: https://lkml.kernel.org/r/20231207161211.2374093-1-ryan.roberts@arm.com Link: https://lkml.kernel.org/r/20231207161211.2374093-2-ryan.roberts@arm.comSigned-off-by:
-
Yosry Ahmed authored
Stats flushing for memcg currently follows the following rules: - Always flush the entire memcg hierarchy (i.e. flush the root). - Only one flusher is allowed at a time. If someone else tries to flush concurrently, they skip and return immediately. - A periodic flusher flushes all the stats every 2 seconds. The reason this approach is followed is because all flushes are serialized by a global rstat spinlock. On the memcg side, flushing is invoked from userspace reads as well as in-kernel flushers (e.g. reclaim, refault, etc). This approach aims to avoid serializing all flushers on the global lock, which can cause a significant performance hit under high concurrency. This approach has the following problems: - Occasionally a userspace read of the stats of a non-root cgroup will be too expensive as it has to flush the entire hierarchy [1]. - Sometimes the stats accuracy are compromised if there is an ongoing flush, and we skip and return before the subtree of interest is actually flushed, yielding stale stats (by up to 2s due to periodic flushing). This is more visible when reading stats from userspace, but can also affect in-kernel flushers. The latter problem is particulary a concern when userspace reads stats after an event occurs, but gets stats from before the event. Examples: - When memory usage / pressure spikes, a userspace OOM handler may look at the stats of different memcgs to select a victim based on various heuristics (e.g. how much private memory will be freed by killing this). Reading stale stats from before the usage spike in this case may cause a wrongful OOM kill. - A proactive reclaimer may read the stats after writing to memory.reclaim to measure the success of the reclaim operation. Stale stats from before reclaim may give a false negative. - Reading the stats of a parent and a child memcg may be inconsistent (child larger than parent), if the flush doesn't happen when the parent is read, but happens when the child is read. As for in-kernel flushers, they will occasionally get stale stats. No regressions are currently known from this, but if there are regressions, they would be very difficult to debug and link to the source of the problem. This patch aims to fix these problems by restoring subtree flushing, and removing the unified/coalesced flushing logic that skips flushing if there is an ongoing flush. This change would introduce a significant regression with global stats flushing thresholds. With per-memcg stats flushing thresholds, this seems to perform really well. The thresholds protect the underlying lock from unnecessary contention. This patch was tested in two ways to ensure the latency of flushing is up to par, on a machine with 384 cpus: - A synthetic test with 5000 concurrent workers in 500 cgroups doing allocations and reclaim, as well as 1000 readers for memory.stat (variation of [2]). No regressions were noticed in the total runtime. Note that significant regressions in this test are observed with global stats thresholds, but not with per-memcg thresholds. - A synthetic stress test for concurrently reading memcg stats while memory allocation/freeing workers are running in the background, provided by Wei Xu [3]. With 250k threads reading the stats every 100ms in 50k cgroups, 99.9% of reads take <= 50us. Less than 0.01% of reads take more than 1ms, and no reads take more than 100ms. [1] https://lore.kernel.org/lkml/CABWYdi0c6__rh-K7dcM_pkf9BJdTRtAU08M43KO9ME4-dsgfoQ@mail.gmail.com/ [2] https://lore.kernel.org/lkml/CAJD7tka13M-zVZTyQJYL1iUAYvuQ1fcHbCjcOBZcz6POYTV-4g@mail.gmail.com/ [3] https://lore.kernel.org/lkml/CAAPL-u9D2b=iF5Lf_cRnKxUfkiEe0AMDTu6yhrUAzX0b6a6rDg@mail.gmail.com/ [akpm@linux-foundation.org: fix mm/zswap.c] [yosryahmed@google.com: remove stats flushing mutex] Link: https://lkml.kernel.org/r/CAJD7tkZgP3m-VVPn+fF_YuvXeQYK=tZZjJHj=dzD=CcSSpp2qg@mail.gmail.com Link: https://lkml.kernel.org/r/20231129032154.3710765-6-yosryahmed@google.comSigned-off-by:
Yosry Ahmed <yosryahmed@google.com> Tested-by:
Domenico Cerasuolo <cerasuolodomenico@gmail.com> Acked-by:
Shakeel Butt <shakeelb@google.com> Cc: Chris Li <chrisl@kernel.org> Cc: Greg Thelen <gthelen@google.com> Cc: Ivan Babrou <ivan@cloudflare.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Michal Hocko <mhocko@kernel.org> Cc: Michal Koutny <mkoutny@suse.com> Cc: Muchun Song <muchun.song@linux.dev> Cc: Roman Gushchin <roman.gushchin@linux.dev> Cc: Tejun Heo <tj@kernel.org> Cc: Waiman Long <longman@redhat.com> Cc: Wei Xu <weixugc@google.com> Signed-off-by:
-
Yosry Ahmed authored
The workingset code flushes the stats in workingset_refault() to get accurate stats of the eviction memcg. In preparation for more scoped flushed and passing the eviction memcg to the flush call, move the call to workingset_test_recent() where we have a pointer to the eviction memcg. The flush call is sleepable, and cannot be made in an rcu read section. Hence, minimize the rcu read section by also moving it into workingset_test_recent(). Furthermore, instead of holding the rcu read lock throughout workingset_test_recent(), only hold it briefly to get a ref on the eviction memcg. This allows us to make the flush call after we get the eviction memcg. As for workingset_refault(), nothing else there appears to be protected by rcu. The memcg of the faulted folio (which is not necessarily the same as the eviction memcg) is protected by the folio lock, which is held from all callsites. Add a VM_BUG_ON() to make sure this doesn't change from under us. No functional change intended. Link: https://lkml.kernel.org/r/20231129032154.3710765-5-yosryahmed@google.comSigned-off-by:
-
Yosry Ahmed authored
A global counter for the magnitude of memcg stats update is maintained on the memcg side to avoid invoking rstat flushes when the pending updates are not significant. This avoids unnecessary flushes, which are not very cheap even if there isn't a lot of stats to flush. It also avoids unnecessary lock contention on the underlying global rstat lock. Make this threshold per-memcg. The scheme is followed where percpu (now also per-memcg) counters are incremented in the update path, and only propagated to per-memcg atomics when they exceed a certain threshold. This provides two benefits: (a) On large machines with a lot of memcgs, the global threshold can be reached relatively fast, so guarding the underlying lock becomes less effective. Making the threshold per-memcg avoids this. (b) Having a global threshold makes it hard to do subtree flushes, as we cannot reset the global counter except for a full flush. Per-memcg counters removes this as a blocker from doing subtree flushes, which helps avoid unnecessary work when the stats of a small subtree are needed. Nothing is free, of course. This comes at a cost: (a) A new per-cpu counter per memcg, consuming NR_CPUS * NR_MEMCGS * 4 bytes. The extra memory usage is insigificant. (b) More work on the update side, although in the common case it will only be percpu counter updates. The amount of work scales with the number of ancestors (i.e. tree depth). This is not a new concept, adding a cgroup to the rstat tree involves a parent loop, so is charging. Testing results below show no significant regressions. (c) The error margin in the stats for the system as a whole increases from NR_CPUS * MEMCG_CHARGE_BATCH to NR_CPUS * MEMCG_CHARGE_BATCH * NR_MEMCGS. This is probably fine because we have a similar per-memcg error in charges coming from percpu stocks, and we have a periodic flusher that makes sure we always flush all the stats every 2s anyway. This patch was tested to make sure no significant regressions are introduced on the update path as follows. The following benchmarks were ran in a cgroup that is 2 levels deep (/sys/fs/cgroup/a/b/): (1) Running 22 instances of netperf on a 44 cpu machine with hyperthreading disabled. All instances are run in a level 2 cgroup, as well as netserver: # netserver -6 # netperf -6 -H ::1 -l 60 -t TCP_SENDFILE -- -m 10K Averaging 20 runs, the numbers are as follows: Base: 40198.0 mbps Patched: 38629.7 mbps (-3.9%) The regression is minimal, especially for 22 instances in the same cgroup sharing all ancestors (so updating the same atomics). (2) will-it-scale page_fault tests. These tests (specifically per_process_ops in page_fault3 test) detected a 25.9% regression before for a change in the stats update path [1]. These are the numbers from 10 runs (+ is good) on a machine with 256 cpus: LABEL | MEAN | MEDIAN | STDDEV | ------------------------------+-------------+-------------+------------- page_fault1_per_process_ops | | | | (A) base | 270249.164 | 265437.000 | 13451.836 | (B) patched | 261368.709 | 255725.000 | 13394.767 | | -3.29% | -3.66% | | page_fault1_per_thread_ops | | | | (A) base | 242111.345 | 239737.000 | 10026.031 | (B) patched | 237057.109 | 235305.000 | 9769.687 | | -2.09% | -1.85% | | page_fault1_scalability | | | (A) base | 0.034387 | 0.035168 | 0.0018283 | (B) patched | 0.033988 | 0.034573 | 0.0018056 | | -1.16% | -1.69% | | page_fault2_per_process_ops | | | (A) base | 203561.836 | 203301.000 | 2550.764 | (B) patched | 197195.945 | 197746.000 | 2264.263 | | -3.13% | -2.73% | | page_fault2_per_thread_ops | | | (A) base | 171046.473 | 170776.000 | 1509.679 | (B) patched | 166626.327 | 166406.000 | 768.753 | | -2.58% | -2.56% | | page_fault2_scalability | | | (A) base | 0.054026 | 0.053821 | 0.00062121 | (B) patched | 0.053329 | 0.05306 | 0.00048394 | | -1.29% | -1.41% | | page_fault3_per_process_ops | | | (A) base | 1295807.782 | 1297550.000 | 5907.585 | (B) patched | 1275579.873 | 1273359.000 | 8759.160 | | -1.56% | -1.86% | | page_fault3_per_thread_ops | | | (A) base | 391234.164 | 390860.000 | 1760.720 | (B) patched | 377231.273 | 376369.000 | 1874.971 | | -3.58% | -3.71% | | page_fault3_scalability | | | (A) base | 0.60369 | 0.60072 | 0.0083029 | (B) patched | 0.61733 | 0.61544 | 0.009855 | | +2.26% | +2.45% | | All regressions seem to be minimal, and within the normal variance for the benchmark. The fix for [1] assumes that 3% is noise -- and there were no further practical complaints), so hopefully this means that such variations in these microbenchmarks do not reflect on practical workloads. (3) I also ran stress-ng in a nested cgroup and did not observe any obvious regressions. [1]https://lore.kernel.org/all/20190520063534.GB19312@shao2-debian/ Link: https://lkml.kernel.org/r/20231129032154.3710765-4-yosryahmed@google.comSigned-off-by:Johannes Weiner <hannes@cmpxchg.org> Tested-by:
-
Yosry Ahmed authored
The following patch will make use of those structs in the flushing code, so move their definitions (and a few other dependencies) a little bit up to reduce the diff noise in the following patch. No functional change intended. Link: https://lkml.kernel.org/r/20231129032154.3710765-3-yosryahmed@google.comSigned-off-by:
-
Yosry Ahmed authored
Patch series "mm: memcg: subtree stats flushing and thresholds", v4. This series attempts to address shortages in today's approach for memcg stats flushing, namely occasionally stale or expensive stat reads. The series does so by changing the threshold that we use to decide whether to trigger a flush to be per memcg instead of global (patch 3), and then changing flushing to be per memcg (i.e. subtree flushes) instead of global (patch 5). This patch (of 5): flush_next_time is an inaccurate name. It's not the next time that periodic flushing will happen, it's rather the next time that ratelimited flushing can happen if the periodic flusher is late. Simplify its semantics by just storing the timestamp of the last flush instead, flush_last_time. Move the 2*FLUSH_TIME addition to mem_cgroup_flush_stats_ratelimited(), and add a comment explaining it. This way, all the ratelimiting semantics live in one place. No functional change intended. Link: https://lkml.kernel.org/r/20231129032154.3710765-1-yosryahmed@google.com Link: https://lkml.kernel.org/r/20231129032154.3710765-2-yosryahmed@google.comSigned-off-by:
Bagas Sanjaya <bagasdotme@gmail.com> Cc: Greg Thelen <gthelen@google.com> Cc: Ivan Babrou <ivan@cloudflare.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Michal Hocko <mhocko@kernel.org> Cc: Michal Koutny <mkoutny@suse.com> Cc: Muchun Song <muchun.song@linux.dev> Cc: Roman Gushchin <roman.gushchin@linux.dev> Cc: Tejun Heo <tj@kernel.org> Cc: Waiman Long <longman@redhat.com> Cc: Wei Xu <weixugc@google.com> Signed-off-by:
-
Andrew Morton authored
Fixes: 0a97c01c ("list_lru: allow explicit memcg and NUMA node selection) Reported-by:
kernel test robot <lkp@intel.com> Closes: https://lore.kernel.org/oe-kbuild-all/202312141318.q8b5yrAq-lkp@intel.com/ Cc: Nhat Pham <nphamcs@gmail.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Bagas Sanjaya <bagasdotme@gmail.com> Signed-off-by:
-
Andrew Morton authored
Commit 0de56e38 ("maple_tree: use maple state end for write operations") was broken by a later patch "maple_tree: do not preallocate nodes for slot stores". But the later patch was scheduled ahead of 0de56e38, for 6.7-rc. This fixlet undoes the damage. Fixes: 0de56e38 ("maple_tree: use maple state end for write operations") Cc: Liam R. Howlett <Liam.Howlett@oracle.com> Cc: Sidhartha Kumar <sidhartha.kumar@oracle.com> Signed-off-by:
-
