- 07 Sep, 2017 40 commits
-
-
Huang Ying authored
Huge page helps to reduce TLB miss rate, but it has higher cache footprint, sometimes this may cause some issue. For example, when clearing huge page on x86_64 platform, the cache footprint is 2M. But on a Xeon E5 v3 2699 CPU, there are 18 cores, 36 threads, and only 45M LLC (last level cache). That is, in average, there are 2.5M LLC for each core and 1.25M LLC for each thread. If the cache pressure is heavy when clearing the huge page, and we clear the huge page from the begin to the end, it is possible that the begin of huge page is evicted from the cache after we finishing clearing the end of the huge page. And it is possible for the application to access the begin of the huge page after clearing the huge page. To help the above situation, in this patch, when we clear a huge page, the order to clear sub-pages is changed. In quite some situation, we can get the address that the application will access after we clear the huge page, for example, in a page fault handler. Instead of clearing the huge page from begin to end, we will clear the sub-pages farthest from the the sub-page to access firstly, and clear the sub-page to access last. This will make the sub-page to access most cache-hot and sub-pages around it more cache-hot too. If we cannot know the address the application will access, the begin of the huge page is assumed to be the the address the application will access. With this patch, the throughput increases ~28.3% in vm-scalability anon-w-seq test case with 72 processes on a 2 socket Xeon E5 v3 2699 system (36 cores, 72 threads). The test case creates 72 processes, each process mmap a big anonymous memory area and writes to it from the begin to the end. For each process, other processes could be seen as other workload which generates heavy cache pressure. At the same time, the cache miss rate reduced from ~33.4% to ~31.7%, the IPC (instruction per cycle) increased from 0.56 to 0.74, and the time spent in user space is reduced ~7.9% Christopher Lameter suggests to clear bytes inside a sub-page from end to begin too. But tests show no visible performance difference in the tests. May because the size of page is small compared with the cache size. Thanks Andi Kleen to propose to use address to access to determine the order of sub-pages to clear. The hugetlbfs access address could be improved, will do that in another patch. [ying.huang@intel.com: improve readability of clear_huge_page()] Link: http://lkml.kernel.org/r/20170830051842.1397-1-ying.huang@intel.com Link: http://lkml.kernel.org/r/20170815014618.15842-1-ying.huang@intel.comSuggested-by:
Andi Kleen <andi.kleen@intel.com> Signed-off-by:
"Huang, Ying" <ying.huang@intel.com> Acked-by:
Jan Kara <jack@suse.cz> Reviewed-by:
Michal Hocko <mhocko@suse.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com> Cc: Nadia Yvette Chambers <nyc@holomorphy.com> Cc: Matthew Wilcox <mawilcox@microsoft.com> Cc: Hugh Dickins <hughd@google.com> Cc: Minchan Kim <minchan@kernel.org> Cc: Shaohua Li <shli@fb.com> Cc: Christopher Lameter <cl@linux.com> Cc: Mike Kravetz <mike.kravetz@oracle.com> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org>
-
Andrea Arcangeli authored
This is purely required because exit_aio() may block and exit_mmap() may never start, if the oom_reap_task cannot start running on a mm with mm_users == 0. At the same time if the OOM reaper doesn't wait at all for the memory of the current OOM candidate to be freed by exit_mmap->unmap_vmas, it would generate a spurious OOM kill. If it wasn't because of the exit_aio or similar blocking functions in the last mmput, it would be enough to change the oom_reap_task() in the case it finds mm_users == 0, to wait for a timeout or to wait for __mmput to set MMF_OOM_SKIP itself, but it's not just exit_mmap the problem here so the concurrency of exit_mmap and oom_reap_task is apparently warranted. It's a non standard runtime, exit_mmap() runs without mmap_sem, and oom_reap_task runs with the mmap_sem for reading as usual (kind of MADV_DONTNEED). The race between the two is solved with a combination of tsk_is_oom_victim() (serialized by task_lock) and MMF_OOM_SKIP (serialized by a dummy down_write/up_write cycle on the same lines of the ksm_exit method). If the oom_reap_task() may be running concurrently during exit_mmap, exit_mmap will wait it to finish in down_write (before taking down mm structures that would make the oom_reap_task fail with use after free). If exit_mmap comes first, oom_reap_task() will skip the mm if MMF_OOM_SKIP is already set and in turn all memory is already freed and furthermore the mm data structures may already have been taken down by free_pgtables. [aarcange@redhat.com: incremental one liner] Link: http://lkml.kernel.org/r/20170726164319.GC29716@redhat.com [rientjes@google.com: remove unused mmput_async] Link: http://lkml.kernel.org/r/alpine.DEB.2.10.1708141733130.50317@chino.kir.corp.google.com [aarcange@redhat.com: microoptimization] Link: http://lkml.kernel.org/r/20170817171240.GB5066@redhat.com Link: http://lkml.kernel.org/r/20170726162912.GA29716@redhat.com Fixes: 26db62f1 ("oom: keep mm of the killed task available") Signed-off-by:
Andrea Arcangeli <aarcange@redhat.com> Signed-off-by:
David Rientjes <rientjes@google.com> Reported-by:
-
Aaron Lu authored
If the system has more than one swap device and swap device has the node information, we can make use of this information to decide which swap device to use in get_swap_pages() to get better performance. The current code uses a priority based list, swap_avail_list, to decide which swap device to use and if multiple swap devices share the same priority, they are used round robin. This patch changes the previous single global swap_avail_list into a per-numa-node list, i.e. for each numa node, it sees its own priority based list of available swap devices. Swap device's priority can be promoted on its matching node's swap_avail_list. The current swap device's priority is set as: user can set a >=0 value, or the system will pick one starting from -1 then downwards. The priority value in the swap_avail_list is the negated value of the swap device's due to plist being sorted from low to high. The new policy doesn't change the semantics for priority >=0 cases, the previous starting from -1 then downwards now becomes starting from -2 then downwards and -1 is reserved as the promoted value. Take 4-node EX machine as an example, suppose 4 swap devices are available, each sit on a different node: swapA on node 0 swapB on node 1 swapC on node 2 swapD on node 3 After they are all swapped on in the sequence of ABCD. Current behaviour: their priorities will be: swapA: -1 swapB: -2 swapC: -3 swapD: -4 And their position in the global swap_avail_list will be: swapA -> swapB -> swapC -> swapD prio:1 prio:2 prio:3 prio:4 New behaviour: their priorities will be(note that -1 is skipped): swapA: -2 swapB: -3 swapC: -4 swapD: -5 And their positions in the 4 swap_avail_lists[nid] will be: swap_avail_lists[0]: /* node 0's available swap device list */ swapA -> swapB -> swapC -> swapD prio:1 prio:3 prio:4 prio:5 swap_avali_lists[1]: /* node 1's available swap device list */ swapB -> swapA -> swapC -> swapD prio:1 prio:2 prio:4 prio:5 swap_avail_lists[2]: /* node 2's available swap device list */ swapC -> swapA -> swapB -> swapD prio:1 prio:2 prio:3 prio:5 swap_avail_lists[3]: /* node 3's available swap device list */ swapD -> swapA -> swapB -> swapC prio:1 prio:2 prio:3 prio:4 To see the effect of the patch, a test that starts N process, each mmap a region of anonymous memory and then continually write to it at random position to trigger both swap in and out is used. On a 2 node Skylake EP machine with 64GiB memory, two 170GB SSD drives are used as swap devices with each attached to a different node, the result is: runtime=30m/processes=32/total test size=128G/each process mmap region=4G kernel throughput vanilla 13306 auto-binding 15169 +14% runtime=30m/processes=64/total test size=128G/each process mmap region=2G kernel throughput vanilla 11885 auto-binding 14879 +25% [aaron.lu@intel.com: v2] Link: http://lkml.kernel.org/r/20170814053130.GD2369@aaronlu.sh.intel.com Link: http://lkml.kernel.org/r/20170816024439.GA10925@aaronlu.sh.intel.com [akpm@linux-foundation.org: use kmalloc_array()] Link: http://lkml.kernel.org/r/20170814053130.GD2369@aaronlu.sh.intel.com Link: http://lkml.kernel.org/r/20170816024439.GA10925@aaronlu.sh.intel.comSigned-off-by:
Aaron Lu <aaron.lu@intel.com> Cc: "Chen, Tim C" <tim.c.chen@intel.com> Cc: Huang Ying <ying.huang@intel.com> Cc: Andi Kleen <andi@firstfloor.org> Cc: Michal Hocko <mhocko@suse.com> Cc: Minchan Kim <minchan@kernel.org> Cc: Hugh Dickins <hughd@google.com> Signed-off-by:
-
Michal Hocko authored
TIF_MEMDIE is set only to the tasks whick were either directly selected by the OOM killer or passed through mark_oom_victim from the allocator path. tsk_is_oom_victim is more generic and allows to identify all tasks (threads) which share the mm with the oom victim. Please note that the freezer still needs to check TIF_MEMDIE because we cannot thaw tasks which do not participage in oom_victims counting otherwise a !TIF_MEMDIE task could interfere after oom_disbale returns. Link: http://lkml.kernel.org/r/20170810075019.28998-3-mhocko@kernel.orgSigned-off-by:
-
Michal Hocko authored
For ages we have been relying on TIF_MEMDIE thread flag to mark OOM victims and then, among other things, to give these threads full access to memory reserves. There are few shortcomings of this implementation, though. First of all and the most serious one is that the full access to memory reserves is quite dangerous because we leave no safety room for the system to operate and potentially do last emergency steps to move on. Secondly this flag is per task_struct while the OOM killer operates on mm_struct granularity so all processes sharing the given mm are killed. Giving the full access to all these task_structs could lead to a quick memory reserves depletion. We have tried to reduce this risk by giving TIF_MEMDIE only to the main thread and the currently allocating task but that doesn't really solve this problem while it surely opens up a room for corner cases - e.g. GFP_NO{FS,IO} requests might loop inside the allocator without access to memory reserves because a particular thread was not the group leader. Now that we have the oom reaper and that all oom victims are reapable after 1b51e65e ("oom, oom_reaper: allow to reap mm shared by the kthreads") we can be more conservative and grant only partial access to memory reserves because there are reasonable chances of the parallel memory freeing. We still want some access to reserves because we do not want other consumers to eat up the victim's freed memory. oom victims will still contend with __GFP_HIGH users but those shouldn't be so aggressive to starve oom victims completely. Introduce ALLOC_OOM flag and give all tsk_is_oom_victim tasks access to the half of the reserves. This makes the access to reserves independent on which task has passed through mark_oom_victim. Also drop any usage of TIF_MEMDIE from the page allocator proper and replace it by tsk_is_oom_victim as well which will make page_alloc.c completely TIF_MEMDIE free finally. CONFIG_MMU=n doesn't have oom reaper so let's stick to the original ALLOC_NO_WATERMARKS approach. There is a demand to make the oom killer memcg aware which will imply many tasks killed at once. This change will allow such a usecase without worrying about complete memory reserves depletion. Link: http://lkml.kernel.org/r/20170810075019.28998-2-mhocko@kernel.orgSigned-off-by:Mel Gorman <mgorman@techsingularity.net> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Roman Gushchin <guro@fb.com> Signed-off-by:
-
Vitaly Wool authored
It's been noted that z3fold doesn't scale well when it's run in a large number of threads on many cores, which can be easily reproduced with fio 'randrw' test with --numjobs=32. E.g. the result for 1 cluster (4 cores) is: Run status group 0 (all jobs): READ: io=244785MB, aggrb=496883KB/s, minb=15527KB/s, ... WRITE: io=246735MB, aggrb=500841KB/s, minb=15651KB/s, ... While for 8 cores (2 clusters) the result is: Run status group 0 (all jobs): READ: io=244785MB, aggrb=265942KB/s, minb=8310KB/s, ... WRITE: io=246735MB, aggrb=268060KB/s, minb=8376KB/s, ... The bottleneck here is the pool lock which many threads become waiting upon. To reduce that spin lock contention, z3fold can operate only on the lists local to the current CPU whenever possible. Due to the nature of z3fold unbuddied list handling (it only takes the first entry off the list on a hot path), if the z3fold pool is big enough and balanced well enough, limiting search to only local unbuddied list doesn't lead to a significant compression ratio degrade (2.57x vs 2.65x in our measurements). This patch also introduces two worker threads: one for async in-page object layout optimization and one for releasing freed pages. This is done to speed up z3fold_free() which is often on a hot path. The fio results for 8-core case are now the following: Run status group 0 (all jobs): READ: io=244785MB, aggrb=1568.3MB/s, minb=50182KB/s, ... WRITE: io=246735MB, aggrb=1580.8MB/s, minb=50582KB/s, ... So we're in for almost 6x performance increase. Link: http://lkml.kernel.org/r/20170806181443.f9b65018f8bde25ef990f9e8@gmail.comSigned-off-by:
Vitaly Wool <vitalywool@gmail.com> Cc: Dan Streetman <ddstreet@ieee.org> Signed-off-by:
-
Huang Ying authored
VMA based swap readahead will readahead the virtual pages that is continuous in the virtual address space. While the original swap readahead will readahead the swap slots that is continuous in the swap device. Although VMA based swap readahead is more correct for the swap slots to be readahead, it will trigger more small random readings, which may cause the performance of HDD (hard disk) to degrade heavily, and may finally exceed the benefit. To avoid the issue, in this patch, if the HDD is used as swap, the VMA based swap readahead will be disabled, and the original swap readahead will be used instead. Link: http://lkml.kernel.org/r/20170807054038.1843-6-ying.huang@intel.comSigned-off-by:
-
Huang Ying authored
The sysfs interface to control the VMA based swap readahead is added as follow, /sys/kernel/mm/swap/vma_ra_enabled Enable the VMA based swap readahead algorithm, or use the original global swap readahead algorithm. /sys/kernel/mm/swap/vma_ra_max_order Set the max order of the readahead window size for the VMA based swap readahead algorithm. The corresponding ABI documentation is added too. Link: http://lkml.kernel.org/r/20170807054038.1843-5-ying.huang@intel.comSigned-off-by:
-
Huang Ying authored
The swap readahead is an important mechanism to reduce the swap in latency. Although pure sequential memory access pattern isn't very popular for anonymous memory, the space locality is still considered valid. In the original swap readahead implementation, the consecutive blocks in swap device are readahead based on the global space locality estimation. But the consecutive blocks in swap device just reflect the order of page reclaiming, don't necessarily reflect the access pattern in virtual memory. And the different tasks in the system may have different access patterns, which makes the global space locality estimation incorrect. In this patch, when page fault occurs, the virtual pages near the fault address will be readahead instead of the swap slots near the fault swap slot in swap device. This avoid to readahead the unrelated swap slots. At the same time, the swap readahead is changed to work on per-VMA from globally. So that the different access patterns of the different VMAs could be distinguished, and the different readahead policy could be applied accordingly. The original core readahead detection and scaling algorithm is reused, because it is an effect algorithm to detect the space locality. The test and result is as follow, Common test condition ===================== Test Machine: Xeon E5 v3 (2 sockets, 72 threads, 32G RAM) Swap device: NVMe disk Micro-benchmark with combined access pattern ============================================ vm-scalability, sequential swap test case, 4 processes to eat 50G virtual memory space, repeat the sequential memory writing until 300 seconds. The first round writing will trigger swap out, the following rounds will trigger sequential swap in and out. At the same time, run vm-scalability random swap test case in background, 8 processes to eat 30G virtual memory space, repeat the random memory write until 300 seconds. This will trigger random swap-in in the background. This is a combined workload with sequential and random memory accessing at the same time. The result (for sequential workload) is as follow, Base Optimized ---- --------- throughput 345413 KB/s 414029 KB/s (+19.9%) latency.average 97.14 us 61.06 us (-37.1%) latency.50th 2 us 1 us latency.60th 2 us 1 us latency.70th 98 us 2 us latency.80th 160 us 2 us latency.90th 260 us 217 us latency.95th 346 us 369 us latency.99th 1.34 ms 1.09 ms ra_hit% 52.69% 99.98% The original swap readahead algorithm is confused by the background random access workload, so readahead hit rate is lower. The VMA-base readahead algorithm works much better. Linpack ======= The test memory size is bigger than RAM to trigger swapping. Base Optimized ---- --------- elapsed_time 393.49 s 329.88 s (-16.2%) ra_hit% 86.21% 98.82% The score of base and optimized kernel hasn't visible changes. But the elapsed time reduced and readahead hit rate improved, so the optimized kernel runs better for startup and tear down stages. And the absolute value of readahead hit rate is high, shows that the space locality is still valid in some practical workloads. Link: http://lkml.kernel.org/r/20170807054038.1843-4-ying.huang@intel.comSigned-off-by:
-
Huang Ying authored
In the original implementation, it is possible that the existing pages in the swap cache (not newly readahead) could be marked as the readahead pages. This will cause the statistics of swap readahead be wrong and influence the swap readahead algorithm too. This is fixed via marking a page as the readahead page only if it is newly allocated and read from the disk. When testing with linpack, after the fixing the swap readahead hit rate increased from ~66% to ~86%. Link: http://lkml.kernel.org/r/20170807054038.1843-3-ying.huang@intel.comSigned-off-by:
-
Huang Ying authored
Patch series "mm, swap: VMA based swap readahead", v4. The swap readahead is an important mechanism to reduce the swap in latency. Although pure sequential memory access pattern isn't very popular for anonymous memory, the space locality is still considered valid. In the original swap readahead implementation, the consecutive blocks in swap device are readahead based on the global space locality estimation. But the consecutive blocks in swap device just reflect the order of page reclaiming, don't necessarily reflect the access pattern in virtual memory space. And the different tasks in the system may have different access patterns, which makes the global space locality estimation incorrect. In this patchset, when page fault occurs, the virtual pages near the fault address will be readahead instead of the swap slots near the fault swap slot in swap device. This avoid to readahead the unrelated swap slots. At the same time, the swap readahead is changed to work on per-VMA from globally. So that the different access patterns of the different VMAs could be distinguished, and the different readahead policy could be applied accordingly. The original core readahead detection and scaling algorithm is reused, because it is an effect algorithm to detect the space locality. In addition to the swap readahead changes, some new sysfs interface is added to show the efficiency of the readahead algorithm and some other swap statistics. This new implementation will incur more small random read, on SSD, the improved correctness of estimation and readahead target should beat the potential increased overhead, this is also illustrated in the test results below. But on HDD, the overhead may beat the benefit, so the original implementation will be used by default. The test and result is as follow, Common test condition ===================== Test Machine: Xeon E5 v3 (2 sockets, 72 threads, 32G RAM) Swap device: NVMe disk Micro-benchmark with combined access pattern ============================================ vm-scalability, sequential swap test case, 4 processes to eat 50G virtual memory space, repeat the sequential memory writing until 300 seconds. The first round writing will trigger swap out, the following rounds will trigger sequential swap in and out. At the same time, run vm-scalability random swap test case in background, 8 processes to eat 30G virtual memory space, repeat the random memory write until 300 seconds. This will trigger random swap-in in the background. This is a combined workload with sequential and random memory accessing at the same time. The result (for sequential workload) is as follow, Base Optimized ---- --------- throughput 345413 KB/s 414029 KB/s (+19.9%) latency.average 97.14 us 61.06 us (-37.1%) latency.50th 2 us 1 us latency.60th 2 us 1 us latency.70th 98 us 2 us latency.80th 160 us 2 us latency.90th 260 us 217 us latency.95th 346 us 369 us latency.99th 1.34 ms 1.09 ms ra_hit% 52.69% 99.98% The original swap readahead algorithm is confused by the background random access workload, so readahead hit rate is lower. The VMA-base readahead algorithm works much better. Linpack ======= The test memory size is bigger than RAM to trigger swapping. Base Optimized ---- --------- elapsed_time 393.49 s 329.88 s (-16.2%) ra_hit% 86.21% 98.82% The score of base and optimized kernel hasn't visible changes. But the elapsed time reduced and readahead hit rate improved, so the optimized kernel runs better for startup and tear down stages. And the absolute value of readahead hit rate is high, shows that the space locality is still valid in some practical workloads. This patch (of 5): The statistics for total readahead pages and total readahead hits are recorded and exported via the following sysfs interface. /sys/kernel/mm/swap/ra_hits /sys/kernel/mm/swap/ra_total With them, the efficiency of the swap readahead could be measured, so that the swap readahead algorithm and parameters could be tuned accordingly. [akpm@linux-foundation.org: don't display swap stats if CONFIG_SWAP=n] Link: http://lkml.kernel.org/r/20170807054038.1843-2-ying.huang@intel.comSigned-off-by:
-
Byungchul Park authored
Although llist provides proper APIs, they are not used. Make them used. Link: http://lkml.kernel.org/r/1502095374-16112-1-git-send-email-byungchul.park@lge.comSigned-off-by:
Byungchul Park <byungchul.park@lge.com> Cc: zijun_hu <zijun_hu@htc.com> Cc: Michal Hocko <mhocko@suse.com> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: Joel Fernandes <joelaf@google.com> Cc: Andrey Ryabinin <aryabinin@virtuozzo.com> Signed-off-by:
-
SeongJae Park authored
Comment for pagetypeinfo_showblockcount() is mistakenly duplicated from pagetypeinfo_show_free()'s comment. This commit fixes it. Link: http://lkml.kernel.org/r/20170809185816.11244-1-sj38.park@gmail.com Fixes: 467c996c ("Print out statistics in relation to fragmentation avoidance to /proc/pagetypeinfo") Signed-off-by:
SeongJae Park <sj38.park@gmail.com> Cc: Michal Hocko <mhocko@kernel.org> Cc: Vlastimil Babka <vbabka@suse.cz> Signed-off-by:
-
Mike Kravetz authored
With the addition of hugetlbfs support in memfd_create, the memfd selftests should verify correct functionality with hugetlbfs. Instead of writing a separate memfd hugetlbfs test, modify the memfd_test program to take an optional argument 'hugetlbfs'. If the hugetlbfs argument is specified, basic memfd_create functionality will be exercised on hugetlbfs. If hugetlbfs is not specified, the current functionality of the test is unchanged. Note that many of the tests in memfd_test test file sealing operations. hugetlbfs does not support file sealing, therefore for hugetlbfs all sealing related tests are skipped. In order to test on hugetlbfs, there needs to be preallocated huge pages. A new script (run_tests) is added. This script will first run the existing memfd_create tests. It will then, attempt to allocate the required number of huge pages before running the hugetlbfs test. At the end of testing, it will release any huge pages allocated for testing purposes. Link: http://lkml.kernel.org/r/1502495772-24736-3-git-send-email-mike.kravetz@oracle.comSigned-off-by:
Mike Kravetz <mike.kravetz@oracle.com> Cc: Michal Hocko <mhocko@suse.com> Cc: Hugh Dickins <hughd@google.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: "Kirill A . Shutemov" <kirill.shutemov@linux.intel.com> Signed-off-by:
-
Mike Kravetz authored
This patch came out of discussions in this e-mail thread: http://lkml.kernel.org/r/1499357846-7481-1-git-send-email-mike.kravetz%40oracle.com The Oracle JVM team is developing a new garbage collection model. This new model requires multiple mappings of the same anonymous memory. One straight forward way to accomplish this is with memfd_create. They can use the returned fd to create multiple mappings of the same memory. The JVM today has an option to use (static hugetlb) huge pages. If this option is specified, they would like to use the same garbage collection model requiring multiple mappings to the same memory. Using hugetlbfs, it is possible to explicitly mount a filesystem and specify file paths in order to get an fd that can be used for multiple mappings. However, this introduces additional system admin work and coordination. Ideally they would like to get a hugetlbfs fd without requiring explicit mounting of a filesystem. Today, mmap and shmget can make use of hugetlbfs without explicitly mounting a filesystem. The patch adds this functionality to memfd_create. Add a new flag MFD_HUGETLB to memfd_create() that will specify the file to be created resides in the hugetlbfs filesystem. This is the generic hugetlbfs filesystem not associated with any specific mount point. As with other system calls that request hugetlbfs backed pages, there is the ability to encode huge page size in the flag arguments. hugetlbfs does not support sealing operations, therefore specifying MFD_ALLOW_SEALING with MFD_HUGETLB will result in EINVAL. Of course, the memfd_man page would need updating if this type of functionality moves forward. Link: http://lkml.kernel.org/r/1502149672-7759-2-git-send-email-mike.kravetz@oracle.comSigned-off-by:
-
Dan Williams authored
devm_memremap_pages() records mapped ranges in pgmap_radix with an entry per section's worth of memory (128MB). The key for each of those entries is a section number. This leads to false positives when devm_memremap_pages() is passed a section-unaligned range as lookups in the misalignment fail to return NULL. We can close this hole by using the pfn as the key for entries in the tree. The number of entries required to describe a remapped range is reduced by leveraging multi-order entries. In practice this approach usually yields just one entry in the tree if the size and starting address are of the same power-of-2 alignment. Previously we always needed nr_entries = mapping_size / 128MB. Link: https://lists.01.org/pipermail/linux-nvdimm/2016-August/006666.html Link: http://lkml.kernel.org/r/150215410565.39310.13767886055248249438.stgit@dwillia2-desk3.amr.corp.intel.comSigned-off-by:
Dan Williams <dan.j.williams@intel.com> Reported-by:
Toshi Kani <toshi.kani@hpe.com> Cc: Matthew Wilcox <mawilcox@microsoft.com> Signed-off-by:
-
Wei Yang authored
In pcpu_get_vm_areas(), it checks each range is not overlapped. To make sure it is, only (N^2)/2 comparison is necessary, while current code does N^2 times. By starting from the next range, it achieves the goal and the continue could be removed. Also, - the overlap check of two ranges could be done with one clause - one typo in comment is fixed. Link: http://lkml.kernel.org/r/20170803063822.48702-1-richard.weiyang@gmail.comSigned-off-by:
Wei Yang <richard.weiyang@gmail.com> Acked-by:
Tejun Heo <tj@kernel.org> Cc: Michal Hocko <mhocko@kernel.org> Signed-off-by:
-
Wen Yang authored
When order is -1 or too big, *1UL << order* will be 0, which will cause a divide error. Although it seems that all callers of __fragmentation_index() will only do so with a valid order, the patch can make it more robust. Should prevent reoccurrences of https://bugzilla.kernel.org/show_bug.cgi?id=196555 Link: http://lkml.kernel.org/r/1501751520-2598-1-git-send-email-wen.yang99@zte.com.cnSigned-off-by:
Wen Yang <wen.yang99@zte.com.cn> Reviewed-by:
Jiang Biao <jiang.biao2@zte.com.cn> Suggested-by:
Vlastimil Babka <vbabka@suse.cz> Signed-off-by:
-
Michal Hocko authored
alloc_gigantic_page doesn't consider movability of the gigantic hugetlb when scanning eligible ranges for the allocation. As 1GB hugetlb pages are not movable currently this can break the movable zone assumption that all allocations are migrateable and as such break memory hotplug. Reorganize the code and use the standard zonelist allocations scheme that we use for standard hugetbl pages. htlb_alloc_mask will ensure that only migratable hugetlb pages will ever see a movable zone. Link: http://lkml.kernel.org/r/20170803083549.21407-1-mhocko@kernel.org Fixes: 944d9fec ("hugetlb: add support for gigantic page allocation at runtime") Signed-off-by:
-
Andrea Arcangeli authored
No ABI change, but this will make it more explicit to software that ptid is only available if requested by passing UFFD_FEATURE_THREAD_ID to UFFDIO_API. The fact it's a union will also self document it shouldn't be taken for granted there's a tpid there. Link: http://lkml.kernel.org/r/20170802165145.22628-7-aarcange@redhat.comSigned-off-by:
-
Alexey Perevalov authored
It could be useful for calculating downtime during postcopy live migration per vCPU. Side observer or application itself will be informed about proper task's sleep during userfaultfd processing. Process's thread id is being provided when user requeste it by setting UFFD_FEATURE_THREAD_ID bit into uffdio_api.features. Link: http://lkml.kernel.org/r/20170802165145.22628-6-aarcange@redhat.comSigned-off-by:
Alexey Perevalov <a.perevalov@samsung.com> Signed-off-by:
-
Andrea Arcangeli authored
A __split_vma is not a worthy event to report, and it's definitely not a unmap so it would be incorrect to report unmap for the whole region to the userfaultfd manager if a __split_vma fails. So only call userfaultfd_unmap_prep after the __vma_splitting is over and do_munmap cannot fail anymore. Also add unlikely because it's better to optimize for the vast majority of apps that aren't using userfaultfd in a non cooperative way. Ideally we should also find a way to eliminate the branch entirely if CONFIG_USERFAULTFD=n, but it would complicate things so stick to unlikely for now. Link: http://lkml.kernel.org/r/20170802165145.22628-5-aarcange@redhat.comSigned-off-by:
-
Andrea Arcangeli authored
Showing zero in the output isn't very self explanatory as a successful result. Show a more explicit error output if the test fails. Link: http://lkml.kernel.org/r/20170802165145.22628-4-aarcange@redhat.comSigned-off-by:
-
Andrea Arcangeli authored
This will retry the UFFDIO_COPY/ZEROPAGE to verify it returns -EEXIST at the first invocation and then later every 10 seconds. In the filebacked MAP_SHARED case this also verifies the -EEXIST triggered in the filesystem pagecache insertion, if the offset in the file was not a hole. shmem MAP_SHARED tries to index the newly allocated pagecache in the radix tree before checking the pagetable so it doesn't need any assistance to exercise that case. hugetlbfs checks the pmd to be not none before trying to index the hugetlbfs page in the radix tree, so it requires to run UFFDIO_COPY into an alias mapping (the alternative would be to use MADV_DONTNEED to only zap the pagetables, but that doesn't work on hugetlbfs). [akpm@linux-foundation.org: fix uffdio_zeropage(), per Mike Kravetz] Link: http://lkml.kernel.org/r/20170802165145.22628-3-aarcange@redhat.comSigned-off-by:
-
Prakash Sangappa authored
Add tests for UFFD_FEATURE_SIGBUS feature. The tests will verify signal delivery instead of userfault events. Also, test use of UFFDIO_COPY to allocate memory and retry accessing monitored area after signal delivery. Also fix a bug in uffd_poll_thread() where 'uffd' is leaked. Link: http://lkml.kernel.org/r/1501552446-748335-3-git-send-email-prakash.sangappa@oracle.comSigned-off-by:
Prakash Sangappa <prakash.sangappa@oracle.com> Cc: Shuah Khan <shuah@kernel.org> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Mike Kravetz <mike.kravetz@oracle.com> Cc: Mike Rapoport <rppt@linux.vnet.ibm.com> Signed-off-by:
-
Prakash Sangappa authored
In some cases, userfaultfd mechanism should just deliver a SIGBUS signal to the faulting process, instead of the page-fault event. Dealing with page-fault event using a monitor thread can be an overhead in these cases. For example applications like the database could use the signaling mechanism for robustness purpose. Database uses hugetlbfs for performance reason. Files on hugetlbfs filesystem are created and huge pages allocated using fallocate() API. Pages are deallocated/freed using fallocate() hole punching support. These files are mmapped and accessed by many processes as shared memory. The database keeps track of which offsets in the hugetlbfs file have pages allocated. Any access to mapped address over holes in the file, which can occur due to bugs in the application, is considered invalid and expect the process to simply receive a SIGBUS. However, currently when a hole in the file is accessed via the mapped address, kernel/mm attempts to automatically allocate a page at page fault time, resulting in implicitly filling the hole in the file. This may not be the desired behavior for applications like the database that want to explicitly manage page allocations of hugetlbfs files. Using userfaultfd mechanism with this support to get a signal, database application can prevent pages from being allocated implicitly when processes access mapped address over holes in the file. This patch adds UFFD_FEATURE_SIGBUS feature to userfaultfd mechnism to request for a SIGBUS signal. See following for previous discussion about the database requirement leading to this proposal as suggested by Andrea. http://www.spinics.net/lists/linux-mm/msg129224.html Link: http://lkml.kernel.org/r/1501552446-748335-2-git-send-email-prakash.sangappa@oracle.comSigned-off-by:
Mike Rapoport <rppt@linux.vnet.ibm.com> Reviewed-by:
-
Michal Hocko authored
global_page_state is error prone as a recent bug report pointed out [1]. It only returns proper values for zone based counters as the enum it gets suggests. We already have global_node_page_state so let's rename global_page_state to global_zone_page_state to be more explicit here. All existing users seems to be correct: $ git grep "global_page_state(NR_" | sed 's@.*(\(NR_[A-Z_]*\)).*@\1@' | sort | uniq -c 2 NR_BOUNCE 2 NR_FREE_CMA_PAGES 11 NR_FREE_PAGES 1 NR_KERNEL_STACK_KB 1 NR_MLOCK 2 NR_PAGETABLE This patch shouldn't introduce any functional change. [1] http://lkml.kernel.org/r/201707260628.v6Q6SmaS030814@www262.sakura.ne.jp Link: http://lkml.kernel.org/r/20170801134256.5400-2-hannes@cmpxchg.orgSigned-off-by:Johannes Weiner <hannes@cmpxchg.org> Cc: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp> Cc: Josef Bacik <josef@toxicpanda.com> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Signed-off-by:
-
Mike Kravetz authored
Use the common definitions from hugetlb_encode.h header file for encoding hugetlb size definitions in shmget system call flags. In addition, move these definitions from the internal (kernel) to user (uapi) header file. Link: http://lkml.kernel.org/r/1501527386-10736-4-git-send-email-mike.kravetz@oracle.comSigned-off-by:
Matthew Wilcox <willy@infradead.org> Acked-by:
-
Mike Kravetz authored
A non-default huge page size can be encoded in the flags argument of the mmap system call. The definitions for these encodings are in arch specific header files. However, all architectures use the same values. Consolidate all the definitions in the primary user header file (uapi/linux/mman.h). Include definitions for all known huge page sizes. Use the generic encoding definitions in hugetlb_encode.h as the basis for these definitions. Link: http://lkml.kernel.org/r/1501527386-10736-3-git-send-email-mike.kravetz@oracle.comSigned-off-by:
-
Mike Kravetz authored
Patch series "Consolidate system call hugetlb page size encodings". These patches are the result of discussions in https://lkml.org/lkml/2017/3/8/548. The following changes are made in the patch set: 1) Put all the log2 encoded huge page size definitions in a common header file. The idea is have a set of definitions that can be use as the basis for system call specific definitions such as MAP_HUGE_* and SHM_HUGE_*. 2) Remove MAP_HUGE_* definitions in arch specific files. All these definitions are the same. Consolidate all definitions in the primary user header file (uapi/linux/mman.h). 3) Remove SHM_HUGE_* definitions intended for user space from kernel header file, and add to user (uapi/linux/shm.h) header file. Add definitions for all known huge page size encodings as in mmap. This patch (of 3): If hugetlb pages are requested in mmap or shmget system calls, a huge page size other than default can be requested. This is accomplished by encoding the log2 of the huge page size in the upper bits of the flag argument. asm-generic and arch specific headers all define the same values for these encodings. Put common definitions in a single header file. The primary uapi header files for mmap and shm will use these definitions as a basis for definitions specific to those system calls. Link: http://lkml.kernel.org/r/1501527386-10736-2-git-send-email-mike.kravetz@oracle.comSigned-off-by:
-
Jeff Layton authored
Link: http://lkml.kernel.org/r/20170525102927.6163-1-jlayton@redhat.comSigned-off-by:
Jeff Layton <jlayton@redhat.com> Reviewed-by:
-
Jeff Layton authored
fsync codepath assumes that f_mapping can never be NULL, but sync_file_range has a check for that. Remove the one from sync_file_range as I don't see how you'd ever get a NULL pointer in here. Link: http://lkml.kernel.org/r/20170525110509.9434-1-jlayton@redhat.comSigned-off-by:
-
Mike Rapoport authored
Link: http://lkml.kernel.org/r/1497939652-16528-8-git-send-email-rppt@linux.vnet.ibm.comSigned-off-by:
-
Mike Rapoport authored
Now when shmem VMAs can be filled with zero page via userfaultfd we can report that UFFDIO_ZEROPAGE is available for those VMAs Link: http://lkml.kernel.org/r/1497939652-16528-7-git-send-email-rppt@linux.vnet.ibm.comSigned-off-by:
-
Mike Rapoport authored
For shmem VMAs we can use shmem_mfill_zeropage_pte for UFFDIO_ZEROPAGE Link: http://lkml.kernel.org/r/1497939652-16528-6-git-send-email-rppt@linux.vnet.ibm.comSigned-off-by:
-
Mike Rapoport authored
Shuffle the code a bit to improve readability. Link: http://lkml.kernel.org/r/1497939652-16528-5-git-send-email-rppt@linux.vnet.ibm.comSigned-off-by:
-
Mike Rapoport authored
shmem_mfill_zeropage_pte is the low level routine that implements the userfaultfd UFFDIO_ZEROPAGE command. Since for shmem mappings zero pages are always allocated and accounted, the new method is a slight extension of the existing shmem_mcopy_atomic_pte. Link: http://lkml.kernel.org/r/1497939652-16528-4-git-send-email-rppt@linux.vnet.ibm.comSigned-off-by:
-
Mike Rapoport authored
The shmem_acct_block and the update of used_blocks are following one another in all the places they are used. Combine these two into a helper function. Link: http://lkml.kernel.org/r/1497939652-16528-3-git-send-email-rppt@linux.vnet.ibm.comSigned-off-by:
-
Mike Rapoport authored
Patch series "userfaultfd: enable zeropage support for shmem". These patches enable support for UFFDIO_ZEROPAGE for shared memory. The first two patches are not strictly related to userfaultfd, they are just minor refactoring to reduce amount of code duplication. This patch (of 7): Currently we update inode and shmem_inode_info before verifying that used_blocks will not exceed max_blocks. In case it will, we undo the update. Let's switch the order and move the verification of the blocks count before the inode and shmem_inode_info update. Link: http://lkml.kernel.org/r/1497939652-16528-2-git-send-email-rppt@linux.vnet.ibm.comSigned-off-by:
-
Huang Ying authored
When swapping out THP (Transparent Huge Page), instead of swapping out the THP as a whole, sometimes we have to fallback to split the THP into normal pages before swapping, because no free swap clusters are available, or cgroup limit is exceeded, etc. To count the number of the fallback, a new VM event THP_SWPOUT_FALLBACK is added, and counted when we fallback to split the THP. Link: http://lkml.kernel.org/r/20170724051840.2309-13-ying.huang@intel.comSigned-off-by:
-
