- 26 Oct, 2018 40 commits
-
-
Alexandre Ghiti authored
arm, ia64, sh, x86 architectures use the same version of huge_ptep_get_and_clear, so move this generic implementation into asm-generic/hugetlb.h. Link: http://lkml.kernel.org/r/20180920060358.16606-5-alex@ghiti.frSigned-off-by:
Alexandre Ghiti <alex@ghiti.fr> Reviewed-by:
Luiz Capitulino <lcapitulino@redhat.com> Reviewed-by:
Mike Kravetz <mike.kravetz@oracle.com> Tested-by: Helge Deller <deller@gmx.de> [parisc] Acked-by: Catalin Marinas <catalin.marinas@arm.com> [arm64] Acked-by: Paul Burton <paul.burton@mips.com> [MIPS] Acked-by: Ingo Molnar <mingo@kernel.org> [x86] Cc: Arnd Bergmann <arnd@arndb.de> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: David S. Miller <davem@davemloft.net> Cc: Fenghua Yu <fenghua.yu@intel.com> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: H. Peter Anvin <hpa@zytor.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: James E.J. Bottomley <jejb@parisc-linux.org> Cc: James Hogan <jhogan@kernel.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Michael Ellerman <mpe@ellerman.id.au> Cc: Paul Mackerras <paulus@samba.org> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Rich Felker <dalias@libc.org> Cc: Russell King <linux@armlinux.org.uk> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Tony Luck <tony.luck@intel.com> Cc: Will Deacon <will.deacon@arm.com> Cc: Yoshinori Sato <ysato@users.sourceforge.jp> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org>
-
Alexandre Ghiti authored
arm, ia64, mips, powerpc, sh, x86 architectures use the same version of set_huge_pte_at, so move this generic implementation into asm-generic/hugetlb.h. Link: http://lkml.kernel.org/r/20180920060358.16606-4-alex@ghiti.frSigned-off-by:
-
Alexandre Ghiti authored
arm, arm64, mips, parisc, sh, x86 architectures use the same version of hugetlb_free_pgd_range, so move this generic implementation into asm-generic/hugetlb.h. Link: http://lkml.kernel.org/r/20180920060358.16606-3-alex@ghiti.frSigned-off-by:
-
Alexandre Ghiti authored
In order to reduce copy/paste of functions across architectures and then make riscv hugetlb port (and future ports) simpler and smaller, this patchset intends to factorize the numerous hugetlb primitives that are defined across all the architectures. Except for prepare_hugepage_range, this patchset moves the versions that are just pass-through to standard pte primitives into asm-generic/hugetlb.h by using the same #ifdef semantic that can be found in asm-generic/pgtable.h, i.e. __HAVE_ARCH_***. s390 architecture has not been tackled in this serie since it does not use asm-generic/hugetlb.h at all. This patchset has been compiled on all addressed architectures with success (except for parisc, but the problem does not come from this series). This patch (of 11): asm-generic/hugetlb.h proposes generic implementations of hugetlb related functions: use __HAVE_ARCH_HUGE* defines in order to make arch specific implementations of hugetlb functions consistent with pgtable.h scheme. Link: http://lkml.kernel.org/r/20180920060358.16606-2-alex@ghiti.frSigned-off-by:
-
Wei Yang authored
The local variable `addr' in __get_user_pages_fast() is just a shadow of `start'. Since `start' never changes after assignment to `addr', it is fine to replace `start' with it. Also the meaning of [start, end] is more obvious than [addr, end] when passed to gup_pgd_range(). Link: http://lkml.kernel.org/r/20180925021448.20265-1-richard.weiyang@gmail.comSigned-off-by:
Wei Yang <richard.weiyang@gmail.com> Reviewed-by:
-
Alexander Duyck authored
The ZONE_DEVICE pages were being initialized in two locations. One was with the memory_hotplug lock held and another was outside of that lock. The problem with this is that it was nearly doubling the memory initialization time. Instead of doing this twice, once while holding a global lock and once without, I am opting to defer the initialization to the one outside of the lock. This allows us to avoid serializing the overhead for memory init and we can instead focus on per-node init times. One issue I encountered is that devm_memremap_pages and hmm_devmmem_pages_create were initializing only the pgmap field the same way. One wasn't initializing hmm_data, and the other was initializing it to a poison value. Since this is something that is exposed to the driver in the case of hmm I am opting for a third option and just initializing hmm_data to 0 since this is going to be exposed to unknown third party drivers. [alexander.h.duyck@linux.intel.com: fix reference count for pgmap in devm_memremap_pages] Link: http://lkml.kernel.org/r/20181008233404.1909.37302.stgit@localhost.localdomain Link: http://lkml.kernel.org/r/20180925202053.3576.66039.stgit@localhost.localdomainSigned-off-by:
Alexander Duyck <alexander.h.duyck@linux.intel.com> Reviewed-by:
Pavel Tatashin <pavel.tatashin@microsoft.com> Tested-by:
Dan Williams <dan.j.williams@intel.com> Cc: Dave Hansen <dave.hansen@intel.com> Cc: Michal Hocko <mhocko@suse.com> Signed-off-by:
-
Alexander Duyck authored
It doesn't make much sense to use the atomic SetPageReserved at init time when we are using memset to clear the memory and manipulating the page flags via simple "&=" and "|=" operations in __init_single_page. This patch adds a non-atomic version __SetPageReserved that can be used during page init and shows about a 10% improvement in initialization times on the systems I have available for testing. On those systems I saw initialization times drop from around 35 seconds to around 32 seconds to initialize a 3TB block of persistent memory. I believe the main advantage of this is that it allows for more compiler optimization as the __set_bit operation can be reordered whereas the atomic version cannot. I tried adding a bit of documentation based on f1dd2cd1 ("mm, memory_hotplug: do not associate hotadded memory to zones until online"). Ideally the reserved flag should be set earlier since there is a brief window where the page is initialization via __init_single_page and we have not set the PG_Reserved flag. I'm leaving that for a future patch set as that will require a more significant refactor. Link: http://lkml.kernel.org/r/20180925202018.3576.11607.stgit@localhost.localdomainSigned-off-by:
Michal Hocko <mhocko@suse.com> Cc: Dave Hansen <dave.hansen@intel.com> Signed-off-by:
-
Alexander Duyck authored
Patch series "Address issues slowing persistent memory initialization", v5. The main thing this patch set achieves is that it allows us to initialize each node worth of persistent memory independently. As a result we reduce page init time by about 2 minutes because instead of taking 30 to 40 seconds per node and going through each node one at a time, we process all 4 nodes in parallel in the case of a 12TB persistent memory setup spread evenly over 4 nodes. This patch (of 3): On systems with a large amount of memory it can take a significant amount of time to initialize all of the page structs with the PAGE_POISON_PATTERN value. I have seen it take over 2 minutes to initialize a system with over 12TB of RAM. In order to work around the issue I had to disable CONFIG_DEBUG_VM and then the boot time returned to something much more reasonable as the arch_add_memory call completed in milliseconds versus seconds. However in doing that I had to disable all of the other VM debugging on the system. In order to work around a kernel that might have CONFIG_DEBUG_VM enabled on a system that has a large amount of memory I have added a new kernel parameter named "vm_debug" that can be set to "-" in order to disable it. Link: http://lkml.kernel.org/r/20180925201921.3576.84239.stgit@localhost.localdomainReviewed-by:
-
Shakeel Butt authored
The flag memcg_kmem_skip_account was added during the era of opt-out kmem accounting. There is no need for such flag in the opt-in world as there aren't any __GFP_ACCOUNT allocations within memcg_create_cache_enqueue(). Link: http://lkml.kernel.org/r/20180919004501.178023-1-shakeelb@google.comSigned-off-by:
Shakeel Butt <shakeelb@google.com> Acked-by:
Johannes Weiner <hannes@cmpxchg.org> Cc: Michal Hocko <mhocko@kernel.org> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Cc: Greg Thelen <gthelen@google.com> Signed-off-by:
-
Oscar Salvador authored
This patch, as the previous one, gets rid of the wrong if statements. While at it, I realized that the comments are sometimes very confusing, to say the least, and wrong. For example: ___ zone_last = ZONE_MOVABLE; /* * check whether node_states[N_HIGH_MEMORY] will be changed * If we try to offline the last present @nr_pages from the node, * we can determind we will need to clear the node from * node_states[N_HIGH_MEMORY]. */ for (; zt <= zone_last; zt++) present_pages += pgdat->node_zones[zt].present_pages; if (nr_pages >= present_pages) arg->status_change_nid = zone_to_nid(zone); else arg->status_change_nid = -1; ___ In case the node gets empry, it must be removed from N_MEMORY. We already check N_HIGH_MEMORY a bit above within the CONFIG_HIGHMEM ifdef code. Not to say that status_change_nid is for N_MEMORY, and not for N_HIGH_MEMORY. So I re-wrote some of the comments to what I think is better. [osalvador@suse.de: address feedback from Pavel] Link: http://lkml.kernel.org/r/20180921132634.10103-5-osalvador@techadventures.net Link: http://lkml.kernel.org/r/20180919100819.25518-6-osalvador@techadventures.netSigned-off-by:Oscar Salvador <osalvador@suse.de> Reviewed-by:
-
Oscar Salvador authored
While looking at node_states_check_changes_online, I stumbled upon some confusing things. Right after entering the function, we find this: if (N_MEMORY == N_NORMAL_MEMORY) zone_last = ZONE_MOVABLE; This is wrong. N_MEMORY cannot really be equal to N_NORMAL_MEMORY. My guess is that this wanted to be something like: if (N_NORMAL_MEMORY == N_HIGH_MEMORY) to check if we have CONFIG_HIGHMEM. Later on, in the CONFIG_HIGHMEM block, we have: if (N_MEMORY == N_HIGH_MEMORY) zone_last = ZONE_MOVABLE; Again, this is wrong, and will never be evaluated to true. Besides removing these wrong if statements, I simplified the function a bit. [osalvador@suse.de: address feedback from Pavel] Link: http://lkml.kernel.org/r/20180921132634.10103-4-osalvador@techadventures.net Link: http://lkml.kernel.org/r/20180919100819.25518-5-osalvador@techadventures.netSigned-off-by: -
Oscar Salvador authored
node_states_clear has the following if statements: if ((N_MEMORY != N_NORMAL_MEMORY) && (arg->status_change_nid_high >= 0)) ... if ((N_MEMORY != N_HIGH_MEMORY) && (arg->status_change_nid >= 0)) ... N_MEMORY can never be equal to neither N_NORMAL_MEMORY nor N_HIGH_MEMORY. Similar problem was found in [1]. Since this is wrong, let us get rid of it. [1] https://patchwork.kernel.org/patch/10579155/ Link: http://lkml.kernel.org/r/20180919100819.25518-4-osalvador@techadventures.netSigned-off-by: -
Oscar Salvador authored
In node_states_check_changes_online, we check if the node will have to be set for any of the N_*_MEMORY states after the pages have been onlined. Later on, we perform the activation in node_states_set_node. Currently, in node_states_set_node we set the node to N_MEMORY unconditionally. This means that we call node_set_state for N_MEMORY every time pages go online, but we only need to do it if the node has not yet been set for N_MEMORY. Fix this by checking status_change_nid. Link: http://lkml.kernel.org/r/20180919100819.25518-2-osalvador@techadventures.netSigned-off-by:
-
haiqing.shq authored
Use the variable write_len instead of ov_iter_count(from). Link: http://lkml.kernel.org/r/1537375855-2088-1-git-send-email-leviathan0992@gmail.comSigned-off-by:
haiqing.shq <leviathan0992@gmail.com> Reviewed-by:
-
Yang Shi authored
When unmapping VM_PFNMAP mappings, vm flags need to be updated. Since the vmas have been detached, so it sounds safe to update vm flags with read mmap_sem. Link: http://lkml.kernel.org/r/1537376621-51150-4-git-send-email-yang.shi@linux.alibaba.comSigned-off-by:
Yang Shi <yang.shi@linux.alibaba.com> Reviewed-by:
Matthew Wilcox <willy@infradead.org> Acked-by:
Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by:
Vlastimil Babka <vbabka@suse.cz> Cc: Michal Hocko <mhocko@kernel.org> Cc: Vlastimil Babka <vbabka@suse.cz> Signed-off-by:
-
Yang Shi authored
When unmapping VM_HUGETLB mappings, vm flags need to be updated. Since the vmas have been detached, so it sounds safe to update vm flags with read mmap_sem. Link: http://lkml.kernel.org/r/1537376621-51150-3-git-send-email-yang.shi@linux.alibaba.comSigned-off-by:
-
Yang Shi authored
Patch series "mm: zap pages with read mmap_sem in munmap for large mapping", v11. Background: Recently, when we ran some vm scalability tests on machines with large memory, we ran into a couple of mmap_sem scalability issues when unmapping large memory space, please refer to https://lkml.org/lkml/2017/12/14/733 and https://lkml.org/lkml/2018/2/20/576. History: Then akpm suggested to unmap large mapping section by section and drop mmap_sem at a time to mitigate it (see https://lkml.org/lkml/2018/3/6/784). V1 patch series was submitted to the mailing list per Andrew's suggestion (see https://lkml.org/lkml/2018/3/20/786). Then I received a lot great feedback and suggestions. Then this topic was discussed on LSFMM summit 2018. In the summit, Michal Hocko suggested (also in the v1 patches review) to try "two phases" approach. Zapping pages with read mmap_sem, then doing via cleanup with write mmap_sem (for discussion detail, see https://lwn.net/Articles/753269/) Approach: Zapping pages is the most time consuming part, according to the suggestion from Michal Hocko [1], zapping pages can be done with holding read mmap_sem, like what MADV_DONTNEED does. Then re-acquire write mmap_sem to cleanup vmas. But, we can't call MADV_DONTNEED directly, since there are two major drawbacks: * The unexpected state from PF if it wins the race in the middle of munmap. It may return zero page, instead of the content or SIGSEGV. * Can't handle VM_LOCKED | VM_HUGETLB | VM_PFNMAP and uprobe mappings, which is a showstopper from akpm But, some part may need write mmap_sem, for example, vma splitting. So, the design is as follows: acquire write mmap_sem lookup vmas (find and split vmas) deal with special mappings detach vmas downgrade_write zap pages free page tables release mmap_sem The vm events with read mmap_sem may come in during page zapping, but since vmas have been detached before, they, i.e. page fault, gup, etc, will not be able to find valid vma, then just return SIGSEGV or -EFAULT as expected. If the vma has VM_HUGETLB | VM_PFNMAP, they are considered as special mappings. They will be handled by falling back to regular do_munmap() with exclusive mmap_sem held in this patch since they may update vm flags. But, with the "detach vmas first" approach, the vmas have been detached when vm flags are updated, so it sounds safe to update vm flags with read mmap_sem for this specific case. So, VM_HUGETLB and VM_PFNMAP will be handled by using the optimized path in the following separate patches for bisectable sake. Unmapping uprobe areas may need update mm flags (MMF_RECALC_UPROBES). However it is fine to have false-positive MMF_RECALC_UPROBES according to uprobes developer. So, uprobe unmap will not be handled by the regular path. With the "detach vmas first" approach we don't have to re-acquire mmap_sem again to clean up vmas to avoid race window which might get the address space changed since downgrade_write() doesn't release the lock to lead regression, which simply downgrades to read lock. And, since the lock acquire/release cost is managed to the minimum and almost as same as before, the optimization could be extended to any size of mapping without incurring significant penalty to small mappings. For the time being, just do this in munmap syscall path. Other vm_munmap() or do_munmap() call sites (i.e mmap, mremap, etc) remain intact due to some implementation difficulties since they acquire write mmap_sem from very beginning and hold it until the end, do_munmap() might be called in the middle. But, the optimized do_munmap would like to be called without mmap_sem held so that we can do the optimization. So, if we want to do the similar optimization for mmap/mremap path, I'm afraid we would have to redesign them. mremap might be called on very large area depending on the usecases, the optimization to it will be considered in the future. This patch (of 3): When running some mmap/munmap scalability tests with large memory (i.e. > 300GB), the below hung task issue may happen occasionally. INFO: task ps:14018 blocked for more than 120 seconds. Tainted: G E 4.9.79-009.ali3000.alios7.x86_64 #1 "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. ps D 0 14018 1 0x00000004 ffff885582f84000 ffff885e8682f000 ffff880972943000 ffff885ebf499bc0 ffff8828ee120000 ffffc900349bfca8 ffffffff817154d0 0000000000000040 00ffffff812f872a ffff885ebf499bc0 024000d000948300 ffff880972943000 Call Trace: [<ffffffff817154d0>] ? __schedule+0x250/0x730 [<ffffffff817159e6>] schedule+0x36/0x80 [<ffffffff81718560>] rwsem_down_read_failed+0xf0/0x150 [<ffffffff81390a28>] call_rwsem_down_read_failed+0x18/0x30 [<ffffffff81717db0>] down_read+0x20/0x40 [<ffffffff812b9439>] proc_pid_cmdline_read+0xd9/0x4e0 [<ffffffff81253c95>] ? do_filp_open+0xa5/0x100 [<ffffffff81241d87>] __vfs_read+0x37/0x150 [<ffffffff812f824b>] ? security_file_permission+0x9b/0xc0 [<ffffffff81242266>] vfs_read+0x96/0x130 [<ffffffff812437b5>] SyS_read+0x55/0xc0 [<ffffffff8171a6da>] entry_SYSCALL_64_fastpath+0x1a/0xc5 It is because munmap holds mmap_sem exclusively from very beginning to all the way down to the end, and doesn't release it in the middle. When unmapping large mapping, it may take long time (take ~18 seconds to unmap 320GB mapping with every single page mapped on an idle machine). Zapping pages is the most time consuming part, according to the suggestion from Michal Hocko [1], zapping pages can be done with holding read mmap_sem, like what MADV_DONTNEED does. Then re-acquire write mmap_sem to cleanup vmas. But, some part may need write mmap_sem, for example, vma splitting. So, the design is as follows: acquire write mmap_sem lookup vmas (find and split vmas) deal with special mappings detach vmas downgrade_write zap pages free page tables release mmap_sem The vm events with read mmap_sem may come in during page zapping, but since vmas have been detached before, they, i.e. page fault, gup, etc, will not be able to find valid vma, then just return SIGSEGV or -EFAULT as expected. If the vma has VM_HUGETLB | VM_PFNMAP, they are considered as special mappings. They will be handled by without downgrading mmap_sem in this patch since they may update vm flags. But, with the "detach vmas first" approach, the vmas have been detached when vm flags are updated, so it sounds safe to update vm flags with read mmap_sem for this specific case. So, VM_HUGETLB and VM_PFNMAP will be handled by using the optimized path in the following separate patches for bisectable sake. Unmapping uprobe areas may need update mm flags (MMF_RECALC_UPROBES). However it is fine to have false-positive MMF_RECALC_UPROBES according to uprobes developer. With the "detach vmas first" approach we don't have to re-acquire mmap_sem again to clean up vmas to avoid race window which might get the address space changed since downgrade_write() doesn't release the lock to lead regression, which simply downgrades to read lock. And, since the lock acquire/release cost is managed to the minimum and almost as same as before, the optimization could be extended to any size of mapping without incurring significant penalty to small mappings. For the time being, just do this in munmap syscall path. Other vm_munmap() or do_munmap() call sites (i.e mmap, mremap, etc) remain intact due to some implementation difficulties since they acquire write mmap_sem from very beginning and hold it until the end, do_munmap() might be called in the middle. But, the optimized do_munmap would like to be called without mmap_sem held so that we can do the optimization. So, if we want to do the similar optimization for mmap/mremap path, I'm afraid we would have to redesign them. mremap might be called on very large area depending on the usecases, the optimization to it will be considered in the future. With the patches, exclusive mmap_sem hold time when munmap a 80GB address space on a machine with 32 cores of E5-2680 @ 2.70GHz dropped to us level from second. munmap_test-15002 [008] 594.380138: funcgraph_entry: | __vm_munmap() { munmap_test-15002 [008] 594.380146: funcgraph_entry: !2485684 us | unmap_region(); munmap_test-15002 [008] 596.865836: funcgraph_exit: !2485692 us | } Here the execution time of unmap_region() is used to evaluate the time of holding read mmap_sem, then the remaining time is used with holding exclusive lock. [1] https://lwn.net/Articles/753269/ Link: http://lkml.kernel.org/r/1537376621-51150-2-git-send-email-yang.shi@linux.alibaba.comSigned-off-by:
Yang Shi <yang.shi@linux.alibaba.com>Suggested-by: Michal Hocko <mhocko@kernel.org> Suggested-by:
Kirill A. Shutemov <kirill@shutemov.name> Suggested-by:
-
Andrey Ryabinin authored
Add might_sleep() call to vfree() to catch potential sleep-in-atomic bugs earlier. [aryabinin@virtuozzo.com: drop might_sleep_if() from kvfree()] Link: http://lkml.kernel.org/r/7e19e4df-b1a6-29bd-9ae7-0266d50bef1d@virtuozzo.com Link: http://lkml.kernel.org/r/20180914130512.10394-3-aryabinin@virtuozzo.comSigned-off-by:
Andrey Ryabinin <aryabinin@virtuozzo.com> Reviewed-by:
-
Andrey Ryabinin authored
vfree() might sleep if called not in interrupt context. Explain that in the comment. Link: http://lkml.kernel.org/r/20180914130512.10394-2-aryabinin@virtuozzo.comSigned-off-by:
-
Andrey Ryabinin authored
vfree() might sleep if called not in interrupt context. So does kvfree() too. Fix misleading kvfree()'s comment about allowed context. Link: http://lkml.kernel.org/r/20180914130512.10394-1-aryabinin@virtuozzo.com Fixes: 04b8e946 ("mm/util.c: improve kvfree() kerneldoc") Signed-off-by:
-
zhong jiang authored
match_string() returns the index of an array for a matching string, which can be used intead of open coded implementation. Link: http://lkml.kernel.org/r/1536988365-50310-1-git-send-email-zhongjiang@huawei.comSigned-off-by:
zhong jiang <zhongjiang@huawei.com> Reviewed-by:
-
YueHaibing authored
Remove duplicated include linux/memremap.h Link: http://lkml.kernel.org/r/20180917131308.16420-1-yuehaibing@huawei.comSigned-off-by:
YueHaibing <yuehaibing@huawei.com> Reviewed-by:
-
Michal Hocko authored
should_suppress_show_mem() was introduced to reduce the overhead of show_mem on large NUMA systems. Things have changed since then though. Namely c78e9363 ("mm: do not walk all of system memory during show_mem") has reduced the overhead considerably. Moreover warn_alloc_show_mem clears SHOW_MEM_FILTER_NODES when called from the IRQ context already so we are not printing per node stats. Remove should_suppress_show_mem because we are losing potentially interesting information about allocation failures. We have seen a bug report where system gets unresponsive under memory pressure and there is only kernel: [2032243.696888] qlge 0000:8b:00.1 ql1: Could not get a page chunk, i=8, clean_idx =200 . kernel: [2032243.710725] swapper/7: page allocation failure: order:1, mode:0x1084120(GFP_ATOMIC|__GFP_COLD|__GFP_COMP) without an additional information for debugging. It would be great to see the state of the page allocator at the moment. Link: http://lkml.kernel.org/r/20180907114334.7088-1-mhocko@kernel.orgSigned-off-by:
-
Johannes Weiner authored
The refault stats go better with the page fault stats, and are of higher interest than the stats on LRU operations. In fact they used to be grouped together; when the LRU operation stats were added later on, they were wedged in between. Move them back together. Documentation/admin-guide/cgroup-v2.rst already lists them in the right order. Link: http://lkml.kernel.org/r/20181010140239.GA2527@cmpxchg.orgSigned-off-by:
-
Johannes Weiner authored
The page cache and most shrinkable slab caches hold data that has been read from disk, but there are some caches that only cache CPU work, such as the dentry and inode caches of procfs and sysfs, as well as the subset of radix tree nodes that track non-resident page cache. Currently, all these are shrunk at the same rate: using DEFAULT_SEEKS for the shrinker's seeks setting tells the reclaim algorithm that for every two page cache pages scanned it should scan one slab object. This is a bogus setting. A virtual inode that required no IO to create is not twice as valuable as a page cache page; shadow cache entries with eviction distances beyond the size of memory aren't either. In most cases, the behavior in practice is still fine. Such virtual caches don't tend to grow and assert themselves aggressively, and usually get picked up before they cause problems. But there are scenarios where that's not true. Our database workloads suffer from two of those. For one, their file workingset is several times bigger than available memory, which has the kernel aggressively create shadow page cache entries for the non-resident parts of it. The workingset code does tell the VM that most of these are expendable, but the VM ends up balancing them 2:1 to cache pages as per the seeks setting. This is a huge waste of memory. These workloads also deal with tens of thousands of open files and use /proc for introspection, which ends up growing the proc_inode_cache to absurdly large sizes - again at the cost of valuable cache space, which isn't a reasonable trade-off, given that proc inodes can be re-created without involving the disk. This patch implements a "zero-seek" setting for shrinkers that results in a target ratio of 0:1 between their objects and IO-backed caches. This allows such virtual caches to grow when memory is available (they do cache/avoid CPU work after all), but effectively disables them as soon as IO-backed objects are under pressure. It then switches the shrinkers for procfs and sysfs metadata, as well as excess page cache shadow nodes, to the new zero-seek setting. Link: http://lkml.kernel.org/r/20181009184732.762-5-hannes@cmpxchg.orgSigned-off-by:
Domas Mituzas <dmituzas@fb.com> Reviewed-by:
Rik van Riel <riel@surriel.com> Acked-by:
Peter Zijlstra (Intel) <peterz@infradead.org> Signed-off-by:
-
Johannes Weiner authored
Make it easier to catch bugs in the shadow node shrinker by adding a counter for the shadow nodes in circulation. [akpm@linux-foundation.org: assert that irqs are disabled, for __inc_lruvec_page_state()] [akpm@linux-foundation.org: s/WARN_ON_ONCE/VM_WARN_ON_ONCE/, per Johannes] Link: http://lkml.kernel.org/r/20181009184732.762-4-hannes@cmpxchg.orgSigned-off-by:
-
Johannes Weiner authored
No need to use the preemption-safe lruvec state function inside the reclaim region that has irqs disabled. Link: http://lkml.kernel.org/r/20181009184732.762-3-hannes@cmpxchg.orgSigned-off-by:
-
Johannes Weiner authored
On a system that executes multiple cgrouped jobs and independent workloads, we don't just care about the health of the overall system, but also that of individual jobs, so that we can ensure individual job health, fairness between jobs, or prioritize some jobs over others. This patch implements pressure stall tracking for cgroups. In kernels with CONFIG_PSI=y, cgroup2 groups will have cpu.pressure, memory.pressure, and io.pressure files that track aggregate pressure stall times for only the tasks inside the cgroup. Link: http://lkml.kernel.org/r/20180828172258.3185-10-hannes@cmpxchg.orgSigned-off-by:
Tejun Heo <tj@kernel.org> Acked-by:
Daniel Drake <drake@endlessm.com> Tested-by:
Suren Baghdasaryan <surenb@google.com> Cc: Christopher Lameter <cl@linux.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Johannes Weiner <jweiner@fb.com> Cc: Mike Galbraith <efault@gmx.de> Cc: Peter Enderborg <peter.enderborg@sony.com> Cc: Randy Dunlap <rdunlap@infradead.org> Cc: Shakeel Butt <shakeelb@google.com> Cc: Vinayak Menon <vinmenon@codeaurora.org> Signed-off-by:
-
Johannes Weiner authored
When systems are overcommitted and resources become contended, it's hard to tell exactly the impact this has on workload productivity, or how close the system is to lockups and OOM kills. In particular, when machines work multiple jobs concurrently, the impact of overcommit in terms of latency and throughput on the individual job can be enormous. In order to maximize hardware utilization without sacrificing individual job health or risk complete machine lockups, this patch implements a way to quantify resource pressure in the system. A kernel built with CONFIG_PSI=y creates files in /proc/pressure/ that expose the percentage of time the system is stalled on CPU, memory, or IO, respectively. Stall states are aggregate versions of the per-task delay accounting delays: cpu: some tasks are runnable but not executing on a CPU memory: tasks are reclaiming, or waiting for swapin or thrashing cache io: tasks are waiting for io completions These percentages of walltime can be thought of as pressure percentages, and they give a general sense of system health and productivity loss incurred by resource overcommit. They can also indicate when the system is approaching lockup scenarios and OOMs. To do this, psi keeps track of the task states associated with each CPU and samples the time they spend in stall states. Every 2 seconds, the samples are averaged across CPUs - weighted by the CPUs' non-idle time to eliminate artifacts from unused CPUs - and translated into percentages of walltime. A running average of those percentages is maintained over 10s, 1m, and 5m periods (similar to the loadaverage). [hannes@cmpxchg.org: doc fixlet, per Randy] Link: http://lkml.kernel.org/r/20180828205625.GA14030@cmpxchg.org [hannes@cmpxchg.org: code optimization] Link: http://lkml.kernel.org/r/20180907175015.GA8479@cmpxchg.org [hannes@cmpxchg.org: rename psi_clock() to psi_update_work(), per Peter] Link: http://lkml.kernel.org/r/20180907145404.GB11088@cmpxchg.org [hannes@cmpxchg.org: fix build] Link: http://lkml.kernel.org/r/20180913014222.GA2370@cmpxchg.org Link: http://lkml.kernel.org/r/20180828172258.3185-9-hannes@cmpxchg.orgSigned-off-by: -
Johannes Weiner authored
do_sched_yield() disables IRQs, looks up this_rq() and locks it. The next patch is adding another site with the same pattern, so provide a convenience function for it. Link: http://lkml.kernel.org/r/20180828172258.3185-8-hannes@cmpxchg.orgSigned-off-by:
-
Johannes Weiner authored
kernel/sched/sched.h includes "stats.h" half-way through the file. The next patch introduces users of sched.h's rq locking functions and update_rq_clock() in kernel/sched/stats.h. Move those definitions up in the file so they are available in stats.h. Link: http://lkml.kernel.org/r/20180828172258.3185-7-hannes@cmpxchg.orgSigned-off-by:
-
Johannes Weiner authored
It's going to be used in a later patch. Keep the churn separate. Link: http://lkml.kernel.org/r/20180828172258.3185-6-hannes@cmpxchg.orgSigned-off-by:
-
Johannes Weiner authored
There are several definitions of those functions/macros in places that mess with fixed-point load averages. Provide an official version. [akpm@linux-foundation.org: fix missed conversion in block/blk-iolatency.c] Link: http://lkml.kernel.org/r/20180828172258.3185-5-hannes@cmpxchg.orgSigned-off-by:
-
Johannes Weiner authored
Delay accounting already measures the time a task spends in direct reclaim and waiting for swapin, but in low memory situations tasks spend can spend a significant amount of their time waiting on thrashing page cache. This isn't tracked right now. To know the full impact of memory contention on an individual task, measure the delay when waiting for a recently evicted active cache page to read back into memory. Also update tools/accounting/getdelays.c: [hannes@computer accounting]$ sudo ./getdelays -d -p 1 print delayacct stats ON PID 1 CPU count real total virtual total delay total delay average 50318 745000000 847346785 400533713 0.008ms IO count delay total delay average 435 122601218 0ms SWAP count delay total delay average 0 0 0ms RECLAIM count delay total delay average 0 0 0ms THRASHING count delay total delay average 19 12621439 0ms Link: http://lkml.kernel.org/r/20180828172258.3185-4-hannes@cmpxchg.orgSigned-off-by: -
Johannes Weiner authored
Refaults happen during transitions between workingsets as well as in-place thrashing. Knowing the difference between the two has a range of applications, including measuring the impact of memory shortage on the system performance, as well as the ability to smarter balance pressure between the filesystem cache and the swap-backed workingset. During workingset transitions, inactive cache refaults and pushes out established active cache. When that active cache isn't stale, however, and also ends up refaulting, that's bonafide thrashing. Introduce a new page flag that tells on eviction whether the page has been active or not in its lifetime. This bit is then stored in the shadow entry, to classify refaults as transitioning or thrashing. How many page->flags does this leave us with on 32-bit? 20 bits are always page flags 21 if you have an MMU 23 with the zone bits for DMA, Normal, HighMem, Movable 29 with the sparsemem section bits 30 if PAE is enabled 31 with this patch. So on 32-bit PAE, that leaves 1 bit for distinguishing two NUMA nodes. If that's not enough, the system can switch to discontigmem and re-gain the 6 or 7 sparsemem section bits. Link: http://lkml.kernel.org/r/20180828172258.3185-3-hannes@cmpxchg.orgSigned-off-by:
-
Johannes Weiner authored
Patch series "psi: pressure stall information for CPU, memory, and IO", v4. Overview PSI reports the overall wallclock time in which the tasks in a system (or cgroup) wait for (contended) hardware resources. This helps users understand the resource pressure their workloads are under, which allows them to rootcause and fix throughput and latency problems caused by overcommitting, underprovisioning, suboptimal job placement in a grid; as well as anticipate major disruptions like OOM. Real-world applications We're using the data collected by PSI (and its previous incarnation, memdelay) quite extensively at Facebook, and with several success stories. One usecase is avoiding OOM hangs/livelocks. The reason these happen is because the OOM killer is triggered by reclaim not being able to free pages, but with fast flash devices there is *always* some clean and uptodate cache to reclaim; the OOM killer never kicks in, even as tasks spend 90% of the time thrashing the cache pages of their own executables. There is no situation where this ever makes sense in practice. We wrote a <100 line POC python script to monitor memory pressure and kill stuff way before such pathological thrashing leads to full system losses that would require forcible hard resets. We've since extended and deployed this code into other places to guarantee latency and throughput SLAs, since they're usually violated way before the kernel OOM killer would ever kick in. It is available here: https://github.com/facebookincubator/oomd Eventually we probably want to trigger the in-kernel OOM killer based on extreme sustained pressure as well, so that Linux can avoid memory livelocks - which technically aren't deadlocks, but to the user indistinguishable from them - out of the box. We'd continue using OOMD as the first line of defense to ensure workload health and implement complex kill policies that are beyond the scope of the kernel. We also use PSI memory pressure for loadshedding. Our batch job infrastructure used to use heuristics based on various VM stats to anticipate OOM situations, with lackluster success. We switched it to PSI and managed to anticipate and avoid OOM kills and lockups fairly reliably. The reduction of OOM outages in the worker pool raised the pool's aggregate productivity, and we were able to switch that service to smaller machines. Lastly, we use cgroups to isolate a machine's main workload from maintenance crap like package upgrades, logging, configuration, as well as to prevent multiple workloads on a machine from stepping on each others' toes. We were not able to configure this properly without the pressure metrics; we would see latency or bandwidth drops, but it would often be hard to impossible to rootcause it post-mortem. We now log and graph pressure for the containers in our fleet and can trivially link latency spikes and throughput drops to shortages of specific resources after the fact, and fix the job config/scheduling. PSI has also received testing, feedback, and feature requests from Android and EndlessOS for the purpose of low-latency OOM killing, to intervene in pressure situations before the UI starts hanging. How do you use this feature? A kernel with CONFIG_PSI=y will create a /proc/pressure directory with 3 files: cpu, memory, and io. If using cgroup2, cgroups will also have cpu.pressure, memory.pressure and io.pressure files, which simply aggregate task stalls at the cgroup level instead of system-wide. The cpu file contains one line: some avg10=2.04 avg60=0.75 avg300=0.40 total=157656722 The averages give the percentage of walltime in which one or more tasks are delayed on the runqueue while another task has the CPU. They're recent averages over 10s, 1m, 5m windows, so you can tell short term trends from long term ones, similarly to the load average. The total= value gives the absolute stall time in microseconds. This allows detecting latency spikes that might be too short to sway the running averages. It also allows custom time averaging in case the 10s/1m/5m windows aren't adequate for the usecase (or are too coarse with future hardware). What to make of this "some" metric? If CPU utilization is at 100% and CPU pressure is 0, it means the system is perfectly utilized, with one runnable thread per CPU and nobody waiting. At two or more runnable tasks per CPU, the system is 100% overcommitted and the pressure average will indicate as much. From a utilization perspective this is a great state of course: no CPU cycles are being wasted, even when 50% of the threads were to go idle (as most workloads do vary). From the perspective of the individual job it's not great, however, and they would do better with more resources. Depending on what your priority and options are, raised "some" numbers may or may not require action. The memory file contains two lines: some avg10=70.24 avg60=68.52 avg300=69.91 total=3559632828 full avg10=57.59 avg60=58.06 avg300=60.38 total=3300487258 The some line is the same as for cpu, the time in which at least one task is stalled on the resource. In the case of memory, this includes waiting on swap-in, page cache refaults and page reclaim. The full line, however, indicates time in which *nobody* is using the CPU productively due to pressure: all non-idle tasks are waiting for memory in one form or another. Significant time spent in there is a good trigger for killing things, moving jobs to other machines, or dropping incoming requests, since neither the jobs nor the machine overall are making too much headway. The io file is similar to memory. Because the block layer doesn't have a concept of hardware contention right now (how much longer is my IO request taking due to other tasks?), it reports CPU potential lost on all IO delays, not just the potential lost due to competition. FAQ Q: How is PSI's CPU component different from the load average? A: There are several quirks in the load average that make it hard to impossible to tell how overcommitted the CPU really is. 1. The load average is reported as a raw number of active tasks. You need to know how many CPUs there are in the system, how many CPUs the workload is allowed to use, then think about what the proportion between load and the number of CPUs mean for the tasks trying to run. PSI reports the percentage of wallclock time in which tasks are waiting for a CPU to run on. It doesn't matter how many CPUs are present or usable. The number always tells the quality of life of tasks in the system or in a particular cgroup. 2. The shortest averaging window is 1m, which is extremely coarse, and it's sampled in 5s intervals. A *lot* can happen on a CPU in 5 seconds. This *may* be able to identify persistent long-term trends and very clear and obvious overloads, but it's unusable for latency spikes and more subtle overutilization. PSI's shortest window is 10s. It also exports the cumulative stall times (in microseconds) of synchronously recorded events. 3. On Linux, the load average for historical reasons includes all TASK_UNINTERRUPTIBLE tasks. This gives a broader sense of how busy the system is, but on the flipside it doesn't distinguish whether tasks are likely to contend over the CPU or IO - which obviously requires very different interventions from a sys admin or a job scheduler. PSI reports independent metrics for CPU and IO. You can tell which resource is making the tasks wait, but in conjunction still see how overloaded the system is overall. Q: What's the cost / performance impact of this feature? A: PSI's primary cost is in the scheduler, in particular task wakeups and sleeps. I benchmarked this code using Facebook's two most scheduling sensitive workloads: memcache and webserver. They handle a ton of small requests - lots of wakeups and sleeps with little actual work in between - so they tend to be canaries for scheduler regressions. In the tests, the boxes were handling live traffic over the course of several hours. Half the machines, the control, ran with CONFIG_PSI=n. For memcache I used eight machines total. They're 2-socket, 14 core, 56 thread boxes. The test runs for half the test period, flips the test and control kernels on the hardware to rule out HW factors, DC location etc., then runs the other half of the test. For the webservers, I used 32 machines total. They're single socket, 16 core, 32 thread machines. During the memcache test, CPU load was nopsi=78.05% psi=78.98% in the first half and nopsi=77.52% psi=78.25%, so PSI added between 0.7 and 0.9 percentage points to the CPU load, a difference of about 1%. UPDATE: I re-ran this test with the v3 version of this patch set and the CPU utilization was equivalent between test and control. UPDATE: v4 is on par with v3. As far as end-to-end request latency from the client perspective goes, we don't sample those finely enough to capture the requests going to those particular machines during the test, but we know the p50 turnaround time in this workload is 54us, and perf bench sched pipe on those machines show nopsi=5.232666 us/op and psi=5.587347 us/op, so this doesn't add much here either. The profile for the pipe benchmark shows: 0.87% sched-pipe [kernel.vmlinux] [k] psi_group_change 0.83% perf.real [kernel.vmlinux] [k] psi_group_change 0.82% perf.real [kernel.vmlinux] [k] psi_task_change 0.58% sched-pipe [kernel.vmlinux] [k] psi_task_change The webserver load is running inside 4 nested cgroup levels. The CPU load with both nopsi and psi kernels was indistinguishable at 81%. For comparison, we had to disable the cgroup cpu controller on the webservers because it added 4 percentage points to the CPU% during this same exact test. Versions of this accounting code now run on 80% of our fleet. None of our workloads have reported regressions during the rollout. Daniel Drake said: : I just retested the latest version at : http://git.cmpxchg.org/cgit.cgi/linux-psi.git (Linux 4.18) and the results : are great. : : Test setup: : Endless OS : GeminiLake N4200 low end laptop : 2GB RAM : swap (and zram swap) disabled : : Baseline test: open a handful of large-ish apps and several website : tabs in Google Chrome. : : Results: after a couple of minutes, system is excessively thrashing, mouse : cursor can barely be moved, UI is not responding to mouse clicks, so it's : impractical to recover from this situation as an ordinary user : : Add my simple killer: : https://gist.github.com/dsd/a8988bf0b81a6163475988120fe8d9cd : : Results: when the thrashing causes the UI to become sluggish, the killer : steps in and kills something (usually a chrome tab), and the system : remains usable. I repeatedly opened more apps and more websites over a 15 : minute period but I wasn't able to get the system to a point of UI : unresponsiveness. Suren said: : Backported to 4.9 and retested on ARMv8 8 code system running Android. : Signals behave as expected reacting to memory pressure, no jumps in : "total" counters that would indicate an overflow/underflow issues. Nicely : done! This patch (of 9): If we keep just enough refault information to match the *current* page cache during reclaim time, we could lose a lot of events when there is only a temporary spike in non-cache memory consumption that pushes out all the cache. Once cache comes back, we won't see those refaults. They might not be actionable for LRU aging, but we want to know about them for measuring memory pressure. [hannes@cmpxchg.org: switch to NUMA-aware lru and slab counters] Link: http://lkml.kernel.org/r/20181009184732.762-2-hannes@cmpxchg.org Link: http://lkml.kernel.org/r/20180828172258.3185-2-hannes@cmpxchg.orgSigned-off-by:
Johannes Weiner <jweiner@fb.com> Acked-by:
-
Vlastimil Babka authored
Kmalloc cache names can get quite long for large object sizes, when the sizes are expressed in bytes. Use 'k' and 'M' prefixes to make the names as short as possible e.g. in /proc/slabinfo. This works, as we mostly use power-of-two sizes, with exceptions only below 1k. Example: 'kmalloc-4194304' becomes 'kmalloc-4M' Link: http://lkml.kernel.org/r/20180731090649.16028-7-vbabka@suse.czSuggested-by:
Mel Gorman <mgorman@techsingularity.net> Acked-by:
Christoph Lameter <cl@linux.com> Acked-by:
Roman Gushchin <guro@fb.com> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Laura Abbott <labbott@redhat.com> Cc: Michal Hocko <mhocko@kernel.org> Cc: Sumit Semwal <sumit.semwal@linaro.org> Cc: Vijayanand Jitta <vjitta@codeaurora.org> Signed-off-by:
-
Vlastimil Babka authored
The vmstat NR_KERNEL_MISC_RECLAIMABLE counter is for kernel non-slab allocations that can be reclaimed via shrinker. In /proc/meminfo, we can show the sum of all reclaimable kernel allocations (including slab) as "KReclaimable". Add the same counter also to per-node meminfo under /sys With this counter, users will have more complete information about kernel memory usage. Non-slab reclaimable pages (currently just the ION allocator) will not be missing from /proc/meminfo, making users wonder where part of their memory went. More precisely, they already appear in MemAvailable, but without the new counter, it's not obvious why the value in MemAvailable doesn't fully correspond with the sum of other counters participating in it. Link: http://lkml.kernel.org/r/20180731090649.16028-6-vbabka@suse.czSigned-off-by:
-
Vlastimil Babka authored
The vmstat counter NR_INDIRECTLY_RECLAIMABLE_BYTES was introduced by commit eb592546 ("mm: introduce NR_INDIRECTLY_RECLAIMABLE_BYTES") with the goal of accounting objects that can be reclaimed, but cannot be allocated via a SLAB_RECLAIM_ACCOUNT cache. This is now possible via kmalloc() with __GFP_RECLAIMABLE flag, and the dcache external names user is converted. The counter is however still useful for accounting direct page allocations (i.e. not slab) with a shrinker, such as the ION page pool. So keep it, and: - change granularity to pages to be more like other counters; sub-page allocations should be able to use kmalloc - rename the counter to NR_KERNEL_MISC_RECLAIMABLE - expose the counter again in vmstat as "nr_kernel_misc_reclaimable"; we can again remove the check for not printing "hidden" counters Link: http://lkml.kernel.org/r/20180731090649.16028-5-vbabka@suse.czSigned-off-by:
-
Vlastimil Babka authored
We can use the newly introduced kmalloc-reclaimable-X caches, to allocate external names in dcache, which will take care of the proper accounting automatically, and also improve anti-fragmentation page grouping. This effectively reverts commit f1782c9b ("dcache: account external names as indirectly reclaimable memory") and instead passes __GFP_RECLAIMABLE to kmalloc(). The accounting thus moves from NR_INDIRECTLY_RECLAIMABLE_BYTES to NR_SLAB_RECLAIMABLE, which is also considered in MemAvailable calculation and overcommit decisions. Link: http://lkml.kernel.org/r/20180731090649.16028-4-vbabka@suse.czSigned-off-by:
-
