- 03 Sep, 2021 40 commits
-
-
Greg Kroah-Hartman authored
fault_in_pages_writeable() and fault_in_pages_readable() treat the size parameter as unsigned, doing pointer math with the value, so make this explicit and set it to be a size_t type which all callers currently treat it as anyway. This solves the issue where static checkers get nervous seeing pointer arithmetic happening with a signed value. Link: https://lkml.kernel.org/r/20210727111136.457638-1-gregkh@linuxfoundation.orgSigned-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org> Reported-by:
Jordy Zomer <jordy@pwning.systems> Cc: Matthew Wilcox <willy@infradead.org> Cc: David Howells <dhowells@redhat.com> Cc: William Kucharski <william.kucharski@oracle.com> Cc: "Darrick J. Wong" <djwong@kernel.org> Cc: Hugh Dickins <hughd@google.com> Cc: Mauro Carvalho Chehab <mchehab+huawei@kernel.org> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org>
-
Huang Ying authored
Before commit c5b5a3dd ("mm: thp: refactor NUMA fault handling"), the TLB flushing is done in do_huge_pmd_numa_page() itself via flush_tlb_range(). But after commit c5b5a3dd ("mm: thp: refactor NUMA fault handling"), the TLB flushing is done in migrate_pages() as in the following code path anyway. do_huge_pmd_numa_page migrate_misplaced_page migrate_pages So now, the TLB flushing code in do_huge_pmd_numa_page() becomes unnecessary. So the code is deleted in this patch to simplify the code. This is only code cleanup, there's no visible performance difference. The mmu_notifier_invalidate_range() in do_huge_pmd_numa_page() is deleted too. Because migrate_pages() takes care of that too when CPU TLB is flushed. Link: https://lkml.kernel.org/r/20210720065529.716031-1-ying.huang@intel.comSigned-off-by:
"Huang, Ying" <ying.huang@intel.com> Reviewed-by:
Zi Yan <ziy@nvidia.com> Reviewed-by:
Yang Shi <shy828301@gmail.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Christian Borntraeger <borntraeger@de.ibm.com> Cc: Gerald Schaefer <gerald.schaefer@linux.ibm.com> Cc: Heiko Carstens <hca@linux.ibm.com> Cc: Hugh Dickins <hughd@google.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Cc: Michal Hocko <mhocko@suse.com> Cc: Vasily Gorbik <gor@linux.ibm.com> Cc: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by:
-
Christoph Hellwig authored
flush_kernel_dcache_page is a rather confusing interface that implements a subset of flush_dcache_page by not being able to properly handle page cache mapped pages. The only callers left are in the exec code as all other previous callers were incorrect as they could have dealt with page cache pages. Replace the calls to flush_kernel_dcache_page with calls to flush_dcache_page, which for all architectures does either exactly the same thing, can contains one or more of the following: 1) an optimization to defer the cache flush for page cache pages not mapped into userspace 2) additional flushing for mapped page cache pages if cache aliases are possible Link: https://lkml.kernel.org/r/20210712060928.4161649-7-hch@lst.deSigned-off-by:Christoph Hellwig <hch@lst.de> Acked-by:
Ira Weiny <ira.weiny@intel.com> Cc: Alex Shi <alexs@kernel.org> Cc: Geoff Levand <geoff@infradead.org> Cc: Greentime Hu <green.hu@gmail.com> Cc: Guo Ren <guoren@kernel.org> Cc: Helge Deller <deller@gmx.de> Cc: "James E.J. Bottomley" <James.Bottomley@HansenPartnership.com> Cc: Nick Hu <nickhu@andestech.com> Cc: Paul Cercueil <paul@crapouillou.net> Cc: Rich Felker <dalias@libc.org> Cc: Russell King <linux@armlinux.org.uk> Cc: Thomas Bogendoerfer <tsbogend@alpha.franken.de> Cc: Ulf Hansson <ulf.hansson@linaro.org> Cc: Vincent Chen <deanbo422@gmail.com> Cc: Yoshinori Sato <ysato@users.osdn.me> Signed-off-by:
-
Christoph Hellwig authored
Pages used in scatterlist can be mapped page cache pages (and often are), so we must use flush_dcache_page here instead of the more limited flush_kernel_dcache_page that is intended for highmem pages only. Also remove the PageSlab check given that page_mapping_file as used by the flush_dcache_page implementations already contains that check. Link: https://lkml.kernel.org/r/20210712060928.4161649-5-hch@lst.deSigned-off-by:
-
Christoph Hellwig authored
Pages passed to block drivers can be mapped page cache pages, so we must use flush_dcache_page here instead of the more limited flush_kernel_dcache_page that is intended for highmem pages only. Link: https://lkml.kernel.org/r/20210712060928.4161649-3-hch@lst.deSigned-off-by:
-
Christoph Hellwig authored
Patch series "_kernel_dcache_page fixes and removal". While looking to convert the block layer away from kmap_atomic towards kmap_local_page and prefeably the helpers that abstract it away I noticed that a few block drivers directly or implicitly call flush_kernel_dcache_page before kunmapping a page that has been written to. flush_kernel_dcache_page is documented to to be used in such cases, but flush_dcache_page is actually required when the page could be in the page cache and mapped to userspace, which is pretty much always the case when kmapping an arbitrary page. Unfortunately the documentation doesn't exactly make that clear, which lead to this misused. And it turns out that only the copy_strings / copy_string_kernel in the exec code were actually correct users of flush_kernel_dcache_page, which is why I think we should just remove it and eat the very minor overhead in exec rather than confusing poor driver writers. This patch (of 6): MIPS now implements flush_kernel_dcache_page (as an alias to flush_dcache_page). Link: https://lkml.kernel.org/r/20210712060928.4161649-1-hch@lst.de Link: https://lkml.kernel.org/r/20210712060928.4161649-2-hch@lst.deSigned-off-by:
-
Colin Ian King authored
There is a spelling mistake in an error message. Fix it. Link: https://lkml.kernel.org/r/20210826121217.12885-1-colin.king@canonical.comSigned-off-by:
Colin Ian King <colin.king@canonical.com> Signed-off-by:
-
Po-Hsu Lin authored
There are several test cases in the vm directory are still using exit 0 when they need to be skipped. Use the kselftest framework to skip code instead so it can help us to distinguish the return status. Criterion to filter out what should be fixed in vm directory: grep -r "exit 0" -B1 | grep -i skip This change might cause some false-positives if people are running these test scripts directly and only checking their return codes, which will change from 0 to 4. However I think the impact should be small as most of our scripts here are already using this skip code. And there will be no such issue if running them with the kselftest framework. Link: https://lkml.kernel.org/r/20210823073433.37653-1-po-hsu.lin@canonical.comSigned-off-by:
Po-Hsu Lin <po-hsu.lin@canonical.com> Cc: Shuah Khan <shuah@kernel.org> Signed-off-by:
-
Shakeel Butt authored
The memcg->event_list_lock is usually taken in the normal context but when the userspace closes the corresponding eventfd, eventfd_release through memcg_event_wake takes memcg->event_list_lock with interrupts disabled. This is not an issue on its own but it creates a nested dependency from eventfd_ctx->wqh.lock to memcg->event_list_lock. Independently, for unrelated eventfd, eventfd_signal() can be called in the irq context, thus making eventfd_ctx->wqh.lock an irq lock. For example, FPGA DFL driver, VHOST VPDA driver and couple of VFIO drivers. This will force memcg->event_list_lock to be an irqsafe lock as well. One way to break the nested dependency between eventfd_ctx->wqh.lock and memcg->event_list_lock is to add an indirection. However the simplest solution would be to make memcg->event_list_lock irqsafe. This is cgroup v1 feature, is in maintenance and may get deprecated in near future. So, no need to add more code. BTW this has been discussed previously [1] but there weren't irq users of eventfd_signal() at the time. [1] https://www.spinics.net/lists/cgroups/msg06248.html Link: https://lkml.kernel.org/r/20210830172953.207257-1-shakeelb@google.comSigned-off-by:
Shakeel Butt <shakeelb@google.com> Cc: Tejun Heo <tj@kernel.org> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Michal Hocko <mhocko@kernel.org> Signed-off-by:
-
Michal Hocko authored
Thomas and Vlastimil have noticed that the comment in drain_local_stock doesn't quite make sense. It talks about a synchronization with the memory hotplug but there is no actual memory hotplug involvement here. I meant to talk about cpu hotplug here. Fix that up and hopefuly make the comment more helpful by referencing the cpu hotplug callback as well. Link: https://lkml.kernel.org/r/YRDwOhVglJmY7ES5@dhcp22.suse.czSigned-off-by:
Michal Hocko <mhocko@suse.com> Acked-by:
Vlastimil Babka <vbabka@suse.cz> Cc: Thomas Gleixner <tglx@linutronix.de> Signed-off-by:
-
Miaohe Lin authored
Add 'else' to save some atomic ops in obj_stock_flush_required() when flush is already true. No functional change intended here. Link: https://lkml.kernel.org/r/20210807082835.61281-3-linmiaohe@huawei.comSigned-off-by:
Miaohe Lin <linmiaohe@huawei.com> Reviewed-by:
Muchun Song <songmuchun@bytedance.com> Acked-by:
Roman Gushchin <guro@fb.com> Acked-by:
-
Miaohe Lin authored
Since commit 2d146aa3 ("mm: memcontrol: switch to rstat"), last user of memcg_stat_item_in_bytes() is gone. And since commit fa40d1ee ("mm: vmscan: memcontrol: remove mem_cgroup_select_victim_node()"), only the declaration of mem_cgroup_select_victim_node() is remained here. Remove them. Link: https://lkml.kernel.org/r/20210807082835.61281-2-linmiaohe@huawei.comSigned-off-by:
-
Baolin Wang authored
Since commit c843966c ("mm: allow swappiness that prefers reclaiming anon over the file workingset") has expended the swappiness value to make swap to be preferred in some systems. We should also change the memcg swappiness restriction to allow memcg swap-preferred. Link: https://lkml.kernel.org/r/d77469b90c45c49953ccbc51e54a1d465bc18f70.1627626255.git.baolin.wang@linux.alibaba.com Fixes: c843966c ("mm: allow swappiness that prefers reclaiming anon over the file workingset") Signed-off-by:
Baolin Wang <baolin.wang@linux.alibaba.com> Acked-by:
-
Vasily Averin authored
set_active_memcg() uses in_interrupt() check to select proper storage for cgroup: pointer on task struct or per-cpu pointer. It isn't fully correct: obsoleted in_interrupt() includes tasks with disabled BH. It's better to use '!in_task()' instead. Link: https://lkml.org/lkml/2021/7/26/487 Link: https://lkml.kernel.org/r/ed4448b0-4970-616f-7368-ef9dd3cb628d@virtuozzo.com Fixes: 37d5985c ("mm: kmem: prepare remote memcg charging infra for interrupt contexts") Signed-off-by:
Vasily Averin <vvs@virtuozzo.com> Reviewed-by:
-
Shakeel Butt authored
We used to have per-cpu memcg and lruvec stats and the readers have to traverse and sum the stats from each cpu. This summing was racy and may expose transient negative values. So, an explicit check was added to avoid such scenarios. Now these stats are moved to rstat infrastructure and are no more per-cpu, so we can remove the fixup for transient negative values. Link: https://lkml.kernel.org/r/20210728012243.3369123-1-shakeelb@google.comSigned-off-by:
David Hildenbrand <david@redhat.com> Acked-by:
-
Vasily Averin authored
Each task can request own LDT and force the kernel to allocate up to 64Kb memory per-mm. There are legitimate workloads with hundreds of processes and there can be hundreds of workloads running on large machines. The unaccounted memory can cause isolation issues between the workloads particularly on highly utilized machines. It makes sense to account for this objects to restrict the host's memory consumption from inside the memcg-limited container. Link: https://lkml.kernel.org/r/38010594-50fe-c06d-7cb0-d1f77ca422f3@virtuozzo.comSigned-off-by:
Borislav Petkov <bp@suse.de> Reviewed-by:
-
Vasily Averin authored
A program may create multiple interval timers using timer_create(). For each timer the kernel preallocates a "queued real-time signal", Consequently, the number of timers is limited by the RLIMIT_SIGPENDING resource limit. The allocated object is quite small, ~250 bytes, but even the default signal limits allow to consume up to 100 megabytes per user. It makes sense to account for them to limit the host's memory consumption from inside the memcg-limited container. Link: https://lkml.kernel.org/r/57795560-025c-267c-6b1a-dea852d95530@virtuozzo.comSigned-off-by:
Thomas Gleixner <tglx@linutronix.de> Reviewed-by:
-
Vasily Averin authored
When a user send a signal to any another processes it forces the kernel to allocate memory for 'struct sigqueue' objects. The number of signals is limited by RLIMIT_SIGPENDING resource limit, but even the default settings allow each user to consume up to several megabytes of memory. It makes sense to account for these allocations to restrict the host's memory consumption from inside the memcg-limited container. Link: https://lkml.kernel.org/r/e34e958c-e785-712e-a62a-2c7b66c646c7@virtuozzo.comSigned-off-by:
-
Vasily Averin authored
When user creates IPC objects it forces kernel to allocate memory for these long-living objects. It makes sense to account them to restrict the host's memory consumption from inside the memcg-limited container. This patch enables accounting for IPC shared memory segments, messages semaphores and semaphore's undo lists. Link: https://lkml.kernel.org/r/d6507b06-4df6-78f8-6c54-3ae86e3b5339@virtuozzo.comSigned-off-by:
-
Vasily Averin authored
Container admin can create new namespaces and force kernel to allocate up to several pages of memory for the namespaces and its associated structures. Net and uts namespaces have enabled accounting for such allocations. It makes sense to account for rest ones to restrict the host's memory consumption from inside the memcg-limited container. Link: https://lkml.kernel.org/r/5525bcbf-533e-da27-79b7-158686c64e13@virtuozzo.comSigned-off-by:
Serge Hallyn <serge@hallyn.com> Acked-by:
Christian Brauner <christian.brauner@ubuntu.com> Acked-by:
Kirill Tkhai <ktkhai@virtuozzo.com> Reviewed-by:
-
Vasily Averin authored
fasync_struct is used by almost all character device drivers to set up the fasync queue, and for regular files by the file lease code. This structure is quite small but long-living and it can be assigned for any open file. It makes sense to account for its allocations to restrict the host's memory consumption from inside the memcg-limited container. Link: https://lkml.kernel.org/r/1b408625-d71c-0b26-b0b6-9baf00f93e69@virtuozzo.comSigned-off-by:
-
Vasily Averin authored
User can create file locks for each open file and force kernel to allocate small but long-living objects per each open file. It makes sense to account for these objects to limit the host's memory consumption from inside the memcg-limited container. Link: https://lkml.kernel.org/r/b009f4c7-f0ab-c0ec-8e83-918f47d677da@virtuozzo.comSigned-off-by:
-
Vasily Averin authored
User can call select/poll system calls with a large number of assigned file descriptors and force kernel to allocate up to several pages of memory till end of these sleeping system calls. We have here long-living unaccounted per-task allocations. It makes sense to account for these allocations to restrict the host's memory consumption from inside the memcg-limited container. Link: https://lkml.kernel.org/r/56e31cb5-6e1e-bdba-d7ca-be64b9842363@virtuozzo.comSigned-off-by:
-
Vasily Averin authored
Patch series "memcg accounting from OpenVZ", v7. OpenVZ uses memory accounting 20+ years since v2.2.x linux kernels. Initially we used our own accounting subsystem, then partially committed it to upstream, and a few years ago switched to cgroups v1. Now we're rebasing again, revising our old patches and trying to push them upstream. We try to protect the host system from any misuse of kernel memory allocation triggered by untrusted users inside the containers. Patch-set is addressed mostly to cgroups maintainers and cgroups@ mailing list, though I would be very grateful for any comments from maintainersi of affected subsystems or other people added in cc: Compared to the upstream, we additionally account the following kernel objects: - network devices and its Tx/Rx queues - ipv4/v6 addresses and routing-related objects - inet_bind_bucket cache objects - VLAN group arrays - ipv6/sit: ip_tunnel_prl - scm_fp_list objects used by SCM_RIGHTS messages of Unix sockets - nsproxy and namespace objects itself - IPC objects: semaphores, message queues and share memory segments - mounts - pollfd and select bits arrays - signals and posix timers - file lock - fasync_struct used by the file lease code and driver's fasync queues - tty objects - per-mm LDT We have an incorrect/incomplete/obsoleted accounting for few other kernel objects: sk_filter, af_packets, netlink and xt_counters for iptables. They require rework and probably will be dropped at all. Also we're going to add an accounting for nft, however it is not ready yet. We have not tested performance on upstream, however, our performance team compares our current RHEL7-based production kernel and reports that they are at least not worse as the according original RHEL7 kernel. This patch (of 10): The kernel allocates ~400 bytes of 'struct mount' for any new mount. Creating a new mount namespace clones most of the parent mounts, and this can be repeated many times. Additionally, each mount allocates up to PATH_MAX=4096 bytes for mnt->mnt_devname. It makes sense to account for these allocations to restrict the host's memory consumption from inside the memcg-limited container. Link: https://lkml.kernel.org/r/045db11f-4a45-7c9b-2664-5b32c2b44943@virtuozzo.comSigned-off-by:
-
Yutian Yang authored
This patch adds accounting flags to fs_context and legacy_fs_context allocation sites so that kernel could correctly charge these objects. We have written a PoC to demonstrate the effect of the missing-charging bugs. The PoC takes around 1,200MB unaccounted memory, while it is charged for only 362MB memory usage. We evaluate the PoC on QEMU x86_64 v5.2.90 + Linux kernel v5.10.19 + Debian buster. All the limitations including ulimits and sysctl variables are set as default. Specifically, the hard NOFILE limit and nr_open in sysctl are both 1,048,576. /*------------------------- POC code ----------------------------*/ #define _GNU_SOURCE #include <sys/types.h> #include <sys/file.h> #include <time.h> #include <sys/wait.h> #include <stdint.h> #include <stdlib.h> #include <unistd.h> #include <stdio.h> #include <signal.h> #include <sched.h> #include <fcntl.h> #include <linux/mount.h> #define errExit(msg) do { perror(msg); exit(EXIT_FAILURE); \ } while (0) #define STACK_SIZE (8 * 1024) #ifndef __NR_fsopen #define __NR_fsopen 430 #endif static inline int fsopen(const char *fs_name, unsigned int flags) { return syscall(__NR_fsopen, fs_name, flags); } static char thread_stack[512][STACK_SIZE]; int thread_fn(void* arg) { for (int i = 0; i< 800000; ++i) { int fsfd = fsopen("nfs", FSOPEN_CLOEXEC); if (fsfd == -1) { errExit("fsopen"); } } while(1); return 0; } int main(int argc, char *argv[]) { int thread_pid; for (int i = 0; i < 1; ++i) { thread_pid = clone(thread_fn, thread_stack[i] + STACK_SIZE, \ SIGCHLD, NULL); } while(1); return 0; } /*-------------------------- end --------------------------------*/ Link: https://lkml.kernel.org/r/1626517201-24086-1-git-send-email-nglaive@gmail.comSigned-off-by:Yutian Yang <nglaive@gmail.com> Reviewed-by:
-
Shakeel Butt authored
At the moment memcg stats are read in four contexts: 1. memcg stat user interfaces 2. dirty throttling 3. page fault 4. memory reclaim Currently the kernel flushes the stats for first two cases. Flushing the stats for remaining two casese may have performance impact. Always flushing the memcg stats on the page fault code path may negatively impacts the performance of the applications. In addition flushing in the memory reclaim code path, though treated as slowpath, can become the source of contention for the global lock taken for stat flushing because when system or memcg is under memory pressure, many tasks may enter the reclaim path. This patch uses following mechanisms to solve these challenges: 1. Periodically flush the stats from root memcg every 2 seconds. This will time limit the out of sync stats. 2. Asynchronously flush the stats after fixed number of stat updates. In the worst case the stat can be out of sync by O(nr_cpus * BATCH) for 2 seconds. 3. For avoiding thundering herd to flush the stats particularly from the memory reclaim context, introduce memcg local spinlock and let only one flusher active at a time. This could have been done through cgroup_rstat_lock lock but that lock is used by other subsystem and for userspace reading memcg stats. So, it is better to keep flushers introduced by this patch decoupled from cgroup_rstat_lock. However we would have to use irqsafe version of rstat flush but that is fine as this code path will be flushing for whole tree and do the work for everyone. No one will be waiting for that worker. [shakeelb@google.com: fix sleep-in-wrong context bug] Link: https://lkml.kernel.org/r/20210716212137.1391164-2-shakeelb@google.com Link: https://lkml.kernel.org/r/20210714013948.270662-2-shakeelb@google.comSigned-off-by:
Marek Szyprowski <m.szyprowski@samsung.com> Cc: Hillf Danton <hdanton@sina.com> Cc: Huang Ying <ying.huang@intel.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Michal Hocko <mhocko@kernel.org> Cc: Michal Koutný <mkoutny@suse.com> Cc: Muchun Song <songmuchun@bytedance.com> Cc: Roman Gushchin <guro@fb.com> Cc: Tejun Heo <tj@kernel.org> Signed-off-by:
-
Shakeel Butt authored
The commit 2d146aa3 ("mm: memcontrol: switch to rstat") switched memcg stats to rstat infrastructure but skipped the conversion of the lruvec stats as such stats are read in the performance critical code paths and flushing stats may have impacted the performances of the applications. This patch converts the lruvec stats to rstat and later patches add mechanisms to keep the performance impact to minimum. The rstat conversion comes with the price i.e. memory cost. Effectively this patch reverts the savings done by the commit f3344adf ("mm: memcontrol: optimize per-lruvec stats counter memory usage"). However this cost is justified due to negative impact of the inaccurate lruvec stats on many heuristics. One such case is reported in [1]. The memory reclaim code is filled with plethora of heuristics and many of those heuristics reads the lruvec stats. So, inaccurate stats can make such heuristics ineffective. [1] reports the impact of inaccurate lruvec stats on the "cache trim mode" heuristic. Inaccurate lruvec stats can impact the deactivation and aging anon heuristics as well. [1] https://lore.kernel.org/linux-mm/20210311004449.1170308-1-ying.huang@intel.com/ Link: https://lkml.kernel.org/r/20210716212137.1391164-1-shakeelb@google.com Link: https://lkml.kernel.org/r/20210714013948.270662-1-shakeelb@google.comSigned-off-by:
-
Vasily Averin authored
Commit 5d097056 ("kmemcg: account certain kmem allocations to memcg") enabled memcg accounting for pids allocated from init_pid_ns.pid_cachep, but forgot to adjust the setting for nested pid namespaces. As a result, pid memory is not accounted exactly where it is really needed, inside memcg-limited containers with their own pid namespaces. Pid was one the first kernel objects enabled for memcg accounting. init_pid_ns.pid_cachep marked by SLAB_ACCOUNT and we can expect that any new pids in the system are memcg-accounted. Though recently I've noticed that it is wrong. nested pid namespaces creates own slab caches for pid objects, nested pids have increased size because contain id both for all parent and for own pid namespaces. The problem is that these slab caches are _NOT_ marked by SLAB_ACCOUNT, as a result any pids allocated in nested pid namespaces are not memcg-accounted. Pid struct in nested pid namespace consumes up to 500 bytes memory, 100000 such objects gives us up to ~50Mb unaccounted memory, this allow container to exceed assigned memcg limits. Link: https://lkml.kernel.org/r/8b6de616-fd1a-02c6-cbdb-976ecdcfa604@virtuozzo.com Fixes: 5d097056 ("kmemcg: account certain kmem allocations to memcg") Cc: stable@vger.kernel.org Signed-off-by:
Michal Koutný <mkoutny@suse.com> Reviewed-by:
-
Suren Baghdasaryan authored
Inline mem_cgroup_try_charge_swap, mem_cgroup_uncharge_swap and cgroup_throttle_swaprate functions to perform mem_cgroup_disabled static key check inline before calling the main body of the function. This minimizes the memcg overhead in the pagefault and exit_mmap paths when memcgs are disabled using cgroup_disable=memory command-line option. This change results in ~1% overhead reduction when running PFT test [1] comparing {CONFIG_MEMCG=n} against {CONFIG_MEMCG=y, cgroup_disable=memory} configuration on an 8-core ARM64 Android device. [1] https://lkml.org/lkml/2006/8/29/294 also used in mmtests suite Link: https://lkml.kernel.org/r/20210713010934.299876-3-surenb@google.comSigned-off-by:Suren Baghdasaryan <surenb@google.com> Reviewed-by:
Johannes Weiner <hannes@cmpxchg.org> Acked-by:
-
Suren Baghdasaryan authored
Inline mem_cgroup_{charge/uncharge} and mem_cgroup_uncharge_list functions functions to perform mem_cgroup_disabled static key check inline before calling the main body of the function. This minimizes the memcg overhead in the pagefault and exit_mmap paths when memcgs are disabled using cgroup_disable=memory command-line option. This change results in ~0.4% overhead reduction when running PFT test [1] comparing {CONFIG_MEMCG=n} against {CONFIG_MEMCG=y, cgroup_disable=memory} configuration on an 8-core ARM64 Android device. [1] https://lkml.org/lkml/2006/8/29/294 also used in mmtests suite Link: https://lkml.kernel.org/r/20210713010934.299876-2-surenb@google.comSigned-off-by: -
Suren Baghdasaryan authored
Add mem_cgroup_disabled check in vmpressure, mem_cgroup_uncharge_swap and cgroup_throttle_swaprate functions. This minimizes the memcg overhead in the pagefault and exit_mmap paths when memcgs are disabled using cgroup_disable=memory command-line option. This change results in ~2.1% overhead reduction when running PFT test [1] comparing {CONFIG_MEMCG=n, CONFIG_MEMCG_SWAP=n} against {CONFIG_MEMCG=y, CONFIG_MEMCG_SWAP=y, cgroup_disable=memory} configuration on an 8-core ARM64 Android device. [1] https://lkml.org/lkml/2006/8/29/294 also used in mmtests suite Link: https://lkml.kernel.org/r/20210713010934.299876-1-surenb@google.comSigned-off-by: -
Hugh Dickins authored
drivers/gpu/drm/i915/gem/i915_gem_shmem.c contains a shmem_writeback() which calls shmem_writepage() from a shrinker: that usually works well enough; but if /sys/kernel/mm/transparent_hugepage/shmem_enabled has been set to "always" (intended to be usable) or "force" (forces huge everywhere for easy testing), shmem_writepage() is surprised to be called with a huge page, and crashes on the VM_BUG_ON_PAGE(PageCompound) (I did not find out where the crash happens when CONFIG_DEBUG_VM is off). LRU page reclaim always splits the shmem huge page first: I'd prefer not to demand that of i915, so check and split compound in shmem_writepage(). Patch history: when first sent last year http://lkml.kernel.org/r/alpine.LSU.2.11.2008301401390.5954@eggly.anvils https://lore.kernel.org/linux-mm/20200919042009.bomzxmrg7%25akpm@linux-foundation.org/ Matthew Wilcox noticed that tail pages were wrongly left clean. This version brackets the split with Set and Clear PageDirty as he suggested: which works very well, even if it falls short of our aspirations. And recently I realized that the crash is not limited to the testing option "force", but affects "always" too: which is more important to fix. Link: https://lkml.kernel.org/r/bac6158c-8b3d-4dca-cffc-4982f58d9794@google.com Fixes: 2d6692e6 ("drm/i915: Start writeback from the shrinker") Signed-off-by:
Hugh Dickins <hughd@google.com> Reviewed-by:
-
Hugh Dickins authored
4.18 commit 89fdcd26 ("mm: shmem: make stat.st_blksize return huge page size if THP is on") added is_huge_enabled() to decide st_blksize: if hugeness is to be defined per file, that will need to be replaced by shmem_is_huge(). This does give a different answer (No) for small files on a "huge=within_size" mount: but that can be considered a minor bugfix. And a different answer (No) for default files on a "huge=advise" mount: I'm reluctant to complicate it, just to reproduce the same debatable answer as before. Link: https://lkml.kernel.org/r/af7fb3f9-4415-9e8e-fdac-b1a5253ad21@google.comSigned-off-by:
-
Hugh Dickins authored
Extend shmem_huge_enabled(vma) to shmem_is_huge(vma, inode, index), so that a consistent set of checks can be applied, even when the inode is accessed through read/write syscalls (with NULL vma) instead of mmaps (the index argument is seldom of interest, but required by mount option "huge=within_size"). Clean up and rearrange the checks a little. This then replaces the checks which shmem_fault() and shmem_getpage_gfp() were making, and eliminates the SGP_HUGE and SGP_NOHUGE modes. Replace a couple of 0s by explicit SHMEM_HUGE_NEVERs; and replace the obscure !shmem_mapping() symlink check by explicit S_ISLNK() - nothing else needs that symlink check, so leave it there in shmem_getpage_gfp(). Link: https://lkml.kernel.org/r/23a77889-2ddc-b030-75cd-44ca27fd4d1@google.comSigned-off-by:
-
Hugh Dickins authored
khugepaged's collapse_file() currently uses SGP_NOHUGE to tell shmem_getpage() not to try allocating a huge page, in the very unlikely event that a racing hole-punch removes the swapped or fallocated page as soon as i_pages lock is dropped. We want to consolidate shmem's huge decisions, removing SGP_HUGE and SGP_NOHUGE; but cannot quite persuade ourselves that it's okay to regress the protection in this case - Yang Shi points out that the huge page would remain indefinitely, charged to root instead of the intended memcg. collapse_file() should not even allocate a small page in this case: why proceed if someone is punching a hole? SGP_READ is almost the right flag here, except that it optimizes away from a fallocated page, with NULL to tell caller to fill with zeroes (like a hole); whereas collapse_file()'s sequence relies on using a cache page. Add SGP_NOALLOC just for this. There are too many consecutive "if (page"s there in shmem_getpage_gfp(): group it better; and fix the outdated "bring it back from swap" comment. Link: https://lkml.kernel.org/r/1355343b-acf-4653-ef79-6aee40214ac5@google.comSigned-off-by:
-
Hugh Dickins authored
shmem_huge_enabled() is about to be enhanced into shmem_is_huge(), so that it can be used more widely throughout: before making functional changes, shift it to its final position (to avoid forward declaration). Link: https://lkml.kernel.org/r/16fec7b7-5c84-415a-8586-69d8bf6a6685@google.comSigned-off-by:
-
Hugh Dickins authored
5.14 commit e6be37b2 ("mm/huge_memory.c: add missing read-only THP checking in transparent_hugepage_enabled()") added transhuge_vma_enabled() as a wrapper for two very different checks (one check is whether the app has marked its address range not to use THPs, the other check is whether the app is running in a hierarchy that has been marked never to use THPs). shmem_huge_enabled() prefers to show those two checks explicitly, as before. Link: https://lkml.kernel.org/r/45e5338-18d-c6f9-c17e-34f510bc1728@google.comSigned-off-by:
-
Hugh Dickins authored
There's a block of code in shmem_setattr() to add the inode to shmem_unused_huge_shrink()'s shrinklist when lowering i_size: it dates from before 5.7 changed truncation to do split_huge_page() for itself, and should have been removed at that time. I am over-stating that: split_huge_page() can fail (notably if there's an extra reference to the page at that time), so there might be value in retrying. But there were already retries as truncation worked through the tails, and this addition risks repeating unsuccessful retries indefinitely: I'd rather remove it now, and work on reducing the chance of split_huge_page() failures separately, if we need to. Link: https://lkml.kernel.org/r/b73b3492-8822-18f9-83e2-938528cdde94@google.com Fixes: 71725ed1 ("mm: huge tmpfs: try to split_huge_page() when punching hole") Signed-off-by:
-
Hugh Dickins authored
A successful shmem_fallocate() guarantees that the extent has been reserved, even beyond i_size when the FALLOC_FL_KEEP_SIZE flag was used. But that guarantee is broken by shmem_unused_huge_shrink()'s attempts to split huge pages and free their excess beyond i_size; and by other uses of split_huge_page() near i_size. It's sad to add a shmem inode field just for this, but I did not find a better way to keep the guarantee. A flag to say KEEP_SIZE has been used would be cheaper, but I'm averse to unclearable flags. The fallocend field is not perfect either (many disjoint ranges might be fallocated), but good enough; and gains another use later on. Link: https://lkml.kernel.org/r/ca9a146-3a59-6cd3-7f28-e9a044bb1052@google.com Fixes: 779750d2 ("shmem: split huge pages beyond i_size under memory pressure") Signed-off-by:
-
Hugh Dickins authored
Patch series "huge tmpfs: shmem_is_huge() fixes and cleanups". A series of huge tmpfs fixes and cleanups. This patch (of 9): shmem_fallocate() goes to a lot of trouble to leave its newly allocated pages !Uptodate, partly to identify and undo them on failure, partly to leave the overhead of clearing them until later. But the huge page case did not skip to the end of the extent, walked through the tail pages one by one, and appeared to work just fine: but in doing so, cleared and Uptodated the huge page, so there was no way to undo it on failure. And by setting Uptodate too soon, it messed up both its nr_falloced and nr_unswapped counts, so that the intended "time to give up" heuristic did not work at all. Now advance immediately to the end of the huge extent, with a comment on why this is more than just an optimization. But although this speeds up huge tmpfs fallocation, it does leave the clearing until first use, and some users may have come to appreciate slow fallocate but fast first use: if they complain, then we can consider adding a pass to clear at the end. Link: https://lkml.kernel.org/r/da632211-8e3e-6b1-aee-ab24734429a0@google.com Link: https://lkml.kernel.org/r/16201bd2-70e-37e2-e89b-5f929430da@google.com Fixes: 800d8c63 ("shmem: add huge pages support") Signed-off-by:
-
