- 12 Nov, 2018 4 commits
-
-
Patrick Bellasi authored
The following pattern: var -= min_t(typeof(var), var, val); is used multiple times in fair.c. The existing sub_positive() already captures that pattern, but it also adds an explicit load-store to properly support lockless observations. In other cases the pattern above is used to update local, and/or not concurrently accessed, variables. Let's add a simpler version of sub_positive(), targeted at local variables updates, which gives the same readability benefits at calling sites, without enforcing {READ,WRITE}_ONCE() barriers. Signed-off-by:Patrick Bellasi <patrick.bellasi@arm.com> Signed-off-by:
Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Dietmar Eggemann <dietmar.eggemann@arm.com> Cc: Juri Lelli <juri.lelli@redhat.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Morten Rasmussen <morten.rasmussen@arm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Quentin Perret <quentin.perret@arm.com> Cc: Steve Muckle <smuckle@google.com> Cc: Suren Baghdasaryan <surenb@google.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Todd Kjos <tkjos@google.com> Cc: Vincent Guittot <vincent.guittot@linaro.org> Link: https://lore.kernel.org/lkml/20181031184527.GA3178@hirez.programming.kicks-ass.netSigned-off-by:
Ingo Molnar <mingo@kernel.org>
-
Patrick Bellasi authored
The _task_util_est() is mainly used to add/remove the task contribution to/from the rq's estimated utilization at task enqueue/dequeue time. In both cases we ensure the UTIL_AVG_UNCHANGED flag is set to keep consistency between enqueue and dequeue time while still being transparent to update_load_avg calls which will eventually reset the flag. Let's move the flag forcing within _task_util_est() itself so that we can simplify calling code by hiding that estimated utilization implementation detail into one of its internal functions. This will affect also the "public" API task_util_est() but we know that the flag will (eventually) impact just on the LSB of the estimated utilization, thus it's certainly acceptable. Signed-off-by:
-
Ingo Molnar authored
Signed-off-by: -
Patrick Bellasi authored
A ~10% regression has been reported for UnixBench's execl throughput test by Aaron Lu and Ye Xiaolong: https://lkml.org/lkml/2018/10/30/765 That test is pretty simple, it does a "recursive" execve() syscall on the same binary. Starting from the syscall, this sequence is possible: do_execve() do_execveat_common() __do_execve_file() sched_exec() select_task_rq_fair() <==| Task already enqueued find_idlest_cpu() find_idlest_group() capacity_spare_wake() <==| Functions not called from cpu_util_wake() | the wakeup path which means we can end up calling cpu_util_wake() not only from the "wakeup path", as its name would suggest. Indeed, the task doing an execve() syscall is already enqueued on the CPU we want to get the cpu_util_wake() for. The estimated utilization for a CPU computed in cpu_util_wake() was written under the assumption that function can be called only from the wakeup path. If instead the task is already enqueued, we end up with a utilization which does not remove the current task's contribution from the estimated utilization of the CPU. This will wrongly assume a reduced spare capacity on the current CPU and increase the chances to migrate the task on execve. The regression is tracked down to: commit d519329f ("sched/fair: Update util_est only on util_avg updates") because in that patch we turn on by default the UTIL_EST sched feature. However, the real issue is introduced by: commit f9be3e59 ("sched/fair: Use util_est in LB and WU paths") Let's fix this by ensuring to always discount the task estimated utilization from the CPU's estimated utilization when the task is also the current one. The same benchmark of the bug report, executed on a dual socket 40 CPUs Intel(R) Xeon(R) CPU E5-2690 v2 @ 3.00GHz machine, reports these "Execl Throughput" figures (higher the better): mainline : 48136.5 lps mainline+fix : 55376.5 lps which correspond to a 15% speedup. Moreover, since {cpu_util,capacity_spare}_wake() are not really only used from the wakeup path, let's remove this ambiguity by using a better matching name: {cpu_util,capacity_spare}_without(). Since we are at that, let's also improve the existing documentation. Reported-by:
Aaron Lu <aaron.lu@intel.com> Reported-by:
Ye Xiaolong <xiaolong.ye@intel.com> Tested-by:
-
- 05 Nov, 2018 1 commit
-
-
Yi Wang authored
Duplicated 'case it'. Signed-off-by:
Yi Wang <wang.yi59@zte.com.cn> Reviewed-by:
Xi Xu <xu.xi8@zte.com.cn> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: zhong.weidong@zte.com.cn Link: http://lkml.kernel.org/r/1541379013-11352-1-git-send-email-wang.yi59@zte.com.cnSigned-off-by:
-
- 03 Nov, 2018 5 commits
-
-
Muchun Song authored
When we pick the next task, we will do the following for the task: 1) p->se.exec_start = rq_clock_task(rq); 2) dequeue_pushable(_dl)_task(rq, p); When we call set_curr_task(), we also need to do the same thing above. In rt.c, the code at 1) is in the _pick_next_task_rt() and the code at 2) is in the pick_next_task_rt(). If we put two operations in one function, maybe better. So, we introduce a new function set_next_task(), which is responsible for doing the above. By introducing the function we can get rid of calling the dequeue_pushable(_dl)_task() directly(We can call set_next_task()) in pick_next_task() and have better code readability and reuse. In set_curr_task_rt(), we also can call set_next_task(). Do this things such that we end up with: static struct task_struct *pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf) { /* do something else ... */ put_prev_task(rq, prev); /* pick next task p */ set_next_task(rq, p); /* do something else ... */ } put_prev_task() can match set_next_task(), which can make the code more readable. Signed-off-by:Muchun Song <smuchun@gmail.com> Signed-off-by:
-
Valentin Schneider authored
When load_balance() fails to move some load because of task affinity, we end up increasing sd->balance_interval to delay the next periodic balance in the hopes that next time we look, that annoying pinned task(s) will be gone. However, idle_balance() pays no attention to sd->balance_interval, yet it will still lead to an increase in balance_interval in case of pinned tasks. If we're going through several newidle balances (e.g. we have a periodic task), this can lead to a huge increase of the balance_interval in a very small amount of time. To prevent that, don't increase the balance interval when going through a newidle balance. This is a similar approach to what is done in commit 58b26c4c ("sched: Increment cache_nice_tries only on periodic lb"), where we disregard newidle balance and rely on periodic balance for more stable results. Signed-off-by:
Valentin Schneider <valentin.schneider@arm.com> Signed-off-by:
-
Valentin Schneider authored
The alignment of the condition is off, clean that up. Also, logical operators have lower precedence than bitwise/relational operators, so remove one layer of parentheses to make the condition a bit simpler to follow. Signed-off-by:
-
Valentin Schneider authored
When running on linux-next (8c60c36d0b8c ("Add linux-next specific files for 20181019")) + CONFIG_PROVE_LOCKING=y on a big.LITTLE system (e.g. Juno or HiKey960), we get the following report: [ 0.748225] Call trace: [ 0.750685] lockdep_assert_cpus_held+0x30/0x40 [ 0.755236] static_key_enable_cpuslocked+0x20/0xc8 [ 0.760137] build_sched_domains+0x1034/0x1108 [ 0.764601] sched_init_domains+0x68/0x90 [ 0.768628] sched_init_smp+0x30/0x80 [ 0.772309] kernel_init_freeable+0x278/0x51c [ 0.776685] kernel_init+0x10/0x108 [ 0.780190] ret_from_fork+0x10/0x18 The static_key in question is 'sched_asym_cpucapacity' introduced by commit: df054e84 ("sched/topology: Add static_key for asymmetric CPU capacity optimizations") In this particular case, we enable it because smp_prepare_cpus() will end up fetching the capacity-dmips-mhz entry from the devicetree, so we already have some asymmetry detected when entering sched_init_smp(). This didn't get detected in tip/sched/core because we were missing: commit cb538267 ("jump_label/lockdep: Assert we hold the hotplug lock for _cpuslocked() operations") Calls to build_sched_domains() post sched_init_smp() will hold the hotplug lock, it just so happens that this very first call is a special case. As stated by a comment in sched_init_smp(), "There's no userspace yet to cause hotplug operations" so this is a harmless warning. However, to both respect the semantics of underlying callees and make lockdep happy, take the hotplug lock in sched_init_smp(). This also satisfies the comment atop sched_init_domains() that says "Callers must hold the hotplug lock". Reported-by:Sudeep Holla <sudeep.holla@arm.com> Tested-by:
-
Peter Zijlstra authored
With the addition of the NUMA identity level, we increased @level by one and will run off the end of the array in the distance sort loop. Fixed: 051f3ca0 ("sched/topology: Introduce NUMA identity node sched domain") Signed-off-by:
-
- 29 Oct, 2018 1 commit
-
-
Muchun Song authored
Commit: f4ebcbc0 ("sched/rt: Substract number of tasks of throttled queues from rq->nr_running") added a new rt_rq->rt_queued field, which is used to indicate the status of rq->rt enqueue or dequeue. So, the ->rt_nr_running check was removed and we now check ->rt_queued instead. Fix the comment in pick_next_task_rt() as well, which was still referencing the old logic. Signed-off-by:
-
- 27 Oct, 2018 4 commits
-
-
Linus Torvalds authored
Commit 9ee3e066 ("HID: i2c-hid: override HID descriptors for certain devices") added a new dmi_system_id quirk table to override certain HID report descriptors for some systems that lack them. But the table wasn't properly terminated, causing the dmi matching to walk off into la-la-land, and starting to treat random data as dmi descriptor pointers, causing boot-time oopses if you were at all unlucky. Terminate the array. We really should have some way to just statically check that arrays that should be terminated by an empty entry actually are so. But the HID people really should have caught this themselves, rather than have me deal with an oops during the merge window. Tssk, tssk. Cc: Julian Sax <jsbc@gmx.de> Cc: Benjamin Tissoires <benjamin.tissoires@redhat.com> Cc: Jiri Kosina <jkosina@suse.cz> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org>
-
Linus Torvalds authored
Merge updates from Andrew Morton: - a few misc things - ocfs2 updates - most of MM * emailed patches from Andrew Morton <akpm@linux-foundation.org>: (132 commits) hugetlbfs: dirty pages as they are added to pagecache mm: export add_swap_extent() mm: split SWP_FILE into SWP_ACTIVATED and SWP_FS tools/testing/selftests/vm/map_fixed_noreplace.c: add test for MAP_FIXED_NOREPLACE mm: thp: relocate flush_cache_range() in migrate_misplaced_transhuge_page() mm: thp: fix mmu_notifier in migrate_misplaced_transhuge_page() mm: thp: fix MADV_DONTNEED vs migrate_misplaced_transhuge_page race condition mm/kasan/quarantine.c: make quarantine_lock a raw_spinlock_t mm/gup: cache dev_pagemap while pinning pages Revert "x86/e820: put !E820_TYPE_RAM regions into memblock.reserved" mm: return zero_resv_unavail optimization mm: zero remaining unavailable struct pages tools/testing/selftests/vm/gup_benchmark.c: add MAP_HUGETLB option tools/testing/selftests/vm/gup_benchmark.c: add MAP_SHARED option tools/testing/selftests/vm/gup_benchmark.c: allow user specified file tools/testing/selftests/vm/gup_benchmark.c: fix 'write' flag usage mm/gup_benchmark.c: add additional pinning methods mm/gup_benchmark.c: time put_page() mm: don't raise MEMCG_OOM event due to failed high-order allocation mm/page-writeback.c: fix range_cyclic writeback vs writepages deadlock ...
-
git://git.kernel.org/pub/scm/linux/kernel/git/davem/netLinus Torvalds authored
Pull networking fixes from David Miller: "What better way to start off a weekend than with some networking bug fixes: 1) net namespace leak in dump filtering code of ipv4 and ipv6, fixed by David Ahern and Bjørn Mork. 2) Handle bad checksums from hardware when using CHECKSUM_COMPLETE properly in UDP, from Sean Tranchetti. 3) Remove TCA_OPTIONS from policy validation, it turns out we don't consistently use nested attributes for this across all packet schedulers. From David Ahern. 4) Fix SKB corruption in cadence driver, from Tristram Ha. 5) Fix broken WoL handling in r8169 driver, from Heiner Kallweit. 6) Fix OOPS in pneigh_dump_table(), from Eric Dumazet" * git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (28 commits) net/neigh: fix NULL deref in pneigh_dump_table() net: allow traceroute with a specified interface in a vrf bridge: do not add port to router list when receives query with source 0.0.0.0 net/smc: fix smc_buf_unuse to use the lgr pointer ipv6/ndisc: Preserve IPv6 control buffer if protocol error handlers are called net/{ipv4,ipv6}: Do not put target net if input nsid is invalid lan743x: Remove SPI dependency from Microchip group. drivers: net: remove <net/busy_poll.h> inclusion when not needed net: phy: genphy_10g_driver: Avoid NULL pointer dereference r8169: fix broken Wake-on-LAN from S5 (poweroff) octeontx2-af: Use GFP_ATOMIC under spin lock net: ethernet: cadence: fix socket buffer corruption problem net/ipv6: Allow onlink routes to have a device mismatch if it is the default route net: sched: Remove TCA_OPTIONS from policy ice: Poll for link status change ice: Allocate VF interrupts and set queue map ice: Introduce ice_dev_onetime_setup net: hns3: Fix for warning uninitialized symbol hw_err_lst3 octeontx2-af: Copy the right amount of memory net: udp: fix handling of CHECKSUM_COMPLETE packets ... -
git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparcLinus Torvalds authored
Pull sparc fixes from David Miller: "Some more sparc fixups, mostly aimed at getting the allmodconfig build up and clean again" * git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc: sparc64: Rework xchg() definition to avoid warnings. sparc64: Export __node_distance. sparc64: Make corrupted user stacks more debuggable.
-
- 26 Oct, 2018 25 commits
-
-
Mike Kravetz authored
Some test systems were experiencing negative huge page reserve counts and incorrect file block counts. This was traced to /proc/sys/vm/drop_caches removing clean pages from hugetlbfs file pagecaches. When non-hugetlbfs explicit code removes the pages, the appropriate accounting is not performed. This can be recreated as follows: fallocate -l 2M /dev/hugepages/foo echo 1 > /proc/sys/vm/drop_caches fallocate -l 2M /dev/hugepages/foo grep -i huge /proc/meminfo AnonHugePages: 0 kB ShmemHugePages: 0 kB HugePages_Total: 2048 HugePages_Free: 2047 HugePages_Rsvd: 18446744073709551615 HugePages_Surp: 0 Hugepagesize: 2048 kB Hugetlb: 4194304 kB ls -lsh /dev/hugepages/foo 4.0M -rw-r--r--. 1 root root 2.0M Oct 17 20:05 /dev/hugepages/foo To address this issue, dirty pages as they are added to pagecache. This can easily be reproduced with fallocate as shown above. Read faulted pages will eventually end up being marked dirty. But there is a window where they are clean and could be impacted by code such as drop_caches. So, just dirty them all as they are added to the pagecache. Link: http://lkml.kernel.org/r/b5be45b8-5afe-56cd-9482-28384699a049@oracle.com Fixes: 6bda666a ("hugepages: fold find_or_alloc_pages into huge_no_page()") Signed-off-by:
Mike Kravetz <mike.kravetz@oracle.com> Acked-by:
Mihcla Hocko <mhocko@suse.com> Reviewed-by:
Khalid Aziz <khalid.aziz@oracle.com> Cc: Hugh Dickins <hughd@google.com> Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com> Cc: "Aneesh Kumar K . V" <aneesh.kumar@linux.vnet.ibm.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: "Kirill A . Shutemov" <kirill.shutemov@linux.intel.com> Cc: Davidlohr Bueso <dave@stgolabs.net> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: <stable@vger.kernel.org> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
-
Omar Sandoval authored
Btrfs currently does not support swap files because swap's use of bmap does not work with copy-on-write and multiple devices. See 35054394 ("Btrfs: stop providing a bmap operation to avoid swapfile corruptions"). However, the swap code has a mechanism for the filesystem to manually add swap extents using add_swap_extent() from the ->swap_activate() aop. iomap has done this since 67482129 ("iomap: add a swapfile activation function"). Btrfs will do the same in a later patch, so export add_swap_extent(). Link: http://lkml.kernel.org/r/bb1208575e02829aae51b538709476964f97b1ea.1536704650.git.osandov@fb.comSigned-off-by:
Omar Sandoval <osandov@fb.com> Reviewed-by:
-
Omar Sandoval authored
The SWP_FILE flag serves two purposes: to make swap_{read,write}page() go through the filesystem, and to make swapoff() call ->swap_deactivate(). For Btrfs, we want the latter but not the former, so split this flag into two. This makes us always call ->swap_deactivate() if ->swap_activate() succeeded, not just if it didn't add any swap extents itself. This also resolves the issue of the very misleading name of SWP_FILE, which is only used for swap files over NFS. Link: http://lkml.kernel.org/r/6d63d8668c4287a4f6d203d65696e96f80abdfc7.1536704650.git.osandov@fb.comSigned-off-by:Nikolay Borisov <nborisov@suse.com> Reviewed-by:
-
Michael Ellerman authored
Add a test for MAP_FIXED_NOREPLACE, based on some code originally by Jann Horn. This would have caught the overlap bug reported by Daniel Micay. I originally suggested to Michal that we create MAP_FIXED_NOREPLACE, but instead of writing a selftest I spent my time bike-shedding whether it should be called MAP_FIXED_SAFE/NOCLOBBER/WEAK/NEW .. mea culpa. Link: http://lkml.kernel.org/r/20181013133929.28653-1-mpe@ellerman.id.auSigned-off-by:
Michael Ellerman <mpe@ellerman.id.au> Reviewed-by:
Kees Cook <keescook@chromium.org> Reviewed-by:
-
Andrea Arcangeli authored
There should be no cache left by the time we overwrite the old transhuge pmd with the new one. It's already too late to flush through the virtual address because we already copied the page data to the new physical address. So flush the cache before the data copy. Also delete the "end" variable to shutoff a "unused variable" warning on x86 where flush_cache_range() is a noop. Link: http://lkml.kernel.org/r/20181015202311.7209-1-aarcange@redhat.comSigned-off-by:
Andrea Arcangeli <aarcange@redhat.com> Acked-by:
Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Cc: Aaron Tomlin <atomlin@redhat.com> Cc: Jerome Glisse <jglisse@redhat.com> Cc: Mel Gorman <mgorman@suse.de> Signed-off-by:
-
Andrea Arcangeli authored
change_huge_pmd() after arming the numa/protnone pmd doesn't flush the TLB right away. do_huge_pmd_numa_page() flushes the TLB before calling migrate_misplaced_transhuge_page(). By the time do_huge_pmd_numa_page() runs some CPU could still access the page through the TLB. change_huge_pmd() before arming the numa/protnone transhuge pmd calls mmu_notifier_invalidate_range_start(). So there's no need of mmu_notifier_invalidate_range_start()/mmu_notifier_invalidate_range_only_end() sequence in migrate_misplaced_transhuge_page() too, because by the time migrate_misplaced_transhuge_page() runs, the pmd mapping has already been invalidated in the secondary MMUs. It has to or if a secondary MMU can still write to the page, the migrate_page_copy() would lose data. However an explicit mmu_notifier_invalidate_range() is needed before migrate_misplaced_transhuge_page() starts copying the data of the transhuge page or the below can happen for MMU notifier users sharing the primary MMU pagetables and only implementing ->invalidate_range: CPU0 CPU1 GPU sharing linux pagetables using only ->invalidate_range ----------- ------------ --------- GPU secondary MMU writes to the page mapped by the transhuge pmd change_pmd_range() mmu..._range_start() ->invalidate_range_start() noop change_huge_pmd() set_pmd_at(numa/protnone) pmd_unlock() do_huge_pmd_numa_page() CPU TLB flush globally (1) CPU cannot write to page migrate_misplaced_transhuge_page() GPU writes to the page... migrate_page_copy() ...GPU stops writing to the page CPU TLB flush (2) mmu..._range_end() (3) ->invalidate_range_stop() noop ->invalidate_range() GPU secondary MMU is invalidated and cannot write to the page anymore (too late) Just like we need a CPU TLB flush (1) because the TLB flush (2) arrives too late, we also need a mmu_notifier_invalidate_range() before calling migrate_misplaced_transhuge_page(), because the ->invalidate_range() in (3) also arrives too late. This requirement is the result of the lazy optimization in change_huge_pmd() that releases the pmd_lock without first flushing the TLB and without first calling mmu_notifier_invalidate_range(). Even converting the removed mmu_notifier_invalidate_range_only_end() into a mmu_notifier_invalidate_range_end() would not have been enough to fix this, because it run after migrate_page_copy(). After the hugepage data copy is done migrate_misplaced_transhuge_page() can proceed and call set_pmd_at without having to flush the TLB nor any secondary MMUs because the secondary MMU invalidate, just like the CPU TLB flush, has to happen before the migrate_page_copy() is called or it would be a bug in the first place (and it was for drivers using ->invalidate_range()). KVM is unaffected because it doesn't implement ->invalidate_range(). The standard PAGE_SIZEd migrate_misplaced_page is less accelerated and uses the generic migrate_pages which transitions the pte from numa/protnone to a migration entry in try_to_unmap_one() and flushes TLBs and all mmu notifiers there before copying the page. Link: http://lkml.kernel.org/r/20181013002430.698-3-aarcange@redhat.comSigned-off-by:Mel Gorman <mgorman@suse.de> Acked-by:
Aaron Tomlin <atomlin@redhat.com> Cc: Jerome Glisse <jglisse@redhat.com> Signed-off-by:
-
Andrea Arcangeli authored
Patch series "migrate_misplaced_transhuge_page race conditions". Aaron found a new instance of the THP MADV_DONTNEED race against pmdp_clear_flush* variants, that was apparently left unfixed. While looking into the race found by Aaron, I may have found two more issues in migrate_misplaced_transhuge_page. These race conditions would not cause kernel instability, but they'd corrupt userland data or leave data non zero after MADV_DONTNEED. I did only minor testing, and I don't expect to be able to reproduce this (especially the lack of ->invalidate_range before migrate_page_copy, requires the latest iommu hardware or infiniband to reproduce). The last patch is noop for x86 and it needs further review from maintainers of archs that implement flush_cache_range() (not in CC yet). To avoid confusion, it's not the first patch that introduces the bug fixed in the second patch, even before removing the pmdp_huge_clear_flush_notify, that _notify suffix was called after migrate_page_copy already run. This patch (of 3): This is a corollary of ced10803 ("thp: fix MADV_DONTNEED vs. numa balancing race"), 58ceeb6b ("thp: fix MADV_DONTNEED vs. MADV_FREE race") and 5b7abeae ("thp: fix MADV_DONTNEED vs clear soft dirty race). When the above three fixes where posted Dave asked https://lkml.kernel.org/r/929b3844-aec2-0111-fef7-8002f9d4e2b9@intel.com but apparently this was missed. The pmdp_clear_flush* in migrate_misplaced_transhuge_page() was introduced in a54a407f ("mm: Close races between THP migration and PMD numa clearing"). The important part of such commit is only the part where the page lock is not released until the first do_huge_pmd_numa_page() finished disarming the pagenuma/protnone. The addition of pmdp_clear_flush() wasn't beneficial to such commit and there's no commentary about such an addition either. I guess the pmdp_clear_flush() in such commit was added just in case for safety, but it ended up introducing the MADV_DONTNEED race condition found by Aaron. At that point in time nobody thought of such kind of MADV_DONTNEED race conditions yet (they were fixed later) so the code may have looked more robust by adding the pmdp_clear_flush(). This specific race condition won't destabilize the kernel, but it can confuse userland because after MADV_DONTNEED the memory won't be zeroed out. This also optimizes the code and removes a superfluous TLB flush. [akpm@linux-foundation.org: reflow comment to 80 cols, fix grammar and typo (beacuse)] Link: http://lkml.kernel.org/r/20181013002430.698-2-aarcange@redhat.comSigned-off-by:
-
Clark Williams authored
The static lock quarantine_lock is used in quarantine.c to protect the quarantine queue datastructures. It is taken inside quarantine queue manipulation routines (quarantine_put(), quarantine_reduce() and quarantine_remove_cache()), with IRQs disabled. This is not a problem on a stock kernel but is problematic on an RT kernel where spin locks are sleeping spinlocks, which can sleep and can not be acquired with disabled interrupts. Convert the quarantine_lock to a raw spinlock_t. The usage of quarantine_lock is confined to quarantine.c and the work performed while the lock is held is used for debug purpose. [bigeasy@linutronix.de: slightly altered the commit message] Link: http://lkml.kernel.org/r/20181010214945.5owshc3mlrh74z4b@linutronix.deSigned-off-by:
Clark Williams <williams@redhat.com> Signed-off-by:
Sebastian Andrzej Siewior <bigeasy@linutronix.de> Acked-by:
Dmitry Vyukov <dvyukov@google.com> Signed-off-by:
-
Keith Busch authored
Getting pages from ZONE_DEVICE memory needs to check the backing device's live-ness, which is tracked in the device's dev_pagemap metadata. This metadata is stored in a radix tree and looking it up adds measurable software overhead. This patch avoids repeating this relatively costly operation when dev_pagemap is used by caching the last dev_pagemap while getting user pages. The gup_benchmark kernel self test reports this reduces time to get user pages to as low as 1/3 of the previous time. Link: http://lkml.kernel.org/r/20181012173040.15669-1-keith.busch@intel.comSigned-off-by:
Keith Busch <keith.busch@intel.com> Reviewed-by:
Dan Williams <dan.j.williams@intel.com> Acked-by:
-
Masayoshi Mizuma authored
commit 124049de ("x86/e820: put !E820_TYPE_RAM regions into memblock.reserved") breaks movable_node kernel option because it changed the memory gap range to reserved memblock. So, the node is marked as Normal zone even if the SRAT has Hot pluggable affinity. ===================================================================== kernel: BIOS-e820: [mem 0x0000180000000000-0x0000180fffffffff] usable kernel: BIOS-e820: [mem 0x00001c0000000000-0x00001c0fffffffff] usable ... kernel: reserved[0x12]#011[0x0000181000000000-0x00001bffffffffff], 0x000003f000000000 bytes flags: 0x0 ... kernel: ACPI: SRAT: Node 2 PXM 6 [mem 0x180000000000-0x1bffffffffff] hotplug kernel: ACPI: SRAT: Node 3 PXM 7 [mem 0x1c0000000000-0x1fffffffffff] hotplug ... kernel: Movable zone start for each node kernel: Node 3: 0x00001c0000000000 kernel: Early memory node ranges ... ===================================================================== The original issue is fixed by the former patches, so let's revert commit 124049de ("x86/e820: put !E820_TYPE_RAM regions into memblock.reserved"). Link: http://lkml.kernel.org/r/20181002143821.5112-4-msys.mizuma@gmail.comSigned-off-by:
Masayoshi Mizuma <m.mizuma@jp.fujitsu.com> Reviewed-by:
Pavel Tatashin <pavel.tatashin@microsoft.com> Acked-by:
-
Pavel Tatashin authored
When checking for valid pfns in zero_resv_unavail(), it is not necessary to verify that pfns within pageblock_nr_pages ranges are valid, only the first one needs to be checked. This is because memory for pages are allocated in contiguous chunks that contain pageblock_nr_pages struct pages. Link: http://lkml.kernel.org/r/20181002143821.5112-3-msys.mizuma@gmail.comSigned-off-by:
Naoya Horiguchi <n-horiguchi@ah.jp.nec.com> Reviewed-by:
Oscar Salvador <osalvador@suse.de> Cc: Ingo Molnar <mingo@kernel.org> Cc: Michal Hocko <mhocko@kernel.org> Cc: Thomas Gleixner <tglx@linutronix.de> Signed-off-by:
-
Naoya Horiguchi authored
Patch series "mm: Fix for movable_node boot option", v3. This patch series contains a fix for the movable_node boot option issue which was introduced by commit 124049de ("x86/e820: put !E820_TYPE_RAM regions into memblock.reserved"). The commit breaks the option because it changed the memory gap range to reserved memblock. So, the node is marked as Normal zone even if the SRAT has Hot pluggable affinity. First and second patch fix the original issue which the commit tried to fix, then revert the commit. This patch (of 3): There is a kernel panic that is triggered when reading /proc/kpageflags on the kernel booted with kernel parameter 'memmap=nn[KMG]!ss[KMG]': BUG: unable to handle kernel paging request at fffffffffffffffe PGD 9b20e067 P4D 9b20e067 PUD 9b210067 PMD 0 Oops: 0000 [#1] SMP PTI CPU: 2 PID: 1728 Comm: page-types Not tainted 4.17.0-rc6-mm1-v4.17-rc6-180605-0816-00236-g2dfb086ef02c+ #160 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.11.0-2.fc28 04/01/2014 RIP: 0010:stable_page_flags+0x27/0x3c0 Code: 00 00 00 0f 1f 44 00 00 48 85 ff 0f 84 a0 03 00 00 41 54 55 49 89 fc 53 48 8b 57 08 48 8b 2f 48 8d 42 ff 83 e2 01 48 0f 44 c7 <48> 8b 00 f6 c4 01 0f 84 10 03 00 00 31 db 49 8b 54 24 08 4c 89 e7 RSP: 0018:ffffbbd44111fde0 EFLAGS: 00010202 RAX: fffffffffffffffe RBX: 00007fffffffeff9 RCX: 0000000000000000 RDX: 0000000000000001 RSI: 0000000000000202 RDI: ffffed1182fff5c0 RBP: ffffffffffffffff R08: 0000000000000001 R09: 0000000000000001 R10: ffffbbd44111fed8 R11: 0000000000000000 R12: ffffed1182fff5c0 R13: 00000000000bffd7 R14: 0000000002fff5c0 R15: ffffbbd44111ff10 FS: 00007efc4335a500(0000) GS:ffff93a5bfc00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: fffffffffffffffe CR3: 00000000b2a58000 CR4: 00000000001406e0 Call Trace: kpageflags_read+0xc7/0x120 proc_reg_read+0x3c/0x60 __vfs_read+0x36/0x170 vfs_read+0x89/0x130 ksys_pread64+0x71/0x90 do_syscall_64+0x5b/0x160 entry_SYSCALL_64_after_hwframe+0x44/0xa9 RIP: 0033:0x7efc42e75e23 Code: 09 00 ba 9f 01 00 00 e8 ab 81 f4 ff 66 2e 0f 1f 84 00 00 00 00 00 90 83 3d 29 0a 2d 00 00 75 13 49 89 ca b8 11 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 34 c3 48 83 ec 08 e8 db d3 01 00 48 89 04 24 According to kernel bisection, this problem became visible due to commit f7f99100 which changes how struct pages are initialized. Memblock layout affects the pfn ranges covered by node/zone. Consider that we have a VM with 2 NUMA nodes and each node has 4GB memory, and the default (no memmap= given) memblock layout is like below: MEMBLOCK configuration: memory size = 0x00000001fff75c00 reserved size = 0x000000000300c000 memory.cnt = 0x4 memory[0x0] [0x0000000000001000-0x000000000009efff], 0x000000000009e000 bytes on node 0 flags: 0x0 memory[0x1] [0x0000000000100000-0x00000000bffd6fff], 0x00000000bfed7000 bytes on node 0 flags: 0x0 memory[0x2] [0x0000000100000000-0x000000013fffffff], 0x0000000040000000 bytes on node 0 flags: 0x0 memory[0x3] [0x0000000140000000-0x000000023fffffff], 0x0000000100000000 bytes on node 1 flags: 0x0 ... If you give memmap=1G!4G (so it just covers memory[0x2]), the range [0x100000000-0x13fffffff] is gone: MEMBLOCK configuration: memory size = 0x00000001bff75c00 reserved size = 0x000000000300c000 memory.cnt = 0x3 memory[0x0] [0x0000000000001000-0x000000000009efff], 0x000000000009e000 bytes on node 0 flags: 0x0 memory[0x1] [0x0000000000100000-0x00000000bffd6fff], 0x00000000bfed7000 bytes on node 0 flags: 0x0 memory[0x2] [0x0000000140000000-0x000000023fffffff], 0x0000000100000000 bytes on node 1 flags: 0x0 ... This causes shrinking node 0's pfn range because it is calculated by the address range of memblock.memory. So some of struct pages in the gap range are left uninitialized. We have a function zero_resv_unavail() which does zeroing the struct pages outside memblock.memory, but currently it covers only the reserved unavailable range (i.e. memblock.memory && !memblock.reserved). This patch extends it to cover all unavailable range, which fixes the reported issue. Link: http://lkml.kernel.org/r/20181002143821.5112-2-msys.mizuma@gmail.com Fixes: f7f99100 ("mm: stop zeroing memory during allocation in vmemmap") Signed-off-by:
-
Keith Busch authored
Add a new option '-H' to the gup benchmark to help understand how hugetlb mapping pages compare with the default. Link: http://lkml.kernel.org/r/20181010195605.10689-6-keith.busch@intel.comSigned-off-by:
-
Keith Busch authored
Add a new benchmark option, -S, to request MAP_SHARED. This can be used to compare with MAP_PRIVATE, or for files that require this option, like dax. Link: http://lkml.kernel.org/r/20181010195605.10689-5-keith.busch@intel.comSigned-off-by:
-
Keith Busch authored
Allow a user to specify a file to map by adding a new option, '-f', providing a means to test various file backings. If not specified, the benchmark will use a private mapping of /dev/zero, which produces an anonymous mapping as before. [akpm@linux-foundation.org: avoid using comma operator] Link: http://lkml.kernel.org/r/20181010195605.10689-4-keith.busch@intel.comSigned-off-by:
-
Keith Busch authored
If the '-w' parameter was provided, the benchmark would exit due to a mssing 'break'. Link: http://lkml.kernel.org/r/20181010195605.10689-3-keith.busch@intel.comSigned-off-by:
-
Keith Busch authored
Provide new gup benchmark ioctl commands to run different user page pinning methods, get_user_pages_longterm() and get_user_pages(), in addition to the existing get_user_pages_fast(). Link: http://lkml.kernel.org/r/20181010195605.10689-2-keith.busch@intel.comSigned-off-by:
-
Keith Busch authored
We'd like to measure time to unpin user pages, so this adds a second benchmark timer on put_page, separate from get_page. Adding the field breaks this ioctl ABI, but should be okay since this an in-tree kernel selftest. [akpm@linux-foundation.org: add expansion to struct gup_benchmark for future use] Link: http://lkml.kernel.org/r/20181010195605.10689-1-keith.busch@intel.comSigned-off-by:
-
Roman Gushchin authored
It was reported that on some of our machines containers were restarted with OOM symptoms without an obvious reason. Despite there were almost no memory pressure and plenty of page cache, MEMCG_OOM event was raised occasionally, causing the container management software to think, that OOM has happened. However, no tasks have been killed. The following investigation showed that the problem is caused by a failing attempt to charge a high-order page. In such case, the OOM killer is never invoked. As shown below, it can happen under conditions, which are very far from a real OOM: e.g. there is plenty of clean page cache and no memory pressure. There is no sense in raising an OOM event in this case, as it might confuse a user and lead to wrong and excessive actions (e.g. restart the workload, as in my case). Let's look at the charging path in try_charge(). If the memory usage is about memory.max, which is absolutely natural for most memory cgroups, we try to reclaim some pages. Even if we were able to reclaim enough memory for the allocation, the following check can fail due to a race with another concurrent allocation: if (mem_cgroup_margin(mem_over_limit) >= nr_pages) goto retry; For regular pages the following condition will save us from triggering the OOM: if (nr_reclaimed && nr_pages <= (1 << PAGE_ALLOC_COSTLY_ORDER)) goto retry; But for high-order allocation this condition will intentionally fail. The reason behind is that we'll likely fall to regular pages anyway, so it's ok and even preferred to return ENOMEM. In this case the idea of raising MEMCG_OOM looks dubious. Fix this by moving MEMCG_OOM raising to mem_cgroup_oom() after allocation order check, so that the event won't be raised for high order allocations. This change doesn't affect regular pages allocation and charging. Link: http://lkml.kernel.org/r/20181004214050.7417-1-guro@fb.comSigned-off-by:Roman Gushchin <guro@fb.com> Acked-by:
David Rientjes <rientjes@google.com> Acked-by:
Michal Hocko <mhocko@kernel.org> Acked-by:
Johannes Weiner <hannes@cmpxchg.org> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Signed-off-by:
-
Dave Chinner authored
We've recently seen a workload on XFS filesystems with a repeatable deadlock between background writeback and a multi-process application doing concurrent writes and fsyncs to a small range of a file. range_cyclic writeback Process 1 Process 2 xfs_vm_writepages write_cache_pages writeback_index = 2 cycled = 0 .... find page 2 dirty lock Page 2 ->writepage page 2 writeback page 2 clean page 2 added to bio no more pages write() locks page 1 dirties page 1 locks page 2 dirties page 1 fsync() .... xfs_vm_writepages write_cache_pages start index 0 find page 1 towrite lock Page 1 ->writepage page 1 writeback page 1 clean page 1 added to bio find page 2 towrite lock Page 2 page 2 is writeback <blocks> write() locks page 1 dirties page 1 fsync() .... xfs_vm_writepages write_cache_pages start index 0 !done && !cycled sets index to 0, restarts lookup find page 1 dirty find page 1 towrite lock Page 1 page 1 is writeback <blocks> lock Page 1 <blocks> DEADLOCK because: - process 1 needs page 2 writeback to complete to make enough progress to issue IO pending for page 1 - writeback needs page 1 writeback to complete so process 2 can progress and unlock the page it is blocked on, then it can issue the IO pending for page 2 - process 2 can't make progress until process 1 issues IO for page 1 The underlying cause of the problem here is that range_cyclic writeback is processing pages in descending index order as we hold higher index pages in a structure controlled from above write_cache_pages(). The write_cache_pages() caller needs to be able to submit these pages for IO before write_cache_pages restarts writeback at mapping index 0 to avoid wcp inverting the page lock/writeback wait order. generic_writepages() is not susceptible to this bug as it has no private context held across write_cache_pages() - filesystems using this infrastructure always submit pages in ->writepage immediately and so there is no problem with range_cyclic going back to mapping index 0. However: mpage_writepages() has a private bio context, exofs_writepages() has page_collect fuse_writepages() has fuse_fill_wb_data nfs_writepages() has nfs_pageio_descriptor xfs_vm_writepages() has xfs_writepage_ctx All of these ->writepages implementations can hold pages under writeback in their private structures until write_cache_pages() returns, and hence they are all susceptible to this deadlock. Also worth noting is that ext4 has it's own bastardised version of write_cache_pages() and so it /may/ have an equivalent deadlock. I looked at the code long enough to understand that it has a similar retry loop for range_cyclic writeback reaching the end of the file and then promptly ran away before my eyes bled too much. I'll leave it for the ext4 developers to determine if their code is actually has this deadlock and how to fix it if it has. There's a few ways I can see avoid this deadlock. There's probably more, but these are the first I've though of: 1. get rid of range_cyclic altogether 2. range_cyclic always stops at EOF, and we start again from writeback index 0 on the next call into write_cache_pages() 2a. wcp also returns EAGAIN to ->writepages implementations to indicate range cyclic has hit EOF. writepages implementations can then flush the current context and call wpc again to continue. i.e. lift the retry into the ->writepages implementation 3. range_cyclic uses trylock_page() rather than lock_page(), and it skips pages it can't lock without blocking. It will already do this for pages under writeback, so this seems like a no-brainer 3a. all non-WB_SYNC_ALL writeback uses trylock_page() to avoid blocking as per pages under writeback. I don't think #1 is an option - range_cyclic prevents frequently dirtied lower file offset from starving background writeback of rarely touched higher file offsets. #2 is simple, and I don't think it will have any impact on performance as going back to the start of the file implies an immediate seek. We'll have exactly the same number of seeks if we switch writeback to another inode, and then come back to this one later and restart from index 0. #2a is pretty much "status quo without the deadlock". Moving the retry loop up into the wcp caller means we can issue IO on the pending pages before calling wcp again, and so avoid locking or waiting on pages in the wrong order. I'm not convinced we need to do this given that we get the same thing from #2 on the next writeback call from the writeback infrastructure. #3 is really just a band-aid - it doesn't fix the access/wait inversion problem, just prevents it from becoming a deadlock situation. I'd prefer we fix the inversion, not sweep it under the carpet like this. #3a is really an optimisation that just so happens to include the band-aid fix of #3. So it seems that the simplest way to fix this issue is to implement solution #2 Link: http://lkml.kernel.org/r/20181005054526.21507-1-david@fromorbit.comSigned-off-by:Dave Chinner <dchinner@redhat.com> Reviewed-by:
Jan Kara <jack@suse.de> Cc: Nicholas Piggin <npiggin@gmail.com> Signed-off-by:
-
Pavel Tatashin authored
memmap_init_zone, is getting complex, because it is called from different contexts: hotplug, and during boot, and also because it must handle some architecture quirks. One of them is mirrored memory. Move the code that decides whether to skip mirrored memory outside of memmap_init_zone, into a separate function. [pasha.tatashin@oracle.com: uninline overlap_memmap_init()] Link: http://lkml.kernel.org/r/20180726193509.3326-4-pasha.tatashin@oracle.com Link: http://lkml.kernel.org/r/20180724235520.10200-4-pasha.tatashin@oracle.comSigned-off-by:
Pavel Tatashin <pasha.tatashin@oracle.com> Reviewed-by:
-
Pavel Tatashin authored
update_defer_init() should be called only when struct page is about to be initialized. Because it counts number of initialized struct pages, but there we may skip struct pages if there is some mirrored memory. So move, update_defer_init() after checking for mirrored memory. Also, rename update_defer_init() to defer_init() and reverse the return boolean to emphasize that this is a boolean function, that tells that the reset of memmap initialization should be deferred. Make this function self-contained: do not pass number of already initialized pages in this zone by using static counters. I found this bug by reading the code. The effect is that fewer than expected struct pages are initialized early in boot, and it is possible that in some corner cases we may fail to boot when mirrored pages are used. The deferred on demand code should somewhat mitigate this. But this still brings some inconsistencies compared to when booting without mirrored pages, so it is better to fix. [pasha.tatashin@oracle.com: add comment about defer_init's lack of locking] Link: http://lkml.kernel.org/r/20180726193509.3326-3-pasha.tatashin@oracle.com [akpm@linux-foundation.org: make defer_init non-inline, __meminit] Link: http://lkml.kernel.org/r/20180724235520.10200-3-pasha.tatashin@oracle.comSigned-off-by:
-
Pavel Tatashin authored
memmap_init is sometimes a macro sometimes a function based on __HAVE_ARCH_MEMMAP_INIT. It is only a function on ia64. Make memmap_init a weak function instead, and let ia64 redefine it. Link: http://lkml.kernel.org/r/20180724235520.10200-2-pasha.tatashin@oracle.comSigned-off-by:
-
Kirill Tkhai authored
This will allow to use generic refcount_t interfaces to check counters overflow instead of currently existing VM_BUG_ON(). The only difference after the patch is VM_BUG_ON() may cause BUG(), while refcount_t fires with WARN(). But this seems not to be significant here, since such the problems are usually caught by syzbot with panic-on-warn enabled. Link: http://lkml.kernel.org/r/153910718919.7006.13400779039257185427.stgit@localhost.localdomainSigned-off-by:
Kirill Tkhai <ktkhai@virtuozzo.com> Reviewed-by:
-
David Rientjes authored
If move_freepages_block() returns 0 because !zone_spans_pfn(), *num_movable can hold the value from the stack because it does not get initialized in move_freepages(). Move the initialization to move_freepages_block() to guarantee the value actually makes sense. This currently doesn't affect its only caller where num_movable != NULL, so no bug fix, but just more robust. Link: http://lkml.kernel.org/r/alpine.DEB.2.21.1810051355490.212229@chino.kir.corp.google.comSigned-off-by:
Vlastimil Babka <vbabka@suse.cz> Cc: Greg Thelen <gthelen@google.com> Signed-off-by:
-
