- 14 Apr, 2015 40 commits
-
-
Kirill A. Shutemov authored
We would want to use number of page table level to define mm_struct. Let's expose it as CONFIG_PGTABLE_LEVELS. We need to define PGTABLE_LEVELS before sourcing init/Kconfig: arch/Kconfig will define default value and it's sourced from init/Kconfig. Signed-off-by:
Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Cc: Tony Luck <tony.luck@intel.com> Cc: Fenghua Yu <fenghua.yu@intel.com> Tested-by:
Guenter Roeck <linux@roeck-us.net> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org>
-
Kirill A. Shutemov authored
We would want to use number of page table level to define mm_struct. Let's expose it as CONFIG_PGTABLE_LEVELS. Signed-off-by:
-
Kirill A. Shutemov authored
We would want to use number of page table level to define mm_struct. Let's expose it as CONFIG_PGTABLE_LEVELS. ARM64_PGTABLE_LEVELS is renamed to PGTABLE_LEVELS and defined before sourcing init/Kconfig: arch/Kconfig will define default value and it's sourced from init/Kconfig. Signed-off-by:
Catalin Marinas <catalin.marinas@arm.com> Cc: Will Deacon <will.deacon@arm.com> Tested-by:
-
Kirill A. Shutemov authored
I've implemented accounting for pmd page tables as we have for pte (see mm->nr_ptes). It's requires a new counter in mm_struct: mm->nr_pmds. But the feature doesn't make any sense if an architecture has PMD level folded and it would be nice get rid of the counter in this case. The problem is that we cannot use __PAGETABLE_PMD_FOLDED in <linux/mm_types.h> due to circular dependencies: <linux/mm_types> -> <asm/pgtable.h> -> <linux/mm_types.h> In most cases <asm/pgtable.h> wants <linux/mm_types.h> to get definition of struct page and struct vm_area_struct. I've tried to split mm_struct into separate header file to be able to user <asm/pgtable.h> there. But it doesn't fly on some architectures, like ARM: it wants mm_struct <asm/pgtable.h> to implement tlb flushing. I don't see how to fix it without massive de-inlining or coverting a lot for inline functions to macros. This is other approach: expose number of page tables in use via Kconfig and use it in <linux/mm_types.h> instead of __PAGETABLE_PMD_FOLDED from <asm/pgtable.h>. This patch (of 19): We would want to use number of page table level to define mm_struct. Let's expose it as CONFIG_PGTABLE_LEVELS. Signed-off-by:
Richard Henderson <rth@twiddle.net> Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru> Cc: Matt Turner <mattst88@gmail.com> Tested-by:
-
Konstantin Khlebnikov authored
It seems nobody needs this. Signed-off-by:
Konstantin Khlebnikov <koct9i@gmail.com> Cc: Michal Hocko <mhocko@suse.cz> Cc: David Rientjes <rientjes@google.com> Signed-off-by:
-
Konstantin Khlebnikov authored
This makes show_mem() much less verbose on huge machines. Instead of huge and almost useless dump of counters for each per-zone per-cpu lists this patch prints the sum of these counters for each zone (free_pcp) and size of per-cpu list for current cpu (local_pcp). The filter flag SHOW_MEM_PERCPU_LISTS reverts to the old verbose mode. [akpm@linux-foundation.org: update show_free_areas comment] Signed-off-by:
Konstantin Khlebnikov <khlebnikov@yandex-team.ru> Acked-by:
Michal Hocko <mhocko@suse.cz> Signed-off-by:
-
Konstantin Khlebnikov authored
This patch replaces cancel_dirty_page() with a helper function account_page_cleaned() which only updates counters. It's called from truncate_complete_page() and from try_to_free_buffers() (hack for ext3). Page is locked in both cases, page-lock protects against concurrent dirtiers: see commit 2d6d7f98 ("mm: protect set_page_dirty() from ongoing truncation"). Delete_from_page_cache() shouldn't be called for dirty pages, they must be handled by caller (either written or truncated). This patch treats final dirty accounting fixup at the end of __delete_from_page_cache() as a debug check and adds WARN_ON_ONCE() around it. If something removes dirty pages without proper handling that might be a bug and unwritten data might be lost. Hugetlbfs has no dirty pages accounting, ClearPageDirty() is enough here. cancel_dirty_page() in nfs_wb_page_cancel() is redundant. This is helper for nfs_invalidate_page() and it's called only in case complete invalidation. The mess was started in v2.6.20 after commits 46d2277c ("Clean up and make try_to_free_buffers() not race with dirty pages") and 3e67c098 ("truncate: clear page dirtiness before running try_to_free_buffers()") first was reverted right in v2.6.20 in commit ecdfc978 ("Resurrect 'try_to_free_buffers()' VM hackery"), second in v2.6.25 commit a2b34564 ("Fix dirty page accounting leak with ext3 data=journal"). Custom fixes were introduced between these points. NFS in v2.6.23, commit 1b3b4a1a ("NFS: Fix a write request leak in nfs_invalidate_page()"). Kludge in __delete_from_page_cache() in v2.6.24, commit 3a692790 ("Do dirty page accounting when removing a page from the page cache"). Since v2.6.25 all of them are redundant. [akpm@linux-foundation.org: coding-style fixes] Signed-off-by:
-
Ebru Akagunduz authored
This patch improves THP collapse rates, by allowing zero pages. Currently THP can collapse 4kB pages into a THP when there are up to khugepaged_max_ptes_none pte_none ptes in a 2MB range. This patch counts pte none and mapped zero pages with the same variable. The patch was tested with a program that allocates 800MB of memory, and performs interleaved reads and writes, in a pattern that causes some 2MB areas to first see read accesses, resulting in the zero pfn being mapped there. To simulate memory fragmentation at allocation time, I modified do_huge_pmd_anonymous_page to return VM_FAULT_FALLBACK for read faults. Without the patch, only %50 of the program was collapsed into THP and the percentage did not increase over time. With this patch after 10 minutes of waiting khugepaged had collapsed %99 of the program's memory. [aarcange@redhat.com: fix bogus BUG()] Signed-off-by:
Ebru Akagunduz <ebru.akagunduz@gmail.com> Reviewed-by:
Rik van Riel <riel@redhat.com> Reviewed-by:
Andrea Arcangeli <aarcange@redhat.com> Acked-by:
Vlastimil Babka <vbabka@suse.cz> Cc: Hugh Dickins <hughd@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Sasha Levin <sasha.levin@oracle.com> Cc: David Rientjes <rientjes@google.com> Cc: Mel Gorman <mel@csn.ul.ie> Signed-off-by:
-
Joonsoo Kim authored
Compaction has anti fragmentation algorithm. It is that freepage should be more than pageblock order to finish the compaction if we don't find any freepage in requested migratetype buddy list. This is for mitigating fragmentation, but, there is a lack of migratetype consideration and it is too excessive compared to page allocator's anti fragmentation algorithm. Not considering migratetype would cause premature finish of compaction. For example, if allocation request is for unmovable migratetype, freepage with CMA migratetype doesn't help that allocation and compaction should not be stopped. But, current logic regards this situation as compaction is no longer needed, so finish the compaction. Secondly, condition is too excessive compared to page allocator's logic. We can steal freepage from other migratetype and change pageblock migratetype on more relaxed conditions in page allocator. This is designed to prevent fragmentation and we can use it here. Imposing hard constraint only to the compaction doesn't help much in this case since page allocator would cause fragmentation again. To solve these problems, this patch borrows anti fragmentation logic from page allocator. It will reduce premature compaction finish in some cases and reduce excessive compaction work. stress-highalloc test in mmtests with non movable order 7 allocation shows considerable increase of compaction success rate. Compaction success rate (Compaction success * 100 / Compaction stalls, %) 31.82 : 42.20 I tested it on non-reboot 5 runs stress-highalloc benchmark and found that there is no more degradation on allocation success rate than before. That roughly means that this patch doesn't result in more fragmentations. Vlastimil suggests additional idea that we only test for fallbacks when migration scanner has scanned a whole pageblock. It looked good for fragmentation because chance of stealing increase due to making more free pages in certain pageblock. So, I tested it, but, it results in decreased compaction success rate, roughly 38.00. I guess the reason that if system is low memory condition, watermark check could be failed due to not enough order 0 free page and so, sometimes, we can't reach a fallback check although migrate_pfn is aligned to pageblock_nr_pages. I can insert code to cope with this situation but it makes code more complicated so I don't include his idea at this patch. [akpm@linux-foundation.org: fix CONFIG_CMA=n build] Signed-off-by:
Joonsoo Kim <iamjoonsoo.kim@lge.com> Acked-by:
-
Joonsoo Kim authored
This is preparation step to use page allocator's anti fragmentation logic in compaction. This patch just separates fallback freepage checking part from fallback freepage management part. Therefore, there is no functional change. Signed-off-by:
-
Joonsoo Kim authored
Freepage with MIGRATE_CMA can be used only for MIGRATE_MOVABLE and they should not be expanded to other migratetype buddy list to protect them from unmovable/reclaimable allocation. Implementing these requirements in __rmqueue_fallback(), that is, finding largest possible block of freepage has bad effect that high order freepage with MIGRATE_CMA are broken continually although there are suitable order CMA freepage. Reason is that they are not be expanded to other migratetype buddy list and next __rmqueue_fallback() invocation try to finds another largest block of freepage and break it again. So, MIGRATE_CMA fallback should be handled separately. This patch introduces __rmqueue_cma_fallback(), that just wrapper of __rmqueue_smallest() and call it before __rmqueue_fallback() if migratetype == MIGRATE_MOVABLE. This results in unintended behaviour change that MIGRATE_CMA freepage is always used first rather than other migratetype as movable allocation's fallback. But, as already mentioned above, MIGRATE_CMA can be used only for MIGRATE_MOVABLE, so it is better to use MIGRATE_CMA freepage first as much as possible. Otherwise, we needlessly take up precious freepages with other migratetype and increase chance of fragmentation. Signed-off-by:
-
David Rientjes authored
There's a deadlock when concurrently hot-adding memory through the probe interface and switching a memory block from offline to online. When hot-adding memory via the probe interface, add_memory() first takes mem_hotplug_begin() and then device_lock() is later taken when registering the newly initialized memory block. This creates a lock dependency of (1) mem_hotplug.lock (2) dev->mutex. When switching a memory block from offline to online, dev->mutex is first grabbed in device_online() when the write(2) transitions an existing memory block from offline to online, and then online_pages() will take mem_hotplug_begin(). This creates a lock inversion between mem_hotplug.lock and dev->mutex. Vitaly reports that this deadlock can happen when kworker handling a probe event races with systemd-udevd switching a memory block's state. This patch requires the state transition to take mem_hotplug_begin() before dev->mutex. Hot-adding memory via the probe interface creates a memory block while holding mem_hotplug_begin(), there is no way to take dev->mutex first in this case. online_pages() and offline_pages() are only called when transitioning memory block state. We now require that mem_hotplug_begin() is taken before calling them -- this requires exporting the mem_hotplug_begin() and mem_hotplug_done() to generic code. In all hot-add and hot-remove cases, mem_hotplug_begin() is done prior to device_online(). This is all that is needed to avoid the deadlock. Signed-off-by:
David Rientjes <rientjes@google.com> Reported-by:
Vitaly Kuznetsov <vkuznets@redhat.com> Tested-by:
-
Sasha Levin authored
Document the structure and files under the new debugfs interface. Signed-off-by:
Sasha Levin <sasha.levin@oracle.com> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Laura Abbott <lauraa@codeaurora.org> Cc: Marek Szyprowski <m.szyprowski@samsung.com> Signed-off-by:
-
Andrew Morton authored
s/CONFIG_CMA_ALIGNMENT/0/, per Joonsoo Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Laura Abbott <lauraa@codeaurora.org> Cc: Marek Szyprowski <m.szyprowski@samsung.com> Cc: Sasha Levin <sasha.levin@oracle.com> Signed-off-by:
-
Sasha Levin authored
Provides a userspace interface to trigger a CMA release. Usage: echo [pages] > free This would provide testing/fuzzing access to the CMA release paths. [akpm@linux-foundation.org: coding-style fixes] [mhocko@suse.cz: fix build] Signed-off-by: -
Sasha Levin authored
Provides a userspace interface to trigger a CMA allocation. Usage: echo [pages] > alloc This would provide testing/fuzzing access to the CMA allocation paths. Signed-off-by: -
Sasha Levin authored
I've noticed that there is no interfaces exposed by CMA which would let me fuzz what's going on in there. This small patchset exposes some information out to userspace, plus adds the ability to trigger allocation and freeing from userspace. This patch (of 3): Implement a simple debugfs interface to expose information about CMA areas in the system. Useful for testing/sanity checks for CMA since it was impossible to previously retrieve this information in userspace. Signed-off-by:
-
Sheng Yong authored
Use macro section_nr_to_pfn() to switch between section and pfn, instead of open-coding it. No semantic changes. Signed-off-by:
Sheng Yong <shengyong1@huawei.com> Signed-off-by:
-
Johannes Weiner authored
Add myself to the list of copyright holders. Signed-off-by:
Johannes Weiner <hannes@cmpxchg.org> Acked-by:
-
Baoquan He authored
A small cleanup. Seems in e3239ff9 ("memblock: Rename memblock_region to memblock_type and memblock_property to memblock_region") this one was missed. Signed-off-by:
Baoquan He <bhe@redhat.com> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by:
-
Kirill A. Shutemov authored
It's odd that we have populate_vma_page_range() and __mm_populate() in mm/mlock.c. It's implementation of generic memory population and mlocking is one of possible side effect, if VM_LOCKED is set. __get_user_pages() is core of the implementation. Let's move the code into mm/gup.c. Signed-off-by:
-
Kirill A. Shutemov authored
This is praparation to moving mm_populate()-related code out of mm/mlock.c. Signed-off-by:
-
Kirill A. Shutemov authored
__mlock_vma_pages_range() doesn't necessarily mlock pages. It depends on vma flags. The same codepath is used for MAP_POPULATE. Let's rename __mlock_vma_pages_range() to populate_vma_page_range(). This patch also drops mlock_vma_pages_range() references from documentation. It has gone in cea10a19 ("mm: directly use __mlock_vma_pages_range() in find_extend_vma()"). Signed-off-by:
-
Kirill A. Shutemov authored
After commit a1fde08c ("VM: skip the stack guard page lookup in get_user_pages only for mlock") FOLL_MLOCK has lost its original meaning: we don't necessarily mlock the page if the flags is set -- we also take VM_LOCKED into consideration. Since we use the same codepath for __mm_populate(), let's rename FOLL_MLOCK to FOLL_POPULATE. Signed-off-by:
-
Fabian Frederick authored
slob_alloc_node() is only used in slob.c. Remove the EXPORT_SYMBOL and make slob_alloc_node() static. Signed-off-by:
Fabian Frederick <fabf@skynet.be> Cc: Christoph Lameter <cl@linux.com> Cc: Pekka Enberg <penberg@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Signed-off-by:
-
Joe Perches authored
Use the normal return values for bool functions Signed-off-by:
Joe Perches <joe@perches.com> Cc: Christoph Lameter <cl@linux.com> Cc: Pekka Enberg <penberg@kernel.org> Acked-by:
-
David Rientjes authored
CONFIG_SLAB_DEBUG doesn't exist, CONFIG_DEBUG_SLAB does. Signed-off-by:
-
Chris J Arges authored
By moving the O option detection into the switch statement, we allow this parameter to be combined with other options correctly. Previously options like slub_debug=OFZ would only detect the 'o' and use DEBUG_DEFAULT_FLAGS to fill in the rest of the flags. Signed-off-by:
Chris J Arges <chris.j.arges@canonical.com> Cc: Christoph Lameter <cl@linux.com> Cc: Pekka Enberg <penberg@kernel.org> Acked-by:
-
Geert Uytterhoeven authored
With gcc version 4.7.3 (Ubuntu/Linaro 4.7.3-12ubuntu1) : mm/migrate.c: In function `migrate_pages': mm/migrate.c:1148:1: internal compiler error: in push_minipool_fix, at config/arm/arm.c:13500 Please submit a full bug report, with preprocessed source if appropriate. See <file:///usr/share/doc/gcc-4.7/README.Bugs> for instructions. Preprocessed source stored into /tmp/ccPoM1tr.out file, please attach this to your bugreport. make[1]: *** [mm/migrate.o] Error 1 make: *** [mm/migrate.o] Error 2 Mark unmap_and_move() (which is used in a single place only) "noinline" to work around this compiler bug. [akpm@linux-foundation.org: make it conditional on gcc-4.7.3 and arm] [khilman@kernel.org: fine-tune compiler versions] [akpm@linux-foundation.org: fix comment] Signed-off-by:Geert Uytterhoeven <geert+renesas@glider.be> Reported-by:
Kevin Hilman <khilman@kernel.org> Cc: Marc Zyngier <marc.zyngier@arm.com> Tested-by:
Kevin Hilman <khilman@linaro.org> Tested-by:
Lina Iyer <lina.iyer@linaro.org> Signed-off-by:
-
Ulrich Obergfell authored
Have kvm_guest_init() use hardlockup_detector_disable() instead of watchdog_enable_hardlockup_detector(false). Remove the watchdog_hardlockup_detector_is_enabled() and the watchdog_enable_hardlockup_detector() function which are no longer needed. Signed-off-by:
Ulrich Obergfell <uobergfe@redhat.com> Signed-off-by:
Don Zickus <dzickus@redhat.com> Cc: Ingo Molnar <mingo@elte.hu> Signed-off-by:
-
Ulrich Obergfell authored
Rename the update_timers*() functions to update_watchdog*(). Remove the boolean argument from watchdog_enable_all_cpus() because update_watchdog_all_cpus() is now a generic function to change the run state of the lockup detectors and to have the lockup detectors use a new sample period. Signed-off-by:
-
Ulrich Obergfell authored
With the current user interface of the watchdog mechanism it is only possible to disable or enable both lockup detectors at the same time. This series introduces new kernel parameters and changes the semantics of some existing kernel parameters, so that the hard lockup detector and the soft lockup detector can be disabled or enabled individually. With this series applied, the user interface is as follows. - parameters in /proc/sys/kernel . soft_watchdog This is a new parameter to control and examine the run state of the soft lockup detector. . nmi_watchdog The semantics of this parameter have changed. It can now be used to control and examine the run state of the hard lockup detector. . watchdog This parameter is still available to control the run state of both lockup detectors at the same time. If this parameter is examined, it shows the logical OR of soft_watchdog and nmi_watchdog. . watchdog_thresh The semantics of this parameter are not affected by the patch. - kernel command line parameters . nosoftlockup The semantics of this parameter have changed. It can now be used to disable the soft lockup detector at boot time. . nmi_watchdog=0 or nmi_watchdog=1 Disable or enable the hard lockup detector at boot time. The patch introduces '=1' as a new option. . nowatchdog The semantics of this parameter are not affected by the patch. It is still available to disable both lockup detectors at boot time. Also, remove the proc_dowatchdog() function which is no longer needed. [dzickus@redhat.com: wrote changelog] [dzickus@redhat.com: update documentation for kernel params and sysctl] Signed-off-by: -
Ulrich Obergfell authored
If watchdog_nmi_enable() fails to set up the hardware perf event of one CPU, the entire hard lockup detector is deemed unreliable. Hence, disable the hard lockup detector and shut down the hardware perf events on all CPUs. [dzickus@redhat.com: update comments to explain some code] Signed-off-by:
-
Ulrich Obergfell authored
Separate handlers for each watchdog parameter in /proc/sys/kernel replace the proc_dowatchdog() function. Three of those handlers merely call proc_watchdog_common() with one different argument. Signed-off-by:
-
Ulrich Obergfell authored
Three of four handlers for the watchdog parameters in /proc/sys/kernel essentially have to do the same thing. if the parameter is being read { return the state of the corresponding bit(s) in 'watchdog_enabled' } else { set/clear the state of the corresponding bit(s) in 'watchdog_enabled' update the run state of the lockup detector(s) } Hence, introduce a common function that can be called by those handlers. The callers pass a 'bit mask' to this function to indicate which bit(s) should be set/cleared in 'watchdog_enabled'. This function handles an uncommon race with watchdog_nmi_enable() where a concurrent update of 'watchdog_enabled' is possible. We use 'cmpxchg' to detect the concurrency. [This avoids introducing a new spinlock or a mutex to synchronize updates of 'watchdog_enabled'. Using the same lock or mutex in watchdog thread context and in system call context needs to be considered carefully because it can make the code prone to deadlock situations in connection with parking/unparking the watchdog threads.] Signed-off-by: -
Ulrich Obergfell authored
This series removes proc_dowatchdog(). Since multiple new functions need the 'watchdog_proc_mutex' to serialize access to the watchdog parameters in /proc/sys/kernel, move the mutex outside of any function. Signed-off-by:
-
Ulrich Obergfell authored
This series introduces a separate handler for each watchdog parameter in /proc/sys/kernel. The separate handlers need a common function that they can call to update the run state of the lockup detectors, or to have the lockup detectors use a new sample period. Signed-off-by:
-
Ulrich Obergfell authored
The hardlockup and softockup had always been tied together. Due to the request of KVM folks, they had a need to have one enabled but not the other. Internally rework the code to split things apart more cleanly. There is a bunch of churn here, but the end result should be code that should be easier to maintain and fix without knowing the internals of what is going on. This patch (of 9): Introduce new definitions and variables to separate the user interface in /proc/sys/kernel from the internal run state of the lockup detectors. The internal run state is represented by two bits in a new variable that is named 'watchdog_enabled'. This helps simplify the code, for example: - In order to check if any of the two lockup detectors is enabled, it is sufficient to check if 'watchdog_enabled' is not zero. - In order to enable/disable one or both lockup detectors, it is sufficient to set/clear one or both bits in 'watchdog_enabled'. - Concurrent updates of 'watchdog_enabled' need not be synchronized via a spinlock or a mutex. Updates can either be atomic or concurrency can be detected by using 'cmpxchg'. Signed-off-by:
-
Andrew Morton authored
ocfs2 does mlog_errno(v); return v; in many places. Change mlog_errno() so we can do return mlog_errno(v); For some weird reason this patch reduces the size of ocfs2 by 6k: akpm3:/usr/src/25> size fs/ocfs2/ocfs2.ko text data bss dec hex filename 1146613 82767 832192 2061572 1f7504 fs/ocfs2/ocfs2.ko-before 1140857 82767 832192 2055816 1f5e88 fs/ocfs2/ocfs2.ko-after [dan.carpenter@oracle.com: double evaluation concerns in mlog_errno()] Cc: Mark Fasheh <mfasheh@suse.com> Cc: Joel Becker <jlbec@evilplan.org> Cc: alex chen <alex.chen@huawei.com> Signed-off-by:Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by:
-
alex chen authored
If ocfs2 lockres has not been initialized before calling ocfs2_dlm_lock, the lock won't be dropped and then will lead umount hung. The case is described below: ocfs2_mknod ocfs2_mknod_locked __ocfs2_mknod_locked ocfs2_journal_access_di Failed because of -ENOMEM or other reasons, the inode lockres has not been initialized yet. iput(inode) ocfs2_evict_inode ocfs2_delete_inode ocfs2_inode_lock ocfs2_inode_lock_full_nested __ocfs2_cluster_lock Succeeds and allocates a new dlm lockres. ocfs2_clear_inode ocfs2_open_unlock ocfs2_drop_inode_locks ocfs2_drop_lock Since lockres has not been initialized, the lock can't be dropped and the lockres can't be migrated, thus umount will hang forever. Signed-off-by:Alex Chen <alex.chen@huawei.com> Reviewed-by:
Joseph Qi <joseph.qi@huawei.com> Reviewed-by:
joyce.xue <xuejiufei@huawei.com> Cc: Mark Fasheh <mfasheh@suse.com> Cc: Joel Becker <jlbec@evilplan.org> Signed-off-by:
-
