- 12 Feb, 2015 40 commits
-
-
Kirill A. Shutemov authored
Dave noticed that unprivileged process can allocate significant amount of memory -- >500 MiB on x86_64 -- and stay unnoticed by oom-killer and memory cgroup. The trick is to allocate a lot of PMD page tables. Linux kernel doesn't account PMD tables to the process, only PTE. The use-cases below use few tricks to allocate a lot of PMD page tables while keeping VmRSS and VmPTE low. oom_score for the process will be 0. #include <errno.h> #include <stdio.h> #include <stdlib.h> #include <unistd.h> #include <sys/mman.h> #include <sys/prctl.h> #define PUD_SIZE (1UL << 30) #define PMD_SIZE (1UL << 21) #define NR_PUD 130000 int main(void) { char *addr = NULL; unsigned long i; prctl(PR_SET_THP_DISABLE); for (i = 0; i < NR_PUD ; i++) { addr = mmap(addr + PUD_SIZE, PUD_SIZE, PROT_WRITE|PROT_READ, MAP_ANONYMOUS|MAP_PRIVATE, -1, 0); if (addr == MAP_FAILED) { perror("mmap"); break; } *addr = 'x'; munmap(addr, PMD_SIZE); mmap(addr, PMD_SIZE, PROT_WRITE|PROT_READ, MAP_ANONYMOUS|MAP_PRIVATE|MAP_FIXED, -1, 0); if (addr == MAP_FAILED) perror("re-mmap"), exit(1); } printf("PID %d consumed %lu KiB in PMD page tables\n", getpid(), i * 4096 >> 10); return pause(); } The patch addresses the issue by account PMD tables to the process the same way we account PTE. The main place where PMD tables is accounted is __pmd_alloc() and free_pmd_range(). But there're few corner cases: - HugeTLB can share PMD page tables. The patch handles by accounting the table to all processes who share it. - x86 PAE pre-allocates few PMD tables on fork. - Architectures with FIRST_USER_ADDRESS > 0. We need to adjust sanity check on exit(2). Accounting only happens on configuration where PMD page table's level is present (PMD is not folded). As with nr_ptes we use per-mm counter. The counter value is used to calculate baseline for badness score by oom-killer. Signed-off-by:Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Reported-by:
Dave Hansen <dave.hansen@linux.intel.com> Cc: Hugh Dickins <hughd@google.com> Reviewed-by:
Cyrill Gorcunov <gorcunov@openvz.org> Cc: Pavel Emelyanov <xemul@openvz.org> Cc: David Rientjes <rientjes@google.com> Tested-by:
Sedat Dilek <sedat.dilek@gmail.com> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org>
-
Kirill A. Shutemov authored
ARM uses custom implementation of PMD folding in 2-level page table case. Generic code expects to see __PAGETABLE_PMD_FOLDED to be defined if PMD is folded, but ARM doesn't do this. Let's fix it. Defining __PAGETABLE_PMD_FOLDED will drop out unused __pmd_alloc(). It also fixes problems with recently-introduced pmd accounting on ARM without LPAE. Signed-off-by:
Nishanth Menon <nm@ti.com> Reported-by:
Simon Horman <horms@verge.net.au> Tested-by:
Simon Horman <horms+renesas@verge.net.au> Tested-by:
Fabio Estevam <festevam@gmail.com> Tested-by:
Felipe Balbi <balbi@ti.com> Tested-by:
Peter Ujfalusi <peter.ujfalusi@ti.com> Tested-by:
Krzysztof Kozlowski <k.kozlowski@samsung.com> Tested-by:
Geert Uytterhoeven <geert+renesas@glider.be> Cc: Dave Hansen <dave.hansen@linux.intel.com> Cc: Hugh Dickins <hughd@google.com> Cc: Cyrill Gorcunov <gorcunov@openvz.org> Cc: Pavel Emelyanov <xemul@openvz.org> Cc: David Rientjes <rientjes@google.com> Cc: Russell King <linux@arm.linux.org.uk> Signed-off-by:
-
Kirill A. Shutemov authored
If an architecure uses <asm-generic/4level-fixup.h>, build fails if we try to use PUD_SHIFT in generic code: In file included from arch/microblaze/include/asm/bug.h:1:0, from include/linux/bug.h:4, from include/linux/thread_info.h:11, from include/asm-generic/preempt.h:4, from arch/microblaze/include/generated/asm/preempt.h:1, from include/linux/preempt.h:18, from include/linux/spinlock.h:50, from include/linux/mmzone.h:7, from include/linux/gfp.h:5, from include/linux/slab.h:14, from mm/mmap.c:12: mm/mmap.c: In function 'exit_mmap': >> mm/mmap.c:2858:46: error: 'PUD_SHIFT' undeclared (first use in this function) round_up(FIRST_USER_ADDRESS, PUD_SIZE) >> PUD_SHIFT); ^ include/asm-generic/bug.h:86:25: note: in definition of macro 'WARN_ON' int __ret_warn_on = !!(condition); \ ^ mm/mmap.c:2858:46: note: each undeclared identifier is reported only once for each function it appears in round_up(FIRST_USER_ADDRESS, PUD_SIZE) >> PUD_SHIFT); ^ include/asm-generic/bug.h:86:25: note: in definition of macro 'WARN_ON' int __ret_warn_on = !!(condition); \ ^ As with <asm-generic/pgtable-nopud.h>, let's define PUD_SHIFT to PGDIR_SHIFT. Signed-off-by:Wu Fengguang <fengguang.wu@intel.com> Signed-off-by:
-
Kirill A. Shutemov authored
LKP has triggered a compiler warning after my recent patch "mm: account pmd page tables to the process": mm/mmap.c: In function 'exit_mmap': >> mm/mmap.c:2857:2: warning: right shift count >= width of type [enabled by default] The code: > 2857 WARN_ON(mm_nr_pmds(mm) > 2858 round_up(FIRST_USER_ADDRESS, PUD_SIZE) >> PUD_SHIFT); In this, on tile, we have FIRST_USER_ADDRESS defined as 0. round_up() has the same type -- int. PUD_SHIFT. I think the best way to fix it is to define FIRST_USER_ADDRESS as unsigned long. On every arch for consistency. Signed-off-by: -
Kirill A. Shutemov authored
Microblaze uses custom implementation of PMD folding, but doesn't define __PAGETABLE_PMD_FOLDED, which generic code expects to see. Let's fix it. Defining __PAGETABLE_PMD_FOLDED will drop out unused __pmd_alloc(). It also fixes problems with recently-introduced pmd accounting. Signed-off-by:
Guenter Roeck <linux@roeck-us.net> Tested-by:
-
Johannes Weiner authored
The swap controller code is scattered all over the file. Gather all the code that isn't directly needed by the memory controller at the end of the file in its own CONFIG_MEMCG_SWAP section. Signed-off-by:
Johannes Weiner <hannes@cmpxchg.org> Cc: Michal Hocko <mhocko@suse.cz> Reviewed-by:
Vladimir Davydov <vdavydov@parallels.com> Signed-off-by:
-
Johannes Weiner authored
The initialization code for the per-cpu charge stock and the soft limit tree is compact enough to inline it into mem_cgroup_init(). Signed-off-by:
Michal Hocko <mhocko@suse.cz> Reviewed-by:
-
Johannes Weiner authored
- No need to test the node for N_MEMORY. node_online() is enough for node fallback to work in slab, use NUMA_NO_NODE for everything else. - Remove the BUG_ON() for allocation failure. A NULL pointer crash is just as descriptive, and the absent return value check is obvious. - Move local variables to the inner-most blocks. - Point to the tree structure after its initialized, not before, it's just more logical that way. Signed-off-by:
-
George G. Davis authored
The totalcma_pages variable is not updated to account for CMA regions defined via device tree reserved-memory sub-nodes. Fix this omission by moving the calculation of totalcma_pages into cma_init_reserved_mem() instead of cma_declare_contiguous() such that it will include reserved memory used by all CMA regions. Signed-off-by:
George G. Davis <george_davis@mentor.com> Cc: Marek Szyprowski <m.szyprowski@samsung.com> Acked-by:
Michal Nazarewicz <mina86@mina86.com> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com> Cc: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com> Signed-off-by:
-
Michal Hocko authored
Commit 5695be14 ("OOM, PM: OOM killed task shouldn't escape PM suspend") has left a race window when OOM killer manages to note_oom_kill after freeze_processes checks the counter. The race window is quite small and really unlikely and partial solution deemed sufficient at the time of submission. Tejun wasn't happy about this partial solution though and insisted on a full solution. That requires the full OOM and freezer's task freezing exclusion, though. This is done by this patch which introduces oom_sem RW lock and turns oom_killer_disable() into a full OOM barrier. oom_killer_disabled check is moved from the allocation path to the OOM level and we take oom_sem for reading for both the check and the whole OOM invocation. oom_killer_disable() takes oom_sem for writing so it waits for all currently running OOM killer invocations. Then it disable all the further OOMs by setting oom_killer_disabled and checks for any oom victims. Victims are counted via mark_tsk_oom_victim resp. unmark_oom_victim. The last victim wakes up all waiters enqueued by oom_killer_disable(). Therefore this function acts as the full OOM barrier. The page fault path is covered now as well although it was assumed to be safe before. As per Tejun, "We used to have freezing points deep in file system code which may be reacheable from page fault." so it would be better and more robust to not rely on freezing points here. Same applies to the memcg OOM killer. out_of_memory tells the caller whether the OOM was allowed to trigger and the callers are supposed to handle the situation. The page allocation path simply fails the allocation same as before. The page fault path will retry the fault (more on that later) and Sysrq OOM trigger will simply complain to the log. Normally there wouldn't be any unfrozen user tasks after try_to_freeze_tasks so the function will not block. But if there was an OOM killer racing with try_to_freeze_tasks and the OOM victim didn't finish yet then we have to wait for it. This should complete in a finite time, though, because - the victim cannot loop in the page fault handler (it would die on the way out from the exception) - it cannot loop in the page allocator because all the further allocation would fail and __GFP_NOFAIL allocations are not acceptable at this stage - it shouldn't be blocked on any locks held by frozen tasks (try_to_freeze expects lockless context) and kernel threads and work queues are not frozen yet Signed-off-by:
Tejun Heo <tj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net> Signed-off-by:
-
Michal Hocko authored
While touching this area let's convert printk to pr_*. This also makes the printing of continuation lines done properly. Signed-off-by:
-
Michal Hocko authored
While touching this area let's convert printk to pr_*. This also makes the printing of continuation lines done properly. Signed-off-by:
-
Michal Hocko authored
oom_kill_process only sets TIF_MEMDIE flag and sends a signal to the victim. This is basically noop when the task is frozen though because the task sleeps in the uninterruptible sleep. The victim is eventually thawed later when oom_scan_process_thread meets the task again in a later OOM invocation so the OOM killer doesn't live lock. But this is less than optimal. Let's add __thaw_task into mark_tsk_oom_victim after we set TIF_MEMDIE to the victim. We are not checking whether the task is frozen because that would be racy and __thaw_task does that already. oom_scan_process_thread doesn't need to care about freezer anymore as TIF_MEMDIE and freezer are excluded completely now. Signed-off-by:
-
Michal Hocko authored
This patchset addresses a race which was described in the changelog for 5695be14 ("OOM, PM: OOM killed task shouldn't escape PM suspend"): : PM freezer relies on having all tasks frozen by the time devices are : getting frozen so that no task will touch them while they are getting : frozen. But OOM killer is allowed to kill an already frozen task in order : to handle OOM situtation. In order to protect from late wake ups OOM : killer is disabled after all tasks are frozen. This, however, still keeps : a window open when a killed task didn't manage to die by the time : freeze_processes finishes. The original patch hasn't closed the race window completely because that would require a more complex solution as it can be seen by this patchset. The primary motivation was to close the race condition between OOM killer and PM freezer _completely_. As Tejun pointed out, even though the race condition is unlikely the harder it would be to debug weird bugs deep in the PM freezer when the debugging options are reduced considerably. I can only speculate what might happen when a task is still runnable unexpectedly. On a plus side and as a side effect the oom enable/disable has a better (full barrier) semantic without polluting hot paths. I have tested the series in KVM with 100M RAM: - many small tasks (20M anon mmap) which are triggering OOM continually - s2ram which resumes automatically is triggered in a loop echo processors > /sys/power/pm_test while true do echo mem > /sys/power/state sleep 1s done - simple module which allocates and frees 20M in 8K chunks. If it sees freezing(current) then it tries another round of allocation before calling try_to_freeze - debugging messages of PM stages and OOM killer enable/disable/fail added and unmark_oom_victim is delayed by 1s after it clears TIF_MEMDIE and before it wakes up waiters. - rebased on top of the current mmotm which means some necessary updates in mm/oom_kill.c. mark_tsk_oom_victim is now called under task_lock but I think this should be OK because __thaw_task shouldn't interfere with any locking down wake_up_process. Oleg? As expected there are no OOM killed tasks after oom is disabled and allocations requested by the kernel thread are failing after all the tasks are frozen and OOM disabled. I wasn't able to catch a race where oom_killer_disable would really have to wait but I kinda expected the race is really unlikely. [ 242.609330] Killed process 2992 (mem_eater) total-vm:24412kB, anon-rss:2164kB, file-rss:4kB [ 243.628071] Unmarking 2992 OOM victim. oom_victims: 1 [ 243.636072] (elapsed 2.837 seconds) done. [ 243.641985] Trying to disable OOM killer [ 243.643032] Waiting for concurent OOM victims [ 243.644342] OOM killer disabled [ 243.645447] Freezing remaining freezable tasks ... (elapsed 0.005 seconds) done. [ 243.652983] Suspending console(s) (use no_console_suspend to debug) [ 243.903299] kmem_eater: page allocation failure: order:1, mode:0x204010 [...] [ 243.992600] PM: suspend of devices complete after 336.667 msecs [ 243.993264] PM: late suspend of devices complete after 0.660 msecs [ 243.994713] PM: noirq suspend of devices complete after 1.446 msecs [ 243.994717] ACPI: Preparing to enter system sleep state S3 [ 243.994795] PM: Saving platform NVS memory [ 243.994796] Disabling non-boot CPUs ... The first 2 patches are simple cleanups for OOM. They should go in regardless the rest IMO. Patches 3 and 4 are trivial printk -> pr_info conversion and they should go in ditto. The main patch is the last one and I would appreciate acks from Tejun and Rafael. I think the OOM part should be OK (except for __thaw_task vs. task_lock where a look from Oleg would appreciated) but I am not so sure I haven't screwed anything in the freezer code. I have found several surprises there. This patch (of 5): This patch is just a preparatory and it doesn't introduce any functional change. Note: I am utterly unhappy about lowmemory killer abusing TIF_MEMDIE just to wait for the oom victim and to prevent from new killing. This is just a side effect of the flag. The primary meaning is to give the oom victim access to the memory reserves and that shouldn't be necessary here. Signed-off-by:
-
Johannes Weiner authored
Turn the move type enum into flags and give the flags field a shorter name. Once that is done, move_anon() and move_file() are simple enough to just fold them into the callsites. [akpm@linux-foundation.org: tweak MOVE_MASK definition, per Michal] Signed-off-by:
-
Johannes Weiner authored
Introduce the basic control files to account, partition, and limit memory using cgroups in default hierarchy mode. This interface versioning allows us to address fundamental design issues in the existing memory cgroup interface, further explained below. The old interface will be maintained indefinitely, but a clearer model and improved workload performance should encourage existing users to switch over to the new one eventually. The control files are thus: - memory.current shows the current consumption of the cgroup and its descendants, in bytes. - memory.low configures the lower end of the cgroup's expected memory consumption range. The kernel considers memory below that boundary to be a reserve - the minimum that the workload needs in order to make forward progress - and generally avoids reclaiming it, unless there is an imminent risk of entering an OOM situation. - memory.high configures the upper end of the cgroup's expected memory consumption range. A cgroup whose consumption grows beyond this threshold is forced into direct reclaim, to work off the excess and to throttle new allocations heavily, but is generally allowed to continue and the OOM killer is not invoked. - memory.max configures the hard maximum amount of memory that the cgroup is allowed to consume before the OOM killer is invoked. - memory.events shows event counters that indicate how often the cgroup was reclaimed while below memory.low, how often it was forced to reclaim excess beyond memory.high, how often it hit memory.max, and how often it entered OOM due to memory.max. This allows users to identify configuration problems when observing a degradation in workload performance. An overcommitted system will have an increased rate of low boundary breaches, whereas increased rates of high limit breaches, maximum hits, or even OOM situations will indicate internally overcommitted cgroups. For existing users of memory cgroups, the following deviations from the current interface are worth pointing out and explaining: - The original lower boundary, the soft limit, is defined as a limit that is per default unset. As a result, the set of cgroups that global reclaim prefers is opt-in, rather than opt-out. The costs for optimizing these mostly negative lookups are so high that the implementation, despite its enormous size, does not even provide the basic desirable behavior. First off, the soft limit has no hierarchical meaning. All configured groups are organized in a global rbtree and treated like equal peers, regardless where they are located in the hierarchy. This makes subtree delegation impossible. Second, the soft limit reclaim pass is so aggressive that it not just introduces high allocation latencies into the system, but also impacts system performance due to overreclaim, to the point where the feature becomes self-defeating. The memory.low boundary on the other hand is a top-down allocated reserve. A cgroup enjoys reclaim protection when it and all its ancestors are below their low boundaries, which makes delegation of subtrees possible. Secondly, new cgroups have no reserve per default and in the common case most cgroups are eligible for the preferred reclaim pass. This allows the new low boundary to be efficiently implemented with just a minor addition to the generic reclaim code, without the need for out-of-band data structures and reclaim passes. Because the generic reclaim code considers all cgroups except for the ones running low in the preferred first reclaim pass, overreclaim of individual groups is eliminated as well, resulting in much better overall workload performance. - The original high boundary, the hard limit, is defined as a strict limit that can not budge, even if the OOM killer has to be called. But this generally goes against the goal of making the most out of the available memory. The memory consumption of workloads varies during runtime, and that requires users to overcommit. But doing that with a strict upper limit requires either a fairly accurate prediction of the working set size or adding slack to the limit. Since working set size estimation is hard and error prone, and getting it wrong results in OOM kills, most users tend to err on the side of a looser limit and end up wasting precious resources. The memory.high boundary on the other hand can be set much more conservatively. When hit, it throttles allocations by forcing them into direct reclaim to work off the excess, but it never invokes the OOM killer. As a result, a high boundary that is chosen too aggressively will not terminate the processes, but instead it will lead to gradual performance degradation. The user can monitor this and make corrections until the minimal memory footprint that still gives acceptable performance is found. In extreme cases, with many concurrent allocations and a complete breakdown of reclaim progress within the group, the high boundary can be exceeded. But even then it's mostly better to satisfy the allocation from the slack available in other groups or the rest of the system than killing the group. Otherwise, memory.max is there to limit this type of spillover and ultimately contain buggy or even malicious applications. - The original control file names are unwieldy and inconsistent in many different ways. For example, the upper boundary hit count is exported in the memory.failcnt file, but an OOM event count has to be manually counted by listening to memory.oom_control events, and lower boundary / soft limit events have to be counted by first setting a threshold for that value and then counting those events. Also, usage and limit files encode their units in the filename. That makes the filenames very long, even though this is not information that a user needs to be reminded of every time they type out those names. To address these naming issues, as well as to signal clearly that the new interface carries a new configuration model, the naming conventions in it necessarily differ from the old interface. - The original limit files indicate the state of an unset limit with a very high number, and a configured limit can be unset by echoing -1 into those files. But that very high number is implementation and architecture dependent and not very descriptive. And while -1 can be understood as an underflow into the highest possible value, -2 or -10M etc. do not work, so it's not inconsistent. memory.low, memory.high, and memory.max will use the string "infinity" to indicate and set the highest possible value. [akpm@linux-foundation.org: use seq_puts() for basic strings] Signed-off-by: -
Johannes Weiner authored
The unified hierarchy interface for memory cgroups will no longer use "-1" to mean maximum possible resource value. In preparation for this, make the string an argument and let the caller supply it. Signed-off-by:
-
Juergen Gross authored
Especially on 32 bit kernels memory node ranges are printed with 32 bit wide addresses only. Use u64 types and %llx specifiers to print full width of addresses. Signed-off-by:
Juergen Gross <jgross@suse.com> Signed-off-by:
-
Greg Thelen authored
Use BUILD_BUG_ON() to compile assert that memcg string tables are in sync with corresponding enums. There aren't currently any issues with these tables. This is just defensive. Signed-off-by:
Greg Thelen <gthelen@google.com> Acked-by:
-
Vladimir Davydov authored
Since commit b2052564 ("mm: memcontrol: continue cache reclaim from offlined groups") pages charged to a memory cgroup are not reparented when the cgroup is removed. Instead, they are supposed to be reclaimed in a regular way, along with pages accounted to online memory cgroups. However, an lruvec of an offline memory cgroup will sooner or later get so small that it will be scanned only at low scan priorities (see get_scan_count()). Therefore, if there are enough reclaimable pages in big lruvecs, pages accounted to offline memory cgroups will never be scanned at all, wasting memory. Fix this by unconditionally forcing scanning dead lruvecs from kswapd. [akpm@linux-foundation.org: fix build] Signed-off-by:
-
Kirill A. Shutemov authored
Although it was not called, destroy_compound_page() did some potentially useful checks. Let's re-introduce them in free_pages_prepare(), where they can be actually triggered when CONFIG_DEBUG_VM=y. compound_order() assert is already in free_pages_prepare(). We have few checks for tail pages left. Signed-off-by:
-
Kirill A. Shutemov authored
The only caller is __free_one_page(). By the time we should have page->flags to be cleared already: - for 0-order pages though PCP list: free_hot_cold_page() free_pages_prepare() free_pages_check() page->flags &= ~PAGE_FLAGS_CHECK_AT_PREP; <put the page to PCP list> free_pcppages_bulk() page = <withdraw pages from PCP list> __free_one_page(page) - for non-0-order pages: __free_pages_ok() free_pages_prepare() free_pages_check() page->flags &= ~PAGE_FLAGS_CHECK_AT_PREP; free_one_page() __free_one_page() So there's no way PageCompound() will return true in __free_one_page(). Let's remove dead destroy_compound_page() and put assert for page->flags there instead. Signed-off-by:
-
Vlastimil Babka authored
next_zones_zonelist() returns a zoneref pointer, as well as a zone pointer via extra parameter. Since the latter can be trivially obtained by dereferencing the former, the overhead of the extra parameter is unjustified. This patch thus removes the zone parameter from next_zones_zonelist(). Both callers happen to be in the same header file, so it's simple to add the zoneref dereference inline. We save some bytes of code size. add/remove: 0/0 grow/shrink: 0/3 up/down: 0/-105 (-105) function old new delta nr_free_zone_pages 129 115 -14 __alloc_pages_nodemask 2300 2285 -15 get_page_from_freelist 2652 2576 -76 add/remove: 0/0 grow/shrink: 1/0 up/down: 10/0 (10) function old new delta try_to_compact_pages 569 579 +10 Signed-off-by:
Vlastimil Babka <vbabka@suse.cz> Cc: Mel Gorman <mgorman@suse.de> Cc: Zhang Yanfei <zhangyanfei@cn.fujitsu.com> Cc: Minchan Kim <minchan@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Rik van Riel <riel@redhat.com> Cc: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com> Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Michal Hocko <mhocko@suse.cz> Signed-off-by:
-
Vlastimil Babka authored
Expand the usage of the struct alloc_context introduced in the previous patch also for calling try_to_compact_pages(), to reduce the number of its parameters. Since the function is in different compilation unit, we need to move alloc_context definition in the shared mm/internal.h header. With this change we get simpler code and small savings of code size and stack usage: add/remove: 0/0 grow/shrink: 0/1 up/down: 0/-27 (-27) function old new delta __alloc_pages_direct_compact 283 256 -27 add/remove: 0/0 grow/shrink: 0/1 up/down: 0/-13 (-13) function old new delta try_to_compact_pages 582 569 -13 Stack usage of __alloc_pages_direct_compact goes from 24 to none (per scripts/checkstack.pl). Signed-off-by:
-
Vlastimil Babka authored
Introduce struct alloc_context to accumulate the numerous parameters passed between the alloc_pages* family of functions and get_page_from_freelist(). This excludes gfp_flags and alloc_info, which mutate too much along the way, and allocation order, which is conceptually different. The result is shorter function signatures, as well as overal code size and stack usage reductions. bloat-o-meter: add/remove: 0/0 grow/shrink: 1/2 up/down: 127/-310 (-183) function old new delta get_page_from_freelist 2525 2652 +127 __alloc_pages_direct_compact 329 283 -46 __alloc_pages_nodemask 2564 2300 -264 checkstack.pl: function old new __alloc_pages_nodemask 248 200 get_page_from_freelist 168 184 __alloc_pages_direct_compact 40 24 Signed-off-by:
-
Vlastimil Babka authored
The possibility of replacing the numerous parameters of alloc_pages* functions with a single structure has been discussed when Minchan proposed to expand the x86 kernel stack [1]. This series implements the change, along with few more cleanups/microoptimizations. The series is based on next-20150108 and I used gcc 4.8.3 20140627 on openSUSE 13.2 for compiling. Config includess NUMA and COMPACTION. The core change is the introduction of a new struct alloc_context, which looks like this: struct alloc_context { struct zonelist *zonelist; nodemask_t *nodemask; struct zone *preferred_zone; int classzone_idx; int migratetype; enum zone_type high_zoneidx; }; All the contents is mostly constant, except that __alloc_pages_slowpath() changes preferred_zone, classzone_idx and potentially zonelist. But that's not a problem in case control returns to retry_cpuset: in __alloc_pages_nodemask(), those will be reset to initial values again (although it's a bit subtle). On the other hand, gfp_flags and alloc_info mutate so much that it doesn't make sense to put them into alloc_context. Still, the result is one parameter instead of up to 7. This is all in Patch 2. Patch 3 is a step to expand alloc_context usage out of page_alloc.c itself. The function try_to_compact_pages() can also much benefit from the parameter reduction, but it means the struct definition has to be moved to a shared header. Patch 1 should IMHO be included even if the rest is deemed not useful enough. It improves maintainability and also has some code/stack reduction. Patch 4 is OTOH a tiny optimization. Overall bloat-o-meter results: add/remove: 0/0 grow/shrink: 0/4 up/down: 0/-460 (-460) function old new delta nr_free_zone_pages 129 115 -14 __alloc_pages_direct_compact 329 256 -73 get_page_from_freelist 2670 2576 -94 __alloc_pages_nodemask 2564 2285 -279 try_to_compact_pages 582 579 -3 Overall stack sizes per ./scripts/checkstack.pl: old new delta get_page_from_freelist: 184 184 0 __alloc_pages_nodemask 248 200 -48 __alloc_pages_direct_c 40 - -40 try_to_compact_pages 72 72 0 -88 [1] http://marc.info/?l=linux-mm&m=140142462528257&w=2 This patch (of 4): prep_new_page() sets almost everything in the struct page of the page being allocated, except page->pfmemalloc. This is not obvious and has at least once led to a bug where page->pfmemalloc was forgotten to be set correctly, see commit 8fb74b9f ("mm: compaction: partially revert capture of suitable high-order page"). This patch moves the pfmemalloc setting to prep_new_page(), which means it needs to gain alloc_flags parameter. The call to prep_new_page is moved from buffered_rmqueue() to get_page_from_freelist(), which also leads to simpler code. An obsolete comment for buffered_rmqueue() is replaced. In addition to better maintainability there is a small reduction of code and stack usage for get_page_from_freelist(), which inlines the other functions involved. add/remove: 0/0 grow/shrink: 0/1 up/down: 0/-145 (-145) function old new delta get_page_from_freelist 2670 2525 -145 Stack usage is reduced from 184 to 168 bytes. Signed-off-by: -
Kirill A. Shutemov authored
32-bit sparc uses swap instruction to implement set_pte(). It called using GCC inline assembler. But it misses the "memory" clobber to indicate that pte value will be updated in memory. As result GCC doesn't know that it cannot postpone pte pointer dereference which occurs before set_pte() to post-set_pte() time. It leads to real-world bugs -- [1]. In this situation we have code: ptent = ptep_modify_prot_start(mm, addr, pte); ptent = pte_modify(ptent, newprot); ... ptep_modify_prot_commit(mm, addr, pte, ptent); ptep_modify_prot_start() in sparc case is just 'pte' dereference plus pte_clear(). pte_clear() calls broken set_pte(). GCC thinks it's valid to dereference 'pte' again on pte_modify() and gets cleared pte. ptep_modify_prot_commit() puts 'pteent' with pfn==0 back to page table, which eventually leads to the crash. [1] http://lkml.kernel.org/r/54C06B19.8060305@roeck-us.netSigned-off-by:
-
Naoya Horiguchi authored
If __unmap_hugepage_range() tries to unmap the address range over which hugepage migration is on the way, we get the wrong page because pte_page() doesn't work for migration entries. This patch simply clears the pte for migration entries as we do for hwpoison entries. Fixes: 290408d4 ("hugetlb: hugepage migration core") Signed-off-by:
Naoya Horiguchi <n-horiguchi@ah.jp.nec.com> Cc: Hugh Dickins <hughd@google.com> Cc: James Hogan <james.hogan@imgtec.com> Cc: David Rientjes <rientjes@google.com> Cc: Mel Gorman <mel@csn.ul.ie> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Michal Hocko <mhocko@suse.cz> Cc: Rik van Riel <riel@redhat.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Luiz Capitulino <lcapitulino@redhat.com> Cc: Nishanth Aravamudan <nacc@linux.vnet.ibm.com> Cc: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Steve Capper <steve.capper@linaro.org> Cc: <stable@vger.kernel.org> [2.6.36+] Signed-off-by:
-
Naoya Horiguchi authored
There is a race condition between hugepage migration and change_protection(), where hugetlb_change_protection() doesn't care about migration entries and wrongly overwrites them. That causes unexpected results like kernel crash. HWPoison entries also can cause the same problem. This patch adds is_hugetlb_entry_(migration|hwpoisoned) check in this function to do proper actions. Fixes: 290408d4 ("hugetlb: hugepage migration core") Signed-off-by:
-
Naoya Horiguchi authored
When running the test which causes the race as shown in the previous patch, we can hit the BUG "get_page() on refcount 0 page" in hugetlb_fault(). This race happens when pte turns into migration entry just after the first check of is_hugetlb_entry_migration() in hugetlb_fault() passed with false. To fix this, we need to check pte_present() again after huge_ptep_get(). This patch also reorders taking ptl and doing pte_page(), because pte_page() should be done in ptl. Due to this reordering, we need use trylock_page() in page != pagecache_page case to respect locking order. Fixes: 66aebce7 ("hugetlb: fix race condition in hugetlb_fault()") Signed-off-by:
-
Naoya Horiguchi authored
We have a race condition between move_pages() and freeing hugepages, where move_pages() calls follow_page(FOLL_GET) for hugepages internally and tries to get its refcount without preventing concurrent freeing. This race crashes the kernel, so this patch fixes it by moving FOLL_GET code for hugepages into follow_huge_pmd() with taking the page table lock. This patch intentionally removes page==NULL check after pte_page. This is justified because pte_page() never returns NULL for any architectures or configurations. This patch changes the behavior of follow_huge_pmd() for tail pages and then tail pages can be pinned/returned. So the caller must be changed to properly handle the returned tail pages. We could have a choice to add the similar locking to follow_huge_(addr|pud) for consistency, but it's not necessary because currently these functions don't support FOLL_GET flag, so let's leave it for future development. Here is the reproducer: $ cat movepages.c #include <stdio.h> #include <stdlib.h> #include <numaif.h> #define ADDR_INPUT 0x700000000000UL #define HPS 0x200000 #define PS 0x1000 int main(int argc, char *argv[]) { int i; int nr_hp = strtol(argv[1], NULL, 0); int nr_p = nr_hp * HPS / PS; int ret; void **addrs; int *status; int *nodes; pid_t pid; pid = strtol(argv[2], NULL, 0); addrs = malloc(sizeof(char *) * nr_p + 1); status = malloc(sizeof(char *) * nr_p + 1); nodes = malloc(sizeof(char *) * nr_p + 1); while (1) { for (i = 0; i < nr_p; i++) { addrs[i] = (void *)ADDR_INPUT + i * PS; nodes[i] = 1; status[i] = 0; } ret = numa_move_pages(pid, nr_p, addrs, nodes, status, MPOL_MF_MOVE_ALL); if (ret == -1) err("move_pages"); for (i = 0; i < nr_p; i++) { addrs[i] = (void *)ADDR_INPUT + i * PS; nodes[i] = 0; status[i] = 0; } ret = numa_move_pages(pid, nr_p, addrs, nodes, status, MPOL_MF_MOVE_ALL); if (ret == -1) err("move_pages"); } return 0; } $ cat hugepage.c #include <stdio.h> #include <sys/mman.h> #include <string.h> #define ADDR_INPUT 0x700000000000UL #define HPS 0x200000 int main(int argc, char *argv[]) { int nr_hp = strtol(argv[1], NULL, 0); char *p; while (1) { p = mmap((void *)ADDR_INPUT, nr_hp * HPS, PROT_READ | PROT_WRITE, MAP_PRIVATE | MAP_ANONYMOUS | MAP_HUGETLB, -1, 0); if (p != (void *)ADDR_INPUT) { perror("mmap"); break; } memset(p, 0, nr_hp * HPS); munmap(p, nr_hp * HPS); } } $ sysctl vm.nr_hugepages=40 $ ./hugepage 10 & $ ./movepages 10 $(pgrep -f hugepage) Fixes: e632a938 ("mm: migrate: add hugepage migration code to move_pages()") Signed-off-by:Hugh Dickins <hughd@google.com> Cc: James Hogan <james.hogan@imgtec.com> Cc: David Rientjes <rientjes@google.com> Cc: Mel Gorman <mel@csn.ul.ie> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Michal Hocko <mhocko@suse.cz> Cc: Rik van Riel <riel@redhat.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Luiz Capitulino <lcapitulino@redhat.com> Cc: Nishanth Aravamudan <nacc@linux.vnet.ibm.com> Cc: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Steve Capper <steve.capper@linaro.org> Cc: <stable@vger.kernel.org> [3.12+] Signed-off-by:
-
Naoya Horiguchi authored
Migrating hugepages and hwpoisoned hugepages are considered as non-present hugepages, and they are referenced via migration entries and hwpoison entries in their page table slots. This behavior causes race condition because pmd_huge() doesn't tell non-huge pages from migrating/hwpoisoned hugepages. follow_page_mask() is one example where the kernel would call follow_page_pte() for such hugepage while this function is supposed to handle only normal pages. To avoid this, this patch makes pmd_huge() return true when pmd_none() is true *and* pmd_present() is false. We don't have to worry about mixing up non-present pmd entry with normal pmd (pointing to leaf level pte entry) because pmd_present() is true in normal pmd. The same race condition could happen in (x86-specific) gup_pmd_range(), where this patch simply adds pmd_present() check instead of pmd_huge(). This is because gup_pmd_range() is fast path. If we have non-present hugepage in this function, we will go into gup_huge_pmd(), then return 0 at flag mask check, and finally fall back to the slow path. Fixes: 290408d4 ("hugetlb: hugepage migration core") Signed-off-by:
-
Naoya Horiguchi authored
Currently we have many duplicates in definitions around follow_huge_addr(), follow_huge_pmd(), and follow_huge_pud(), so this patch tries to remove the m. The basic idea is to put the default implementation for these functions in mm/hugetlb.c as weak symbols (regardless of CONFIG_ARCH_WANT_GENERAL_HUGETL B), and to implement arch-specific code only when the arch needs it. For follow_huge_addr(), only powerpc and ia64 have their own implementation, and in all other architectures this function just returns ERR_PTR(-EINVAL). So this patch sets returning ERR_PTR(-EINVAL) as default. As for follow_huge_(pmd|pud)(), if (pmd|pud)_huge() is implemented to always return 0 in your architecture (like in ia64 or sparc,) it's never called (the callsite is optimized away) no matter how implemented it is. So in such architectures, we don't need arch-specific implementation. In some architecture (like mips, s390 and tile,) their current arch-specific follow_huge_(pmd|pud)() are effectively identical with the common code, so this patch lets these architecture use the common code. One exception is metag, where pmd_huge() could return non-zero but it expects follow_huge_pmd() to always return NULL. This means that we need arch-specific implementation which returns NULL. This behavior looks strange to me (because non-zero pmd_huge() implies that the architecture supports PMD-based hugepage, so follow_huge_pmd() can/should return some relevant value,) but that's beyond this cleanup patch, so let's keep it. Justification of non-trivial changes: - in s390, follow_huge_pmd() checks !MACHINE_HAS_HPAGE at first, and this patch removes the check. This is OK because we can assume MACHINE_HAS_HPAGE is true when follow_huge_pmd() can be called (note that pmd_huge() has the same check and always returns 0 for !MACHINE_HAS_HPAGE.) - in s390 and mips, we use HPAGE_MASK instead of PMD_MASK as done in common code. This patch forces these archs use PMD_MASK, but it's OK because they are identical in both archs. In s390, both of HPAGE_SHIFT and PMD_SHIFT are 20. In mips, HPAGE_SHIFT is defined as (PAGE_SHIFT + PAGE_SHIFT - 3) and PMD_SHIFT is define as (PAGE_SHIFT + PAGE_SHIFT + PTE_ORDER - 3), but PTE_ORDER is always 0, so these are identical. Signed-off-by:
-
Vlastimil Babka authored
Kswapd in balance_pgdate() currently uses wake_up() on processes waiting in throttle_direct_reclaim(), which only wakes up a single process. This might leave processes waiting for longer than necessary, until the check is reached in the next loop iteration. Processes might also be left waiting if zone was fully balanced in single iteration. Note that the comment in balance_pgdat() also says "Wake them", so waking up a single process does not seem intentional. Thus, replace wake_up() with wake_up_all(). Signed-off-by:
Rik van Riel <riel@redhat.com> Signed-off-by:
-
Baoquan He authored
Found it when I want to jump to the definition of MIGRATE_RESERVE ctags. Signed-off-by:
Baoquan He <bhe@redhat.com> Signed-off-by:
-
Xishi Qiu authored
Now kmemcheck_pagealloc_alloc() is only called by __alloc_pages_slowpath(). __alloc_pages_nodemask() __alloc_pages_slowpath() kmemcheck_pagealloc_alloc() And the page will not be tracked by kmemcheck in the following path. __alloc_pages_nodemask() get_page_from_freelist() So move kmemcheck_pagealloc_alloc() into __alloc_pages_nodemask(), like this: __alloc_pages_nodemask() ... get_page_from_freelist() if (!page) __alloc_pages_slowpath() kmemcheck_pagealloc_alloc() ... Signed-off-by:
Xishi Qiu <qiuxishi@huawei.com> Cc: Vegard Nossum <vegard.nossum@oracle.com> Cc: Pekka Enberg <penberg@kernel.org> Cc: Li Zefan <lizefan@huawei.com> Signed-off-by:
-
Andrew Morton authored
__alloc_pages_nodemask() strips __GFP_IO when retrying the page allocation. But it does this by altering the function-wide variable gfp_mask. This will cause subsequent allocation attempts to inadvertently use the modified gfp_mask. Also, pass the correct mask (the mask we actually used) into trace_mm_page_alloc(). Cc: Ming Lei <ming.lei@canonical.com> Cc: Mel Gorman <mel@csn.ul.ie> Cc: Johannes Weiner <hannes@cmpxchg.org> Reviewed-by:
Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com> Cc: David Rientjes <rientjes@google.com> Acked-by:
-
Johannes Weiner authored
The complexity of memcg page stat synchronization is currently leaking into the callsites, forcing them to keep track of the move_lock state and the IRQ flags. Simplify the API by tracking it in the memcg. Signed-off-by:
-
Vladimir Davydov authored
The body of this function was removed by commit 0a31bc97 ("mm: memcontrol: rewrite uncharge API"). Signed-off-by:
-
Michal Hocko authored
OOM killer tries to exclude tasks which do not have mm_struct associated because killing such a task wouldn't help much. The OOM victim gets TIF_MEMDIE set to disable OOM killer while the current victim releases the memory and then enables the OOM killer again by dropping the flag. oom_kill_process is currently prone to a race condition when the OOM victim is already exiting and TIF_MEMDIE is set after the task releases its address space. This might theoretically lead to OOM livelock if the OOM victim blocks on an allocation later during exiting because it wouldn't kill any other process and the exiting one won't be able to exit. The situation is highly unlikely because the OOM victim is expected to release some memory which should help to sort out OOM situation. Fix this by checking task->mm and setting TIF_MEMDIE flag under task_lock which will serialize the OOM killer with exit_mm which sets task->mm to NULL. Setting the flag for current is not necessary because check and set is not racy. Reported-by:
Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Signed-off-by:
-
