- 13 Nov, 2013 40 commits
-
-
KOSAKI Motohiro authored
When __rmqueue_fallback() doesn't find a free block with the required size it splits a larger page and puts the rest of the page onto the free list. But it has one serious mistake. When putting back, __rmqueue_fallback() always use start_migratetype if type is not CMA. However, __rmqueue_fallback() is only called when all of the start_migratetype queue is empty. That said, __rmqueue_fallback always puts back memory to the wrong queue except try_to_steal_freepages() changed pageblock type (i.e. requested size is smaller than half of page block). The end result is that the antifragmentation framework increases fragmenation instead of decreasing it. Mel's original anti fragmentation does the right thing. But commit 47118af0 ("mm: mmzone: MIGRATE_CMA migration type added") broke it. This patch restores sane and old behavior. It also removes an incorrect comment which was introduced by commit fef903ef ("mm/page_alloc.c: restructure free-page stealing code and fix a bug"). Signed-off-by:
KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Michal Nazarewicz <mina86@mina86.com> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org>
-
KOSAKI Motohiro authored
In general, every tracepoint should be zero overhead if it is disabled. However, trace_mm_page_alloc_extfrag() is one of exception. It evaluate "new_type == start_migratetype" even if tracepoint is disabled. However, the code can be moved into tracepoint's TP_fast_assign() and TP_fast_assign exist exactly such purpose. This patch does it. Signed-off-by:
Mel Gorman <mgorman@suse.de> Signed-off-by:
-
KOSAKI Motohiro authored
Currently, set_pageblock_migratetype() screws up MIGRATE_CMA and MIGRATE_ISOLATE if page_group_by_mobility_disabled is true. It rewrites the argument to MIGRATE_UNMOVABLE and we lost these attribute. The problem was introduced by commit 49255c61 ("page allocator: move check for disabled anti-fragmentation out of fastpath"). So a 4 year old issue may mean that nobody uses page_group_by_mobility_disabled. But anyway, this patch fixes the problem. Signed-off-by:
-
Damien Ramonda authored
The kernel's readahead algorithm sometimes interprets random read accesses as sequential and triggers unnecessary data prefecthing from storage device (impacting random read average latency). In order to identify sequential cache read misses, the readahead algorithm intends to check whether offset - previous offset == 1 (trivial sequential reads) or offset - previous offset == 0 (sequential reads not aligned on page boundary): if (offset - (ra->prev_pos >> PAGE_CACHE_SHIFT) <= 1UL) The current offset is stored in the "offset" variable of type "pgoff_t" (unsigned long), while previous offset is stored in "ra->prev_pos" of type "loff_t" (long long). Therefore, operands of the if statement are implicitly converted to type long long. Consequently, when previous offset > current offset (which happens on random pattern), the if condition is true and access is wrongly interpeted as sequential. An unnecessary data prefetching is triggered, impacting the average random read latency. Storing the previous offset value in a "pgoff_t" variable (unsigned long) fixes the sequential read detection logic. Signed-off-by:
Damien Ramonda <damien.ramonda@intel.com> Reviewed-by:
Fengguang Wu <fengguang.wu@intel.com> Acked-by:
Pierre Tardy <pierre.tardy@intel.com> Acked-by:
David Cohen <david.a.cohen@linux.intel.com> Signed-off-by:
-
Mel Gorman authored
It has been reported on very large machines that show_mem is taking almost 5 minutes to display information. This is a serious problem if there is an OOM storm. The bulk of the cost is in show_mem doing a very expensive PFN walk to give us the following information Total RAM: Also available as totalram_pages Highmem pages: Also available as totalhigh_pages Reserved pages: Can be inferred from the zone structure Shared pages: PFN walk required Unshared pages: PFN walk required Quick pages: Per-cpu walk required Only the shared/unshared pages requires a full PFN walk but that information is useless. It is also inaccurate as page pins of unshared pages would be accounted for as shared. Even if the information was accurate, I'm struggling to think how the shared/unshared information could be useful for debugging OOM conditions. Maybe it was useful before rmap existed when reclaiming shared pages was costly but it is less relevant today. The PFN walk could be optimised a bit but why bother as the information is useless. This patch deletes the PFN walker and infers the total RAM, highmem and reserved pages count from struct zone. It omits the shared/unshared page usage on the grounds that it is useless. It also corrects the reporting of HighMem as HighMem/MovableOnly as ZONE_MOVABLE has similar problems to HighMem with respect to lowmem/highmem exhaustion. Signed-off-by:
-
Daeseok Youn authored
Signed-off-by:
Daeseok Youn <daeseok.youn@gmail.com> Signed-off-by:
-
Toshi Kani authored
vmstat_cpuup_callback() is a CPU notifier callback, which marks N_CPU to a node at CPU online event. However, it does not update this N_CPU info at CPU offline event. Changed vmstat_cpuup_callback() to clear N_CPU when the last CPU in the node is put into offline, i.e. the node no longer has any online CPU. Signed-off-by:
Toshi Kani <toshi.kani@hp.com> Acked-by:
Christoph Lameter <cl@linux.com> Reviewed-by:
Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com> Tested-by:
-
Toshi Kani authored
After a system booted, N_CPU is not set to any node as has_cpu shows an empty line. # cat /sys/devices/system/node/has_cpu (show-empty-line) setup_vmstat() registers its CPU notifier callback, vmstat_cpuup_callback(), which marks N_CPU to a node when a CPU is put into online. However, setup_vmstat() is called after all CPUs are launched in the boot sequence. Changed setup_vmstat() to mark N_CPU to the nodes with online CPUs at boot, which is consistent with other operations in vmstat_cpuup_callback(), i.e. start_cpu_timer() and refresh_zone_stat_thresholds(). Also added get_online_cpus() to protect the for_each_online_cpu() loop. Signed-off-by:
-
Tang Chen authored
The hot-Pluggable field in SRAT specifies which memory is hotpluggable. As we mentioned before, if hotpluggable memory is used by the kernel, it cannot be hot-removed. So memory hotplug users may want to set all hotpluggable memory in ZONE_MOVABLE so that the kernel won't use it. Memory hotplug users may also set a node as movable node, which has ZONE_MOVABLE only, so that the whole node can be hot-removed. But the kernel cannot use memory in ZONE_MOVABLE. By doing this, the kernel cannot use memory in movable nodes. This will cause NUMA performance down. And other users may be unhappy. So we need a way to allow users to enable and disable this functionality. In this patch, we introduce movable_node boot option to allow users to choose to not to consume hotpluggable memory at early boot time and later we can set it as ZONE_MOVABLE. To achieve this, the movable_node boot option will control the memblock allocation direction. That said, after memblock is ready, before SRAT is parsed, we should allocate memory near the kernel image as we explained in the previous patches. So if movable_node boot option is set, the kernel does the following: 1. After memblock is ready, make memblock allocate memory bottom up. 2. After SRAT is parsed, make memblock behave as default, allocate memory top down. Users can specify "movable_node" in kernel commandline to enable this functionality. For those who don't use memory hotplug or who don't want to lose their NUMA performance, just don't specify anything. The kernel will work as before. Signed-off-by:
Tang Chen <tangchen@cn.fujitsu.com> Signed-off-by:
Zhang Yanfei <zhangyanfei@cn.fujitsu.com> Suggested-by:
Kamezawa Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Suggested-by:
Ingo Molnar <mingo@kernel.org> Acked-by:
Tejun Heo <tj@kernel.org> Acked-by:
-
Tang Chen authored
Memory reserved for crashkernel could be large. So we should not allocate this memory bottom up from the end of kernel image. When SRAT is parsed, we will be able to know which memory is hotpluggable, and we can avoid allocating this memory for the kernel. So reorder reserve_crashkernel() after SRAT is parsed. Signed-off-by:
-
Tang Chen authored
The Linux kernel cannot migrate pages used by the kernel. As a result, kernel pages cannot be hot-removed. So we cannot allocate hotpluggable memory for the kernel. In a memory hotplug system, any numa node the kernel resides in should be unhotpluggable. And for a modern server, each node could have at least 16GB memory. So memory around the kernel image is highly likely unhotpluggable. ACPI SRAT (System Resource Affinity Table) contains the memory hotplug info. But before SRAT is parsed, memblock has already started to allocate memory for the kernel. So we need to prevent memblock from doing this. So direct memory mapping page tables setup is the case. init_mem_mapping() is called before SRAT is parsed. To prevent page tables being allocated within hotpluggable memory, we will use bottom-up direction to allocate page tables from the end of kernel image to the higher memory. Note: As for allocating page tables in lower memory, TJ said: : This is an optional behavior which is triggered by a very specific kernel : boot param, which I suspect is gonna need to stick around to support : memory hotplug in the current setup unless we add another layer of address : translation to support memory hotplug. As for page tables may occupy too much lower memory if using 4K mapping (CONFIG_DEBUG_PAGEALLOC and CONFIG_KMEMCHECK both disable using >4k pages), TJ said: : But as I said in the same paragraph, parsing SRAT earlier doesn't solve : the problem in itself either. Ignoring the option if 4k mapping is : required and memory consumption would be prohibitive should work, no? : Something like that would be necessary if we're gonna worry about cases : like this no matter how we implement it, but, frankly, I'm not sure this : is something worth worrying about. Signed-off-by:
-
Tang Chen authored
Create a new function memory_map_top_down to factor out of the top-down direct memory mapping pagetable setup. This is also a preparation for the following patch, which will introduce the bottom-up memory mapping. That said, we will put the two ways of pagetable setup into separate functions, and choose to use which way in init_mem_mapping, which makes the code more clear. Signed-off-by:
-
Tang Chen authored
The Linux kernel cannot migrate pages used by the kernel. As a result, kernel pages cannot be hot-removed. So we cannot allocate hotpluggable memory for the kernel. ACPI SRAT (System Resource Affinity Table) contains the memory hotplug info. But before SRAT is parsed, memblock has already started to allocate memory for the kernel. So we need to prevent memblock from doing this. In a memory hotplug system, any numa node the kernel resides in should be unhotpluggable. And for a modern server, each node could have at least 16GB memory. So memory around the kernel image is highly likely unhotpluggable. So the basic idea is: Allocate memory from the end of the kernel image and to the higher memory. Since memory allocation before SRAT is parsed won't be too much, it could highly likely be in the same node with kernel image. The current memblock can only allocate memory top-down. So this patch introduces a new bottom-up allocation mode to allocate memory bottom-up. And later when we use this allocation direction to allocate memory, we will limit the start address above the kernel. Signed-off-by:
-
Tang Chen authored
[Problem] The current Linux cannot migrate pages used by the kernel because of the kernel direct mapping. In Linux kernel space, va = pa + PAGE_OFFSET. When the pa is changed, we cannot simply update the pagetable and keep the va unmodified. So the kernel pages are not migratable. There are also some other issues will cause the kernel pages not migratable. For example, the physical address may be cached somewhere and will be used. It is not to update all the caches. When doing memory hotplug in Linux, we first migrate all the pages in one memory device somewhere else, and then remove the device. But if pages are used by the kernel, they are not migratable. As a result, memory used by the kernel cannot be hot-removed. Modifying the kernel direct mapping mechanism is too difficult to do. And it may cause the kernel performance down and unstable. So we use the following way to do memory hotplug. [What we are doing] In Linux, memory in one numa node is divided into several zones. One of the zones is ZONE_MOVABLE, which the kernel won't use. In order to implement memory hotplug in Linux, we are going to arrange all hotpluggable memory in ZONE_MOVABLE so that the kernel won't use these memory. To do this, we need ACPI's help. In ACPI, SRAT(System Resource Affinity Table) contains NUMA info. The memory affinities in SRAT record every memory range in the system, and also, flags specifying if the memory range is hotpluggable. (Please refer to ACPI spec 5.0 5.2.16) With the help of SRAT, we have to do the following two things to achieve our goal: 1. When doing memory hot-add, allow the users arranging hotpluggable as ZONE_MOVABLE. (This has been done by the MOVABLE_NODE functionality in Linux.) 2. when the system is booting, prevent bootmem allocator from allocating hotpluggable memory for the kernel before the memory initialization finishes. The problem 2 is the key problem we are going to solve. But before solving it, we need some preparation. Please see below. [Preparation] Bootloader has to load the kernel image into memory. And this memory must be unhotpluggable. We cannot prevent this anyway. So in a memory hotplug system, we can assume any node the kernel resides in is not hotpluggable. Before SRAT is parsed, we don't know which memory ranges are hotpluggable. But memblock has already started to work. In the current kernel, memblock allocates the following memory before SRAT is parsed: setup_arch() |->memblock_x86_fill() /* memblock is ready */ |...... |->early_reserve_e820_mpc_new() /* allocate memory under 1MB */ |->reserve_real_mode() /* allocate memory under 1MB */ |->init_mem_mapping() /* allocate page tables, about 2MB to map 1GB memory */ |->dma_contiguous_reserve() /* specified by user, should be low */ |->setup_log_buf() /* specified by user, several mega bytes */ |->relocate_initrd() /* could be large, but will be freed after boot, should reorder */ |->acpi_initrd_override() /* several mega bytes */ |->reserve_crashkernel() /* could be large, should reorder */ |...... |->initmem_init() /* Parse SRAT */ According to Tejun's advice, before SRAT is parsed, we should try our best to allocate memory near the kernel image. Since the whole node the kernel resides in won't be hotpluggable, and for a modern server, a node may have at least 16GB memory, allocating several mega bytes memory around the kernel image won't cross to hotpluggable memory. [About this patchset] So this patchset is the preparation for the problem 2 that we want to solve. It does the following: 1. Make memblock be able to allocate memory bottom up. 1) Keep all the memblock APIs' prototype unmodified. 2) When the direction is bottom up, keep the start address greater than the end of kernel image. 2. Improve init_mem_mapping() to support allocate page tables in bottom up direction. 3. Introduce "movable_node" boot option to enable and disable this functionality. This patch (of 6): Create a new function __memblock_find_range_top_down to factor out of top-down allocation from memblock_find_in_range_node. This is a preparation because we will introduce a new bottom-up allocation mode in the following patch. Signed-off-by: -
Heiko Carstens authored
Implement mmap base randomization for the bottom up direction, so ASLR works for both mmap layouts on s390. See also commit df54d6fa ("x86 get_unmapped_area(): use proper mmap base for bottom-up direction"). Signed-off-by:
Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Radu Caragea <sinaelgl@gmail.com> Cc: Michel Lespinasse <walken@google.com> Cc: Catalin Marinas <catalin.marinas@arm.com> Cc: Will Deacon <will.deacon@arm.com> Cc: Chris Metcalf <cmetcalf@tilera.com> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Signed-off-by:
-
Heiko Carstens authored
This is more or less the generic variant of commit 41aacc1e ("x86 get_unmapped_area: Access mmap_legacy_base through mm_struct member"). So effectively architectures which use an own arch_pick_mmap_layout() implementation but call the generic arch_get_unmapped_area() now can also randomize their mmap_base. All architectures which have an own arch_pick_mmap_layout() and call the generic arch_get_unmapped_area() (arm64, s390, tile) currently set mmap_base to TASK_UNMAPPED_BASE. This is also true for the generic arch_pick_mmap_layout() function. So this change is a no-op currently. Signed-off-by:
-
Weijie Yang authored
Add SetPageReclaim() before __swap_writepage() so that page can be moved to the tail of the inactive list, which can avoid unnecessary page scanning as this page was reclaimed by swap subsystem before. Signed-off-by:
Weijie Yang <weijie.yang@samsung.com> Reviewed-by:
Bob Liu <bob.liu@oracle.com> Reviewed-by:
Minchan Kim <minchan@kernel.org> Acked-by:
Seth Jennings <sjenning@linux.vnet.ibm.com> Signed-off-by:
-
Jan Kara authored
When there are processes heavily creating small files while sync(2) is running, it can easily happen that quite some new files are created between WB_SYNC_NONE and WB_SYNC_ALL pass of sync(2). That can happen especially if there are several busy filesystems (remember that sync traverses filesystems sequentially and waits in WB_SYNC_ALL phase on one fs before starting it on another fs). Because WB_SYNC_ALL pass is slow (e.g. causes a transaction commit and cache flush for each inode in ext3), resulting sync(2) times are rather large. The following script reproduces the problem: function run_writers { for (( i = 0; i < 10; i++ )); do mkdir $1/dir$i for (( j = 0; j < 40000; j++ )); do dd if=/dev/zero of=$1/dir$i/$j bs=4k count=4 &>/dev/null done & done } for dir in "$@"; do run_writers $dir done sleep 40 time sync Fix the problem by disregarding inodes dirtied after sync(2) was called in the WB_SYNC_ALL pass. To allow for this, sync_inodes_sb() now takes a time stamp when sync has started which is used for setting up work for flusher threads. To give some numbers, when above script is run on two ext4 filesystems on simple SATA drive, the average sync time from 10 runs is 267.549 seconds with standard deviation 104.799426. With the patched kernel, the average sync time from 10 runs is 2.995 seconds with standard deviation 0.096. Signed-off-by:Jan Kara <jack@suse.cz> Reviewed-by:
Dave Chinner <dchinner@redhat.com> Signed-off-by:
-
Naoya Horiguchi authored
Soft dirty bit allows us to track which pages are written since the last clear_ref (by "echo 4 > /proc/pid/clear_refs".) This is useful for userspace applications to know their memory footprints. Note that the kernel exposes this flag via bit[55] of /proc/pid/pagemap, and the semantics is not a default one (scheduled to be the default in the near future.) However, it shifts to the new semantics at the first clear_ref, and the users of soft dirty bit always do it before utilizing the bit, so that's not a big deal. Users must avoid relying on the bit in page-types before the first clear_ref. Signed-off-by:
Naoya Horiguchi <n-horiguchi@ah.jp.nec.com> Cc: Wu Fengguang <fengguang.wu@intel.com> Cc: Pavel Emelyanov <xemul@parallels.com> Cc: Cyrill Gorcunov <gorcunov@openvz.org> Signed-off-by:
-
Naoya Horiguchi authored
This flag shows that the VMA is "newly created" and thus represents "dirty" in the task's VM. You can clear it by "echo 4 > /proc/pid/clear_refs." Signed-off-by:
Cyrill Gorcunov <gorcunov@openvz.org> Signed-off-by:
-
Zhang Yanfei authored
Signed-off-by:
-
Krzysztof Kozlowski authored
During swapoff the frontswap_map was NULL-ified before calling frontswap_invalidate_area(). However the frontswap_invalidate_area() exits early if frontswap_map is NULL. Invalidate was never called during swapoff. This patch moves frontswap_map_set() in swapoff just after calling frontswap_invalidate_area() so outside of locks (swap_lock and swap_info_struct->lock). This shouldn't be a problem as during swapon the frontswap_map_set() is called also outside of any locks. Signed-off-by:
Krzysztof Kozlowski <k.kozlowski@samsung.com> Reviewed-by:
-
Seth Jennings authored
Signed-off-by:
-
Catalin Marinas authored
Commit 248ac0e1 ("mm/vmalloc: remove guard page from between vmap blocks") had the side effect of making vmap_area.va_end member point to the next vmap_area.va_start. This was creating an artificial reference to vmalloc'ed objects and kmemleak was rarely reporting vmalloc() leaks. This patch marks the vmap_area containing pointers explicitly and reduces the min ref_count to 2 as vm_struct still contains a reference to the vmalloc'ed object. The kmemleak add_scan_area() function has been improved to allow a SIZE_MAX argument covering the rest of the object (for simpler calling sites). Signed-off-by:
Catalin Marinas <catalin.marinas@arm.com> Signed-off-by:
-
Zhang Yanfei authored
We pass the number of pages which hold page structs of a memory section to free_map_bootmem(). This is right when !CONFIG_SPARSEMEM_VMEMMAP but wrong when CONFIG_SPARSEMEM_VMEMMAP. When CONFIG_SPARSEMEM_VMEMMAP, we should pass the number of pages of a memory section to free_map_bootmem. So the fix is removing the nr_pages parameter. When CONFIG_SPARSEMEM_VMEMMAP, we directly use the prefined marco PAGES_PER_SECTION in free_map_bootmem. When !CONFIG_SPARSEMEM_VMEMMAP, we calculate page numbers needed to hold the page structs for a memory section and use the value in free_map_bootmem(). This was found by reading the code. And I have no machine that support memory hot-remove to test the bug now. Signed-off-by:
Wanpeng Li <liwanp@linux.vnet.ibm.com> Cc: Wen Congyang <wency@cn.fujitsu.com> Cc: Tang Chen <tangchen@cn.fujitsu.com> Cc: Toshi Kani <toshi.kani@hp.com> Cc: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com> Cc: Yinghai Lu <yinghai@kernel.org> Cc: Yasunori Goto <y-goto@jp.fujitsu.com> Cc: Andy Whitcroft <apw@shadowen.org> Signed-off-by:
-
Zhang Yanfei authored
For below functions, - sparse_add_one_section() - kmalloc_section_memmap() - __kmalloc_section_memmap() - __kfree_section_memmap() they are always invoked to operate on one memory section, so it is redundant to always pass a nr_pages parameter, which is the page numbers in one section. So we can directly use predefined macro PAGES_PER_SECTION instead of passing the parameter. Signed-off-by:
-
Ying Han authored
The memory.numa_stat file was not hierarchical. Memory charged to the children was not shown in parent's numa_stat. This change adds the "hierarchical_" stats to the existing stats. The new hierarchical stats include the sum of all children's values in addition to the value of the memcg. Tested: Create cgroup a, a/b and run workload under b. The values of b are included in the "hierarchical_*" under a. $ cd /sys/fs/cgroup $ echo 1 > memory.use_hierarchy $ mkdir a a/b Run workload in a/b: $ (echo $BASHPID >> a/b/cgroup.procs && cat /some/file && bash) & The hierarchical_ fields in parent (a) show use of workload in a/b: $ cat a/memory.numa_stat total=0 N0=0 N1=0 N2=0 N3=0 file=0 N0=0 N1=0 N2=0 N3=0 anon=0 N0=0 N1=0 N2=0 N3=0 unevictable=0 N0=0 N1=0 N2=0 N3=0 hierarchical_total=908 N0=552 N1=317 N2=39 N3=0 hierarchical_file=850 N0=549 N1=301 N2=0 N3=0 hierarchical_anon=58 N0=3 N1=16 N2=39 N3=0 hierarchical_unevictable=0 N0=0 N1=0 N2=0 N3=0 $ cat a/b/memory.numa_stat total=908 N0=552 N1=317 N2=39 N3=0 file=850 N0=549 N1=301 N2=0 N3=0 anon=58 N0=3 N1=16 N2=39 N3=0 unevictable=0 N0=0 N1=0 N2=0 N3=0 hierarchical_total=908 N0=552 N1=317 N2=39 N3=0 hierarchical_file=850 N0=549 N1=301 N2=0 N3=0 hierarchical_anon=58 N0=3 N1=16 N2=39 N3=0 hierarchical_unevictable=0 N0=0 N1=0 N2=0 N3=0 Signed-off-by:
Ying Han <yinghan@google.com> Signed-off-by:
Greg Thelen <gthelen@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Michal Hocko <mhocko@suse.cz> Signed-off-by:
-
Greg Thelen authored
Refactor mem_control_numa_stat_show() to use a new stats structure for smaller and simpler code. This consolidates nearly identical code. text data bss dec hex filename 8,137,679 1,703,496 1,896,448 11,737,623 b31a17 vmlinux.before 8,136,911 1,703,496 1,896,448 11,736,855 b31717 vmlinux.after Signed-off-by: -
Jianguo Wu authored
Use more appropriate NUMA_NO_NODE instead of -1 Signed-off-by:
Jianguo Wu <wujianguo@huawei.com> Acked-by:
David Rientjes <rientjes@google.com> Signed-off-by:
-
Bob Liu authored
Khugepaged will scan/free HPAGE_PMD_NR normal pages and replace with a hugepage which is allocated from the node of the first scanned normal page, but this policy is too rough and may end with unexpected result to upper users. The problem is the original page-balancing among all nodes will be broken after hugepaged started. Thinking about the case if the first scanned normal page is allocated from node A, most of other scanned normal pages are allocated from node B or C.. But hugepaged will always allocate hugepage from node A which will cause extra memory pressure on node A which is not the situation before khugepaged started. This patch try to fix this problem by making khugepaged allocate hugepage from the node which have max record of scaned normal pages hit, so that the effect to original page-balancing can be minimized. The other problem is if normal scanned pages are equally allocated from Node A,B and C, after khugepaged started Node A will still suffer extra memory pressure. Andrew Davidoff reported a related issue several days ago. He wanted his application interleaving among all nodes and "numactl --interleave=all ./test" was used to run the testcase, but the result wasn't not as expected. cat /proc/2814/numa_maps: 7f50bd440000 interleave:0-3 anon=51403 dirty=51403 N0=435 N1=435 N2=435 N3=50098 The end result showed that most pages are from Node3 instead of interleave among node0-3 which was unreasonable. This patch also fix this issue by allocating hugepage round robin from all nodes have the same record, after this patch the result was as expected: 7f78399c0000 interleave:0-3 anon=51403 dirty=51403 N0=12723 N1=12723 N2=13235 N3=12722 The simple testcase is like this: int main() { char *p; int i; int j; for (i=0; i < 200; i++) { p = (char *)malloc(1048576); printf("malloc done\n"); if (p == 0) { printf("Out of memory\n"); return 1; } for (j=0; j < 1048576; j++) { p[j] = 'A'; } printf("touched memory\n"); sleep(1); } printf("enter sleep\n"); while(1) { sleep(100); } } [akpm@linux-foundation.org: make last_khugepaged_target_node local to khugepaged_find_target_node()] Reported-by:Andrew Davidoff <davidoff@qedmf.net> Tested-by:
-
Bob Liu authored
Move alloc_hugepage() to a better place, no need for a seperate #ifndef CONFIG_NUMA Signed-off-by:
Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Mel Gorman <mel@csn.ul.ie> Cc: Andrew Davidoff <davidoff@qedmf.net> Cc: Wanpeng Li <liwanp@linux.vnet.ibm.com> Signed-off-by:
-
Christian Hesse authored
Signed-off-by:
Christian Hesse <mail@eworm.de> Acked-by:
-
Wanpeng Li authored
Don't warn twice in __vmalloc_area_node and __vmalloc_node_range if __vmalloc_area_node allocation failure. This patch reverts commit 46c001a2 ("mm/vmalloc.c: emit the failure message before return"). Signed-off-by:
-
Wanpeng Li authored
The VM_UNINITIALIZED/VM_UNLIST flag introduced by f5252e00 ("mm: avoid null pointer access in vm_struct via /proc/vmallocinfo") is used to avoid accessing the pages field with unallocated page when show_numa_info() is called. This patch moves the check just before show_numa_info in order that some messages still can be dumped via /proc/vmallocinfo. This patch reverts commit d157a558 ("mm/vmalloc.c: check VM_UNINITIALIZED flag in s_show instead of show_numa_info"); Reviewed-by:
-
Wanpeng Li authored
There is a race window between vmap_area tear down and show vmap_area information. A B remove_vm_area spin_lock(&vmap_area_lock); va->vm = NULL; va->flags &= ~VM_VM_AREA; spin_unlock(&vmap_area_lock); spin_lock(&vmap_area_lock); if (va->flags & (VM_LAZY_FREE | VM_LAZY_FREEZING)) return 0; if (!(va->flags & VM_VM_AREA)) { seq_printf(m, "0x%pK-0x%pK %7ld vm_map_ram\n", (void *)va->va_start, (void *)va->va_end, va->va_end - va->va_start); return 0; } free_unmap_vmap_area(va); flush_cache_vunmap free_unmap_vmap_area_noflush unmap_vmap_area free_vmap_area_noflush va->flags |= VM_LAZY_FREE The assumption !VM_VM_AREA represents vm_map_ram allocation is introduced by d4033afd ("mm, vmalloc: iterate vmap_area_list, instead of vmlist, in vmallocinfo()"). However, !VM_VM_AREA also represents vmap_area is being tear down in race window mentioned above. This patch fix it by don't dump any information for !VM_VM_AREA case and also remove (VM_LAZY_FREE | VM_LAZY_FREEING) check since they are not possible for !VM_VM_AREA case. Suggested-by:Joonsoo Kim <iamjoonsoo.kim@lge.com> Acked-by:
-
Wanpeng Li authored
The caller address has already been set in set_vmalloc_vm(), there's no need to set it again in __vmalloc_area_node. Reviewed-by:
-
David Rientjes authored
mpol_to_str() should not fail. Currently, it either fails because the string buffer is too small or because a string hasn't been defined for a mempolicy mode. If a new mempolicy mode is introduced and no string is defined for it, just warn and return "unknown". If the buffer is too small, just truncate the string and return, the same behavior as snprintf(). This also fixes a bug where there was no NULL-byte termination when doing *p++ = '=' and *p++ ':' and maxlen has been reached. Signed-off-by:
-
Jianguo Wu authored
Use more appropriate NUMA_NO_NODE instead of -1 in all archs' module_alloc() Signed-off-by:
-
Naoya Horiguchi authored
Chen Gong pointed out that set/unset_migratetype_isolate() was done in different functions in mm/memory-failure.c, which makes the code less readable/maintainable. So this patch does it in soft_offline_page(). With this patch, we get to hold lock_memory_hotplug() longer but it's not a problem because races between memory hotplug and soft offline are very rare. Signed-off-by:
Chen, Gong <gong.chen@linux.intel.com> Acked-by:
Andi Kleen <ak@linux.intel.com> Reviewed-by:
-
Toshi Kani authored
cpu_up() has #ifdef CONFIG_MEMORY_HOTPLUG code blocks, which call mem_online_node() to put its node online if offlined and then call build_all_zonelists() to initialize the zone list. These steps are specific to memory hotplug, and should be managed in mm/memory_hotplug.c. lock_memory_hotplug() should also be held for the whole steps. For this reason, this patch replaces mem_online_node() with try_online_node(), which performs the whole steps with lock_memory_hotplug() held. try_online_node() is named after try_offline_node() as they have similar purpose. There is no functional change in this patch. Signed-off-by:
-
