linux/mm
Steve Capper 2667f50e8b mm: introduce a general RCU get_user_pages_fast()
This series implements general forms of get_user_pages_fast and
__get_user_pages_fast in core code and activates them for arm and arm64.

These are required for Transparent HugePages to function correctly, as a
futex on a THP tail will otherwise result in an infinite loop (due to the
core implementation of __get_user_pages_fast always returning 0).

Unfortunately, a futex on THP tail can be quite common for certain
workloads; thus THP is unreliable without a __get_user_pages_fast
implementation.

This series may also be beneficial for direct-IO heavy workloads and
certain KVM workloads.

This patch (of 6):

get_user_pages_fast() attempts to pin user pages by walking the page
tables directly and avoids taking locks.  Thus the walker needs to be
protected from page table pages being freed from under it, and needs to
block any THP splits.

One way to achieve this is to have the walker disable interrupts, and rely
on IPIs from the TLB flushing code blocking before the page table pages
are freed.

On some platforms we have hardware broadcast of TLB invalidations, thus
the TLB flushing code doesn't necessarily need to broadcast IPIs; and
spuriously broadcasting IPIs can hurt system performance if done too
often.

This problem has been solved on PowerPC and Sparc by batching up page
table pages belonging to more than one mm_user, then scheduling an
rcu_sched callback to free the pages.  This RCU page table free logic has
been promoted to core code and is activated when one enables
HAVE_RCU_TABLE_FREE.  Unfortunately, these architectures implement their
own get_user_pages_fast routines.

The RCU page table free logic coupled with an IPI broadcast on THP split
(which is a rare event), allows one to protect a page table walker by
merely disabling the interrupts during the walk.

This patch provides a general RCU implementation of get_user_pages_fast
that can be used by architectures that perform hardware broadcast of TLB
invalidations.

It is based heavily on the PowerPC implementation by Nick Piggin.

[akpm@linux-foundation.org: various comment fixes]
Signed-off-by: Steve Capper <steve.capper@linaro.org>
Tested-by: Dann Frazier <dann.frazier@canonical.com>
Reviewed-by: Catalin Marinas <catalin.marinas@arm.com>
Acked-by: Hugh Dickins <hughd@google.com>
Cc: Russell King <rmk@arm.linux.org.uk>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Christoffer Dall <christoffer.dall@linaro.org>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-10-09 22:26:00 -04:00
..
backing-dev.c mm: clean up zone flags 2014-10-09 22:25:57 -04:00
balloon_compaction.c
bootmem.c mm/bootmem.c: use include/linux/ headers 2014-10-09 22:26:00 -04:00
cleancache.c
cma.c mm: cma: adjust address limit to avoid hitting low/high memory boundary 2014-10-09 22:25:53 -04:00
compaction.c mm/compaction.c: fix warning of 'flags' may be used uninitialized 2014-10-09 22:25:57 -04:00
debug-pagealloc.c
debug.c mm/debug.c: use pr_emerg() 2014-10-09 22:25:59 -04:00
dmapool.c mm/dmapool.c: fixed a brace coding style issue 2014-10-09 22:26:00 -04:00
early_ioremap.c mm: create generic early_ioremap() support 2014-04-07 16:36:15 -07:00
fadvise.c
failslab.c
filemap_xip.c
filemap.c mm/filemap.c: remove trailing whitespace 2014-10-09 22:26:00 -04:00
fremap.c mm: mark remap_file_pages() syscall as deprecated 2014-06-06 16:08:17 -07:00
frontswap.c swap: change swap_list_head to plist, add swap_avail_head 2014-06-04 16:54:07 -07:00
gup.c mm: introduce a general RCU get_user_pages_fast() 2014-10-09 22:26:00 -04:00
highmem.c mm/highmem: make kmap cache coloring aware 2014-08-06 18:01:22 -07:00
huge_memory.c mm: use VM_BUG_ON_MM where possible 2014-10-09 22:25:58 -04:00
hugetlb_cgroup.c hugetlb_cgroup: use lockdep_assert_held rather than spin_is_locked 2014-08-29 16:28:16 -07:00
hugetlb.c mm: convert a few VM_BUG_ON callers to VM_BUG_ON_VMA 2014-10-09 22:25:57 -04:00
hwpoison-inject.c mm/hwpoison-inject.c: remove unnecessary null test before debugfs_remove_recursive 2014-08-06 18:01:19 -07:00
init-mm.c
internal.h mm, compaction: pass gfp mask to compact_control 2014-10-09 22:25:55 -04:00
interval_tree.c mm: convert a few VM_BUG_ON callers to VM_BUG_ON_VMA 2014-10-09 22:25:57 -04:00
iov_iter.c fuse: honour max_read and max_write in direct_io mode 2014-09-26 21:16:51 -04:00
Kconfig mm: introduce a general RCU get_user_pages_fast() 2014-10-09 22:26:00 -04:00
Kconfig.debug
kmemcheck.c mm/slab_common: move kmem_cache definition to internal header 2014-10-09 22:25:50 -04:00
kmemleak-test.c mm/kmemleak-test.c: use pr_fmt for logging 2014-06-06 16:08:18 -07:00
kmemleak.c mm: introduce kmemleak_update_trace() 2014-06-06 16:08:17 -07:00
ksm.c mm: ksm use pr_err instead of printk 2014-10-09 22:26:00 -04:00
list_lru.c mm: keep page cache radix tree nodes in check 2014-04-03 16:21:01 -07:00
maccess.c
madvise.c mm: update the description for madvise_remove 2014-08-06 18:01:18 -07:00
Makefile mm: move debug code out of page_alloc.c 2014-10-09 22:25:58 -04:00
memblock.c mem-hotplug: let memblock skip the hotpluggable memory regions in __next_mem_range() 2014-09-10 15:42:12 -07:00
memcontrol.c memcg: zap memcg_can_account_kmem 2014-10-09 22:26:00 -04:00
memory_hotplug.c memory-hotplug: add sysfs valid_zones attribute 2014-10-09 22:25:52 -04:00
memory-failure.c hwpoison: fix race with changing page during offlining 2014-08-06 18:01:19 -07:00
memory.c mm: softdirty: keep bit when zapping file pte 2014-09-26 08:10:35 -07:00
mempolicy.c mempolicy: unexport get_vma_policy() and remove its "task" arg 2014-10-09 22:25:56 -04:00
mempool.c mm/mempool.c: update the kmemleak stack trace for mempool allocations 2014-06-06 16:08:17 -07:00
migrate.c mm: migrate: Close race between migration completion and mprotect 2014-10-02 11:57:18 -07:00
mincore.c mm + fs: prepare for non-page entries in page cache radix trees 2014-04-03 16:21:00 -07:00
mlock.c mm: use VM_BUG_ON_MM where possible 2014-10-09 22:25:58 -04:00
mm_init.c
mmap.c mm: use VM_BUG_ON_MM where possible 2014-10-09 22:25:58 -04:00
mmu_context.c
mmu_notifier.c kvm: Fix page ageing bugs 2014-09-24 14:07:58 +02:00
mmzone.c
mprotect.c mm: move mmu notifier call from change_protection to change_pmd_range 2014-04-07 16:35:50 -07:00
mremap.c mm/mremap.c: use linux headers 2014-10-09 22:26:00 -04:00
msync.c msync: fix incorrect fstart calculation 2014-07-03 09:21:53 -07:00
nobootmem.c mem-hotplug: let memblock skip the hotpluggable memory regions in __next_mem_range() 2014-09-10 15:42:12 -07:00
nommu.c arm64,ia64,ppc,s390,sh,tile,um,x86,mm: remove default gate area 2014-08-08 15:57:27 -07:00
oom_kill.c mm: clean up zone flags 2014-10-09 22:25:57 -04:00
page_alloc.c mm: move debug code out of page_alloc.c 2014-10-09 22:25:58 -04:00
page_cgroup.c mm/page_cgroup.c: mark functions as static 2014-04-03 16:21:02 -07:00
page_io.c fix __swap_writepage() compile failure on old gcc versions 2014-06-14 19:30:48 -05:00
page_isolation.c
page-writeback.c mm/page-writeback.c: use min3/max3 macros to avoid shadow warnings 2014-10-09 22:25:57 -04:00
pagewalk.c mm: use VM_BUG_ON_MM where possible 2014-10-09 22:25:58 -04:00
percpu-km.c
percpu-vm.c percpu: perform tlb flush after pcpu_map_pages() failure 2014-08-15 16:06:10 -04:00
percpu.c percpu: free percpu allocation info for uniprocessor system 2014-08-16 08:59:02 -04:00
pgtable-generic.c mm: actually clear pmd_numa before invalidating 2014-08-29 16:28:15 -07:00
process_vm_access.c start adding the tag to iov_iter 2014-05-06 17:32:49 -04:00
quicklist.c
readahead.c mm/readahead.c: remove unused file_ra_state from count_history_pages 2014-08-06 18:01:15 -07:00
rmap.c mm: convert a few VM_BUG_ON callers to VM_BUG_ON_VMA 2014-10-09 22:25:57 -04:00
shmem.c include/linux/migrate.h: remove migrate_page #define 2014-10-09 22:25:56 -04:00
slab_common.c memcg: move memcg_update_cache_size() to slab_common.c 2014-10-09 22:25:59 -04:00
slab.c mm/slab.c: use __seq_open_private() instead of seq_open() 2014-10-09 22:25:57 -04:00
slab.h mm/slab: use percpu allocator for cpu cache 2014-10-09 22:25:51 -04:00
slob.c mm/sl[ao]b: always track caller in kmalloc_(node_)track_caller() 2014-10-09 22:25:50 -04:00
slub.c mm/slab_common: commonize slab merge logic 2014-10-09 22:25:51 -04:00
sparse-vmemmap.c
sparse.c mm: use macros from compiler.h instead of __attribute__((...)) 2014-04-07 16:35:54 -07:00
swap_state.c mm: memcontrol: do not kill uncharge batching in free_pages_and_swap_cache 2014-10-09 22:25:59 -04:00
swap.c mm: memcontrol: do not kill uncharge batching in free_pages_and_swap_cache 2014-10-09 22:25:59 -04:00
swapfile.c mm: memcontrol: rewrite uncharge API 2014-08-08 15:57:17 -07:00
truncate.c mm: memcontrol: rewrite uncharge API 2014-08-08 15:57:17 -07:00
util.c proc/maps: make vm_is_stack() logic namespace-friendly 2014-10-09 22:25:50 -04:00
vmacache.c mm,vmacache: optimize overflow system-wide flushing 2014-06-04 16:53:57 -07:00
vmalloc.c mm/vmalloc.c: use seq_open_private() instead of seq_open() 2014-10-09 22:25:56 -04:00
vmpressure.c
vmscan.c mm: memcontrol: fix transparent huge page allocations under pressure 2014-10-09 22:25:59 -04:00
vmstat.c mm: vmscan: only update per-cpu thresholds for online CPU 2014-08-06 18:01:20 -07:00
workingset.c mm: keep page cache radix tree nodes in check 2014-04-03 16:21:01 -07:00
zbud.c mm/zpool: use prefixed module loading 2014-08-29 16:28:16 -07:00
zpool.c mm/zpool: use prefixed module loading 2014-08-29 16:28:16 -07:00
zsmalloc.c mm/zpool: use prefixed module loading 2014-08-29 16:28:16 -07:00
zswap.c mm/zswap.c: add __init to zswap_entry_cache_destroy() 2014-08-08 15:57:18 -07:00