While working on s390 support for gigantic hugepages I ran into the
following "Bad page state" warning when freeing gigantic pages:
BUG: Bad page state in process bash pfn:580001
page:000003d116000040 count:0 mapcount:0 mapping:ffffffff00000000 index:0x0
flags: 0x7fffc0000000000()
page dumped because: non-NULL mapping
This is because page->compound_mapcount, which is part of a union with
page->mapping, is initialized with -1 in prep_compound_gigantic_page(),
and not cleared again during destroy_compound_gigantic_page(). Fix this
by clearing the compound_mapcount in destroy_compound_gigantic_page()
before clearing compound_head.
Interestingly enough, the warning will not show up on x86_64, although
this should not be architecture specific. Apparently there is an
endianness issue, combined with the fact that the union contains both a
64 bit ->mapping pointer and a 32 bit atomic_t ->compound_mapcount as
members. The resulting bogus page->mapping on x86_64 therefore contains
00000000ffffffff instead of ffffffff00000000 on s390, which will falsely
trigger the PageAnon() check in free_pages_prepare() because
page->mapping & PAGE_MAPPING_ANON is true on little-endian architectures
like x86_64 in this case (the page is not compound anymore,
->compound_head was already cleared before). As a result, page->mapping
will be cleared before doing the checks in free_pages_check().
Not sure if the bogus "PageAnon() returning true" on x86_64 for the
first tail page of a gigantic page (at this stage) has other theoretical
implications, but they would also be fixed with this patch.
Link: http://lkml.kernel.org/r/1466612719-5642-1-git-send-email-gerald.schaefer@de.ibm.com
Signed-off-by: Gerald Schaefer <gerald.schaefer@de.ibm.com>
Reviewed-by: Mike Kravetz <mike.kravetz@oracle.com>
Cc: Luiz Capitulino <lcapitulino@redhat.com>
Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Hillf Danton <hillf.zj@alibaba-inc.com>
Cc: "Kirill A . Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Paul Gortmaker <paul.gortmaker@windriver.com>
Cc: "Aneesh Kumar K . V" <aneesh.kumar@linux.vnet.ibm.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Currently we can have compound pages held on per cpu pagevecs, which
leads to a lot of memory unavailable for reclaim when needed. In the
systems with hundreads of processors it can be GBs of memory.
On of the way of reproducing the problem is to not call munmap
explicitly on all mapped regions (i.e. after receiving SIGTERM). After
that some pages (with THP enabled also huge pages) may end up on
lru_add_pvec, example below.
void main() {
#pragma omp parallel
{
size_t size = 55 * 1000 * 1000; // smaller than MEM/CPUS
void *p = mmap(NULL, size, PROT_READ | PROT_WRITE,
MAP_PRIVATE | MAP_ANONYMOUS , -1, 0);
if (p != MAP_FAILED)
memset(p, 0, size);
//munmap(p, size); // uncomment to make the problem go away
}
}
When we run it with THP enabled it will leave significant amount of
memory on lru_add_pvec. This memory will be not reclaimed if we hit
OOM, so when we run above program in a loop:
for i in `seq 100`; do ./a.out; done
many processes (95% in my case) will be killed by OOM.
The primary point of the LRU add cache is to save the zone lru_lock
contention with a hope that more pages will belong to the same zone and
so their addition can be batched. The huge page is already a form of
batched addition (it will add 512 worth of memory in one go) so skipping
the batching seems like a safer option when compared to a potential
excess in the caching which can be quite large and much harder to fix
because lru_add_drain_all is way to expensive and it is not really clear
what would be a good moment to call it.
Similarly we can reproduce the problem on lru_deactivate_pvec by adding:
madvise(p, size, MADV_FREE); after memset.
This patch flushes lru pvecs on compound page arrival making the problem
less severe - after applying it kill rate of above example drops to 0%,
due to reducing maximum amount of memory held on pvec from 28MB (with
THP) to 56kB per CPU.
Suggested-by: Michal Hocko <mhocko@suse.com>
Link: http://lkml.kernel.org/r/1466180198-18854-1-git-send-email-lukasz.odzioba@intel.com
Signed-off-by: Lukasz Odzioba <lukasz.odzioba@intel.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Kirill Shutemov <kirill.shutemov@linux.intel.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Vladimir Davydov <vdavydov@parallels.com>
Cc: Ming Li <mingli199x@qq.com>
Cc: Minchan Kim <minchan@kernel.org>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
We account HugeTLB's shared page table to all processes who share it.
The accounting happens during huge_pmd_share().
If somebody populates pud entry under us, we should decrease pagetable's
refcount and decrease nr_pmds of the process.
By mistake, I increase nr_pmds again in this case. :-/ It will lead to
"BUG: non-zero nr_pmds on freeing mm: 2" on process' exit.
Let's fix this by increasing nr_pmds only when we're sure that the page
table will be used.
Link: http://lkml.kernel.org/r/20160617122506.GC6534@node.shutemov.name
Fixes: dc6c9a35b6 ("mm: account pmd page tables to the process")
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Reported-by: zhongjiang <zhongjiang@huawei.com>
Reviewed-by: Mike Kravetz <mike.kravetz@oracle.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Commit d0164adc89 ("mm, page_alloc: distinguish between being unable
to sleep, unwilling to sleep and avoiding waking kswapd") modified
__GFP_WAIT to explicitly identify the difference between atomic callers
and those that were unwilling to sleep. Later the definition was
removed entirely.
The GFP_RECLAIM_MASK is the set of flags that affect watermark checking
and reclaim behaviour but __GFP_ATOMIC was never added. Without it,
atomic users of the slab allocator strip the __GFP_ATOMIC flag and
cannot access the page allocator atomic reserves. This patch addresses
the problem.
The user-visible impact depends on the workload but potentially atomic
allocations unnecessarily fail without this path.
Link: http://lkml.kernel.org/r/20160610093832.GK2527@techsingularity.net
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Reported-by: Marcin Wojtas <mw@semihalf.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: <stable@vger.kernel.org> [4.4+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Currently we may put reserved by mempool elements into quarantine via
kasan_kfree(). This is totally wrong since quarantine may really free
these objects. So when mempool will try to use such element,
use-after-free will happen. Or mempool may decide that it no longer
need that element and double-free it.
So don't put object into quarantine in kasan_kfree(), just poison it.
Rename kasan_kfree() to kasan_poison_kfree() to respect that.
Also, we shouldn't use kasan_slab_alloc()/kasan_krealloc() in
kasan_unpoison_element() because those functions may update allocation
stacktrace. This would be wrong for the most of the remove_element call
sites.
(The only call site where we may want to update alloc stacktrace is
in mempool_alloc(). Kmemleak solves this by calling
kmemleak_update_trace(), so we could make something like that too.
But this is out of scope of this patch).
Fixes: 55834c5909 ("mm: kasan: initial memory quarantine implementation")
Link: http://lkml.kernel.org/r/575977C3.1010905@virtuozzo.com
Signed-off-by: Andrey Ryabinin <aryabinin@virtuozzo.com>
Reported-by: Kuthonuzo Luruo <kuthonuzo.luruo@hpe.com>
Acked-by: Alexander Potapenko <glider@google.com>
Cc: Dmitriy Vyukov <dvyukov@google.com>
Cc: Kostya Serebryany <kcc@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Export these symbols such that UBIFS can implement
->migratepage.
Cc: stable@vger.kernel.org
Signed-off-by: Richard Weinberger <richard@nod.at>
Acked-by: Christoph Hellwig <hch@lst.de>
Pull percpu fixes from Tejun Heo:
"While adding GFP_ATOMIC support to the percpu allocator, the
synchronization for the fast-path which doesn't require external
allocations was separated into pcpu_lock.
Unfortunately, it incorrectly decoupled async paths and percpu
chunks could get destroyed while still being operated on. This
contains two patches to fix the bug"
* 'for-4.7-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/percpu:
percpu: fix synchronization between synchronous map extension and chunk destruction
percpu: fix synchronization between chunk->map_extend_work and chunk destruction
Pull block layer fixes from Jens Axboe:
"A small collection of fixes for the current series. This contains:
- Two fixes for xen-blkfront, from Bob Liu.
- A bug fix for NVMe, releasing only the specific resources we
requested.
- Fix for a debugfs flags entry for nbd, from Josef.
- Plug fix from Omar, fixing up a case of code being switched between
two functions.
- A missing bio_put() for the new discard callers of
submit_bio_wait(), fixing a regression causing a leak of the bio.
From Shaun.
- Improve dirty limit calculation precision in the writeback code,
fixing a case where setting a limit lower than 1% of memory would
end up being zero. From Tejun"
* 'for-linus' of git://git.kernel.dk/linux-block:
NVMe: Only release requested regions
xen-blkfront: fix resume issues after a migration
xen-blkfront: don't call talk_to_blkback when already connected to blkback
nbd: pass the nbd pointer for flags debugfs
block: missing bio_put following submit_bio_wait
blk-mq: really fix plug list flushing for nomerge queues
writeback: use higher precision calculation in domain_dirty_limits()
I noticed that the logic in the fadvise64_64 syscall is incorrect for
partial pages. While first page of the region is correctly skipped if
it is partial, the last page of the region is mistakenly discarded.
This leads to problems for applications that read data in
non-page-aligned chunks discarding already processed data between the
reads.
A somewhat misguided application that does something like write(XX bytes
(non-page-alligned)); drop the data it just wrote; repeat gets a
significant penalty in performance as a result.
Link: http://lkml.kernel.org/r/1464917140-1506698-1-git-send-email-green@linuxhacker.ru
Signed-off-by: Oleg Drokin <green@linuxhacker.ru>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This patch is based on https://patchwork.ozlabs.org/patch/574623/.
Tejun submitted commit 23d11a58a9 ("workqueue: skip flush dependency
checks for legacy workqueues") for the legacy create*_workqueue()
interface.
But some workq created by alloc_workqueue still reports warning on
memory reclaim, e.g nvme_workq with flag WQ_MEM_RECLAIM set:
workqueue: WQ_MEM_RECLAIM nvme:nvme_reset_work is flushing !WQ_MEM_RECLAIM events:lru_add_drain_per_cpu
------------[ cut here ]------------
WARNING: CPU: 0 PID: 6 at SoC/linux/kernel/workqueue.c:2448 check_flush_dependency+0xb4/0x10c
...
check_flush_dependency+0xb4/0x10c
flush_work+0x54/0x140
lru_add_drain_all+0x138/0x188
migrate_prep+0xc/0x18
alloc_contig_range+0xf4/0x350
cma_alloc+0xec/0x1e4
dma_alloc_from_contiguous+0x38/0x40
__dma_alloc+0x74/0x25c
nvme_alloc_queue+0xcc/0x36c
nvme_reset_work+0x5c4/0xda8
process_one_work+0x128/0x2ec
worker_thread+0x58/0x434
kthread+0xd4/0xe8
ret_from_fork+0x10/0x50
That's because lru_add_drain_all() will schedule the drain work on
system_wq, whose flag is set to 0, !WQ_MEM_RECLAIM.
Introduce a dedicated WQ_MEM_RECLAIM workqueue to do
lru_add_drain_all(), aiding in getting memory freed.
Link: http://lkml.kernel.org/r/1464917521-9775-1-git-send-email-shhuiw@foxmail.com
Signed-off-by: Wang Sheng-Hui <shhuiw@foxmail.com>
Acked-by: Tejun Heo <tj@kernel.org>
Cc: Keith Busch <keith.busch@intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thierry Reding <treding@nvidia.com>
Cc: Ingo Molnar <mingo@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Christian Borntraeger reported a kernel panic after corrupt page counts,
and it turned out to be a regression introduced with commit aa88b68c3b
("thp: keep huge zero page pinned until tlb flush"), at least on s390.
put_huge_zero_page() was moved over from zap_huge_pmd() to
release_pages(), and it was replaced by tlb_remove_page(). However,
release_pages() might not always be triggered by (the arch-specific)
tlb_remove_page().
On s390 we call free_page_and_swap_cache() from tlb_remove_page(), and
not tlb_flush_mmu() -> free_pages_and_swap_cache() like the generic
version, because we don't use the MMU-gather logic. Although both
functions have very similar names, they are doing very unsimilar things,
in particular free_page_xxx is just doing a put_page(), while
free_pages_xxx calls release_pages().
This of course results in very harmful put_page()s on the huge zero
page, on architectures where tlb_remove_page() is implemented in this
way. It seems to affect only s390 and sh, but sh doesn't have THP
support, so the problem (currently) probably only exists on s390.
The following quick hack fixed the issue:
Link: http://lkml.kernel.org/r/20160602172141.75c006a9@thinkpad
Signed-off-by: Gerald Schaefer <gerald.schaefer@de.ibm.com>
Reported-by: Christian Borntraeger <borntraeger@de.ibm.com>
Tested-by: Christian Borntraeger <borntraeger@de.ibm.com>
Cc: "Kirill A. Shutemov" <kirill@shutemov.name>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Hugh Dickins <hughd@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: <stable@vger.kernel.org> [4.6.x]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
When creating a private mapping of a hugetlbfs file, it is possible to
unmap pages via ftruncate or fallocate hole punch. If subsequent faults
repopulate these mappings, the reserve counts will go negative. This is
because the code currently assumes all faults to private mappings will
consume reserves. The problem can be recreated as follows:
- mmap(MAP_PRIVATE) a file in hugetlbfs filesystem
- write fault in pages in the mapping
- fallocate(FALLOC_FL_PUNCH_HOLE) some pages in the mapping
- write fault in pages in the hole
This will result in negative huge page reserve counts and negative
subpool usage counts for the hugetlbfs. Note that this can also be
recreated with ftruncate, but fallocate is more straight forward.
This patch modifies the routines vma_needs_reserves and vma_has_reserves
to examine the reserve map associated with private mappings similar to
that for shared mappings. However, the reserve map semantics for
private and shared mappings are very different. This results in subtly
different code that is explained in the comments.
Link: http://lkml.kernel.org/r/1464720957-15698-1-git-send-email-mike.kravetz@oracle.com
Signed-off-by: Mike Kravetz <mike.kravetz@oracle.com>
Acked-by: Hillf Danton <hillf.zj@alibaba-inc.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Kirill Shutemov <kirill.shutemov@linux.intel.com>
Cc: Michal Hocko <mhocko@suse.cz>
Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Aneesh Kumar <aneesh.kumar@linux.vnet.ibm.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The optimistic fast path may use cpuset_current_mems_allowed instead of
of a NULL nodemask supplied by the caller for cpuset allocations. The
preferred zone is calculated on this basis for statistic purposes and as
a starting point in the zonelist iterator.
However, if the context can ignore memory policies due to being atomic
or being able to ignore watermarks then the starting point in the
zonelist iterator is no longer correct. This patch resets the zonelist
iterator in the allocator slowpath if the context can ignore memory
policies. This will alter the zone used for statistics but only after
it is known that it makes sense for that context. Resetting it before
entering the slowpath would potentially allow an ALLOC_CPUSET allocation
to be accounted for against the wrong zone. Note that while nodemask is
not explicitly set to the original nodemask, it would only have been
overwritten if cpuset_enabled() and it was reset before the slowpath was
entered.
Link: http://lkml.kernel.org/r/20160602103936.GU2527@techsingularity.net
Fixes: c33d6c06f6 ("mm, page_alloc: avoid looking up the first zone in a zonelist twice")
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Reported-by: Geert Uytterhoeven <geert@linux-m68k.org>
Tested-by: Geert Uytterhoeven <geert@linux-m68k.org>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Geert Uytterhoeven reported the following problem that bisected to
commit c33d6c06f6 ("mm, page_alloc: avoid looking up the first zone
in a zonelist twice") on m68k/ARAnyM
BUG: scheduling while atomic: cron/668/0x10c9a0c0
Modules linked in:
CPU: 0 PID: 668 Comm: cron Not tainted 4.6.0-atari-05133-gc33d6c06f60f710f #364
Call Trace: [<0003d7d0>] __schedule_bug+0x40/0x54
__schedule+0x312/0x388
__schedule+0x0/0x388
prepare_to_wait+0x0/0x52
schedule+0x64/0x82
schedule_timeout+0xda/0x104
set_next_entity+0x18/0x40
pick_next_task_fair+0x78/0xda
io_schedule_timeout+0x36/0x4a
bit_wait_io+0x0/0x40
bit_wait_io+0x12/0x40
__wait_on_bit+0x46/0x76
wait_on_page_bit_killable+0x64/0x6c
bit_wait_io+0x0/0x40
wake_bit_function+0x0/0x4e
__lock_page_or_retry+0xde/0x124
do_scan_async+0x114/0x17c
lookup_swap_cache+0x24/0x4e
handle_mm_fault+0x626/0x7de
find_vma+0x0/0x66
down_read+0x0/0xe
wait_on_page_bit_killable_timeout+0x77/0x7c
find_vma+0x16/0x66
do_page_fault+0xe6/0x23a
res_func+0xa3c/0x141a
buserr_c+0x190/0x6d4
res_func+0xa3c/0x141a
buserr+0x20/0x28
res_func+0xa3c/0x141a
buserr+0x20/0x28
The relationship is not obvious but it's due to a failure to rescan the
full zonelist after the fair zone allocation policy exhausts the batch
count. While this is a functional problem, it's also a performance
issue. A page allocator microbenchmark showed the following
4.7.0-rc1 4.7.0-rc1
vanilla reset-v1r2
Min alloc-odr0-1 327.00 ( 0.00%) 326.00 ( 0.31%)
Min alloc-odr0-2 235.00 ( 0.00%) 235.00 ( 0.00%)
Min alloc-odr0-4 198.00 ( 0.00%) 198.00 ( 0.00%)
Min alloc-odr0-8 170.00 ( 0.00%) 170.00 ( 0.00%)
Min alloc-odr0-16 156.00 ( 0.00%) 156.00 ( 0.00%)
Min alloc-odr0-32 150.00 ( 0.00%) 150.00 ( 0.00%)
Min alloc-odr0-64 146.00 ( 0.00%) 146.00 ( 0.00%)
Min alloc-odr0-128 145.00 ( 0.00%) 145.00 ( 0.00%)
Min alloc-odr0-256 155.00 ( 0.00%) 155.00 ( 0.00%)
Min alloc-odr0-512 168.00 ( 0.00%) 165.00 ( 1.79%)
Min alloc-odr0-1024 175.00 ( 0.00%) 174.00 ( 0.57%)
Min alloc-odr0-2048 180.00 ( 0.00%) 180.00 ( 0.00%)
Min alloc-odr0-4096 187.00 ( 0.00%) 186.00 ( 0.53%)
Min alloc-odr0-8192 190.00 ( 0.00%) 190.00 ( 0.00%)
Min alloc-odr0-16384 191.00 ( 0.00%) 191.00 ( 0.00%)
Min alloc-odr1-1 736.00 ( 0.00%) 445.00 ( 39.54%)
Min alloc-odr1-2 343.00 ( 0.00%) 335.00 ( 2.33%)
Min alloc-odr1-4 277.00 ( 0.00%) 270.00 ( 2.53%)
Min alloc-odr1-8 238.00 ( 0.00%) 233.00 ( 2.10%)
Min alloc-odr1-16 224.00 ( 0.00%) 218.00 ( 2.68%)
Min alloc-odr1-32 210.00 ( 0.00%) 208.00 ( 0.95%)
Min alloc-odr1-64 207.00 ( 0.00%) 203.00 ( 1.93%)
Min alloc-odr1-128 276.00 ( 0.00%) 202.00 ( 26.81%)
Min alloc-odr1-256 206.00 ( 0.00%) 202.00 ( 1.94%)
Min alloc-odr1-512 207.00 ( 0.00%) 202.00 ( 2.42%)
Min alloc-odr1-1024 208.00 ( 0.00%) 205.00 ( 1.44%)
Min alloc-odr1-2048 213.00 ( 0.00%) 212.00 ( 0.47%)
Min alloc-odr1-4096 218.00 ( 0.00%) 216.00 ( 0.92%)
Min alloc-odr1-8192 341.00 ( 0.00%) 219.00 ( 35.78%)
Note that order-0 allocations are unaffected but higher orders get a
small boost from this patch and a large reduction in system CPU usage
overall as can be seen here:
4.7.0-rc1 4.7.0-rc1
vanilla reset-v1r2
User 85.32 86.31
System 2221.39 2053.36
Elapsed 2368.89 2202.47
Fixes: c33d6c06f6 ("mm, page_alloc: avoid looking up the first zone in a zonelist twice")
Link: http://lkml.kernel.org/r/20160531100848.GR2527@techsingularity.net
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Reported-by: Geert Uytterhoeven <geert@linux-m68k.org>
Tested-by: Geert Uytterhoeven <geert@linux-m68k.org>
Tested-by: Mikulas Patocka <mpatocka@redhat.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Dmitry Vyukov
Gerald Schaefer
Lukasz Odzioba
Tejun Heo
Kirill A. Shutemov
Mel Gorman
Andrey Ryabinin
Anthony Romano
Tetsuo Handa
Richard Weinberger
Linus Torvalds
Oleg Drokin
Wang Sheng-Hui
Andrew Morton
Shuah Khan
Mike Kravetz
Michal Hocko