A-Team_Digital_Solutions/android_kernel_motorola_sm6375
Watch
2
0
Fork
You've already forked android_kernel_motorola_sm6375
 0
mirror of https://github.com/BobTheBlinker/android_kernel_motorola_sm6375.git synced 2026-10-29 03:01:39 -04:00
Code Issues Projects Releases Packages Wiki Activity

memcg: fix split_huge_page_refcounts()

Browse source 
This patch started off as a cleanup: __split_huge_page_refcounts() has to
cope with two scenarios, when the hugepage being split is already on LRU,
and when it is not; but why does it have to split that accounting across
three different sites?  Consolidate it in lru_add_page_tail(), handling
evictable and unevictable alike, and use standard add_page_to_lru_list()
when accounting is needed (when the head is not yet on LRU).

But a recent regression in -next, I guess the removal of PageCgroupAcctLRU
test from mem_cgroup_split_huge_fixup(), makes this now a necessary fix:
under load, the MEM_CGROUP_ZSTAT count was wrapping to a huge number,
messing up reclaim calculations and causing a freeze at rmdir of cgroup.

Add a VM_BUG_ON to mem_cgroup_lru_del_list() when we're about to wrap that
count - this has not been the only such incident.  Document that
lru_add_page_tail() is for Transparent HugePages by #ifdef around it.

Signed-off-by: Hugh Dickins <[email protected]>
Cc: Daisuke Nishimura <[email protected]>
Cc: KAMEZAWA Hiroyuki <[email protected]>
Cc: Johannes Weiner <[email protected]>
Cc: Michal Hocko <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
This commit is contained in:
Hugh Dickins 2012-01-12 17:19:52 -08:00 • committed by Linus Torvalds
commit 12d2710786
3 changed files with 21 additions and 30 deletions
Download patch file Download diff file Expand all files Collapse all files

12
mm/memcontrol.c
View file

@ -1071,6 +1071,7 @@ void mem_cgroup_lru_del_list(struct page *page, enum lru_list lru)
VM_BUG_ON(!memcg);
mz = page_cgroup_zoneinfo(memcg, page);
/* huge page split is done under lru_lock. so, we have no races. */
VM_BUG_ON(MEM_CGROUP_ZSTAT(mz, lru) < (1 << compound_order(page)));
MEM_CGROUP_ZSTAT(mz, lru) -= 1 << compound_order(page);
}
@ -2465,9 +2466,7 @@ static void __mem_cgroup_commit_charge(struct mem_cgroup *memcg,
void mem_cgroup_split_huge_fixup(struct page *head)
{
struct page_cgroup *head_pc = lookup_page_cgroup(head);
struct mem_cgroup_per_zone *mz;
struct page_cgroup *pc;
enum lru_list lru;
int i;
if (mem_cgroup_disabled())
@ -2478,15 +2477,8 @@ void mem_cgroup_split_huge_fixup(struct page *head)
smp_wmb();/* see __commit_charge() */
pc->flags = head_pc->flags & ~PCGF_NOCOPY_AT_SPLIT;
}
/*
* Tail pages will be added to LRU.
* We hold lru_lock,then,reduce counter directly.
*/
lru = page_lru(head);
mz = page_cgroup_zoneinfo(head_pc->mem_cgroup, head);
MEM_CGROUP_ZSTAT(mz, lru) -= HPAGE_PMD_NR - 1;
}
#endif
#endif /* CONFIG_TRANSPARENT_HUGEPAGE */
/**
* mem_cgroup_move_account - move account of the page