(2012/04/04 10:56), David Rientjes wrote:
> On COW, a new hugepage is allocated and charged to the memcg. If the
> memcg is oom, however, this charge will fail and will return VM_FAULT_OOM
> to the page fault handler which results in an oom kill.
>
> Instead, it's possible to fallback to splitting the hugepage so that the
> COW results only in an order-0 page being charged to the memcg which has
> a higher liklihood to succeed. This is expensive because the hugepage
> must be split in the page fault handler, but it is much better than
> unnecessarily oom killing a process.
>
> Signed-off-by: David Rientjes <rientjes@xxxxxxxxxx>
> ---
> mm/huge_memory.c | 1 +
> mm/memory.c | 18 +++++++++++++++---
> 2 files changed, 16 insertions(+), 3 deletions(-)
>
> diff --git a/mm/huge_memory.c b/mm/huge_memory.c
> --- a/mm/huge_memory.c
> +++ b/mm/huge_memory.c
> @@ -959,6 +959,7 @@ int do_huge_pmd_wp_page(struct mm_struct *mm, struct vm_area_struct *vma,
>
> if (unlikely(mem_cgroup_newpage_charge(new_page, mm, GFP_KERNEL))) {
> put_page(new_page);
> + split_huge_page(page);
> put_page(page);
> ret |= VM_FAULT_OOM;
> goto out;
?? how about
==
if (transparent_hugepage_enabled(vma) &&
!transparent_hugepage_debug_cow())
new_page = alloc_hugepage_vma(transparent_hugepage_defrag(vma),
vma, haddr, numa_node_id(), 0);
else
new_page = NULL;
if (unlikely(mem_cgroup_newpage_charge(new_page, mm, GFP_KERNEL))) {
put_page(new_page);
new_page = NULL; /* never OOM, just cause fallback */
}
if (unlikely(!new_page)) {
count_vm_event(THP_FAULT_FALLBACK);
ret = do_huge_pmd_wp_page_fallback(mm, vma, address,
pmd, orig_pmd, page, haddr);
put_page(page);
goto out;
}
==
?
Thanks,
-Kame
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxx. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>