This is a note to let you know that I've just added the patch titled
mm: fix negative commitlimit when gigantic hugepages are
to the 2.6.33-longterm tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/longterm/longterm-queue-2.6.33.git;a=summary
The filename of the patch is:
mm-fix-negative-commitlimit-when-gigantic-hugepages-are.patch
and it can be found in the queue-2.6.33 subdirectory.
If you, or anyone else, feels it should not be added to the 2.6.33 longterm
tree,
please let <[email protected]> know about it.
>From b0320c7b7d1ac1bd5c2d9dff3258524ab39bad32 Mon Sep 17 00:00:00 2001
From: Rafael Aquini <[email protected]>
Date: Wed, 15 Jun 2011 15:08:39 -0700
Subject: mm: fix negative commitlimit when gigantic hugepages are
allocated
From: Rafael Aquini <[email protected]>
commit b0320c7b7d1ac1bd5c2d9dff3258524ab39bad32 upstream.
When 1GB hugepages are allocated on a system, free(1) reports less
available memory than what really is installed in the box. Also, if the
total size of hugepages allocated on a system is over half of the total
memory size, CommitLimit becomes a negative number.
The problem is that gigantic hugepages (order > MAX_ORDER) can only be
allocated at boot with bootmem, thus its frames are not accounted to
'totalram_pages'. However, they are accounted to hugetlb_total_pages()
What happens to turn CommitLimit into a negative number is this
calculation, in fs/proc/meminfo.c:
allowed = ((totalram_pages - hugetlb_total_pages())
* sysctl_overcommit_ratio / 100) + total_swap_pages;
A similar calculation occurs in __vm_enough_memory() in mm/mmap.c.
Also, every vm statistic which depends on 'totalram_pages' will render
confusing values, as if system were 'missing' some part of its memory.
Impact of this bug:
When gigantic hugepages are allocated and sysctl_overcommit_memory ==
OVERCOMMIT_NEVER. In a such situation, __vm_enough_memory() goes through
the mentioned 'allowed' calculation and might end up mistakenly returning
-ENOMEM, thus forcing the system to start reclaiming pages earlier than it
would be ususal, and this could cause detrimental impact to overall
system's performance, depending on the workload.
Besides the aforementioned scenario, I can only think of this causing
annoyances with memory reports from /proc/meminfo and free(1).
[[email protected]: standardize comment layout]
Reported-by: Russ Anderson <[email protected]>
Signed-off-by: Rafael Aquini <[email protected]>
Acked-by: Russ Anderson <[email protected]>
Cc: Andrea Arcangeli <[email protected]>
Cc: Christoph Lameter <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
---
mm/hugetlb.c | 8 ++++++++
1 file changed, 8 insertions(+)
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -1105,6 +1105,14 @@ static void __init gather_bootmem_preall
WARN_ON(page_count(page) != 1);
prep_compound_huge_page(page, h->order);
prep_new_huge_page(h, page, page_to_nid(page));
+ /*
+ * If we had gigantic hugepages allocated at boot time, we need
+ * to restore the 'stolen' pages to totalram_pages in order to
+ * fix confusing memory reports from free(1) and another
+ * side-effects, like CommitLimit going negative.
+ */
+ if (h->order > (MAX_ORDER - 1))
+ totalram_pages += 1 << h->order;
}
}
Patches currently in longterm-queue-2.6.33 which might be from [email protected]
are
/home/gregkh/linux/longterm/longterm-queue-2.6.33/queue-2.6.33/mm-fix-negative-commitlimit-when-gigantic-hugepages-are.patch
_______________________________________________
stable mailing list
[email protected]
http://linux.kernel.org/mailman/listinfo/stable
