lkml.org 
[lkml]   [2011]   [Jun]   [21]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH 3/3] mm: print information when THP is disabled automatically
On Tue, Jun 21, 2011 at 01:26:58AM +0800, Cong Wang wrote:
> ??? 2011???06???21??? 01:01, Mel Gorman ??????:
> >On Tue, Jun 21, 2011 at 12:34:30AM +0800, Amerigo Wang wrote:
> >>Print information when THP is disabled automatically so that
> >>users can find this info in dmesg.
> >>
> >>Signed-off-by: WANG Cong<amwang@redhat.com>
> >>---
> >> mm/huge_memory.c | 5 ++++-
> >> 1 files changed, 4 insertions(+), 1 deletions(-)
> >>
> >>diff --git a/mm/huge_memory.c b/mm/huge_memory.c
> >>index 7fb44cc..07679da 100644
> >>--- a/mm/huge_memory.c
> >>+++ b/mm/huge_memory.c
> >>@@ -544,8 +544,11 @@ static int __init hugepage_init(void)
> >> * where the extra memory used could hurt more than TLB overhead
> >> * is likely to save. The admin can still enable it through /sys.
> >> */
> >>- if (totalram_pages< (CONFIG_TRANSPARENT_HUGEPAGE_THRESHOLD<< (20 - PAGE_SHIFT)))
> >>+ if (totalram_pages< (CONFIG_TRANSPARENT_HUGEPAGE_THRESHOLD
> >>+ << (20 - PAGE_SHIFT))) {
> >>+ printk(KERN_INFO "hugepage: disabled auotmatically\n");
> >> transparent_hugepage_flags = 0;
> >>+ }
> >>
> >> start_khugepaged();
> >>
> >
> >Guess this doesn't hurt. You misspelled automatically though and
> >mentioning "hugepage" could be confused with hugetlbfs.
> >
>
> Yeah, sorry for the typo.
>
> But, there are many printk messages in the same file start with "hugepage:".
> :-) I can send a patch to replace all of them with "THP" if you want...
>

My bad. They are in a memory allocation failure path that should be
impossible to trigger but I should still have spotted the prefix at
the time and complained. Changing them to THP would be nice.

--
Mel Gorman
SUSE Labs


\
 
 \ /
  Last update: 2011-06-21 11:43    [from the cache]
©2003-2011 Jasper Spaans