From: Linus Torvalds <torvalds@linux-foundation.org>
Date: Wed, 13 May 2009 15:29:12 +0000 (-0700)
Subject: Revert "Ignore madvise(MADV_WILLNEED) for hugetlbfs-backed regions"
X-Git-Url: https://git.karo-electronics.de/?a=commitdiff_plain;h=0f181328287db30671e9997329cff71395d4af8b;p=linux-beck.git

Revert "Ignore madvise(MADV_WILLNEED) for hugetlbfs-backed regions"

This reverts commit a425a638c858fd10370b573bde81df3ba500e271.

Now that the previous commit removed the "readpage" actor for hugetlb
files, read-ahead will no longer mess up the mapping, and there's no
longer any reason to treat hugetlbfs mappings specially.

Tested-and-acked-by: Mel Gorman <mel@csn.ul.ie>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---

diff --git a/mm/madvise.c b/mm/madvise.c
index 36d6ea2b6340..b9ce574827c8 100644
--- a/mm/madvise.c
+++ b/mm/madvise.c
@@ -112,14 +112,6 @@ static long madvise_willneed(struct vm_area_struct * vma,
 	if (!file)
 		return -EBADF;
 
-	/*
-	 * Page cache readahead assumes page cache pages are order-0 which
-	 * is not the case for hugetlbfs. Do not give a bad return value
-	 * but ignore the advice.
-	 */
-	if (vma->vm_flags & VM_HUGETLB)
-		return 0;
-
 	if (file->f_mapping->a_ops->get_xip_mem) {
 		/* no bad return value, but ignore advice */
 		return 0;