Minor hugepage bugfix

[David Gibson]

Andrew, please apply:

add_to_page_cache() locks the given page if and only if it suceeds.
The hugepage code (every arch), however, does an unlock_page() after
add_to_page_cache() before checking the return code, which could trip
the BUG() in unlock_page() if add_to_page_cache() failed.

In practice we've never hit this bug, because the only ways
add_to_page_cache() can fail are when we fail to allocate a radix tree
node (very rare), or when there is already a page at that offset in
the radix tree, which never happens during prefault, obviously.  We
should probably fix it anyway, though.

The analagous bug in some of the patches floating about to
demand-allocation of hugepages is more of a problem, because multiple
processes can race to instantiate a particular page in the radix tree
- that's been hit at least once (which is how I found this).

Index: working-2.6/arch/ppc64/mm/hugetlbpage.c
===================================================================
--- working-2.6.orig/arch/ppc64/mm/hugetlbpage.c	2004-05-05 10:05:52.000000000 +1000
+++ working-2.6/arch/ppc64/mm/hugetlbpage.c	2004-05-07 15:52:28.536883064 +1000
@@ -452,8 +452,9 @@
 				goto out;
 			}
 			ret = add_to_page_cache(page, mapping, idx, GFP_ATOMIC);
-			unlock_page(page);
-			if (ret) {
+			if (! ret) {
+				unlock_page(page);
+			} else {
 				hugetlb_put_quota(mapping);
 				free_huge_page(page);
 				goto out;
Index: working-2.6/arch/i386/mm/hugetlbpage.c
===================================================================
--- working-2.6.orig/arch/i386/mm/hugetlbpage.c	2004-04-28 09:37:18.000000000 +1000
+++ working-2.6/arch/i386/mm/hugetlbpage.c	2004-05-07 15:52:28.537882912 +1000
@@ -264,8 +264,9 @@
 				goto out;
 			}
 			ret = add_to_page_cache(page, mapping, idx, GFP_ATOMIC);
-			unlock_page(page);
-			if (ret) {
+			if (! ret) {
+				unlock_page(page);
+			} else {
 				hugetlb_put_quota(mapping);
 				free_huge_page(page);
 				goto out;
Index: working-2.6/arch/ia64/mm/hugetlbpage.c
===================================================================
--- working-2.6.orig/arch/ia64/mm/hugetlbpage.c	2004-04-28 09:37:18.000000000 +1000
+++ working-2.6/arch/ia64/mm/hugetlbpage.c	2004-05-07 15:53:19.858935352 +1000
@@ -293,8 +293,9 @@
 				goto out;
 			}
 			ret = add_to_page_cache(page, mapping, idx, GFP_ATOMIC);
-			unlock_page(page);
-			if (ret) {
+			if (! ret) {
+				unlock_page(page);
+			} else {
 				hugetlb_put_quota(mapping);
 				free_huge_page(page);
 				goto out;
Index: working-2.6/arch/sparc64/mm/hugetlbpage.c
===================================================================
--- working-2.6.orig/arch/sparc64/mm/hugetlbpage.c	2004-04-28 09:37:19.000000000 +1000
+++ working-2.6/arch/sparc64/mm/hugetlbpage.c	2004-05-07 15:53:54.482863760 +1000
@@ -244,8 +244,9 @@
 				goto out;
 			}
 			ret = add_to_page_cache(page, mapping, idx, GFP_ATOMIC);
-			unlock_page(page);
-			if (ret) {
+			if (! ret) {
+				unlock_page(page);
+			} else {
 				hugetlb_put_quota(mapping);
 				free_huge_page(page);
 				goto out;
Index: working-2.6/arch/sh/mm/hugetlbpage.c
===================================================================
--- working-2.6.orig/arch/sh/mm/hugetlbpage.c	2004-04-28 09:37:19.000000000 +1000
+++ working-2.6/arch/sh/mm/hugetlbpage.c	2004-05-07 15:54:19.282954200 +1000
@@ -248,8 +248,9 @@
 				goto out;
 			}
 			ret = add_to_page_cache(page, mapping, idx, GFP_ATOMIC);
-			unlock_page(page);
-			if (ret) {
+			if (! ret) {
+				unlock_page(page);
+			} else {
 				hugetlb_put_quota(mapping);
 				free_huge_page(page);
 				goto out;



-- 
David Gibson			| For every complex problem there is a
david AT gibson.dropbear.id.au	| solution which is simple, neat and
				| wrong.
http://www.ozlabs.org/people/dgibson
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/