Author: Vegard Nossum <vegard.nossum@xxxxxxxxx>
AuthorDate: Thu, 26 Feb 2009 14:36:37 +0100
Commit: Vegard Nossum <vegard.nossum@xxxxxxxxx>
CommitDate: Thu, 26 Feb 2009 14:36:37 +0100
kmemcheck (bitfields): don't actually initialize the bitfield
If we do a memset() on the bitfield, it will be cleared. What we
want is to mark it initialized, so do (only) that. This way, we
also don't change the behaviour of the code itself -- for example,
a cache constructor could decide to preserve a bitfield across
alloc/free, which would be bad for us, since we reset the actual
contents of the bitfield too. So fix that.
Signed-off-by: Vegard Nossum <vegard.nossum@xxxxxxxxx>
---
include/linux/kmemcheck.h | 6 +++---
1 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/include/linux/kmemcheck.h b/include/linux/kmemcheck.h
index d1237f0..54c053b 100644
--- a/include/linux/kmemcheck.h
+++ b/include/linux/kmemcheck.h
@@ -68,9 +68,9 @@ void kmemcheck_mark_initialized_pages(struct page *p, unsigned int n);
int kmemcheck_show_addr(unsigned long address);
int kmemcheck_hide_addr(unsigned long address);
-#define kmemcheck_annotate_bitfield(field) \
- do { \
- memset(&(field), 0, sizeof(field)); \
+#define kmemcheck_annotate_bitfield(field) \
+ do { \
+ kmemcheck_mark_initialized(&(field), sizeof(field)); \
} while (0)
#else
#define kmemcheck_enabled 0
--
To unsubscribe from this list: send the line "unsubscribe linux-tip-commits" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
|