diff options
| author | Andi Kleen <andi@firstfloor.org> | 2009-12-16 12:20:00 +0100 |
|---|---|---|
| committer | Andi Kleen <ak@linux.intel.com> | 2009-12-16 12:20:00 +0100 |
| commit | d15f107d97bd74c74d8f5144843d372666ddbdac (patch) | |
| tree | 6c12d3c738c3118ccab724b5b984fc07c7e58082 /mm | |
| parent | 1bfe5febe34d2be2120803c10720e179186357c9 (diff) | |
| download | lwn-d15f107d97bd74c74d8f5144843d372666ddbdac.tar.gz lwn-d15f107d97bd74c74d8f5144843d372666ddbdac.zip | |
HWPOISON: Use get_user_page_fast in hwpoison madvise
The previous version didn't take the mmap_sem before calling gup(),
which is racy.
Use get_user_pages_fast() instead which doesn't need any locks.
This is also faster of course, but then it doesn't really matter
because this is just a testing path.
Based on report from Nick Piggin.
Cc: npiggin@suse.de
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Diffstat (limited to 'mm')
| -rw-r--r-- | mm/madvise.c | 3 |
1 files changed, 1 insertions, 2 deletions
diff --git a/mm/madvise.c b/mm/madvise.c index 6ca34f0cd4aa..7964e36ba915 100644 --- a/mm/madvise.c +++ b/mm/madvise.c @@ -230,8 +230,7 @@ static int madvise_hwpoison(unsigned long start, unsigned long end) return -EPERM; for (; start < end; start += PAGE_SIZE) { struct page *p; - int ret = get_user_pages(current, current->mm, start, 1, - 0, 0, &p, NULL); + int ret = get_user_pages_fast(start, 1, 0, &p); if (ret != 1) return ret; printk(KERN_INFO "Injecting memory failure for page %lx at %lx\n", |