tracing: Fix unmapping loop in tracing_mark_write
authorVikram Mulukutla <markivx@codeaurora.org>
Thu, 18 Dec 2014 02:50:56 +0000 (18:50 -0800)
committerSteven Rostedt <rostedt@goodmis.org>
Mon, 9 Feb 2015 23:47:09 +0000 (18:47 -0500)
Commit 6edb2a8a385f0cdef51dae37ff23e74d76d8a6ce introduced
an array map_pages that contains the addresses returned by
kmap_atomic. However, when unmapping those pages, map_pages[0]
is unmapped before map_pages[1], breaking the nesting requirement
as specified in the documentation for kmap_atomic/kunmap_atomic.

This was caught by the highmem debug code present in kunmap_atomic.
Fix the loop to do the unmapping properly.

Link: http://lkml.kernel.org/r/1418871056-6614-1-git-send-email-markivx@codeaurora.org
Cc: stable@vger.kernel.org # 3.5+
Reviewed-by: Stephen Boyd <sboyd@codeaurora.org>
Reported-by: Lime Yang <limey@codeaurora.org>
Signed-off-by: Vikram Mulukutla <markivx@codeaurora.org>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
kernel/trace/trace.c

index 5afce60e1b68e46bca602eeaf87569064260b1c8..2078b86750e0507d27bb1655bd5ef59819cfa338 100644 (file)
@@ -4928,7 +4928,7 @@ tracing_mark_write(struct file *filp, const char __user *ubuf,
        *fpos += written;
 
  out_unlock:
-       for (i = 0; i < nr_pages; i++){
+       for (i = nr_pages - 1; i >= 0; i--) {
                kunmap_atomic(map_page[i]);
                put_page(pages[i]);
        }