2012-06-22 18:50:13

by Vaibhav Nagarnaik

Subject: [PATCH] tracing: Fix crash due to uninitialized new_pages list head

The new_pages list head in the cpu_buffer is not initialized. When
adding pages to the ring buffer, if the memory allocation fails in
ring_buffer_resize, the clean up handler tries to free up the allocated
pages from all the cpu buffers. The panic is caused by referencing the
uninitialized new_pages list head.

Initializing the new_pages list head in rb_allocate_cpu_buffer fixes
this.

Signed-off-by: Vaibhav Nagarnaik <[email protected]>
---
kernel/trace/ring_buffer.c | 1 +
1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c
index a2bec4c..c5a5479 100644
--- a/kernel/trace/ring_buffer.c
+++ b/kernel/trace/ring_buffer.c
@@ -1075,6 +1075,7 @@ rb_allocate_cpu_buffer(struct ring_buffer *buffer, int nr_pages, int cpu)
rb_init_page(bpage->page);

INIT_LIST_HEAD(&cpu_buffer->reader_page->list);
+ INIT_LIST_HEAD(&cpu_buffer->new_pages);

ret = rb_allocate_pages(cpu_buffer, nr_pages);
if (ret < 0)
--
1.7.7.3

2012-06-22 18:55:17

by Vaibhav Nagarnaik

Subject: Re: [PATCH] tracing: Fix crash due to uninitialized new_pages list head

On Fri, Jun 22, 2012 at 11:50 AM, Vaibhav Nagarnaik
<[email protected]> wrote:
> The new_pages list head in the cpu_buffer is not initialized. When
> adding pages to the ring buffer, if the memory allocation fails in
> ring_buffer_resize, the clean up handler tries to free up the allocated
> pages from all the cpu buffers. The panic is caused by referencing the
> uninitialized new_pages list head.
>
> Initializing the new_pages list head in rb_allocate_cpu_buffer fixes
> this.
>
> Signed-off-by: Vaibhav Nagarnaik <[email protected]>
> ---

Hi Steven,

I believe this fix should be pushed to 3.5 along with the other patch
I sent earlier:
tracing: Update entries counter when removing pages

These two patches fix issues with the recently added atomic resize patches.


Thanks

Vaibhav Nagarnaik