Am 23.07.2014 11:07, schrieb Michel Dänzer:
On 23.07.2014 16:47, Christian König wrote:
From: Christian König christian.koenig@amd.com
We must mask out the overflow bit as well, otherwise the wptr will never match the rptr again and the interrupt handler will loop forever.
Signed-off-by: Christian König christian.koenig@amd.com
Reviewed-by: Michel Dänzer michel.daenzer@amd.com
Cc: stable@vger.kernel.org
Not sure this is justified though. Were you able to reproduce the problem and verify the fix?
Not yet, reproducing ring buffer overflows aren't easy to do. But I've had that multiple times with CIK and userptr support work and I can try if I can reproduce it once more.
From reading the code it looks like a rather fundamental problem to me and because of this I've added the stable Cc.
Christian.