Re: [PATCH] trace: Set oom_score_adj to maximum for ring bufferallocating process

[Vaibhav Nagarnaik]

On Thu, May 26, 2011 at 2:00 PM, Steven Rostedt <rostedt@xxxxxxxxxxx> wrote:
> I'm thinking the oom killer used here got lucky. As it killed this task,
> we were still out of memory, and the ring buffer failed to get the
> memory it needed and freed up everything that it previously allocated,
> and returned. Then the process calling this function would be killed by

I have tested this multiple times and I don't see any other process
being killed, regardless of who invoked oom-killer. The oom-killer
always selects the "echo" process to kill because of its maximum
oom_score_adj and sets TIF_MEMDIE. Since the ring buffer is still
getting allocated, any further allocation from this process fails
because of this flag and the fatal signal pending. This causes the ring
buffer memory to be freed.

I don't see the oom-killer getting lucky in my test cases.


Vaibhav Nagarnaik
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/