Messages in this thread | | | Date | Mon, 18 Jan 2010 17:41:30 +0100 | From | Frederic Weisbecker <> | Subject | Re: [PATCH 1/3] perf_event: fix race in perf_swevent_get_recursion_context() |
| |
On Mon, Jan 18, 2010 at 09:42:34PM +0800, Xiao Guangrong wrote: > It only disable preemption in perf_swevent_get_recursion_context() > it can't avoid race of hard-irq and NMI > > In this patch, we use atomic operation to avoid it and reduce > cpu_ctx->recursion size, it also make this patch no need diable > preemption > > Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
I don't understand what is racy in what we have currently.
> int perf_swevent_get_recursion_context(void) > { > - struct perf_cpu_context *cpuctx = &get_cpu_var(perf_cpu_context); > + struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context); > int rctx; > > if (in_nmi()) > @@ -3933,13 +3933,8 @@ int perf_swevent_get_recursion_context(void) > else > rctx = 0; > > - if (cpuctx->recursion[rctx]) { > - put_cpu_var(perf_cpu_context); > + if (test_and_set_bit(rctx, &cpuctx->recursion)) > return -1;
This looks broken. We don't call back perf_swevent_put_recursion_context in fail case, so the bit won't ever be cleared once we recurse.
| |