Messages in this thread |  | | | Date | Tue, 19 Jan 2010 17:06:55 +0800 | | From | Xiao Guangrong <> | | Subject | Re: [PATCH 1/3 v2] perf_event: fix race in perf_swevent_get_recursion_context() |
| |
Peter Zijlstra wrote: > On Tue, 2010-01-19 at 16:37 +0800, Xiao Guangrong wrote: >> It only disable preemption in perf_swevent_get_recursion_context(), >> it can't avoid race of hard-irq and NMI since they are nested that >> will re-enter this path and make the recursion value chaotic >> >> In this patch, we use atomic operation to avoid it and reduce >> cpu_ctx->recursion size, it also make this patch no need disable >> preemption > > NAK >
Sorry, i forget the nesting of hard-irq is sequential, and not pollute recursion value, thanks for you point out.
Hi Ingo,
Please ignore this patch, it not pollute other 2 patches in this patchset
Thanks, Xiao
|  |