Messages in this thread | | | Date | Thu, 13 Dec 2012 17:17:09 +0100 | From | Oleg Nesterov <> | Subject | Re: [RFC PATCH v4 1/9] CPU hotplug: Provide APIs to prevent CPU offline from atomic context |
| |
On 12/13, Srivatsa S. Bhat wrote: > > On 12/13/2012 12:42 AM, Srivatsa S. Bhat wrote: > > > > Even I don't spot anything wrong with it. But I'll give it some more > > thought.. > > Since an interrupt handler can also run get_online_cpus_atomic(), we > cannot use the __this_cpu_* versions for modifying reader_percpu_refcnt, > right?
Hmm. I thought that __this_cpu_* must be safe under preempt_disable(). IOW, I thought that, say, this_cpu_inc() is "equal" to preempt_disable + __this_cpu_inc() correctness-wise.
And. I thought that this_cpu_inc() is safe wrt interrupt, like local_t.
But when I try to read the comments percpu.h, I am starting to think that even this_cpu_inc() is not safe if irq handler can do the same?
Confused...
I am shy to ask... will, say, DEFINE_PER_CPU(local_t) and local_inc(__this_cpu_ptr(...)) work??
> But still, this scheme is better, because the reader doesn't have to spin > on the read_lock() with interrupts disabled.
Yes, but my main concern is that irq_disable/enable itself is not that cheap.
Oleg.
| |