On Mon, Sep 26, 2011 at 10:20 AM, Peter Zijlstra <a.p.zijlstra@chello.nl> wrote:
> On Sun, 2011-09-25 at 11:54 +0300, Gilad Ben-Yossef wrote:
>> This first version creates an on_each_cpu_mask infrastructure API
>
> But we already have the existing smp_call_function_many() doing that.

I might be wrong but my understanding is that smp_call_function_many()
does not invoke the IPI handler on the current processor. The original
code I replaced uses on_each_cpu() which does, so I figured a wrapper
was in order and then I discovered the same wrapper in arch specific
code.

> The on_each_cpu() thing is mostly a hysterical relic and could be
> completely depricated

Wont this require each caller to call smp_call_function_* and then
check to see if it needs to also invoke the IPI handler locally ? I
thought that was the reason for on_each_cpu existence... What have I
missed?

Thanks,
Gilad





-- 
Gilad Ben-Yossef
Chief Coffee Drinker
gilad@benyossef.com
Israel Cell: +972-52-8260388
US Cell: +1-973-8260388
http://benyossef.com

"I've seen things you people wouldn't believe. Goto statements used to
implement co-routines. I watched C structures being stored in
registers. All those moments will be lost in time... like tears in
rain... Time to die. "