Prev: perf/perf_events: misleading number of samples due to mmap()
Next: [PATCH] usb: conexant: fixed spacing and brace coding style issues
From: Cyrill Gorcunov on 16 Jun 2010 13:00
On Wed, Jun 16, 2010 at 06:00:32PM +0200, Peter Zijlstra wrote:
> Since the current perf_disable() usage is only an optimization, remove
> it for now. This eases the removal of the weak hw_perf_enable
> interface.
>
> [ Some comments talk about NMI races, but I cannot find any such
> things. ]
>

The race was there and was fixed by commit

| commit 34adc8062227f41b04ade0ff3fbd1dbe3002669e
| Author: Ingo Molnar <mingo(a)elte.hu>
| Date: Wed May 20 20:13:28 2009 +0200

then you fixed it further in commit

| commit 3fb2b8ddcc6a7aa62af6bd2cb939edfd4c460506
| Author: Peter Zijlstra <a.p.zijlstra(a)chello.nl>
| Date: Mon Mar 8 13:51:01 2010 +0100

which I understand as we don't need perf_disable here for sure.
Just for the record :)

-- Cyrill
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
 | 
Pages: 1
Prev: perf/perf_events: misleading number of samples due to mmap()
Next: [PATCH] usb: conexant: fixed spacing and brace coding style issues