Re: [RFC GIT PULL] perf/trace/lock optimization/scalabilityimprovements

[Frederic Weisbecker]

On Sun, Feb 07, 2010 at 06:10:27PM +0100, Peter Zijlstra wrote:
> One thing we could do is add a PERF_TYPE_COLLECTION and have its read()
> method do something else than regular counters would, that way we can
> use read()/lseek()/stat() fd interface. The only trouble with that is
> that doesn't use the 'regular' output buffer.


But would be nice too. The drawback is indeed a separation/implementation
treatment against other events.


> /me goes try and come up with yet another solution..


Not sure there is another solution than the above or an inject/mmap
thing.

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/