[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <4D6D1A4C.1090009@cisco.com>
Date: Tue, 01 Mar 2011 09:09:48 -0700
From: David Ahern <daahern@...co.com>
To: Peter Zijlstra <peterz@...radead.org>
CC: linux-perf-users@...r.kernel.org, linux-kernel@...r.kernel.org,
acme@...stprotocols.net, mingo@...e.hu, fweisbec@...il.com,
paulus@...ba.org, tglx@...utronix.de
Subject: Re: [PATCH 3/6] perf record: add time-of-day option
On 03/01/2011 09:00 AM, Peter Zijlstra wrote:
> On Tue, 2011-03-01 at 08:41 -0700, David Ahern wrote:
>
>> realtime-clock counter is created with a period N. How do you correlate
>> perf_clock to time-of-day until now + N?
>
> You scan fwd? Since its at most N time ahead you're scanning fwd its
> still constant time.
So I potentially scan a multi-gigabyte file looking for a single event
-- that may not even exist. From a usability perspective the time string
should appear if the clock event exists and not if it does not exist.
This should happen automatically.
>
> Thing is, I'm really not convinced this is a very useful ioctl() in
> generic, and ioctl()s really need a good justification to add.
>
I did not code it create samples of a realtime-clock event; I coded it
to add an event to the stream on demand -- whatever event the fd is tied to.
David
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists