[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20100510113857.GA5131@brick.ozlabs.ibm.com>
Date: Mon, 10 May 2010 21:38:57 +1000
From: Paul Mackerras <paulus@...ba.org>
To: Peter Zijlstra <peterz@...radead.org>
Cc: mingo@...hat.com, hpa@...or.com, eranian@...gle.com,
linux-kernel@...r.kernel.org, davem@...emloft.net,
fweisbec@...il.com, ming.m.lin@...el.com, tglx@...utronix.de,
mingo@...e.hu, linux-tip-commits@...r.kernel.org
Subject: Re: [tip:perf/core] perf: Add group scheduling transactional APIs
On Mon, May 10, 2010 at 10:11:55AM +0200, Peter Zijlstra wrote:
> Can't you compute n_txn_start by subtracting n_added from n_events ?
Well, n_added says how many have been added since hw_perf_disable, and
we could get multiple transactions inside a hw_perf_disable/enable
pair, if we're scheduling multiple groups on. So n_events - n_added
doesn't tell us how many have been added in this transaction -- unless
I'm mistaken and there is a one transaction per hw_perf_disable/enable
rule, but I didn't see that written anywhere, and it does seem to be
explicitly one transaction per group.
Paul.
--
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