[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <AANLkTilwlZKx0k-Z_6LHVXIM561JLcWV4mX4Wcwgb7r0@mail.gmail.com>
Date: Tue, 25 May 2010 18:10:06 +0200
From: Stephane Eranian <eranian@...gle.com>
To: Peter Zijlstra <peterz@...radead.org>
Cc: eranian@...il.com, linux-kernel@...r.kernel.org, mingo@...e.hu,
paulus@...ba.org, davem@...emloft.net, fweisbec@...il.com,
acme@...radead.org, ming.m.lin@...el.com,
perfmon2-devel@...ts.sf.net
Subject: Re: [PATCH] perf_events: fix event scheduling issues introduced by
transactional API (take 2)
> Index: linux-2.6/kernel/perf_event.c
> ===================================================================
> --- linux-2.6.orig/kernel/perf_event.c
> +++ linux-2.6/kernel/perf_event.c
> @@ -668,15 +668,9 @@ group_sched_in(struct perf_event *group_
> }
> }
>
> - if (!txn)
> + if (!txn || !pmu->commit_txn(pmu))
> return 0;
>
> - ret = pmu->commit_txn(pmu);
> - if (!ret) {
> - pmu->cancel_txn(pmu);
> - return 0;
> - }
> -
> group_error:
> /*
> * Groups can be scheduled in as one unit only, so undo any
>
Looks okay.
I believe you can also drop the txn test in group_sched_in() after group_error:,
given you have the if !(txn) return 0.
Powered by blists - more mailing lists