Date: Thu, 16 Apr 2015 17:43:57 +0100
From: Mark Rutland <mark.rutland@arm.com>
To: Kan Liang <kan.liang@intel.com>
Cc: "acme@kernel.org" <acme@kernel.org>,
        "a.p.zijlstra@chello.nl" <a.p.zijlstra@chello.nl>,
        "eranian@google.com" <eranian@google.com>,
        "andi@firstfloor.org" <andi@firstfloor.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH V2 1/6] perf,core: allow invalid context events to be
 part of sw/hw groups
Message-ID: <20150416164356.GD19775@leverpostej>
References: <1429084576-1078-1-git-send-email-kan.liang@intel.com>
 <20150416163140.GA19775@leverpostej>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20150416163140.GA19775@leverpostej>
User-Agent: Mutt/1.5.21 (2010-09-15)
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 989
Lines: 23

> Even if you ignore the fundamental semantics of groups, there are other
> problems with allowing shared contexts:
> 
> * The *_txn functions only get called on the group leader's PMU. If your
>   system PMU has these functions, they are not called.
> 
> * Event rotation is per ctx, but now you could have some events in a CPU
>   PMU's context, and some in the uncore PMU's context. So those can race
>   with each other.
> 
> * Throttling is also per-context. So those can race with each other too.

There's also a break down of behaviour: events in the uncore context
will get migrated to another CPU in the event of a hot unplug, while
events that are grouped with CPU events (and hence live in the CPU
context) will be destroyed.

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