Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
IronPort-SDR: 898Xm2NgKOIHwiVJ3iaYeL2/igbzwnruKElMWUXt4FANA0LKougCVeyQ6CSnoE8NJwvVMtTyOH
 aT8nSNwoYtJg==
IronPort-SDR: pbFnx7iBOrbMc2QALC+mxcPEvXgco+DtPPXnpBxnGJEXO8ORlqKUl8v7VeObfdKSR1zv+kH540
 W9wM4FhOcuSA==
Date:   Mon, 8 Mar 2021 08:54:37 -0800
From:   Andi Kleen <ak@linux.intel.com>
To:     Adrian Hunter <adrian.hunter@intel.com>
Cc:     Arnaldo Carvalho de Melo <acme@kernel.org>,
        Jiri Olsa <jolsa@redhat.com>, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] perf auxtrace: Fix auxtrace queue conflict
Message-ID: <20210308165437.GA203350@tassilo.jf.intel.com>
References: <20210308151143.18338-1-adrian.hunter@intel.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20210308151143.18338-1-adrian.hunter@intel.com>
Precedence: bulk

On Mon, Mar 08, 2021 at 05:11:43PM +0200, Adrian Hunter wrote:
> The only requirement of an auxtrace queue is that the buffers are in
> time order.  That is achieved by making separate queues for separate
> perf buffer or AUX area buffer mmaps.
> 
> That generally means a separate queue per cpu for per-cpu contexts,
> and a separate queue per thread for per-task contexts.
> 
> When buffers are added to a queue, perf checks that the buffer cpu
> and thread id (tid) match the queue cpu and thread id.
> 
> However, generally, that need not be true, and perf will queue
> buffers correctly anyway, so the check is not needed.
> 
> In addition, the check gets erroneously hit when using sample mode
> to trace multiple threads.
> 
> Consequently, fix that case by removing the check.

Thanks!

Reviewed-by: Andi Kleen <ak@linux.intel.com>

-Andi