Re: [PATCH 4/5] perf record: mmap output file - v5

From: David Ahern <dsahern@gmail.com>
To: Peter Zijlstra <peterz@infradead.org>, Ingo Molnar <mingo@kernel.org>
Cc: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>,
	linux-kernel@vger.kernel.org, jolsa@redhat.com,
	Frederic Weisbecker <fweisbec@gmail.com>,
	Namhyung Kim <namhyung@kernel.org>,
	Mike Galbraith <efault@gmx.de>,
	Stephane Eranian <eranian@google.com>
Subject: Re: [PATCH 4/5] perf record: mmap output file - v5
Date: Fri, 15 Nov 2013 09:41:53 -0700	[thread overview]
Message-ID: <52864ED1.1080607@gmail.com> (raw)
In-Reply-To: <20131113113439.GI21461@twins.programming.kicks-ass.net>

On 11/13/13, 4:34 AM, Peter Zijlstra wrote:
>>> one option here is not allow page faults and system wide system calls.
>>> system wide tracing needs mmap; page faults for a task can use write().
>>> I left that option in case something like this came up.
>>
>> So maybe splice() sounds like the right long term solution after all? :-/
>
> Right until you put a tracepoint (kprobe) somewhere in whatever function
> is used to transfer a single page into/from a splice pipe.
>
> You can always screw yourself over using this stuff, no exceptions.
>

What now? Can we add the mmap path as an option? Leave write by default 
and users can select mmap if they want? e.g., have out_size default to 0 
and 'perf trace record' can set --out-pages to 64M. It bypasses all 
system calls at the expense of more page faults.

David