Date:	Wed, 24 Sep 2008 18:11:25 +0200
From:	Peter Zijlstra <peterz@...radead.org>
To:	Martin Bligh <mbligh@...gle.com>
Cc:	Steven Rostedt <rostedt@...dmis.org>, linux-kernel@...r.kernel.org,
	Ingo Molnar <mingo@...e.hu>,
	Thomas Gleixner <tglx@...utronix.de>,
	Andrew Morton <akpm@...ux-foundation.org>,
	prasad@...ux.vnet.ibm.com,
	Linus Torvalds <torvalds@...ux-foundation.org>,
	Mathieu Desnoyers <compudj@...stal.dyndns.org>,
	"Frank Ch. Eigler" <fche@...hat.com>,
	David Wilder <dwilder@...ibm.com>, hch@....de,
	Tom Zanussi <zanussi@...cast.net>,
	Steven Rostedt <srostedt@...hat.com>
Subject: Re: [RFC PATCH 1/3] Unified trace buffer

On Wed, 2008-09-24 at 08:47 -0700, Martin Bligh wrote:
> Thanks for creating this so quickly ;-)
> 
> >> We can record either the fast way of reserving a part of the buffer:
> >>
> >> event = ring_buffer_lock_reserve(buffer, event_id, length, &flags);
> >> event->data = record_this_data;
> >> ring_buffer_unlock_commit(buffer, event, flags);
> >
> > This can, in generic, not work. Due to the simple fact that we might
> > straddle a page boundary. Therefore I think its best to limit our self
> > to the write interface below, so that it can handle that.
> 
> I'm not sure why this is any harder to deal with in write, than it is
> in reserve? We should be able to make reserve handle this just
> as well?

No, imagine the mentioned case where we're straddling a page boundary.

A----|   |----B
    ^------|

So when we reserve we get a pointer into page A, but our reserve length
will run over into page B. A write() method will know how to check for
this and break up the memcpy to copy up-to the end of A and continue
into B.

You cannot expect the reserve/commit interface users to do this
correctly - it would also require one to expose too much internals,
you'd need to be able to locate page B for starters.

> If you use write rather than reserve, you have to copy all the data
> twice for every event.

Well, once. I'm not seeing where the second copy comes from.

> > On top of that foundation build an eventbuffer, which knows about
> > encoding/decoding/printing events.
> >
> > This too needs to be a flexible layer -
> 
> That would be nice. However, we need to keep at least the length
> and timestamp fields common so we can do parsing and the mergesort?

And here I was thinking you guys bit encoded the event id into the
timestamp delta :-)

> +struct ring_buffer_event {
> +       unsigned long long counter;
> +       short type;
> +       short length;
> +       char body[];
> +} __attribute__((__packed__))
> 
> So type would move into the body here?

All of it would, basically I have no notion of an event in the
ringbuffer API. You write $something and your read routine would need to
be smart enough to figure it out.

The trivial case is a fixed size entry, in which case you always know
how much to read. A slightly more involved but still easy to understand
example might be a 7bit encoding and using the 8th bit for continuation.

Another option is to start out with a fixed sized header that contains a
length field.

But the raw ringbuffer layer, the one concerned with fiddling the pages
and writing/reading thereto need not be aware of anything else.

> > as I suspect the google guys
> > will want their ultra-compressed events back.
> 
> Is useful when gathering GB of data across 10,000 machines ;-)
> Also reduces general overhead for everyone to keep events small.

Exactly - which is why a flexible encoding layer makes sense to me -
aside from the abstraction itself.

--
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