From: Peter Zijlstra on
On Mon, 2010-06-28 at 19:08 +0100, Eric B Munson wrote:
> This patch adds a new software event for munmaps. It will allows
> users to profile changes to address space. munmaps will be tracked
> with mmaps.

Why?
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Eric B Munson on
On Mon, 28 Jun 2010, Peter Zijlstra wrote:

> On Mon, 2010-06-28 at 19:08 +0100, Eric B Munson wrote:
> > This patch adds a new software event for munmaps. It will allows
> > users to profile changes to address space. munmaps will be tracked
> > with mmaps.
>
> Why?
>

It is going to be used by a tool that will model memory usage over the
lifetime of a process.

--
Eric B Munson
IBM Linux Technology Center
ebmunson(a)us.ibm.com

From: Peter Zijlstra on
On Tue, 2010-06-29 at 09:33 +0100, Eric B Munson wrote:
> On Mon, 28 Jun 2010, Peter Zijlstra wrote:
>
> > On Mon, 2010-06-28 at 19:08 +0100, Eric B Munson wrote:
> > > This patch adds a new software event for munmaps. It will allows
> > > users to profile changes to address space. munmaps will be tracked
> > > with mmaps.
> >
> > Why?
> >
>
> It is going to be used by a tool that will model memory usage over the
> lifetime of a process.

Wouldn't it be better to use some tracepoints for that instead? I want
to keep the sideband data to a minimum required to interpret the sample
data, and you don't need unmap events for that.


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Eric B Munson on
On Tue, 29 Jun 2010, Peter Zijlstra wrote:

> On Tue, 2010-06-29 at 09:33 +0100, Eric B Munson wrote:
> > On Mon, 28 Jun 2010, Peter Zijlstra wrote:
> >
> > > On Mon, 2010-06-28 at 19:08 +0100, Eric B Munson wrote:
> > > > This patch adds a new software event for munmaps. It will allows
> > > > users to profile changes to address space. munmaps will be tracked
> > > > with mmaps.
> > >
> > > Why?
> > >
> >
> > It is going to be used by a tool that will model memory usage over the
> > lifetime of a process.
>
> Wouldn't it be better to use some tracepoints for that instead? I want
> to keep the sideband data to a minimum required to interpret the sample
> data, and you don't need unmap events for that.
>
>

Sure, I will get it moved to a tracepoint event instead.

--
Eric B Munson
IBM Linux Technology Center
ebmunson(a)us.ibm.com