From: Frank Ch. Eigler on
Hi -

On Wed, May 19, 2010 at 11:08:06PM +0530, Srikar Dronamraju wrote:

> [...] For global tracing we can't use access_process_vm. [...]

Well, we have already demonstrated functional "global tracing" via
access_process_vm, with the trade-off that affected pages are
replicated on a per-process basis. (KSM could end up sharing them
again, one supposes.)

- FChE
--
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: Peter Zijlstra on
On Wed, 2010-05-19 at 23:08 +0530, Srikar Dronamraju wrote:

> Or do you suggest any alternative methods for global tracing?

COW the page for each existing and future mapping.

--
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: Srikar Dronamraju on
* Peter Zijlstra <peterz(a)infradead.org> [2010-05-19 19:54:20]:

> On Wed, 2010-05-19 at 23:08 +0530, Srikar Dronamraju wrote:
>
> > Or do you suggest any alternative methods for global tracing?
>
> COW the page for each existing and future mapping.

Yes, we could certainly do this. However we could end up with multiple
copies of the same page. And probably equal number of uprobe and probept
structures.

--
Thanks and Regards
Srikar

>
--
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/