Sorry for late reply,
On Mon, Oct 21, 2019 at 6:12 PM Darcy Beurle <
[email protected]> wrote:
> Is there anyway to access the alignment through a type trait of the
> matrix class? Passing the correct alignment of the underlying storage
> to the Eigen::Map would then be a simple task, albeit ugly.
>
I don't see how this could help in this context, but you can use some
internal mechanisms (not guaranteed to be API stable):
internal::traits<MatrixXf>::Alignment
or for statically allocated matrix types:
alignof(Matrix4f)
Gaël
>
> Cheers,
>
> Darcy
>
> On Mon, 2019-10-21 at 16:35 +0200, Christoph Hertzberg wrote:
> > If you can ensure that the allocated memory is aligned, you can use
> > `Map<Matrix<...>, Aligned>`. Also, we do allow vectorization of
> > non-aligned data (with newer CPUs there is barely a performance
> > difference, except when loads/store are split between cache-lines).
> >
> > Btw: There is a macro `ei_declare_aligned_stack_constructed_variable`
> > in
> > Eigen/src/Core/util/Memory.h which already does most of what you need
> > to
> > do (grep the source code to see where/how it is used).
> >
> > Cheers,
> > Christoph
> >
> >
> > On 21/10/2019 16.21, Rob Conde wrote:
> > > Thanks...I will think about this and maybe try to prototype
> > > something.
> > >
> > > Using the Map approach...I think this would disable SIMD right?
> > > Because you don't know if the mapped data is aligned?
> > >
> > > Rob
> > > ________________________________
> > > From: Christoph Hertzberg <[email protected]>
> > > Sent: Monday, October 21, 2019 10:15 AM
> > > To: [email protected] <[email protected]>
> > > Subject: Re: [eigen] Small dynamic matrix optimization
> > >
> > > This is already done for some temporaries (like blocks in
> > > cache-optimized GEMM). Doing that for custom types would be
> > > interesting,
> > > but not trivial (e.g., move-behavior is completely different, if
> > > you
> > > want to move the data to an object higher in the stack).
> > >
> > > If you are ok with managing data yourself, you can manually achieve
> > > this
> > > by alloca/allocate-ing data yourself and use an Eigen::Map to refer
> > > to
> > > the data.
> > >
> > > If you have a proof-of-concept how this could be integrated into
> > > the
> > > Eigen API, we are open for suggestions!
> > >
> > > Another idea would be to allow custom allocators for matrices -- in
> > > that
> > > case the custom allocator could very easily be a stack-allocator
> > > (which
> > > would not even need to use the "main" stack). But that's also not
> > > too
> > > easy to integrate without breaking the current API (adding a new
> > > type
> > > would of course be an option for that).
> > >
> > > Cheers,
> > > Christoph
> > >
> > >
> > >
> > > On 21/10/2019 15.39, Rob Conde wrote:
> > > > I'm wondering if a small dynamic matrix optimization has been
> > > > considered. In other words, allowing a stack allocation for
> > > > matrices of unknown size up to a certain limit and using a heap
> > > > allocation beyond that?
> > > >
> > > > Rob Conde
> > > >
> > >
> > > --
> > > Dr.-Ing. Christoph Hertzberg
> > >
> > > Besuchsadresse der Nebengeschäftsstelle:
> > > DFKI GmbH
> > > Robotics Innovation Center
> > > Robert-Hooke-Straße 5
> > > 28359 Bremen, Germany
> > >
> > > Postadresse der Hauptgeschäftsstelle Standort Bremen:
> > > DFKI GmbH
> > > Robotics Innovation Center
> > > Robert-Hooke-Straße 1
> > > 28359 Bremen, Germany
> > >
> > > Tel.: +49 421 178 45-4021
> > > Zentrale: +49 421 178 45-0
> > > E-Mail: [email protected]
> > >
> > > Weitere Informationen: http://www.dfki.de/robotik
> > > -------------------------------------------------------------
> > > Deutsches Forschungszentrum für Künstliche Intelligenz GmbH
> > > Trippstadter Strasse 122, D-67663 Kaiserslautern, Germany
> > >
> > > Geschäftsführung:
> > > Prof. Dr. Jana Koehler (Vorsitzende)
> > > Dr. Walter Olthoff
> > >
> > > Vorsitzender des Aufsichtsrats:
> > > Prof. Dr. h.c. Hans A. Aukes
> > > Amtsgericht Kaiserslautern, HRB 2313
> > > -------------------------------------------------------------
> > >
> > >
> > >
> > >
>
>
>
>
