Re: [eigen] integer types - something's odd...

[ Thread Index | Date Index | More lists.tuxfamily.org/eigen Archives ]


On Tue, Jun 8, 2010 at 11:08 AM, Gael Guennebaud
<gael.guennebaud@xxxxxxxxx> wrote:
> This fix makes sense to me. What does not make sense is that changing s1*m2
> to (s1*m2).eval() fix the issue while I would expect the contrary since the
> former is just an expression which does not require alignment while the
> later is a Matrix which does require alignment. Maybe it is just a
> coincidence.

This is true. I am also puzzled why I am unable to reproduce the issue
with GCC -- it is not even possible after switching to debug. I did
not look up what kind of expression s1*m1 returns but I am wondering
why it is causing trouble in this case since as you said, we are just
passing an expression. Maybe the expression is holding a fixed size
member? Like the matrix-matrix products.

- Hauke



Mail converted by MHonArc 2.6.19+ http://listengine.tuxfamily.org/