[llvm] r200607 - Update a .fill test to use the updated semantics.

[David Majnemer]

Fill only supports 32-bit patterns.  If you ask for more than four bytes,
you get zeros beyond that.  Implementing .fill as pushing out a four byte
number followed by some number of zeros gets us behavior compatible with
.fill under big endian targets.


On Mon, Feb 3, 2014 at 12:58 PM, Rafael Espíndola <
rafael.espindola at gmail.com> wrote:

> >  # CHECK: TEST2:
> > -# CHECK: .quad 4
> > +# CHECK: .long 4
> > +# CHECK: .long 0
> >  TEST2:
>
> Is this desirable? The old code was using a .quad in a target that
> doesn't support it?
>
> Cheers,
> Rafael
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20140203/6a3087ed/attachment.html>