This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [PATCH, rs6000] Don't convert a vector constant load into a splat illegally


On Thu, Oct 17, 2013 at 10:43 PM, Bill Schmidt
<wschmidt@linux.vnet.ibm.com> wrote:
> Hi,
>
> In little endian mode, we managed to convert a load of the V4SI vector
> {3, 3, 3, 7} into a vspltisw of 3, apparently taking offense at the
> number 7.  It turns out we only looked at the first N-1 elements of an
> N-element vector in little endian mode, and verified the zeroth element
> twice.  Adjusting the loop boundaries fixes the problem.
>
> Currently bootstrapping for powerpc64{,le}-unknown-linux-gnu.  Ok to
> commit to trunk if no regressions?

This patch does not make sense. It biases the loop bounds based on
BYTES_BIG_ENDIAN, but the body of the loop biases the index variable
based on BYTES_BIG_ENDIAN in one of the two uses.  The changes seem to
compute the same value for the first use of index "i" for both
BYTES_BIG_ENDIAN and !BYTES_BIG_ENDIAN in a convoluted way.  It looks
like something should be able to be simplified.

Thanks, David


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]