Richard Biener writes:
> Index: gcc/testsuite/gcc.dg/vect/bb-slp-32.c
> ===
> *** gcc/testsuite/gcc.dg/vect/bb-slp-32.c (revision 0)
> --- gcc/testsuite/gcc.dg/vect/bb-slp-32.c (working copy)
> ***
> *** 0
>
The patch breaks compiling Open MPI with default compilation flags (for
that file: -O3 -fno-strict-aliasing):
../../../../../ompi/mca/btl/tcp/btl_tcp_frag.c:100:6: internal compiler
error: in operator[], at vec.h:815
bool mca_btl_tcp_frag_send(mca_btl_tcp_frag_t* frag, int sd)
See PR 57453
This properly accounts for remaining unvectorized scalar uses of
stmts we vectorize during BB vectorization. Previously we
happily retained all of the scalar loads and computations, just
optimizing the final stores, alongside the vectorized code.
To account for scalar uses properly the following
