On 14 November 2014 10:46, Alan Lawrence wrote:
> This patch replaces the inline asm for vld1_dup intrinsics with a vdup_n_
> and a load from the pointer. The existing *aarch64_simd_ld1r insn,
> combiner, etc., are quite capable of generating the expected single ld1r
> instruction from this. (I've
This patch replaces the inline asm for vld1_dup intrinsics with a vdup_n_ and a
load from the pointer. The existing *aarch64_simd_ld1r insn, combiner,
etc., are quite capable of generating the expected single ld1r instruction from
this. (I've verified by inspecting assembler output.)
gcc/Chang