Yuliang Wang writes:
> Hi Richard,
>
> Thanks for your comments and advice; I have applied the relevant changes.
>
> Regards,
> Yuliang
>
>
> UPDATE:
>
> Added new tests. Built and regression tested on aarch64-none-elf and
> aarch64-linux-gnu.
>
> gcc/ChangeLog:
>
> 2019-09-1 Yuliang Wang
>
>
):
Return true for AArch64 without SVE2.
-Original Message-
From: Richard Sandiford
Sent: 30 August 2019 12:49
To: Yuliang Wang
Cc: gcc-patches@gcc.gnu.org; nd
Subject: Re: [PATCH][AArch64] Vectorize MULH(R)S patterns with SVE2 instructions
Thanks for doing this. The patch looks
Thanks for doing this. The patch looks good, so this review is mostly a
list of very minor formatting comments, sorry.
Yuliang Wang writes:
> 2019-08-22 Yuliang Wang
>
Please add a line here pointing at the PR:
PR tree-optimization/89386
The commit hooks pick this up automatically
This patch allows for more efficient SVE2 vectorization of Multiply High with
Round and Scale (MULHRS) patterns.
The example snippet:
uint16_t a[N], b[N], c[N];
void foo_round (void)
{
for (int i = 0; i < N; i++)
a[i] = int32_t)b[i] * (int32_t)c[i]) >> 14) +