On Fri, 22 Mar 2019 13:43:20 PDT (-0700), c...@braap.org wrote:
From: Kito Cheng
Before falling back to softfloat FMA, we do not restore the original
values of inputs A and C. Fix it.
This bug was caught by running gcc's testsuite on RISC-V qemu.
Note that this change gives a small perf incre
On 3/22/19 1:43 PM, Emilio G. Cota wrote:
> From: Kito Cheng
>
> Before falling back to softfloat FMA, we do not restore the original
> values of inputs A and C. Fix it.
Reviewed-by: Richard Henderson
r~
Emilio G. Cota writes:
> From: Kito Cheng
>
> Before falling back to softfloat FMA, we do not restore the original
> values of inputs A and C. Fix it.
>
> This bug was caught by running gcc's testsuite on RISC-V qemu.
>
> Note that this change gives a small perf increase for fp-bench:
>
> Ho
From: Kito Cheng
Before falling back to softfloat FMA, we do not restore the original
values of inputs A and C. Fix it.
This bug was caught by running gcc's testsuite on RISC-V qemu.
Note that this change gives a small perf increase for fp-bench:
Host: Intel(R) Core(TM) i7-4790K CPU @ 4.00GH