Xiaolin Zhao e6cc9d228a cmd/compile: implement FMA codegen for loong64 ...

Benchmark results on Loongson 3A5000 and 3A6000:

goos: linux
goarch: loong64
pkg: math
cpu: Loongson-3A6000 @ 2500.00MHz
    |  bench.old   |              bench.new              |
    |    sec/op    |   sec/op     vs base                |
FMA   25.930n ± 0%   2.002n ± 0%  -92.28% (p=0.000 n=10)

goos: linux
goarch: loong64
pkg: math
cpu: Loongson-3A5000 @ 2500.00MHz
    |  bench.old   |              bench.new              |
    |    sec/op    |   sec/op     vs base                |
FMA   32.840n ± 0%   2.002n ± 0%  -93.90% (p=0.000 n=10)

Updates #59120

This patch is a copy of CL 483355.
Co-authored-by: WANG Xuerui <git@xen0n.name>

Change-Id: I88b89d23f00864f9173a182a47ee135afec7ed6e
Reviewed-on: https://go-review.googlesource.com/c/go/+/625335
Reviewed-by: abner chenc <chenguoqi@loongson.cn>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: David Chase <drchase@google.com>
Reviewed-by: Carlos Amedee <carlos@golang.org>

2024-11-08 01:05:48 +00:00

5.4 KiB

Raw Permalink Blame History

View Raw