]> Cypherpunks repositories - gostls13.git/commit
cmd/compile: implement FMA codegen for loong64
authorXiaolin Zhao <zhaoxiaolin@loongson.cn>
Tue, 5 Nov 2024 07:30:45 +0000 (15:30 +0800)
committerabner chenc <chenguoqi@loongson.cn>
Fri, 8 Nov 2024 01:05:48 +0000 (01:05 +0000)
commite6cc9d228a913b871fb20a3fa23021bd5231dbc0
treeae79e9601b394e6a8a7cb9aa975918158ff3e220
parent2751443e92d9ac6adee2ce0433253fd7b2c04b0f
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>
src/cmd/compile/internal/loong64/ssa.go
src/cmd/compile/internal/ssa/_gen/LOONG64.rules
src/cmd/compile/internal/ssa/_gen/LOONG64Ops.go
src/cmd/compile/internal/ssa/opGen.go
src/cmd/compile/internal/ssa/rewriteLOONG64.go
src/cmd/compile/internal/ssagen/intrinsics.go
src/cmd/compile/internal/ssagen/intrinsics_test.go
test/codegen/floats.go
test/codegen/math.go