qbit/go
1
0
Fork 0
mirror of https://github.com/golang/go synced 2024-09-29 19:24:33 -06:00
Code Releases Activity

math/big: initial vector arithmetic in riscv64 assembly

Browse Source 
Provide an assembly implementation of mulWW - for now all others run the
Go code.

Change-Id: Icb594c31048255f131bdea8d64f56784fc9db4d1
Reviewed-on: https://go-review.googlesource.com/c/go/+/220919
Reviewed-by: Robert Griesemer <gri@golang.org>
This commit is contained in:
Joel Sing 2020-02-26 03:09:59 +11:00 committed by Robert Griesemer
parent 89f249a40d
commit fe70838598
3 changed files with 53 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
src/math/big/arith_decl.go
View File

@ -2,7 +2,7 @@
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build !math_big_pure_go,!riscv64
// +build !math_big_pure_go
package big

2
src/math/big/arith_decl_pure.go
View File

@ -2,7 +2,7 @@
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build math_big_pure_go riscv64
// +build math_big_pure_go
package big

51
src/math/big/arith_riscv64.s Normal file
View File

@ -0,0 +1,51 @@
// Copyright 2020 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build !math_big_pure_go,riscv64
#include "textflag.h"
// This file provides fast assembly versions for the elementary
// arithmetic operations on vectors implemented in arith.go.
// func mulWW(x, y Word) (z1, z0 Word)
TEXT ·mulWW(SB),NOSPLIT,$0
MOV x+0(FP), X5
MOV y+8(FP), X6
MULHU X5, X6, X7
MUL X5, X6, X8
MOV X7, z1+16(FP)
MOV X8, z0+24(FP)
RET
// func divWW(x1, x0, y Word) (q, r Word)
TEXT ·divWW(SB),NOSPLIT,$0
JMP ·divWW_g(SB) // riscv64 has no multiword division
TEXT ·addVV(SB),NOSPLIT,$0
JMP ·addVV_g(SB)
TEXT ·subVV(SB),NOSPLIT,$0
JMP ·subVV_g(SB)
TEXT ·addVW(SB),NOSPLIT,$0
JMP ·addVW_g(SB)
TEXT ·subVW(SB),NOSPLIT,$0
JMP ·subVW_g(SB)
TEXT ·shlVU(SB),NOSPLIT,$0
JMP ·shlVU_g(SB)
TEXT ·shrVU(SB),NOSPLIT,$0
JMP ·shrVU_g(SB)
TEXT ·mulAddVWW(SB),NOSPLIT,$0
JMP ·mulAddVWW_g(SB)
TEXT ·addMulVVW(SB),NOSPLIT,$0
JMP ·addMulVVW_g(SB)
TEXT ·divWVW(SB),NOSPLIT,$0
JMP ·divWVW_g(SB)