github.com/emmansun/gmsm@v0.29.1/internal/bigmod/nat_s390x.s (about)

     1  // Copyright 2016 The Go Authors. All rights reserved.
     2  // Use of this source code is governed by a BSD-style
     3  // license that can be found in the LICENSE file.
     4  
     5  //go:build !purego
     6  
     7  #include "textflag.h"
     8  
     9  // func addMulVVW256(z, x *uint, y uint) (c uint)
    10  TEXT ·addMulVVW256(SB), $0-32
    11  	MOVD	$4, R5
    12  	JMP		addMulVVWy(SB)
    13  
    14  // func addMulVVW1024(z, x *uint, y uint) (c uint)
    15  TEXT ·addMulVVW1024(SB), $0-32
    16  	MOVD	$16, R5
    17  	JMP		addMulVVWy(SB)
    18  
    19  // func addMulVVW1536(z, x *uint, y uint) (c uint)
    20  TEXT ·addMulVVW1536(SB), $0-32
    21  	MOVD	$24, R5
    22  	JMP		addMulVVWy(SB)
    23  
    24  // func addMulVVW2048(z, x *uint, y uint) (c uint)
    25  TEXT ·addMulVVW2048(SB), $0-32
    26  	MOVD	$32, R5
    27  	JMP		addMulVVWy(SB)
    28  
    29  TEXT addMulVVWy(SB), NOFRAME|NOSPLIT, $0
    30  	MOVD z+0(FP), R2
    31  	MOVD x+8(FP), R8
    32  	MOVD y+16(FP), R9
    33  
    34  	MOVD $0, R1 // i*8 = 0
    35  	MOVD $0, R7 // i = 0
    36  	MOVD $0, R0 // make sure it's zero
    37  	MOVD $0, R4 // c = 0
    38  
    39  	MOVD   R5, R12
    40  	AND    $-2, R12
    41  	CMPBGE R5, $2, A6
    42  	BR     E6
    43  
    44  A6:
    45  	MOVD   (R8)(R1*1), R6
    46  	MULHDU R9, R6
    47  	MOVD   (R2)(R1*1), R10
    48  	ADDC   R10, R11        // add to low order bits
    49  	ADDE   R0, R6
    50  	ADDC   R4, R11
    51  	ADDE   R0, R6
    52  	MOVD   R6, R4
    53  	MOVD   R11, (R2)(R1*1)
    54  
    55  	MOVD   (8)(R8)(R1*1), R6
    56  	MULHDU R9, R6
    57  	MOVD   (8)(R2)(R1*1), R10
    58  	ADDC   R10, R11           // add to low order bits
    59  	ADDE   R0, R6
    60  	ADDC   R4, R11
    61  	ADDE   R0, R6
    62  	MOVD   R6, R4
    63  	MOVD   R11, (8)(R2)(R1*1)
    64  
    65  	ADD $16, R1 // i*8 + 8
    66  	ADD $2, R7  // i++
    67  
    68  	CMPBLT R7, R12, A6
    69  	BR     E6
    70  
    71  L6:
    72  	// TODO: drop unused single-step loop.
    73  	MOVD   (R8)(R1*1), R6
    74  	MULHDU R9, R6
    75  	MOVD   (R2)(R1*1), R10
    76  	ADDC   R10, R11        // add to low order bits
    77  	ADDE   R0, R6
    78  	ADDC   R4, R11
    79  	ADDE   R0, R6
    80  	MOVD   R6, R4
    81  	MOVD   R11, (R2)(R1*1)
    82  
    83  	ADD $8, R1 // i*8 + 8
    84  	ADD $1, R7 // i++
    85  
    86  E6:
    87  	CMPBLT R7, R5, L6 // i < n
    88  
    89  	MOVD R4, c+24(FP)
    90  	RET