cybercyst/go
1
0
Fork 0
mirror of https://github.com/golang/go.git synced 2025-05-05 15:43:04 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

math/big: fix incorrect register allocation for mipsx/mips64x

Browse Source 
According to the MIPS ABI, R26/R27 are reserved for OS kernel, and may be clobbered by it. They must not be used by user mode.

See Figure 3-18 of MIPS ELF ABI specification: https://refspecs.linuxfoundation.org/elf/mipsabi.pdf

Fixes #73472

Change-Id: Ifda692a803176bfaab2c70d6623636c5d135f42e
Reviewed-on: https://go-review.googlesource.com/c/go/+/667816
Reviewed-by: Alan Donovan <adonovan@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
Reviewed-by: Keith Randall <khr@google.com>
This commit is contained in:
Julian Zhu 2025-04-25 20:59:13 +08:00 committed by Keith Randall
parent b48e52b428
commit 6109185cf9
4 changed files with 186 additions and 186 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

178
src/math/big/arith_mips64x.s
View File

@ -17,18 +17,18 @@ TEXT ·addVV(SB), NOSPLIT, $0
// compute unrolled loop lengths // compute unrolled loop lengths
AND $3, R1, R5 AND $3, R1, R5
SRLV $2, R1 SRLV $2, R1
XOR R26, R26 // clear carry XOR R24, R24 // clear carry
loop1: loop1:
BEQ R5, loop1done BEQ R5, loop1done
loop1cont: loop1cont:
// unroll 1X // unroll 1X
MOVV 0(R2), R6 MOVV 0(R2), R6
MOVV 0(R3), R7 MOVV 0(R3), R7
ADDVU R7, R6 // ADCS R7, R6, R6 (cr=R26) ADDVU R7, R6 // ADCS R7, R6, R6 (cr=R24)
SGTU R7, R6, R23 // ... SGTU R7, R6, R23 // ...
ADDVU R26, R6 // ... ADDVU R24, R6 // ...
SGTU R26, R6, R26 // ... SGTU R24, R6, R24 // ...
ADDVU R23, R26 // ... ADDVU R23, R24 // ...
MOVV R6, 0(R4) MOVV R6, 0(R4)
ADDVU $8, R2 ADDVU $8, R2
ADDVU $8, R3 ADDVU $8, R3
@ -48,26 +48,26 @@ loop4cont:
MOVV 8(R3), R10 MOVV 8(R3), R10
MOVV 16(R3), R11 MOVV 16(R3), R11
MOVV 24(R3), R12 MOVV 24(R3), R12
ADDVU R9, R5 // ADCS R9, R5, R5 (cr=R26) ADDVU R9, R5 // ADCS R9, R5, R5 (cr=R24)
SGTU R9, R5, R23 // ... SGTU R9, R5, R23 // ...
ADDVU R26, R5 // ... ADDVU R24, R5 // ...
SGTU R26, R5, R26 // ... SGTU R24, R5, R24 // ...
ADDVU R23, R26 // ... ADDVU R23, R24 // ...
ADDVU R10, R6 // ADCS R10, R6, R6 (cr=R26) ADDVU R10, R6 // ADCS R10, R6, R6 (cr=R24)
SGTU R10, R6, R23 // ... SGTU R10, R6, R23 // ...
ADDVU R26, R6 // ... ADDVU R24, R6 // ...
SGTU R26, R6, R26 // ... SGTU R24, R6, R24 // ...
ADDVU R23, R26 // ... ADDVU R23, R24 // ...
ADDVU R11, R7 // ADCS R11, R7, R7 (cr=R26) ADDVU R11, R7 // ADCS R11, R7, R7 (cr=R24)
SGTU R11, R7, R23 // ... SGTU R11, R7, R23 // ...
ADDVU R26, R7 // ... ADDVU R24, R7 // ...
SGTU R26, R7, R26 // ... SGTU R24, R7, R24 // ...
ADDVU R23, R26 // ... ADDVU R23, R24 // ...
ADDVU R12, R8 // ADCS R12, R8, R8 (cr=R26) ADDVU R12, R8 // ADCS R12, R8, R8 (cr=R24)
SGTU R12, R8, R23 // ... SGTU R12, R8, R23 // ...
ADDVU R26, R8 // ... ADDVU R24, R8 // ...
SGTU R26, R8, R26 // ... SGTU R24, R8, R24 // ...
ADDVU R23, R26 // ... ADDVU R23, R24 // ...
MOVV R5, 0(R4) MOVV R5, 0(R4)
MOVV R6, 8(R4) MOVV R6, 8(R4)
MOVV R7, 16(R4) MOVV R7, 16(R4)
@ -78,7 +78,7 @@ loop4cont:
SUBVU $1, R1 SUBVU $1, R1
BNE R1, loop4cont BNE R1, loop4cont
loop4done: loop4done:
MOVV R26, c+72(FP) MOVV R24, c+72(FP)
RET RET
// func subVV(z, x, y []Word) (c Word) // func subVV(z, x, y []Word) (c Word)
@ -90,18 +90,18 @@ TEXT ·subVV(SB), NOSPLIT, $0
// compute unrolled loop lengths // compute unrolled loop lengths
AND $3, R1, R5 AND $3, R1, R5
SRLV $2, R1 SRLV $2, R1
XOR R26, R26 // clear carry XOR R24, R24 // clear carry
loop1: loop1:
BEQ R5, loop1done BEQ R5, loop1done
loop1cont: loop1cont:
// unroll 1X // unroll 1X
MOVV 0(R2), R6 MOVV 0(R2), R6
MOVV 0(R3), R7 MOVV 0(R3), R7
SGTU R26, R6, R23 // SBCS R7, R6, R6 SGTU R24, R6, R23 // SBCS R7, R6, R6
SUBVU R26, R6 // ... SUBVU R24, R6 // ...
SGTU R7, R6, R26 // ... SGTU R7, R6, R24 // ...
SUBVU R7, R6 // ... SUBVU R7, R6 // ...
ADDVU R23, R26 // ... ADDVU R23, R24 // ...
MOVV R6, 0(R4) MOVV R6, 0(R4)
ADDVU $8, R2 ADDVU $8, R2
ADDVU $8, R3 ADDVU $8, R3
@ -121,26 +121,26 @@ loop4cont:
MOVV 8(R3), R10 MOVV 8(R3), R10
MOVV 16(R3), R11 MOVV 16(R3), R11
MOVV 24(R3), R12 MOVV 24(R3), R12
SGTU R26, R5, R23 // SBCS R9, R5, R5 SGTU R24, R5, R23 // SBCS R9, R5, R5
SUBVU R26, R5 // ... SUBVU R24, R5 // ...
SGTU R9, R5, R26 // ... SGTU R9, R5, R24 // ...
SUBVU R9, R5 // ... SUBVU R9, R5 // ...
ADDVU R23, R26 // ... ADDVU R23, R24 // ...
SGTU R26, R6, R23 // SBCS R10, R6, R6 SGTU R24, R6, R23 // SBCS R10, R6, R6
SUBVU R26, R6 // ... SUBVU R24, R6 // ...
SGTU R10, R6, R26 // ... SGTU R10, R6, R24 // ...
SUBVU R10, R6 // ... SUBVU R10, R6 // ...
ADDVU R23, R26 // ... ADDVU R23, R24 // ...
SGTU R26, R7, R23 // SBCS R11, R7, R7 SGTU R24, R7, R23 // SBCS R11, R7, R7
SUBVU R26, R7 // ... SUBVU R24, R7 // ...
SGTU R11, R7, R26 // ... SGTU R11, R7, R24 // ...
SUBVU R11, R7 // ... SUBVU R11, R7 // ...
ADDVU R23, R26 // ... ADDVU R23, R24 // ...
SGTU R26, R8, R23 // SBCS R12, R8, R8 SGTU R24, R8, R23 // SBCS R12, R8, R8
SUBVU R26, R8 // ... SUBVU R24, R8 // ...
SGTU R12, R8, R26 // ... SGTU R12, R8, R24 // ...
SUBVU R12, R8 // ... SUBVU R12, R8 // ...
ADDVU R23, R26 // ... ADDVU R23, R24 // ...
MOVV R5, 0(R4) MOVV R5, 0(R4)
MOVV R6, 8(R4) MOVV R6, 8(R4)
MOVV R7, 16(R4) MOVV R7, 16(R4)
@ -151,7 +151,7 @@ loop4cont:
SUBVU $1, R1 SUBVU $1, R1
BNE R1, loop4cont BNE R1, loop4cont
loop4done: loop4done:
MOVV R26, c+72(FP) MOVV R24, c+72(FP)
RET RET
// func lshVU(z, x []Word, s uint) (c Word) // func lshVU(z, x []Word, s uint) (c Word)
@ -316,9 +316,9 @@ loop1cont:
MULVU R1, R7 MULVU R1, R7
MOVV LO, R8 MOVV LO, R8
MOVV HI, R9 MOVV HI, R9
ADDVU R2, R8, R7 // ADDS R2, R8, R7 (cr=R26) ADDVU R2, R8, R7 // ADDS R2, R8, R7 (cr=R24)
SGTU R2, R7, R26 // ... SGTU R2, R7, R24 // ...
ADDVU R26, R9, R2 // ADC $0, R9, R2 ADDVU R24, R9, R2 // ADC $0, R9, R2
MOVV R7, 0(R5) MOVV R7, 0(R5)
ADDVU $8, R4 ADDVU $8, R4
ADDVU $8, R5 ADDVU $8, R5
@ -337,27 +337,27 @@ loop4cont:
MULVU R1, R6 MULVU R1, R6
MOVV LO, R10 MOVV LO, R10
MOVV HI, R11 MOVV HI, R11
ADDVU R2, R10, R6 // ADDS R2, R10, R6 (cr=R26) ADDVU R2, R10, R6 // ADDS R2, R10, R6 (cr=R24)
SGTU R2, R6, R26 // ... SGTU R2, R6, R24 // ...
ADDVU R26, R11, R2 // ADC $0, R11, R2 ADDVU R24, R11, R2 // ADC $0, R11, R2
MULVU R1, R7 MULVU R1, R7
MOVV LO, R10 MOVV LO, R10
MOVV HI, R11 MOVV HI, R11
ADDVU R2, R10, R7 // ADDS R2, R10, R7 (cr=R26) ADDVU R2, R10, R7 // ADDS R2, R10, R7 (cr=R24)
SGTU R2, R7, R26 // ... SGTU R2, R7, R24 // ...
ADDVU R26, R11, R2 // ADC $0, R11, R2 ADDVU R24, R11, R2 // ADC $0, R11, R2
MULVU R1, R8 MULVU R1, R8
MOVV LO, R10 MOVV LO, R10
MOVV HI, R11 MOVV HI, R11
ADDVU R2, R10, R8 // ADDS R2, R10, R8 (cr=R26) ADDVU R2, R10, R8 // ADDS R2, R10, R8 (cr=R24)
SGTU R2, R8, R26 // ... SGTU R2, R8, R24 // ...
ADDVU R26, R11, R2 // ADC $0, R11, R2 ADDVU R24, R11, R2 // ADC $0, R11, R2
MULVU R1, R9 MULVU R1, R9
MOVV LO, R10 MOVV LO, R10
MOVV HI, R11 MOVV HI, R11
ADDVU R2, R10, R9 // ADDS R2, R10, R9 (cr=R26) ADDVU R2, R10, R9 // ADDS R2, R10, R9 (cr=R24)
SGTU R2, R9, R26 // ... SGTU R2, R9, R24 // ...
ADDVU R26, R11, R2 // ADC $0, R11, R2 ADDVU R24, R11, R2 // ADC $0, R11, R2
MOVV R6, 0(R5) MOVV R6, 0(R5)
MOVV R7, 8(R5) MOVV R7, 8(R5)
MOVV R8, 16(R5) MOVV R8, 16(R5)
@ -391,12 +391,12 @@ loop1cont:
MULVU R1, R9 MULVU R1, R9
MOVV LO, R10 MOVV LO, R10
MOVV HI, R11 MOVV HI, R11
ADDVU R8, R10 // ADDS R8, R10, R10 (cr=R26) ADDVU R8, R10 // ADDS R8, R10, R10 (cr=R24)
SGTU R8, R10, R26 // ... SGTU R8, R10, R24 // ...
ADDVU R26, R11 // ADC $0, R11, R11 ADDVU R24, R11 // ADC $0, R11, R11
ADDVU R2, R10, R9 // ADDS R2, R10, R9 (cr=R26) ADDVU R2, R10, R9 // ADDS R2, R10, R9 (cr=R24)
SGTU R2, R9, R26 // ... SGTU R2, R9, R24 // ...
ADDVU R26, R11, R2 // ADC $0, R11, R2 ADDVU R24, R11, R2 // ADC $0, R11, R2
MOVV R9, 0(R6) MOVV R9, 0(R6)
ADDVU $8, R4 ADDVU $8, R4
ADDVU $8, R5 ADDVU $8, R5
@ -420,39 +420,39 @@ loop4cont:
MULVU R1, R11 MULVU R1, R11
MOVV LO, R15 MOVV LO, R15
MOVV HI, R16 MOVV HI, R16
ADDVU R7, R15 // ADDS R7, R15, R15 (cr=R26) ADDVU R7, R15 // ADDS R7, R15, R15 (cr=R24)
SGTU R7, R15, R26 // ... SGTU R7, R15, R24 // ...
ADDVU R26, R16 // ADC $0, R16, R16 ADDVU R24, R16 // ADC $0, R16, R16
ADDVU R2, R15, R11 // ADDS R2, R15, R11 (cr=R26) ADDVU R2, R15, R11 // ADDS R2, R15, R11 (cr=R24)
SGTU R2, R11, R26 // ... SGTU R2, R11, R24 // ...
ADDVU R26, R16, R2 // ADC $0, R16, R2 ADDVU R24, R16, R2 // ADC $0, R16, R2
MULVU R1, R12 MULVU R1, R12
MOVV LO, R15 MOVV LO, R15
MOVV HI, R16 MOVV HI, R16
ADDVU R8, R15 // ADDS R8, R15, R15 (cr=R26) ADDVU R8, R15 // ADDS R8, R15, R15 (cr=R24)
SGTU R8, R15, R26 // ... SGTU R8, R15, R24 // ...
ADDVU R26, R16 // ADC $0, R16, R16 ADDVU R24, R16 // ADC $0, R16, R16
ADDVU R2, R15, R12 // ADDS R2, R15, R12 (cr=R26) ADDVU R2, R15, R12 // ADDS R2, R15, R12 (cr=R24)
SGTU R2, R12, R26 // ... SGTU R2, R12, R24 // ...
ADDVU R26, R16, R2 // ADC $0, R16, R2 ADDVU R24, R16, R2 // ADC $0, R16, R2
MULVU R1, R13 MULVU R1, R13
MOVV LO, R15 MOVV LO, R15
MOVV HI, R16 MOVV HI, R16
ADDVU R9, R15 // ADDS R9, R15, R15 (cr=R26) ADDVU R9, R15 // ADDS R9, R15, R15 (cr=R24)
SGTU R9, R15, R26 // ... SGTU R9, R15, R24 // ...
ADDVU R26, R16 // ADC $0, R16, R16 ADDVU R24, R16 // ADC $0, R16, R16
ADDVU R2, R15, R13 // ADDS R2, R15, R13 (cr=R26) ADDVU R2, R15, R13 // ADDS R2, R15, R13 (cr=R24)
SGTU R2, R13, R26 // ... SGTU R2, R13, R24 // ...
ADDVU R26, R16, R2 // ADC $0, R16, R2 ADDVU R24, R16, R2 // ADC $0, R16, R2
MULVU R1, R14 MULVU R1, R14
MOVV LO, R15 MOVV LO, R15
MOVV HI, R16 MOVV HI, R16
ADDVU R10, R15 // ADDS R10, R15, R15 (cr=R26) ADDVU R10, R15 // ADDS R10, R15, R15 (cr=R24)
SGTU R10, R15, R26 // ... SGTU R10, R15, R24 // ...
ADDVU R26, R16 // ADC $0, R16, R16 ADDVU R24, R16 // ADC $0, R16, R16
ADDVU R2, R15, R14 // ADDS R2, R15, R14 (cr=R26) ADDVU R2, R15, R14 // ADDS R2, R15, R14 (cr=R24)
SGTU R2, R14, R26 // ... SGTU R2, R14, R24 // ...
ADDVU R26, R16, R2 // ADC $0, R16, R2 ADDVU R24, R16, R2 // ADC $0, R16, R2
MOVV R11, 0(R6) MOVV R11, 0(R6)
MOVV R12, 8(R6) MOVV R12, 8(R6)
MOVV R13, 16(R6) MOVV R13, 16(R6)

178
src/math/big/arith_mipsx.s
View File

@ -17,18 +17,18 @@ TEXT ·addVV(SB), NOSPLIT, $0
// compute unrolled loop lengths // compute unrolled loop lengths
AND $3, R1, R5 AND $3, R1, R5
SRL $2, R1 SRL $2, R1
XOR R26, R26 // clear carry XOR R24, R24 // clear carry
loop1: loop1:
BEQ R5, loop1done BEQ R5, loop1done
loop1cont: loop1cont:
// unroll 1X // unroll 1X
MOVW 0(R2), R6 MOVW 0(R2), R6
MOVW 0(R3), R7 MOVW 0(R3), R7
ADDU R7, R6 // ADCS R7, R6, R6 (cr=R26) ADDU R7, R6 // ADCS R7, R6, R6 (cr=R24)
SGTU R7, R6, R23 // ... SGTU R7, R6, R23 // ...
ADDU R26, R6 // ... ADDU R24, R6 // ...
SGTU R26, R6, R26 // ... SGTU R24, R6, R24 // ...
ADDU R23, R26 // ... ADDU R23, R24 // ...
MOVW R6, 0(R4) MOVW R6, 0(R4)
ADDU $4, R2 ADDU $4, R2
ADDU $4, R3 ADDU $4, R3
@ -48,26 +48,26 @@ loop4cont:
MOVW 4(R3), R10 MOVW 4(R3), R10
MOVW 8(R3), R11 MOVW 8(R3), R11
MOVW 12(R3), R12 MOVW 12(R3), R12
ADDU R9, R5 // ADCS R9, R5, R5 (cr=R26) ADDU R9, R5 // ADCS R9, R5, R5 (cr=R24)
SGTU R9, R5, R23 // ... SGTU R9, R5, R23 // ...
ADDU R26, R5 // ... ADDU R24, R5 // ...
SGTU R26, R5, R26 // ... SGTU R24, R5, R24 // ...
ADDU R23, R26 // ... ADDU R23, R24 // ...
ADDU R10, R6 // ADCS R10, R6, R6 (cr=R26) ADDU R10, R6 // ADCS R10, R6, R6 (cr=R24)
SGTU R10, R6, R23 // ... SGTU R10, R6, R23 // ...
ADDU R26, R6 // ... ADDU R24, R6 // ...
SGTU R26, R6, R26 // ... SGTU R24, R6, R24 // ...
ADDU R23, R26 // ... ADDU R23, R24 // ...
ADDU R11, R7 // ADCS R11, R7, R7 (cr=R26) ADDU R11, R7 // ADCS R11, R7, R7 (cr=R24)
SGTU R11, R7, R23 // ... SGTU R11, R7, R23 // ...
ADDU R26, R7 // ... ADDU R24, R7 // ...
SGTU R26, R7, R26 // ... SGTU R24, R7, R24 // ...
ADDU R23, R26 // ... ADDU R23, R24 // ...
ADDU R12, R8 // ADCS R12, R8, R8 (cr=R26) ADDU R12, R8 // ADCS R12, R8, R8 (cr=R24)
SGTU R12, R8, R23 // ... SGTU R12, R8, R23 // ...
ADDU R26, R8 // ... ADDU R24, R8 // ...
SGTU R26, R8, R26 // ... SGTU R24, R8, R24 // ...
ADDU R23, R26 // ... ADDU R23, R24 // ...
MOVW R5, 0(R4) MOVW R5, 0(R4)
MOVW R6, 4(R4) MOVW R6, 4(R4)
MOVW R7, 8(R4) MOVW R7, 8(R4)
@ -78,7 +78,7 @@ loop4cont:
SUBU $1, R1 SUBU $1, R1
BNE R1, loop4cont BNE R1, loop4cont
loop4done: loop4done:
MOVW R26, c+36(FP) MOVW R24, c+36(FP)
RET RET
// func subVV(z, x, y []Word) (c Word) // func subVV(z, x, y []Word) (c Word)
@ -90,18 +90,18 @@ TEXT ·subVV(SB), NOSPLIT, $0
// compute unrolled loop lengths // compute unrolled loop lengths
AND $3, R1, R5 AND $3, R1, R5
SRL $2, R1 SRL $2, R1
XOR R26, R26 // clear carry XOR R24, R24 // clear carry
loop1: loop1:
BEQ R5, loop1done BEQ R5, loop1done
loop1cont: loop1cont:
// unroll 1X // unroll 1X
MOVW 0(R2), R6 MOVW 0(R2), R6
MOVW 0(R3), R7 MOVW 0(R3), R7
SGTU R26, R6, R23 // SBCS R7, R6, R6 SGTU R24, R6, R23 // SBCS R7, R6, R6
SUBU R26, R6 // ... SUBU R24, R6 // ...
SGTU R7, R6, R26 // ... SGTU R7, R6, R24 // ...
SUBU R7, R6 // ... SUBU R7, R6 // ...
ADDU R23, R26 // ... ADDU R23, R24 // ...
MOVW R6, 0(R4) MOVW R6, 0(R4)
ADDU $4, R2 ADDU $4, R2
ADDU $4, R3 ADDU $4, R3
@ -121,26 +121,26 @@ loop4cont:
MOVW 4(R3), R10 MOVW 4(R3), R10
MOVW 8(R3), R11 MOVW 8(R3), R11
MOVW 12(R3), R12 MOVW 12(R3), R12
SGTU R26, R5, R23 // SBCS R9, R5, R5 SGTU R24, R5, R23 // SBCS R9, R5, R5
SUBU R26, R5 // ... SUBU R24, R5 // ...
SGTU R9, R5, R26 // ... SGTU R9, R5, R24 // ...
SUBU R9, R5 // ... SUBU R9, R5 // ...
ADDU R23, R26 // ... ADDU R23, R24 // ...
SGTU R26, R6, R23 // SBCS R10, R6, R6 SGTU R24, R6, R23 // SBCS R10, R6, R6
SUBU R26, R6 // ... SUBU R24, R6 // ...
SGTU R10, R6, R26 // ... SGTU R10, R6, R24 // ...
SUBU R10, R6 // ... SUBU R10, R6 // ...
ADDU R23, R26 // ... ADDU R23, R24 // ...
SGTU R26, R7, R23 // SBCS R11, R7, R7 SGTU R24, R7, R23 // SBCS R11, R7, R7
SUBU R26, R7 // ... SUBU R24, R7 // ...
SGTU R11, R7, R26 // ... SGTU R11, R7, R24 // ...
SUBU R11, R7 // ... SUBU R11, R7 // ...
ADDU R23, R26 // ... ADDU R23, R24 // ...
SGTU R26, R8, R23 // SBCS R12, R8, R8 SGTU R24, R8, R23 // SBCS R12, R8, R8
SUBU R26, R8 // ... SUBU R24, R8 // ...
SGTU R12, R8, R26 // ... SGTU R12, R8, R24 // ...
SUBU R12, R8 // ... SUBU R12, R8 // ...
ADDU R23, R26 // ... ADDU R23, R24 // ...
MOVW R5, 0(R4) MOVW R5, 0(R4)
MOVW R6, 4(R4) MOVW R6, 4(R4)
MOVW R7, 8(R4) MOVW R7, 8(R4)
@ -151,7 +151,7 @@ loop4cont:
SUBU $1, R1 SUBU $1, R1
BNE R1, loop4cont BNE R1, loop4cont
loop4done: loop4done:
MOVW R26, c+36(FP) MOVW R24, c+36(FP)
RET RET
// func lshVU(z, x []Word, s uint) (c Word) // func lshVU(z, x []Word, s uint) (c Word)
@ -316,9 +316,9 @@ loop1cont:
MULU R1, R7 MULU R1, R7
MOVW LO, R8 MOVW LO, R8
MOVW HI, R9 MOVW HI, R9
ADDU R2, R8, R7 // ADDS R2, R8, R7 (cr=R26) ADDU R2, R8, R7 // ADDS R2, R8, R7 (cr=R24)
SGTU R2, R7, R26 // ... SGTU R2, R7, R24 // ...
ADDU R26, R9, R2 // ADC $0, R9, R2 ADDU R24, R9, R2 // ADC $0, R9, R2
MOVW R7, 0(R5) MOVW R7, 0(R5)
ADDU $4, R4 ADDU $4, R4
ADDU $4, R5 ADDU $4, R5
@ -337,27 +337,27 @@ loop4cont:
MULU R1, R6 MULU R1, R6
MOVW LO, R10 MOVW LO, R10
MOVW HI, R11 MOVW HI, R11
ADDU R2, R10, R6 // ADDS R2, R10, R6 (cr=R26) ADDU R2, R10, R6 // ADDS R2, R10, R6 (cr=R24)
SGTU R2, R6, R26 // ... SGTU R2, R6, R24 // ...
ADDU R26, R11, R2 // ADC $0, R11, R2 ADDU R24, R11, R2 // ADC $0, R11, R2
MULU R1, R7 MULU R1, R7
MOVW LO, R10 MOVW LO, R10
MOVW HI, R11 MOVW HI, R11
ADDU R2, R10, R7 // ADDS R2, R10, R7 (cr=R26) ADDU R2, R10, R7 // ADDS R2, R10, R7 (cr=R24)
SGTU R2, R7, R26 // ... SGTU R2, R7, R24 // ...
ADDU R26, R11, R2 // ADC $0, R11, R2 ADDU R24, R11, R2 // ADC $0, R11, R2
MULU R1, R8 MULU R1, R8
MOVW LO, R10 MOVW LO, R10
MOVW HI, R11 MOVW HI, R11
ADDU R2, R10, R8 // ADDS R2, R10, R8 (cr=R26) ADDU R2, R10, R8 // ADDS R2, R10, R8 (cr=R24)
SGTU R2, R8, R26 // ... SGTU R2, R8, R24 // ...
ADDU R26, R11, R2 // ADC $0, R11, R2 ADDU R24, R11, R2 // ADC $0, R11, R2
MULU R1, R9 MULU R1, R9
MOVW LO, R10 MOVW LO, R10
MOVW HI, R11 MOVW HI, R11
ADDU R2, R10, R9 // ADDS R2, R10, R9 (cr=R26) ADDU R2, R10, R9 // ADDS R2, R10, R9 (cr=R24)
SGTU R2, R9, R26 // ... SGTU R2, R9, R24 // ...
ADDU R26, R11, R2 // ADC $0, R11, R2 ADDU R24, R11, R2 // ADC $0, R11, R2
MOVW R6, 0(R5) MOVW R6, 0(R5)
MOVW R7, 4(R5) MOVW R7, 4(R5)
MOVW R8, 8(R5) MOVW R8, 8(R5)
@ -391,12 +391,12 @@ loop1cont:
MULU R1, R9 MULU R1, R9
MOVW LO, R10 MOVW LO, R10
MOVW HI, R11 MOVW HI, R11
ADDU R8, R10 // ADDS R8, R10, R10 (cr=R26) ADDU R8, R10 // ADDS R8, R10, R10 (cr=R24)
SGTU R8, R10, R26 // ... SGTU R8, R10, R24 // ...
ADDU R26, R11 // ADC $0, R11, R11 ADDU R24, R11 // ADC $0, R11, R11
ADDU R2, R10, R9 // ADDS R2, R10, R9 (cr=R26) ADDU R2, R10, R9 // ADDS R2, R10, R9 (cr=R24)
SGTU R2, R9, R26 // ... SGTU R2, R9, R24 // ...
ADDU R26, R11, R2 // ADC $0, R11, R2 ADDU R24, R11, R2 // ADC $0, R11, R2
MOVW R9, 0(R6) MOVW R9, 0(R6)
ADDU $4, R4 ADDU $4, R4
ADDU $4, R5 ADDU $4, R5
@ -420,39 +420,39 @@ loop4cont:
MULU R1, R11 MULU R1, R11
MOVW LO, R15 MOVW LO, R15
MOVW HI, R16 MOVW HI, R16
ADDU R7, R15 // ADDS R7, R15, R15 (cr=R26) ADDU R7, R15 // ADDS R7, R15, R15 (cr=R24)
SGTU R7, R15, R26 // ... SGTU R7, R15, R24 // ...
ADDU R26, R16 // ADC $0, R16, R16 ADDU R24, R16 // ADC $0, R16, R16
ADDU R2, R15, R11 // ADDS R2, R15, R11 (cr=R26) ADDU R2, R15, R11 // ADDS R2, R15, R11 (cr=R24)
SGTU R2, R11, R26 // ... SGTU R2, R11, R24 // ...
ADDU R26, R16, R2 // ADC $0, R16, R2 ADDU R24, R16, R2 // ADC $0, R16, R2
MULU R1, R12 MULU R1, R12
MOVW LO, R15 MOVW LO, R15
MOVW HI, R16 MOVW HI, R16
ADDU R8, R15 // ADDS R8, R15, R15 (cr=R26) ADDU R8, R15 // ADDS R8, R15, R15 (cr=R24)
SGTU R8, R15, R26 // ... SGTU R8, R15, R24 // ...
ADDU R26, R16 // ADC $0, R16, R16 ADDU R24, R16 // ADC $0, R16, R16
ADDU R2, R15, R12 // ADDS R2, R15, R12 (cr=R26) ADDU R2, R15, R12 // ADDS R2, R15, R12 (cr=R24)
SGTU R2, R12, R26 // ... SGTU R2, R12, R24 // ...
ADDU R26, R16, R2 // ADC $0, R16, R2 ADDU R24, R16, R2 // ADC $0, R16, R2
MULU R1, R13 MULU R1, R13
MOVW LO, R15 MOVW LO, R15
MOVW HI, R16 MOVW HI, R16
ADDU R9, R15 // ADDS R9, R15, R15 (cr=R26) ADDU R9, R15 // ADDS R9, R15, R15 (cr=R24)
SGTU R9, R15, R26 // ... SGTU R9, R15, R24 // ...
ADDU R26, R16 // ADC $0, R16, R16 ADDU R24, R16 // ADC $0, R16, R16
ADDU R2, R15, R13 // ADDS R2, R15, R13 (cr=R26) ADDU R2, R15, R13 // ADDS R2, R15, R13 (cr=R24)
SGTU R2, R13, R26 // ... SGTU R2, R13, R24 // ...
ADDU R26, R16, R2 // ADC $0, R16, R2 ADDU R24, R16, R2 // ADC $0, R16, R2
MULU R1, R14 MULU R1, R14
MOVW LO, R15 MOVW LO, R15
MOVW HI, R16 MOVW HI, R16
ADDU R10, R15 // ADDS R10, R15, R15 (cr=R26) ADDU R10, R15 // ADDS R10, R15, R15 (cr=R24)
SGTU R10, R15, R26 // ... SGTU R10, R15, R24 // ...
ADDU R26, R16 // ADC $0, R16, R16 ADDU R24, R16 // ADC $0, R16, R16
ADDU R2, R15, R14 // ADDS R2, R15, R14 (cr=R26) ADDU R2, R15, R14 // ADDS R2, R15, R14 (cr=R24)
SGTU R2, R14, R26 // ... SGTU R2, R14, R24 // ...
ADDU R26, R16, R2 // ADC $0, R16, R2 ADDU R24, R16, R2 // ADC $0, R16, R2
MOVW R11, 0(R6) MOVW R11, 0(R6)
MOVW R12, 4(R6) MOVW R12, 4(R6)
MOVW R13, 8(R6) MOVW R13, 8(R6)

8
src/math/big/internal/asmgen/mips.go
View File

@ -14,19 +14,19 @@ var ArchMIPS = &Arch{
regs: []string{ regs: []string{
// R0 is 0 // R0 is 0
// R23 is the assembler/linker temporary (which we use too). // R23 is the assembler/linker temporary (which we use too).
// R26 and R27 are our virtual carry flags. // R24 and R25 are our virtual carry flags.
// R28 is SB. // R28 is SB.
// R29 is SP. // R29 is SP.
// R30 is g. // R30 is g.
// R31 is LR. // R31 is LR.
"R1", "R2", "R3", "R4", "R5", "R6", "R7", "R8", "R9", "R1", "R2", "R3", "R4", "R5", "R6", "R7", "R8", "R9",
"R10", "R11", "R12", "R13", "R14", "R15", "R16", "R17", "R18", "R19", "R10", "R11", "R12", "R13", "R14", "R15", "R16", "R17", "R18", "R19",
"R20", "R21", "R22", "R24", "R25", "R26", "R27", "R20", "R21", "R22", "R24", "R25",
}, },
reg0: "R0", reg0: "R0",
regTmp: "R23", regTmp: "R23",
regCarry: "R26", regCarry: "R24",
regAltCarry: "R27", regAltCarry: "R25",
mov: "MOVW", mov: "MOVW",
add: "ADDU", add: "ADDU",

8
src/math/big/internal/asmgen/mips64.go
View File

@ -14,19 +14,19 @@ var ArchMIPS64x = &Arch{
regs: []string{ regs: []string{
// R0 is 0 // R0 is 0
// R23 is the assembler/linker temporary (which we use too). // R23 is the assembler/linker temporary (which we use too).
// R26 and R27 are our virtual carry flags. // R24 and R25 are our virtual carry flags.
// R28 is SB. // R28 is SB.
// R29 is SP. // R29 is SP.
// R30 is g. // R30 is g.
// R31 is LR. // R31 is LR.
"R1", "R2", "R3", "R4", "R5", "R6", "R7", "R8", "R9", "R1", "R2", "R3", "R4", "R5", "R6", "R7", "R8", "R9",
"R10", "R11", "R12", "R13", "R14", "R15", "R16", "R17", "R18", "R19", "R10", "R11", "R12", "R13", "R14", "R15", "R16", "R17", "R18", "R19",
"R20", "R21", "R22", "R24", "R25", "R26", "R27", "R20", "R21", "R22", "R24", "R25",
}, },
reg0: "R0", reg0: "R0",
regTmp: "R23", regTmp: "R23",
regCarry: "R26", regCarry: "R24",
regAltCarry: "R27", regAltCarry: "R25",
mov: "MOVV", mov: "MOVV",
add: "ADDVU", add: "ADDVU",