5
0
mirror of https://github.com/cwinfo/matterbridge.git synced 2024-11-29 12:51:36 +00:00
matterbridge/vendor/github.com/klauspost/compress/huff0/decompress_amd64.s

866 lines
16 KiB
ArmAsm
Raw Normal View History

2022-06-24 22:36:16 +00:00
// Code generated by command: go run gen.go -out ../decompress_amd64.s -pkg=huff0. DO NOT EDIT.
//go:build amd64 && !appengine && !noasm && gc
// +build amd64,!appengine,!noasm,gc
// func decompress4x_main_loop_amd64(ctx *decompress4xContext)
TEXT ·decompress4x_main_loop_amd64(SB), $8-8
XORQ DX, DX
// Preload values
MOVQ ctx+0(FP), AX
MOVBQZX 32(AX), SI
MOVQ 40(AX), DI
MOVQ DI, BX
MOVQ 72(AX), CX
MOVQ CX, (SP)
MOVQ 48(AX), R8
MOVQ 56(AX), R9
MOVQ (AX), R10
MOVQ 8(AX), R11
MOVQ 16(AX), R12
MOVQ 24(AX), R13
// Main loop
2022-04-11 22:30:21 +00:00
main_loop:
2022-06-24 22:36:16 +00:00
MOVQ BX, DI
CMPQ DI, (SP)
SETGE DL
// br0.fillFast32()
MOVQ 32(R10), R14
MOVBQZX 40(R10), R15
CMPQ R15, $0x20
JBE skip_fill0
MOVQ 24(R10), AX
SUBQ $0x20, R15
SUBQ $0x04, AX
MOVQ (R10), BP
2022-04-11 22:30:21 +00:00
// b.value |= uint64(low) << (b.bitsRead & 63)
2022-06-24 22:36:16 +00:00
MOVL (AX)(BP*1), BP
MOVQ R15, CX
SHLQ CL, BP
MOVQ AX, 24(R10)
ORQ BP, R14
2022-04-11 22:30:21 +00:00
// exhausted = exhausted || (br0.off < 4)
2022-06-24 22:36:16 +00:00
CMPQ AX, $0x04
SETLT AL
ORB AL, DL
2022-04-11 22:30:21 +00:00
skip_fill0:
// val0 := br0.peekTopBits(peekBits)
2022-06-24 22:36:16 +00:00
MOVQ R14, BP
MOVQ SI, CX
SHRQ CL, BP
2022-04-11 22:30:21 +00:00
// v0 := table[val0&mask]
2022-06-24 22:36:16 +00:00
MOVW (R9)(BP*2), CX
2022-04-11 22:30:21 +00:00
2022-06-24 22:36:16 +00:00
// br0.advance(uint8(v0.entry)
MOVB CH, AL
SHLQ CL, R14
ADDB CL, R15
2022-04-11 22:30:21 +00:00
// val1 := br0.peekTopBits(peekBits)
2022-06-24 22:36:16 +00:00
MOVQ SI, CX
MOVQ R14, BP
SHRQ CL, BP
2022-04-11 22:30:21 +00:00
// v1 := table[val1&mask]
2022-06-24 22:36:16 +00:00
MOVW (R9)(BP*2), CX
2022-04-11 22:30:21 +00:00
// br0.advance(uint8(v1.entry))
2022-06-24 22:36:16 +00:00
MOVB CH, AH
SHLQ CL, R14
ADDB CL, R15
2022-04-11 22:30:21 +00:00
// these two writes get coalesced
2022-06-24 22:36:16 +00:00
// out[id * dstEvery + 0] = uint8(v0.entry >> 8)
// out[id * dstEvery + 1] = uint8(v1.entry >> 8)
MOVW AX, (DI)
2022-04-11 22:30:21 +00:00
// update the bitrader reader structure
2022-06-24 22:36:16 +00:00
MOVQ R14, 32(R10)
MOVB R15, 40(R10)
ADDQ R8, DI
// br1.fillFast32()
MOVQ 32(R11), R14
MOVBQZX 40(R11), R15
CMPQ R15, $0x20
JBE skip_fill1
MOVQ 24(R11), AX
SUBQ $0x20, R15
SUBQ $0x04, AX
MOVQ (R11), BP
2022-04-11 22:30:21 +00:00
// b.value |= uint64(low) << (b.bitsRead & 63)
2022-06-24 22:36:16 +00:00
MOVL (AX)(BP*1), BP
MOVQ R15, CX
SHLQ CL, BP
MOVQ AX, 24(R11)
ORQ BP, R14
2022-04-11 22:30:21 +00:00
// exhausted = exhausted || (br1.off < 4)
2022-06-24 22:36:16 +00:00
CMPQ AX, $0x04
SETLT AL
ORB AL, DL
2022-04-11 22:30:21 +00:00
skip_fill1:
// val0 := br1.peekTopBits(peekBits)
2022-06-24 22:36:16 +00:00
MOVQ R14, BP
MOVQ SI, CX
SHRQ CL, BP
2022-04-11 22:30:21 +00:00
// v0 := table[val0&mask]
2022-06-24 22:36:16 +00:00
MOVW (R9)(BP*2), CX
2022-04-11 22:30:21 +00:00
2022-06-24 22:36:16 +00:00
// br1.advance(uint8(v0.entry)
MOVB CH, AL
SHLQ CL, R14
ADDB CL, R15
2022-04-11 22:30:21 +00:00
// val1 := br1.peekTopBits(peekBits)
2022-06-24 22:36:16 +00:00
MOVQ SI, CX
MOVQ R14, BP
SHRQ CL, BP
2022-04-11 22:30:21 +00:00
// v1 := table[val1&mask]
2022-06-24 22:36:16 +00:00
MOVW (R9)(BP*2), CX
2022-04-11 22:30:21 +00:00
// br1.advance(uint8(v1.entry))
2022-06-24 22:36:16 +00:00
MOVB CH, AH
SHLQ CL, R14
ADDB CL, R15
2022-04-11 22:30:21 +00:00
// these two writes get coalesced
2022-06-24 22:36:16 +00:00
// out[id * dstEvery + 0] = uint8(v0.entry >> 8)
// out[id * dstEvery + 1] = uint8(v1.entry >> 8)
MOVW AX, (DI)
2022-04-11 22:30:21 +00:00
// update the bitrader reader structure
2022-06-24 22:36:16 +00:00
MOVQ R14, 32(R11)
MOVB R15, 40(R11)
ADDQ R8, DI
// br2.fillFast32()
MOVQ 32(R12), R14
MOVBQZX 40(R12), R15
CMPQ R15, $0x20
JBE skip_fill2
MOVQ 24(R12), AX
SUBQ $0x20, R15
SUBQ $0x04, AX
MOVQ (R12), BP
2022-04-11 22:30:21 +00:00
// b.value |= uint64(low) << (b.bitsRead & 63)
2022-06-24 22:36:16 +00:00
MOVL (AX)(BP*1), BP
MOVQ R15, CX
SHLQ CL, BP
MOVQ AX, 24(R12)
ORQ BP, R14
2022-04-11 22:30:21 +00:00
// exhausted = exhausted || (br2.off < 4)
2022-06-24 22:36:16 +00:00
CMPQ AX, $0x04
SETLT AL
ORB AL, DL
2022-04-11 22:30:21 +00:00
skip_fill2:
// val0 := br2.peekTopBits(peekBits)
2022-06-24 22:36:16 +00:00
MOVQ R14, BP
MOVQ SI, CX
SHRQ CL, BP
2022-04-11 22:30:21 +00:00
// v0 := table[val0&mask]
2022-06-24 22:36:16 +00:00
MOVW (R9)(BP*2), CX
2022-04-11 22:30:21 +00:00
2022-06-24 22:36:16 +00:00
// br2.advance(uint8(v0.entry)
MOVB CH, AL
SHLQ CL, R14
ADDB CL, R15
2022-04-11 22:30:21 +00:00
// val1 := br2.peekTopBits(peekBits)
2022-06-24 22:36:16 +00:00
MOVQ SI, CX
MOVQ R14, BP
SHRQ CL, BP
2022-04-11 22:30:21 +00:00
// v1 := table[val1&mask]
2022-06-24 22:36:16 +00:00
MOVW (R9)(BP*2), CX
2022-04-11 22:30:21 +00:00
// br2.advance(uint8(v1.entry))
2022-06-24 22:36:16 +00:00
MOVB CH, AH
SHLQ CL, R14
ADDB CL, R15
2022-04-11 22:30:21 +00:00
// these two writes get coalesced
2022-06-24 22:36:16 +00:00
// out[id * dstEvery + 0] = uint8(v0.entry >> 8)
// out[id * dstEvery + 1] = uint8(v1.entry >> 8)
MOVW AX, (DI)
2022-04-11 22:30:21 +00:00
// update the bitrader reader structure
2022-06-24 22:36:16 +00:00
MOVQ R14, 32(R12)
MOVB R15, 40(R12)
ADDQ R8, DI
// br3.fillFast32()
MOVQ 32(R13), R14
MOVBQZX 40(R13), R15
CMPQ R15, $0x20
JBE skip_fill3
MOVQ 24(R13), AX
SUBQ $0x20, R15
SUBQ $0x04, AX
MOVQ (R13), BP
2022-04-11 22:30:21 +00:00
// b.value |= uint64(low) << (b.bitsRead & 63)
2022-06-24 22:36:16 +00:00
MOVL (AX)(BP*1), BP
MOVQ R15, CX
SHLQ CL, BP
MOVQ AX, 24(R13)
ORQ BP, R14
2022-04-11 22:30:21 +00:00
// exhausted = exhausted || (br3.off < 4)
2022-06-24 22:36:16 +00:00
CMPQ AX, $0x04
SETLT AL
ORB AL, DL
2022-04-11 22:30:21 +00:00
skip_fill3:
// val0 := br3.peekTopBits(peekBits)
2022-06-24 22:36:16 +00:00
MOVQ R14, BP
MOVQ SI, CX
SHRQ CL, BP
2022-04-11 22:30:21 +00:00
2022-06-24 22:36:16 +00:00
// v0 := table[val0&mask]
MOVW (R9)(BP*2), CX
2022-04-11 22:30:21 +00:00
2022-06-24 22:36:16 +00:00
// br3.advance(uint8(v0.entry)
MOVB CH, AL
SHLQ CL, R14
ADDB CL, R15
2022-04-11 22:30:21 +00:00
2022-06-24 22:36:16 +00:00
// val1 := br3.peekTopBits(peekBits)
MOVQ SI, CX
MOVQ R14, BP
SHRQ CL, BP
2022-04-11 22:30:21 +00:00
2022-06-24 22:36:16 +00:00
// v1 := table[val1&mask]
MOVW (R9)(BP*2), CX
2022-04-11 22:30:21 +00:00
2022-06-24 22:36:16 +00:00
// br3.advance(uint8(v1.entry))
MOVB CH, AH
SHLQ CL, R14
ADDB CL, R15
2022-04-11 22:30:21 +00:00
2022-06-24 22:36:16 +00:00
// these two writes get coalesced
// out[id * dstEvery + 0] = uint8(v0.entry >> 8)
// out[id * dstEvery + 1] = uint8(v1.entry >> 8)
MOVW AX, (DI)
2022-04-11 22:30:21 +00:00
2022-06-24 22:36:16 +00:00
// update the bitrader reader structure
MOVQ R14, 32(R13)
MOVB R15, 40(R13)
ADDQ $0x02, BX
TESTB DL, DL
JZ main_loop
MOVQ ctx+0(FP), AX
MOVQ 40(AX), CX
MOVQ BX, DX
SUBQ CX, DX
SHLQ $0x02, DX
MOVQ DX, 64(AX)
RET
2022-04-11 22:30:21 +00:00
2022-06-24 22:36:16 +00:00
// func decompress4x_8b_main_loop_amd64(ctx *decompress4xContext)
TEXT ·decompress4x_8b_main_loop_amd64(SB), $16-8
XORQ DX, DX
// Preload values
MOVQ ctx+0(FP), CX
MOVBQZX 32(CX), BX
MOVQ 40(CX), SI
MOVQ SI, (SP)
MOVQ 72(CX), DX
MOVQ DX, 8(SP)
MOVQ 48(CX), DI
MOVQ 56(CX), R8
MOVQ (CX), R9
MOVQ 8(CX), R10
MOVQ 16(CX), R11
MOVQ 24(CX), R12
// Main loop
main_loop:
MOVQ (SP), SI
CMPQ SI, 8(SP)
SETGE DL
// br1000.fillFast32()
MOVQ 32(R9), R13
MOVBQZX 40(R9), R14
CMPQ R14, $0x20
JBE skip_fill1000
MOVQ 24(R9), R15
SUBQ $0x20, R14
SUBQ $0x04, R15
MOVQ (R9), BP
2022-04-11 22:30:21 +00:00
2022-06-24 22:36:16 +00:00
// b.value |= uint64(low) << (b.bitsRead & 63)
MOVL (R15)(BP*1), BP
MOVQ R14, CX
SHLQ CL, BP
MOVQ R15, 24(R9)
ORQ BP, R13
// exhausted = exhausted || (br1000.off < 4)
CMPQ R15, $0x04
SETLT AL
ORB AL, DL
skip_fill1000:
// val0 := br0.peekTopBits(peekBits)
MOVQ R13, R15
MOVQ BX, CX
SHRQ CL, R15
2022-04-11 22:30:21 +00:00
2022-06-24 22:36:16 +00:00
// v0 := table[val0&mask]
MOVW (R8)(R15*2), CX
2022-04-11 22:30:21 +00:00
2022-06-24 22:36:16 +00:00
// br0.advance(uint8(v0.entry)
MOVB CH, AL
SHLQ CL, R13
ADDB CL, R14
2022-04-11 22:30:21 +00:00
2022-06-24 22:36:16 +00:00
// val1 := br0.peekTopBits(peekBits)
MOVQ R13, R15
MOVQ BX, CX
SHRQ CL, R15
// v1 := table[val0&mask]
MOVW (R8)(R15*2), CX
// br0.advance(uint8(v1.entry)
MOVB CH, AH
SHLQ CL, R13
ADDB CL, R14
BSWAPL AX
// val2 := br0.peekTopBits(peekBits)
MOVQ R13, R15
MOVQ BX, CX
SHRQ CL, R15
// v2 := table[val0&mask]
MOVW (R8)(R15*2), CX
// br0.advance(uint8(v2.entry)
MOVB CH, AH
SHLQ CL, R13
ADDB CL, R14
// val3 := br0.peekTopBits(peekBits)
MOVQ R13, R15
MOVQ BX, CX
SHRQ CL, R15
// v3 := table[val0&mask]
MOVW (R8)(R15*2), CX
// br0.advance(uint8(v3.entry)
MOVB CH, AL
SHLQ CL, R13
ADDB CL, R14
BSWAPL AX
// these four writes get coalesced
// out[id * dstEvery + 0] = uint8(v0.entry >> 8)
// out[id * dstEvery + 1] = uint8(v1.entry >> 8)
// out[id * dstEvery + 3] = uint8(v2.entry >> 8)
// out[id * dstEvery + 4] = uint8(v3.entry >> 8)
MOVL AX, (SI)
// update the bitreader reader structure
MOVQ R13, 32(R9)
MOVB R14, 40(R9)
ADDQ DI, SI
// br1001.fillFast32()
MOVQ 32(R10), R13
MOVBQZX 40(R10), R14
CMPQ R14, $0x20
JBE skip_fill1001
MOVQ 24(R10), R15
SUBQ $0x20, R14
SUBQ $0x04, R15
MOVQ (R10), BP
2022-04-11 22:30:21 +00:00
2022-06-24 22:36:16 +00:00
// b.value |= uint64(low) << (b.bitsRead & 63)
MOVL (R15)(BP*1), BP
MOVQ R14, CX
SHLQ CL, BP
MOVQ R15, 24(R10)
ORQ BP, R13
// exhausted = exhausted || (br1001.off < 4)
CMPQ R15, $0x04
SETLT AL
ORB AL, DL
skip_fill1001:
// val0 := br1.peekTopBits(peekBits)
MOVQ R13, R15
MOVQ BX, CX
SHRQ CL, R15
2022-04-11 22:30:21 +00:00
2022-06-24 22:36:16 +00:00
// v0 := table[val0&mask]
MOVW (R8)(R15*2), CX
2022-04-11 22:30:21 +00:00
2022-06-24 22:36:16 +00:00
// br1.advance(uint8(v0.entry)
MOVB CH, AL
SHLQ CL, R13
ADDB CL, R14
2022-04-11 22:30:21 +00:00
2022-06-24 22:36:16 +00:00
// val1 := br1.peekTopBits(peekBits)
MOVQ R13, R15
MOVQ BX, CX
SHRQ CL, R15
// v1 := table[val0&mask]
MOVW (R8)(R15*2), CX
// br1.advance(uint8(v1.entry)
MOVB CH, AH
SHLQ CL, R13
ADDB CL, R14
BSWAPL AX
// val2 := br1.peekTopBits(peekBits)
MOVQ R13, R15
MOVQ BX, CX
SHRQ CL, R15
// v2 := table[val0&mask]
MOVW (R8)(R15*2), CX
// br1.advance(uint8(v2.entry)
MOVB CH, AH
SHLQ CL, R13
ADDB CL, R14
// val3 := br1.peekTopBits(peekBits)
MOVQ R13, R15
MOVQ BX, CX
SHRQ CL, R15
// v3 := table[val0&mask]
MOVW (R8)(R15*2), CX
// br1.advance(uint8(v3.entry)
MOVB CH, AL
SHLQ CL, R13
ADDB CL, R14
BSWAPL AX
// these four writes get coalesced
// out[id * dstEvery + 0] = uint8(v0.entry >> 8)
// out[id * dstEvery + 1] = uint8(v1.entry >> 8)
// out[id * dstEvery + 3] = uint8(v2.entry >> 8)
// out[id * dstEvery + 4] = uint8(v3.entry >> 8)
MOVL AX, (SI)
// update the bitreader reader structure
MOVQ R13, 32(R10)
MOVB R14, 40(R10)
ADDQ DI, SI
// br1002.fillFast32()
MOVQ 32(R11), R13
MOVBQZX 40(R11), R14
CMPQ R14, $0x20
JBE skip_fill1002
MOVQ 24(R11), R15
SUBQ $0x20, R14
SUBQ $0x04, R15
MOVQ (R11), BP
2022-04-11 22:30:21 +00:00
2022-06-24 22:36:16 +00:00
// b.value |= uint64(low) << (b.bitsRead & 63)
MOVL (R15)(BP*1), BP
MOVQ R14, CX
SHLQ CL, BP
MOVQ R15, 24(R11)
ORQ BP, R13
// exhausted = exhausted || (br1002.off < 4)
CMPQ R15, $0x04
SETLT AL
ORB AL, DL
skip_fill1002:
// val0 := br2.peekTopBits(peekBits)
MOVQ R13, R15
MOVQ BX, CX
SHRQ CL, R15
2022-04-11 22:30:21 +00:00
2022-06-24 22:36:16 +00:00
// v0 := table[val0&mask]
MOVW (R8)(R15*2), CX
2022-04-11 22:30:21 +00:00
2022-06-24 22:36:16 +00:00
// br2.advance(uint8(v0.entry)
MOVB CH, AL
SHLQ CL, R13
ADDB CL, R14
2022-04-11 22:30:21 +00:00
2022-06-24 22:36:16 +00:00
// val1 := br2.peekTopBits(peekBits)
MOVQ R13, R15
MOVQ BX, CX
SHRQ CL, R15
// v1 := table[val0&mask]
MOVW (R8)(R15*2), CX
// br2.advance(uint8(v1.entry)
MOVB CH, AH
SHLQ CL, R13
ADDB CL, R14
BSWAPL AX
// val2 := br2.peekTopBits(peekBits)
MOVQ R13, R15
MOVQ BX, CX
SHRQ CL, R15
// v2 := table[val0&mask]
MOVW (R8)(R15*2), CX
// br2.advance(uint8(v2.entry)
MOVB CH, AH
SHLQ CL, R13
ADDB CL, R14
// val3 := br2.peekTopBits(peekBits)
MOVQ R13, R15
MOVQ BX, CX
SHRQ CL, R15
// v3 := table[val0&mask]
MOVW (R8)(R15*2), CX
// br2.advance(uint8(v3.entry)
MOVB CH, AL
SHLQ CL, R13
ADDB CL, R14
BSWAPL AX
// these four writes get coalesced
// out[id * dstEvery + 0] = uint8(v0.entry >> 8)
// out[id * dstEvery + 1] = uint8(v1.entry >> 8)
// out[id * dstEvery + 3] = uint8(v2.entry >> 8)
// out[id * dstEvery + 4] = uint8(v3.entry >> 8)
MOVL AX, (SI)
// update the bitreader reader structure
MOVQ R13, 32(R11)
MOVB R14, 40(R11)
ADDQ DI, SI
// br1003.fillFast32()
MOVQ 32(R12), R13
MOVBQZX 40(R12), R14
CMPQ R14, $0x20
JBE skip_fill1003
MOVQ 24(R12), R15
SUBQ $0x20, R14
SUBQ $0x04, R15
MOVQ (R12), BP
2022-04-11 22:30:21 +00:00
2022-06-24 22:36:16 +00:00
// b.value |= uint64(low) << (b.bitsRead & 63)
MOVL (R15)(BP*1), BP
MOVQ R14, CX
SHLQ CL, BP
MOVQ R15, 24(R12)
ORQ BP, R13
// exhausted = exhausted || (br1003.off < 4)
CMPQ R15, $0x04
SETLT AL
ORB AL, DL
skip_fill1003:
// val0 := br3.peekTopBits(peekBits)
MOVQ R13, R15
MOVQ BX, CX
SHRQ CL, R15
2022-04-11 22:30:21 +00:00
2022-06-24 22:36:16 +00:00
// v0 := table[val0&mask]
MOVW (R8)(R15*2), CX
2022-04-11 22:30:21 +00:00
2022-06-24 22:36:16 +00:00
// br3.advance(uint8(v0.entry)
MOVB CH, AL
SHLQ CL, R13
ADDB CL, R14
2022-04-11 22:30:21 +00:00
2022-06-24 22:36:16 +00:00
// val1 := br3.peekTopBits(peekBits)
MOVQ R13, R15
MOVQ BX, CX
SHRQ CL, R15
// v1 := table[val0&mask]
MOVW (R8)(R15*2), CX
// br3.advance(uint8(v1.entry)
MOVB CH, AH
SHLQ CL, R13
ADDB CL, R14
BSWAPL AX
// val2 := br3.peekTopBits(peekBits)
MOVQ R13, R15
MOVQ BX, CX
SHRQ CL, R15
// v2 := table[val0&mask]
MOVW (R8)(R15*2), CX
// br3.advance(uint8(v2.entry)
MOVB CH, AH
SHLQ CL, R13
ADDB CL, R14
// val3 := br3.peekTopBits(peekBits)
MOVQ R13, R15
MOVQ BX, CX
SHRQ CL, R15
// v3 := table[val0&mask]
MOVW (R8)(R15*2), CX
// br3.advance(uint8(v3.entry)
MOVB CH, AL
SHLQ CL, R13
ADDB CL, R14
BSWAPL AX
// these four writes get coalesced
// out[id * dstEvery + 0] = uint8(v0.entry >> 8)
// out[id * dstEvery + 1] = uint8(v1.entry >> 8)
// out[id * dstEvery + 3] = uint8(v2.entry >> 8)
// out[id * dstEvery + 4] = uint8(v3.entry >> 8)
MOVL AX, (SI)
// update the bitreader reader structure
MOVQ R13, 32(R12)
MOVB R14, 40(R12)
ADDQ $0x04, (SP)
TESTB DL, DL
JZ main_loop
MOVQ ctx+0(FP), AX
MOVQ 40(AX), CX
MOVQ (SP), DX
SUBQ CX, DX
SHLQ $0x02, DX
MOVQ DX, 64(AX)
2022-04-11 22:30:21 +00:00
RET
2022-06-24 22:36:16 +00:00
// func decompress1x_main_loop_amd64(ctx *decompress1xContext)
TEXT ·decompress1x_main_loop_amd64(SB), $0-8
MOVQ ctx+0(FP), CX
MOVQ 16(CX), DX
MOVQ 24(CX), BX
CMPQ BX, $0x04
JB error_max_decoded_size_exeeded
LEAQ (DX)(BX*1), BX
MOVQ (CX), SI
MOVQ (SI), R8
MOVQ 24(SI), R9
MOVQ 32(SI), R10
MOVBQZX 40(SI), R11
MOVQ 32(CX), SI
MOVBQZX 8(CX), DI
JMP loop_condition
2022-04-11 22:30:21 +00:00
2022-06-24 22:36:16 +00:00
main_loop:
// Check if we have room for 4 bytes in the output buffer
LEAQ 4(DX), CX
CMPQ CX, BX
JGE error_max_decoded_size_exeeded
// Decode 4 values
CMPQ R11, $0x20
JL bitReader_fillFast_1_end
SUBQ $0x20, R11
SUBQ $0x04, R9
MOVL (R8)(R9*1), R12
MOVQ R11, CX
SHLQ CL, R12
ORQ R12, R10
bitReader_fillFast_1_end:
MOVQ DI, CX
MOVQ R10, R12
SHRQ CL, R12
MOVW (SI)(R12*2), CX
MOVB CH, AL
MOVBQZX CL, CX
ADDQ CX, R11
SHLQ CL, R10
MOVQ DI, CX
MOVQ R10, R12
SHRQ CL, R12
MOVW (SI)(R12*2), CX
MOVB CH, AH
MOVBQZX CL, CX
ADDQ CX, R11
SHLQ CL, R10
BSWAPL AX
CMPQ R11, $0x20
JL bitReader_fillFast_2_end
SUBQ $0x20, R11
SUBQ $0x04, R9
MOVL (R8)(R9*1), R12
MOVQ R11, CX
SHLQ CL, R12
ORQ R12, R10
bitReader_fillFast_2_end:
MOVQ DI, CX
MOVQ R10, R12
SHRQ CL, R12
MOVW (SI)(R12*2), CX
MOVB CH, AH
MOVBQZX CL, CX
ADDQ CX, R11
SHLQ CL, R10
MOVQ DI, CX
MOVQ R10, R12
SHRQ CL, R12
MOVW (SI)(R12*2), CX
MOVB CH, AL
MOVBQZX CL, CX
ADDQ CX, R11
SHLQ CL, R10
BSWAPL AX
// Store the decoded values
MOVL AX, (DX)
ADDQ $0x04, DX
loop_condition:
CMPQ R9, $0x08
JGE main_loop
// Update ctx structure
MOVQ ctx+0(FP), AX
MOVQ DX, CX
MOVQ 16(AX), DX
SUBQ DX, CX
MOVQ CX, 40(AX)
MOVQ (AX), AX
MOVQ R9, 24(AX)
MOVQ R10, 32(AX)
MOVB R11, 40(AX)
RET
2022-04-11 22:30:21 +00:00
2022-06-24 22:36:16 +00:00
// Report error
error_max_decoded_size_exeeded:
MOVQ ctx+0(FP), AX
MOVQ $-1, CX
MOVQ CX, 40(AX)
RET
// func decompress1x_main_loop_bmi2(ctx *decompress1xContext)
// Requires: BMI2
TEXT ·decompress1x_main_loop_bmi2(SB), $0-8
MOVQ ctx+0(FP), CX
MOVQ 16(CX), DX
MOVQ 24(CX), BX
CMPQ BX, $0x04
JB error_max_decoded_size_exeeded
LEAQ (DX)(BX*1), BX
MOVQ (CX), SI
MOVQ (SI), R8
MOVQ 24(SI), R9
MOVQ 32(SI), R10
MOVBQZX 40(SI), R11
MOVQ 32(CX), SI
MOVBQZX 8(CX), DI
JMP loop_condition
main_loop:
// Check if we have room for 4 bytes in the output buffer
LEAQ 4(DX), CX
CMPQ CX, BX
JGE error_max_decoded_size_exeeded
// Decode 4 values
CMPQ R11, $0x20
JL bitReader_fillFast_1_end
SUBQ $0x20, R11
SUBQ $0x04, R9
MOVL (R8)(R9*1), CX
SHLXQ R11, CX, CX
ORQ CX, R10
bitReader_fillFast_1_end:
SHRXQ DI, R10, CX
MOVW (SI)(CX*2), CX
MOVB CH, AL
MOVBQZX CL, CX
ADDQ CX, R11
SHLXQ CX, R10, R10
SHRXQ DI, R10, CX
MOVW (SI)(CX*2), CX
MOVB CH, AH
MOVBQZX CL, CX
ADDQ CX, R11
SHLXQ CX, R10, R10
BSWAPL AX
CMPQ R11, $0x20
JL bitReader_fillFast_2_end
SUBQ $0x20, R11
SUBQ $0x04, R9
MOVL (R8)(R9*1), CX
SHLXQ R11, CX, CX
ORQ CX, R10
bitReader_fillFast_2_end:
SHRXQ DI, R10, CX
MOVW (SI)(CX*2), CX
MOVB CH, AH
MOVBQZX CL, CX
ADDQ CX, R11
SHLXQ CX, R10, R10
SHRXQ DI, R10, CX
MOVW (SI)(CX*2), CX
MOVB CH, AL
MOVBQZX CL, CX
ADDQ CX, R11
SHLXQ CX, R10, R10
BSWAPL AX
// Store the decoded values
MOVL AX, (DX)
ADDQ $0x04, DX
loop_condition:
CMPQ R9, $0x08
JGE main_loop
// Update ctx structure
MOVQ ctx+0(FP), AX
MOVQ DX, CX
MOVQ 16(AX), DX
SUBQ DX, CX
MOVQ CX, 40(AX)
MOVQ (AX), AX
MOVQ R9, 24(AX)
MOVQ R10, 32(AX)
MOVB R11, 40(AX)
RET
// Report error
error_max_decoded_size_exeeded:
MOVQ ctx+0(FP), AX
MOVQ $-1, CX
MOVQ CX, 40(AX)
RET