aboutsummaryrefslogtreecommitdiff
path: root/tools/rasm/decrunch/dzx0_fast.asm
blob: 55f4388aa1ed2d4fbdabf9309c3355d56fb42037 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
;
;  Speed-optimized ZX0 decompressor by spke (187 bytes)
;
;  ver.00 by spke (27/01-23/03/2021, 191 bytes)
;  ver.01 by spke (24/03/2021, 193(+2) bytes - fixed a bug in the initialization)
;  ver.01patch2 by uniabis (25/03/2021, 191(-2) bytes - fixed a bug with elias over 8bits)
;  ver.01patch9 by uniabis (10/09/2021, 187(-4) bytes - support for new v2 format)
;
;  Original ZX0 decompressors were written by Einar Saukas
;
;  This decompressor was written on the basis of "Standard" decompressor by
;  Einar Saukas and optimized for speed by spke. This decompressor is
;  about 5% faster than the "Turbo" decompressor, which is 128 bytes long.
;  It has about the same speed as the 412 bytes version of the "Mega" decompressor.
;
;  The decompressor uses AF, BC, DE, HL and IX and relies upon self-modified code.
;
;  The decompression is done in the standard way:
;
;  ld hl,FirstByteOfCompressedData
;  ld de,FirstByteOfMemoryForDecompressedData
;  call DecompressZX0
;
;  Of course, ZX0 compression algorithms are (c) 2021 Einar Saukas,
;  see https://github.com/einar-saukas/ZX0 for more information
;
;  Drop me an email if you have any comments/ideas/suggestions: zxintrospec@gmail.com
;
;  This software is provided 'as-is', without any express or implied
;  warranty.  In no event will the authors be held liable for any damages
;  arising from the use of this software.
;
;  Permission is granted to anyone to use this software for any purpose,
;  including commercial applications, and to alter it and redistribute it
;  freely, subject to the following restrictions:
;
;  1. The origin of this software must not be misrepresented; you must not
;     claim that you wrote the original software. If you use this software
;     in a product, an acknowledgment in the product documentation would be
;     appreciated but is not required.
;  2. Altered source versions must be plainly marked as such, and must not be
;     misrepresented as being the original software.
;  3. This notice may not be removed or altered from any source distribution.

macro DecompressZX0

        ld      ix, @CopyMatch1
        ld      bc, $ffff
        ld      (@PrevOffset+1), bc      ; default offset is -1
        inc     bc
        ld      a, $80
        jr      @RunOfLiterals           ; BC is assumed to contains 0 most of the time

@ShorterOffsets:
        ld      b, $ff                  ; the top byte of the offset is always $FF
        ld      c, (hl)
        inc     hl
        rr      c
        ld      (@PrevOffset+1), bc
        jr      nc, @LongerMatch

@CopyMatch2:                             ; the case of matches with len=2
        ld      bc, 2

        ; the faster match copying code
@CopyMatch1:
        push    hl                      ; preserve source

@PrevOffset:
        ld      hl, $ffff               ; restore offset (default offset is -1)
        add     hl, de                  ; HL = dest - offset
        ldir
        pop     hl                      ; restore source

        ; after a match you can have either
        ; 0 + <elias length> = run of literals, or
        ; 1 + <elias offset msb> + [7-bits of offset lsb + 1-bit of length] + <elias length> = another match
@AfterMatch1:
        add     a, a
        jr      nc, @RunOfLiterals

@UsualMatch:                             ; this is the case of usual match+offset
        add     a, a
        jr      nc, @LongerOffets
        jr      nz, @ShorterOffsets      ; NZ after NC == "confirmed C"
        
        ld      a, (hl)                 ; reload bits
        inc     hl
        rla

        jr      c, @ShorterOffsets

@LongerOffets:
        ld      c, $fe

        add     a, a                    ; inline read gamma
        rl      c
        add     a, a
        jr      nc, $-4

        call    z, @ReloadReadGamma

@ProcessOffset:

        inc     c
        ret     z                       ; end-of-data marker (only checked for longer offsets)
        rr      c
        ld      b, c
        ld      c, (hl)
        inc     hl
        rr      c
        ld      (@PrevOffset+1), bc

        ; lowest bit is the first bit of the gamma code for length
        jr      c, @CopyMatch2

@LongerMatch:
        ld      bc, 1

        add     a, a                    ; inline read gamma
        rl      c
        add     a, a
        jr      nc, $-4

        call    z,@ReloadReadGamma

@CopyMatch3:
        push    hl                      ; preserve source
        ld      hl, (@PrevOffset+1)      ; restore offset
        add     hl, de                  ; HL = dest - offset

        ; because BC>=3-1, we can do 2 x LDI safely
        ldi
        ldir
        inc     c
        ldi
        pop     hl                      ; restore source

        ; after a match you can have either
        ; 0 + <elias length> = run of literals, or
        ; 1 + <elias offset msb> + [7-bits of offset lsb + 1-bit of length] + <elias length> = another match
@AfterMatch3:
        add     a, a
        jr      c, @UsualMatch

@RunOfLiterals:
        inc     c
        add     a, a
        jr      nc, @LongerRun
        jr      nz, @CopyLiteral         ; NZ after NC == "confirmed C"
        
        ld      a, (hl)                 ; reload bits
        inc     hl
        rla

        jr      c, @CopyLiteral

@LongerRun:
        add     a, a                    ; inline read gamma
        rl      c
        add     a, a
        jr      nc, $-4

        jr      nz, @CopyLiterals
        
        ld      a, (hl)                 ; reload bits
        inc     hl
        rla

        call    nc, @ReadGammaAligned

@CopyLiterals:
        ldi

@CopyLiteral:
        ldir

        ; after a literal run you can have either
        ; 0 + <elias length> = match using a repeated offset, or
        ; 1 + <elias offset msb> + [7-bits of offset lsb + 1-bit of length] + <elias length> = another match
        add     a, a
        jr      c, @UsualMatch

@RepMatch:
        inc     c
        add     a, a
        jr      nc, @LongerRepMatch
        jr      nz, @CopyMatch1          ; NZ after NC == "confirmed C"

        ld      a, (hl)                 ; reload bits
        inc     hl
        rla

        jr      c, @CopyMatch1

@LongerRepMatch:
        add     a, a                    ; inline read gamma
        rl      c
        add     a, a
        jr      nc, $-4

        jp      nz, @CopyMatch1

        ; this is a crafty equivalent of CALL ReloadReadGamma : JP CopyMatch1
        push    ix

        ;  the subroutine for reading the remainder of the partly read Elias gamma code.
        ;  it has two entry points: ReloadReadGamma first refills the bit reservoir in A,
        ;  while ReadGammaAligned assumes that the bit reservoir has just been refilled.
@ReloadReadGamma:
        ld      a, (hl)                 ; reload bits
        inc     hl
        rla

        ret     c
@ReadGammaAligned:
        add     a, a
        rl      c
        add     a, a
        ret     c
        add     a, a
        rl      c
        add     a, a
@ReadingLongGamma:                       ; this loop does not need unrolling, as it does not get much use anyway
        ret     c
        add     a, a
        rl      c
        rl      b
        add     a, a
        jr      nz, @ReadingLongGamma

        ld      a, (hl)                 ; reload bits
        inc     hl
        rla
        jr      @ReadingLongGamma
mend