z80count

This is a simple tool that parses Z80 assembler using regular expressions (I know!) and adds comments to the code with the cycles used by the instruction.

It needs testing and probably a proper Z80 parser, but it works for me and the Z80 assembler syntax I use.

Demo with vim

Installation

To install for production you can use pip:

pip install --user z80count

Or you can download the package from z80count's releases tab, unpack and run:

python setup.py install

To install for developmet run:

git clone https://github.com/reidrac/z80count.git
cd z80count
pip install -e ".[dev]"

Usage

You can use it with:

z80count.py  < file.asm > file_c.asm

Or inside vim you can:

:% !z80count.py -su

With -s the tool adds a subtotal to the comments and -u tries to update existing comments generated by the tool.

Example:

    push hl
    pop bc
    ld hl, $5800

    ld e, 7
.fade_out_all_loop0
    push hl
    push bc

    halt
.fade_out_all_loop1
    ld a, (hl)
    and 7
    jr z, no_fade_all_ink
    dec a
.no_fade_all_ink

    ld d, a

    ld a, (hl)
    and $38
    jr z, no_fade_all_paper
    sub 8
.no_fade_all_paper

    or d
    ld d, a

    ld a, (hl)
    and $c0
    or d

    ld (hl), a
    inc hl

    dec bc
    ld a, b
    or c
    jr nz, fade_out_all_loop1

    pop bc
    pop hl
    dec e
    jr nz, fade_out_all_loop0

Processed with z80count.py -s results in:

    push hl             ; [11 .. 11]
    pop bc              ; [10 .. 21]
    ld hl, $5800            ; [10 .. 31]

    ld e, 7             ; [7 .. 38]
.fade_out_all_loop0
    push hl             ; [11 .. 49]
    push bc             ; [11 .. 60]

    halt                ; [4 .. 64]
.fade_out_all_loop1
    ld a, (hl)          ; [7 .. 71]
    and 7               ; [7 .. 78]
    jr z, no_fade_all_ink       ; [12/7 .. 97/85]
    dec a               ; [4 .. 89]
.no_fade_all_ink

    ld d, a             ; [4 .. 93]

    ld a, (hl)          ; [7 .. 100]
    and $38             ; [7 .. 107]
    jr z, no_fade_all_paper     ; [12/7 .. 126/114]
    sub 8               ; [7 .. 121]
.no_fade_all_paper

    or d                ; [4 .. 125]
    ld d, a             ; [4 .. 129]

    ld a, (hl)          ; [7 .. 136]
    and $c0             ; [7 .. 143]
    or d                ; [4 .. 147]

    ld (hl), a          ; [7 .. 154]
    inc hl              ; [6 .. 160]

    dec bc              ; [6 .. 166]
    ld a, b             ; [4 .. 170]
    or c                ; [4 .. 174]
    jr nz, fade_out_all_loop1   ; [12/7 .. 193/181]

    pop bc              ; [10 .. 191]
    pop hl              ; [10 .. 201]
    dec e               ; [4 .. 205]
    jr nz, fade_out_all_loop0   ; [12/7 .. 224/212]

Comments show subtotals, and there are two types: - [A .. T0] - [B/A .. T1/T0]

Where A, B, T0 and T1 are: - A is the number of cycles of current instruction. In case of a conditional instruction, this is the value when the condition is not met. - B is the number of cycles of current instruction when the condition is met. - T0 is the subtotal when the conditional is not met. - T1 is the subtotal when the conditional is met.

Troubleshooting

Here be dragons!

Use -d flag if you think one instruction is not correctly parsed.

Feel free to open a PR if you find a bug!

Authors

Juan J. Martinez jjm@usebox.net
Alexis Roda (https://github.com/patxoca)