From 1fb29f3a1b4e55f440446860a81e12a3cbbfad85 Mon Sep 17 00:00:00 2001
From: "Peter Helcmanovsky (Ped)" <ped@7gods.org>
Date: Sun, 18 Sep 2022 23:41:51 +0200
Subject: [PATCH] z80_unpacker: optimisation: -1B and -1T in decode_bit = 173B

---
 z80_unpacker/example/example.sna | Bin 49179 -> 49179 bytes
 z80_unpacker/unpack.asm          |   6 +++---
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/z80_unpacker/example/example.sna b/z80_unpacker/example/example.sna
index 515f5d68690562b02a92e501cbcea636569d53df..f9d8ac6a21f3660ab798d06a222eb4ffeaf3405a 100644
GIT binary patch
delta 149
zcmbQ;z&yKwd4u>u#`BXU7KZU{c_>}A`NZ4c^oM*3XSYnQS?J6o|Ll*vy!^Qj@dvpk
zPd>BINJs9Gg#gc4xksl34SrVM6q3FfAn~$-W!DwCM+Q7RXA>SlOb~elG$CQK)*>Y)
ol}D2oE|gP1Hi)U|;z_~(^%9(5gQQMN1L@7Zi=N9&F515m0NL(DH~;_u

delta 151
zcmbQ;z&yKwd4u>u#tV}r7KZU{eJEYE`NZ4c^oM*3XSYtSS?J6gFaPY1yuAFm50l@z
z9GrZ5p^=XKBMSkZv+|El3mW{ayeTAoGeF{H1<S50@{bI7c+Mt0f*2wC2xvsoWX(lN
pOsbD2FI*_6h++~`)5Vj5|LY|<!6r$amIl(ByB9r|nOw4eIRM06Mv?#k

diff --git a/z80_unpacker/unpack.asm b/z80_unpacker/unpack.asm
index 26e47b7..cb458e4 100644
--- a/z80_unpacker/unpack.asm
+++ b/z80_unpacker/unpack.asm
@@ -259,16 +259,16 @@ decode_bit:
     ; so 7*128 > 250 and thus edge case `ADD hl=(7*128+0),de=(-250)` => CF=1
 .bit_is_0_2:
  ; *** adjust probs[context_index]
-    ld      e,a                     ; preserve prob
     rra                             ; + (bit<<4) ; part of -prob_offset, needs another -16
     and     $FC                     ; clear/keep correct bits to get desired (prob>>4) + extras, CF=0
     rra
     rra
     rra                             ; A = (bit<<4) + (prob>>4), CF=(prob & 8)
     adc     a,-16                   ; A = (bit<<4) - 16 + ((prob + 8)>>4) ; -prob_offset = (bit<<4) - 16
-    sub     e                       ; A = (bit<<4) - 16 + ((prob + 8)>>4) - prob ; = ((prob + 8)>>4) - prob_offset - prob
-    neg                             ; A = prob_offset + prob - ((prob + 8)>>4)
+    ld      e,a
     pop     bc
+    ld      a,(bc)                  ; A = prob (cheaper + shorter to re-read again from memory)
+    sub     e                       ; A = 16 - (bit<<4) + prob - ((prob + 8)>>4) ; = prob_offset + prob - ((prob + 8)>>4)
     ld      (bc),a                  ; probs[context_index] = prob_offset + prob - ((prob + 8) >> 4);
     add     a,d                     ; restore CF = bit (D = bit ? $FF : $00 && A > 0)
     pop     de