updated s/h arm asm code. Untested.

[picodrive.git] / pico / draw_arm.s

diff --git a/pico/draw_arm.s b/pico/draw_arm.s
index 5074a63..c7e0897 100644 (file)
--- a/pico/draw_arm.s
+++ b/pico/draw_arm.s
@@ -72,7 +72,7 @@
 .endif\r
     ldreqb  r4, [r1,#\offs]\r
     orrne   r4, r3, r4\r
-    andeq   r4, r4, #0x3f\r
+    andeq   r4, r4, #0xbf\r
     strb    r4, [r1,#\offs]\r
 .endm\r
 \r
@@ -163,14 +163,12 @@
 .endif\r
     beq     0f\r
     cmp     r4, #0xe\r
-    ldrgeb  r4, [r1,#\ofs]\r
-    orrlt   r4, r3, r4            @ normal\r
-\r
-    biceq   r4, r4, #0xc0         @ hilight\r
-    orreq   r4, r4, #0x80\r
-    orrgt   r4, r4, #0xc0         @ shadow\r
+    ldrgeb  r7, [r1,#\ofs]\r
+    orrlt   r7, r3, r4            @ normal\r
 \r
-    strb    r4, [r1,#\ofs]\r
+    bicge   r7, r7, #0xc0\r
+    orrge   r7, r7, r4, lsl #6\r
+    strb    r7, [r1,#\ofs]\r
 0:\r
 .endm\r
 \r
@@ -198,38 +196,41 @@
     TileDoShGenPixel 16,  7 @ #0x000f0000\r
 .endm\r
 \r
-.macro TileDoShGenPixel_noop shift ofs\r
+.macro TileDoShGenPixel_markop shift ofs\r
 .if \shift\r
-    and     r4, r12, r2, lsr #\shift\r
+    ands    r4, r12, r2, lsr #\shift\r
 .else\r
-    and     r4, r12, r2\r
+    ands    r4, r12, r2\r
 .endif\r
-    sub     r7, r4, #1\r
-    cmp     r7, #0xd\r
-    orrcc   r4, r3, r4           @ 0-0xc (was 1-0xd)\r
-    strccb  r4, [r1,#\ofs]\r
+    beq     0f\r
+    cmp     r4, #0xe\r
+    ldrgeb  r4, [r1,#\ofs]\r
+    orrlt   r4, r3, r4\r
+    orrge   r4, r3, 0x80\r
+    strb    r4, [r1,#\ofs]\r
+0:\r
 .endm\r
 \r
 .macro TileFlipSh_noop\r
-    TileDoShGenPixel_noop 16,  0 @ #0x000f0000\r
-    TileDoShGenPixel_noop 20,  1 @ #0x00f00000\r
-    TileDoShGenPixel_noop 24,  2 @ #0x0f000000\r
-    TileDoShGenPixel_noop 28,  3 @ #0xf0000000\r
-    TileDoShGenPixel_noop  0,  4 @ #0x0000000f\r
-    TileDoShGenPixel_noop  4,  5 @ #0x000000f0\r
-    TileDoShGenPixel_noop  8,  6 @ #0x00000f00\r
-    TileDoShGenPixel_noop 12,  7 @ #0x0000f000\r
+    TileDoShGenPixel_markop 16,  0 @ #0x000f0000\r
+    TileDoShGenPixel_markop 20,  1 @ #0x00f00000\r
+    TileDoShGenPixel_markop 24,  2 @ #0x0f000000\r
+    TileDoShGenPixel_markop 28,  3 @ #0xf0000000\r
+    TileDoShGenPixel_markop  0,  4 @ #0x0000000f\r
+    TileDoShGenPixel_markop  4,  5 @ #0x000000f0\r
+    TileDoShGenPixel_markop  8,  6 @ #0x00000f00\r
+    TileDoShGenPixel_markop 12,  7 @ #0x0000f000\r
 .endm\r
 \r
 .macro TileNormSh_noop\r
-    TileDoShGenPixel_noop 12,  0 @ #0x0000f000\r
-    TileDoShGenPixel_noop  8,  1 @ #0x00000f00\r
-    TileDoShGenPixel_noop  4,  2 @ #0x000000f0\r
-    TileDoShGenPixel_noop  0,  3 @ #0x0000000f\r
-    TileDoShGenPixel_noop 28,  4 @ #0xf0000000\r
-    TileDoShGenPixel_noop 24,  5 @ #0x0f000000\r
-    TileDoShGenPixel_noop 20,  6 @ #0x00f00000\r
-    TileDoShGenPixel_noop 16,  7 @ #0x000f0000\r
+    TileDoShGenPixel_markop 12,  0 @ #0x0000f000\r
+    TileDoShGenPixel_markop  8,  1 @ #0x00000f00\r
+    TileDoShGenPixel_markop  4,  2 @ #0x000000f0\r
+    TileDoShGenPixel_markop  0,  3 @ #0x0000000f\r
+    TileDoShGenPixel_markop 28,  4 @ #0xf0000000\r
+    TileDoShGenPixel_markop 24,  5 @ #0x0f000000\r
+    TileDoShGenPixel_markop 20,  6 @ #0x00f00000\r
+    TileDoShGenPixel_markop 16,  7 @ #0x000f0000\r
 .endm\r
 \r
 .macro TileDoShGenPixel_onlyop_lp shift ofs\r
@@ -239,14 +240,13 @@
     ands    r7, r12, r2\r
 .endif\r
     ldrneb  r4, [r1,#\ofs]\r
-    tstne   r4, #0x40\r
-    beq     0f\r
-\r
     cmp     r7, #0xe\r
-    biceq   r4, r4, #0xc0         @ hilight\r
-    orreq   r4, r4, #0x80\r
-    orrgt   r4, r4, #0xc0         @ shadow\r
-    strgeb  r4, [r1,#\ofs]\r
+    blt     0f\r
+\r
+    tst     r4, #0xc0\r
+    bicne   r4, r4, #0xc0\r
+    orrne   r4, r4, r7, lsl #6\r
+    strneb  r4, [r1,#\ofs]\r
 0:\r
 .endm\r
 \r
@@ -873,8 +873,8 @@ DrawTilesFromCache:
 .dtfc_shadow_blank:\r
     tst     r1, #1\r
     ldrneb  r4, [r1]\r
-    mov     r6, #0x3f\r
-    and     r4, r4, #0x3f\r
+    mov     r6, #0xbf\r
+    and     r4, r4, #0xbf\r
     strneb  r4, [r1], #1\r
     ldrh    r4, [r1]\r
     orr     r6, r6, r6, lsl #8\r
@@ -925,7 +925,7 @@ DrawTilesFromCache:
 \r
     add     r1, r11,#8\r
     mov     r3, #320/4/4\r
-    mov     r6, #0x3f\r
+    mov     r6, #0xbf\r
     orr     r6, r6, r6, lsl #8\r
     orr     r6, r6, r6, lsl #16\r
 .dtfc_loop_shprep:\r