X-Git-Url: https://notaz.gp2x.de/cgi-bin/gitweb.cgi?a=blobdiff_plain;f=pandora%2Fasm_utils.s;h=74dd3fdcca0bbe59afe43c1ad36b56519d924bf6;hb=e99d80480196f551f7e6c06deae3750de19b4c0e;hp=7d57c54e816273e946254ce02cfed1e7d5c8964c;hpb=5d239ae7a46330845af01f9e5f429e78a099fbd9;p=libpicofe.git

diff --git a/pandora/asm_utils.s b/pandora/asm_utils.s
index 7d57c54..74dd3fd 100644
--- a/pandora/asm_utils.s
+++ b/pandora/asm_utils.s
@@ -1,17 +1,10 @@
 @ vim:filetype=armasm
 
 
-.global cache_flush_d_inval_i
+@ FIXME: handle dual issue
+.global clut_line2x2 @ void *dest, void *src, unsigned short *pal, int pixels_mask
 
-cache_flush_d_inval_i:
-    mov     r2, #0x0  @ must be 0 ???
-    swi     0x9f0002
-    bx      lr
-
-
-.global clut_line @ void *dest, void *src, unsigned short *pal, int pixels_mask
-
-clut_line:
+clut_line2x2:
     stmfd   sp!, {r4-r11,lr}
 
     and     lr, r3, #0xff0000
@@ -20,7 +13,7 @@ clut_line:
     and     r3, r3, #0xff          @ counter
     add     r11,r0, #800*2
 
-clut_line_loop:
+clut_line_2x2_loop:
     ldmia   r1!, {r10,r12}
 
     and     r4, lr, r10, lsl #1
@@ -58,7 +51,58 @@ clut_line_loop:
 
     stmia   r0!, {r4-r10,r12}
     stmia   r11!,{r4-r10,r12}
-    bne     clut_line_loop
+    bne     clut_line_2x2_loop
+
+    ldmfd   sp!, {r4-r11,pc}
+
+
+@   00 01 11 22 23 33
+@   r4 r5 r6 r7 r8 r9
+
+.macro do_4_to_12 rs
+    and     r4, lr, \rs, lsl #1
+    and     r6, lr, \rs, lsr #7
+    and     r7, lr, \rs, lsr #15
+    and     r9, lr, \rs, lsr #23
+    ldrh    r4, [r2, r4]
+    ldrh    r6, [r2, r6]
+    ldrh    r7, [r2, r7]
+    ldrh    r9, [r2, r9]
+
+    orr     r5, r4, r6,  lsl #16
+    orr     r4, r4, r4,  lsl #16
+    orr     r6, r6, r6,  lsl #16
+
+    orr     r8, r7, r9,  lsl #16
+    orr     r7, r7, r7,  lsl #16
+    orr     r9, r9, r9,  lsl #16
+.endm
+
+
+.global clut_line3x2 @ void *dest, void *src, unsigned short *pal, int pixels_mask
+
+clut_line3x2:
+    stmfd   sp!, {r4-r11,lr}
+
+    and     lr, r3, #0xff0000
+    mov     lr, lr, lsr #15        @ mask
+    mov     r3, r3, lsr #3
+    and     r3, r3, #0xff          @ counter
+    add     r11,r0, #800*2
+
+clut_line3x2_loop:
+    ldmia   r1!, {r10,r12}
+
+    do_4_to_12 r10
+    stmia   r0!, {r4-r9}
+    stmia   r11!,{r4-r9}
+
+    do_4_to_12 r12
+    subs    r3, r3, #1
+    stmia   r0!, {r4-r9}
+    stmia   r11!,{r4-r9}
+
+    bne     clut_line3x2_loop
 
     ldmfd   sp!, {r4-r11,pc}