4 .global flushcache @ beginning_addr, end_addr, flags
11 /* buggy and slow, probably because function call overhead
12 @ renderer helper, based on bitbank's method
13 .global draw8pix @ uint8 *P, uint8 *C, uint8 *PALRAM @ dest, src, pal
18 ldrb r3, [r1] @ get bit 0 pixels
20 orr r12,r12,r12,lsl #8
21 orr r12,r12,r12,lsl #16
22 ldrb r1, [r1, #8] @ get bit 1 pixels
23 orr r3, r3, r3, lsl #9 @ shift them over 1 byte + 1 bit
24 orr r3, r3, r3, lsl #18 @ now 4 pixels take up 4 bytes
25 and r4, r12,r3, lsr #7 @ mask off the upper nibble pixels we want
26 and r5, r12,r3, lsr #3 @ mask off the lower nibble pixels we want
29 orr r1, r1, r1, lsl #9 @ process the bit 1 pixels
30 orr r1, r1, r1, lsl #18
31 and r3, r12,r1, lsr #7 @ mask off the upper nibble pixels we want
32 and r1, r12,r1, lsr #3 @ mask off the lower nibble
33 orr r4, r4, r3, lsl #1
34 orr r5, r5, r1, lsl #5
36 @ can this be avoided?
37 mov r4, r4, lsl #3 @ *8
50 mov r5, r5, lsl #3 @ *8