+++ /dev/null
-@ vim:filetype=armasm\r
-@ some color conversion and blitting routines\r
-\r
-@ (c) Copyright 2006, 2007 notaz\r
-@ All Rights Reserved\r
-\r
-@ vim:filetype=armasm\r
-\r
-.text\r
-.align 4\r
-\r
-@ Convert 0000bbb0 ggg0rrr0 0000bbb0 ggg0rrr0\r
-@ to 00000000 rrr00000 ggg00000 bbb00000 ...\r
-\r
-@ lr = 0x00e000e0, out: r3=lower_pix, r2=higher_pix; trashes rin\r
-.macro convRGB32_2 rin sh=0\r
- and r2, lr, \rin, lsr #4 @ blue\r
- and r3, \rin, lr\r
- orr r2, r2, r3, lsl #8 @ g0b0g0b0\r
-\r
- mov r3, r2, lsl #16 @ g0b00000\r
- and \rin,lr, \rin, ror #12 @ 00r000r0 (reversed)\r
- orr r3, r3, \rin, lsr #16 @ g0b000r0\r
-.if \sh == 1\r
- mov r3, r3, ror #17 @ shadow mode\r
-.elseif \sh == 2\r
- adds r3, r3, #0x40000000 @ green\r
- orrcs r3, r3, #0xe0000000\r
- mov r3, r3, ror #8\r
- adds r3, r3, #0x40000000\r
- orrcs r3, r3, #0xe0000000\r
- mov r3, r3, ror #16\r
- adds r3, r3, #0x40000000\r
- orrcs r3, r3, #0xe0000000\r
- mov r3, r3, ror #24\r
-.else\r
- mov r3, r3, ror #16 @ r3=low\r
-.endif\r
-\r
- orr r3, r3, r3, lsr #3\r
- str r3, [r0], #4\r
-\r
- mov r2, r2, lsr #16\r
- orr r2, r2, \rin, lsl #16\r
-.if \sh == 1\r
- mov r2, r2, lsr #1\r
-.elseif \sh == 2\r
- mov r2, r2, ror #8\r
- adds r2, r2, #0x40000000 @ blue\r
- orrcs r2, r2, #0xe0000000\r
- mov r2, r2, ror #8\r
- adds r2, r2, #0x40000000\r
- orrcs r2, r2, #0xe0000000\r
- mov r2, r2, ror #8\r
- adds r2, r2, #0x40000000\r
- orrcs r2, r2, #0xe0000000\r
- mov r2, r2, ror #8\r
-.endif\r
-\r
- orr r2, r2, r2, lsr #3\r
-.if \sh == 1\r
- str r2, [r0, #0x40*2*4]\r
-.endif\r
- str r2, [r0], #4\r
-.endm\r
-\r
-\r
-.global bgr444_to_rgb32 @ void *to, void *from\r
-\r
-bgr444_to_rgb32:\r
- stmfd sp!, {r4-r7,lr}\r
-\r
- mov r12, #0x40>>3 @ repeats\r
- mov lr, #0x00e00000\r
- orr lr, lr, #0x00e0\r
-\r
-.loopRGB32:\r
- subs r12, r12, #1\r
-\r
- ldmia r1!, {r4-r7}\r
- convRGB32_2 r4\r
- convRGB32_2 r5\r
- convRGB32_2 r6\r
- convRGB32_2 r7\r
- bgt .loopRGB32\r
-\r
- ldmfd sp!, {r4-r7,pc}\r
-\r
-\r
-.global bgr444_to_rgb32_sh @ void *to, void *from\r
-\r
-bgr444_to_rgb32_sh:\r
- stmfd sp!, {r4-r7,lr}\r
-\r
- mov r12, #0x40>>3 @ repeats\r
- add r0, r0, #0x40*4\r
- mov lr, #0x00e00000\r
- orr lr, lr, #0x00e0\r
-\r
-.loopRGB32sh:\r
- subs r12, r12, #1\r
-\r
- ldmia r1!, {r4-r7}\r
- convRGB32_2 r4, 1\r
- convRGB32_2 r5, 1\r
- convRGB32_2 r6, 1\r
- convRGB32_2 r7, 1\r
- bgt .loopRGB32sh\r
-\r
- mov r12, #0x40>>3 @ repeats\r
- sub r1, r1, #0x40*2\r
-\r
-.loopRGB32hi:\r
- ldmia r1!, {r4-r7}\r
- convRGB32_2 r4, 2\r
- convRGB32_2 r5, 2\r
- convRGB32_2 r6, 2\r
- convRGB32_2 r7, 2\r
-\r
- subs r12, r12, #1\r
- bgt .loopRGB32hi\r
-\r
- ldmfd sp!, {r4-r7,lr}\r
- bx lr\r
-\r
-\r
-@ @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@\r
-\r
-\r
-@ mode2 blitter\r
-.global vidcpy_m2 @ void *dest, void *src, int m32col, int with_32c_border\r
-vidcpy_m2:\r
- stmfd sp!, {r4-r6,lr}\r
-\r
- mov r12, #224 @ lines\r
- add r0, r0, #320*8\r
- add r1, r1, #8\r
- mov lr, #0\r
-\r
- tst r2, r2\r
- movne lr, #64\r
- tstne r3, r3\r
- addne r0, r0, #32\r
-\r
-vidCpyM2_loop_out:\r
- mov r6, #10\r
- sub r6, r6, lr, lsr #5 @ -= 2 in 32col mode\r
-vidCpyM2_loop:\r
- subs r6, r6, #1\r
- ldmia r1!, {r2-r5}\r
- stmia r0!, {r2-r5}\r
- ldmia r1!, {r2-r5}\r
- stmia r0!, {r2-r5}\r
- bne vidCpyM2_loop\r
-\r
- subs r12,r12,#1\r
- add r0, r0, lr\r
- add r1, r1, #8\r
- add r1, r1, lr\r
- bne vidCpyM2_loop_out\r
-\r
- ldmfd sp!, {r4-r6,pc}\r
-\r
-\r
-.global vidcpy_m2_rot @ void *dest, void *src, int m32col, int with_32c_border\r
-vidcpy_m2_rot:\r
- stmfd sp!,{r4-r8,lr}\r
- add r1, r1, #8\r
- tst r2, r2\r
- subne r1, r1, #32 @ adjust\r
-\r
- mov r4, r0\r
- mov r5, r1\r
- mov r6, r2\r
- mov r7, #8+4\r
-\r
-vidcpy_m2_rot_loop:\r
- @ a bit lame but oh well..\r
- mov r0, r4\r
- mov r1, r5\r
- mov r2, r7\r
- mov r3, r6\r
- mov r8, #328\r
- adr lr, after_rot_blit8\r
- stmfd sp!,{r4-r8,lr}\r
- b rotated_blit8_2\r
-\r
-after_rot_blit8:\r
- add r5, r5, #328*4\r
- add r7, r7, #4\r
- cmp r7, #224+8+4\r
- ldmgefd sp!,{r4-r8,pc}\r
- b vidcpy_m2_rot_loop\r
-\r
-\r
-.global rotated_blit8 @ void *dst, void *linesx4, u32 y, int is_32col\r
-rotated_blit8:\r
- stmfd sp!,{r4-r8,lr}\r
- mov r8, #320\r
-\r
-rotated_blit8_2:\r
- add r0, r0, #(240*320)\r
- sub r0, r0, #(240+4) @ y starts from 4\r
- add r0, r0, r2\r
-\r
- tst r3, r3\r
- subne r0, r0, #(240*32)\r
- addne r1, r1, #32\r
- movne lr, #256/4\r
- moveq lr, #320/4\r
-\r
-rotated_blit_loop8:\r
- mov r6, r1\r
- ldr r2, [r6], r8\r
- ldr r3, [r6], r8\r
- ldr r4, [r6], r8\r
- ldr r5, [r6], r8\r
-\r
- mov r6, r2, lsl #24\r
- mov r6, r6, lsr #8\r
- orr r6, r6, r3, lsl #24\r
- mov r6, r6, lsr #8\r
- orr r6, r6, r4, lsl #24\r
- mov r6, r6, lsr #8\r
- orr r6, r6, r5, lsl #24\r
- str r6, [r0], #-240\r
-\r
- and r6, r3, #0xff00\r
- and r7, r2, #0xff00\r
- orr r6, r6, r7, lsr #8\r
- and r7, r4, #0xff00\r
- orr r6, r6, r7, lsl #8\r
- and r7, r5, #0xff00\r
- orr r6, r6, r7, lsl #16\r
- str r6, [r0], #-240\r
-\r
- and r6, r4, #0xff0000\r
- and r7, r2, #0xff0000\r
- orr r6, r6, r7, lsr #16\r
- and r7, r3, #0xff0000\r
- orr r6, r6, r7, lsr #8\r
- and r7, r5, #0xff0000\r
- orr r6, r6, r7, lsl #8\r
- str r6, [r0], #-240\r
-\r
- mov r6, r5, lsr #24\r
- mov r6, r6, lsl #8\r
- orr r6, r6, r4, lsr #24\r
- mov r6, r6, lsl #8\r
- orr r6, r6, r3, lsr #24\r
- mov r6, r6, lsl #8\r
- orr r6, r6, r2, lsr #24\r
- str r6, [r0], #-240\r
-\r
- subs lr, lr, #1\r
- add r1, r1, #4\r
- bne rotated_blit_loop8\r
-\r
- ldmfd sp!,{r4-r8,pc}\r
-\r
-\r
-@ input: r2-r5\r
-@ output: r7,r8\r
-@ trash: r6\r
-.macro rb_line_low\r
- mov r6, r2, lsl #16\r
- mov r7, r3, lsl #16\r
- orr r7, r7, r6, lsr #16\r
- mov r6, r4, lsl #16\r
- mov r8, r5, lsl #16\r
- orr r8, r8, r6, lsr #16\r
-.endm\r
-\r
-.macro rb_line_hi\r
- mov r6, r2, lsr #16\r
- mov r7, r3, lsr #16\r
- orr r7, r6, r7, lsl #16\r
- mov r6, r4, lsr #16\r
- mov r8, r5, lsr #16\r
- orr r8, r6, r8, lsl #16\r
-.endm\r
-\r
-.global rotated_blit16 @ void *dst, void *linesx4, u32 y, int is_32col\r
-rotated_blit16:\r
- stmfd sp!,{r4-r8,lr}\r
-\r
- add r0, r0, #(240*320)*2\r
- sub r0, r0, #(240+4)*2 @ y starts from 4\r
- add r0, r0, r2, lsl #1\r
-\r
- tst r3, r3\r
- subne r0, r0, #(240*32)*2\r
- addne r1, r1, #32*2\r
- movne lr, #256/4\r
- moveq lr, #320/4\r
-\r
-rotated_blit_loop16:\r
- ldr r2, [r1, #320*0*2]\r
- ldr r3, [r1, #320*1*2]\r
- ldr r4, [r1, #320*2*2]\r
- ldr r5, [r1, #320*3*2]\r
- rb_line_low\r
- stmia r0, {r7,r8}\r
- sub r0, r0, #240*2\r
- rb_line_hi\r
- stmia r0, {r7,r8}\r
- sub r0, r0, #240*2\r
-\r
- ldr r2, [r1, #320*0*2+4]\r
- ldr r3, [r1, #320*1*2+4]\r
- ldr r4, [r1, #320*2*2+4]\r
- ldr r5, [r1, #320*3*2+4]\r
- rb_line_low\r
- stmia r0, {r7,r8}\r
- sub r0, r0, #240*2\r
- rb_line_hi\r
- stmia r0, {r7,r8}\r
- sub r0, r0, #240*2\r
-\r
- subs lr, lr, #1\r
- add r1, r1, #8\r
- bne rotated_blit_loop16\r
-\r
- ldmfd sp!,{r4-r8,pc}\r
-\r
-\r
-.global spend_cycles @ c\r
-\r
-spend_cycles:\r
- mov r0, r0, lsr #2 @ 4 cycles/iteration\r
- sub r0, r0, #2 @ entry/exit/init\r
-.sc_loop:\r
- subs r0, r0, #1\r
- bpl .sc_loop\r
-\r
- bx lr\r
-\r
notaz.gp2x.de / libpicofe.git / blobdiff