X-Git-Url: https://notaz.gp2x.de/cgi-bin/gitweb.cgi?a=blobdiff_plain;f=common%2Farm_utils.s;h=b93b99f0ad33cc96e1793d9386465681f5464638;hb=2f3ca01638b740bc8ca5530dac8f3e2d2e069cb1;hp=9d516493901b2020c66997b90d9fd35e4621f316;hpb=095a240bd91548be4b9770d28d5d6c3e97637094;p=libpicofe.git

diff --git a/common/arm_utils.s b/common/arm_utils.s
index 9d51649..b93b99f 100644
--- a/common/arm_utils.s
+++ b/common/arm_utils.s
@@ -137,69 +137,201 @@ vidConvCpyRGB32hi:
 @ @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
 
 
-@ mode2 blitter for 40 cols
-.global vidCpyM2_40col @ void *dest, void *src
-
-vidCpyM2_40col:
+@ mode2 blitter
+.global vidcpy_m2 @ void *dest, void *src, int m32col, int with_32c_border
+vidcpy_m2:
     stmfd   sp!, {r4-r6,lr}
 
     mov     r12, #224       @ lines
+    add     r0, r0, #320*8
     add     r1, r1, #8
+    mov     lr, #0
+
+    tst     r2, r2
+    movne   lr, #64
+    tstne   r3, r3
+    addne   r0, r0, #32
 
-vidCpyM2_40_loop_out:
+vidCpyM2_loop_out:
     mov     r6, #10
-vidCpyM2_40_loop:
+    sub     r6, r6, lr, lsr #5	@ -= 2 in 32col mode
+vidCpyM2_loop:
     subs    r6, r6, #1
-     ldmia    r1!, {r2-r5}
-     stmia    r0!, {r2-r5}
-     ldmia    r1!, {r2-r5}
-     stmia    r0!, {r2-r5}
-    bne     vidCpyM2_40_loop
+    ldmia   r1!, {r2-r5}
+    stmia   r0!, {r2-r5}
+    ldmia   r1!, {r2-r5}
+    stmia   r0!, {r2-r5}
+    bne     vidCpyM2_loop
+
     subs    r12,r12,#1
+    add     r0, r0, lr
     add     r1, r1, #8
-    bne     vidCpyM2_40_loop_out
-
-    ldmfd   sp!, {r4-r6,lr}
-    bx      lr
+    add     r1, r1, lr
+    bne     vidCpyM2_loop_out
 
+    ldmfd   sp!, {r4-r6,pc}
 
-@ mode2 blitter for 32 cols
-.global vidCpyM2_32col @ void *dest, void *src
 
-vidCpyM2_32col:
-    stmfd   sp!, {r4-r6,lr}
-
-    mov     r12, #224       @ lines
+.global vidcpy_m2_rot @ void *dest, void *src, int m32col, int with_32c_border
+vidcpy_m2_rot:
+    stmfd   sp!,{r4-r8,lr}
     add     r1, r1, #8
-    add     r0, r0, #32
-
-vidCpyM2_32_loop_out:
-    mov     r6, #8
-vidCpyM2_32_loop:
-    subs    r6, r6, #1
-     ldmia    r1!, {r2-r5}
-     stmia    r0!, {r2-r5}
-     ldmia    r1!, {r2-r5}
-     stmia    r0!, {r2-r5}
-    bne     vidCpyM2_32_loop
-    subs    r12,r12,#1
-    add     r0, r0, #64
-    add     r1, r1, #8+64
-    bne     vidCpyM2_32_loop_out
-
-    ldmfd   sp!, {r4-r6,lr}
-    bx      lr
-
-
-@ mode2 blitter for 32 cols with no borders
-.global vidCpyM2_32col_nobord @ void *dest, void *src
+    tst     r2, r2
+    subne   r1, r1, #32		@ adjust
+
+    mov     r4, r0
+    mov     r5, r1
+    mov     r6, r2
+    mov     r7, #8+4
+
+vidcpy_m2_rot_loop:
+    @ a bit lame but oh well..
+    mov     r0, r4
+    mov     r1, r5
+    mov     r2, r7
+    mov     r3, r6
+    mov     r8, #328
+    adr     lr, after_rot_blit8
+    stmfd   sp!,{r4-r8,lr}
+    b       rotated_blit8_2
+
+after_rot_blit8:
+    add     r5, r5, #328*4
+    add     r7, r7, #4
+    cmp     r7, #224+8+4
+    ldmgefd sp!,{r4-r8,pc}
+    b       vidcpy_m2_rot_loop
+
+
+.global rotated_blit8 @ void *dst, void *linesx4, u32 y, int is_32col
+rotated_blit8:
+    stmfd   sp!,{r4-r8,lr}
+    mov     r8, #320
+
+rotated_blit8_2:
+    add     r0, r0, #(240*320)
+    sub     r0, r0, #(240+4)	@ y starts from 4
+    add     r0, r0, r2
+
+    tst     r3, r3
+    subne   r0, r0, #(240*32)
+    addne   r1, r1, #32
+    movne   lr, #256/4
+    moveq   lr, #320/4
+
+rotated_blit_loop8:
+    mov     r6, r1
+    ldr     r2, [r6], r8
+    ldr     r3, [r6], r8
+    ldr     r4, [r6], r8
+    ldr     r5, [r6], r8
+
+    mov     r6, r2, lsl #24
+    mov     r6, r6, lsr #8
+    orr     r6, r6, r3, lsl #24
+    mov     r6, r6, lsr #8
+    orr     r6, r6, r4, lsl #24
+    mov     r6, r6, lsr #8
+    orr     r6, r6, r5, lsl #24
+    str     r6, [r0], #-240
+
+    and     r6, r3, #0xff00
+    and     r7, r2, #0xff00
+    orr     r6, r6, r7, lsr #8
+    and     r7, r4, #0xff00
+    orr     r6, r6, r7, lsl #8
+    and     r7, r5, #0xff00
+    orr     r6, r6, r7, lsl #16
+    str     r6, [r0], #-240
+
+    and     r6, r4, #0xff0000
+    and     r7, r2, #0xff0000
+    orr     r6, r6, r7, lsr #16
+    and     r7, r3, #0xff0000
+    orr     r6, r6, r7, lsr #8
+    and     r7, r5, #0xff0000
+    orr     r6, r6, r7, lsl #8
+    str     r6, [r0], #-240
+
+    mov     r6, r5, lsr #24
+    mov     r6, r6, lsl #8
+    orr     r6, r6, r4, lsr #24
+    mov     r6, r6, lsl #8
+    orr     r6, r6, r3, lsr #24
+    mov     r6, r6, lsl #8
+    orr     r6, r6, r2, lsr #24
+    str     r6, [r0], #-240
+
+    subs    lr, lr, #1
+    add     r1, r1, #4
+    bne     rotated_blit_loop8
+
+    ldmfd   sp!,{r4-r8,pc}
+
+
+@ input: r2-r5
+@ output: r7,r8
+@ trash: r6
+.macro rb_line_low
+    mov     r6, r2, lsl #16
+    mov     r7, r3, lsl #16
+    orr     r7, r7, r6, lsr #16
+    mov     r6, r4, lsl #16
+    mov     r8, r5, lsl #16
+    orr     r8, r8, r6, lsr #16
+.endm
 
-vidCpyM2_32col_nobord:
-    stmfd   sp!, {r4-r6,lr}
+.macro rb_line_hi
+    mov     r6, r2, lsr #16
+    mov     r7, r3, lsr #16
+    orr     r7, r6, r7, lsl #16
+    mov     r6, r4, lsr #16
+    mov     r8, r5, lsr #16
+    orr     r8, r6, r8, lsl #16
+.endm
 
-    mov     r12, #224       @ lines
+.global rotated_blit16 @ void *dst, void *linesx4, u32 y, int is_32col
+rotated_blit16:
+    stmfd   sp!,{r4-r8,lr}
+
+    add     r0, r0, #(240*320)*2
+    sub     r0, r0, #(240+4)*2	@ y starts from 4
+    add     r0, r0, r2, lsl #1
+
+    tst     r3, r3
+    subne   r0, r0, #(240*32)*2
+    addne   r1, r1, #32*2
+    movne   lr, #256/4
+    moveq   lr, #320/4
+
+rotated_blit_loop16:
+    ldr     r2, [r1, #320*0*2]
+    ldr     r3, [r1, #320*1*2]
+    ldr     r4, [r1, #320*2*2]
+    ldr     r5, [r1, #320*3*2]
+    rb_line_low
+    stmia   r0, {r7,r8}
+    sub     r0, r0, #240*2
+    rb_line_hi
+    stmia   r0, {r7,r8}
+    sub     r0, r0, #240*2
+
+    ldr     r2, [r1, #320*0*2+4]
+    ldr     r3, [r1, #320*1*2+4]
+    ldr     r4, [r1, #320*2*2+4]
+    ldr     r5, [r1, #320*3*2+4]
+    rb_line_low
+    stmia   r0, {r7,r8}
+    sub     r0, r0, #240*2
+    rb_line_hi
+    stmia   r0, {r7,r8}
+    sub     r0, r0, #240*2
+
+    subs    lr, lr, #1
     add     r1, r1, #8
-    b       vidCpyM2_32_loop_out
+    bne     rotated_blit_loop16
+
+    ldmfd   sp!,{r4-r8,pc}
 
 
 .global spend_cycles @ c
@@ -213,4 +345,3 @@ spend_cycles:
 
     bx      lr
 
-