gpsp/gp2x/upscale_aspect.s

580 lines
14 KiB
ArmAsm

/*
* 240x160 -> 320x213 upscaler for ARM with interpolation
*
* Written by Gražvydas "notaz" Ignotas
* Prototyped by Rokas
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions are met:
* * Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* * Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* * Neither the name of the organization nor the
* names of its contributors may be used to endorse or promote products
* derived from this software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
* AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
* SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
* CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
* OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*
* 0 1 2 : 3 4 5
* 6 7 8 : 9 10 11
* 12 13 14 : 15 16 17
* v
* 0 1 2 3 : 4 5 6 7
* 8 9 10 11 : 12 13 14 15
* 16 17 18 19 : 20 21 22 23
* 24 25 26 27 : 28 29 30 31
*/
.macro unpack_hi dst, src
mov \dst, \src, lsr #16
orr \dst, \dst, \dst, lsl #16
and \dst, \dst, lr
.endm
.macro unpack_lo dst, src
mov \dst, \src, lsl #16
orr \dst, \dst, \dst, lsr #16
and \dst, \dst, lr
.endm
@ do 3:5 summing: r2 = (s1*3 + s2*5 + 4) / 8
@ s2 != r2
.macro do_3_5 s1, s2
add r2,\s1,\s1, lsl #1 @ r2 = s1 * 3
add r2, r2,\s2, lsl #2
add r2, r2,\s2 @ r2 += s2 * 5
add r2, r2, r12,lsl #2 @ sum += round * 4
and r2, lr, r2, lsr #3 @ mask_to_unpacked(sum / 8)
.endm
@ do 14:7:7:4: r2 = (s1*14 + s2*7 + s3*7 + s4*4 + 16) / 32
@ {s2,s3,s4} != r2
.macro do_14_7_7_4 s1, s2, s3, s4
mov r2,\s1, lsl #4
sub r2, r2,\s1, lsl #1 @ r2 = s1 * 14
add r2, r2,\s2, lsl #3
sub r2, r2,\s2 @ r2 += s2 * 7
add r2, r2,\s3, lsl #3
sub r2, r2,\s3 @ r2 += s3 * 7
add r2, r2,\s4, lsl #2 @ r2 += s4 * 4
add r2, r2, r12,lsl #3 @ sum += round * 16
and r2, lr, r2, lsr #5 @ mask_to_unpacked(sum / 32)
.endm
.global upscale_aspect @ u16 *dst, u16 *src
upscale_aspect:
stmfd sp!,{r4-r11,lr}
mov lr, #0x0000001f
orr lr, lr, #0x0000f800 @ for "unpacked" form of
orr lr, lr, #0x07e00000 @ 00000ggg'ggg00000'rrrrr000'000bbbbb
mov r12, #0x00000001
orr r12,r12,#0x00000800
orr r12,r12,#0x00200000 @ rounding constant
mov r8, #((240/6)-1) << 24 @ cols
orr r8, r8, #160/3 @ rows
add r0, r0, #320*2*13
loop1:
ldr r10,[r1]
ldr r11,[r1, #320*2*1]
unpack_lo r4, r10
unpack_hi r5, r10
unpack_lo r6, r11
unpack_hi r7, r11
ldr r11,[r1, #4]
do_3_5 r4, r5
orr r2, r2, r2, lsr #16
mov r3, r10, lsl #16
mov r3, r3, lsr #16
orr r2, r3, r2, lsl #16
str r2, [r0] @ 0,1
ldr r10,[r1, #320*2*2]
do_3_5 r4, r6
orr r3, r2, r2, lsl #16
mov r3, r3, lsr #16 @ 8
do_14_7_7_4 r7, r5, r6, r4
orr r2, r2, r2, lsr #16
orr r2, r3, r2, lsl #16
str r2, [r0, #320*2*1] @ 8,9
unpack_lo r4, r10
unpack_hi r9, r10
do_3_5 r4, r6
orr r3, r2, r2, lsl #16
mov r3, r3, lsr #16
do_14_7_7_4 r7, r9, r6, r4
orr r2, r2, r2, lsr #16
orr r2, r3, r2, lsl #16
str r2, [r0, #320*2*2] @ 16,17
do_3_5 r4, r9
orr r2, r2, r2, lsr #16
mov r3, r10, lsl #16
mov r3, r3, lsr #16
orr r2, r3, r2, lsl #16
str r2, [r0, #320*2*3] @ 24,25
ldr r10,[r1, #320*2*1+4]
unpack_lo r6, r11
unpack_lo r4, r10
do_3_5 r6, r5
orr r2, r2, r2, lsl #16
mov r3, r11, lsl #16
orr r2, r3, r2, lsr #16
str r2, [r0, #4] @ 2,3
do_14_7_7_4 r7, r4, r5, r6
orr r2, r2, r2, lsl #16
mov r3, r2, lsr #16
ldr r5, [r1, #320*2*2+4]
do_3_5 r6, r4
orr r2, r2, r2, lsr #16
orr r2, r3, r2, lsl #16
str r2, [r0, #320*2*1+4] @ 10,11
unpack_lo r6, r5
do_14_7_7_4 r7, r4, r9, r6
orr r2, r2, r2, lsl #16
mov r3, r2, lsr #16
do_3_5 r6, r4
orr r2, r2, r2, lsr #16
orr r2, r3, r2, lsl #16
str r2, [r0, #320*2*2+4] @ 18,19
unpack_hi r4, r10
ldr r10,[r1, #8]
do_3_5 r6, r9
orr r2, r2, r2, lsl #16
mov r2, r2, lsr #16
orr r2, r2, r5, lsl #16
str r2, [r0, #320*2*3+4] @ 26,27
unpack_hi r6, r11
unpack_lo r7, r10
do_3_5 r6, r7
orr r2, r2, r2, lsr #16
mov r2, r2, lsl #16
orr r2, r2, r11,lsr #16
str r2, [r0, #8] @ 4,5
ldr r11,[r1, #320*2*1+8]
unpack_hi r9, r10
do_3_5 r9, r7
orr r2, r2, r2, lsr #16
mov r2, r2, lsl #16
orr r2, r2, r10,lsr #16
mov r2, r2, ror #16
str r2, [r0, #12] @ 6,7
unpack_lo r10,r11
do_3_5 r6, r4
orr r2, r2, r2, lsl #16
mov r3, r2, lsr #16
do_14_7_7_4 r10, r4, r7, r6
orr r2, r2, r2, lsr #16
orr r2, r3, r2, lsl #16
str r2, [r0, #320*2*1+8] @ 12,13
unpack_hi r6, r11
ldr r11,[r1, #320*2*2+8]
do_14_7_7_4 r10, r6, r7, r9
orr r2, r2, r2, lsl #16
mov r3, r2, lsr #16
do_3_5 r9, r6
orr r2, r2, r2, lsr #16
orr r2, r3, r2, lsl #16
str r2, [r0, #320*2*1+12] @ 14,15
unpack_hi r7, r5
unpack_lo r9, r11
do_3_5 r7, r4
orr r2, r2, r2, lsl #16
mov r3, r2, lsr #16
do_14_7_7_4 r10, r4, r9, r7
orr r2, r2, r2, lsr #16
orr r2, r3, r2, lsl #16
str r2, [r0, #320*2*2+8] @ 20,21
do_3_5 r7, r9
orr r2, r2, r2, lsr #16
mov r2, r2, lsl #16
orr r2, r2, r5, lsr #16
str r2, [r0, #320*2*3+8] @ 28,29
unpack_hi r5, r11
do_14_7_7_4 r10, r6, r9, r5
orr r2, r2, r2, lsl #16
mov r3, r2, lsr #16
do_3_5 r5, r6
orr r2, r2, r2, lsr #16
orr r2, r3, r2, lsl #16
str r2, [r0, #320*2*2+12] @ 22,23
do_3_5 r5, r9
orr r2, r2, r2, lsr #16
mov r3, r11, lsr #16
orr r2, r3, r2, lsl #16
mov r2, r2, ror #16
str r2, [r0, #320*2*3+12] @ 30,31
add r0, r0, #16
add r1, r1, #12
subs r8, r8, #1<<24
bpl loop1
add r0, r0, #320*3*2
add r1, r1, #(320*2+80)*2
sub r8, r8, #1
tst r8, #0xff
add r8, r8, #(240/6) << 24 @ cols
bne loop1
@@ last line
mov r8, #240/6
loop2:
ldmia r1!,{r9,r10,r11}
unpack_lo r4, r9
unpack_hi r5, r9
do_3_5 r4, r5
orr r2, r2, r2, lsr #16
mov r3, r9, lsl #16
mov r3, r3, lsr #16
orr r2, r3, r2, lsl #16
str r2, [r0], #4
unpack_lo r6, r10
unpack_hi r7, r10
do_3_5 r6, r5
orr r2, r2, r2, lsl #16
mov r2, r2, lsr #16
orr r2, r2, r10,lsl #16
str r2, [r0], #4
unpack_lo r4, r11
unpack_hi r5, r11
do_3_5 r7, r4
orr r2, r2, r2, lsr #16
mov r3, r10, lsr #16
orr r2, r3, r2, lsl #16
str r2, [r0], #4
do_3_5 r5, r4
orr r2, r2, r2, lsr #16
mov r3, r11, lsr #16
orr r2, r3, r2, lsl #16
mov r2, r2, ror #16
str r2, [r0], #4
subs r8, r8, #1
bne loop2
ldmfd sp!,{r4-r11,pc}
.global upscale_aspect_row @ void *dst, void *linesx4, u32 row
upscale_aspect_row:
stmfd sp!,{r4-r11,lr}
mov lr, #0x0000001f
orr lr, lr, #0x0000f800 @ for "unpacked" form of
orr lr, lr, #0x07e00000 @ 00000ggg'ggg00000'rrrrr000'000bbbbb
mov r12, #0x00000001
orr r12,r12,#0x00000800
orr r12,r12,#0x00200000 @ rounding constant
mov r8, #(240/6) @ cols
add r0, r0, #(240*320)*2
add r0, r0, #12*2
add r0, r0, r2, lsl #3
uar_loop:
ldr r10,[r1]
ldr r11,[r1, #240*2*1]
unpack_lo r4, r10
unpack_hi r5, r10
unpack_lo r6, r11
unpack_hi r7, r11
ldr r11,[r1, #240*2*2]
do_3_5 r4, r6
orr r2, r2, r2, lsr #16
mov r3, r10, lsl #16
mov r3, r3, lsr #16
orr r2, r3, r2, lsl #16
str r2, [r0, #-240*2]! @ 0,8
unpack_lo r10,r11
unpack_hi r9, r11
do_3_5 r10,r6
orr r2, r2, r2, lsl #16
mov r3, r11, lsl #16
orr r2, r3, r2, lsr #16
str r2, [r0, #4] @ 16,24
do_3_5 r4, r5
orr r3, r2, r2, lsl #16
do_14_7_7_4 r7, r5, r6, r4
orr r2, r2, r2, lsr #16
mov r3, r3, lsr #16
orr r2, r3, r2, lsl #16
str r2, [r0, #-240*2]! @ 1,9
ldr r11,[r1, #4]
do_14_7_7_4 r7, r6, r9, r10
orr r3, r2, r2, lsl #16
do_3_5 r10,r9
orr r2, r2, r2, lsr #16
mov r3, r3, lsr #16
orr r2, r3, r2, lsl #16
str r2, [r0, #4] @ 17,25
ldr r10,[r1, #240*2*1+4]
unpack_lo r4, r11
unpack_lo r6, r10
do_3_5 r4, r5
orr r3, r2, r2, lsl #16
do_14_7_7_4 r7, r5, r6, r4
orr r2, r2, r2, lsr #16
mov r3, r3, lsr #16
orr r2, r3, r2, lsl #16
str r2, [r0, #-240*2]! @ 2,10
do_3_5 r4, r6
ldr r4, [r1, #240*2*2+4]
orr r2, r2, r2, lsr #16
mov r3, r11, lsl #16
mov r3, r3, lsr #16
orr r2, r3, r2, lsl #16
str r2, [r0, #-240*2] @ 3,11
unpack_lo r5, r4
do_14_7_7_4 r7, r6, r9, r5
orr r3, r2, r2, lsl #16
do_3_5 r5, r9
orr r2, r2, r2, lsr #16
mov r2, r2, lsl #16
orr r2, r2, r3, lsr #16
str r2, [r0, #4] @ 18,26
do_3_5 r5, r6
orr r2, r2, r2, lsl #16
mov r3, r4, lsl #16
orr r2, r3, r2, lsr #16
str r2, [r0, #-240*2+4] @ 19,27
unpack_hi r5, r11
unpack_hi r6, r10
unpack_hi r7, r4
ldr r10,[r1, #8]
do_3_5 r5, r6
orr r2, r2, r2, lsr #16
mov r3, r11, lsr #16
orr r2, r3, r2, lsl #16
str r2, [r0, #-240*2*2]! @ 4,12
ldr r11,[r1, #240*2*1+8]
do_3_5 r7, r6
orr r2, r2, r2, lsl #16
mov r3, r4, lsr #16
mov r3, r3, lsl #16
orr r2, r3, r2, lsr #16
str r2, [r0, #4] @ 20,28
unpack_lo r4, r10
unpack_lo r9, r11
ldr r11,[r1, #240*2*2+8]
do_3_5 r5, r4
orr r3, r2, r2, lsl #16
do_14_7_7_4 r9, r4, r6, r5
orr r2, r2, r2, lsr #16
mov r2, r2, lsl #16
orr r2, r2, r3, lsr #16
str r2, [r0, #-240*2]! @ 5,13
unpack_lo r5, r11
do_14_7_7_4 r9, r5, r6, r7
orr r3, r2, r2, lsl #16
do_3_5 r7, r5
orr r2, r2, r2, lsr #16
mov r3, r3, lsr #16
orr r2, r3, r2, lsl #16
str r2, [r0, #4] @ 21,29
ldr r7, [r1, #240*2*1+8]
unpack_hi r6, r10
unpack_hi r7, r7
do_3_5 r6, r4
orr r3, r2, r2, lsl #16
do_14_7_7_4 r9, r4, r7, r6
orr r2, r2, r2, lsr #16
mov r2, r2, lsl #16
orr r2, r2, r3, lsr #16
str r2, [r0, #-240*2]! @ 6,14
unpack_hi r4, r11
do_14_7_7_4 r9, r5, r7, r4
orr r3, r2, r2, lsl #16
do_3_5 r4, r5
orr r2, r2, r2, lsr #16
mov r3, r3, lsr #16
orr r2, r3, r2, lsl #16
str r2, [r0, #4] @ 22,30
do_3_5 r6, r7
orr r2, r2, r2, lsr #16
mov r3, r10, lsr #16
orr r2, r3, r2, lsl #16
str r2, [r0, #-240*2]! @ 7,15
do_3_5 r4, r7
orr r2, r2, r2, lsl #16
mov r3, r11, lsr #16
mov r3, r3, lsl #16
orr r2, r3, r2, lsr #16
str r2, [r0, #4] @ 23,31
subs r8, r8, #1
add r1, r1, #12
bne uar_loop
ldmfd sp!,{r4-r11,pc}
@ bonus function
@ 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
.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
.global do_rotated_blit @ void *dst, void *linesx4, u32 y
do_rotated_blit:
stmfd sp!,{r4-r8,lr}
add r0, r0, #(240*320)*2
sub r0, r0, #(240*40)*2
sub r0, r0, #(240-40+4)*2 @ y starts from 4
add r0, r0, r2, lsl #1
mov lr, #240/4
rotated_blit_loop:
ldr r2, [r1, #240*0*2]
ldr r3, [r1, #240*1*2]
ldr r4, [r1, #240*2*2]
ldr r5, [r1, #240*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, #240*0*2+4]
ldr r3, [r1, #240*1*2+4]
ldr r4, [r1, #240*2*2+4]
ldr r5, [r1, #240*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
bne rotated_blit_loop
ldmfd sp!,{r4-r8,pc}
@ vim:filetype=armasm