Originally committed as revision 15725 to svn://svn.ffmpeg.org/ffmpeg/trunktags/v0.5
@@ -20,6 +20,7 @@ | |||||
@ | @ | ||||
#include "config.h" | #include "config.h" | ||||
#include "asm.S" | |||||
#ifndef HAVE_PLD | #ifndef HAVE_PLD | ||||
.macro pld reg | .macro pld reg | ||||
@@ -79,8 +80,7 @@ | |||||
@ ---------------------------------------------------------------- | @ ---------------------------------------------------------------- | ||||
.align 8 | .align 8 | ||||
.global put_pixels16_arm | |||||
put_pixels16_arm: | |||||
function put_pixels16_arm, export=1 | |||||
@ void func(uint8_t *block, const uint8_t *pixels, int line_size, int h) | @ void func(uint8_t *block, const uint8_t *pixels, int line_size, int h) | ||||
@ block = word aligned, pixles = unaligned | @ block = word aligned, pixles = unaligned | ||||
pld [r1] | pld [r1] | ||||
@@ -138,11 +138,11 @@ put_pixels16_arm: | |||||
.word 2b | .word 2b | ||||
.word 3b | .word 3b | ||||
.word 4b | .word 4b | ||||
.endfunc | |||||
@ ---------------------------------------------------------------- | @ ---------------------------------------------------------------- | ||||
.align 8 | .align 8 | ||||
.global put_pixels8_arm | |||||
put_pixels8_arm: | |||||
function put_pixels8_arm, export=1 | |||||
@ void func(uint8_t *block, const uint8_t *pixels, int line_size, int h) | @ void func(uint8_t *block, const uint8_t *pixels, int line_size, int h) | ||||
@ block = word aligned, pixles = unaligned | @ block = word aligned, pixles = unaligned | ||||
pld [r1] | pld [r1] | ||||
@@ -200,11 +200,11 @@ put_pixels8_arm: | |||||
.word 2b | .word 2b | ||||
.word 3b | .word 3b | ||||
.word 4b | .word 4b | ||||
.endfunc | |||||
@ ---------------------------------------------------------------- | @ ---------------------------------------------------------------- | ||||
.align 8 | .align 8 | ||||
.global put_pixels8_x2_arm | |||||
put_pixels8_x2_arm: | |||||
function put_pixels8_x2_arm, export=1 | |||||
@ void func(uint8_t *block, const uint8_t *pixels, int line_size, int h) | @ void func(uint8_t *block, const uint8_t *pixels, int line_size, int h) | ||||
@ block = word aligned, pixles = unaligned | @ block = word aligned, pixles = unaligned | ||||
pld [r1] | pld [r1] | ||||
@@ -270,10 +270,10 @@ put_pixels8_x2_arm: | |||||
.word 2b | .word 2b | ||||
.word 3b | .word 3b | ||||
.word 4b | .word 4b | ||||
.endfunc | |||||
.align 8 | .align 8 | ||||
.global put_no_rnd_pixels8_x2_arm | |||||
put_no_rnd_pixels8_x2_arm: | |||||
function put_no_rnd_pixels8_x2_arm, export=1 | |||||
@ void func(uint8_t *block, const uint8_t *pixels, int line_size, int h) | @ void func(uint8_t *block, const uint8_t *pixels, int line_size, int h) | ||||
@ block = word aligned, pixles = unaligned | @ block = word aligned, pixles = unaligned | ||||
pld [r1] | pld [r1] | ||||
@@ -339,12 +339,12 @@ put_no_rnd_pixels8_x2_arm: | |||||
.word 2b | .word 2b | ||||
.word 3b | .word 3b | ||||
.word 4b | .word 4b | ||||
.endfunc | |||||
@ ---------------------------------------------------------------- | @ ---------------------------------------------------------------- | ||||
.align 8 | .align 8 | ||||
.global put_pixels8_y2_arm | |||||
put_pixels8_y2_arm: | |||||
function put_pixels8_y2_arm, export=1 | |||||
@ void func(uint8_t *block, const uint8_t *pixels, int line_size, int h) | @ void func(uint8_t *block, const uint8_t *pixels, int line_size, int h) | ||||
@ block = word aligned, pixles = unaligned | @ block = word aligned, pixles = unaligned | ||||
pld [r1] | pld [r1] | ||||
@@ -450,10 +450,10 @@ put_pixels8_y2_arm: | |||||
.word 2b | .word 2b | ||||
.word 3b | .word 3b | ||||
.word 4b | .word 4b | ||||
.endfunc | |||||
.align 8 | .align 8 | ||||
.global put_no_rnd_pixels8_y2_arm | |||||
put_no_rnd_pixels8_y2_arm: | |||||
function put_no_rnd_pixels8_y2_arm, export=1 | |||||
@ void func(uint8_t *block, const uint8_t *pixels, int line_size, int h) | @ void func(uint8_t *block, const uint8_t *pixels, int line_size, int h) | ||||
@ block = word aligned, pixles = unaligned | @ block = word aligned, pixles = unaligned | ||||
pld [r1] | pld [r1] | ||||
@@ -558,6 +558,7 @@ put_no_rnd_pixels8_y2_arm: | |||||
.word 2b | .word 2b | ||||
.word 3b | .word 3b | ||||
.word 4b | .word 4b | ||||
.endfunc | |||||
@ ---------------------------------------------------------------- | @ ---------------------------------------------------------------- | ||||
.macro RND_XY2_IT align | .macro RND_XY2_IT align | ||||
@@ -625,8 +626,7 @@ put_no_rnd_pixels8_y2_arm: | |||||
.endm | .endm | ||||
.align 8 | .align 8 | ||||
.global put_pixels8_xy2_arm | |||||
put_pixels8_xy2_arm: | |||||
function put_pixels8_xy2_arm, export=1 | |||||
@ void func(uint8_t *block, const uint8_t *pixels, int line_size, int h) | @ void func(uint8_t *block, const uint8_t *pixels, int line_size, int h) | ||||
@ block = word aligned, pixles = unaligned | @ block = word aligned, pixles = unaligned | ||||
pld [r1] | pld [r1] | ||||
@@ -659,10 +659,10 @@ put_pixels8_xy2_arm: | |||||
.word 0x02020202 | .word 0x02020202 | ||||
.word 0xFCFCFCFC >> 2 | .word 0xFCFCFCFC >> 2 | ||||
.word 0x0F0F0F0F | .word 0x0F0F0F0F | ||||
.endfunc | |||||
.align 8 | .align 8 | ||||
.global put_no_rnd_pixels8_xy2_arm | |||||
put_no_rnd_pixels8_xy2_arm: | |||||
function put_no_rnd_pixels8_xy2_arm, export=1 | |||||
@ void func(uint8_t *block, const uint8_t *pixels, int line_size, int h) | @ void func(uint8_t *block, const uint8_t *pixels, int line_size, int h) | ||||
@ block = word aligned, pixles = unaligned | @ block = word aligned, pixles = unaligned | ||||
pld [r1] | pld [r1] | ||||
@@ -695,3 +695,4 @@ put_no_rnd_pixels8_xy2_arm: | |||||
.word 0x01010101 | .word 0x01010101 | ||||
.word 0xFCFCFCFC >> 2 | .word 0xFCFCFCFC >> 2 | ||||
.word 0x0F0F0F0F | .word 0x0F0F0F0F | ||||
.endfunc |
@@ -24,6 +24,9 @@ | |||||
CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. | CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. | ||||
*/ | */ | ||||
#include "asm.S" | |||||
#define FIX_0_298631336 2446 | #define FIX_0_298631336 2446 | ||||
#define FIX_0_541196100 4433 | #define FIX_0_541196100 4433 | ||||
#define FIX_0_765366865 6270 | #define FIX_0_765366865 6270 | ||||
@@ -54,8 +57,7 @@ | |||||
.text | .text | ||||
.align | .align | ||||
.global j_rev_dct_ARM | |||||
j_rev_dct_ARM: | |||||
function j_rev_dct_ARM, export=1 | |||||
stmdb sp!, { r4 - r12, lr } @ all callee saved regs | stmdb sp!, { r4 - r12, lr } @ all callee saved regs | ||||
sub sp, sp, #4 @ reserve some space on the stack | sub sp, sp, #4 @ reserve some space on the stack | ||||
@@ -24,6 +24,8 @@ | |||||
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA | * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA | ||||
*/ | */ | ||||
#include "asm.S" | |||||
/* useful constants for the algorithm, they are save in __constant_ptr__ at */ | /* useful constants for the algorithm, they are save in __constant_ptr__ at */ | ||||
/* the end of the source code.*/ | /* the end of the source code.*/ | ||||
#define W1 22725 | #define W1 22725 | ||||
@@ -53,10 +55,8 @@ | |||||
.text | .text | ||||
.align | |||||
.global simple_idct_ARM | |||||
simple_idct_ARM: | |||||
function simple_idct_ARM, export=1 | |||||
@@ void simple_idct_ARM(int16_t *block) | @@ void simple_idct_ARM(int16_t *block) | ||||
@@ save stack for reg needed (take all of them), | @@ save stack for reg needed (take all of them), | ||||
@@ R0-R3 are scratch regs, so no need to save them, but R0 contains the pointer to block | @@ R0-R3 are scratch regs, so no need to save them, but R0 contains the pointer to block | ||||
@@ -21,6 +21,8 @@ | |||||
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA | * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA | ||||
*/ | */ | ||||
#include "asm.S" | |||||
#define W1 22725 /* cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5 */ | #define W1 22725 /* cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5 */ | ||||
#define W2 21407 /* cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5 */ | #define W2 21407 /* cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5 */ | ||||
#define W3 19266 /* cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5 */ | #define W3 19266 /* cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5 */ | ||||
@@ -41,10 +43,7 @@ w13: .long W13 | |||||
w26: .long W26 | w26: .long W26 | ||||
w57: .long W57 | w57: .long W57 | ||||
.align | |||||
.type idct_row_armv5te, %function | |||||
.func idct_row_armv5te | |||||
idct_row_armv5te: | |||||
function idct_row_armv5te | |||||
str lr, [sp, #-4]! | str lr, [sp, #-4]! | ||||
ldrd v1, [a1, #8] | ldrd v1, [a1, #8] | ||||
@@ -262,10 +261,7 @@ row_dc_only: | |||||
sub fp, fp, a4 | sub fp, fp, a4 | ||||
.endm | .endm | ||||
.align | |||||
.type idct_col_armv5te, %function | |||||
.func idct_col_armv5te | |||||
idct_col_armv5te: | |||||
function idct_col_armv5te | |||||
str lr, [sp, #-4]! | str lr, [sp, #-4]! | ||||
idct_col | idct_col | ||||
@@ -337,10 +333,7 @@ idct_col_armv5te: | |||||
ldr pc, [sp], #4 | ldr pc, [sp], #4 | ||||
.endfunc | .endfunc | ||||
.align | |||||
.type idct_col_put_armv5te, %function | |||||
.func idct_col_put_armv5te | |||||
idct_col_put_armv5te: | |||||
function idct_col_put_armv5te | |||||
str lr, [sp, #-4]! | str lr, [sp, #-4]! | ||||
idct_col | idct_col | ||||
@@ -457,10 +450,7 @@ idct_col_put_armv5te: | |||||
ldr pc, [sp], #4 | ldr pc, [sp], #4 | ||||
.endfunc | .endfunc | ||||
.align | |||||
.type idct_col_add_armv5te, %function | |||||
.func idct_col_add_armv5te | |||||
idct_col_add_armv5te: | |||||
function idct_col_add_armv5te | |||||
str lr, [sp, #-4]! | str lr, [sp, #-4]! | ||||
idct_col | idct_col | ||||
@@ -610,11 +600,7 @@ idct_col_add_armv5te: | |||||
ldr pc, [sp], #4 | ldr pc, [sp], #4 | ||||
.endfunc | .endfunc | ||||
.align | |||||
.global simple_idct_armv5te | |||||
.type simple_idct_armv5te, %function | |||||
.func simple_idct_armv5te | |||||
simple_idct_armv5te: | |||||
function simple_idct_armv5te, export=1 | |||||
stmfd sp!, {v1, v2, v3, v4, v5, v6, v7, fp, lr} | stmfd sp!, {v1, v2, v3, v4, v5, v6, v7, fp, lr} | ||||
bl idct_row_armv5te | bl idct_row_armv5te | ||||
@@ -646,11 +632,7 @@ simple_idct_armv5te: | |||||
ldmfd sp!, {v1, v2, v3, v4, v5, v6, v7, fp, pc} | ldmfd sp!, {v1, v2, v3, v4, v5, v6, v7, fp, pc} | ||||
.endfunc | .endfunc | ||||
.align | |||||
.global simple_idct_add_armv5te | |||||
.type simple_idct_add_armv5te, %function | |||||
.func simple_idct_add_armv5te | |||||
simple_idct_add_armv5te: | |||||
function simple_idct_add_armv5te, export=1 | |||||
stmfd sp!, {a1, a2, v1, v2, v3, v4, v5, v6, v7, fp, lr} | stmfd sp!, {a1, a2, v1, v2, v3, v4, v5, v6, v7, fp, lr} | ||||
mov a1, a3 | mov a1, a3 | ||||
@@ -685,11 +667,7 @@ simple_idct_add_armv5te: | |||||
ldmfd sp!, {v1, v2, v3, v4, v5, v6, v7, fp, pc} | ldmfd sp!, {v1, v2, v3, v4, v5, v6, v7, fp, pc} | ||||
.endfunc | .endfunc | ||||
.align | |||||
.global simple_idct_put_armv5te | |||||
.type simple_idct_put_armv5te, %function | |||||
.func simple_idct_put_armv5te | |||||
simple_idct_put_armv5te: | |||||
function simple_idct_put_armv5te, export=1 | |||||
stmfd sp!, {a1, a2, v1, v2, v3, v4, v5, v6, v7, fp, lr} | stmfd sp!, {a1, a2, v1, v2, v3, v4, v5, v6, v7, fp, lr} | ||||
mov a1, a3 | mov a1, a3 | ||||
@@ -21,6 +21,8 @@ | |||||
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA | * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA | ||||
*/ | */ | ||||
#include "asm.S" | |||||
#define W1 22725 /* cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5 */ | #define W1 22725 /* cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5 */ | ||||
#define W2 21407 /* cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5 */ | #define W2 21407 /* cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5 */ | ||||
#define W3 19266 /* cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5 */ | #define W3 19266 /* cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5 */ | ||||
@@ -190,10 +192,7 @@ w57: .long W57 | |||||
a1 = source | a1 = source | ||||
a2 = dest | a2 = dest | ||||
*/ | */ | ||||
.align | |||||
.type idct_row_armv6, %function | |||||
.func idct_row_armv6 | |||||
idct_row_armv6: | |||||
function idct_row_armv6 | |||||
str lr, [sp, #-4]! | str lr, [sp, #-4]! | ||||
ldr lr, [a1, #12] /* lr = row[7,5] */ | ldr lr, [a1, #12] /* lr = row[7,5] */ | ||||
@@ -245,10 +244,7 @@ idct_row_armv6: | |||||
a1 = source | a1 = source | ||||
a2 = dest | a2 = dest | ||||
*/ | */ | ||||
.align | |||||
.type idct_col_armv6, %function | |||||
.func idct_col_armv6 | |||||
idct_col_armv6: | |||||
function idct_col_armv6 | |||||
stmfd sp!, {a2, lr} | stmfd sp!, {a2, lr} | ||||
ldr a3, [a1] /* a3 = row[2,0] */ | ldr a3, [a1] /* a3 = row[2,0] */ | ||||
@@ -276,10 +272,7 @@ idct_col_armv6: | |||||
a2 = dest | a2 = dest | ||||
a3 = line size | a3 = line size | ||||
*/ | */ | ||||
.align | |||||
.type idct_col_put_armv6, %function | |||||
.func idct_col_put_armv6 | |||||
idct_col_put_armv6: | |||||
function idct_col_put_armv6 | |||||
stmfd sp!, {a2, a3, lr} | stmfd sp!, {a2, a3, lr} | ||||
ldr a3, [a1] /* a3 = row[2,0] */ | ldr a3, [a1] /* a3 = row[2,0] */ | ||||
@@ -309,10 +302,7 @@ idct_col_put_armv6: | |||||
a2 = dest | a2 = dest | ||||
a3 = line size | a3 = line size | ||||
*/ | */ | ||||
.align | |||||
.type idct_col_add_armv6, %function | |||||
.func idct_col_add_armv6 | |||||
idct_col_add_armv6: | |||||
function idct_col_add_armv6 | |||||
stmfd sp!, {a2, a3, lr} | stmfd sp!, {a2, a3, lr} | ||||
ldr a3, [a1] /* a3 = row[2,0] */ | ldr a3, [a1] /* a3 = row[2,0] */ | ||||
@@ -393,12 +383,8 @@ idct_col_add_armv6: | |||||
sub a1, a1, #(16*7) | sub a1, a1, #(16*7) | ||||
.endm | .endm | ||||
.align | |||||
.global ff_simple_idct_armv6 | |||||
.type ff_simple_idct_armv6, %function | |||||
.func ff_simple_idct_armv6 | |||||
/* void ff_simple_idct_armv6(DCTELEM *data); */ | /* void ff_simple_idct_armv6(DCTELEM *data); */ | ||||
ff_simple_idct_armv6: | |||||
function ff_simple_idct_armv6, export=1 | |||||
stmfd sp!, {v1, v2, v3, v4, v5, v6, v7, fp, lr} | stmfd sp!, {v1, v2, v3, v4, v5, v6, v7, fp, lr} | ||||
sub sp, sp, #128 | sub sp, sp, #128 | ||||
@@ -412,12 +398,8 @@ ff_simple_idct_armv6: | |||||
ldmfd sp!, {v1, v2, v3, v4, v5, v6, v7, fp, pc} | ldmfd sp!, {v1, v2, v3, v4, v5, v6, v7, fp, pc} | ||||
.endfunc | .endfunc | ||||
.align | |||||
.global ff_simple_idct_add_armv6 | |||||
.type ff_simple_idct_add_armv6, %function | |||||
.func ff_simple_idct_add_armv6 | |||||
/* ff_simple_idct_add_armv6(uint8_t *dest, int line_size, DCTELEM *data); */ | /* ff_simple_idct_add_armv6(uint8_t *dest, int line_size, DCTELEM *data); */ | ||||
ff_simple_idct_add_armv6: | |||||
function ff_simple_idct_add_armv6, export=1 | |||||
stmfd sp!, {a1, a2, v1, v2, v3, v4, v5, v6, v7, fp, lr} | stmfd sp!, {a1, a2, v1, v2, v3, v4, v5, v6, v7, fp, lr} | ||||
sub sp, sp, #128 | sub sp, sp, #128 | ||||
@@ -433,12 +415,8 @@ ff_simple_idct_add_armv6: | |||||
ldmfd sp!, {v1, v2, v3, v4, v5, v6, v7, fp, pc} | ldmfd sp!, {v1, v2, v3, v4, v5, v6, v7, fp, pc} | ||||
.endfunc | .endfunc | ||||
.align | |||||
.global ff_simple_idct_put_armv6 | |||||
.type ff_simple_idct_put_armv6, %function | |||||
.func ff_simple_idct_put_armv6 | |||||
/* ff_simple_idct_put_armv6(uint8_t *dest, int line_size, DCTELEM *data); */ | /* ff_simple_idct_put_armv6(uint8_t *dest, int line_size, DCTELEM *data); */ | ||||
ff_simple_idct_put_armv6: | |||||
function ff_simple_idct_put_armv6, export=1 | |||||
stmfd sp!, {a1, a2, v1, v2, v3, v4, v5, v6, v7, fp, lr} | stmfd sp!, {a1, a2, v1, v2, v3, v4, v5, v6, v7, fp, lr} | ||||
sub sp, sp, #128 | sub sp, sp, #128 | ||||