cpet/freebsd-ports
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Pull in upstream commits to our libx264 to deal with the clang 3.5

Browse Source 
-integrated-as defaults for ARMv6 compilation.

Handling for koobs@ while he's on vacation.

PR:		196591
Submitted by:	mikael.urankar@gmail.com
Approved by:	mentor (implicit)
This commit is contained in:
Sean Bruno 2015-01-07 18:54:54 +00:00
parent 201b450d44
commit 99a42b0dad
Notes: svn2git 2021-03-31 03:12:20 +00:00 
svn path=/head/; revision=376488
10 changed files with 1356 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
multimedia/libx264/Makefile
View File

@ -5,7 +5,7 @@
PORTNAME= libx264
PORTVERSION= 0.${X264_BUILD}.${X264_REV}
PORTREVISION= 1
PORTREVISION= 2
CATEGORIES= multimedia
MASTER_SITES= ftp://ftp.videolan.org/pub/videolan/x264/snapshots/ \
http://samples.mplayerhq.hu/yuv4mpeg2/:pgo

40
multimedia/libx264/files/patch-common_arm_asm.S Normal file
View File

@ -0,0 +1,40 @@
--- common/arm/asm.S.orig 2014-08-27 20:45:08 UTC
+++ common/arm/asm.S
@@ -40,6 +40,12 @@
# define ELF @
#endif
+#if HAVE_AS_FUNC
+# define FUNC
+#else
+# define FUNC @
+#endif
+
.macro require8, val=1
ELF .eabi_attribute 24, \val
.endm
@@ -49,17 +55,22 @@ ELF .eabi_attribute 25, \val
.endm
.macro function name, export=1
+ .macro endfunc
+ELF .size \name, . - \name
+FUNC .endfunc
+ .purgem endfunc
+ .endm
.align 2
.if \export == 1
.global EXTERN_ASM\name
ELF .hidden EXTERN_ASM\name
ELF .type EXTERN_ASM\name, %function
- .func EXTERN_ASM\name
+FUNC .func EXTERN_ASM\name
EXTERN_ASM\name:
.else
ELF .hidden \name
ELF .type \name, %function
- .func \name
+FUNC .func \name
\name:
.endif
.endm

34
multimedia/libx264/files/patch-common_arm_cpu-a.S Normal file
View File

@ -0,0 +1,34 @@
--- common/arm/cpu-a.S.orig 2014-08-27 20:45:08 UTC
+++ common/arm/cpu-a.S
@@ -33,7 +33,7 @@
function x264_cpu_neon_test
vadd.i16 q0, q0, q0
bx lr
-.endfunc
+endfunc
// return: 0 on success
// 1 if counters were already enabled
@@ -49,14 +49,14 @@ function x264_cpu_enable_armv7_counter,
mov r2, #1 << 31 // enable cycle counter
mcr p15, 0, r2, c9, c12, 1 // write CNTENS
bx lr
-.endfunc
+endfunc
function x264_cpu_disable_armv7_counter, export=0
mrc p15, 0, r0, c9, c12, 0 // read PMNC
bic r0, r0, #1 // disable counters
mcr p15, 0, r0, c9, c12, 0 // write PMNC
bx lr
-.endfunc
+endfunc
.macro READ_TIME r
@@ -106,4 +106,4 @@ average_loop:
cmp r0, #10
movgt r0, #0
pop {r4-r6,pc}
-.endfunc
+endfunc

161
multimedia/libx264/files/patch-common_arm_dct-a.S Normal file
View File

@ -0,0 +1,161 @@
--- common/arm/dct-a.S.orig 2014-08-27 20:45:08 UTC
+++ common/arm/dct-a.S
@@ -82,7 +82,7 @@ function x264_dct4x4dc_neon
vrhadd.s16 d3, d6, d7
vst1.64 {d0-d3}, [r0,:128]
bx lr
-.endfunc
+endfunc
function x264_idct4x4dc_neon
vld1.64 {d0-d3}, [r0,:128]
@@ -94,7 +94,7 @@ function x264_idct4x4dc_neon
HADAMARD 2, sumsub, d3, d2, d6, d7
vst1.64 {d0-d3}, [r0,:128]
bx lr
-.endfunc
+endfunc
.macro DCT_1D d0 d1 d2 d3 d4 d5 d6 d7
@@ -129,7 +129,7 @@ function x264_sub4x4_dct_neon
DCT_1D d4, d5, d6, d7, d0, d1, d2, d3
vst1.64 {d4-d7}, [r0,:128]
bx lr
-.endfunc
+endfunc
function x264_sub8x4_dct_neon, export=0
vld1.64 {d0}, [r1,:64], r3
@@ -165,7 +165,7 @@ function x264_sub8x4_dct_neon, export=0
vst1.64 {d4-d5}, [r0,:128]!
vst1.64 {d6-d7}, [r0,:128]!
bx lr
-.endfunc
+endfunc
function x264_sub8x8_dct_neon
push {lr}
@@ -174,7 +174,7 @@ function x264_sub8x8_dct_neon
bl x264_sub8x4_dct_neon
pop {lr}
b x264_sub8x4_dct_neon
-.endfunc
+endfunc
function x264_sub16x16_dct_neon
push {lr}
@@ -195,7 +195,7 @@ function x264_sub16x16_dct_neon
bl x264_sub8x4_dct_neon
pop {lr}
b x264_sub8x4_dct_neon
-.endfunc
+endfunc
.macro DCT8_1D type
@@ -279,7 +279,7 @@ function x264_sub8x8_dct8_neon
vst1.64 {d24-d27}, [r0,:128]!
vst1.64 {d28-d31}, [r0,:128]!
bx lr
-.endfunc
+endfunc
function x264_sub16x16_dct8_neon
push {lr}
@@ -294,7 +294,7 @@ function x264_sub16x16_dct8_neon
sub r1, r1, #FENC_STRIDE*8 - 8
sub r2, r2, #FDEC_STRIDE*8 - 8
b X(x264_sub8x8_dct8_neon)
-.endfunc
+endfunc
// First part of IDCT (minus final SUMSUB_BA)
@@ -336,7 +336,7 @@ function x264_add4x4_idct_neon
vst1.32 {d2[1]}, [r0,:32], r2
vst1.32 {d2[0]}, [r0,:32], r2
bx lr
-.endfunc
+endfunc
function x264_add8x4_idct_neon, export=0
vld1.64 {d0-d3}, [r1,:128]!
@@ -376,7 +376,7 @@ function x264_add8x4_idct_neon, export=0
vst1.32 {d2}, [r0,:64], r2
vst1.32 {d3}, [r0,:64], r2
bx lr
-.endfunc
+endfunc
function x264_add8x8_idct_neon
mov r2, #FDEC_STRIDE
@@ -384,7 +384,7 @@ function x264_add8x8_idct_neon
bl x264_add8x4_idct_neon
mov lr, ip
b x264_add8x4_idct_neon
-.endfunc
+endfunc
function x264_add16x16_idct_neon
mov r2, #FDEC_STRIDE
@@ -401,7 +401,7 @@ function x264_add16x16_idct_neon
bl x264_add8x4_idct_neon
mov lr, ip
b x264_add8x4_idct_neon
-.endfunc
+endfunc
.macro IDCT8_1D type
@@ -498,7 +498,7 @@ function x264_add8x8_idct8_neon
vst1.64 {d6}, [r0,:64], r2
vst1.64 {d7}, [r0,:64], r2
bx lr
-.endfunc
+endfunc
function x264_add16x16_idct8_neon
mov ip, lr
@@ -510,7 +510,7 @@ function x264_add16x16_idct8_neon
sub r0, r0, #8*FDEC_STRIDE-8
mov lr, ip
b X(x264_add8x8_idct8_neon)
-.endfunc
+endfunc
function x264_add8x8_idct_dc_neon
@@ -562,7 +562,7 @@ function x264_add8x8_idct_dc_neon
vst1.64 {d6}, [r0,:64], r2
vst1.64 {d7}, [r0,:64], r2
bx lr
-.endfunc
+endfunc
.macro ADD16x4_IDCT_DC dc
vld1.64 {d16-d17}, [r0,:128], r3
@@ -610,7 +610,7 @@ function x264_add16x16_idct_dc_neon
ADD16x4_IDCT_DC d2
ADD16x4_IDCT_DC d3
bx lr
-.endfunc
+endfunc
function x264_sub8x8_dct_dc_neon
mov r3, #FENC_STRIDE
@@ -658,7 +658,7 @@ function x264_sub8x8_dct_dc_neon
vpadd.s16 d0, d0, d1
vst1.64 {d0}, [r0,:64]
bx lr
-.endfunc
+endfunc
function x264_zigzag_scan_4x4_frame_neon
@@ -671,4 +671,4 @@ function x264_zigzag_scan_4x4_frame_neon
vtbl.8 d7, {d2-d3}, d19
vst1.64 {d4-d7}, [r0,:128]
bx lr
-.endfunc
+endfunc

44
multimedia/libx264/files/patch-common_arm_deblock-a.S Normal file
View File

@ -0,0 +1,44 @@
--- common/arm/deblock-a.S.orig 2014-08-27 20:45:08 UTC
+++ common/arm/deblock-a.S
@@ -142,7 +142,7 @@ function x264_deblock_v_luma_neon
align_pop_regs
bx lr
-.endfunc
+endfunc
function x264_deblock_h_luma_neon
h264_loop_filter_start
@@ -194,7 +194,7 @@ function x264_deblock_h_luma_neon
align_pop_regs
bx lr
-.endfunc
+endfunc
.macro h264_loop_filter_chroma
vdup.8 q11, r2 // alpha
@@ -255,7 +255,7 @@ function x264_deblock_v_chroma_neon
vst2.8 {d0, d1}, [r0,:128], r1
bx lr
-.endfunc
+endfunc
function x264_deblock_h_chroma_neon
h264_loop_filter_start
@@ -303,7 +303,7 @@ function x264_deblock_h_chroma_neon
vst1.8 {d3}, [r0], r1
bx lr
-.endfunc
+endfunc
function x264_deblock_strength_neon
ldr ip, [sp]
@@ -409,4 +409,4 @@ lists:
vst1.8 {q8}, [r3,:128] @ bs[0]
bx lr
-.endfunc
+endfunc

386
multimedia/libx264/files/patch-common_arm_mc-a.S Normal file
View File

@ -0,0 +1,386 @@
--- common/arm/mc-a.S.orig 2014-08-27 20:45:08 UTC
+++ common/arm/mc-a.S
@@ -50,7 +50,7 @@ function x264_prefetch_ref_arm
pld [r3, r1, lsl #1]
pld [r3, r2]
bx lr
-.endfunc
+endfunc
// void prefetch_fenc( uint8_t *pix_y, intptr_t stride_y,
// uint8_t *pix_uv, intptr_t stride_uv, int mb_x )
@@ -76,7 +76,7 @@ function x264_prefetch_fenc_arm
pld [ip]
pld [ip, r3]
pop {pc}
-.endfunc
+endfunc
// void *x264_memcpy_aligned( void *dst, const void *src, size_t n )
@@ -85,7 +85,7 @@ function x264_memcpy_aligned_neon
movrel ip, memcpy_table
and r3, r3, #0xc
ldr pc, [ip, r3]
-.endfunc
+endfunc
.macro MEMCPY_ALIGNED srcalign dstalign
function memcpy_aligned_\dstalign\()_\srcalign\()_neon, export=0
@@ -127,7 +127,7 @@ function memcpy_aligned_\dstalign\()_\sr
vst1.64 {d0}, [r3,:64]!
.endif
bx lr
-.endfunc
+endfunc
.endm
MEMCPY_ALIGNED 16, 16
@@ -156,7 +156,7 @@ memzero_loop:
.endr
bgt memzero_loop
bx lr
-.endfunc
+endfunc
// void pixel_avg( uint8_t *dst, intptr_t dst_stride,
@@ -175,7 +175,7 @@ function x264_pixel_avg_\w\()x\h\()_neon
cmp ip, #0
bge x264_pixel_avg_weight_w\w\()_add_add_neon
b x264_pixel_avg_weight_w\w\()_sub_add_neon // weight < 0
-.endfunc
+endfunc
.endm
AVGH 4, 2
@@ -253,7 +253,7 @@ function x264_pixel_avg_weight_w4_\ext\(
vst1.32 {d1[0]}, [r0,:32], r1
bgt 1b
pop {r4-r6,pc}
-.endfunc
+endfunc
function x264_pixel_avg_weight_w8_\ext\()_neon, export=0
load_weights_\ext
@@ -277,7 +277,7 @@ function x264_pixel_avg_weight_w8_\ext\(
vst1.64 {d3}, [r0,:64], r1
bgt 1b
pop {r4-r6,pc}
-.endfunc
+endfunc
function x264_pixel_avg_weight_w16_\ext\()_neon, export=0
load_weights_\ext
@@ -297,7 +297,7 @@ function x264_pixel_avg_weight_w16_\ext\
vst1.64 {d2-d3}, [r0,:128], r1
bgt 1b
pop {r4-r6,pc}
-.endfunc
+endfunc
.endm
AVG_WEIGHT add_add
@@ -316,7 +316,7 @@ function x264_pixel_avg_w4_neon, export=
vst1.32 {d1[0]}, [r0,:32], r1
bgt x264_pixel_avg_w4_neon
pop {r4-r6,pc}
-.endfunc
+endfunc
function x264_pixel_avg_w8_neon, export=0
subs lr, lr, #4
@@ -338,7 +338,7 @@ function x264_pixel_avg_w8_neon, export=
vst1.64 {d3}, [r0,:64], r1
bgt x264_pixel_avg_w8_neon
pop {r4-r6,pc}
-.endfunc
+endfunc
function x264_pixel_avg_w16_neon, export=0
subs lr, lr, #4
@@ -360,7 +360,7 @@ function x264_pixel_avg_w16_neon, export
vst1.64 {d6-d7}, [r0,:128], r1
bgt x264_pixel_avg_w16_neon
pop {r4-r6,pc}
-.endfunc
+endfunc
function x264_pixel_avg2_w4_neon
@@ -379,7 +379,7 @@ avg2_w4_loop:
vst1.32 {d1[0]}, [r0,:32], r1
bgt avg2_w4_loop
pop {pc}
-.endfunc
+endfunc
function x264_pixel_avg2_w8_neon
ldr ip, [sp, #4]
@@ -397,7 +397,7 @@ avg2_w8_loop:
vst1.64 {d1}, [r0,:64], r1
bgt avg2_w8_loop
pop {pc}
-.endfunc
+endfunc
function x264_pixel_avg2_w16_neon
ldr ip, [sp, #4]
@@ -415,7 +415,7 @@ avg2_w16_loop:
vst1.64 {d4-d5}, [r0,:128], r1
bgt avg2_w16_loop
pop {pc}
-.endfunc
+endfunc
function x264_pixel_avg2_w20_neon
ldr ip, [sp, #4]
@@ -438,7 +438,7 @@ avg2_w20_loop:
vst1.32 {d6[0]}, [r0,:32], r1
bgt avg2_w20_loop
pop {pc}
-.endfunc
+endfunc
.macro weight_prologue type
@@ -499,7 +499,7 @@ weight20_loop:
vst1.32 {d20[1]}, [r0,:32], r1
bgt weight20_loop
pop {r4-r5,pc}
-.endfunc
+endfunc
function x264_mc_weight_w16_neon
weight_prologue full
@@ -531,7 +531,7 @@ weight16_loop:
vst1.8 {d18-d19}, [r0,:128], r1
bgt weight16_loop
pop {r4-r5,pc}
-.endfunc
+endfunc
function x264_mc_weight_w8_neon
weight_prologue full
@@ -553,7 +553,7 @@ weight8_loop:
vst1.8 {d18}, [r0,:64], r1
bgt weight8_loop
pop {r4-r5,pc}
-.endfunc
+endfunc
function x264_mc_weight_w4_neon
weight_prologue full
@@ -572,7 +572,7 @@ weight4_loop:
vst1.32 {d16[1]}, [r0,:32], r1
bgt weight4_loop
pop {r4-r5,pc}
-.endfunc
+endfunc
function x264_mc_weight_w20_nodenom_neon
weight_prologue nodenom
@@ -609,7 +609,7 @@ weight20_nodenom_loop:
vst1.32 {d20[1]}, [r0,:32], r1
bgt weight20_nodenom_loop
pop {r4-r5,pc}
-.endfunc
+endfunc
function x264_mc_weight_w16_nodenom_neon
weight_prologue nodenom
@@ -637,7 +637,7 @@ weight16_nodenom_loop:
vst1.8 {d18-d19}, [r0,:128], r1
bgt weight16_nodenom_loop
pop {r4-r5,pc}
-.endfunc
+endfunc
function x264_mc_weight_w8_nodenom_neon
weight_prologue nodenom
@@ -657,7 +657,7 @@ weight8_nodenom_loop:
vst1.8 {d17}, [r0,:64], r1
bgt weight8_nodenom_loop
pop {r4-r5,pc}
-.endfunc
+endfunc
function x264_mc_weight_w4_nodenom_neon
weight_prologue nodenom
@@ -675,7 +675,7 @@ weight4_nodenom_loop:
vst1.32 {d16[1]}, [r0,:32], r1
bgt weight4_nodenom_loop
pop {r4-r5,pc}
-.endfunc
+endfunc
.macro weight_simple_prologue
push {lr}
@@ -699,7 +699,7 @@ weight20_\name\()_loop:
vst1.8 {d19-d21}, [r0,:64], r1
bgt weight20_\name\()_loop
pop {pc}
-.endfunc
+endfunc
function x264_mc_weight_w16_\name\()_neon
weight_simple_prologue
@@ -713,7 +713,7 @@ weight16_\name\()_loop:
vst1.8 {d18-d19}, [r0,:128], r1
bgt weight16_\name\()_loop
pop {pc}
-.endfunc
+endfunc
function x264_mc_weight_w8_\name\()_neon
weight_simple_prologue
@@ -726,7 +726,7 @@ weight8_\name\()_loop:
vst1.8 {d17}, [r0,:64], r1
bgt weight8_\name\()_loop
pop {pc}
-.endfunc
+endfunc
function x264_mc_weight_w4_\name\()_neon
weight_simple_prologue
@@ -739,7 +739,7 @@ weight4_\name\()_loop:
vst1.32 {d17[0]}, [r0,:32], r1
bgt weight4_\name\()_loop
pop {pc}
-.endfunc
+endfunc
.endm
weight_simple offsetadd, vqadd.u8
@@ -761,7 +761,7 @@ copy_w4_loop:
vst1.32 {d3[0]}, [r0,:32], r1
bgt copy_w4_loop
bx lr
-.endfunc
+endfunc
function x264_mc_copy_w8_neon
ldr ip, [sp]
@@ -777,7 +777,7 @@ copy_w8_loop:
vst1.32 {d3}, [r0,:64], r1
bgt copy_w8_loop
bx lr
-.endfunc
+endfunc
function x264_mc_copy_w16_neon
ldr ip, [sp]
@@ -793,7 +793,7 @@ copy_w16_loop:
vst1.32 {d6-d7}, [r0,:128], r1
bgt copy_w16_loop
bx lr
-.endfunc
+endfunc
function x264_mc_copy_w16_aligned_neon
ldr ip, [sp]
@@ -809,7 +809,7 @@ copy_w16_aligned_loop:
vst1.32 {d6-d7}, [r0,:128], r1
bgt copy_w16_aligned_loop
bx lr
-.endfunc
+endfunc
// void x264_mc_chroma_neon( uint8_t *dst, intptr_t i_dst_stride,
@@ -1159,7 +1159,7 @@ mc_chroma_w8:
vpop {d8-d11}
pop {r4-r8, pc}
-.endfunc
+endfunc
// hpel_filter_v( uint8_t *dst, uint8_t *src, int16_t *buf, intptr_t stride, int width )
@@ -1200,7 +1200,7 @@ filter_v_loop:
vst1.64 {d0-d1}, [r0,:128]!
bgt filter_v_loop
pop {pc}
-.endfunc
+endfunc
// hpel_filter_c( uint8_t *dst, int16_t *buf, int width );
function x264_hpel_filter_c_neon
@@ -1285,7 +1285,7 @@ filter_c_loop:
vst1.64 {d30-d31}, [r0,:128]!
bgt filter_c_loop
bx lr
-.endfunc
+endfunc
// hpel_filter_h( uint8_t *dst, uint8_t *src, int width );
function x264_hpel_filter_h_neon
@@ -1372,7 +1372,7 @@ filter_h_loop:
vst1.64 {d6-d7}, [r0,:128]!
bgt filter_h_loop
bx lr
-.endfunc
+endfunc
// frame_init_lowres_core( uint8_t *src0, uint8_t *dst0, uint8_t *dsth, uint8_t *dstv,
@@ -1464,7 +1464,7 @@ lowres_xloop_end:
vpop {d8-d15}
pop {r4-r10,pc}
-.endfunc
+endfunc
function x264_load_deinterleave_chroma_fdec_neon
mov ip, #FDEC_STRIDE/2
@@ -1477,7 +1477,7 @@ function x264_load_deinterleave_chroma_f
bgt 1b
bx lr
-.endfunc
+endfunc
function x264_load_deinterleave_chroma_fenc_neon
mov ip, #FENC_STRIDE/2
@@ -1490,7 +1490,7 @@ function x264_load_deinterleave_chroma_f
bgt 1b
bx lr
-.endfunc
+endfunc
function x264_plane_copy_deinterleave_neon
push {r4-r7, lr}
@@ -1516,7 +1516,7 @@ block:
bgt block
pop {r4-r7, pc}
-.endfunc
+endfunc
function x264_plane_copy_deinterleave_rgb_neon
push {r4-r8, r10, r11, lr}
@@ -1568,7 +1568,7 @@ block4:
bgt block4
pop {r4-r8, r10, r11, pc}
-.endfunc
+endfunc
function x264_plane_copy_interleave_neon
push {r4-r7, lr}
@@ -1595,7 +1595,7 @@ blocki:
bgt blocki
pop {r4-r7, pc}
-.endfunc
+endfunc
function x264_store_interleave_chroma_neon
push {lr}
@@ -1609,4 +1609,4 @@ function x264_store_interleave_chroma_ne
bgt 1b
pop {pc}
-.endfunc
+endfunc

275
multimedia/libx264/files/patch-common_arm_pixel-a.S Normal file
View File

@ -0,0 +1,275 @@
--- common/arm/pixel-a.S.orig 2014-08-27 20:45:08 UTC
+++ common/arm/pixel-a.S
@@ -29,10 +29,12 @@
.section .rodata
.align 4
-.rept 16 .byte 0xff
+.rept 16
+ .byte 0xff
.endr
mask_ff:
-.rept 16 .byte 0
+.rept 16
+ .byte 0
.endr
mask_ac4:
@@ -60,7 +62,7 @@ function x264_pixel_sad_4x\h\()_armv6
.endr
usada8 r0, r6, lr, ip
pop {r4-r6,pc}
-.endfunc
+endfunc
.endm
SAD4_ARMV6 4
@@ -137,7 +139,7 @@ function x264_pixel_sad\name\()_\w\()x\h
vpaddl.u16 d0, d0
vmov.u32 r0, d0[0]
bx lr
-.endfunc
+endfunc
.endm
SAD_FUNC 4, 4
@@ -222,7 +224,7 @@ function x264_pixel_sad_aligned_\w\()x\h
vpaddl.u16 d0, d0
vmov.u32 r0, d0[0]
bx lr
-.endfunc
+endfunc
.endm
SAD_FUNC_DUAL 8, 4
@@ -368,7 +370,7 @@ function x264_pixel_sad_x\x\()_\w\()x\h\
vst1.32 {d0-d1}, [r7]
.endif
pop {r6-r7,pc}
-.endfunc
+endfunc
.endm
SAD_X_FUNC 3, 4, 4
@@ -477,7 +479,7 @@ function x264_pixel_ssd_\w\()x\h\()_neon
vpadd.s32 d0, d0, d0
vmov.32 r0, d0[0]
bx lr
-.endfunc
+endfunc
.endm
SSD_FUNC 4, 4
@@ -517,7 +519,7 @@ function x264_pixel_var_8x8_neon
vld1.64 {d26}, [r0,:64], r1
VAR_SQR_SUM q2, q10, q15, d26
b x264_var_end
-.endfunc
+endfunc
function x264_pixel_var_8x16_neon
vld1.64 {d16}, [r0,:64], r1
@@ -549,7 +551,7 @@ function x264_pixel_var_8x16_neon
2:
VAR_SQR_SUM q2, q13, q15, d22
b x264_var_end
-.endfunc
+endfunc
function x264_pixel_var_16x16_neon
vld1.64 {d16-d17}, [r0,:128], r1
@@ -573,7 +575,7 @@ var16_loop:
VAR_SQR_SUM q1, q12, q14, d18
VAR_SQR_SUM q2, q13, q15, d19
bgt var16_loop
-.endfunc
+endfunc
function x264_var_end, export=0
vpaddl.u16 q8, q14
@@ -588,7 +590,7 @@ function x264_var_end, export=0
vmov r0, r1, d0
bx lr
-.endfunc
+endfunc
.macro DIFF_SUM diff da db lastdiff
vld1.64 {\da}, [r0,:64], r1
@@ -633,7 +635,7 @@ function x264_pixel_var2_8x8_neon
mul r0, r0, r0
sub r0, r1, r0, lsr #6
bx lr
-.endfunc
+endfunc
function x264_pixel_var2_8x16_neon
vld1.64 {d16}, [r0,:64], r1
@@ -677,7 +679,7 @@ function x264_pixel_var2_8x16_neon
mul r0, r0, r0
sub r0, r1, r0, lsr #7
bx lr
-.endfunc
+endfunc
.macro LOAD_DIFF_8x4 q0 q1 q2 q3
vld1.32 {d1}, [r2], r3
@@ -714,7 +716,7 @@ function x264_pixel_satd_4x4_neon
HORIZ_ADD d0, d0, d1
vmov.32 r0, d0[0]
bx lr
-.endfunc
+endfunc
function x264_pixel_satd_4x8_neon
vld1.32 {d1[]}, [r2], r3
@@ -741,7 +743,7 @@ function x264_pixel_satd_4x8_neon
vsubl.u8 q3, d6, d7
SUMSUB_AB q10, q11, q2, q3
b x264_satd_4x8_8x4_end_neon
-.endfunc
+endfunc
function x264_pixel_satd_8x4_neon
vld1.64 {d1}, [r2], r3
@@ -758,7 +760,7 @@ function x264_pixel_satd_8x4_neon
vld1.64 {d6}, [r0,:64], r1
vsubl.u8 q3, d6, d7
SUMSUB_AB q10, q11, q2, q3
-.endfunc
+endfunc
function x264_satd_4x8_8x4_end_neon, export=0
vadd.s16 q0, q8, q10
@@ -785,7 +787,7 @@ function x264_satd_4x8_8x4_end_neon, exp
HORIZ_ADD d0, d0, d1
vmov.32 r0, d0[0]
bx lr
-.endfunc
+endfunc
function x264_pixel_satd_8x8_neon
mov ip, lr
@@ -799,7 +801,7 @@ function x264_pixel_satd_8x8_neon
mov lr, ip
vmov.32 r0, d0[0]
bx lr
-.endfunc
+endfunc
function x264_pixel_satd_8x16_neon
vpush {d8-d11}
@@ -821,7 +823,7 @@ function x264_pixel_satd_8x16_neon
mov lr, ip
vmov.32 r0, d0[0]
bx lr
-.endfunc
+endfunc
function x264_satd_8x8_neon, export=0
LOAD_DIFF_8x4 q8, q9, q10, q11
@@ -841,7 +843,7 @@ function x264_satd_8x8_neon, export=0
SUMSUB_AB q9, q11, q1, q3
vld1.64 {d0}, [r0,:64], r1
vsubl.u8 q15, d0, d1
-.endfunc
+endfunc
// one vertical hadamard pass and two horizontal
function x264_satd_8x4v_8x8h_neon, export=0
@@ -870,7 +872,7 @@ function x264_satd_8x4v_8x8h_neon, expor
vmax.s16 q14, q8, q10
vmax.s16 q15, q9, q11
bx lr
-.endfunc
+endfunc
function x264_pixel_satd_16x8_neon
vpush {d8-d11}
@@ -892,7 +894,7 @@ function x264_pixel_satd_16x8_neon
mov lr, ip
vmov.32 r0, d0[0]
bx lr
-.endfunc
+endfunc
function x264_pixel_satd_16x16_neon
vpush {d8-d11}
@@ -926,7 +928,7 @@ function x264_pixel_satd_16x16_neon
mov lr, ip
vmov.32 r0, d0[0]
bx lr
-.endfunc
+endfunc
function x264_satd_16x4_neon, export=0
vld1.64 {d2-d3}, [r2], r3
@@ -950,7 +952,7 @@ function x264_satd_16x4_neon, export=0
SUMSUB_AB q2, q3, q10, q11
SUMSUB_ABCD q8, q10, q9, q11, q0, q2, q1, q3
b x264_satd_8x4v_8x8h_neon
-.endfunc
+endfunc
function x264_pixel_sa8d_8x8_neon
@@ -963,7 +965,7 @@ function x264_pixel_sa8d_8x8_neon
add r0, r0, #1
lsr r0, r0, #1
bx lr
-.endfunc
+endfunc
function x264_pixel_sa8d_16x16_neon
vpush {d8-d11}
@@ -995,7 +997,7 @@ function x264_pixel_sa8d_16x16_neon
add r0, r0, #1
lsr r0, r0, #1
bx lr
-.endfunc
+endfunc
.macro HADAMARD4_V r1, r2, r3, r4, t1, t2, t3, t4
SUMSUB_ABCD \t1, \t2, \t3, \t4, \r1, \r2, \r3, \r4
@@ -1058,7 +1060,7 @@ function x264_sa8d_8x8_neon, export=0
vadd.i16 q8, q8, q9
vadd.i16 q9, q10, q11
bx lr
-.endfunc
+endfunc
.macro HADAMARD_AC w h
@@ -1094,7 +1096,7 @@ function x264_pixel_hadamard_ac_\w\()x\h
lsr r0, r0, #1
lsr r1, r1, #2
bx lr
-.endfunc
+endfunc
.endm
HADAMARD_AC 8, 8
@@ -1189,7 +1191,7 @@ function x264_hadamard_ac_8x8_neon, expo
vadd.s16 q2, q2, q14
vpadal.u16 q5, q2
bx lr
-.endfunc
+endfunc
.macro SSIM_ITER n ssa s12 ssb lastssa lasts12 lastssb da db dnext
@@ -1243,7 +1245,7 @@ function x264_pixel_ssim_4x4x2_core_neon
vst4.32 {d0-d3}, [ip]
bx lr
-.endfunc
+endfunc
// FIXME: see about doing 16x16 -> 32 bit multiplies for s1/s2
function x264_pixel_ssim_end4_neon
@@ -1314,4 +1316,4 @@ ssim_skip:
vpadd.f32 d0, d0, d0
vmov.32 r0, d0[0]
bx lr
-.endfunc
+endfunc

242
multimedia/libx264/files/patch-common_arm_predict-a.S Normal file
View File

@ -0,0 +1,242 @@
--- common/arm/predict-a.S.orig 2014-08-27 20:45:08 UTC
+++ common/arm/predict-a.S
@@ -77,7 +77,7 @@ function x264_predict_4x4_h_armv6
add ip, ip, ip, lsl #16
str ip, [r0, #3*FDEC_STRIDE]
bx lr
-.endfunc
+endfunc
function x264_predict_4x4_v_armv6
ldr r1, [r0, #0 - 1 * FDEC_STRIDE]
@@ -86,7 +86,7 @@ function x264_predict_4x4_v_armv6
str r1, [r0, #0 + 2 * FDEC_STRIDE]
str r1, [r0, #0 + 3 * FDEC_STRIDE]
bx lr
-.endfunc
+endfunc
function x264_predict_4x4_dc_armv6
mov ip, #0
@@ -109,7 +109,7 @@ function x264_predict_4x4_dc_armv6
str r1, [r0, #2*FDEC_STRIDE]
str r1, [r0, #3*FDEC_STRIDE]
bx lr
-.endfunc
+endfunc
function x264_predict_4x4_dc_top_neon
mov r12, #FDEC_STRIDE
@@ -124,7 +124,7 @@ function x264_predict_4x4_dc_top_neon
vst1.32 d1[0], [r0,:32], r12
vst1.32 d1[0], [r0,:32], r12
bx lr
-.endfunc
+endfunc
// return a1 = (a1+2*b1+c1+2)>>2 a2 = (a2+2*b2+c2+2)>>2
.macro PRED4x4_LOWPASS a1 b1 c1 a2 b2 c2 pb_1
@@ -167,7 +167,7 @@ function x264_predict_4x4_ddr_armv6
add r5, r5, r4, lsr #8
str r5, [r0, #3*FDEC_STRIDE]
pop {r4-r6,pc}
-.endfunc
+endfunc
function x264_predict_4x4_ddl_neon
sub r0, #FDEC_STRIDE
@@ -186,7 +186,7 @@ function x264_predict_4x4_ddl_neon
vst1.32 {d2[0]}, [r0,:32], ip
vst1.32 {d3[0]}, [r0,:32], ip
bx lr
-.endfunc
+endfunc
function x264_predict_8x8_dc_neon
mov ip, #0
@@ -210,7 +210,7 @@ function x264_predict_8x8_dc_neon
vst1.64 {d0}, [r0,:64], ip
.endr
pop {r4-r5,pc}
-.endfunc
+endfunc
function x264_predict_8x8_h_neon
add r1, r1, #7
@@ -233,7 +233,7 @@ function x264_predict_8x8_h_neon
vst1.64 {d6}, [r0,:64], ip
vst1.64 {d7}, [r0,:64], ip
bx lr
-.endfunc
+endfunc
function x264_predict_8x8_v_neon
add r1, r1, #16
@@ -243,7 +243,7 @@ function x264_predict_8x8_v_neon
vst1.8 {d0}, [r0,:64], r12
.endr
bx lr
-.endfunc
+endfunc
function x264_predict_8x8_ddl_neon
add r1, #16
@@ -271,7 +271,7 @@ function x264_predict_8x8_ddl_neon
vst1.8 d2, [r0,:64], r12
vst1.8 d1, [r0,:64], r12
bx lr
-.endfunc
+endfunc
function x264_predict_8x8_ddr_neon
vld1.8 {d0-d3}, [r1,:128]
@@ -301,7 +301,7 @@ function x264_predict_8x8_ddr_neon
vst1.8 {d4}, [r0,:64], r12
vst1.8 {d5}, [r0,:64], r12
bx lr
-.endfunc
+endfunc
function x264_predict_8x8_vl_neon
add r1, #16
@@ -332,7 +332,7 @@ function x264_predict_8x8_vl_neon
vst1.8 {d3}, [r0,:64], r12
vst1.8 {d2}, [r0,:64], r12
bx lr
-.endfunc
+endfunc
function x264_predict_8x8_vr_neon
add r1, #8
@@ -364,7 +364,7 @@ function x264_predict_8x8_vr_neon
vst1.8 {d6}, [r0,:64], r12
vst1.8 {d3}, [r0,:64], r12
bx lr
-.endfunc
+endfunc
function x264_predict_8x8_hd_neon
mov r12, #FDEC_STRIDE
@@ -397,7 +397,7 @@ function x264_predict_8x8_hd_neon
vst1.8 {d16}, [r0,:64], r12
bx lr
-.endfunc
+endfunc
function x264_predict_8x8_hu_neon
mov r12, #FDEC_STRIDE
@@ -430,7 +430,7 @@ function x264_predict_8x8_hu_neon
vst1.8 {d7}, [r0,:64], r12
vst1.8 {d17}, [r0,:64]
bx lr
-.endfunc
+endfunc
function x264_predict_8x8c_dc_top_neon
sub r2, r0, #FDEC_STRIDE
@@ -443,7 +443,7 @@ function x264_predict_8x8c_dc_top_neon
vdup.8 d0, d0[0]
vtrn.32 d0, d1
b pred8x8_dc_end
-.endfunc
+endfunc
function x264_predict_8x8c_dc_left_neon
mov r1, #FDEC_STRIDE
@@ -455,7 +455,7 @@ function x264_predict_8x8c_dc_left_neon
vdup.8 d1, d0[1]
vdup.8 d0, d0[0]
b pred8x8_dc_end
-.endfunc
+endfunc
function x264_predict_8x8c_dc_neon
sub r2, r0, #FDEC_STRIDE
@@ -481,7 +481,7 @@ pred8x8_dc_end:
vst1.8 {d1}, [r2,:64], r1
.endr
bx lr
-.endfunc
+endfunc
function x264_predict_8x8c_h_neon
sub r1, r0, #1
@@ -493,7 +493,7 @@ function x264_predict_8x8c_h_neon
vst1.64 {d2}, [r0,:64], ip
.endr
bx lr
-.endfunc
+endfunc
function x264_predict_8x8c_v_neon
sub r0, r0, #FDEC_STRIDE
@@ -503,7 +503,7 @@ function x264_predict_8x8c_v_neon
vst1.64 {d0}, [r0,:64], ip
.endr
bx lr
-.endfunc
+endfunc
function x264_predict_8x8c_p_neon
sub r3, r0, #FDEC_STRIDE
@@ -556,7 +556,7 @@ function x264_predict_8x8c_p_neon
subs r3, r3, #1
bne 1b
bx lr
-.endfunc
+endfunc
function x264_predict_16x16_dc_top_neon
@@ -567,7 +567,7 @@ function x264_predict_16x16_dc_top_neon
vrshrn.u16 d0, q0, #4
vdup.8 q0, d0[0]
b pred16x16_dc_end
-.endfunc
+endfunc
function x264_predict_16x16_dc_left_neon
mov r1, #FDEC_STRIDE
@@ -578,7 +578,7 @@ function x264_predict_16x16_dc_left_neon
vrshrn.u16 d0, q0, #4
vdup.8 q0, d0[0]
b pred16x16_dc_end
-.endfunc
+endfunc
function x264_predict_16x16_dc_neon
sub r3, r0, #FDEC_STRIDE
@@ -616,7 +616,7 @@ pred16x16_dc_end:
vst1.64 {d0-d1}, [r0,:128], r1
.endr
bx lr
-.endfunc
+endfunc
function x264_predict_16x16_h_neon
sub r1, r0, #1
@@ -630,7 +630,7 @@ function x264_predict_16x16_h_neon
vst1.64 {d2-d3}, [r0,:128], ip
.endr
bx lr
-.endfunc
+endfunc
function x264_predict_16x16_v_neon
sub r0, r0, #FDEC_STRIDE
@@ -640,7 +640,7 @@ function x264_predict_16x16_v_neon
vst1.64 {d0-d1}, [r0,:128], ip
.endr
bx lr
-.endfunc
+endfunc
function x264_predict_16x16_p_neon
sub r3, r0, #FDEC_STRIDE
@@ -697,4 +697,4 @@ function x264_predict_16x16_p_neon
subs r3, r3, #1
bne 1b
bx lr
-.endfunc
+endfunc

98
multimedia/libx264/files/patch-common_arm_quant-a.S Normal file
View File

@ -0,0 +1,98 @@
--- common/arm/quant-a.S.orig 2014-08-27 20:45:08 UTC
+++ common/arm/quant-a.S
@@ -80,7 +80,7 @@ function x264_quant_2x2_dc_neon
vsub.s16 d3, d3, d0
vst1.64 {d3}, [r0,:64]
QUANT_END d3
-.endfunc
+endfunc
// quant_4x4_dc( int16_t dct[16], int mf, int bias )
function x264_quant_4x4_dc_neon
@@ -92,7 +92,7 @@ function x264_quant_4x4_dc_neon
QUANT_TWO q0, q0, d4, d5, d4, d5, q0
vorr d0, d0, d1
QUANT_END d0
-.endfunc
+endfunc
// quant_4x4( int16_t dct[16], uint16_t mf[16], uint16_t bias[16] )
function x264_quant_4x4_neon
@@ -104,7 +104,7 @@ function x264_quant_4x4_neon
QUANT_TWO q0, q1, d4, d5, d6, d7, q0
vorr d0, d0, d1
QUANT_END d0
-.endfunc
+endfunc
// quant_4x4x4( int16_t dct[4][16], uint16_t mf[16], uint16_t bias[16] )
function x264_quant_4x4x4_neon
@@ -145,7 +145,7 @@ function x264_quant_4x4x4_neon
orrne r0, #8
vpop {d8-d15}
bx lr
-.endfunc
+endfunc
// quant_8x8( int16_t dct[64], uint16_t mf[64], uint16_t bias[64] )
function x264_quant_8x8_neon
@@ -165,7 +165,7 @@ function x264_quant_8x8_neon
.endr
vorr d0, d0, d1
QUANT_END d0
-.endfunc
+endfunc
.macro DEQUANT_START mf_size offset dc=no
mov r3, #0x2b
@@ -257,7 +257,7 @@ dequant_\size\()_rshift_loop:
bgt dequant_\size\()_rshift_loop
.endif
bx lr
-.endfunc
+endfunc
.endm
DEQUANT 4x4, 4
@@ -307,7 +307,7 @@ dequant_4x4_dc_rshift:
vmovn.s32 d3, q13
vst1.16 {d0-d3}, [r0,:128]
bx lr
-.endfunc
+endfunc
// int coeff_last( int16_t *l )
@@ -319,7 +319,7 @@ function x264_coeff_last4_arm
lsrs r2, r2, #16
addne r0, r0, #1
bx lr
-.endfunc
+endfunc
function x264_coeff_last8_arm
ldrd r2, r3, [r0, #8]
@@ -333,7 +333,7 @@ function x264_coeff_last8_arm
lsrs r2, r2, #16
addne r0, r0, #1
bx lr
-.endfunc
+endfunc
.macro COEFF_LAST_1x size
function x264_coeff_last\size\()_neon
@@ -358,7 +358,7 @@ function x264_coeff_last\size\()_neon
subslt r0, r3, r0, lsr #2
movlt r0, #0
bx lr
-.endfunc
+endfunc
.endm
COEFF_LAST_1x 15
@@ -407,4 +407,4 @@ function x264_coeff_last64_neon
subslt r0, ip, r0
movlt r0, #0
bx lr
-.endfunc
+endfunc

84
multimedia/libx264/files/patch-configure
View File

@ -1,20 +1,74 @@
From 9463ec0004f1bddc49c05ed8e38430a4ce1738fb Mon Sep 17 00:00:00 2001
From: Janne Grunau <janne-x264@jannau.net>
Date: Sun, 20 Jul 2014 13:40:28 +0200
Subject: [PATCH] arm/ppc: use $CC as default assembler
--- configure.orig 2014-08-27 20:45:08 UTC
+++ configure
@@ -648,7 +648,7 @@ case $host_cpu in
--- configure.orig 2014-08-27 22:45:08.000000000 +0200
+++ configure 2015-01-07 13:42:20.868702616 +0100
@@ -197,8 +197,9 @@
as_check() {
log_check "whether $AS supports $1"
- echo "$1" > conftest.asm
- if $AS conftest.asm $ASFLAGS $2 -o conftest.o >conftest.log 2>&1; then
+ echo "$1" > conftest$AS_EXT
+ as_cmd="$AS conftest$AS_EXT $ASFLAGS $2 -o conftest.o"
+ if $as_cmd >conftest.log 2>&1; then
res=$?
log_ok
else
@@ -206,12 +207,12 @@
log_fail
log_msg "Failed commandline was:"
log_msg "--------------------------------------------------"
- log_msg "$AS conftest.asm $ASFLAGS $2 -o conftest.o"
+ log_msg "$as_cmd"
cat conftest.log >> config.log
log_msg "--------------------------------------------------"
log_msg "Failed program was:"
log_msg "--------------------------------------------------"
- cat conftest.asm >> config.log
+ cat conftest$AS_EXT >> config.log
log_msg "--------------------------------------------------"
fi
return $res
@@ -302,10 +303,13 @@
cross_prefix=""
EXE=""
+AS_EXT=".S"
+NL="
+"
# list of all preprocessor HAVE values we can define
CONFIG_HAVE="MALLOC_H ALTIVEC ALTIVEC_H MMX ARMV6 ARMV6T2 NEON BEOSTHREAD POSIXTHREAD WIN32THREAD THREAD LOG2F SWSCALE \
- LAVF FFMS GPAC AVS GPL VECTOREXT INTERLACED CPU_COUNT OPENCL THP LSMASH X86_INLINE_ASM"
+ LAVF FFMS GPAC AVS GPL VECTOREXT INTERLACED CPU_COUNT OPENCL THP LSMASH X86_INLINE_ASM AS_FUNC"
# parse options
@@ -586,6 +590,7 @@
i*86)
ARCH="X86"
AS="yasm"
+ AS_EXT=".asm"
ASFLAGS="$ASFLAGS -O2 -DARCH_X86_64=0 -I\$(SRCPATH)/common/x86/"
if [ $compiler = GNU ]; then
if [[ "$asm" == auto && "$CFLAGS" != *-march* ]]; then
@@ -626,6 +631,7 @@
x86_64)
ARCH="X86_64"
AS="yasm"
+ AS_EXT=".asm"
ASFLAGS="$ASFLAGS -DARCH_X86_64=1 -I\$(SRCPATH)/common/x86/"
[ $compiler = GNU ] && CFLAGS="-m64 $CFLAGS" && LDFLAGS="-m64 $LDFLAGS"
if [ "$SYS" = MACOSX ]; then
@@ -648,7 +654,8 @@
ARCH="PPC"
if [ $asm = auto ] ; then
define HAVE_ALTIVEC
- AS="${AS-${cross_prefix}gcc}"
+ AS="${AS-${CC}}"
+ AS_EXT=".c"
if [ $SYS = MACOSX ] ; then
CFLAGS="$CFLAGS -faltivec -fastf -mcpu=G4"
else
@@ -674,7 +674,7 @@ case $host_cpu in
@@ -674,7 +681,7 @@
LDFLAGS="$LDFLAGS -arch armv7"
fi
else
@ -23,3 +77,15 @@ Subject: [PATCH] arm/ppc: use $CC as default assembler
fi
;;
s390|s390x)
@@ -760,6 +767,11 @@
fi
fi
+if [ $asm = auto -a $ARCH = ARM ] ; then
+ # check if the assembler supports '.func' (clang 3.5 does not)
+ as_check ".func test${NL}.endfunc" && define HAVE_AS_FUNC 1
+fi
+
[ $asm = no ] && AS=""
[ "x$AS" = x ] && asm="no" || asm="yes"