| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/common/arm/neon/ |
| D | loopfilter_neon.asm | 21 ; r1 int pitch
29 sub r2, r0, r1, lsl #2 ; move src pointer down by 4 lines
31 add r12, r2, r1
32 add r1, r1, r1
36 vld1.u8 {q3}, [r2@128], r1 ; p3
37 vld1.u8 {q4}, [r12@128], r1 ; p2
38 vld1.u8 {q5}, [r2@128], r1 ; p1
39 vld1.u8 {q6}, [r12@128], r1 ; p0
40 vld1.u8 {q7}, [r2@128], r1 ; q0
41 vld1.u8 {q8}, [r12@128], r1 ; q1
[all …]
|
| D | mbloopfilter_neon.asm | 25 ; r1 int pitch,
31 add r1, r1, r1 ; double stride
33 sub r0, r0, r1, lsl #1 ; move src pointer down by 4 lines
35 add r12, r0, r1, lsr #1 ; move src pointer up by 1 line
37 vld1.u8 {q3}, [r0@128], r1 ; p3
38 vld1.u8 {q4}, [r12@128], r1 ; p2
39 vld1.u8 {q5}, [r0@128], r1 ; p1
40 vld1.u8 {q6}, [r12@128], r1 ; p0
41 vld1.u8 {q7}, [r0@128], r1 ; q0
42 vld1.u8 {q8}, [r12@128], r1 ; q1
[all …]
|
| D | buildintrapredictorsmby_neon.asm | 21 ; r1 unsigned char *ypred_ptr
129 vst1.u8 {q0}, [r1]!
130 vst1.u8 {q0}, [r1]!
131 vst1.u8 {q0}, [r1]!
132 vst1.u8 {q0}, [r1]!
133 vst1.u8 {q0}, [r1]!
134 vst1.u8 {q0}, [r1]!
135 vst1.u8 {q0}, [r1]!
136 vst1.u8 {q0}, [r1]!
137 vst1.u8 {q0}, [r1]!
[all …]
|
| D | sad8_neon.asm | 28 vld1.8 {d0}, [r0], r1
31 vld1.8 {d2}, [r0], r1
36 vld1.8 {d4}, [r0], r1
41 vld1.8 {d6}, [r0], r1
46 vld1.8 {d0}, [r0], r1
51 vld1.8 {d2}, [r0], r1
56 vld1.8 {d4}, [r0], r1
61 vld1.8 {d6}, [r0], r1
85 vld1.8 {d0}, [r0], r1
88 vld1.8 {d2}, [r0], r1
[all …]
|
| D | sad16_neon.asm | 22 ; r1 int src_stride
27 vld1.8 {q0}, [r0], r1
30 vld1.8 {q1}, [r0], r1
36 vld1.8 {q2}, [r0], r1
42 vld1.8 {q3}, [r0], r1
49 vld1.8 {q0}, [r0], r1
55 vld1.8 {q1}, [r0], r1
61 vld1.8 {q2}, [r0], r1
67 vld1.8 {q3}, [r0], r1
74 vld1.8 {q0}, [r0], r1
[all …]
|
| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp9/common/arm/neon/ |
| D | vp9_reconintra_neon.asm | 33 ; r1 ptrdiff_t y_stride
39 vst1.32 {d0[0]}, [r0], r1
40 vst1.32 {d0[0]}, [r0], r1
41 vst1.32 {d0[0]}, [r0], r1
42 vst1.32 {d0[0]}, [r0], r1
50 ; r1 ptrdiff_t y_stride
56 vst1.8 {d0}, [r0], r1
57 vst1.8 {d0}, [r0], r1
58 vst1.8 {d0}, [r0], r1
59 vst1.8 {d0}, [r0], r1
[all …]
|
| D | vp9_mb_lpf_neon.asm | 23 ; r1 int p, /* pitch */
38 sub r8, r0, r1, lsl #3 ; move src pointer down by 8 lines
40 vld1.u8 {d0}, [r8@64], r1 ; p7
41 vld1.u8 {d1}, [r8@64], r1 ; p6
42 vld1.u8 {d2}, [r8@64], r1 ; p5
43 vld1.u8 {d3}, [r8@64], r1 ; p4
44 vld1.u8 {d4}, [r8@64], r1 ; p3
45 vld1.u8 {d5}, [r8@64], r1 ; p2
46 vld1.u8 {d6}, [r8@64], r1 ; p1
47 vld1.u8 {d7}, [r8@64], r1 ; p0
[all …]
|
| D | vp9_copy_neon.asm | 31 sub lr, r1, #32
34 pld [r0, r1, lsl #1]
44 pld [r0, r1, lsl #1]
45 vld1.8 {q0-q1}, [r0], r1
46 pld [r0, r1, lsl #1]
47 vld1.8 {q2-q3}, [r0], r1
55 pld [r0, r1, lsl #1]
56 vld1.8 {q0}, [r0], r1
57 pld [r0, r1, lsl #1]
58 vld1.8 {q1}, [r0], r1
[all …]
|
| D | vp9_idct16x16_1_add_neon.asm | 22 ; r1 uint8_t *dest
39 mov r12, r1 ; save dest
52 vld1.64 {d2}, [r1], r0
53 vld1.64 {d3}, [r1], r2
54 vld1.64 {d4}, [r1], r0
55 vld1.64 {d5}, [r1], r2
56 vld1.64 {d6}, [r1], r0
57 vld1.64 {d7}, [r1], r2
58 vld1.64 {d16}, [r1], r0
59 vld1.64 {d17}, [r1], r2
[all …]
|
| D | vp9_loopfilter_neon.asm | 32 ; r1 int p, /* pitch */
43 add r1, r1, r1 ; double pitch
52 sub r2, r0, r1, lsl #1 ; move src pointer down by 4 lines
53 add r3, r2, r1, lsr #1 ; set to 3 lines down
55 vld1.u8 {d3}, [r2@64], r1 ; p3
56 vld1.u8 {d4}, [r3@64], r1 ; p2
57 vld1.u8 {d5}, [r2@64], r1 ; p1
58 vld1.u8 {d6}, [r3@64], r1 ; p0
59 vld1.u8 {d7}, [r2@64], r1 ; q0
60 vld1.u8 {d16}, [r3@64], r1 ; q1
[all …]
|
| D | vp9_convolve8_neon.asm | 44 ; r1 int src_stride
69 sub r8, r1, r1, lsl #2 ; -src_stride * 3
75 rsb r9, r6, r1, lsl #2 ; reset src for outer loop
82 vld1.8 {d24}, [r0], r1
83 vld1.8 {d25}, [r0], r1
84 vld1.8 {d26}, [r0], r1
91 pld [r0, r1, lsl #2]
107 vld1.32 {d28[]}, [r0], r1
108 vld1.32 {d29[]}, [r0], r1
109 vld1.32 {d31[]}, [r0], r1
[all …]
|
| D | vp9_convolve8_avg_neon.asm | 44 ; r1 int src_stride
69 sub r8, r1, r1, lsl #2 ; -src_stride * 3
75 rsb r9, r6, r1, lsl #2 ; reset src for outer loop
82 vld1.8 {d24}, [r0], r1
83 vld1.8 {d25}, [r0], r1
84 vld1.8 {d26}, [r0], r1
91 pld [r0, r1, lsl #2]
107 vld1.32 {d28[]}, [r0], r1
108 vld1.32 {d29[]}, [r0], r1
109 vld1.32 {d31[]}, [r0], r1
[all …]
|
| D | vp9_avg_neon.asm | 32 sub lr, r1, #32
35 pld [r0, r1, lsl #1]
52 vld1.8 {q0-q1}, [r0], r1
53 vld1.8 {q2-q3}, [r0], r1
58 pld [r0, r1]
71 vld1.8 {q0}, [r0], r1
72 vld1.8 {q1}, [r0], r1
76 pld [r0, r1]
88 vld1.8 {d0}, [r0], r1
89 vld1.8 {d1}, [r0], r1
[all …]
|
| D | vp9_idct16x16_add_neon.asm | 43 ; r1 int16_t *output
258 vst1.64 {d16}, [r1], r2
259 vst1.64 {d17}, [r1], r2
260 vst1.64 {d18}, [r1], r2
261 vst1.64 {d19}, [r1], r2
262 vst1.64 {d20}, [r1], r2
263 vst1.64 {d21}, [r1], r2
264 vst1.64 {d22}, [r1], r2
265 vst1.64 {d23}, [r1], r2
266 vst1.64 {d24}, [r1], r2
[all …]
|
| D | vp9_idct8x8_1_add_neon.asm | 22 ; r1 uint8_t *dest
39 mov r12, r1 ; save dest
50 vld1.64 {d2}, [r1], r2
51 vld1.64 {d3}, [r1], r2
52 vld1.64 {d4}, [r1], r2
53 vld1.64 {d5}, [r1], r2
54 vld1.64 {d6}, [r1], r2
55 vld1.64 {d7}, [r1], r2
56 vld1.64 {d16}, [r1], r2
57 vld1.64 {d17}, [r1]
|
| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/common/arm/armv6/ |
| D | dequant_idct_v6.asm | 16 ; r1 = dq
24 ldr r5, [r1], #4 ;dq
36 ldr r5, [r1], #4 ;dq
47 ldrne r5, [r1], #4
55 mov r1, r0
89 str r6, [r1, #8]
91 str r7, [r1, #16]
92 str r10, [r1, #24]
93 str r9, [r1], #4
97 sub r0, r1, #8
[all …]
|
| D | bilinearfilter_v6.asm | 19 ; r1 unsigned short *dst_ptr,
44 mov r11, r1 ; save dst_ptr for each row
78 strh r6, [r1], r3 ; result is transposed and stored
81 strh r7, [r1], r3
86 strh r8, [r1], r3 ; result is transposed and stored
89 strh r9, [r1], r3
103 mov r1, r11
119 strh r6, [r1], r3 ; store it to immediate buffer
121 strh r7, [r1], r3
123 strh r8, [r1], r3
[all …]
|
| D | dequantize_v6.asm | 18 ; r1 short *DQC
24 ldr r4, [r1] ;load DQC
26 ldr r6, [r1, #4]
37 ldr r4, [r1, #8]
39 ldr r6, [r1, #12]
53 add r1, r1, #16
57 ldrne r4, [r1]
61 ldrne r6, [r1, #4]
|
| D | iwalsh_v6.asm | 72 strh r12, [r1], #32
74 strh lr, [r1], #32
78 strh r2, [r1], #32
80 strh r3, [r1], #32
83 strh r12, [r1], #32
85 strh lr, [r1], #32
89 strh r4, [r1], #32
91 strh r5, [r1], #32
109 strh r12, [r1], #32
111 strh lr, [r1], #32
[all …]
|
| D | filter_v6.asm | 23 ; r1 short *output_ptr
49 str r1, [sp] ; push destination to stack
94 strh lr, [r1], r12 ; result is transposed and stored, which
97 strh r11, [r1], r12
101 ldr r1, [sp] ; load and update dst address
105 add r1, r1, #2 ; move over to next column
106 str r1, [sp]
139 str r1, [sp] ; push destination to stack
184 strh lr, [r1], r12 ; result is transposed and stored, which
187 strh r11, [r1], r12
[all …]
|
| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/encoder/arm/neon/ |
| D | vp8_memcpy_neon.asm | 24 ;pld [r1] ;preload pred data
25 ;pld [r1, #128]
26 ;pld [r1, #256]
27 ;pld [r1, #384]
32 vld1.8 {q0, q1}, [r1]! ;load src data
34 vld1.8 {q2, q3}, [r1]!
36 vld1.8 {q4, q5}, [r1]!
38 vld1.8 {q6, q7}, [r1]!
40 vld1.8 {q8, q9}, [r1]!
42 vld1.8 {q10, q11}, [r1]!
[all …]
|
| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vpx_scale/arm/neon/ |
| D | vp8_vpxyv12_extendframeborders_neon.asm | 29 ldr r1, [r0, #yv12_buffer_config_y_buffer] ; src_ptr1
35 add r6, r1, r3 ; dest_ptr2 = src_ptr2 + 1 (src_ptr1 + plane_width)
37 sub r5, r1, #32 ; dest_ptr1 = src_ptr1 - Border
42 vld1.8 {d0[], d1[]}, [r1], lr
44 vld1.8 {d8[], d9[]}, [r1], lr
46 vld1.8 {d16[], d17[]}, [r1], lr
48 vld1.8 {d24[], d25[]}, [r1], lr
74 ldr r1, [r0, #yv12_buffer_config_y_buffer] ; y_buffer
80 sub r1, r1, #32 ; src_ptr1 = y_buffer - Border
81 …add r6, r1, r8 ; dest_ptr2 = src_ptr2 - plane_stride (src_ptr1 + (plane_h…
[all …]
|
| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/common/ppc/ |
| D | variance_altivec.asm | 35 stwu r1, -32(r1) ;# create space on the stack
45 addi r1, r1, 32 ;# recover stack
89 stvx v8, 0, r1
90 lwz r3, 12(r1)
92 stvx v9, 0, r1
93 lwz r4, 12(r1)
133 stvx v8, 0, r1
134 lwz r3, 12(r1)
136 stvx v9, 0, r1
137 lwz r4, 12(r1)
[all …]
|
| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/encoder/arm/armv6/ |
| D | vp8_fast_quantize_b_armv6.asm | 23 ; r1 BLOCKD *d
25 stmfd sp!, {r1, r4-r11, lr}
30 ldr r6, [r1, #vp8_blockd_qcoeff] ; qcoeff
31 ldr r7, [r1, #vp8_blockd_dqcoeff] ; dqcoeff
32 ldr r8, [r1, #vp8_blockd_dequant] ; dequant
39 mov r1, #0 ; flags for quantized coeffs
44 ldr r10, [r5], #4 ; [r1 | r0]
50 sadd16 r9, r9, r10 ; [x1+r1 | x0+r0]
55 smultt r9, r9, r11 ; [(x1+r1)*q1]
75 orrne r1, r1, r2, lsr #24 ; add flag for nonzero coeffs
[all …]
|
| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/encoder/ppc/ |
| D | rdopt_altivec.asm | 22 stwu r1,-32(r1) ;# create space on the stack
24 stw r5, 12(r1) ;# tranfer dc to vector register
45 stvx v1, 0, r1
46 lwz r3, 12(r1) ;# return value
48 addi r1, r1, 32 ;# recover stack
|