/hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/encoder/arm/armv5te/ |
D | vp8_packtokens_armv5.asm | 46 push {r4-r12, lr} 61 ldr r4, [sp, #8] ; vp8_coef_encodings 63 add r4, r4, r6, lsl #3 ; a = vp8_coef_encodings + t 68 ldr r6, [r4, #vp8_token_value] ; v 69 ldr r8, [r4, #vp8_token_len] ; n 76 rsb r4, r8, #32 ; 32-n 80 lsl r12, r6, r4 ; r12 = v << 32 - n 84 ldrb r4, [r9, lr, asr #1] ; pp [i>>1] 91 mul r6, r4, r7 ; ((range-1) * pp[i>>1])) 99 add r4, r7, r6, lsr #8 ; 1 + (((range-1) * pp[i>>1]) >> 8) [all …]
|
D | vp8_packtokens_mbrow_armv5.asm | 45 push {r4-r12, lr} 49 ldr r4, _VP8_COMP_common_ 51 add r4, r0, r4 53 ldr r5, [r4, r6] ; load up mb_rows 59 ldr r4, _VP8_COMP_tplist_ 60 add r4, r0, r4 61 ldr r7, [r4, #0] ; dereference cpi->tp_list 82 ldr r4, [sp, #20] ; vp8_coef_encodings 84 add r4, r4, r6, lsl #3 ; a = vp8_coef_encodings + t 89 ldr r6, [r4, #vp8_token_value] ; v [all …]
|
D | vp8_packtokens_partitions_armv5.asm | 46 push {r4-r12, lr} 50 ldr r4, _VP8_COMP_common_ 52 add r4, r0, r4 54 ldr r5, [r4, r6] ; load up mb_rows 61 ldr r4, _VP8_COMP_tplist_ 62 add r4, r0, r4 63 ldr r7, [r4, #0] ; dereference cpi->tp_list 111 ldr r4, [sp, #80] ; vp8_coef_encodings 113 add r4, r4, r6, lsl #3 ; a = vp8_coef_encodings + t 118 ldr r6, [r4, #vp8_token_value] ; v [all …]
|
D | boolhuff_armv5te.asm | 60 push {r4-r10, lr} 62 mov r4, r2 71 mul r6, r4, r7 ; ((range-1) * probability) 74 add r4, r7, r6, lsr #8 ; 1 + (((range-1) * probability) >> 8) 76 addne r2, r2, r4 ; if (bit) lowvalue += split 77 subne r4, r5, r4 ; if (bit) range = range-split 80 clz r6, r4 86 lsl r5, r4, r6 ; range <<= shift 90 sub r4, r6, #1 ; offset-1 91 lsls r4, r2, r4 ; if((lowvalue<<(offset-1)) & 0x80000000 ) [all …]
|
/hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/common/arm/armv6/ |
D | copymem8x8_v6.asm | 21 ;push {r4-r5} 22 stmdb sp!, {r4-r5} 29 ands r4, r0, #7 32 ands r4, r0, #3 36 ldrb r4, [r0] 42 strb r4, [r2] 45 ldrb r4, [r0, #2] 50 strb r4, [r2, #2] 53 ldrb r4, [r0, #4] 56 strb r4, [r2, #4] [all …]
|
D | copymem8x4_v6.asm | 21 ;push {r4-r5} 22 stmdb sp!, {r4-r5} 29 ands r4, r0, #7 32 ands r4, r0, #3 36 ldrb r4, [r0] 42 strb r4, [r2] 45 ldrb r4, [r0, #2] 50 strb r4, [r2, #2] 53 ldrb r4, [r0, #4] 56 strb r4, [r2, #4] [all …]
|
D | copymem16x16_v6.asm | 21 stmdb sp!, {r4 - r7} 22 ;push {r4-r7} 27 ands r4, r0, #15 30 ands r4, r0, #7 33 ands r4, r0, #3 37 ldrb r4, [r0] 45 strb r4, [r2] 50 ldrb r4, [r0, #4] 57 strb r4, [r2, #4] 62 ldrb r4, [r0, #8] [all …]
|
D | intra4x4_predict_v6.asm | 34 push {r4-r12, lr} 38 pop {r4-r12, pc} ; default 53 ldrb r4, [r1], r2 ; Left[0] 61 add r4, r4, r5 62 add r4, r4, r6 63 add r4, r4, r7 64 add r4, r4, r12 65 add r4, r4, #4 67 mov r12, r4, asr #3 ; (expected_dc + 4) >> 3 79 pop {r4-r12, pc} [all …]
|
D | vp8_variance_halfpixvar16x16_hv_armv6.asm | 27 stmfd sp!, {r4-r12, lr} 40 ldr r4, [r0, #0] ; load source pixels a, row N 47 uhsub8 r4, r4, r6 48 eor r4, r4, r10 55 uhsub8 r4, r4, r5 57 eor r4, r4, r10 59 usub8 r6, r4, r5 ; calculate difference 62 usub8 r6, r5, r4 ; calculate difference with reversed operands 67 usad8 r4, r7, lr ; calculate sum of positive differences 71 adds r8, r8, r4 ; add positive differences to sum [all …]
|
D | dc_only_idct_add_v6.asm | 24 stmdb sp!, {r4 - r7} 28 ldr r4, [r1], r2 35 uxtab16 r5, r0, r4 ; a1+2 | a1+0 36 uxtab16 r4, r0, r4, ror #8 ; a1+3 | a1+1 40 usat16 r4, #8, r4 43 orr r5, r5, r4, lsl #8 45 ldr r4, [r1], r2 50 uxtab16 r5, r0, r4 51 uxtab16 r4, r0, r4, ror #8 55 usat16 r4, #8, r4 [all …]
|
D | vp8_variance_halfpixvar16x16_h_armv6.asm | 27 stmfd sp!, {r4-r12, lr} 39 ldr r4, [r0, #0] ; load 4 src pixels 45 uhsub8 r4, r4, r6 46 eor r4, r4, r10 48 usub8 r6, r4, r5 ; calculate difference 51 usub8 r6, r5, r4 ; calculate difference with reversed operands 56 usad8 r4, r7, lr ; calculate sum of positive differences 60 adds r8, r8, r4 ; add positive differences to sum 69 ldr r4, [r0, #4] ; load 4 src pixels 75 uhsub8 r4, r4, r6 [all …]
|
D | vp8_variance_halfpixvar16x16_v_armv6.asm | 27 stmfd sp!, {r4-r12, lr} 40 ldr r4, [r0, #0] ; load 4 src pixels 46 uhsub8 r4, r4, r6 47 eor r4, r4, r10 49 usub8 r6, r4, r5 ; calculate difference 52 usub8 r6, r5, r4 ; calculate difference with reversed operands 57 usad8 r4, r7, lr ; calculate sum of positive differences 61 adds r8, r8, r4 ; add positive differences to sum 70 ldr r4, [r0, #4] ; load 4 src pixels 76 uhsub8 r4, r4, r6 [all …]
|
D | simpleloopfilter_v6.asm | 59 stmdb sp!, {r4 - r11, lr} 63 ldr r4, [src, -pstep] ; p0 173 pkhbt r9, r3, r4, lsl #16 176 ;transpose r7, r8, r9, r10 to r3, r4, r5, r6 177 TRANSPOSE_MATRIX r7, r8, r9, r10, r3, r4, r5, r6 181 uqsub8 r9, r4, r5 ; p0 - q0 182 uqsub8 r10, r5, r4 ; q0 - p0 200 eor r4, r4, r2 ; p0 offset to convert to a signed value 204 qsub8 r6, r5, r4 ; q0 - p0 229 qadd8 r4, r4, r9 ; u = p0 + Filter2 [all …]
|
D | iwalsh_v6.asm | 22 stmdb sp!, {r4 - r12, lr} 26 ldr r4, [r0, #8] ; [5 | 4] 34 qadd16 r11, r4, r6 ; b1 [5+9 | 4+8] 35 qsub16 r12, r4, r6 ; c1 [5-9 | 4-8] 39 qadd16 r4, r12, lr ; c1 + d1 [5 | 4] 57 qsubaddx r12, r4, r5 ; [c1|a1] [5-6 | 4+7] 58 qaddsubx lr, r4, r5 ; [b1|d1] [5+6 | 4-7] 63 qaddsubx r4, r12, lr ; [b2|c2] [c1+d1 | a1-b1] 68 qadd16 r4, r4, r10 ; [b2+3|c2+3] 84 asr lr, r4, #19 ; [5] [all …]
|
D | vp8_sad16x16_armv6.asm | 26 stmfd sp!, {r4-r12, lr} 33 mov r4, #0 ; sad = 0; 45 usada8 r4, r8, r6, r4 ; calculate sad for 4 pixels 57 usada8 r4, r10, r12, r4 ; calculate sad for 4 pixels 62 add r4, r4, r8 ; add partial sad values 70 usada8 r4, r6, r8, r4 ; calculate sad for 4 pixels 79 usada8 r4, r10, r12, r4 ; calculate sad for 4 pixels 86 add r4, r4, r8 ; add partial sad values 90 mov r0, r4 ; return sad 91 ldmfd sp!, {r4-r12, pc}
|
D | vp8_variance16x16_armv6.asm | 27 stmfd sp!, {r4-r12, lr} 38 ldr r4, [r0, #0] ; load 4 src pixels 43 usub8 r6, r4, r5 ; calculate difference 46 usub8 r9, r5, r4 ; calculate difference with reversed operands 51 usad8 r4, r7, lr ; calculate sum of positive differences 55 adds r8, r8, r4 ; add positive differences to sum 64 ldr r4, [r0, #4] ; load 4 src pixels 68 usub8 r6, r4, r5 ; calculate difference 70 usub8 r9, r5, r4 ; calculate difference with reversed operands 74 usad8 r4, r7, lr ; calculate sum of positive differences [all …]
|
D | dequantize_v6.asm | 21 stmdb sp!, {r4-r9, lr} 24 ldr r4, [r1] ;load DQC 31 smulbb r7, r3, r4 ;multiply 32 smultt r8, r3, r4 37 ldr r4, [r1, #8] 42 smulbb r7, r3, r4 ;multiply 44 smultt r8, r3, r4 57 ldrne r4, [r1] 66 ldmia sp!, {r4-r9, pc}
|
D | dequant_idct_v6.asm | 21 stmdb sp!, {r4-r11, lr} 23 ldr r4, [r0] ;input 32 smulbb r6, r4, r5 33 smultt r7, r4, r5 35 ldr r4, [r0, #4] ;input 41 smulbb r6, r4, r5 42 smultt r7, r4, r5 46 ldrne r4, [r0, #4] 59 ldr r4, sinpi8sqrt2 67 smulwt r10, r4, r6 [all …]
|
/hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/common/arm/neon/ |
D | buildintrapredictorsmby_neon.asm | 28 push {r4-r8, lr} 40 ldr r4, [sp, #24] ; Up 51 adds r7, r4, r5 55 cmp r4, #0 64 vmov.32 r4, d8[0] 67 add r12, r4, r6 80 ldrb r4, [r0], r2 85 add r12, r12, r4 90 ldrb r4, [r0], r2 95 add r12, r12, r4 [all …]
|
/hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/encoder/arm/armv6/ |
D | vp8_short_fdct4x4_armv6.asm | 21 stmfd sp!, {r4 - r12, lr} 26 ldrd r4, r5, [r0] ; [i1 | i0] [i3 | i2] 34 qadd16 r6, r4, r5 ; [i1+i2 | i0+i3] = [b1 | a1] without shift 35 qsub16 r7, r4, r5 ; [i1-i2 | i0-i3] = [c1 | d1] without shift 43 smuad r4, r6, lr ; o0 = (i1+i2)*8 + (i0+i3)*8 51 pkhbt r3, r4, r6, lsl #4 ; [o1 | o0], keep in register for PART 2 74 ldrd r4, r5, [r0] ; [i9 | i8] [i11 | i10] 84 qadd16 r6, r4, r5 ; [i9+i10 | i8+i11]=[b1 | a1] without shift 85 qsub16 r7, r4, r5 ; [i9-i10 | i8-i11]=[c1 | d1] without shift 99 ldrd r4, r5, [r0] ; [i13 | i12] [i15 | i14] [all …]
|
D | vp8_mse16x16_armv6.asm | 29 push {r4-r9, lr} 35 mov r4, #0 ; initialize sse = 0 61 smlad r4, r6, r6, r4 ; dual signed multiply, add and accumulate (1) 65 smlad r4, r7, r7, r4 ; dual signed multiply, add and accumulate (2) 80 smlad r4, r6, r6, r4 ; dual signed multiply, add and accumulate (1) 84 smlad r4, r7, r7, r4 ; dual signed multiply, add and accumulate (2) 101 smlad r4, r6, r6, r4 ; dual signed multiply, add and accumulate (1) 105 smlad r4, r7, r7, r4 ; dual signed multiply, add and accumulate (2) 124 smlad r4, r6, r6, r4 ; dual signed multiply, add and accumulate (1) 125 smlad r4, r7, r7, r4 ; dual signed multiply, add and accumulate (2) [all …]
|
/hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/encoder/arm/neon/ |
D | subtract_neon.asm | 26 stmfd sp!, {r4-r7} 29 ldr r4, [r0, #vp8_block_src] 33 add r3, r3, r4 ; src = *base_src + src 57 ldmfd sp!, {r4-r7} 67 push {r4-r7} 69 ldr r4, [sp, #16] ; pred_stride 75 vld1.8 {q1}, [r3], r4 ;load pred 77 vld1.8 {q3}, [r3], r4 79 vld1.8 {q5}, [r3], r4 81 vld1.8 {q7}, [r3], r4 [all …]
|
/hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/encoder/ppc/ |
D | encodemb_altivec.asm | 16 ;# r4 unsigned char *usrc 37 lvsl v5, 0, r4 ;# permutate value for alignment 38 lvx v1, 0, r4 ;# src 41 add r4, r4, r7 49 lvsl v5, 0, r4 ;# permutate value for alignment 50 lvx v1, 0, r4 ;# src 52 add r4, r4, r7 113 ;# r4 unsigned char *src 127 lvx v1, 0, r4 ;# src 130 add r4, r4, r6
|
/hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/common/ppc/ |
D | variance_altivec.asm | 79 add r3, r3, r4 93 lwz r4, 12(r1) 98 stw r4, 0(r7) ;# sse 102 subf r3, r3, r4 ;# sse - ((sum*sum) >> DS) 112 add r3, r3, r4 120 add r3, r3, r4 137 lwz r4, 12(r1) 142 stw r4, 0(r7) ;# sse 146 subf r3, r3, r4 ;# sse - ((sum*sum) >> 8) 151 ;# r4 int source_stride [all …]
|
/hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp9/common/arm/neon/ |
D | vp9_avg_neon.asm | 19 push {r4-r6, lr} 20 ldrd r4, r5, [sp, #32] 23 cmp r4, #32 26 cmp r4, #8 33 sub r4, r3, #32 40 vld1.8 {q10-q11}, [r6@128], r4 46 vst1.8 {q2-q3}, [r2@128], r4 49 pop {r4-r6, pc} 68 pop {r4-r6, pc} 85 pop {r4-r6, pc} [all …]
|