/hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/encoder/arm/armv6/ |
D | walsh_v6.asm | 22 ; r2 int pitch 27 ldrd r4, r5, [r0], r2 29 ldrd r6, r7, [r0], r2 35 ldrd r8, r9, [r0], r2 50 lsls r2, r3, #16 54 lsls r2, r7, #16 61 lsls r2, r5, #16 65 lsls r2, r9, #16 66 smuad r2, r9, lr ; D0 = a1<<2 + d1<<2 67 addne r2, r2, #1 ; D0 += (a1!=0) [all …]
|
D | vp8_fast_quantize_b_armv6.asm | 34 ldr r2, loop_count ; loop_count=0x1000000. 'lsls' instruction 57 ldr r10, [r5], #4 ; [r3 | r2] 65 sadd16 r12, r12, r10 ; [x3+r3 | x2+r2] 69 smulbb r10, r12, r9 ; [(x2+r2)*q2] 75 orrne r1, r1, r2, lsr #24 ; add flag for nonzero coeffs 85 orrne r1, r1, r2, lsr #23 ; add flag for nonzero coeffs 96 lsls r2, r2, #2 ; update loop counter 131 ldrh r2, [r0, #30] ; rc=15, i=15 133 cmp r2, #0 142 ldrh r2, [r0, #22] ; rc=11, i=13 [all …]
|
/hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/common/arm/neon/ |
D | buildintrapredictorsmby_neon.asm | 22 ; r2 int y_stride 58 sub r6, r0, r2 79 ldrb r3, [r0], r2 80 ldrb r4, [r0], r2 81 ldrb r5, [r0], r2 82 ldrb r6, [r0], r2 89 ldrb r3, [r0], r2 90 ldrb r4, [r0], r2 91 ldrb r5, [r0], r2 92 ldrb r6, [r0], r2 [all …]
|
D | sad8_neon.asm | 29 vld1.8 {d8}, [r2], r3 32 vld1.8 {d10}, [r2], r3 37 vld1.8 {d12}, [r2], r3 42 vld1.8 {d14}, [r2], r3 47 vld1.8 {d8}, [r2], r3 52 vld1.8 {d10}, [r2], r3 57 vld1.8 {d12}, [r2], r3 62 vld1.8 {d14}, [r2], r3 86 vld1.8 {d8}, [r2], r3 89 vld1.8 {d10}, [r2], r3 [all …]
|
D | sad16_neon.asm | 23 ; r2 unsigned char *ref_ptr 28 vld1.8 {q4}, [r2], r3 31 vld1.8 {q5}, [r2], r3 37 vld1.8 {q6}, [r2], r3 43 vld1.8 {q7}, [r2], r3 50 vld1.8 {q4}, [r2], r3 56 vld1.8 {q5}, [r2], r3 62 vld1.8 {q6}, [r2], r3 68 vld1.8 {q7}, [r2], r3 75 vld1.8 {q4}, [r2], r3 [all …]
|
D | iwalsh_neon.asm | 61 mov r2, #64 64 vst1.i16 d0[0], [r1],r2 65 vst1.i16 d1[0], [r3],r2 66 vst1.i16 d2[0], [r1],r2 67 vst1.i16 d3[0], [r3],r2 69 vst1.i16 d0[1], [r1],r2 70 vst1.i16 d1[1], [r3],r2 71 vst1.i16 d2[1], [r1],r2 72 vst1.i16 d3[1], [r3],r2 74 vst1.i16 d0[2], [r1],r2 [all …]
|
D | idct_dequant_0_2x_neon.asm | 21 ; r2 *dst 26 add r12, r2, #4 27 vld1.32 {d2[0]}, [r2], r3 29 vld1.32 {d2[1]}, [r2], r3 31 vld1.32 {d4[0]}, [r2], r3 33 vld1.32 {d4[1]}, [r2], r3 58 sub r2, r2, r3, lsl #2 ; dst - 4*stride 59 add r0, r2, #4 66 vst1.32 {d2[0]}, [r2], r3 ; lo 68 vst1.32 {d2[1]}, [r2], r3 [all …]
|
D | loopfilter_neon.asm | 22 ; r2 unsigned char blimit 27 vdup.u8 q0, r2 ; duplicate blimit 29 sub r2, r0, r1, lsl #2 ; move src pointer down by 4 lines 31 add r12, r2, r1 36 vld1.u8 {q3}, [r2@128], r1 ; p3 38 vld1.u8 {q5}, [r2@128], r1 ; p1 40 vld1.u8 {q7}, [r2@128], r1 ; q0 42 vld1.u8 {q9}, [r2@128] ; q2 45 sub r2, r2, r1, lsl #1 50 vst1.u8 {q5}, [r2@128], r1 ; store op1 [all …]
|
/hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/common/arm/armv6/ |
D | copymem16x16_v6.asm | 45 strb r4, [r2] 46 strb r5, [r2, #1] 47 strb r6, [r2, #2] 48 strb r7, [r2, #3] 57 strb r4, [r2, #4] 58 strb r5, [r2, #5] 59 strb r6, [r2, #6] 60 strb r7, [r2, #7] 67 strb r4, [r2, #8] 68 strb r5, [r2, #9] [all …]
|
D | copymem8x4_v6.asm | 42 strb r4, [r2] 43 strb r5, [r2, #1] 50 strb r4, [r2, #2] 51 strb r5, [r2, #3] 56 strb r4, [r2, #4] 57 strb r5, [r2, #5] 64 strb r4, [r2, #6] 65 strb r5, [r2, #7] 67 add r2, r2, r3 89 str r4, [r2] [all …]
|
D | copymem8x8_v6.asm | 42 strb r4, [r2] 43 strb r5, [r2, #1] 50 strb r4, [r2, #2] 51 strb r5, [r2, #3] 56 strb r4, [r2, #4] 57 strb r5, [r2, #5] 64 strb r4, [r2, #6] 65 strb r5, [r2, #7] 67 add r2, r2, r3 89 str r4, [r2] [all …]
|
D | vp8_sad16x16_armv6.asm | 22 ; r2 const unsigned char *ref_ptr 29 pld [r2, r3, lsl #0] 31 pld [r2, r3, lsl #1] 39 ldr r8, [r2, #0x0] ; load 4 ref pixels (1A) 41 ldr r9, [r2, #0x4] ; load 4 ref pixels (1A) 48 ldr r12, [r2, #0x8] ; load 4 ref pixels (1B) 49 ldr lr, [r2, #0xC] ; load 4 ref pixels (1B) 52 add r2, r2, r3 ; set dst pointer to next row 55 pld [r2, r3, lsl #1] 65 ldr r8, [r2, #0x0] ; load 4 ref pixels (2A) [all …]
|
D | sixtappredict8x4_v6.asm | 18 ; r2 int xoffset, 30 cmp r2, #0 ;skip first_pass filter if xoffset=0 41 add r2, r12, r2, lsl #4 ;calculate filter location 44 ldr r3, [r2] ; load up packed filter coefficients 45 ldr r4, [r2, #4] 46 ldr r5, [r2, #8] 48 mov r2, #0x90000 ; height=9 is top part of counter 59 orr r2, r2, #0x4 ; construct loop counter. width=8=4x2 82 sub r2, r2, #1 85 tst r2, #0xff ; test loop counter [all …]
|
D | iwalsh_v6.asm | 24 ldr r2, [r0, #0] ; [1 | 0] 33 qadd16 r10, r2, r8 ; a1 [1+13 | 0+12] 36 qsub16 lr, r2, r8 ; d1 [1-13 | 0-12] 38 qadd16 r2, r10, r11 ; a1 + b1 [1 | 0] 55 qsubaddx r10, r2, r3 ; [c1|a1] [1-2 | 0+3] 56 qaddsubx r11, r2, r3 ; [b1|d1] [1+2 | 0-3] 60 qaddsubx r2, r10, r11 ; [b2|c2] [c1+d1 | a1-b1] 66 qadd16 r2, r2, r10 ; [b2+3|c2+3] 73 asr lr, r2, #19 ; [1] 75 sxth r2, r2 [all …]
|
/hardware/samsung_slsi/exynos5/libswconverter/ |
D | csc_tiled_to_linear_y_neon.s | 57 @r2 width 74 bic r10, r2, #0xF @ aligned_width = width & (~0xF) 75 add r11, r2, #15 @ tiled_width = ((width + 15) >> 4) << 4 91 mul r12, r2, r5 @ temp1 = width * i + j; 101 vst1.8 {q0}, [r7], r2 102 vst1.8 {q1}, [r7], r2 103 vst1.8 {q2}, [r7], r2 104 vst1.8 {q3}, [r7], r2 105 vst1.8 {q4}, [r7], r2 106 vst1.8 {q5}, [r7], r2 [all …]
|
D | csc_tiled_to_linear_uv_neon.s | 56 @r2 width 73 bic r10, r2, #0xF @ aligned_width = width & (~0xF) 74 add r11, r2, #15 @ tiled_width = ((width + 15) >> 4) << 4 86 mul r12, r2, r5 @ temp1 = width * i + j; 94 vst1.8 {q0}, [r7], r2 95 vst1.8 {q1}, [r7], r2 96 vst1.8 {q2}, [r7], r2 97 vst1.8 {q3}, [r7], r2 98 vst1.8 {q4}, [r7], r2 99 vst1.8 {q5}, [r7], r2 [all …]
|
/hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/encoder/arm/armv5te/ |
D | boolhuff_armv5te.asm | 32 ldr r2, [r0, #vp8_writer_buffer_end] 42 ; r2 unsigned char *source_end 44 str r2, [r0, #vp8_writer_buffer_end] 47 mvn r2, #23 50 str r2, [r0, #vp8_writer_count] 58 ; r2 int probability 62 mov r4, r2 64 ldr r2, [r0, #vp8_writer_lowvalue] 76 addne r2, r2, r4 ; if (bit) lowvalue += split 91 lsls r4, r2, r4 ; if((lowvalue<<(offset-1)) & 0x80000000 ) [all …]
|
D | vp8_packtokens_armv5.asm | 30 ldr r2, [r0, #vp8_writer_buffer_end] 41 ; r2 int xcount 51 add r2, r1, r2, lsl #3 ; stop = p + xcount*sizeof(TOKENEXTRA) 52 str r2, [sp, #0] 54 ldr r2, [r0, #vp8_writer_lowvalue] 101 addcs r2, r2, r4 ; if (bb) lowvalue += split 116 lsls r4, r2, r4 ; if((lowvalue<<(offset-1)) & 0x80000000 ) 140 lsr r7, r2, r4 ; lowvalue >> (24-offset) 142 lsl r2, r2, r6 ; lowvalue <<= offset 145 bic r2, r2, #0xff000000 ; lowvalue &= 0xffffff [all …]
|
D | vp8_packtokens_partitions_armv5.asm | 29 ldr r2, [r0, #vp8_writer_buffer_end] 39 ; r2 const unsigned char *cx_data_end 59 str r2, [sp, #8] ; save cx_data_end 73 ldr r2, _vp8_writer_sz_ ; load up sizeof(vp8_writer) 74 add r0, r2 ; bc[i + 1] 89 mov r2, #0 ; vp8_writer_lowvalue 93 str r2, [r0, #vp8_writer_pos] 151 addcs r2, r2, r4 ; if (bb) lowvalue += split 166 lsls r4, r2, r4 ; if((lowvalue<<(offset-1)) & 0x80000000 ) 190 lsr r7, r2, r4 ; lowvalue >> (24-offset) [all …]
|
/hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp9/common/arm/neon/ |
D | vp9_idct16x16_1_add_neon.asm | 23 ; r2 int dest_stride) 49 sub r2, #8 53 vld1.64 {d3}, [r1], r2 55 vld1.64 {d5}, [r1], r2 57 vld1.64 {d7}, [r1], r2 59 vld1.64 {d17}, [r1], r2 70 vst1.64 {d3}, [r12], r2 72 vst1.64 {d31}, [r12], r2 83 vst1.64 {d3}, [r12], r2 85 vst1.64 {d31}, [r12], r2 [all …]
|
D | vp9_idct8x8_1_add_neon.asm | 23 ; r2 int dest_stride) 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 67 vst1.64 {d2}, [r12], r2 68 vst1.64 {d3}, [r12], r2 [all …]
|
D | vp9_loopfilter_neon.asm | 33 ; r2 const uint8_t *blimit, 40 vld1.8 {d0[]}, [r2] ; duplicate *blimit 42 ldr r2, [sp, #4] ; load thresh 49 vld1.8 {d2[]}, [r2] ; duplicate *thresh 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 57 vld1.u8 {d5}, [r2@64], r1 ; p1 59 vld1.u8 {d7}, [r2@64], r1 ; q0 61 vld1.u8 {d17}, [r2@64] ; q2 [all …]
|
/hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vpx_scale/arm/neon/ |
D | vp8_vpxyv12_copysrcframe_func_neon.asm | 36 ldr r2, [r0, #yv12_buffer_config_y_buffer] ;srcptr1 39 add r10, r2, r6 ;second row src 53 vld1.8 {q0, q1}, [r2]! 55 vld1.8 {q2, q3}, [r2]! 57 vld1.8 {q8, q9}, [r2]! 59 vld1.8 {q10, q11}, [r2]! 77 vld1.8 {d0}, [r2]! 89 ldrb r8, [r2], #1 98 add r2, r2, r6 110 vld1.8 {q0, q1}, [r2]! [all …]
|
D | vp8_vpxyv12_copyframe_func_neon.asm | 40 ldr r2, [r0, #yv12_buffer_config_y_buffer] ;srcptr1 52 mov r8, r2 54 add r10, r2, r6 83 add r2, r2, r6, lsl #1 91 ldr r2, [r0, #yv12_buffer_config_y_buffer] ;srcptr1 97 ldr r2, [sp] ;srcptr1 112 mov r8, r2 114 add r10, r2, r6 135 add r2, r2, r6, lsl #1 143 ldr r2, [sp] ;srcptr1 [all …]
|
/hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/encoder/arm/neon/ |
D | subtract_neon.asm | 37 vld1.8 {d1}, [r7], r2 ;load pred 39 vld1.8 {d3}, [r7], r2 41 vld1.8 {d5}, [r7], r2 43 vld1.8 {d7}, [r7], r2 50 mov r2, r2, lsl #1 52 vst1.16 {d20}, [r5], r2 ;store diff 53 vst1.16 {d22}, [r5], r2 54 vst1.16 {d24}, [r5], r2 55 vst1.16 {d26}, [r5], r2 74 vld1.8 {q0}, [r1], r2 ;load src [all …]
|