/hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/common/arm/armv6/ |
D | copymem16x16_v6.asm | 25 pld [r0, #31] ; preload for next 16x16 block 27 ands r4, r0, #15 30 ands r4, r0, #7 33 ands r4, r0, #3 37 ldrb r4, [r0] 38 ldrb r5, [r0, #1] 39 ldrb r6, [r0, #2] 40 ldrb r7, [r0, #3] 50 ldrb r4, [r0, #4] 51 ldrb r5, [r0, #5] [all …]
|
D | copymem8x4_v6.asm | 25 pld [r0] 26 pld [r0, r1] 27 pld [r0, r1, lsl #1] 29 ands r4, r0, #7 32 ands r4, r0, #3 36 ldrb r4, [r0] 37 ldrb r5, [r0, #1] 45 ldrb r4, [r0, #2] 46 ldrb r5, [r0, #3] 53 ldrb r4, [r0, #4] [all …]
|
D | copymem8x8_v6.asm | 25 pld [r0] 26 pld [r0, r1] 27 pld [r0, r1, lsl #1] 29 ands r4, r0, #7 32 ands r4, r0, #3 36 ldrb r4, [r0] 37 ldrb r5, [r0, #1] 45 ldrb r4, [r0, #2] 46 ldrb r5, [r0, #3] 53 ldrb r4, [r0, #4] [all …]
|
D | sixtappredict8x4_v6.asm | 16 ; r0 unsigned char *src_ptr, 36 sub r0, r0, r1, lsl #1 39 pld [r0, r3] 42 add r0, r0, #3 ;adjust src only for loading convinience 53 ldrb r6, [r0, #-5] ; load source data 54 ldrb r7, [r0, #-4] 55 ldrb r8, [r0, #-3] 56 ldrb r9, [r0, #-2] 57 ldrb r10, [r0, #-1] 71 ldrb r6, [r0], #1 [all …]
|
D | filter_v6.asm | 22 ; r0 unsigned char *src_ptr 54 ldrb r8, [r0, #-2] ; load source data 55 ldrb r9, [r0, #-1] 56 ldrb r10, [r0], #2 60 ldrb r11, [r0, #-1] 65 ldrb r9, [r0] 73 ldrb r10, [r0, #1] 75 ldrb r11, [r0, #2] 88 ldrneb r8, [r0, #-2] ; load data for next loop 91 ldrneb r9, [r0, #-1] [all …]
|
D | bilinearfilter_v6.asm | 18 ; r0 unsigned char *src_ptr, 35 pld [r0, r7] 50 ldrb r6, [r0] ; load source data 51 ldrb r7, [r0, #1] 52 ldrb r8, [r0, #2] 56 ldrb r9, [r0, #3] 57 ldrb r10, [r0, #4] 70 add r0, r0, #4 88 ldrneb r6, [r0] ; load source data 91 ldrneb r7, [r0, #1] [all …]
|
D | dequant_idct_v6.asm | 15 ; r0 = q 23 ldr r4, [r0] ;input 35 ldr r4, [r0, #4] ;input 38 strh r6, [r0], #2 39 strh r7, [r0], #2 46 ldrne r4, [r0, #4] 49 strh r6, [r0], #2 50 strh r7, [r0], #2 54 sub r0, r0, #32 55 mov r1, r0 [all …]
|
D | dc_only_idct_add_v6.asm | 17 ; r0 input_dc 26 add r0, r0, #4 ; input_dc += 4 29 and r0, r12, r0, asr #3 ; input_dc >> 3 + mask 31 orr r0, r0, r0, lsl #16 ; a1 | a1 35 uxtab16 r5, r0, r4 ; a1+2 | a1+0 36 uxtab16 r4, r0, r4, ror #8 ; a1+3 | a1+1 37 uxtab16 r7, r0, r6 38 uxtab16 r6, r0, r6, ror #8 50 uxtab16 r5, r0, r4 51 uxtab16 r4, r0, r4, ror #8 [all …]
|
D | idct_v6.asm | 19 ; r0 short* input 38 ldr r6, [r0, #(4*2)] ; i5 | i4 39 ldr r12, [r0, #(12*2)] ; i13|i12 40 ldr r14, [r0, #(8*2)] ; i9 | i8 59 ldr r11, [r0] ; i1 | i0 74 str r6, [r0, #(4*2)] ; o5 | o4 75 str r7, [r0, #(8*2)] ; o9 | o8 76 str r10,[r0, #(12*2)] ; o13|o12 77 str r9, [r0], #4 ; o1 | o0 81 sub r0, r0, #8 ; reset input/output [all …]
|
/hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/common/ppc/ |
D | recon_altivec.asm | 21 vaddshs v2, v2, v3 ;# v2 = r0..r7 26 vpkshus v2, v2, v3 ;# v2 = 8-bit r0..r15 38 mfspr r0, 256 ;# get old VRSAVE 39 stw r0, -8(r1) ;# save old VRSAVE to stack 40 oris r0, r0, 0xf000 41 mtspr 256,r0 ;# set VRSAVE 60 vaddshs v2, v2, v3 ;# v2 = r0..r7 64 vpkshus v2, v2, v3 ;# v3 = 8-bit r0..r15 66 lwz r0, 0(r10) 68 stw r0, 0(\Dst) [all …]
|
/hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp9/common/arm/neon/ |
D | vp9_reconintra_neon.asm | 32 ; r0 uint8_t *dst 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 49 ; r0 uint8_t *dst 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_idct32x32_1_add_neon.asm | 70 ; r0 int16_t input 78 ldrsh r0, [r0] 85 mul r0, r0, r12 ; input[0] * cospi_16_64 86 add r0, r0, #0x2000 ; +(1 << ((DCT_CONST_BITS) - 1)) 87 asr r0, r0, #14 ; >> DCT_CONST_BITS 90 mul r0, r0, r12 ; out * cospi_16_64 92 add r0, r0, #0x2000 ; +(1 << ((DCT_CONST_BITS) - 1)) 93 asr r0, r0, #14 ; >> DCT_CONST_BITS 96 add r0, r0, #32 ; + (1 <<((6) - 1)) 97 asrs r0, r0, #6 ; >> 6 [all …]
|
D | vp9_idct16x16_1_add_neon.asm | 21 ; r0 int16_t input 26 ldrsh r0, [r0] 33 mul r0, r0, r12 ; input[0] * cospi_16_64 34 add r0, r0, #0x2000 ; +(1 << ((DCT_CONST_BITS) - 1)) 35 asr r0, r0, #14 ; >> DCT_CONST_BITS 38 mul r0, r0, r12 ; out * cospi_16_64 40 add r0, r0, #0x2000 ; +(1 << ((DCT_CONST_BITS) - 1)) 41 asr r0, r0, #14 ; >> DCT_CONST_BITS 44 add r0, r0, #32 ; + (1 <<((6) - 1)) 45 asr r0, r0, #6 ; >> 6 [all …]
|
D | vp9_idct4x4_1_add_neon.asm | 21 ; r0 int16_t input 26 ldrsh r0, [r0] 33 mul r0, r0, r12 ; input[0] * cospi_16_64 34 add r0, r0, #0x2000 ; +(1 << ((DCT_CONST_BITS) - 1)) 35 asr r0, r0, #14 ; >> DCT_CONST_BITS 38 mul r0, r0, r12 ; out * cospi_16_64 40 add r0, r0, #0x2000 ; +(1 << ((DCT_CONST_BITS) - 1)) 41 asr r0, r0, #14 ; >> DCT_CONST_BITS 44 add r0, r0, #8 ; + (1 <<((4) - 1)) 45 asr r0, r0, #4 ; >> 4 [all …]
|
D | vp9_dc_only_idct_add_neon.asm | 21 ; r0 int input_dc 34 mul r0, r0, r12 ; input_dc * cospi_16_64 35 add r0, r0, #0x2000 ; +(1 << ((DCT_CONST_BITS) - 1)) 36 asr r0, r0, #14 ; >> DCT_CONST_BITS 39 mul r0, r0, r12 ; out * cospi_16_64 40 add r0, r0, #0x2000 ; +(1 << ((DCT_CONST_BITS) - 1)) 41 asr r0, r0, #14 ; >> DCT_CONST_BITS 44 add r0, r0, #8 ; + (1 <<((4) - 1)) 45 asr r0, r0, #4 ; >> 4 47 vdup.16 q0, r0; ; duplicate a1
|
D | vp9_idct8x8_1_add_neon.asm | 21 ; r0 int16_t input 26 ldrsh r0, [r0] 33 mul r0, r0, r12 ; input[0] * cospi_16_64 34 add r0, r0, #0x2000 ; +(1 << ((DCT_CONST_BITS) - 1)) 35 asr r0, r0, #14 ; >> DCT_CONST_BITS 38 mul r0, r0, r12 ; out * cospi_16_64 40 add r0, r0, #0x2000 ; +(1 << ((DCT_CONST_BITS) - 1)) 41 asr r0, r0, #14 ; >> DCT_CONST_BITS 44 add r0, r0, #16 ; + (1 <<((5) - 1)) 45 asr r0, r0, #5 ; >> 5 [all …]
|
D | vp9_copy_neon.asm | 34 pld [r0, r1, lsl #1] 35 vld1.8 {q0-q1}, [r0]! 36 vld1.8 {q2-q3}, [r0], lr 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] [all …]
|
/hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/common/arm/neon/ |
D | idct_dequant_0_2x_neon.asm | 19 ; r0 *q 36 ldrh r12, [r0] ; lo q 37 ldrh r4, [r0, #32] ; hi q 39 strh r5, [r0] 40 strh r5, [r0, #32] 43 mul r0, r12, r1 44 add r0, r0, #4 45 asr r0, r0, #3 46 vdup.16 q0, r0 48 mul r0, r4, r1 [all …]
|
D | buildintrapredictorsmby_neon.asm | 20 ; r0 unsigned char *y_buffer 58 sub r6, r0, r2 76 sub r0, r0, #1 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 [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 71 vmov.32 r0, d0[0] 85 vld1.8 {d0}, [r0], r1 [all …]
|
D | mbloopfilter_neon.asm | 24 ; r0 unsigned char *src, 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 39 vld1.u8 {q5}, [r0@128], r1 ; p1 41 vld1.u8 {q7}, [r0@128], r1 ; q0 43 vld1.u8 {q9}, [r0@128], r1 ; q2 49 add r0, r12, r1, lsr #1 52 vst1.u8 {q5}, [r0@128],r1 ; store op1 54 vst1.u8 {q7}, [r0@128],r1 ; store oq0 [all …]
|
D | loopfiltersimpleverticaledge_neon.asm | 20 ; r0 unsigned char *s, PRESERVE 25 sub r0, r0, #2 ; move src pointer down by 2 columns 27 add r3, r0, r1 29 vld4.8 {d6[0], d7[0], d8[0], d9[0]}, [r0], r12 31 vld4.8 {d6[2], d7[2], d8[2], d9[2]}, [r0], r12 33 vld4.8 {d6[4], d7[4], d8[4], d9[4]}, [r0], r12 35 vld4.8 {d6[6], d7[6], d8[6], d9[6]}, [r0], r12 38 vld4.8 {d10[0], d11[0], d12[0], d13[0]}, [r0], r12 40 vld4.8 {d10[2], d11[2], d12[2], d13[2]}, [r0], r12 42 vld4.8 {d10[4], d11[4], d12[4], d13[4]}, [r0], r12 [all …]
|
D | sad16_neon.asm | 21 ; r0 unsigned char *src_ptr 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/vp8/encoder/arm/armv6/ |
D | vp8_fast_quantize_b_armv6.asm | 22 ; r0 BLOCK *b 27 ldr r3, [r0, #vp8_block_coeff] ; coeff 28 ldr r4, [r0, #vp8_block_quant_fast] ; quant_fast 29 ldr r5, [r0, #vp8_block_round] ; round 44 ldr r10, [r5], #4 ; [r1 | r0] 50 sadd16 r9, r9, r10 ; [x1+r1 | x0+r0] 54 smulbb r0, r9, r11 ; [(x0+r0)*q0] 61 pkhtb r0, r9, r0, asr #16 ; [y1 | y0] 67 eor r0, r0, lr ; [(y1 ^ sz1) | (y0 ^ sz0)] 72 ssub16 r0, r0, lr ; x = (y ^ sz) - sz [all …]
|
/hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/encoder/arm/armv5te/ |
D | boolhuff_armv5te.asm | 27 ; needs vp8_writer in r0 31 push {r0-r3, r12, lr} ; rest of regs are preserved by subroutine call 32 ldr r2, [r0, #vp8_writer_buffer_end] 33 ldr r3, [r0, #vp8_writer_error] 35 mov r0, $start 37 pop {r0-r3, r12, lr} 40 ; r0 BOOL_CODER *br 44 str r2, [r0, #vp8_writer_buffer_end] 48 str r12, [r0, #vp8_writer_lowvalue] 49 str r3, [r0, #vp8_writer_range] [all …]
|