| /external/libhevc/common/arm/ |
| ihevc_deblk_luma_horz.s | 201 vaddw.u8 q4,q3,d27
205 vmul.i16 q6,q4,d0[0]
222 vaddw.u8 q7,q4,d28
271 vaddw.u8 q4,q3,d26
277 vmul.i16 q6,q4,d0[0]
289 vaddw.u8 q7,q4,d23
373 vaddl.u8 q4,d23,d22
388 vmul.i16 q4,q4,d0[0]
390 vadd.i16 q8,q4,q
[all...] |
| ihevc_weighted_pred_bi_default.s | 289 vqadd.s16 q4,q6,q7 @vaddq_s32(i4_tmp2_t1, i4_tmp2_t2) iv iteration
291 vqadd.s16 q4,q4,q0
293 vqshrun.s16 d31,q4,#7
393 vld1.s16 {q4},[r1]! @load and increment the pi2_src2 ii iteration
401 vqadd.s16 q12,q3,q4
417 vld1.s16 {q4},[r1]! @load and increment the pi2_src2 ii iteration
435 vqadd.s16 q12,q3,q4
456 vld1.s16 {q4},[r1]! @load and increment the pi2_src2 ii iteration
|
| /external/libavc/common/arm/ |
| ih264_iquant_itrans_recon_a9.s | 525 vld1.32 {q4}, [r6]! @ Scaling factors row 4
539 vmul.s16 q10, q4, q13 @ Dequant row4*scale matrix row 4
547 vmull.s16 q4, d28, d20 @ i = 32..35
559 vshl.s32 q4, q4, q15 @
567 vqrshrn.s32 d8, q4, #0x6 @ D8 = c[i] = ((q[i] + 32) >> 6) where i = 32..35
582 vzip.s16 q4, q5 @Transpose
598 vzip.s32 q4, q6 @
603 vswp d5, d12 @ Q4/Q5 = Row order x4/x5
606 vswp q1, q4
[all...] |
| ih264_inter_pred_luma_vert_qpel_a9q.s | 133 vld1.u32 {q4}, [r0], r2 @ Vector load from src[4_0]
226 vaddl.u8 q4, d0, d5 @ temp = src[0_0] + src[5_0]
228 vmla.u16 q4, q3, q11 @ temp += temp1 * 20
233 vmls.u16 q4, q5, q12 @ temp -= temp2 * 5
240 vqrshrun.s16 d26, q4, #5 @ dst[0_0] = CLIP_U8( (temp + 16) >> 5)
247 vrhadd.u8 q13, q4, q13 @ Interpolation step for qpel calculation
281 vaddl.u8 q4, d0, d5 @ temp = src[0_0] + src[5_0]
283 vmla.u16 q4, q3, q11 @ temp += temp1 * 20
288 vmls.u16 q4, q5, q12 @ temp -= temp2 * 5
295 vqrshrun.s16 d26, q4, #5 @ dst[0_0] = CLIP_U8( (temp + 16) >> 5
[all...] |
| ih264_default_weighted_pred_a9q.s | 183 vld1.8 {q4}, [r0], r3 @load row 5 in source 1
188 vrhadd.u8 q4, q4, q12
202 vst1.8 {q4}, [r2], r5 @load row 5 in destination
333 vld1.8 {q4}, [r1], r4 @load row 1 in source 2
335 vrhadd.u8 q0, q0, q4
|
| ih264_ihadamard_scaling_a9.s | 120 vsub.s32 q4, q12, q13 @pi4_tmp_ptr[2] = x0 - x1
124 vtrn.32 q4, q5
126 vswp d5, d8 @Q2 = x4, Q4 = x6
131 vadd.s32 q13, q3, q4 @x1 = x5+x6
132 vsub.s32 q14, q3, q4 @x2 = x5-x6
|
| /external/libvpx/libvpx/vpx_dsp/arm/ |
| highbd_loopfilter_neon.c | 93 const uint16x8_t q3, const uint16x8_t q4,
103 flat2 = vmaxq_u16(flat2, vabdq_u16(q4, q0));
196 const uint16x8_t q3, const uint16x8_t q4, const uint16x8_t q5,
217 *op2 = apply_15_tap_filter_kernel(flat2, p7, p3, p2, q4, *op2, &sum);
224 *oq4 = apply_15_tap_filter_kernel(flat2, p3, q3, q4, q7, q4, &sum);
225 *oq5 = apply_15_tap_filter_kernel(flat2, p2, q4, q5, q7, q5, &sum);
320 const uint16x8_t q4, const uint16x8_t q5, const uint16x8_t q6,
338 q4, q5, q6, q7, op6, op5, op4, op3, op2, op1, op0,
542 const uint16x8_t q4, const uint16x8_t q5
670 q3, q4, q5, q6, q7, op6, op5, op4, op3, op2, op1, op0, oq0, oq1, oq2, oq3, local
693 q3, q4, q5, q6, q7, op6, op5, op4, op3, op2, op1, op0, oq0, oq1, oq2, oq3, local
[all...] |
| loopfilter_16_neon.asm | 54 vld1.u8 {d12}, [r8@64], r1 ; q4
192 vtrn.32 q4, q6
197 vtrn.16 q4, q5
364 ; d12 q4
425 ; flatmask5(1, p7, p6, p5, p4, p0, q0, q4, q5, q6, q7)
427 vabd.u8 d23, d12, d8 ; abs(q4 - q0)
436 vmax.u8 d22, d22, d23 ; max(abs(p4 - p0), abs(q4 - q0))
593 vaddw.u8 q15, d12 ; op2 += q4
621 vaddl.u8 q4, d10, d15
628 vadd.i16 q15, q4
[all...] |
| /frameworks/native/libs/math/tests/ |
| quat_test.cpp | 71 quatd q4(q3.xyz, 42);
72 EXPECT_EQ(q4.x, 2);
73 EXPECT_EQ(q4.y, 3);
74 EXPECT_EQ(q4.z, 4);
75 EXPECT_EQ(q4.w, 42);
|
| /external/libvpx/libvpx/vpx_dsp/mips/ |
| loopfilter_filters_dspr2.h | 514 const uint32_t q4 = *oq4, q5 = *oq5, q6 = *oq6, q7 = *oq7; local
522 /* addition of p6,p5,p4,p3,p2,p1,p0,q0,q1,q2,q3,q4,q5,q6
534 "addu.ph %[add_p6toq6], %[add_p6toq6], %[q4] \n\t"
542 [q3] "r"(q3), [q4] "r"(q4), [q5] "r"(q5), [q6] "r"(q6),
555 "subu.ph %[res_op6], %[res_op6], %[q4] \n\t"
569 "subu.ph %[res_op5], %[res_op5], %[q4] \n\t"
581 "subu.ph %[res_op4], %[res_op4], %[q4] \n\t"
591 "subu.ph %[res_op3], %[res_op3], %[q4] \n\t"
597 p0 + q0 + q1 + q2 + q3 + q4, 4) *
[all...] |
| loopfilter_mb_vert_dspr2.c | 30 uint32_t p7, p6, p5, p4, p3, p2, p1, p0, q0, q1, q2, q3, q4, q5, q6, q7; local
84 "lw %[q4], +4(%[s4]) \n\t"
87 [q7] "=&r"(q7), [q6] "=&r"(q6), [q5] "=&r"(q5), [q4] "=&r"(q4)
198 /* transpose q4, q5, q6, q7
204 q4 q4_0 q4_1 q4_2 q4_3
211 q4 q4_0 q5_0 q26_0 q7_0
216 "precrq.qb.ph %[prim3], %[q5], %[q4] \n\t"
217 "precr.qb.ph %[prim4], %[q5], %[q4] \n\t"
220 "precr.qb.ph %[q4], %[prim1], %[prim2] \n\t
[all...] |
| /external/llvm/test/MC/ARM/ |
| neon-mul-encoding.s | 118 vmul.s32 q4, d3[1]
134 vmul.s32 q5, q4, d3[1]
135 vmul.u32 q4, q5, d4[0]
150 @ CHECK: vmul.i32 q4, q4, d3[1] @ encoding: [0x63,0x88,0xa8,0xf3]
166 @ CHECK: vmul.i32 q5, q4, d3[1] @ encoding: [0x63,0xa8,0xa8,0xf3]
167 @ CHECK: vmul.i32 q4, q5, d4[0] @ encoding: [0x44,0x88,0xaa,0xf3]
|
| fullfp16-neon-neg.s | 22 vmul.f16 q4, q5, d6[3]
57 vceq.f16 q2, q3, q4
67 vcge.f16 q2, q3, q4
77 vcgt.f16 q2, q3, q4
87 vcle.f16 q2, q3, q4
97 vclt.f16 q2, q3, q4
|
| neon-sub-encoding.s | 22 vsub.i64 q4, q7
44 @ CHECK: vsub.i64 q4, q4, q7 @ encoding: [0x4e,0x88,0x38,0xf3]
145 vhsub.u8 q4, q9
158 @ CHECK: vhsub.u8 q4, q4, q9 @ encoding: [0x62,0x82,0x08,0xf3]
|
| fullfp16-neon.s | 26 vmul.f16 q4, q5, d6[3]
28 @ ARM: vmul.f16 q4, q5, d6[3] @ encoding: [0x6e,0x89,0x9a,0xf3]
30 @ THUMB: vmul.f16 q4, q5, d6[3] @ encoding: [0x9a,0xff,0x6e,0x89]
75 vceq.f16 q2, q3, q4
77 @ ARM: vceq.f16 q2, q3, q4 @ encoding: [0x48,0x4e,0x16,0xf2]
79 @ THUMB: vceq.f16 q2, q3, q4 @ encoding: [0x16,0xef,0x48,0x4e]
89 vcge.f16 q2, q3, q4
91 @ ARM: vcge.f16 q2, q3, q4 @ encoding: [0x48,0x4e,0x16,0xf3]
93 @ THUMB: vcge.f16 q2, q3, q4 @ encoding: [0x16,0xff,0x48,0x4e]
103 vcgt.f16 q2, q3, q4
[all...] |
| /external/boringssl/ios-arm/crypto/fipsmodule/ |
| aesv8-armx32.S | 471 veor q4,q6,q7
513 veor q4,q4,q0
517 vst1.8 {q4},[r1]!
651 .byte 0x80,0x83,0xb0,0xf3 @ aesmc q4,q0
660 .byte 0x22,0x83,0xb0,0xf3 @ aese q4,q9
661 .byte 0x88,0x83,0xb0,0xf3 @ aesmc q4,q4
670 .byte 0x28,0x83,0xb0,0xf3 @ aese q4,q12
671 .byte 0x88,0x83,0xb0,0xf3 @ aesmc q4,q
[all...] |
| /external/boringssl/ios-arm/crypto/chacha/ |
| chacha-armv4.S | 828 vmov q4,q0
840 vmov q4,q0
866 vadd.i32 q4,q4,q5
872 veor q7,q7,q4
908 vadd.i32 q4,q4,q5
914 veor q13,q7,q4
974 vadd.i32 q4,q4,q
[all...] |
| /external/boringssl/linux-arm/crypto/chacha/ |
| chacha-armv4.S | 825 vmov q4,q0
837 vmov q4,q0
863 vadd.i32 q4,q4,q5
869 veor q7,q7,q4
905 vadd.i32 q4,q4,q5
911 veor q13,q7,q4
971 vadd.i32 q4,q4,q
[all...] |
| /external/libavc/encoder/arm/ |
| ih264e_evaluate_intra4x4_modes_a9q.s | 176 vmov.32 q4, q3
177 vtrn.8 q3, q4
200 vext.8 q4, q0, q0, #5
201 vaddl.u8 q4, d0, d8
217 vdup.8 q4, r4
499 vmov.32 q15, q4
|
| ih264e_half_pel.s | 106 vaddl.u8 q4, d31, d2 @// a0 + a5 (column1,row0)
119 vmlal.u8 q4, d31, d1 @// a0 + a5 + 20a2 (column1,row0)
132 vmlal.u8 q4, d31, d1 @// a0 + a5 + 20a2 + 20a3 (column1,row0)
145 vmlsl.u8 q4, d31, d0 @// a0 + a5 + 20a2 + 20a3 - 5a1 (column1,row0)
158 vmlsl.u8 q4, d31, d0 @// a0 + a5 + 20a2 + 20a3 - 5a1 - 5a4 (column1,row0)
170 vqrshrun.s16 d20, q4, #5 @// (a0 + a5 + 20a2 + 20a3 - 5a1 - 5a4 + 16) >> 5 (column1,row0)
558 vaddl.s16 q4, d31, d22 @// a0 + a5 (set3)
560 vmlal.s16 q4, d30, d0[1] @// a0 + a5 + 20a2 (set3)
561 vmlal.s16 q4, d29, d0[1] @// a0 + a5 + 20a2 + 20a3 (set3)
562 vmlsl.s16 q4, d28, d0[0] @// a0 + a5 + 20a2 + 20a3 - 5a1 (set3
[all...] |
| /device/linaro/bootloader/edk2/ArmPkg/Library/BaseMemoryLibVstm/Arm/ |
| SetMem.S | 61 vmov q4,q0
|
| SetMem.asm | 59 vmov q4,q0
|
| /frameworks/av/media/libstagefright/codecs/on2/h264dec/omxdl/arm_neon/vc/m4p10/src_gcc/ |
| omxVCM4P10_PredictIntraChroma_8x8_s.S | 192 VDUP.16 q4,d6[2]
200 VADD.I16 q4,q2,q4
208 VQRSHRUN.S16 d2,q4,#5
|
| omxVCM4P10_DequantTransformResidualFromPairAndAdd_s.S | 122 VADDW.U8 q4,q1,d5
124 VQMOVUN.S16 d1,q4
|
| /toolchain/binutils/binutils-2.25/gas/testsuite/gas/arm/ |
| neon-psyn.s | 61 el6 .qn q4.16
|
