| /external/llvm/test/MC/ARM/ |
| H A D | thumb-neon-crypto.s | 3 aesd.8 q0, q1
4 @ CHECK: aesd.8 q0, q1 @ encoding: [0xb0,0xff,0x42,0x03]
5 aese.8 q0, q1
6 @ CHECK: aese.8 q0, q1 @ encoding: [0xb0,0xff,0x02,0x03]
7 aesimc.8 q0, q1
8 @ CHECK: aesimc.8 q0, q1 @ encoding: [0xb0,0xff,0xc2,0x03]
9 aesmc.8 q0, q1
10 @ CHECK: aesmc.8 q0, q1 @ encoding: [0xb0,0xff,0x82,0x03]
12 sha1h.32 q0, q1
13 @ CHECK: sha1h.32 q0, q
[all...] |
| H A D | neon-crypto.s | 4 aesd.8 q0, q1
5 aese.8 q0, q1
6 aesimc.8 q0, q1
7 aesmc.8 q0, q1
8 @ CHECK: aesd.8 q0, q1 @ encoding: [0x42,0x03,0xb0,0xf3]
9 @ CHECK: aese.8 q0, q1 @ encoding: [0x02,0x03,0xb0,0xf3]
10 @ CHECK: aesimc.8 q0, q1 @ encoding: [0xc2,0x03,0xb0,0xf3]
11 @ CHECK: aesmc.8 q0, q1 @ encoding: [0x82,0x03,0xb0,0xf3]
17 sha1h.32 q0, q1
18 sha1su1.32 q0, q
[all...] |
| H A D | directive-arch_extension-crypto.s | 19 vmull.p64 q0, d0, d1
22 aesd.8 q0, q1
24 aese.8 q0, q1
26 aesimc.8 q0, q1
28 aesmc.8 q0, q1
31 sha1h.32 q0, q1
33 sha1su1.32 q0, q1
35 sha256su0.32 q0, q1
38 sha1c.32 q0, q1, q2
40 sha1m.32 q0, q
[all...] |
| H A D | invalid-neon-v8.s | 28 aesd.8 q0, s1
34 aesmc.8 q0, d1
36 aesdge.8 q0, q1
41 sha1su1.32 q0, s1
45 sha1heq.32 q0, q1
50 sha1m.32 q0, s1, q2
56 sha256h.32 q0, s1, q2
58 sha256h2.32 q0, q1, s2
62 sha256su1lt.32 q0, d1, q2
65 vmull.p64 q0, s
[all...] |
| /external/chromium_org/third_party/skia/src/opts/ |
| H A D | memset32_neon.S | 31 vdup.32 q0, r1
32 vmov q1, q0
43 vst1.64 {q0, q1}, [r0]!
44 vst1.64 {q0, q1}, [r0]!
45 vst1.64 {q0, q1}, [r0]!
46 vst1.64 {q0, q1}, [r0]!
47 vst1.64 {q0, q1}, [r0]!
48 vst1.64 {q0, q1}, [r0]!
49 vst1.64 {q0, q1}, [r0]!
50 vst1.64 {q0, q
[all...] |
| H A D | memset16_neon.S | 78 * build q0 and q1 depending on if we'll need it, so that's
87 vmov q1, q0
93 vst1.64 {q0, q1}, [r0]!
94 vst1.64 {q0, q1}, [r0]!
95 vst1.64 {q0, q1}, [r0]!
96 vst1.64 {q0, q1}, [r0]!
106 vst1.64 {q0, q1}, [r0]!
115 vst1.32 {q0}, [r0]!
|
| /external/skia/src/opts/ |
| H A D | memset32_neon.S | 32 vdup.32 q0, r1
33 vmov q1, q0
44 vst1.64 {q0, q1}, [r0]!
45 vst1.64 {q0, q1}, [r0]!
46 vst1.64 {q0, q1}, [r0]!
47 vst1.64 {q0, q1}, [r0]!
48 vst1.64 {q0, q1}, [r0]!
49 vst1.64 {q0, q1}, [r0]!
50 vst1.64 {q0, q1}, [r0]!
51 vst1.64 {q0, q
[all...] |
| H A D | memset16_neon.S | 77 * build q0 and q1 depending on if we'll need it, so that's
86 vmov q1, q0
92 vst1.64 {q0, q1}, [r0]!
93 vst1.64 {q0, q1}, [r0]!
94 vst1.64 {q0, q1}, [r0]!
95 vst1.64 {q0, q1}, [r0]!
105 vst1.64 {q0, q1}, [r0]!
114 vst1.32 {q0}, [r0]!
|
| /external/openssl/crypto/aes/asm/ |
| H A D | aesv8-armx.S | 21 veor q0,q0,q0
34 vext.8 q9,q0,q3,#12
36 .byte 0x00,0x43,0xf0,0xf3 @ aese q10,q0
40 vext.8 q9,q0,q9,#12
42 vext.8 q9,q0,q9,#12
53 vext.8 q9,q0,q3,#12
55 .byte 0x00,0x43,0xf0,0xf3 @ aese q10,q0
58 vext.8 q9,q0,q
[all...] |
| /external/chromium_org/third_party/libvpx/source/libvpx/vp9/common/arm/neon/ |
| H A D | vp9_reconintra_neon.asm | 76 vld1.8 {q0}, [r2]
77 vst1.8 {q0}, [r0], r1
78 vst1.8 {q0}, [r0], r1
79 vst1.8 {q0}, [r0], r1
80 vst1.8 {q0}, [r0], r1
81 vst1.8 {q0}, [r0], r1
82 vst1.8 {q0}, [r0], r1
83 vst1.8 {q0}, [r0], r1
84 vst1.8 {q0}, [r0], r1
85 vst1.8 {q0}, [r
[all...] |
| H A D | vp9_avg_neon.asm | 36 vld1.8 {q0-q1}, [r0]!
41 vrhadd.u8 q0, q0, q8
45 vst1.8 {q0-q1}, [r2@128]!
52 vld1.8 {q0-q1}, [r0], r1
57 vrhadd.u8 q0, q0, q8
64 vst1.8 {q0-q1}, [r2@128], r3
71 vld1.8 {q0}, [r0], r1
77 vrhadd.u8 q0, q
[all...] |
| /external/libvpx/libvpx/vp9/common/arm/neon/ |
| H A D | vp9_reconintra_neon.asm | 76 vld1.8 {q0}, [r2]
77 vst1.8 {q0}, [r0], r1
78 vst1.8 {q0}, [r0], r1
79 vst1.8 {q0}, [r0], r1
80 vst1.8 {q0}, [r0], r1
81 vst1.8 {q0}, [r0], r1
82 vst1.8 {q0}, [r0], r1
83 vst1.8 {q0}, [r0], r1
84 vst1.8 {q0}, [r0], r1
85 vst1.8 {q0}, [r
[all...] |
| H A D | vp9_avg_neon.asm | 36 vld1.8 {q0-q1}, [r0]!
41 vrhadd.u8 q0, q0, q8
45 vst1.8 {q0-q1}, [r2@128]!
52 vld1.8 {q0-q1}, [r0], r1
57 vrhadd.u8 q0, q0, q8
64 vst1.8 {q0-q1}, [r2@128], r3
71 vld1.8 {q0}, [r0], r1
77 vrhadd.u8 q0, q
[all...] |
| /external/valgrind/main/none/tests/arm/ |
| H A D | neon128.c | 355 TESTINSN_imm("vmov.i32 q0", q0, 0x7);
365 TESTINSN_imm("vmov.f32 q0", q0, 0.328125);
366 TESTINSN_imm("vmov.f32 q0", q0, -0.328125);
369 TESTINSN_imm("vmvn.i32 q0", q0, 0x7);
381 TESTINSN_imm("vorr.i32 q0", q0,
[all...] |
| /external/llvm/test/MC/AArch64/ |
| H A D | neon-crypto.s | 30 sha1c q0, s1, v2.4s
31 sha1p q0, s1, v2.4s
32 sha1m q0, s1, v2.4s
34 sha256h q0, q1, v2.4s
35 sha256h2 q0, q1, v2.4s
38 // CHECK: sha1c q0, s1, v2.4s // encoding: [0x20,0x00,0x02,0x5e]
39 // CHECK: sha1p q0, s1, v2.4s // encoding: [0x20,0x10,0x02,0x5e]
40 // CHECK: sha1m q0, s1, v2.4s // encoding: [0x20,0x20,0x02,0x5e]
42 // CHECK: sha256h q0, q1, v2.4s // encoding: [0x20,0x40,0x02,0x5e]
43 // CHECK: sha256h2 q0, q
[all...] |
| H A D | elf-reloc-ldstunsimm.s | 8 str q0, [sp, #:lo12:some_label]
|
| /external/libvpx/libvpx/vp8/common/arm/neon/ |
| H A D | buildintrapredictorsmby_neon.asm | 45 vdup.u8 q0, r12
126 vdup.u8 q0, r5
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}, [r
[all...] |
| /external/openssl/crypto/modes/asm/ |
| H A D | ghashv8-armx.S | 54 vld1.64 {q0},[r0] @ load [rotated] Xi
60 vext.8 q0,q0,q0,#8
65 vrev64.8 q0,q0
74 vext.8 q10,q0,q0,#8
75 veor q3,q3,q0 @ inp^=Xi
79 .byte 0x86,0x0e,0xa8,0xf2 @ pmull q0,q1
[all...] |
| /external/chromium_org/third_party/libvpx/source/libvpx/vp9/common/mips/dspr2/ |
| H A D | vp9_loopfilter_masks_dspr2.h | 30 uint32_t q0, uint32_t q1,
65 /* mask |= (abs(q1 - q0) > limit)
66 * hev |= (abs(q1 - q0) > thresh)
68 "subu_s.qb %[c], %[q1], %[q0] \n\t"
69 "subu_s.qb %[r_k], %[q0], %[q1] \n\t"
94 [p1] "r" (p1), [p0] "r" (p0), [q1] "r" (q1), [q0] "r" (q0),
99 /* abs(p0 - q0) */
100 "subu_s.qb %[c], %[p0], %[q0] \n\t"
101 "subu_s.qb %[r_k], %[q0],
27 vp9_filter_hev_mask_dspr2(uint32_t limit, uint32_t flimit, uint32_t p1, uint32_t p0, uint32_t p3, uint32_t p2, uint32_t q0, uint32_t q1, uint32_t q2, uint32_t q3, uint32_t thresh, uint32_t *hev, uint32_t *mask) argument
132 vp9_filter_hev_mask_flatmask4_dspr2(uint32_t limit, uint32_t flimit, uint32_t thresh, uint32_t p1, uint32_t p0, uint32_t p3, uint32_t p2, uint32_t q0, uint32_t q1, uint32_t q2, uint32_t q3, uint32_t *hev, uint32_t *mask, uint32_t *flat) argument
282 vp9_flatmask5(uint32_t p4, uint32_t p3, uint32_t p2, uint32_t p1, uint32_t p0, uint32_t q0, uint32_t q1, uint32_t q2, uint32_t q3, uint32_t q4, uint32_t *flat2) argument
[all...] |
| /external/libvpx/libvpx/vp9/common/mips/dspr2/ |
| H A D | vp9_loopfilter_masks_dspr2.h | 30 uint32_t q0, uint32_t q1,
65 /* mask |= (abs(q1 - q0) > limit)
66 * hev |= (abs(q1 - q0) > thresh)
68 "subu_s.qb %[c], %[q1], %[q0] \n\t"
69 "subu_s.qb %[r_k], %[q0], %[q1] \n\t"
94 [p1] "r" (p1), [p0] "r" (p0), [q1] "r" (q1), [q0] "r" (q0),
99 /* abs(p0 - q0) */
100 "subu_s.qb %[c], %[p0], %[q0] \n\t"
101 "subu_s.qb %[r_k], %[q0],
27 vp9_filter_hev_mask_dspr2(uint32_t limit, uint32_t flimit, uint32_t p1, uint32_t p0, uint32_t p3, uint32_t p2, uint32_t q0, uint32_t q1, uint32_t q2, uint32_t q3, uint32_t thresh, uint32_t *hev, uint32_t *mask) argument
132 vp9_filter_hev_mask_flatmask4_dspr2(uint32_t limit, uint32_t flimit, uint32_t thresh, uint32_t p1, uint32_t p0, uint32_t p3, uint32_t p2, uint32_t q0, uint32_t q1, uint32_t q2, uint32_t q3, uint32_t *hev, uint32_t *mask, uint32_t *flat) argument
282 vp9_flatmask5(uint32_t p4, uint32_t p3, uint32_t p2, uint32_t p1, uint32_t p0, uint32_t q0, uint32_t q1, uint32_t q2, uint32_t q3, uint32_t q4, uint32_t *flat2) argument
[all...] |
| /external/chromium_org/third_party/libwebp/dsp/ |
| H A D | dec_neon.c | 68 uint8x16_t* const q0, uint8x16_t* const q1) {
69 // row0 = p1[0..7]|p0[0..7]|q0[0..7]|q1[0..7]
70 // row8 = p1[8..15]|p0[8..15]|q0[8..15]|q1[8..15]
75 *q0 = vcombine_u8(row0.val[2], row8.val[2]);
88 uint8x16_t* const q0, uint8x16_t* const q1) {
121 *q0 = vreinterpretq_u8_u16(row02.val[1]);
132 uint8x16_t* const q0, uint8x16_t* const q1,
135 Load4x16(src + 2, stride, q0, q1, q2, q3);
140 uint8x16_t* const q0, uint8x16_t* const q1) {
143 *q0
66 Load4x16(const uint8_t* const src, int stride, uint8x16_t* const p1, uint8x16_t* const p0, uint8x16_t* const q0, uint8x16_t* const q1) argument
86 Load4x16(const uint8_t* src, int stride, uint8x16_t* const p1, uint8x16_t* const p0, uint8x16_t* const q0, uint8x16_t* const q1) argument
129 Load8x16(const uint8_t* const src, int stride, uint8x16_t* const p3, uint8x16_t* const p2, uint8x16_t* const p1, uint8x16_t* const p0, uint8x16_t* const q0, uint8x16_t* const q1, uint8x16_t* const q2, uint8x16_t* const q3) argument
138 Load16x4(const uint8_t* const src, int stride, uint8x16_t* const p1, uint8x16_t* const p0, uint8x16_t* const q0, uint8x16_t* const q1) argument
147 Load16x8(const uint8_t* const src, int stride, uint8x16_t* const p3, uint8x16_t* const p2, uint8x16_t* const p1, uint8x16_t* const p0, uint8x16_t* const q0, uint8x16_t* const q1, uint8x16_t* const q2, uint8x16_t* const q3) argument
156 Load8x8x2(const uint8_t* const u, const uint8_t* const v, int stride, uint8x16_t* const p3, uint8x16_t* const p2, uint8x16_t* const p1, uint8x16_t* const p0, uint8x16_t* const q0, uint8x16_t* const q1, uint8x16_t* const q2, uint8x16_t* const q3) argument
180 Load8x8x2T(const uint8_t* const u, const uint8_t* const v, int stride, uint8x16_t* const p3, uint8x16_t* const p2, uint8x16_t* const p1, uint8x16_t* const p0, uint8x16_t* const q0, uint8x16_t* const q1, uint8x16_t* const q2, uint8x16_t* const q3) argument
253 Store2x16(const uint8x16_t p0, const uint8x16_t q0, uint8_t* const dst, int stride) argument
277 Store4x16(const uint8x16_t p1, const uint8x16_t p0, const uint8x16_t q0, const uint8x16_t q1, uint8_t* const dst, int stride) argument
292 Store16x2(const uint8x16_t p0, const uint8x16_t q0, uint8_t* const dst, int stride) argument
298 Store16x4(const uint8x16_t p1, const uint8x16_t p0, const uint8x16_t q0, const uint8x16_t q1, uint8_t* const dst, int stride) argument
305 Store8x2x2(const uint8x16_t p0, const uint8x16_t q0, uint8_t* const u, uint8_t* const v, int stride) argument
315 Store8x4x2(const uint8x16_t p1, const uint8x16_t p0, const uint8x16_t q0, const uint8x16_t q1, uint8_t* const u, uint8_t* const v, int stride) argument
332 Store6x8x2(const uint8x16_t p2, const uint8x16_t p1, const uint8x16_t p0, const uint8x16_t q0, const uint8x16_t q1, const uint8x16_t q2, uint8_t* u, uint8_t* v, int stride) argument
361 Store4x8x2(const uint8x16_t p1, const uint8x16_t p0, const uint8x16_t q0, const uint8x16_t q1, uint8_t* const u, uint8_t* const v, int stride) argument
440 NeedsFilter(const uint8x16_t p1, const uint8x16_t p0, const uint8x16_t q0, const uint8x16_t q1, int thresh) argument
463 GetBaseDelta(const int8x16_t p1, const int8x16_t p0, const int8x16_t q0, const int8x16_t q1) argument
473 GetBaseDelta0(const int8x16_t p0, const int8x16_t q0) argument
499 DoFilter2(const uint8x16_t p1, const uint8x16_t p0, const uint8x16_t q0, const uint8x16_t q1, const uint8x16_t mask, uint8x16_t* const op0, uint8x16_t* const oq0) argument
513 uint8x16_t p1, p0, q0, q1, op0, oq0; local
523 uint8x16_t p1, p0, q0, q1, oq0, op0; local
650 NeedsHev(const uint8x16_t p1, const uint8x16_t p0, const uint8x16_t q0, const uint8x16_t q1, int hev_thresh) argument
662 NeedsFilter2(const uint8x16_t p3, const uint8x16_t p2, const uint8x16_t p1, const uint8x16_t p0, const uint8x16_t q0, const uint8x16_t q1, const uint8x16_t q2, const uint8x16_t q3, int ithresh, int thresh) argument
687 ApplyFilter4( const int8x16_t p1, const int8x16_t p0, const int8x16_t q0, const int8x16_t q1, const int8x16_t delta0, uint8x16_t* const op1, uint8x16_t* const op0, uint8x16_t* const oq0, uint8x16_t* const oq1) argument
706 DoFilter4( const uint8x16_t p1, const uint8x16_t p0, const uint8x16_t q0, const uint8x16_t q1, const uint8x16_t mask, const uint8x16_t hev_mask, uint8x16_t* const op1, uint8x16_t* const op0, uint8x16_t* const oq0, uint8x16_t* const oq1) argument
744 ApplyFilter6( const int8x16_t p2, const int8x16_t p1, const int8x16_t p0, const int8x16_t q0, const int8x16_t q1, const int8x16_t q2, const int8x16_t delta, uint8x16_t* const op2, uint8x16_t* const op1, uint8x16_t* const op0, uint8x16_t* const oq0, uint8x16_t* const oq1, uint8x16_t* const oq2) argument
780 DoFilter6( const uint8x16_t p2, const uint8x16_t p1, const uint8x16_t p0, const uint8x16_t q0, const uint8x16_t q1, const uint8x16_t q2, const uint8x16_t mask, const uint8x16_t hev_mask, uint8x16_t* const op2, uint8x16_t* const op1, uint8x16_t* const op0, uint8x16_t* const oq0, uint8x16_t* const oq1, uint8x16_t* const oq2) argument
822 uint8x16_t p3, p2, p1, p0, q0, q1, q2, q3; local
839 uint8x16_t p3, p2, p1, p0, q0, q1, q2, q3; local
861 uint8x16_t q0, q1, q2, q3; local
885 uint8x16_t q0, q1, q2, q3; local
904 uint8x16_t p3, p2, p1, p0, q0, q1, q2, q3; local
920 uint8x16_t p3, p2, p1, p0, q0, q1, q2, q3; local
937 uint8x16_t p3, p2, p1, p0, q0, q1, q2, q3; local
952 uint8x16_t p3, p2, p1, p0, q0, q1, q2, q3; local
[all...] |
| H A D | dec_sse2.c | 325 const __m128i* const q0,
330 const __m128i t_2 = MM_ABS(*q1, *q0);
334 const __m128i t_4 = _mm_subs_epu8(t_2, h); // abs(q1 - q0) - hev_tresh
343 const __m128i* const q0,
348 const __m128i q0_p0 = _mm_subs_epi8(*q0, *p0); // q0 - p0
349 const __m128i s1 = _mm_adds_epi8(p1_q1, q0_p0); // p1 - q1 + 1 * (q0 - p0)
350 const __m128i s2 = _mm_adds_epi8(q0_p0, s1); // p1 - q1 + 2 * (q0 - p0)
351 const __m128i s3 = _mm_adds_epi8(q0_p0, s2); // p1 - q1 + 3 * (q0 - p0)
356 static WEBP_INLINE void DoSimpleFilter(__m128i* const p0, __m128i* const q0, argument
323 GetNotHEV(const __m128i* const p1, const __m128i* const p0, const __m128i* const q0, const __m128i* const q1, int hev_thresh, __m128i* const not_hev) argument
341 GetBaseDelta(const __m128i* const p1, const __m128i* const p0, const __m128i* const q0, const __m128i* const q1, __m128i* const delta) argument
386 NeedsFilter(const __m128i* const p1, const __m128i* const p0, const __m128i* const q0, const __m128i* const q1, int thresh, __m128i* const mask) argument
409 DoFilter2(__m128i* const p1, __m128i* const p0, __m128i* const q0, __m128i* const q1, int thresh) argument
428 DoFilter4(__m128i* const p1, __m128i* const p0, __m128i* const q0, __m128i* const q1, const __m128i* const mask, int hev_thresh) argument
473 DoFilter6(__m128i* const p2, __m128i* const p1, __m128i* const p0, __m128i* const q0, __m128i* const q1, __m128i* const q2, const __m128i* const mask, int hev_thresh) argument
566 Load16x4(const uint8_t* const r0, const uint8_t* const r8, int stride, __m128i* const p1, __m128i* const p0, __m128i* const q0, __m128i* const q1) argument
611 Store16x4(const __m128i* const p1, const __m128i* const p0, const __m128i* const q0, const __m128i* const q1, uint8_t* r0, uint8_t* r8, int stride) argument
659 __m128i q0 = _mm_loadu_si128((__m128i*)&p[0]); local
670 __m128i p1, p0, q0, q1; local
736 ComplexMask(const __m128i* const p1, const __m128i* const p0, const __m128i* const q0, const __m128i* const q1, int thresh, int ithresh, __m128i* const mask) argument
755 __m128i p2, p1, p0, q0, q1, q2; local
780 __m128i p3, p2, p1, p0, q0, q1, q2, q3; local
862 __m128i t1, p2, p1, p0, q0, q1, q2; local
887 __m128i p3, p2, p1, p0, q0, q1, q2, q3; local
907 __m128i t1, t2, p1, p0, q0, q1; local
933 __m128i t1, t2, p1, p0, q0, q1; local
[all...] |
| /external/webp/src/dsp/ |
| H A D | dec_neon.c | 68 uint8x16_t* const q0, uint8x16_t* const q1) {
69 // row0 = p1[0..7]|p0[0..7]|q0[0..7]|q1[0..7]
70 // row8 = p1[8..15]|p0[8..15]|q0[8..15]|q1[8..15]
75 *q0 = vcombine_u8(row0.val[2], row8.val[2]);
88 uint8x16_t* const q0, uint8x16_t* const q1) {
121 *q0 = vreinterpretq_u8_u16(row02.val[1]);
132 uint8x16_t* const q0, uint8x16_t* const q1,
135 Load4x16(src + 2, stride, q0, q1, q2, q3);
140 uint8x16_t* const q0, uint8x16_t* const q1) {
143 *q0
66 Load4x16(const uint8_t* const src, int stride, uint8x16_t* const p1, uint8x16_t* const p0, uint8x16_t* const q0, uint8x16_t* const q1) argument
86 Load4x16(const uint8_t* src, int stride, uint8x16_t* const p1, uint8x16_t* const p0, uint8x16_t* const q0, uint8x16_t* const q1) argument
129 Load8x16(const uint8_t* const src, int stride, uint8x16_t* const p3, uint8x16_t* const p2, uint8x16_t* const p1, uint8x16_t* const p0, uint8x16_t* const q0, uint8x16_t* const q1, uint8x16_t* const q2, uint8x16_t* const q3) argument
138 Load16x4(const uint8_t* const src, int stride, uint8x16_t* const p1, uint8x16_t* const p0, uint8x16_t* const q0, uint8x16_t* const q1) argument
147 Load16x8(const uint8_t* const src, int stride, uint8x16_t* const p3, uint8x16_t* const p2, uint8x16_t* const p1, uint8x16_t* const p0, uint8x16_t* const q0, uint8x16_t* const q1, uint8x16_t* const q2, uint8x16_t* const q3) argument
156 Load8x8x2(const uint8_t* const u, const uint8_t* const v, int stride, uint8x16_t* const p3, uint8x16_t* const p2, uint8x16_t* const p1, uint8x16_t* const p0, uint8x16_t* const q0, uint8x16_t* const q1, uint8x16_t* const q2, uint8x16_t* const q3) argument
180 Load8x8x2T(const uint8_t* const u, const uint8_t* const v, int stride, uint8x16_t* const p3, uint8x16_t* const p2, uint8x16_t* const p1, uint8x16_t* const p0, uint8x16_t* const q0, uint8x16_t* const q1, uint8x16_t* const q2, uint8x16_t* const q3) argument
253 Store2x16(const uint8x16_t p0, const uint8x16_t q0, uint8_t* const dst, int stride) argument
277 Store4x16(const uint8x16_t p1, const uint8x16_t p0, const uint8x16_t q0, const uint8x16_t q1, uint8_t* const dst, int stride) argument
292 Store16x2(const uint8x16_t p0, const uint8x16_t q0, uint8_t* const dst, int stride) argument
298 Store16x4(const uint8x16_t p1, const uint8x16_t p0, const uint8x16_t q0, const uint8x16_t q1, uint8_t* const dst, int stride) argument
305 Store8x2x2(const uint8x16_t p0, const uint8x16_t q0, uint8_t* const u, uint8_t* const v, int stride) argument
315 Store8x4x2(const uint8x16_t p1, const uint8x16_t p0, const uint8x16_t q0, const uint8x16_t q1, uint8_t* const u, uint8_t* const v, int stride) argument
332 Store6x8x2(const uint8x16_t p2, const uint8x16_t p1, const uint8x16_t p0, const uint8x16_t q0, const uint8x16_t q1, const uint8x16_t q2, uint8_t* u, uint8_t* v, int stride) argument
361 Store4x8x2(const uint8x16_t p1, const uint8x16_t p0, const uint8x16_t q0, const uint8x16_t q1, uint8_t* const u, uint8_t* const v, int stride) argument
440 NeedsFilter(const uint8x16_t p1, const uint8x16_t p0, const uint8x16_t q0, const uint8x16_t q1, int thresh) argument
463 GetBaseDelta(const int8x16_t p1, const int8x16_t p0, const int8x16_t q0, const int8x16_t q1) argument
473 GetBaseDelta0(const int8x16_t p0, const int8x16_t q0) argument
499 DoFilter2(const uint8x16_t p1, const uint8x16_t p0, const uint8x16_t q0, const uint8x16_t q1, const uint8x16_t mask, uint8x16_t* const op0, uint8x16_t* const oq0) argument
513 uint8x16_t p1, p0, q0, q1, op0, oq0; local
523 uint8x16_t p1, p0, q0, q1, oq0, op0; local
650 NeedsHev(const uint8x16_t p1, const uint8x16_t p0, const uint8x16_t q0, const uint8x16_t q1, int hev_thresh) argument
662 NeedsFilter2(const uint8x16_t p3, const uint8x16_t p2, const uint8x16_t p1, const uint8x16_t p0, const uint8x16_t q0, const uint8x16_t q1, const uint8x16_t q2, const uint8x16_t q3, int ithresh, int thresh) argument
687 ApplyFilter4( const int8x16_t p1, const int8x16_t p0, const int8x16_t q0, const int8x16_t q1, const int8x16_t delta0, uint8x16_t* const op1, uint8x16_t* const op0, uint8x16_t* const oq0, uint8x16_t* const oq1) argument
706 DoFilter4( const uint8x16_t p1, const uint8x16_t p0, const uint8x16_t q0, const uint8x16_t q1, const uint8x16_t mask, const uint8x16_t hev_mask, uint8x16_t* const op1, uint8x16_t* const op0, uint8x16_t* const oq0, uint8x16_t* const oq1) argument
744 ApplyFilter6( const int8x16_t p2, const int8x16_t p1, const int8x16_t p0, const int8x16_t q0, const int8x16_t q1, const int8x16_t q2, const int8x16_t delta, uint8x16_t* const op2, uint8x16_t* const op1, uint8x16_t* const op0, uint8x16_t* const oq0, uint8x16_t* const oq1, uint8x16_t* const oq2) argument
780 DoFilter6( const uint8x16_t p2, const uint8x16_t p1, const uint8x16_t p0, const uint8x16_t q0, const uint8x16_t q1, const uint8x16_t q2, const uint8x16_t mask, const uint8x16_t hev_mask, uint8x16_t* const op2, uint8x16_t* const op1, uint8x16_t* const op0, uint8x16_t* const oq0, uint8x16_t* const oq1, uint8x16_t* const oq2) argument
822 uint8x16_t p3, p2, p1, p0, q0, q1, q2, q3; local
839 uint8x16_t p3, p2, p1, p0, q0, q1, q2, q3; local
861 uint8x16_t q0, q1, q2, q3; local
885 uint8x16_t q0, q1, q2, q3; local
904 uint8x16_t p3, p2, p1, p0, q0, q1, q2, q3; local
920 uint8x16_t p3, p2, p1, p0, q0, q1, q2, q3; local
937 uint8x16_t p3, p2, p1, p0, q0, q1, q2, q3; local
952 uint8x16_t p3, p2, p1, p0, q0, q1, q2, q3; local
[all...] |
| H A D | dec_sse2.c | 325 const __m128i* const q0,
330 const __m128i t_2 = MM_ABS(*q1, *q0);
334 const __m128i t_4 = _mm_subs_epu8(t_2, h); // abs(q1 - q0) - hev_tresh
343 const __m128i* const q0,
348 const __m128i q0_p0 = _mm_subs_epi8(*q0, *p0); // q0 - p0
349 const __m128i s1 = _mm_adds_epi8(p1_q1, q0_p0); // p1 - q1 + 1 * (q0 - p0)
350 const __m128i s2 = _mm_adds_epi8(q0_p0, s1); // p1 - q1 + 2 * (q0 - p0)
351 const __m128i s3 = _mm_adds_epi8(q0_p0, s2); // p1 - q1 + 3 * (q0 - p0)
356 static WEBP_INLINE void DoSimpleFilter(__m128i* const p0, __m128i* const q0, argument
323 GetNotHEV(const __m128i* const p1, const __m128i* const p0, const __m128i* const q0, const __m128i* const q1, int hev_thresh, __m128i* const not_hev) argument
341 GetBaseDelta(const __m128i* const p1, const __m128i* const p0, const __m128i* const q0, const __m128i* const q1, __m128i* const delta) argument
386 NeedsFilter(const __m128i* const p1, const __m128i* const p0, const __m128i* const q0, const __m128i* const q1, int thresh, __m128i* const mask) argument
409 DoFilter2(__m128i* const p1, __m128i* const p0, __m128i* const q0, __m128i* const q1, int thresh) argument
428 DoFilter4(__m128i* const p1, __m128i* const p0, __m128i* const q0, __m128i* const q1, const __m128i* const mask, int hev_thresh) argument
473 DoFilter6(__m128i* const p2, __m128i* const p1, __m128i* const p0, __m128i* const q0, __m128i* const q1, __m128i* const q2, const __m128i* const mask, int hev_thresh) argument
566 Load16x4(const uint8_t* const r0, const uint8_t* const r8, int stride, __m128i* const p1, __m128i* const p0, __m128i* const q0, __m128i* const q1) argument
611 Store16x4(const __m128i* const p1, const __m128i* const p0, const __m128i* const q0, const __m128i* const q1, uint8_t* r0, uint8_t* r8, int stride) argument
659 __m128i q0 = _mm_loadu_si128((__m128i*)&p[0]); local
670 __m128i p1, p0, q0, q1; local
736 ComplexMask(const __m128i* const p1, const __m128i* const p0, const __m128i* const q0, const __m128i* const q1, int thresh, int ithresh, __m128i* const mask) argument
755 __m128i p2, p1, p0, q0, q1, q2; local
780 __m128i p3, p2, p1, p0, q0, q1, q2, q3; local
862 __m128i t1, p2, p1, p0, q0, q1, q2; local
887 __m128i p3, p2, p1, p0, q0, q1, q2, q3; local
907 __m128i t1, t2, p1, p0, q0, q1; local
933 __m128i t1, t2, p1, p0, q0, q1; local
[all...] |
| /external/fdlibm/ |
| H A D | k_rem_pio2.c | 104 * q0 the corresponding exponent of q[0]. Note that the
105 * exponent for q[i] would be q0-24*i.
170 int jz,jx,jv,jp,jk,carry,n,iq[20],i,j,k,m,q0,ih; local
177 /* determine jx,jv,q0, note that 3>q0 */
180 q0 = e0-24*(jv+1);
201 z = ieee_scalbn(z,q0); /* actual value of z */
206 if(q0>0) { /* need iq[jz-1] to determine n */
207 i = (iq[jz-1]>>(24-q0)); n += i;
208 iq[jz-1] -= i<<(24-q0);
[all...] |