ETISS 0.8.0
Extendable Translating Instruction Set Simulator (version 0.8.0)
•All Classes Namespaces Files Functions Variables Typedefs Enumerations Enumerator Friends Macros Modules Pages
tmmintrin.h
Go to the documentation of this file.
1/*===---- tmmintrin.h - Implementation of SSSE3 intrinsics on PowerPC ------===
2 *
3 * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
4 * See https://llvm.org/LICENSE.txt for license information.
5 * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
6 *
7 *===-----------------------------------------------------------------------===
8 */
9
10/* Implemented from the specification included in the Intel C++ Compiler
11 User Guide and Reference, version 9.0. */
12
13#ifndef NO_WARN_X86_INTRINSICS
14/* This header is distributed to simplify porting x86_64 code that
15 makes explicit use of Intel intrinsics to powerpc64le.
16
17 It is the user's responsibility to determine if the results are
18 acceptable and make additional changes as necessary.
19
20 Note that much code that uses Intel intrinsics can be rewritten in
21 standard C or GNU C extensions, which are more portable and better
22 optimized across multiple targets. */
23#endif
24
25#ifndef TMMINTRIN_H_
26#define TMMINTRIN_H_
27
28#if defined(__linux__) && defined(__ppc64__)
29
30#include <altivec.h>
31
32/* We need definitions from the SSE header files. */
33#include <pmmintrin.h>
34
35extern __inline __m128i
36__attribute__((__gnu_inline__, __always_inline__, __artificial__))
37_mm_abs_epi16 (__m128i __A)
38{
39 return (__m128i) vec_abs ((__v8hi) __A);
40}
41
42extern __inline __m128i
43__attribute__((__gnu_inline__, __always_inline__, __artificial__))
44_mm_abs_epi32 (__m128i __A)
45{
46 return (__m128i) vec_abs ((__v4si) __A);
47}
48
49extern __inline __m128i
50__attribute__((__gnu_inline__, __always_inline__, __artificial__))
51_mm_abs_epi8 (__m128i __A)
52{
53 return (__m128i) vec_abs ((__v16qi) __A);
54}
55
56extern __inline __m64
57__attribute__((__gnu_inline__, __always_inline__, __artificial__))
58_mm_abs_pi16 (__m64 __A)
59{
60 __v8hi __B = (__v8hi) (__v2du) { __A, __A };
61 return (__m64) ((__v2du) vec_abs (__B))[0];
62}
63
64extern __inline __m64
65__attribute__((__gnu_inline__, __always_inline__, __artificial__))
66_mm_abs_pi32 (__m64 __A)
67{
68 __v4si __B = (__v4si) (__v2du) { __A, __A };
69 return (__m64) ((__v2du) vec_abs (__B))[0];
70}
71
72extern __inline __m64
73__attribute__((__gnu_inline__, __always_inline__, __artificial__))
74_mm_abs_pi8 (__m64 __A)
75{
76 __v16qi __B = (__v16qi) (__v2du) { __A, __A };
77 return (__m64) ((__v2du) vec_abs (__B))[0];
78}
79
80extern __inline __m128i
81__attribute__((__gnu_inline__, __always_inline__, __artificial__))
82_mm_alignr_epi8 (__m128i __A, __m128i __B, const unsigned int __count)
83{
84 if (__builtin_constant_p (__count) && __count < 16)
85 {
86#ifdef __LITTLE_ENDIAN__
87 __A = (__m128i) vec_reve ((__v16qu) __A);
88 __B = (__m128i) vec_reve ((__v16qu) __B);
89#endif
90 __A = (__m128i) vec_sld ((__v16qu) __B, (__v16qu) __A, __count);
91#ifdef __LITTLE_ENDIAN__
92 __A = (__m128i) vec_reve ((__v16qu) __A);
93#endif
94 return __A;
95 }
96
97 if (__count == 0)
98 return __B;
99
100 if (__count >= 16)
101 {
102 if (__count >= 32)
103 {
104 const __v16qu zero = { 0 };
105 return (__m128i) zero;
106 }
107 else
108 {
109 const __v16qu __shift =
110 vec_splats ((unsigned char) ((__count - 16) * 8));
111#ifdef __LITTLE_ENDIAN__
112 return (__m128i) vec_sro ((__v16qu) __A, __shift);
113#else
114 return (__m128i) vec_slo ((__v16qu) __A, __shift);
115#endif
116 }
117 }
118 else
119 {
120 const __v16qu __shiftA =
121 vec_splats ((unsigned char) ((16 - __count) * 8));
122 const __v16qu __shiftB = vec_splats ((unsigned char) (__count * 8));
123#ifdef __LITTLE_ENDIAN__
124 __A = (__m128i) vec_slo ((__v16qu) __A, __shiftA);
125 __B = (__m128i) vec_sro ((__v16qu) __B, __shiftB);
126#else
127 __A = (__m128i) vec_sro ((__v16qu) __A, __shiftA);
128 __B = (__m128i) vec_slo ((__v16qu) __B, __shiftB);
129#endif
130 return (__m128i) vec_or ((__v16qu) __A, (__v16qu) __B);
131 }
132}
133
134extern __inline __m64
135__attribute__((__gnu_inline__, __always_inline__, __artificial__))
136_mm_alignr_pi8 (__m64 __A, __m64 __B, unsigned int __count)
137{
138 if (__count < 16)
139 {
140 __v2du __C = { __B, __A };
141#ifdef __LITTLE_ENDIAN__
142 const __v4su __shift = { __count << 3, 0, 0, 0 };
143 __C = (__v2du) vec_sro ((__v16qu) __C, (__v16qu) __shift);
144#else
145 const __v4su __shift = { 0, 0, 0, __count << 3 };
146 __C = (__v2du) vec_slo ((__v16qu) __C, (__v16qu) __shift);
147#endif
148 return (__m64) __C[0];
149 }
150 else
151 {
152 const __m64 __zero = { 0 };
153 return __zero;
154 }
155}
156
157extern __inline __m128i
158__attribute__((__gnu_inline__, __always_inline__, __artificial__))
159_mm_hadd_epi16 (__m128i __A, __m128i __B)
160{
161 const __v16qu __P =
162 { 0, 1, 4, 5, 8, 9, 12, 13, 16, 17, 20, 21, 24, 25, 28, 29 };
163 const __v16qu __Q =
164 { 2, 3, 6, 7, 10, 11, 14, 15, 18, 19, 22, 23, 26, 27, 30, 31 };
165 __v8hi __C = vec_perm ((__v8hi) __A, (__v8hi) __B, __P);
166 __v8hi __D = vec_perm ((__v8hi) __A, (__v8hi) __B, __Q);
167 return (__m128i) vec_add (__C, __D);
168}
169
170extern __inline __m128i
171__attribute__((__gnu_inline__, __always_inline__, __artificial__))
172_mm_hadd_epi32 (__m128i __A, __m128i __B)
173{
174 const __v16qu __P =
175 { 0, 1, 2, 3, 8, 9, 10, 11, 16, 17, 18, 19, 24, 25, 26, 27 };
176 const __v16qu __Q =
177 { 4, 5, 6, 7, 12, 13, 14, 15, 20, 21, 22, 23, 28, 29, 30, 31 };
178 __v4si __C = vec_perm ((__v4si) __A, (__v4si) __B, __P);
179 __v4si __D = vec_perm ((__v4si) __A, (__v4si) __B, __Q);
180 return (__m128i) vec_add (__C, __D);
181}
182
183extern __inline __m64
184__attribute__((__gnu_inline__, __always_inline__, __artificial__))
185_mm_hadd_pi16 (__m64 __A, __m64 __B)
186{
187 __v8hi __C = (__v8hi) (__v2du) { __A, __B };
188 const __v16qu __P =
189 { 0, 1, 4, 5, 8, 9, 12, 13, 0, 1, 4, 5, 8, 9, 12, 13 };
190 const __v16qu __Q =
191 { 2, 3, 6, 7, 10, 11, 14, 15, 2, 3, 6, 7, 10, 11, 14, 15 };
192 __v8hi __D = vec_perm (__C, __C, __Q);
193 __C = vec_perm (__C, __C, __P);
194 __C = vec_add (__C, __D);
195 return (__m64) ((__v2du) __C)[1];
196}
197
198extern __inline __m64
199__attribute__((__gnu_inline__, __always_inline__, __artificial__))
200_mm_hadd_pi32 (__m64 __A, __m64 __B)
201{
202 __v4si __C = (__v4si) (__v2du) { __A, __B };
203 const __v16qu __P =
204 { 0, 1, 2, 3, 8, 9, 10, 11, 0, 1, 2, 3, 8, 9, 10, 11 };
205 const __v16qu __Q =
206 { 4, 5, 6, 7, 12, 13, 14, 15, 4, 5, 6, 7, 12, 13, 14, 15 };
207 __v4si __D = vec_perm (__C, __C, __Q);
208 __C = vec_perm (__C, __C, __P);
209 __C = vec_add (__C, __D);
210 return (__m64) ((__v2du) __C)[1];
211}
212
213extern __inline __m128i
214__attribute__((__gnu_inline__, __always_inline__, __artificial__))
215_mm_hadds_epi16 (__m128i __A, __m128i __B)
216{
217 __v4si __C = { 0 }, __D = { 0 };
218 __C = vec_sum4s ((__v8hi) __A, __C);
219 __D = vec_sum4s ((__v8hi) __B, __D);
220 __C = (__v4si) vec_packs (__C, __D);
221 return (__m128i) __C;
222}
223
224extern __inline __m64
225__attribute__((__gnu_inline__, __always_inline__, __artificial__))
226_mm_hadds_pi16 (__m64 __A, __m64 __B)
227{
228 const __v4si __zero = { 0 };
229 __v8hi __C = (__v8hi) (__v2du) { __A, __B };
230 __v4si __D = vec_sum4s (__C, __zero);
231 __C = vec_packs (__D, __D);
232 return (__m64) ((__v2du) __C)[1];
233}
234
235extern __inline __m128i
236__attribute__((__gnu_inline__, __always_inline__, __artificial__))
237_mm_hsub_epi16 (__m128i __A, __m128i __B)
238{
239 const __v16qu __P =
240 { 0, 1, 4, 5, 8, 9, 12, 13, 16, 17, 20, 21, 24, 25, 28, 29 };
241 const __v16qu __Q =
242 { 2, 3, 6, 7, 10, 11, 14, 15, 18, 19, 22, 23, 26, 27, 30, 31 };
243 __v8hi __C = vec_perm ((__v8hi) __A, (__v8hi) __B, __P);
244 __v8hi __D = vec_perm ((__v8hi) __A, (__v8hi) __B, __Q);
245 return (__m128i) vec_sub (__C, __D);
246}
247
248extern __inline __m128i
249__attribute__((__gnu_inline__, __always_inline__, __artificial__))
250_mm_hsub_epi32 (__m128i __A, __m128i __B)
251{
252 const __v16qu __P =
253 { 0, 1, 2, 3, 8, 9, 10, 11, 16, 17, 18, 19, 24, 25, 26, 27 };
254 const __v16qu __Q =
255 { 4, 5, 6, 7, 12, 13, 14, 15, 20, 21, 22, 23, 28, 29, 30, 31 };
256 __v4si __C = vec_perm ((__v4si) __A, (__v4si) __B, __P);
257 __v4si __D = vec_perm ((__v4si) __A, (__v4si) __B, __Q);
258 return (__m128i) vec_sub (__C, __D);
259}
260
261extern __inline __m64
262__attribute__((__gnu_inline__, __always_inline__, __artificial__))
263_mm_hsub_pi16 (__m64 __A, __m64 __B)
264{
265 const __v16qu __P =
266 { 0, 1, 4, 5, 8, 9, 12, 13, 0, 1, 4, 5, 8, 9, 12, 13 };
267 const __v16qu __Q =
268 { 2, 3, 6, 7, 10, 11, 14, 15, 2, 3, 6, 7, 10, 11, 14, 15 };
269 __v8hi __C = (__v8hi) (__v2du) { __A, __B };
270 __v8hi __D = vec_perm (__C, __C, __Q);
271 __C = vec_perm (__C, __C, __P);
272 __C = vec_sub (__C, __D);
273 return (__m64) ((__v2du) __C)[1];
274}
275
276extern __inline __m64
277__attribute__((__gnu_inline__, __always_inline__, __artificial__))
278_mm_hsub_pi32 (__m64 __A, __m64 __B)
279{
280 const __v16qu __P =
281 { 0, 1, 2, 3, 8, 9, 10, 11, 0, 1, 2, 3, 8, 9, 10, 11 };
282 const __v16qu __Q =
283 { 4, 5, 6, 7, 12, 13, 14, 15, 4, 5, 6, 7, 12, 13, 14, 15 };
284 __v4si __C = (__v4si) (__v2du) { __A, __B };
285 __v4si __D = vec_perm (__C, __C, __Q);
286 __C = vec_perm (__C, __C, __P);
287 __C = vec_sub (__C, __D);
288 return (__m64) ((__v2du) __C)[1];
289}
290
291extern __inline __m128i
292__attribute__((__gnu_inline__, __always_inline__, __artificial__))
293_mm_hsubs_epi16 (__m128i __A, __m128i __B)
294{
295 const __v16qu __P =
296 { 0, 1, 4, 5, 8, 9, 12, 13, 16, 17, 20, 21, 24, 25, 28, 29 };
297 const __v16qu __Q =
298 { 2, 3, 6, 7, 10, 11, 14, 15, 18, 19, 22, 23, 26, 27, 30, 31 };
299 __v8hi __C = vec_perm ((__v8hi) __A, (__v8hi) __B, __P);
300 __v8hi __D = vec_perm ((__v8hi) __A, (__v8hi) __B, __Q);
301 return (__m128i) vec_subs (__C, __D);
302}
303
304extern __inline __m64
305__attribute__((__gnu_inline__, __always_inline__, __artificial__))
306_mm_hsubs_pi16 (__m64 __A, __m64 __B)
307{
308 const __v16qu __P =
309 { 0, 1, 4, 5, 8, 9, 12, 13, 0, 1, 4, 5, 8, 9, 12, 13 };
310 const __v16qu __Q =
311 { 2, 3, 6, 7, 10, 11, 14, 15, 2, 3, 6, 7, 10, 11, 14, 15 };
312 __v8hi __C = (__v8hi) (__v2du) { __A, __B };
313 __v8hi __D = vec_perm (__C, __C, __P);
314 __v8hi __E = vec_perm (__C, __C, __Q);
315 __C = vec_subs (__D, __E);
316 return (__m64) ((__v2du) __C)[1];
317}
318
319extern __inline __m128i
320__attribute__((__gnu_inline__, __always_inline__, __artificial__))
321_mm_shuffle_epi8 (__m128i __A, __m128i __B)
322{
323 const __v16qi __zero = { 0 };
324 __vector __bool char __select = vec_cmplt ((__v16qi) __B, __zero);
325 __v16qi __C = vec_perm ((__v16qi) __A, (__v16qi) __A, (__v16qu) __B);
326 return (__m128i) vec_sel (__C, __zero, __select);
327}
328
329extern __inline __m64
330__attribute__((__gnu_inline__, __always_inline__, __artificial__))
331_mm_shuffle_pi8 (__m64 __A, __m64 __B)
332{
333 const __v16qi __zero = { 0 };
334 __v16qi __C = (__v16qi) (__v2du) { __A, __A };
335 __v16qi __D = (__v16qi) (__v2du) { __B, __B };
336 __vector __bool char __select = vec_cmplt ((__v16qi) __D, __zero);
337 __C = vec_perm ((__v16qi) __C, (__v16qi) __C, (__v16qu) __D);
338 __C = vec_sel (__C, __zero, __select);
339 return (__m64) ((__v2du) (__C))[0];
340}
341
342extern __inline __m128i
343__attribute__((__gnu_inline__, __always_inline__, __artificial__))
344_mm_sign_epi8 (__m128i __A, __m128i __B)
345{
346 const __v16qi __zero = { 0 };
347 __v16qi __selectneg = (__v16qi) vec_cmplt ((__v16qi) __B, __zero);
348 __v16qi __selectpos =
349 (__v16qi) vec_neg ((__v16qi) vec_cmpgt ((__v16qi) __B, __zero));
350 __v16qi __conv = vec_add (__selectneg, __selectpos);
351 return (__m128i) vec_mul ((__v16qi) __A, (__v16qi) __conv);
352}
353
354extern __inline __m128i
355__attribute__((__gnu_inline__, __always_inline__, __artificial__))
356_mm_sign_epi16 (__m128i __A, __m128i __B)
357{
358 const __v8hi __zero = { 0 };
359 __v8hi __selectneg = (__v8hi) vec_cmplt ((__v8hi) __B, __zero);
360 __v8hi __selectpos =
361 (__v8hi) vec_neg ((__v8hi) vec_cmpgt ((__v8hi) __B, __zero));
362 __v8hi __conv = vec_add (__selectneg, __selectpos);
363 return (__m128i) vec_mul ((__v8hi) __A, (__v8hi) __conv);
364}
365
366extern __inline __m128i
367__attribute__((__gnu_inline__, __always_inline__, __artificial__))
368_mm_sign_epi32 (__m128i __A, __m128i __B)
369{
370 const __v4si __zero = { 0 };
371 __v4si __selectneg = (__v4si) vec_cmplt ((__v4si) __B, __zero);
372 __v4si __selectpos =
373 (__v4si) vec_neg ((__v4si) vec_cmpgt ((__v4si) __B, __zero));
374 __v4si __conv = vec_add (__selectneg, __selectpos);
375 return (__m128i) vec_mul ((__v4si) __A, (__v4si) __conv);
376}
377
378extern __inline __m64
379__attribute__((__gnu_inline__, __always_inline__, __artificial__))
380_mm_sign_pi8 (__m64 __A, __m64 __B)
381{
382 const __v16qi __zero = { 0 };
383 __v16qi __C = (__v16qi) (__v2du) { __A, __A };
384 __v16qi __D = (__v16qi) (__v2du) { __B, __B };
385 __C = (__v16qi) _mm_sign_epi8 ((__m128i) __C, (__m128i) __D);
386 return (__m64) ((__v2du) (__C))[0];
387}
388
389extern __inline __m64
390__attribute__((__gnu_inline__, __always_inline__, __artificial__))
391_mm_sign_pi16 (__m64 __A, __m64 __B)
392{
393 const __v8hi __zero = { 0 };
394 __v8hi __C = (__v8hi) (__v2du) { __A, __A };
395 __v8hi __D = (__v8hi) (__v2du) { __B, __B };
396 __C = (__v8hi) _mm_sign_epi16 ((__m128i) __C, (__m128i) __D);
397 return (__m64) ((__v2du) (__C))[0];
398}
399
400extern __inline __m64
401__attribute__((__gnu_inline__, __always_inline__, __artificial__))
402_mm_sign_pi32 (__m64 __A, __m64 __B)
403{
404 const __v4si __zero = { 0 };
405 __v4si __C = (__v4si) (__v2du) { __A, __A };
406 __v4si __D = (__v4si) (__v2du) { __B, __B };
407 __C = (__v4si) _mm_sign_epi32 ((__m128i) __C, (__m128i) __D);
408 return (__m64) ((__v2du) (__C))[0];
409}
410
411extern __inline __m128i
412__attribute__((__gnu_inline__, __always_inline__, __artificial__))
413_mm_maddubs_epi16 (__m128i __A, __m128i __B)
414{
415 __v8hi __unsigned = vec_splats ((signed short) 0x00ff);
416 __v8hi __C = vec_and (vec_unpackh ((__v16qi) __A), __unsigned);
417 __v8hi __D = vec_and (vec_unpackl ((__v16qi) __A), __unsigned);
418 __v8hi __E = vec_unpackh ((__v16qi) __B);
419 __v8hi __F = vec_unpackl ((__v16qi) __B);
420 __C = vec_mul (__C, __E);
421 __D = vec_mul (__D, __F);
422 const __v16qu __odds =
423 { 0, 1, 4, 5, 8, 9, 12, 13, 16, 17, 20, 21, 24, 25, 28, 29 };
424 const __v16qu __evens =
425 { 2, 3, 6, 7, 10, 11, 14, 15, 18, 19, 22, 23, 26, 27, 30, 31 };
426 __E = vec_perm (__C, __D, __odds);
427 __F = vec_perm (__C, __D, __evens);
428 return (__m128i) vec_adds (__E, __F);
429}
430
431extern __inline __m64
432__attribute__((__gnu_inline__, __always_inline__, __artificial__))
433_mm_maddubs_pi16 (__m64 __A, __m64 __B)
434{
435 __v8hi __C = (__v8hi) (__v2du) { __A, __A };
436 __C = vec_unpackl ((__v16qi) __C);
437 const __v8hi __unsigned = vec_splats ((signed short) 0x00ff);
438 __C = vec_and (__C, __unsigned);
439 __v8hi __D = (__v8hi) (__v2du) { __B, __B };
440 __D = vec_unpackl ((__v16qi) __D);
441 __D = vec_mul (__C, __D);
442 const __v16qu __odds =
443 { 0, 1, 4, 5, 8, 9, 12, 13, 16, 17, 20, 21, 24, 25, 28, 29 };
444 const __v16qu __evens =
445 { 2, 3, 6, 7, 10, 11, 14, 15, 18, 19, 22, 23, 26, 27, 30, 31 };
446 __C = vec_perm (__D, __D, __odds);
447 __D = vec_perm (__D, __D, __evens);
448 __C = vec_adds (__C, __D);
449 return (__m64) ((__v2du) (__C))[0];
450}
451
452extern __inline __m128i
453__attribute__((__gnu_inline__, __always_inline__, __artificial__))
454_mm_mulhrs_epi16 (__m128i __A, __m128i __B)
455{
456 __v4si __C = vec_unpackh ((__v8hi) __A);
457 __v4si __D = vec_unpackh ((__v8hi) __B);
458 __C = vec_mul (__C, __D);
459 __D = vec_unpackl ((__v8hi) __A);
460 __v4si __E = vec_unpackl ((__v8hi) __B);
461 __D = vec_mul (__D, __E);
462 const __v4su __shift = vec_splats ((unsigned int) 14);
463 __C = vec_sr (__C, __shift);
464 __D = vec_sr (__D, __shift);
465 const __v4si __ones = vec_splats ((signed int) 1);
466 __C = vec_add (__C, __ones);
467 __C = vec_sr (__C, (__v4su) __ones);
468 __D = vec_add (__D, __ones);
469 __D = vec_sr (__D, (__v4su) __ones);
470 return (__m128i) vec_pack (__C, __D);
471}
472
473extern __inline __m64
474__attribute__((__gnu_inline__, __always_inline__, __artificial__))
475_mm_mulhrs_pi16 (__m64 __A, __m64 __B)
476{
477 __v4si __C = (__v4si) (__v2du) { __A, __A };
478 __C = vec_unpackh ((__v8hi) __C);
479 __v4si __D = (__v4si) (__v2du) { __B, __B };
480 __D = vec_unpackh ((__v8hi) __D);
481 __C = vec_mul (__C, __D);
482 const __v4su __shift = vec_splats ((unsigned int) 14);
483 __C = vec_sr (__C, __shift);
484 const __v4si __ones = vec_splats ((signed int) 1);
485 __C = vec_add (__C, __ones);
486 __C = vec_sr (__C, (__v4su) __ones);
487 __v8hi __E = vec_pack (__C, __D);
488 return (__m64) ((__v2du) (__E))[0];
489}
490
491#else
492#include_next <tmmintrin.h>
493#endif /* defined(__linux__) && defined(__ppc64__) */
494
495#endif /* TMMINTRIN_H_ */
__attribute__
_Float16 __2f16 __attribute__((ext_vector_type(2)))
Zeroes the upper 128 bits (bits 255:128) of all YMM registers.
Definition __clang_hip_libdevice_declares.h:294
vec_sr
static __inline__ vector unsigned char __ATTRS_o_ai vec_sr(vector unsigned char __a, vector unsigned char __b)
Definition altivec.h:9543
vec_sro
static __inline__ vector signed char __ATTRS_o_ai vec_sro(vector signed char __a, vector signed char __b)
Definition altivec.h:10073
vec_reve
static __ATTRS_o_ai vector bool char vec_reve(vector bool char __a)
Definition altivec.h:16218
vec_splats
static __inline__ vector signed char __ATTRS_o_ai vec_splats(signed char __a)
Definition altivec.h:13710
vec_sld
static __inline__ vector signed int __ATTRS_o_ai vec_sld(vector signed int, vector signed int, unsigned const int __c)
Definition altivec.h:8309
vec_unpackl
static __inline__ vector short __ATTRS_o_ai vec_unpackl(vector signed char __a)
Definition altivec.h:11807
vec_sum4s
static __inline__ vector int __ATTRS_o_ai vec_sum4s(vector signed char __a, vector int __b)
Definition altivec.h:11531
vec_mul
static __inline__ vector signed char __ATTRS_o_ai vec_mul(vector signed char __a, vector signed char __b)
Definition altivec.h:5638
vec_and
static __inline__ vector signed char __ATTRS_o_ai vec_and(vector signed char __a, vector signed char __b)
Definition altivec.h:810
vec_subs
static __inline__ vector signed char __ATTRS_o_ai vec_subs(vector signed char __a, vector signed char __b)
Definition altivec.h:11232
vec_adds
static __inline__ vector signed char __ATTRS_o_ai vec_adds(vector signed char __a, vector signed char __b)
Definition altivec.h:560
vec_perm
static __inline__ vector signed char __ATTRS_o_ai vec_perm(vector signed char __a, vector signed char __b, vector unsigned char __c)
Definition altivec.h:7320
vec_sel
static __inline__ vector signed char __ATTRS_o_ai vec_sel(vector signed char __a, vector signed char __b, vector unsigned char __c)
Definition altivec.h:7834
vec_cmplt
static __inline__ vector bool char __ATTRS_o_ai vec_cmplt(vector signed char __a, vector signed char __b)
Definition altivec.h:2196
vec_slo
static __inline__ vector signed char __ATTRS_o_ai vec_slo(vector signed char __a, vector signed char __b)
Definition altivec.h:9034
vec_pack
static __inline__ vector signed char __ATTRS_o_ai vec_pack(vector signed short __a, vector signed short __b)
Definition altivec.h:6747
vec_neg
static vector float __ATTRS_o_ai vec_neg(vector float __a)
Definition altivec.h:16763
vec_or
static __inline__ vector signed char __ATTRS_o_ai vec_or(vector signed char __a, vector signed char __b)
Definition altivec.h:6234
vec_unpackh
static __inline__ vector short __ATTRS_o_ai vec_unpackh(vector signed char __a)
Definition altivec.h:11668
vec_add
static __inline__ vector signed char __ATTRS_o_ai vec_add(vector signed char __a, vector signed char __b)
Definition altivec.h:198
vec_abs
static __inline__ vector signed char __ATTRS_o_ai vec_abs(vector signed char __a)
Definition altivec.h:115
vec_cmpgt
static __inline__ vector bool char __ATTRS_o_ai vec_cmpgt(vector signed char __a, vector signed char __b)
Definition altivec.h:1964
vec_packs
static __inline__ vector signed char __ATTRS_o_ai vec_packs(vector short __a, vector short __b)
Definition altivec.h:7073
vec_sub
static __inline__ vector signed char __ATTRS_o_ai vec_sub(vector signed char __a, vector signed char __b)
Definition altivec.h:10963
__D
static __inline__ unsigned int unsigned char __D
Definition ia32intrin.h:283
__C
static __inline__ unsigned char int __C
Definition ia32intrin.h:373
__conv
#define __conv
Definition opencl-c.h:32
_mm_sign_epi16
static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_sign_epi16(__m128i __a, __m128i __b)
For each 16-bit integer in the first source operand, perform one of the following actions as specifie...
Definition tmmintrin.h:659
_mm_hadd_epi32
static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_hadd_epi32(__m128i __a, __m128i __b)
Horizontally adds the adjacent pairs of values contained in 2 packed 128-bit vectors of [4 x i32].
Definition tmmintrin.h:214
_mm_alignr_epi8
#define _mm_alignr_epi8(a, b, n)
Concatenates the two 128-bit integer vector operands, and right-shifts the result by the number of by...
Definition tmmintrin.h:147
_mm_sign_pi16
static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX _mm_sign_pi16(__m64 __a, __m64 __b)
For each 16-bit integer in the first source operand, perform one of the following actions as specifie...
Definition tmmintrin.h:737
_mm_shuffle_pi8
static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX _mm_shuffle_pi8(__m64 __a, __m64 __b)
Copies the 8-bit integers from a 64-bit integer vector to the destination or clears 8-bit values in t...
Definition tmmintrin.h:607
_mm_alignr_pi8
#define _mm_alignr_pi8(a, b, n)
Concatenates the two 64-bit integer vector operands, and right-shifts the result by the number of byt...
Definition tmmintrin.h:170
_mm_hsub_pi16
static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX _mm_hsub_pi16(__m64 __a, __m64 __b)
Horizontally subtracts the adjacent pairs of values contained in 2 packed 64-bit vectors of [4 x i16]...
Definition tmmintrin.h:379
_mm_hsub_epi32
static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_hsub_epi32(__m128i __a, __m128i __b)
Horizontally subtracts the adjacent pairs of values contained in 2 packed 128-bit vectors of [4 x i32...
Definition tmmintrin.h:356
_mm_abs_epi8
static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_abs_epi8(__m128i __a)
Computes the absolute value of each of the packed 8-bit signed integers in the source operand and sto...
Definition tmmintrin.h:50
_mm_mulhrs_epi16
static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_mulhrs_epi16(__m128i __a, __m128i __b)
Multiplies packed 16-bit signed integer values, truncates the 32-bit products to the 18 most signific...
Definition tmmintrin.h:536
_mm_maddubs_pi16
static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX _mm_maddubs_pi16(__m64 __a, __m64 __b)
Multiplies corresponding pairs of packed 8-bit unsigned integer values contained in the first source ...
Definition tmmintrin.h:516
_mm_hadd_pi32
static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX _mm_hadd_pi32(__m64 __a, __m64 __b)
Horizontally adds the adjacent pairs of values contained in 2 packed 64-bit vectors of [2 x i32].
Definition tmmintrin.h:260
_mm_abs_epi16
static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_abs_epi16(__m128i __a)
Computes the absolute value of each of the packed 16-bit signed integers in the source operand and st...
Definition tmmintrin.h:86
_mm_maddubs_epi16
static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_maddubs_epi16(__m128i __a, __m128i __b)
Multiplies corresponding pairs of packed 8-bit unsigned integer values contained in the first source ...
Definition tmmintrin.h:486
_mm_abs_pi8
static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX _mm_abs_pi8(__m64 __a)
Computes the absolute value of each of the packed 8-bit signed integers in the source operand and sto...
Definition tmmintrin.h:32
_mm_sign_epi8
static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_sign_epi8(__m128i __a, __m128i __b)
For each 8-bit integer in the first source operand, perform one of the following actions as specified...
Definition tmmintrin.h:633
_mm_hsub_epi16
static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_hsub_epi16(__m128i __a, __m128i __b)
Horizontally subtracts the adjacent pairs of values contained in 2 packed 128-bit vectors of [8 x i16...
Definition tmmintrin.h:333
_mm_hsubs_epi16
static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_hsubs_epi16(__m128i __a, __m128i __b)
Horizontally subtracts the adjacent pairs of values contained in 2 packed 128-bit vectors of [8 x i16...
Definition tmmintrin.h:427
_mm_abs_pi16
static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX _mm_abs_pi16(__m64 __a)
Computes the absolute value of each of the packed 16-bit signed integers in the source operand and st...
Definition tmmintrin.h:68
_mm_hadds_pi16
static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX _mm_hadds_pi16(__m64 __a, __m64 __b)
Horizontally adds the adjacent pairs of values contained in 2 packed 64-bit vectors of [4 x i16].
Definition tmmintrin.h:310
_mm_sign_pi8
static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX _mm_sign_pi8(__m64 __a, __m64 __b)
For each 8-bit integer in the first source operand, perform one of the following actions as specified...
Definition tmmintrin.h:711
_mm_sign_pi32
static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX _mm_sign_pi32(__m64 __a, __m64 __b)
For each 32-bit integer in the first source operand, perform one of the following actions as specifie...
Definition tmmintrin.h:763
_mm_mulhrs_pi16
static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX _mm_mulhrs_pi16(__m64 __a, __m64 __b)
Multiplies packed 16-bit signed integer values, truncates the 32-bit products to the 18 most signific...
Definition tmmintrin.h:556
_mm_hadd_pi16
static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX _mm_hadd_pi16(__m64 __a, __m64 __b)
Horizontally adds the adjacent pairs of values contained in 2 packed 64-bit vectors of [4 x i16].
Definition tmmintrin.h:237
_mm_hsubs_pi16
static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX _mm_hsubs_pi16(__m64 __a, __m64 __b)
Horizontally subtracts the adjacent pairs of values contained in 2 packed 64-bit vectors of [4 x i16]...
Definition tmmintrin.h:452
_mm_hadds_epi16
static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_hadds_epi16(__m128i __a, __m128i __b)
Horizontally adds the adjacent pairs of values contained in 2 packed 128-bit vectors of [8 x i16].
Definition tmmintrin.h:285
_mm_sign_epi32
static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_sign_epi32(__m128i __a, __m128i __b)
For each 32-bit integer in the first source operand, perform one of the following actions as specifie...
Definition tmmintrin.h:685
_mm_hadd_epi16
static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_hadd_epi16(__m128i __a, __m128i __b)
Horizontally adds the adjacent pairs of values contained in 2 packed 128-bit vectors of [8 x i16].
Definition tmmintrin.h:191
_mm_shuffle_epi8
static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_shuffle_epi8(__m128i __a, __m128i __b)
Copies the 8-bit integers from a 128-bit integer vector to the destination or clears 8-bit values in ...
Definition tmmintrin.h:582
_mm_hsub_pi32
static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX _mm_hsub_pi32(__m64 __a, __m64 __b)
Horizontally subtracts the adjacent pairs of values contained in 2 packed 64-bit vectors of [2 x i32]...
Definition tmmintrin.h:402
_mm_abs_epi32
static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_abs_epi32(__m128i __a)
Computes the absolute value of each of the packed 32-bit signed integers in the source operand and st...
Definition tmmintrin.h:122
_mm_abs_pi32
static __inline__ __m64 __DEFAULT_FN_ATTRS_MMX _mm_abs_pi32(__m64 __a)
Computes the absolute value of each of the packed 32-bit signed integers in the source operand and st...
Definition tmmintrin.h:104
Generated on Mon Feb 17 2025 13:12:28 for ETISS 0.8.0 by doxygen 1.9.8