diff options
Diffstat (limited to 'volk/lib')
| -rw-r--r-- | volk/lib/assembly.h | 67 | ||||
| -rw-r--r-- | volk/lib/cpuid_x86.S | 60 | ||||
| -rw-r--r-- | volk/lib/cpuid_x86_64.S | 54 | ||||
| -rw-r--r-- | volk/lib/gcc_x86_cpuid.h | 178 | ||||
| -rw-r--r-- | volk/lib/qa_32fc_conjugate_dot_prod_unaligned.cc | 138 | ||||
| -rw-r--r-- | volk/lib/qa_32fc_conjugate_dot_prod_unaligned.h | 18 | ||||
| -rw-r--r-- | volk/lib/qa_volk.cc | 213 |
7 files changed, 547 insertions, 181 deletions
diff --git a/volk/lib/assembly.h b/volk/lib/assembly.h deleted file mode 100644 index 8a99aa07c..000000000 --- a/volk/lib/assembly.h +++ /dev/null @@ -1,67 +0,0 @@ -/* -*- c++ -*- */ -/* - * Copyright 2002 Free Software Foundation, Inc. - * - * This file is part of GNU Radio - * - * GNU Radio is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 3, or (at your option) - * any later version. - * - * GNU Radio is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with GNU Radio; see the file COPYING. If not, write to - * the Free Software Foundation, Inc., 51 Franklin Street, - * Boston, MA 02110-1301, USA. - */ - -#ifndef _ASSEMBLY_H_ -#define _ASSEMBLY_H_ - -#if defined (__APPLE__) && defined (__APPLE_CC__) - -// XCode ignores the .scl and .type functions in XCode 2.2.1 and 2.3, -// but creates an error in XCode 2.4. Just ignore them. - -#define GLOB_SYMB(f) _ ## f - -#define DEF_FUNC_HEAD(f) /* none */ - -#define FUNC_TAIL(f) /* none*/ - -#elif !defined (__ELF__) - -/* - * Too bad, the following define does not work as expected --SF - * #define GLOB_SYMB(f) __USER_LABEL_PREFIX__ ## f - */ -#define GLOB_SYMB(f) _ ## f - -#define DEF_FUNC_HEAD(f) \ - .def GLOB_SYMB(f); .scl 2; .type 32; .endef - -#define FUNC_TAIL(f) /* none */ - - -#else /* !__ELF__ */ - - -#define GLOB_SYMB(f) f - -#define DEF_FUNC_HEAD(f) \ - .type GLOB_SYMB(f),@function \ - -#define FUNC_TAIL(f) \ - .Lfe1: \ - .size GLOB_SYMB(f),.Lfe1-GLOB_SYMB(f) - - -#endif /* !__ELF__ */ - - -#endif /* _ASSEMBLY_H_ */ diff --git a/volk/lib/cpuid_x86.S b/volk/lib/cpuid_x86.S deleted file mode 100644 index 4e1a9404f..000000000 --- a/volk/lib/cpuid_x86.S +++ /dev/null @@ -1,60 +0,0 @@ -# -# Copyright 2003 Free Software Foundation, Inc. -# -# This file is part of GNU Radio -# -# GNU Radio is free software; you can redistribute it and/or modify -# it under the terms of the GNU General Public License as published by -# the Free Software Foundation; either version 3, or (at your option) -# any later version. -# -# GNU Radio is distributed in the hope that it will be useful, -# but WITHOUT ANY WARRANTY; without even the implied warranty of -# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -# GNU General Public License for more details. -# -# You should have received a copy of the GNU General Public License -# along with GNU Radio; see the file COPYING. If not, write to -# the Free Software Foundation, Inc., 51 Franklin Street, -# Boston, MA 02110-1301, USA. -# - -# -# execute CPUID instruction, return EAX, EBX, ECX and EDX values in result -# -# void cpuid_x86 (unsigned int op, unsigned int result[4]); -# - -#include "assembly.h" - -.file "cpuid_x86.S" - .version "01.01" -.text -.globl GLOB_SYMB(cpuid_x86) - DEF_FUNC_HEAD(cpuid_x86) -GLOB_SYMB(cpuid_x86): - pushl %ebp - movl %esp, %ebp - pushl %ebx # must save in PIC mode, holds GOT pointer - pushl %esi - - movl 8(%ebp), %eax # op - movl 12(%ebp), %esi # result - cpuid - movl %eax, 0(%esi) - movl %ebx, 4(%esi) - movl %ecx, 8(%esi) - movl %edx, 12(%esi) - - popl %esi - popl %ebx - popl %ebp - ret - -FUNC_TAIL(cpuid_x86) - .ident "Hand coded cpuid assembly" - - -#if defined(__linux__) && defined(__ELF__) -.section .note.GNU-stack,"",%progbits -#endif diff --git a/volk/lib/cpuid_x86_64.S b/volk/lib/cpuid_x86_64.S deleted file mode 100644 index 32b1847cd..000000000 --- a/volk/lib/cpuid_x86_64.S +++ /dev/null @@ -1,54 +0,0 @@ -# -# Copyright 2003,2005 Free Software Foundation, Inc. -# -# This file is part of GNU Radio -# -# GNU Radio is free software; you can redistribute it and/or modify -# it under the terms of the GNU General Public License as published by -# the Free Software Foundation; either version 3, or (at your option) -# any later version. -# -# GNU Radio is distributed in the hope that it will be useful, -# but WITHOUT ANY WARRANTY; without even the implied warranty of -# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -# GNU General Public License for more details. -# -# You should have received a copy of the GNU General Public License -# along with GNU Radio; see the file COPYING. If not, write to -# the Free Software Foundation, Inc., 51 Franklin Street, -# Boston, MA 02110-1301, USA. -# - -# -# execute CPUID instruction, return EAX, EBX, ECX and EDX values in result -# -# void cpuid_x86 (unsigned int op, unsigned int result[4]); -# - -#include "assembly.h" - -.file "cpuid_x86_64.S" - .version "01.01" -.text -.globl GLOB_SYMB(cpuid_x86) - DEF_FUNC_HEAD(cpuid_x86) -GLOB_SYMB(cpuid_x86): - mov %rbx, %r11 # must save in PIC mode, holds GOT pointer - - mov %rdi, %rax # op - cpuid - movl %eax, 0(%rsi) # result - movl %ebx, 4(%rsi) - movl %ecx, 8(%rsi) - movl %edx, 12(%rsi) - - mov %r11, %rbx - retq - -FUNC_TAIL(cpuid_x86) - .ident "Hand coded cpuid64 assembly" - - -#if defined(__linux__) && defined(__ELF__) -.section .note.GNU-stack,"",%progbits -#endif diff --git a/volk/lib/gcc_x86_cpuid.h b/volk/lib/gcc_x86_cpuid.h new file mode 100644 index 000000000..2d0916fb3 --- /dev/null +++ b/volk/lib/gcc_x86_cpuid.h @@ -0,0 +1,178 @@ +/* + * Copyright (C) 2007, 2008, 2009 Free Software Foundation, Inc. + * + * This file is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 3, or (at your option) any + * later version. + * + * This file is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * Under Section 7 of GPL version 3, you are granted additional + * permissions described in the GCC Runtime Library Exception, version + * 3.1, as published by the Free Software Foundation. + * + * You should have received a copy of the GNU General Public License and + * a copy of the GCC Runtime Library Exception along with this program; + * see the files COPYING3 and COPYING.RUNTIME respectively. If not, see + * <http://www.gnu.org/licenses/>. + */ + +/* %ecx */ +#define bit_SSE3 (1 << 0) +#define bit_PCLMUL (1 << 1) +#define bit_SSSE3 (1 << 9) +#define bit_FMA (1 << 12) +#define bit_CMPXCHG16B (1 << 13) +#define bit_SSE4_1 (1 << 19) +#define bit_SSE4_2 (1 << 20) +#define bit_MOVBE (1 << 22) +#define bit_POPCNT (1 << 23) +#define bit_AES (1 << 25) +#define bit_XSAVE (1 << 26) +#define bit_OSXSAVE (1 << 27) +#define bit_AVX (1 << 28) + +/* %edx */ +#define bit_CMPXCHG8B (1 << 8) +#define bit_CMOV (1 << 15) +#define bit_MMX (1 << 23) +#define bit_FXSAVE (1 << 24) +#define bit_SSE (1 << 25) +#define bit_SSE2 (1 << 26) + +/* Extended Features */ +/* %ecx */ +#define bit_LAHF_LM (1 << 0) +#define bit_SSE4a (1 << 6) +#define bit_SSE5 (1 << 11) + +/* %edx */ +#define bit_LM (1 << 29) +#define bit_3DNOWP (1 << 30) +#define bit_3DNOW (1 << 31) + + +#if defined(__i386__) && defined(__PIC__) +/* %ebx may be the PIC register. */ +#if __GNUC__ >= 3 +#define __cpuid(level, a, b, c, d) \ + __asm__ ("xchg{l}\t{%%}ebx, %1\n\t" \ + "cpuid\n\t" \ + "xchg{l}\t{%%}ebx, %1\n\t" \ + : "=a" (a), "=r" (b), "=c" (c), "=d" (d) \ + : "0" (level)) + +#define __cpuid_count(level, count, a, b, c, d) \ + __asm__ ("xchg{l}\t{%%}ebx, %1\n\t" \ + "cpuid\n\t" \ + "xchg{l}\t{%%}ebx, %1\n\t" \ + : "=a" (a), "=r" (b), "=c" (c), "=d" (d) \ + : "0" (level), "2" (count)) +#else +/* Host GCCs older than 3.0 weren't supporting Intel asm syntax + nor alternatives in i386 code. */ +#define __cpuid(level, a, b, c, d) \ + __asm__ ("xchgl\t%%ebx, %1\n\t" \ + "cpuid\n\t" \ + "xchgl\t%%ebx, %1\n\t" \ + : "=a" (a), "=r" (b), "=c" (c), "=d" (d) \ + : "0" (level)) + +#define __cpuid_count(level, count, a, b, c, d) \ + __asm__ ("xchgl\t%%ebx, %1\n\t" \ + "cpuid\n\t" \ + "xchgl\t%%ebx, %1\n\t" \ + : "=a" (a), "=r" (b), "=c" (c), "=d" (d) \ + : "0" (level), "2" (count)) +#endif +#else +#define __cpuid(level, a, b, c, d) \ + __asm__ ("cpuid\n\t" \ + : "=a" (a), "=b" (b), "=c" (c), "=d" (d) \ + : "0" (level)) + +#define __cpuid_count(level, count, a, b, c, d) \ + __asm__ ("cpuid\n\t" \ + : "=a" (a), "=b" (b), "=c" (c), "=d" (d) \ + : "0" (level), "2" (count)) +#endif + +/* Return highest supported input value for cpuid instruction. ext can + be either 0x0 or 0x8000000 to return highest supported value for + basic or extended cpuid information. Function returns 0 if cpuid + is not supported or whatever cpuid returns in eax register. If sig + pointer is non-null, then first four bytes of the signature + (as found in ebx register) are returned in location pointed by sig. */ + +static __inline unsigned int +__get_cpuid_max (unsigned int __ext, unsigned int *__sig) +{ + unsigned int __eax, __ebx, __ecx, __edx; + +#ifndef __x86_64__ +#if __GNUC__ >= 3 + /* See if we can use cpuid. On AMD64 we always can. */ + __asm__ ("pushf{l|d}\n\t" + "pushf{l|d}\n\t" + "pop{l}\t%0\n\t" + "mov{l}\t{%0, %1|%1, %0}\n\t" + "xor{l}\t{%2, %0|%0, %2}\n\t" + "push{l}\t%0\n\t" + "popf{l|d}\n\t" + "pushf{l|d}\n\t" + "pop{l}\t%0\n\t" + "popf{l|d}\n\t" + : "=&r" (__eax), "=&r" (__ebx) + : "i" (0x00200000)); +#else +/* Host GCCs older than 3.0 weren't supporting Intel asm syntax + nor alternatives in i386 code. */ + __asm__ ("pushfl\n\t" + "pushfl\n\t" + "popl\t%0\n\t" + "movl\t%0, %1\n\t" + "xorl\t%2, %0\n\t" + "pushl\t%0\n\t" + "popfl\n\t" + "pushfl\n\t" + "popl\t%0\n\t" + "popfl\n\t" + : "=&r" (__eax), "=&r" (__ebx) + : "i" (0x00200000)); +#endif + + if (!((__eax ^ __ebx) & 0x00200000)) + return 0; +#endif + + /* Host supports cpuid. Return highest supported cpuid input value. */ + __cpuid (__ext, __eax, __ebx, __ecx, __edx); + + if (__sig) + *__sig = __ebx; + + return __eax; +} + +/* Return cpuid data for requested cpuid level, as found in returned + eax, ebx, ecx and edx registers. The function checks if cpuid is + supported and returns 1 for valid cpuid information or 0 for + unsupported cpuid level. All pointers are required to be non-null. */ + +static __inline int +__get_cpuid (unsigned int __level, + unsigned int *__eax, unsigned int *__ebx, + unsigned int *__ecx, unsigned int *__edx) +{ + unsigned int __ext = __level & 0x80000000; + + if (__get_cpuid_max (__ext, 0) < __level) + return 0; + + __cpuid (__level, *__eax, *__ebx, *__ecx, *__edx); + return 1; +} diff --git a/volk/lib/qa_32fc_conjugate_dot_prod_unaligned.cc b/volk/lib/qa_32fc_conjugate_dot_prod_unaligned.cc new file mode 100644 index 000000000..a0680bab6 --- /dev/null +++ b/volk/lib/qa_32fc_conjugate_dot_prod_unaligned.cc @@ -0,0 +1,138 @@ +#include <volk/volk.h> +#include <qa_32fc_conjugate_dot_prod_unaligned.h> +#include <stdlib.h> +#include <math.h> +#include <time.h> + + +#define assertcomplexEqual(expected, actual, delta) \ + CPPUNIT_ASSERT_DOUBLES_EQUAL (std::real(expected), std::real(actual), fabs(std::real(expected)) * delta); \ + CPPUNIT_ASSERT_DOUBLES_EQUAL (std::imag(expected), std::imag(actual), fabs(std::imag(expected))* delta); + +#define ERR_DELTA (1e-4) + +//test for sse + +#if LV_HAVE_SSE && LV_HAVE_64 + +static float uniform() { + return 2.0 * ((float) rand() / RAND_MAX - 0.5); // uniformly (-1, 1) +} + +static void +random_floats (float *buf, unsigned n) +{ + for (unsigned i = 0; i < n; i++) + buf[i] = uniform () * 32767; +} + + +void qa_32fc_conjugate_dot_prod_unaligned::t1() { + const int vlen = 789743; + + volk_environment_init(); + int ret; + + std::complex<float>* input; + std::complex<float>* taps; + + std::complex<float>* result_generic; + std::complex<float>* result; + + ret = posix_memalign((void**)&input, 16, vlen << 3); + ret = posix_memalign((void**)&taps, 16, vlen << 3); + ret = posix_memalign((void**)&result_generic, 16, 8); + ret = posix_memalign((void**)&result, 16, 8); + + + result_generic[0] = std::complex<float>(0,0); + result[0] = std::complex<float>(0,0); + + random_floats((float*)input, vlen * 2); + random_floats((float*)taps, vlen * 2); + + + + volk_32fc_conjugate_dot_prod_unaligned_manual(result_generic, input, taps, vlen * 8, "generic"); + + + volk_32fc_conjugate_dot_prod_unaligned_manual(result, input, taps, vlen * 8, "sse"); + + printf("32fc_conjugate_dot_prod_unaligned\n"); + printf("generic: %f +i%f ... sse: %f +i%f\n", std::real(result_generic[0]), std::imag(result_generic[0]), std::real(result[0]), std::imag(result[0])); + + assertcomplexEqual(result_generic[0], result[0], ERR_DELTA); + + free(input); + free(taps); + free(result_generic); + free(result); + +} + + +#elif LV_HAVE_SSE && LV_HAVE_32 + +static float uniform() { + return 2.0 * ((float) rand() / RAND_MAX - 0.5); // uniformly (-1, 1) +} + +static void +random_floats (float *buf, unsigned n) +{ + for (unsigned i = 0; i < n; i++) + buf[i] = uniform () * 32767; +} + + +void qa_32fc_conjugate_dot_prod_unaligned::t1() { + const int vlen = 789743; + + volk_environment_init(); + int ret; + + std::complex<float>* input; + std::complex<float>* taps; + + std::complex<float>* result_generic; + std::complex<float>* result; + + ret = posix_memalign((void**)&input, 16, vlen << 3); + ret = posix_memalign((void**)&taps, 16, vlen << 3); + ret = posix_memalign((void**)&result_generic, 16, 8); + ret = posix_memalign((void**)&result, 16, 8); + + + result_generic[0] = std::complex<float>(0,0); + result[0] = std::complex<float>(0,0); + + random_floats((float*)input, vlen * 2); + random_floats((float*)taps, vlen * 2); + + + + volk_32fc_conjugate_dot_prod_unaligned_manual(result_generic, input, taps, vlen * 8, "generic"); + + + volk_32fc_conjugate_dot_prod_unaligned_manual(result, input, taps, vlen * 8, "sse_32"); + + printf("32fc_conjugate_dot_prod_unaligned\n"); + printf("generic: %f +i%f ... sse: %f +i%f\n", std::real(result_generic[0]), std::imag(result_generic[0]), std::real(result[0]), std::imag(result[0])); + + assertcomplexEqual(result_generic[0], result[0], ERR_DELTA); + + free(input); + free(taps); + free(result_generic); + free(result); + +} + + +#else + +void qa_32fc_conjugate_dot_prod_unaligned::t1() { + printf("sse not available... no test performed\n"); +} + +#endif /*LV_HAVE_SSE*/ diff --git a/volk/lib/qa_32fc_conjugate_dot_prod_unaligned.h b/volk/lib/qa_32fc_conjugate_dot_prod_unaligned.h new file mode 100644 index 000000000..7aead53a1 --- /dev/null +++ b/volk/lib/qa_32fc_conjugate_dot_prod_unaligned.h @@ -0,0 +1,18 @@ +#ifndef INCLUDED_QA_32FC_CONJUGATE_DOT_PROD_UNALIGNED_H +#define INCLUDED_QA_32FC_CONJUGATE_DOT_PROD_UNALIGNED_H + +#include <cppunit/extensions/HelperMacros.h> +#include <cppunit/TestCase.h> + +class qa_32fc_conjugate_dot_prod_unaligned : public CppUnit::TestCase { + + CPPUNIT_TEST_SUITE (qa_32fc_conjugate_dot_prod_unaligned); + CPPUNIT_TEST (t1); + CPPUNIT_TEST_SUITE_END (); + + private: + void t1 (); +}; + + +#endif /* INCLUDED_QA_32FC_CONJUGATE_DOT_PROD_UNALIGNED_H */ diff --git a/volk/lib/qa_volk.cc b/volk/lib/qa_volk.cc new file mode 100644 index 000000000..98d3e9728 --- /dev/null +++ b/volk/lib/qa_volk.cc @@ -0,0 +1,213 @@ +/* + * Copyright 2008 Free Software Foundation, Inc. + * + * This file is part of GNU Radio + * + * GNU Radio is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 3, or (at your option) + * any later version. + * + * GNU Radio is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with GNU Radio; see the file COPYING. If not, write to + * the Free Software Foundation, Inc., 51 Franklin Street, + * Boston, MA 02110-1301, USA. + */ + +/* + * This class gathers together all the test cases for the example + * directory into a single test suite. As you create new test cases, + * add them here. + */ + +#include <qa_volk.h> +#include <qa_16s_quad_max_star_aligned16.h> +#include <qa_32fc_dot_prod_aligned16.h> +#include <qa_32fc_square_dist_aligned16.h> +#include <qa_32fc_square_dist_scalar_mult_aligned16.h> +#include <qa_32f_sum_of_poly_aligned16.h> +#include <qa_32fc_index_max_aligned16.h> +#include <qa_32f_index_max_aligned16.h> +#include <qa_32fc_conjugate_dot_prod_aligned16.h> +#include <qa_32fc_conjugate_dot_prod_unaligned.h> +#include <qa_16s_permute_and_scalar_add_aligned16.h> +#include <qa_16s_branch_4_state_8_aligned16.h> +#include <qa_16s_max_star_horizontal_aligned16.h> +#include <qa_16s_max_star_aligned16.h> +#include <qa_16s_add_quad_aligned16.h> +#include <qa_32f_add_aligned16.h> +#include <qa_32f_subtract_aligned16.h> +#include <qa_32f_max_aligned16.h> +#include <qa_32f_min_aligned16.h> +#include <qa_64f_max_aligned16.h> +#include <qa_64f_min_aligned16.h> +#include <qa_32s_and_aligned16.h> +#include <qa_32s_or_aligned16.h> +#include <qa_32f_dot_prod_aligned16.h> +#include <qa_32f_dot_prod_unaligned16.h> +#include <qa_32f_fm_detect_aligned16.h> +#include <qa_32fc_32f_multiply_aligned16.h> +#include <qa_32fc_multiply_aligned16.h> +#include <qa_32f_divide_aligned16.h> +#include <qa_32f_multiply_aligned16.h> +#include <qa_32f_sqrt_aligned16.h> +#include <qa_8sc_multiply_conjugate_16sc_aligned16.h> +#include <qa_8sc_multiply_conjugate_32fc_aligned16.h> +#include <qa_32u_popcnt_aligned16.h> +#include <qa_64u_popcnt_aligned16.h> +#include <qa_16u_byteswap_aligned16.h> +#include <qa_32u_byteswap_aligned16.h> +#include <qa_64u_byteswap_aligned16.h> +#include <qa_32f_normalize_aligned16.h> +#include <qa_16sc_deinterleave_16s_aligned16.h> +#include <qa_16sc_deinterleave_32f_aligned16.h> +#include <qa_16sc_deinterleave_real_16s_aligned16.h> +#include <qa_16sc_deinterleave_real_32f_aligned16.h> +#include <qa_16sc_deinterleave_real_8s_aligned16.h> +#include <qa_16sc_magnitude_16s_aligned16.h> +#include <qa_16sc_magnitude_32f_aligned16.h> +#include <qa_32fc_deinterleave_32f_aligned16.h> +#include <qa_32fc_deinterleave_64f_aligned16.h> +#include <qa_32fc_deinterleave_real_16s_aligned16.h> +#include <qa_32fc_deinterleave_real_32f_aligned16.h> +#include <qa_32fc_deinterleave_real_64f_aligned16.h> +#include <qa_32fc_magnitude_16s_aligned16.h> +#include <qa_32fc_magnitude_32f_aligned16.h> +#include <qa_32f_interleave_16sc_aligned16.h> +#include <qa_32f_interleave_32fc_aligned16.h> +#include <qa_8sc_deinterleave_16s_aligned16.h> +#include <qa_8sc_deinterleave_32f_aligned16.h> +#include <qa_8sc_deinterleave_real_16s_aligned16.h> +#include <qa_8sc_deinterleave_real_32f_aligned16.h> +#include <qa_8sc_deinterleave_real_8s_aligned16.h> +#include <qa_16s_convert_32f_aligned16.h> +#include <qa_16s_convert_32f_unaligned16.h> +#include <qa_16s_convert_8s_aligned16.h> +#include <qa_16s_convert_8s_unaligned16.h> +#include <qa_32f_convert_16s_aligned16.h> +#include <qa_32f_convert_16s_unaligned16.h> +#include <qa_32f_convert_32s_aligned16.h> +#include <qa_32f_convert_32s_unaligned16.h> +#include <qa_32f_convert_64f_aligned16.h> +#include <qa_32f_convert_64f_unaligned16.h> +#include <qa_32f_convert_8s_aligned16.h> +#include <qa_32f_convert_8s_unaligned16.h> +#include <qa_32s_convert_32f_aligned16.h> +#include <qa_32s_convert_32f_unaligned16.h> +#include <qa_64f_convert_32f_aligned16.h> +#include <qa_64f_convert_32f_unaligned16.h> +#include <qa_8s_convert_16s_aligned16.h> +#include <qa_8s_convert_16s_unaligned16.h> +#include <qa_8s_convert_32f_aligned16.h> +#include <qa_8s_convert_32f_unaligned16.h> +#include <qa_32fc_32f_power_32fc_aligned16.h> +#include <qa_32f_power_aligned16.h> +#include <qa_32fc_atan2_32f_aligned16.h> +#include <qa_32fc_power_spectral_density_32f_aligned16.h> +#include <qa_32fc_power_spectrum_32f_aligned16.h> +#include <qa_32f_calc_spectral_noise_floor_aligned16.h> +#include <qa_32f_accumulator_aligned16.h> +#include <qa_32f_stddev_aligned16.h> +#include <qa_32f_stddev_and_mean_aligned16.h> + +CppUnit::TestSuite * +qa_volk::suite() +{ + CppUnit::TestSuite *s = new CppUnit::TestSuite("volk"); + + s->addTest(qa_16s_quad_max_star_aligned16::suite()); + s->addTest(qa_32fc_dot_prod_aligned16::suite()); + s->addTest(qa_32fc_square_dist_scalar_mult_aligned16::suite()); + s->addTest(qa_32fc_square_dist_aligned16::suite()); + s->addTest(qa_32f_sum_of_poly_aligned16::suite()); + s->addTest(qa_32fc_index_max_aligned16::suite()); + s->addTest(qa_32f_index_max_aligned16::suite()); + s->addTest(qa_32fc_conjugate_dot_prod_aligned16::suite()); + s->addTest(qa_32fc_conjugate_dot_prod_unaligned::suite()); + s->addTest(qa_16s_permute_and_scalar_add_aligned16::suite()); + s->addTest(qa_16s_branch_4_state_8_aligned16::suite()); + s->addTest(qa_16s_max_star_horizontal_aligned16::suite()); + s->addTest(qa_16s_max_star_aligned16::suite()); + s->addTest(qa_16s_add_quad_aligned16::suite()); + s->addTest(qa_32f_add_aligned16::suite()); + s->addTest(qa_32f_subtract_aligned16::suite()); + s->addTest(qa_32f_max_aligned16::suite()); + s->addTest(qa_32f_min_aligned16::suite()); + s->addTest(qa_64f_max_aligned16::suite()); + s->addTest(qa_64f_min_aligned16::suite()); + s->addTest(qa_32s_and_aligned16::suite()); + s->addTest(qa_32s_or_aligned16::suite()); + s->addTest(qa_32f_dot_prod_aligned16::suite()); + s->addTest(qa_32f_dot_prod_unaligned16::suite()); + s->addTest(qa_32f_fm_detect_aligned16::suite()); + s->addTest(qa_32fc_32f_multiply_aligned16::suite()); + s->addTest(qa_32fc_multiply_aligned16::suite()); + s->addTest(qa_32f_divide_aligned16::suite()); + s->addTest(qa_32f_multiply_aligned16::suite()); + s->addTest(qa_32f_sqrt_aligned16::suite()); + s->addTest(qa_8sc_multiply_conjugate_16sc_aligned16::suite()); + s->addTest(qa_8sc_multiply_conjugate_32fc_aligned16::suite()); + s->addTest(qa_32u_popcnt_aligned16::suite()); + s->addTest(qa_64u_popcnt_aligned16::suite()); + s->addTest(qa_16u_byteswap_aligned16::suite()); + s->addTest(qa_32u_byteswap_aligned16::suite()); + s->addTest(qa_64u_byteswap_aligned16::suite()); + s->addTest(qa_32f_normalize_aligned16::suite()); + s->addTest(qa_16sc_deinterleave_16s_aligned16::suite()); + s->addTest(qa_16sc_deinterleave_32f_aligned16::suite()); + s->addTest(qa_16sc_deinterleave_real_16s_aligned16::suite()); + s->addTest(qa_16sc_deinterleave_real_32f_aligned16::suite()); + s->addTest(qa_16sc_deinterleave_real_8s_aligned16::suite()); + s->addTest(qa_16sc_magnitude_16s_aligned16::suite()); + s->addTest(qa_16sc_magnitude_32f_aligned16::suite()); + s->addTest(qa_32fc_deinterleave_32f_aligned16::suite()); + s->addTest(qa_32fc_deinterleave_64f_aligned16::suite()); + s->addTest(qa_32fc_deinterleave_real_16s_aligned16::suite()); + s->addTest(qa_32fc_deinterleave_real_32f_aligned16::suite()); + s->addTest(qa_32fc_deinterleave_real_64f_aligned16::suite()); + s->addTest(qa_32fc_magnitude_16s_aligned16::suite()); + s->addTest(qa_32fc_magnitude_32f_aligned16::suite()); + s->addTest(qa_32f_interleave_16sc_aligned16::suite()); + s->addTest(qa_32f_interleave_32fc_aligned16::suite()); + s->addTest(qa_8sc_deinterleave_16s_aligned16::suite()); + s->addTest(qa_8sc_deinterleave_32f_aligned16::suite()); + s->addTest(qa_8sc_deinterleave_real_16s_aligned16::suite()); + s->addTest(qa_8sc_deinterleave_real_32f_aligned16::suite()); + s->addTest(qa_8sc_deinterleave_real_8s_aligned16::suite()); + s->addTest(qa_16s_convert_32f_aligned16::suite()); + s->addTest(qa_16s_convert_32f_unaligned16::suite()); + s->addTest(qa_16s_convert_8s_aligned16::suite()); + s->addTest(qa_16s_convert_8s_unaligned16::suite()); + s->addTest(qa_32f_convert_16s_aligned16::suite()); + s->addTest(qa_32f_convert_16s_unaligned16::suite()); + s->addTest(qa_32f_convert_32s_aligned16::suite()); + s->addTest(qa_32f_convert_32s_unaligned16::suite()); + s->addTest(qa_32f_convert_64f_aligned16::suite()); + s->addTest(qa_32f_convert_64f_unaligned16::suite()); + s->addTest(qa_32f_convert_8s_aligned16::suite()); + s->addTest(qa_32f_convert_8s_unaligned16::suite()); + s->addTest(qa_32s_convert_32f_aligned16::suite()); + s->addTest(qa_32s_convert_32f_unaligned16::suite()); + s->addTest(qa_64f_convert_32f_aligned16::suite()); + s->addTest(qa_64f_convert_32f_unaligned16::suite()); + s->addTest(qa_8s_convert_16s_aligned16::suite()); + s->addTest(qa_8s_convert_16s_unaligned16::suite()); + s->addTest(qa_8s_convert_32f_aligned16::suite()); + s->addTest(qa_8s_convert_32f_unaligned16::suite()); + s->addTest(qa_32fc_32f_power_32fc_aligned16::suite()); + s->addTest(qa_32f_power_aligned16::suite()); + s->addTest(qa_32fc_atan2_32f_aligned16::suite()); + s->addTest(qa_32fc_power_spectral_density_32f_aligned16::suite()); + s->addTest(qa_32fc_power_spectrum_32f_aligned16::suite()); + s->addTest(qa_32f_calc_spectral_noise_floor_aligned16::suite()); + s->addTest(qa_32f_accumulator_aligned16::suite()); + s->addTest(qa_32f_stddev_aligned16::suite()); + s->addTest(qa_32f_stddev_and_mean_aligned16::suite()); + + return s; +} |