|
| From: | Ron Economos |
| Subject: | Re: [Discuss-gnuradio] A lot confused by volk_32f_cos_32f() |
| Date: | Fri, 15 Jan 2016 15:04:58 -0800 |
| User-agent: | Mozilla/5.0 (X11; Linux i686; rv:24.0) Gecko/20100101 Thunderbird/24.6.0 |
Version v1.1-22-g99594b12 is too old. It is from before the 2fa96d9 commit. You need Volk version 1.2 or 1.1.2. Ron On 01/15/2016 02:52 PM, Dennis Glatting wrote:
On Fri, 2016-01-15 at 14:28 -0800, Ron Economos wrote:This issue has been fixed just recently. https://github.com/gnuradio/volk/issues/52If I understand https://github.com/n-west/volk/commit/2fa96d970dcb582ac 8a6c65ec2088df2a79747d5 (dated 1Dec2015) correctly, I am running the updated code installed on 12Jan2016: address@hidden:$ volk-config-info -v v1.1-22-g99594b12 address@hidden:$ ldd a.out linux-vdso.so.1 => (0x00007fffc23ea000) libvolk.so.1.1git => /usr/local/lib/libvolk.so.1.1git (0x00007f59b0370000) libstdc++.so.6 => /usr/lib/x86_64-linux-gnu/libstdc++.so.6 (0x00007f59affe8000) libm.so.6 => /lib/x86_64-linux-gnu/libm.so.6 (0x00007f59afce0000) libgcc_s.so.1 => /lib/x86_64-linux-gnu/libgcc_s.so.1 (0x00007f59afac8000) libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x00007f59af6f8000) liborc-0.4.so.0 => /usr/lib/x86_64-linux-gnu/liborc-0.4.so.0 (0x00007f59af470000) /lib64/ld-linux-x86-64.so.2 (0x00007f59b07f8000) libpthread.so.0 => /lib/x86_64-linux-gnu/libpthread.so.0 (0x00007f59af250000) address@hidden:# ls -lh /usr/local/lib/libvolk.so.1.1git /usr/local/lib/libvolk.so lrwxrwxrwx 1 root root 17 Jan 12 10:19 /usr/local/lib/libvolk.so -> libvolk.so.1.1git -rw-r--r-- 1 root root 2.7M Jan 12 10:41 /usr/local/lib/libvolk.so.1.1git There are no odd libvolk* libraries in my system. BTW, I am running: address@hidden:# cat /proc/cpuinfo processor : 0 vendor_id : AuthenticAMD cpu family : 21 model : 2 model name : AMD FX(tm)-9590 Eight-Core Processor stepping : 0 microcode : 0x6000822 cpu MHz : 4721.634 cache size : 2048 KB physical id : 0 siblings : 8 core id : 0 cpu cores : 4 apicid : 16 initial apicid : 0 fpu : yes fpu_exception : yes cpuid level : 13 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good nopl nonstop_tsc extd_apicid aperfmperf pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 popcnt aes xsave avx f16c lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs xop skinit wdt lwp fma4 tce nodeid_msr tbm topoext perfctr_core perfctr_nb arat cpb hw_pstate npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold vmmcall bmi1 bugs : fxsave_leak bogomips : 9443.26 TLB size : 1536 4K pages clflush size : 64 cache_alignment : 64 address sizes : 48 bits physical, 48 bits virtual power management: ts ttp tm 100mhzsteps hwpstate cpb eff_freq_ro processor : 1 vendor_id : AuthenticAMD cpu family : 21 model : 2 model name : AMD FX(tm)-9590 Eight-Core Processor stepping : 0 microcode : 0x6000822 cpu MHz : 4721.634 cache size : 2048 KB physical id : 0 siblings : 8 core id : 1 cpu cores : 4 apicid : 17 initial apicid : 1 fpu : yes fpu_exception : yes cpuid level : 13 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good nopl nonstop_tsc extd_apicid aperfmperf pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 popcnt aes xsave avx f16c lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs xop skinit wdt lwp fma4 tce nodeid_msr tbm topoext perfctr_core perfctr_nb arat cpb hw_pstate npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold vmmcall bmi1 bugs : fxsave_leak bogomips : 9443.26 TLB size : 1536 4K pages clflush size : 64 cache_alignment : 64 address sizes : 48 bits physical, 48 bits virtual power management: ts ttp tm 100mhzsteps hwpstate cpb eff_freq_ro processor : 2 vendor_id : AuthenticAMD cpu family : 21 model : 2 model name : AMD FX(tm)-9590 Eight-Core Processor stepping : 0 microcode : 0x6000822 cpu MHz : 4721.634 cache size : 2048 KB physical id : 0 siblings : 8 core id : 2 cpu cores : 4 apicid : 18 initial apicid : 2 fpu : yes fpu_exception : yes cpuid level : 13 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good nopl nonstop_tsc extd_apicid aperfmperf pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 popcnt aes xsave avx f16c lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs xop skinit wdt lwp fma4 tce nodeid_msr tbm topoext perfctr_core perfctr_nb arat cpb hw_pstate npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold vmmcall bmi1 bugs : fxsave_leak bogomips : 9443.26 TLB size : 1536 4K pages clflush size : 64 cache_alignment : 64 address sizes : 48 bits physical, 48 bits virtual power management: ts ttp tm 100mhzsteps hwpstate cpb eff_freq_ro processor : 3 vendor_id : AuthenticAMD cpu family : 21 model : 2 model name : AMD FX(tm)-9590 Eight-Core Processor stepping : 0 microcode : 0x6000822 cpu MHz : 4721.634 cache size : 2048 KB physical id : 0 siblings : 8 core id : 3 cpu cores : 4 apicid : 19 initial apicid : 3 fpu : yes fpu_exception : yes cpuid level : 13 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good nopl nonstop_tsc extd_apicid aperfmperf pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 popcnt aes xsave avx f16c lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs xop skinit wdt lwp fma4 tce nodeid_msr tbm topoext perfctr_core perfctr_nb arat cpb hw_pstate npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold vmmcall bmi1 bugs : fxsave_leak bogomips : 9443.26 TLB size : 1536 4K pages clflush size : 64 cache_alignment : 64 address sizes : 48 bits physical, 48 bits virtual power management: ts ttp tm 100mhzsteps hwpstate cpb eff_freq_ro processor : 4 vendor_id : AuthenticAMD cpu family : 21 model : 2 model name : AMD FX(tm)-9590 Eight-Core Processor stepping : 0 microcode : 0x6000822 cpu MHz : 4721.634 cache size : 2048 KB physical id : 0 siblings : 8 core id : 4 cpu cores : 4 apicid : 20 initial apicid : 4 fpu : yes fpu_exception : yes cpuid level : 13 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good nopl nonstop_tsc extd_apicid aperfmperf pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 popcnt aes xsave avx f16c lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs xop skinit wdt lwp fma4 tce nodeid_msr tbm topoext perfctr_core perfctr_nb arat cpb hw_pstate npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold vmmcall bmi1 bugs : fxsave_leak bogomips : 9443.26 TLB size : 1536 4K pages clflush size : 64 cache_alignment : 64 address sizes : 48 bits physical, 48 bits virtual power management: ts ttp tm 100mhzsteps hwpstate cpb eff_freq_ro processor : 5 vendor_id : AuthenticAMD cpu family : 21 model : 2 model name : AMD FX(tm)-9590 Eight-Core Processor stepping : 0 microcode : 0x6000822 cpu MHz : 4721.634 cache size : 2048 KB physical id : 0 siblings : 8 core id : 5 cpu cores : 4 apicid : 21 initial apicid : 5 fpu : yes fpu_exception : yes cpuid level : 13 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good nopl nonstop_tsc extd_apicid aperfmperf pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 popcnt aes xsave avx f16c lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs xop skinit wdt lwp fma4 tce nodeid_msr tbm topoext perfctr_core perfctr_nb arat cpb hw_pstate npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold vmmcall bmi1 bugs : fxsave_leak bogomips : 9443.26 TLB size : 1536 4K pages clflush size : 64 cache_alignment : 64 address sizes : 48 bits physical, 48 bits virtual power management: ts ttp tm 100mhzsteps hwpstate cpb eff_freq_ro processor : 6 vendor_id : AuthenticAMD cpu family : 21 model : 2 model name : AMD FX(tm)-9590 Eight-Core Processor stepping : 0 microcode : 0x6000822 cpu MHz : 4721.634 cache size : 2048 KB physical id : 0 siblings : 8 core id : 6 cpu cores : 4 apicid : 22 initial apicid : 6 fpu : yes fpu_exception : yes cpuid level : 13 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good nopl nonstop_tsc extd_apicid aperfmperf pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 popcnt aes xsave avx f16c lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs xop skinit wdt lwp fma4 tce nodeid_msr tbm topoext perfctr_core perfctr_nb arat cpb hw_pstate npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold vmmcall bmi1 bugs : fxsave_leak bogomips : 9443.26 TLB size : 1536 4K pages clflush size : 64 cache_alignment : 64 address sizes : 48 bits physical, 48 bits virtual power management: ts ttp tm 100mhzsteps hwpstate cpb eff_freq_ro processor : 7 vendor_id : AuthenticAMD cpu family : 21 model : 2 model name : AMD FX(tm)-9590 Eight-Core Processor stepping : 0 microcode : 0x6000822 cpu MHz : 4721.634 cache size : 2048 KB physical id : 0 siblings : 8 core id : 7 cpu cores : 4 apicid : 23 initial apicid : 7 fpu : yes fpu_exception : yes cpuid level : 13 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good nopl nonstop_tsc extd_apicid aperfmperf pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 popcnt aes xsave avx f16c lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs xop skinit wdt lwp fma4 tce nodeid_msr tbm topoext perfctr_core perfctr_nb arat cpb hw_pstate npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold vmmcall bmi1 bugs : fxsave_leak bogomips : 9443.26 TLB size : 1536 4K pages clflush size : 64 cache_alignment : 64 address sizes : 48 bits physical, 48 bits virtual power management: ts ttp tm 100mhzsteps hwpstate cpb eff_freq_roRon On 01/15/2016 02:20 PM, Dennis Glatting wrote:I am confused by this function because the output DOES NOT match ::cos() as I believe is demonstrated in the web page: http://libvolk.org/doxygen/volk_32f_cos_32f.html I /think/ I am doing the same thing as the web page. Regardless, my code (below) output is the following with sign mismatches at i=4 and i=14: address@hidden:~/gr-acars-code/3.7.5/lib$ ./a.out Using Volk machine: avx_64_mmx_orc i=0, input=0, volk=1, cos()=1 i=1, input=0.314159, volk=0.951056, cos()=0.951057 i=2, input=0.628319, volk=0.809017, cos()=0.809017 i=3, input=0.942478, volk=0.587785, cos()=0.587785 i=4, input=1.25664, volk=-0.309017, cos()=0.309017 i=5, input=1.5708, volk=-1.19209e-07, cos()=-4.37114e-08 i=6, input=1.88496, volk=-0.309017, cos()=-0.309017 i=7, input=2.19911, volk=-0.587785, cos()=-0.587785 i=8, input=2.51327, volk=-0.809017, cos()=-0.809017 i=9, input=2.82743, volk=-0.951057, cos()=-0.951057 i=10, input=3.14159, volk=-1, cos()=-1 i=11, input=3.45575, volk=-0.951056, cos()=-0.951057 i=12, input=3.76991, volk=-0.809017, cos()=-0.809017 i=13, input=4.08407, volk=-0.587785, cos()=-0.587785 i=14, input=4.39823, volk=0.309017, cos()=-0.309017 i=15, input=4.71239, volk=2.38419e-07, cos()=1.19249e-08 i=16, input=5.02655, volk=0.309017, cos()=0.309017 i=17, input=5.34071, volk=0.587785, cos()=0.587785 i=18, input=5.65487, volk=0.809017, cos()=0.809017 i=19, input=5.96903, volk=0.951057, cos()=0.951057 I am using: address@hidden:~/gr-acars-code/3.7.5/lib$ gnuradio-companion - -version GNU Radio Companion 3.7.10git-31-gb17bcb88 This program is part of GNU Radio GRC comes with ABSOLUTELY NO WARRANTY. This is free software, and you are welcome to redistribute it. On: address@hidden:~/gr-acars-code/3.7.5/lib$ lsb_release -a No LSB modules are available. Distributor ID: Ubuntu Description: Ubuntu 15.10 Release: 15.10 Codename: wily With compiler: gcc version 5.2.1 20151010 (Ubuntu 5.2.1-22ubuntu2) Here's how I compile my code: c++ -std=c++11 -O volk_test.cc -lvolkAnd, of course, the code:#include <iostream> #include <memory> extern "C" { #include <assert.h> } #include <volk/volk.h> #define PPP 20 int main( void ) { std::unique_ptr<float,std::function<void(float*)>> points2400((float*)volk_malloc( sizeof(float*)*( PPP + 1 ), volk_get_alignment()), [](float* p) { if( p ) volk_free((void*)p ); }); assert( points2400.get()); std::unique_ptr<float,std::function<void(float*)>> oBuf((float*)volk_malloc( sizeof(float*)*( PPP + 1 ), volk_get_alignment()), [](float* p) { if( p ) volk_free((void*)p ); }); assert( oBuf.get());for( int i = 0; i < PPP; ++i ) {const float ii = i;points2400.get()[i] = ii * ( 2400.0 / 48000.0 ) * ( 2.0 * M_PI);}volk_32f_cos_32f( oBuf.get(), points2400.get(), PPP ); for( int i = 0; i < PPP; ++i ) std::cout << "i=" << i << ", input=" << points2400.get()[i] << ", volk=" << oBuf.get()[i] << ", cos()=" << ::cos(points2400.get()[i]) <<std::endl;return 0;} _______________________________________________ Discuss-gnuradio mailing list address@hidden https://lists.gnu.org/mailman/listinfo/discuss-gnuradio_______________________________________________ Discuss-gnuradio mailing list address@hidden https://lists.gnu.org/mailman/listinfo/discuss-gnuradio
| [Prev in Thread] | Current Thread | [Next in Thread] |