Commit message (Collapse) | Author | Age | Files | Lines | ||
---|---|---|---|---|---|---|
... | ||||||
| * | Detect presence of malloc.h, fixes anthonix/ffts#40 | Jukka Ojanen | 2015-07-30 | 2 | -0/+8 | |
| | | ||||||
| * | Define [pa] and [pb] as constant input variables, not writable outputs | Jukka Ojanen | 2015-07-16 | 2 | -8/+5 | |
| | | ||||||
| * | Remove unreferenced header | Jukka Ojanen | 2015-07-15 | 1 | -2/+0 | |
| | | ||||||
| * | Improve compiler optimization by turning "patterns.c" to "patterns.h" | Jukka Ojanen | 2015-07-15 | 3 | -233/+506 | |
| | | ||||||
| * | Remove some dead code | Jukka Ojanen | 2015-07-15 | 1 | -19/+0 | |
| | | ||||||
| * | FFTS is no longer depended on any other math library, and this should help ↵ | Jukka Ojanen | 2015-07-14 | 5 | -82/+135 | |
| | | | | | | | | to verify its numerical accuracy. | |||||
| * | Move trigonometric stuff to separate file. | Jukka Ojanen | 2015-07-14 | 5 | -54/+274 | |
| | | | | | | | | Implemented Oscar Buneman's method for generating a sequence of sines and cosines. | |||||
| * | SSE3 detection failed with MSVC 2005 x64 | Jukka Ojanen | 2015-07-10 | 1 | -3/+12 | |
| | | ||||||
| * | Unroll loops to process 64 byte cache line per iteration | Jukka Ojanen | 2015-07-09 | 1 | -39/+205 | |
| | | ||||||
| * | Add new attributes to control/improve branch predictions | Jukka Ojanen | 2015-07-09 | 1 | -0/+12 | |
| | | ||||||
| * | Half the number of calls to sin/cos functions in ffts_init_1d_real | Jukka Ojanen | 2015-07-08 | 1 | -12/+68 | |
| | | ||||||
| * | Add SSE3 optimized version of ffts_execute_1d_real_inv | Jukka Ojanen | 2015-07-07 | 1 | -20/+78 | |
| | | ||||||
| * | Add SSE3 optimized version of ffts_execute_1d_real | Jukka Ojanen | 2015-07-07 | 1 | -13/+80 | |
| | | ||||||
| * | Add detection for SSE3 intrinsics | Jukka Ojanen | 2015-07-07 | 1 | -2/+20 | |
| | | ||||||
| * | To silence warning 'possible loss of data', use explicit casting to float | Jukka Ojanen | 2015-07-06 | 1 | -8/+8 | |
| | | ||||||
| * | SSE optimized versions of ffts_execute_1d_real and ffts_execute_1d_real_inv | Jukka Ojanen | 2015-07-06 | 1 | -4/+100 | |
| | | ||||||
| * | Add new attributes to help auto-vectorization | Jukka Ojanen | 2015-07-06 | 2 | -19/+54 | |
| | | ||||||
| * | Avoid allocating array of single pointer | Jukka Ojanen | 2015-07-06 | 1 | -10/+8 | |
| | | ||||||
| * | Fix ffts_aligned_free MinGW crash | Jukka Ojanen | 2015-07-06 | 1 | -1/+1 | |
| | | ||||||
| * | Incorrect stride with GCC flags "-march=native -ffast-math" | Jukka Ojanen | 2015-07-02 | 2 | -3/+5 | |
| | | | | | | | | Note that N/leaf_N is always a multiply of 2 | |||||
| * | Fix assertion failed in ffts_compare_offsets | Jukka Ojanen | 2015-07-02 | 1 | -4/+5 | |
| | | ||||||
| * | Generate cosine and sine table without using C math library. About 100 times ↵ | Jukka Ojanen | 2015-03-31 | 2 | -15/+54 | |
| | | | | | | | | faster on ARM and 15 times faster on x86. | |||||
| * | ffts_nd.c is using SSE2 intrinsics, detect and include emmintrin.h instead ↵ | Jukka Ojanen | 2015-03-19 | 2 | -4/+12 | |
| | | | | | | | | xmmintrin.h, and fix GCC error: inlining failed in call to always_inline '_mm_load_pd': target specific option mismatch by adding "-msse2" instead of "-msse" | |||||
| * | To support building for Windows with MinGW, don't assume MSVC to be the compiler | Jukka Ojanen | 2015-03-19 | 1 | -1/+1 | |
| | | ||||||
| * | Minimize sin/cos calculations by calculating all factors ones and generate ↵ | Jukka Ojanen | 2015-03-18 | 2 | -17/+31 | |
| | | | | | | | | lookup tables by mapping | |||||
| * | Remove unused sse.s | Jukka Ojanen | 2015-03-18 | 2 | -895/+0 | |
| | | ||||||
| * | Always run-time generate x64 dynamic code | Jukka Ojanen | 2015-03-18 | 1 | -841/+1335 | |
| | | ||||||
| * | Remove dependency on YASM as Windows dynamic code is run-time generated | Jukka Ojanen | 2015-03-17 | 2 | -854/+0 | |
| | | ||||||
| * | Determinate lookup table size using closed-form expression | Jukka Ojanen | 2015-03-16 | 1 | -22/+4 | |
| | | ||||||
| * | Remove dead code | Jukka Ojanen | 2015-03-16 | 1 | -98/+47 | |
| | | ||||||
| * | Don't generate lookup tables when size is less than 32 | Jukka Ojanen | 2015-03-16 | 1 | -5/+5 | |
| | | ||||||
| * | Merge ffts_small with ffts_static, and define small transforms "fully" constant | Jukka Ojanen | 2015-03-16 | 6 | -517/+585 | |
| | | ||||||
| * | Add string.h to fix implicit declaration of function 'memcpy' | Jukka Ojanen | 2015-03-13 | 1 | -0/+4 | |
| | | ||||||
| * | One more macro fix | Jukka Ojanen | 2015-03-13 | 1 | -2/+2 | |
| | | ||||||
| * | Forgot to rename some V macros | Jukka Ojanen | 2015-03-13 | 4 | -26/+28 | |
| | | ||||||
| * | Rename vector V as V4SF; vector of 4 single precision floats. Rename all ↵ | Jukka Ojanen | 2015-03-12 | 8 | -673/+796 | |
| | | | | | | | | | | | | vector V macros accordingly. Redefine ffts_constants as ffts_constants_32f and ffts_constants_64f. | |||||
| * | Replace data_t with float | Jukka Ojanen | 2015-03-12 | 1 | -31/+53 | |
| | | ||||||
| * | Remove unused neon_float.h header | Jukka Ojanen | 2015-03-12 | 1 | -1127/+0 | |
| | | ||||||
| * | Remove unused variable 'i' from 'ffts_generate_func_code' | Jukka Ojanen | 2015-03-12 | 1 | -4/+0 | |
| | | ||||||
| * | Initial steps to support double precision. Replace data_t with float, and ↵ | Jukka Ojanen | 2015-03-12 | 5 | -145/+413 | |
| | | | | | | | | cdata_t with ffts_cpx_32f. | |||||
| * | Try to execute detection quietly without messages | Jukka Ojanen | 2015-03-11 | 1 | -1/+5 | |
| | | ||||||
| * | Fix conflicting types | Jukka Ojanen | 2015-03-11 | 1 | -0/+2 | |
| | | ||||||
| * | Automatically detect ARM FPU and float ABI | Jukka Ojanen | 2015-03-11 | 1 | -18/+73 | |
| | | ||||||
| * | Fix conflicting types for 'temp0' | Jukka Ojanen | 2015-03-11 | 1 | -0/+2 | |
| | | ||||||
| * | Add explicit '.fpu' directive | Jukka Ojanen | 2015-03-11 | 3 | -0/+3 | |
| | | ||||||
| * | Use FFTS_INLINE instead of __INLINE and make functions static | Jukka Ojanen | 2015-03-11 | 1 | -17/+14 | |
| | | ||||||
| * | Added "-mfloat-abi=softfp" as default for ARM | Jukka Ojanen | 2015-03-11 | 1 | -1/+5 | |
| | | ||||||
| * | Now ARM VFP building works with "CMake -DENABLE_VFP=ON". Performance when ↵ | Jukka Ojanen | 2015-03-11 | 1 | -0/+2 | |
| | | | | | | | | "-DDYNAMIC_CODE_DISABLED=ON" is poor, but it can be improved with gcc switches. | |||||
| * | Don't use CMake option name | Jukka Ojanen | 2015-03-10 | 1 | -7/+7 | |
| | | ||||||
| * | ARM compile shall use generic C if dynamic code is disabled | Jukka Ojanen | 2015-03-10 | 1 | -7/+7 | |
| | |