Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | Add initial AUTHORS fileHEADmaster | Timothy Pearson | 2015-12-21 | 1 | -0/+2 |
| | |||||
* | Add VSX double support for POWER8 | Timothy Pearson | 2015-12-21 | 2 | -1/+583 |
| | |||||
* | Initial fixes for POWER8 and GCC | Timothy Pearson | 2015-12-21 | 5 | -8/+12 |
| | |||||
* | Report AVX2 capabilities in capability string | Erik Schnetter | 2015-11-14 | 1 | -1/+6 |
| | |||||
* | Improve optimization barriers | Erik Schnetter | 2015-10-30 | 3 | -29/+29 |
| | |||||
* | Correct syntax error in AVX2 code | Erik Schnetter | 2015-10-29 | 1 | -6/+6 |
| | |||||
* | Update build instructions | Erik Schnetter | 2015-10-29 | 1 | -2/+2 |
| | |||||
* | Correct type name __m256i | Erik Schnetter | 2015-10-29 | 1 | -14/+14 |
| | |||||
* | Use C++03 standard for source code formatting | Erik Schnetter | 2015-10-16 | 3 | -38/+104 |
| | |||||
* | Reformat source code with clang-format | Erik Schnetter | 2015-10-16 | 44 | -17970/+15431 |
| | | | | | | | | | Note: If you have an existing checkout with modifications, then you should: (1) save your current state on a branch (2) run clang-format on this branch (3) compare this branch to a fresh checkout of the clang-formatted master Since both your branch and master have been formatted with clang-format, this will lead to a very small diff, avoiding spurious changes due to formatting differences. | ||||
* | Modern versions of Clang support __builtin_assume_aligned | Erik Schnetter | 2015-10-16 | 1 | -6/+0 |
| | |||||
* | Begin to implement AVX2 support for 16-bit types | Erik Schnetter | 2015-10-16 | 1 | -0/+29 |
| | |||||
* | Add Meson build script | Erik Schnetter | 2015-10-16 | 1 | -0/+15 |
| | |||||
* | Improve fallback static_assert implementation | Erik Schnetter | 2015-10-16 | 1 | -1/+5 |
| | |||||
* | Correct AVX2 to_bool implementation | Erik Schnetter | 2015-10-16 | 2 | -2/+2 |
| | | | | Thanks to Guillaume de Roujoux for finding this and providing a patch. | ||||
* | Use AVX2 intrinsics if available | Erik Schnetter | 2015-09-03 | 5 | -27/+356 |
| | |||||
* | Rename test to selftest | Erik Schnetter | 2015-09-03 | 2 | -1/+1 |
| | |||||
* | Add missing typecast | Erik Schnetter | 2015-02-28 | 1 | -1/+1 |
| | |||||
* | Correct SSE2 min/max | Erik Schnetter | 2015-01-21 | 1 | -2/+2 |
| | |||||
* | Describe exponent offset of float8 and float16 | Erik Schnetter | 2015-01-21 | 1 | -2/+2 |
| | |||||
* | Optimize AVX isignbit | Erik Schnetter | 2014-10-10 | 1 | -6/+1 |
| | |||||
* | Add missing #endif | Erik Schnetter | 2014-06-17 | 1 | -0/+1 |
| | |||||
* | Remove debug output | Erik Schnetter | 2014-06-07 | 1 | -4/+0 |
| | |||||
* | Do not accidentally disable builtin vectors' barrier function | Erik Schnetter | 2014-06-07 | 1 | -1/+1 |
| | |||||
* | Provide integer comparison operafors for AVX fp8 and fp16 vectors | Erik Schnetter | 2014-06-07 | 3 | -0/+25 |
| | |||||
* | Add constructors from double for fp8 and fp16 types | Erik Schnetter | 2014-06-07 | 2 | -0/+23 |
| | |||||
* | Correct calculating inverse of a bit pattern (counteract integer promotion) | Erik Schnetter | 2014-06-07 | 1 | -2/+4 |
| | |||||
* | Add constants for infinity and nan to floating point properties | Erik Schnetter | 2014-06-07 | 1 | -0/+6 |
| | |||||
* | Correct return types of some floating point properties | Erik Schnetter | 2014-06-07 | 1 | -6/+6 |
| | |||||
* | Benchmark Clang's builtin vectors as well | Erik Schnetter | 2014-03-24 | 1 | -6/+6 |
| | |||||
* | Improve log performance | Erik Schnetter | 2013-11-28 | 1 | -43/+30 |
| | |||||
* | Add to-do items | Erik Schnetter | 2013-11-28 | 2 | -0/+5 |
| | |||||
* | Indent | Erik Schnetter | 2013-11-28 | 1 | -2/+8 |
| | |||||
* | Update log coefficients | Erik Schnetter | 2013-11-28 | 2 | -387/+115 |
| | |||||
* | Don’t test builtin vectors; they are too slow | Erik Schnetter | 2013-11-28 | 1 | -2/+30 |
| | |||||
* | Speed up benchmarks | Erik Schnetter | 2013-11-28 | 1 | -1/+1 |
| | |||||
* | Update build instructions for out-of-source-tree building | Erik Schnetter | 2013-11-28 | 1 | -5/+2 |
| | |||||
* | Don’t define << or >> for VML_NO_IOSTREAM | Erik Schnetter | 2013-11-26 | 3 | -0/+6 |
| | |||||
* | vec_builtin is only for Clang | Erik Schnetter | 2013-11-26 | 1 | -45/+6 |
| | |||||
* | Correct syntax errors in NEON vectors | Erik Schnetter | 2013-11-26 | 2 | -2/+2 |
| | |||||
* | Improve auto-detecting of math functions provided by STL | Erik Schnetter | 2013-11-26 | 1 | -33/+40 |
| | |||||
* | Use __restrict__ instead of restrict | Erik Schnetter | 2013-11-26 | 1 | -2/+4 |
| | |||||
* | Some ideas to improve exp on QPX | Erik Schnetter | 2013-11-25 | 1 | -2/+28 |
| | |||||
* | Rename “automatic” vector types from float/double to float32/float64 for ↵ | Erik Schnetter | 2013-11-25 | 4 | -28/+28 |
| | | | | clarity | ||||
* | Run benchmarks of all functions, not just some | Erik Schnetter | 2013-11-25 | 1 | -90/+184 |
| | |||||
* | Begin to support builtin vectors on non-clang compilers | Erik Schnetter | 2013-11-25 | 1 | -7/+48 |
| | |||||
* | Only enable builtin vectors when clang is used | Erik Schnetter | 2013-11-25 | 2 | -1/+21 |
| | |||||
* | Correct clz implementation | Erik Schnetter | 2013-11-25 | 1 | -12/+20 |
| | |||||
* | If VML_NO_IOSTREAM is set, then don’t provide any iostream operations | Erik Schnetter | 2013-11-25 | 5 | -5/+20 |
| | | | | This reduces the size of the library. | ||||
* | Provide lrint | Erik Schnetter | 2013-11-25 | 9 | -1/+77 |
| |