summaryrefslogtreecommitdiffstats
path: root/vec_avx_float8.h
Commit message (Expand)AuthorAgeFilesLines
* Reformat source code with clang-formatErik Schnetter2015-10-161-763/+632
* Correct AVX2 to_bool implementationErik Schnetter2015-10-161-1/+1
* Use AVX2 intrinsics if availableErik Schnetter2015-09-031-2/+72
* Introduce mad(), as low-accuracy version of fma()Erik Schnetter2013-09-101-0/+4
* SSE/AVX: Use nicer type namesErik Schnetter2013-09-021-104/+113
* Provide clean set of libm elemental functions in namespace vml_stdErik Schnetter2013-09-021-10/+12
* Implement more integer functions: abs bitifthen clz isignbit max min popcount...Erik Schnetter2013-09-011-7/+55
* Make frexp take a pointer argument instead of a referenceErik Schnetter2013-08-211-1/+1
* Add potential optimization for SSE/AVX all/any functionsErik Schnetter2013-08-191-1/+1
* Implement ifthen with boolvec argumentsErik Schnetter2013-07-071-4/+8
* Simplify return type specificationsErik Schnetter2013-07-071-4/+4
* Do not require C++11 any moreErik Schnetter2013-07-061-8/+8
* Optimize prod and sumErik Schnetter2013-07-041-3/+9
* Implement maxval and minval reduction operationsErik Schnetter2013-07-041-0/+26
* Do not use type punning for converting between vectors and elements -- gcc 4....Erik Schnetter2013-07-031-6/+22
* Correct operator!= for AVX instructionsErik Schnetter2013-07-031-1/+1
* Whitespace change in asm intrinsicsErik Schnetter2013-07-021-1/+1
* Correct isnan on x86; improve isnan on other architecturesErik Schnetter2013-07-021-1/+8
* Implement frexpErik Schnetter2013-06-281-0/+1
* Rename implementation files to indicate vector sizeErik Schnetter2013-06-171-0/+646
OpenPOWER on IntegriCloud