vecmathlib - vecmathlib forked from https://bitbucket.org/eschnett/vecmathlib for POWER8 improvements

	Commit message (Collapse)	Author	Age	Files	Lines
...
*	Optimize atan and acos	Erik Schnetter	2013-06-20	1	-10/+58
\|
*	Do not benchmark multi-element test vectors (as they are very slow)	Erik Schnetter	2013-06-20	1	-5/+5
\|
*	Choose required accuracy depending on FP_CONTRACT	Erik Schnetter	2013-06-20	1	-1/+6
\|
*	Improve efficiency of round	Erik Schnetter	2013-06-20	1	-1/+2
\|
*	Describe in comment how NEON implements rcp	Erik Schnetter	2013-06-20	1	-0/+2
\|
*	Reduce accuracy of sin when no FP_CONTRACT is required	Erik Schnetter	2013-06-20	1	-0/+19
\|
*	Implement efficient rsqrt. Implement sqrt in terms of rsqrt.	Erik Schnetter	2013-06-20	1	-19/+40
\|
*	Explain what VML_HAVE_FP_CONTRACT means	Erik Schnetter	2013-06-20	1	-1/+7
\|
*	Correct case in comment	Erik Schnetter	2013-06-19	1	-1/+1
\|
*	Test floating point rounding	Erik Schnetter	2013-06-18	1	-2/+11
\|
*	Test barrier function	Erik Schnetter	2013-06-18	1	-0/+6
\|
*	Implement barrier for vec_pseudo	Erik Schnetter	2013-06-18	1	-0/+27
\|
*	Offer VML_BROKEN_STL to decide whether the STL has C++11 features	Erik Schnetter	2013-06-17	2	-4/+4
\|
*	Support Intel compiler	Erik Schnetter	2013-06-17	1	-1/+7
\|
*	Explain why vec_builtin doesn't work	Erik Schnetter	2013-06-17	1	-0/+1
\|
*	Introduce VML_HAVE_FP_CONTRACT (still unused)	Erik Schnetter	2013-06-17	1	-0/+1
\|
*	Add missing std:: namespace qualifier	Erik Schnetter	2013-06-17	2	-4/+4
\|
*	Correct NEON implementation	Erik Schnetter	2013-06-17	2	-12/+26
\|
*	Do not use constructor forwarding	Erik Schnetter	2013-06-17	1	-1/+8
\| \| \| \|	The Intel compiler does not support it.
*	Correct syntax error in include guard	Erik Schnetter	2013-06-17	1	-2/+2
\|
*	Test iota() function	Erik Schnetter	2013-06-17	1	-0/+2
\|
*	Support NEON float32x4_t	Erik Schnetter	2013-06-17	2	-1/+559
\|
*	Rename implementation files to indicate vector size	Erik Schnetter	2013-06-17	13	-49/+52
\|
*	Automatically use a "good" vector type for loop code quality tests	Erik Schnetter	2013-06-13	1	-53/+65
\|
*	Implement fall-back timer if no architecture-specific high-resolution timer ↵	Erik Schnetter	2013-06-13	1	-7/+16
\| \| \| \|	is available
*	Beautify screen output	Erik Schnetter	2013-06-13	1	-3/+3
\|
*	Allow wrap-around when testing fmod and remainder	Erik Schnetter	2013-06-13	1	-5/+5
\|
*	Benchmark NEON as well	Erik Schnetter	2013-06-12	1	-0/+6
\|
*	Correct fma	Erik Schnetter	2013-06-12	1	-1/+5
\|
*	Correct NEON barrier	Erik Schnetter	2013-06-12	1	-1/+1
\|
*	Correct generic barrier on ARM	Erik Schnetter	2013-06-12	1	-1/+1
\|
*	Test NEON as well (first float2 architecture)	Erik Schnetter	2013-06-12	1	-0/+9
\|
*	Correct operator&	Erik Schnetter	2013-06-12	1	-1/+1
\|
*	Correct last (?) syntax error	Erik Schnetter	2013-06-12	1	-1/+1
\|
*	Correct more errors	Erik Schnetter	2013-06-12	1	-5/+12
\|
*	Correct various errors	Erik Schnetter	2013-06-12	1	-26/+36
\|
*	Correct NEON detection	Erik Schnetter	2013-06-12	1	-1/+1
\|
*	Implement barrier for ARM	Erik Schnetter	2013-06-12	1	-0/+2
\|
*	Auto-detect support for unaligned memory access	Erik Schnetter	2013-06-12	1	-0/+8
\|
*	Begin to implement ARM NEON	Erik Schnetter	2013-06-12	2	-0/+534
\|
*	Suggest some additional vector architectures	Erik Schnetter	2013-06-12	1	-1/+4
\|
*	Use #if instead of comments to deactive non-C++11 fallback code	Erik Schnetter	2013-06-12	1	-67/+75
\|
*	Some QPX corrections/optimizations	Erik Schnetter	2013-06-12	1	-3/+7
\|
*	Declare some functions without auto keyword	Erik Schnetter	2013-06-12	3	-5/+5
\|
*	Add more tests for loop code quality	Erik Schnetter	2013-06-12	1	-47/+66
\|
*	Optimize AVX all/any functions	Erik Schnetter	2013-06-12	4	-22/+20
\|
*	Use integer signbit instead of integer comparisons in mask class	Erik Schnetter	2013-06-12	1	-11/+25
\|
*	Implement signbit function for integer vectors	Erik Schnetter	2013-06-12	13	-0/+76
\|
*	Test code quality for loops	Erik Schnetter	2013-06-11	1	-0/+126
\|
*	Implement nextafter	Erik Schnetter	2013-06-09	15	-1/+59
\|