Hal Finkel 4a1535c038 Make the PPC fast-math sqrt expansion safe at 0 ... In fast-math mode sqrt(x) is calculated using the fast expansion of the reciprocal of the reciprocal sqrt expansion. The reciprocal and reciprocal sqrt expansions use the associated estimate instructions along with some Newton iterations. Unfortunately, as a result, sqrt(0) was being calculated as NaN, which is not correct. Now we explicitly return a result of zero if the input is zero. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@190624 91177308-0d34-0410-b5e6-96231b3b80d8		2013-09-12 19:04:12 +00:00
..
AArch64	Implement aarch64 neon instruction set AdvSIMD (3V Diff), covering the following 26 instructions,	2013-09-09 02:20:27 +00:00
ARM	[ARMv8] Prevent generation of deprecated IT blocks on ARMv8 in Thumb mode.	2013-09-09 14:21:49 +00:00
CPP
Generic
Hexagon	Debug Info Testing: use null instead of an empty string in context field.	2013-09-09 00:12:17 +00:00
Inputs	Debug Info: add an identifier field to DICompositeType.	2013-08-26 22:39:55 +00:00
Mips	[mips][msa] Added test cases that were supposed to be part of r190507, r190509, r190512, and r190518.	2013-09-11 12:39:25 +00:00
MSP430
NVPTX
PowerPC	Make the PPC fast-math sqrt expansion safe at 0	2013-09-12 19:04:12 +00:00
R600	R600/SI: expose TBUFFER_STORE_FORMAT_* for OpenGL transform feedback	2013-09-12 02:55:14 +00:00
SPARC	[Sparc] Correctly handle call to functions with ReturnsTwice attribute.	2013-09-05 05:32:16 +00:00
SystemZ	[SystemZ] Add TM and TMY	2013-09-10 10:20:32 +00:00
Thumb	ARM: Use "dmb sy" for barriers on M-class CPUs	2013-08-28 14:39:19 +00:00
Thumb2	[ARMv8] Prevent generation of deprecated IT blocks on ARMv8 in Thumb mode.	2013-09-09 14:21:49 +00:00
X86	AVX-512: implemented extractelement with variable index.	2013-09-12 08:55:00 +00:00
XCore	XCore handling of thread local lowering	2013-09-09 10:42:11 +00:00