Justin Lebar b12cedb551 [NVPTX] Implement NVPTXTargetLowering::getSqrtEstimate. ... Summary: This lets us lower to sqrt.approx and rsqrt.approx under more circumstances. * Now we emit sqrt.approx and rsqrt.approx for calls to @llvm.sqrt.f32, when fast-math is enabled. Previously, we only would emit it for calls to @llvm.nvvm.sqrt.f. (With this patch we no longer emit sqrt.approx for calls to @llvm.nvvm.sqrt.f; we rely on intcombine to simplify llvm.nvvm.sqrt.f into llvm.sqrt.f32.) * Now we emit the ftz version of rsqrt.approx when ftz is enabled. Previously, we only emitted rsqrt.approx when ftz was disabled. Reviewers: hfinkel Subscribers: llvm-commits, tra, jholewinski Differential Revision: https://reviews.llvm.org/D28508 llvm-svn: 293605		2017-01-31 05:58:22 +00:00
..
AArch64	GlobalISel: correctly translate invoke when callee is a register.	2017-01-30 21:45:21 +00:00
AMDGPU	AMDGPU: Generalize matching of v_med3_f32	2017-01-31 03:07:46 +00:00
ARM	DAG: Constant fold fp16_to_fp/fp16_to_fp	2017-01-30 16:57:41 +00:00
AVR
BPF	Revert "In visitSTORE, always use FindBetterChain, rather than only when UseAA is enabled."	2017-01-26 16:46:13 +00:00
Generic	Revert "Do not verify dominator tree if it has no roots"	2017-01-25 17:15:48 +00:00
Hexagon	[Hexagon] Add Hexagon-specific loop idiom recognition pass	2017-01-26 21:41:10 +00:00
Inputs
Lanai
Mips	[mips] Recommit: "N64 static relocation model support"	2017-01-27 11:36:52 +00:00
MIR
MSP430	Revert "In visitSTORE, always use FindBetterChain, rather than only when UseAA is enabled."	2017-01-26 16:46:13 +00:00
NVPTX	[NVPTX] Implement NVPTXTargetLowering::getSqrtEstimate.	2017-01-31 05:58:22 +00:00
PowerPC	[PPC] cleanup of mayLoad/mayStore flags and memory operands.	2017-01-26 18:59:15 +00:00
SPARC
SystemZ	SDAG: Update ChainNodesMatched during UpdateChains if a node is replaced	2017-01-30 18:29:46 +00:00
Thumb	Revert "In visitSTORE, always use FindBetterChain, rather than only when UseAA is enabled."	2017-01-26 16:46:13 +00:00
Thumb2
WebAssembly
WinEH
X86	[X86] Update the broadcast fallback patterns to use shuffle instructions from the appropriate execution domain.	2017-01-31 05:18:29 +00:00
XCore	DAG: Fold fneg into compare with constant into the constant	2017-01-30 17:57:28 +00:00