Mark Heffernan 030525f5bf Use 32-bit divides instead of 64-bit divides where possible. ... For NVPTX, try to use 32-bit division instead of 64-bit division when the dividend and divisor fit in 32 bits. This speeds up some internal benchmarks significantly. The underlying reason is that many index computations are carried out in 64-bits but never actually exceed the capacity of a 32-bit word. llvm-svn: 244684		2015-08-11 22:16:34 +00:00
..
AArch64	[GlobalMerge] Use private linkage for MergedGlobals variables	2015-08-11 15:48:04 +00:00
AMDGPU	AMDGPU: Add pass to lower OpenCL image and sampler arguments.	2015-08-07 23:19:30 +00:00
ARM	[GlobalMerge] Use private linkage for MergedGlobals variables	2015-08-11 15:48:04 +00:00
BPF
CPP
Generic	Update test suite to make "ninja check" succeed without native backend builtin	2015-08-04 06:32:54 +00:00
Hexagon	DI: Disallow uniquable DICompileUnits	2015-08-03 17:26:41 +00:00
Inputs	DI: Disallow uniquable DICompileUnits	2015-08-03 17:26:41 +00:00
Mips	Revert r229675 - [mips] Avoid redundant sign extension of the result of binary bitwise instructions.	2015-08-04 14:26:35 +00:00
MIR	MIR Serialization: Serialize UsedPhysRegMask from the machine register info.	2015-08-11 00:32:49 +00:00
MSP430
NVPTX	Use 32-bit divides instead of 64-bit divides where possible.	2015-08-11 22:16:34 +00:00
PowerPC	StackMap: FastISel: Add an appropriate number of immediate operands to the	2015-08-10 21:27:03 +00:00
SPARC	[Sparc] Implement i64 load/store support for 32-bit sparc.	2015-08-10 19:11:39 +00:00
SystemZ
Thumb	DI: Disallow uniquable DICompileUnits	2015-08-03 17:26:41 +00:00
Thumb2
WebAssembly	WebAssembly: implement comparison.	2015-08-11 21:02:46 +00:00
WinEH
X86	[x86] enable machine combiner reassociations for 128-bit vector single/double multiplies	2015-08-11 20:19:23 +00:00
XCore	DI: Disallow uniquable DICompileUnits	2015-08-03 17:26:41 +00:00