1
0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-10-25 05:52:53 +02:00
llvm-mirror/test/CodeGen
Tom Stellard c738ae1559 AMDGPU/SI: Fix threshold calculation for branching when exec is zero
Summary:
When control flow is implemented using the exec mask, the compiler will
insert branch instructions to skip over the masked section when exec is
zero if the section contains more than a certain number of instructions.

The previous code would only count instructions in successor blocks,
and this patch modifies the code to start counting instructions in all
blocks between the start and end of the branch.

Reviewers: nhaehnle, arsenm

Subscribers: arsenm, llvm-commits

Differential Revision: http://reviews.llvm.org/D18282

llvm-svn: 263969
2016-03-21 18:56:58 +00:00
..
AArch64 [CXX_FAST_TLS] Disable tail call when calling conventions are mismatched. 2016-03-18 23:41:51 +00:00
AMDGPU AMDGPU/SI: Fix threshold calculation for branching when exec is zero 2016-03-21 18:56:58 +00:00
ARM ARM: Better codegen for 64-bit compares. 2016-03-21 18:00:02 +00:00
BPF BPF: emit an error message for unsupported signed division operation 2016-03-18 22:02:47 +00:00
CPP
Generic
Hexagon
Inputs
Mips [mips] MIPS32R6 compact branch support 2016-03-14 16:24:05 +00:00
MIR [MIR] Add a test case for the diagnostic of a wrongly typed generic instruction 2016-03-15 18:31:29 +00:00
MSP430
NVPTX [NVPTX] Adds a new address space inference pass. 2016-03-20 20:59:20 +00:00
PowerPC [PPC, FastISel] Fix ordered/unordered fcmp 2016-03-17 22:27:58 +00:00
SPARC
SystemZ [SystemZ] Avoid LER on z13 due to partial register dependencies 2016-03-14 13:50:03 +00:00
Thumb
Thumb2
WebAssembly [WebAssembly] Stackify code emitted by eliminateFrameIndex and SP writeback 2016-03-17 17:00:29 +00:00
WinEH
X86 Fixed -mcpu flag 2016-03-21 11:06:20 +00:00
XCore