llvm-6502/test
Owen Anderson 3042a65e5f Remove a very old instcombine where we would turn sequences of selects into
logical operations on the i1's driving them.  This is a bad idea for every
target I can think of (confirmed with micro tests on all of: x86-64, ARM,
AArch64, Mips, and PowerPC) because it forces the i1 to be materialized into
a general purpose register, whereas consuming it directly into a select generally
allows it to exist only transiently in a predicate or flags register.

Chandler ran a set of performance tests with this change, and reported no
measurable change on x86-64.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201275 91177308-0d34-0410-b5e6-96231b3b80d8
2014-02-12 23:54:07 +00:00
..
Analysis [Vectorizer] Add a new 'OperandValueKind' in TargetTransformInfo called 2014-02-12 23:43:47 +00:00
Assembler
Bindings
Bitcode
BugPoint
CodeGen [X86] Teach the backend how to lower vector shift left into multiply rather than scalarizing it. 2014-02-12 23:42:28 +00:00
DebugInfo DebugInfo: Demonstrate that we're not currently uniquing address table entries in fission 2014-02-12 23:03:54 +00:00
ExecutionEngine PC-rel implemented in AArch64, test now pass 2014-02-12 17:17:41 +00:00
Feature
FileCheck
Instrumentation [asan] support for FreeBSD, LLVM part. patch by Viktor Kutuzov 2014-02-10 07:37:04 +00:00
Integer
JitListener
Linker
LTO
MC Tidy up a bit. Formatting only. 2014-02-11 20:48:41 +00:00
Object
Other
TableGen
tools
Transforms Remove a very old instcombine where we would turn sequences of selects into 2014-02-12 23:54:07 +00:00
Unit
Verifier
YAMLParser
.clang-format
CMakeLists.txt
lit.cfg
lit.site.cfg.in
Makefile
Makefile.tests
TestRunner.sh