Jingyue Wu 75d77cd179 [MachineSink] Use the real post dominator tree ... Summary: Fixes a FIXME in MachineSinking. Instead of using the simple heuristics in isPostDominatedBy, use the real MachinePostDominatorTree and MachineLoopInfo. The old heuristics caused instructions to sink unnecessarily, and might create register pressure. This is the second try of the fix. The first one (D4814) caused a performance regression due to failing to sink instructions out of loops (PR21115). This patch fixes PR21115 by sinking an instruction from a deeper loop to a shallower one regardless of whether the target block post-dominates the source. Thanks Alexey Volkov for reporting PR21115! Test Plan: Added a NVPTX codegen test to verify that our change prevents the backend from over-sinking. It also shows the unnecessary register pressure caused by over-sinking. Added an X86 test to verify we can sink instructions out of loops regardless of the dominance relationship. This test is reduced from Alexey's test in PR21115. Updated an affected test in X86. Also ran SPEC CINT2006 and llvm-test-suite for compilation time and runtime performance. Results are attached separately in the review thread. Reviewers: Jiangning, resistor, hfinkel Reviewed By: hfinkel Subscribers: hfinkel, bruno, volkalexey, llvm-commits, meheff, eliben, jholewinski Differential Revision: http://reviews.llvm.org/D5633 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@219773 91177308-0d34-0410-b5e6-96231b3b80d8		2014-10-15 03:27:43 +00:00
..
Analysis	[CFL-AA] CFL-AA should not assert on an va_arg instruction	2014-10-14 20:51:26 +00:00
Assembler	Revert "Revert "DI: Fold constant arguments into a single MDString""	2014-10-03 20:01:09 +00:00
Bindings	Revert "Revert "DI: Fold constant arguments into a single MDString""	2014-10-03 20:01:09 +00:00
Bitcode	Revert "Revert "DI: Fold constant arguments into a single MDString""	2014-10-03 20:01:09 +00:00
BugPoint	Revert "Revert "DI: Fold constant arguments into a single MDString""	2014-10-03 20:01:09 +00:00
CodeGen	[MachineSink] Use the real post dominator tree	2014-10-15 03:27:43 +00:00
DebugInfo	[dwarfdump] Prettyprint DW_AT_APPLE_property_attribute bitfield values.	2014-10-10 15:51:10 +00:00
ExecutionEngine	[mips] Remove XFAIL from two XPASS'ing tests on the llvm-mips-linux builder	2014-10-03 08:49:44 +00:00
Feature	Revert "Revert "DI: Fold constant arguments into a single MDString""	2014-10-03 20:01:09 +00:00
FileCheck
Instrumentation	[asan-asm-instrumentation] Fixed memory references which includes %rsp as a base or an index register.	2014-10-13 09:37:47 +00:00
Integer
JitListener	Revert "Revert "DI: Fold constant arguments into a single MDString""	2014-10-03 20:01:09 +00:00
Linker	[dwarfdump] Print the name for referenced specification of abstract_origin DIEs.	2014-10-06 03:36:31 +00:00
LTO	LTO: Add missing target triple from r218784	2014-10-01 18:49:58 +00:00
MC	MC, COFF: Make bigobj test compatible with python3	2014-10-14 22:35:11 +00:00
Object	Suppress llvm-ar's MRI tests for now on win32, since line_iterator is incompatible to CRLF.	2014-10-11 22:21:27 +00:00
Other	[lit] Parse all strings as UTF-8 rather than ASCII.	2014-09-12 16:46:05 +00:00
TableGen	[AVX512] Added intrinsics for VPCMPEQB and VPCMPEQW.	2014-09-30 11:32:22 +00:00
tools	[llvm-objdump] Update error message and add test case for mach-o file with bad library ordinals	2014-10-14 23:29:38 +00:00
Transforms	[LoopVectorize] Ignore @llvm.assume for cost estimates and legality	2014-10-14 22:59:49 +00:00
Unit
Verifier
YAMLParser
.clang-format
CMakeLists.txt
lit.cfg	llvm/test/lit.cfg: Suppress dwarf stuff for targeting x86_64-mingw32 while investigating since r219108.	2014-10-06 23:29:06 +00:00
lit.site.cfg.in
Makefile
Makefile.tests
TestRunner.sh