mirror of
https://github.com/RPCS3/llvm-mirror.git
synced 2024-11-24 11:42:57 +01:00
c5e08120a4
The primary advantage is that loop optimizations will be applied in a stable order. This helps debugging and unit test creation. It is also a better overall implementation without pathologically bad performance on deep functions. On large functions (llvm-stress --size=200000 | opt -loops) Before: 0.1263s After: 0.0225s On deep functions (after tweaking llvm-stress, thanks Nadav): Before: 0.2281s After: 0.0227s See r158790 for more comments. The loop tree is now consistently generated in forward order, but loop passes are applied in reverse order over the program. If we have a loop optimization that prefers forward order, that can easily be achieved by adding a different type of LoopPassManager. llvm-svn: 159183 |
||
|---|---|---|
| .. | ||
| 2006-06-13-SingleEntryPHI.ll | ||
| 2006-06-27-DeadSwitchCase.ll | ||
| 2007-05-09-tl.ll | ||
| 2007-05-09-Unreachable.ll | ||
| 2007-07-12-ExitDomInfo.ll | ||
| 2007-07-13-DomInfo.ll | ||
| 2007-07-18-DomInfo.ll | ||
| 2007-08-01-Dom.ll | ||
| 2007-08-01-LCSSA.ll | ||
| 2007-10-04-DomFrontier.ll | ||
| 2008-06-02-DomInfo.ll | ||
| 2008-06-17-DomFrontier.ll | ||
| 2008-11-03-Invariant.ll | ||
| 2010-11-18-LCSSA.ll | ||
| 2011-06-02-CritSwitch.ll | ||
| 2011-09-26-EHCrash.ll | ||
| 2011-11-18-SimpleSwitch.ll | ||
| 2011-11-18-TwoSwitches-Threshold.ll | ||
| 2011-11-18-TwoSwitches.ll | ||
| 2012-04-02-IndirectBr.ll | ||
| 2012-04-30-LoopUnswitch-LPad-Crash.ll | ||
| 2012-05-20-Phi.ll | ||
| basictest.ll | ||
| crash.ll | ||
| infinite-loop.ll | ||
| lit.local.cfg | ||
| preserve-analyses.ll | ||