search for: r251049

...[GlobalsAA] Disable globals-aa by default. -1% 7. r248638: [SCEV] Reapply 'Teach isLoopBackedgeGuardedByCond to exploit trip counts'. +2% 8. r249802: [SCEV] Call `StrengthenNoWrapFlags` after `GroupByComplexity`; NFCI. +4% 9. r250157: [GlobalsAA] Turn GlobalsAA on again by default. +1% 10. r251049: [SCEV] Mark AddExprs as nsw or nuw if legal. +23% 11. No data 12. r259252: AttributeSetImpl: Summarize existing function attributes in a bitset. -1% r259256: Add LoopSimplifyCFG pass. -2% 13. r262250: Enable LoopLoadElimination by default. +3% 14. r262839: Revert "Enable LoopLoadEliminati...

...efault. -1% >> 7. r248638: [SCEV] Reapply 'Teach isLoopBackedgeGuardedByCond to exploit trip counts'. +2% >> 8. r249802: [SCEV] Call `StrengthenNoWrapFlags` after `GroupByComplexity`; NFCI. +4% >> 9. r250157: [GlobalsAA] Turn GlobalsAA on again by default. +1% >> 10. r251049: [SCEV] Mark AddExprs as nsw or nuw if legal. +23% >> 11. No data >> 12. r259252: AttributeSetImpl: Summarize existing function attributes in a bitset. -1% >> r259256: Add LoopSimplifyCFG pass. -2% >> 13. r262250: Enable LoopLoadElimination by default. +3% >> 14. r...

...t; > 7. r248638: [SCEV] Reapply 'Teach isLoopBackedgeGuardedByCond to exploit > trip counts'. +2% > > 8. r249802: [SCEV] Call `StrengthenNoWrapFlags` after > `GroupByComplexity`; NFCI. +4% > > 9. r250157: [GlobalsAA] Turn GlobalsAA on again by default. +1% > > 10. r251049: [SCEV] Mark AddExprs as nsw or nuw if legal. +23% > > 11. No data > > 12. r259252: AttributeSetImpl: Summarize existing function attributes in > a bitset. -1% > > r259256: Add LoopSimplifyCFG pass. -2% > > 13. r262250: Enable LoopLoadElimination by default. +3% > &...

...4 - changes in Inlining cost model You probably can try reverting them locally to check if my hypothesis is correct. I also looked at earlier data, and on top of the before-mentioned issues I found the following causes of compile-time degradations (and sometimes improvements): 1) Changes in SCEV (r251049, r248637). 2) Changes in LoopUnrolling cost-model: adding a new unroll analyzer, changing Os thresholds, etc. 3) Adding new passes (e.g. LoopLoadElimination). Not everything is recoverable, but there are still some opportunities laying here. For instance, I have some ideas on how to mitigate the c...

First of all, sorry for the long mail. Inspired by the excellent analysis Rui did for lld, I decided to do the same for llvm. I'm personally very interested in build-time for LTO configuration, with particular attention to the time spent in the optimizer. Rafael did something similar back in March, so this can be considered as an update. This tries to include a more accurate high-level