1. d1498ed [CostModel][X86] Fix overcounting arithmetic cost in illegal types in getArithmeticReductionCost/getMinMaxReductionCost by Craig Topper · 7 years ago
  2. 381b4fb [X86] Remove -costmodel-reduxcost=true from the experimental vector reduction intrinsic tests as it appears to be unnecessary. NFC by Craig Topper · 7 years ago
  3. b4719e5 [X86] Add more cost model tests for vector reductions with narrow vector types. NFC by Craig Topper · 7 years ago
  4. 102854f [TTI] Reduction costs only need to include a single extract element cost (REAPPLIED) by Simon Pilgrim · 7 years ago
  5. 8cd9d1b Revert "[TTI] Reduction costs only need to include a single extract element cost" by Fedor Sergeev · 7 years ago
  6. 924f193 [TTI] Reduction costs only need to include a single extract element cost by Simon Pilgrim · 7 years ago
  7. fc8f1d7 [CostModel][X86] SK_ExtractSubvector is free if the subvector is at the start of the source vector by Simon Pilgrim · 7 years ago
  8. 44a9a71 [TTI] Fix uses of SK_ExtractSubvector shuffle costs (PR39368) by Simon Pilgrim · 7 years ago
  9. e612ab0 [CostModel][X86] Add integer vector reduction cost tests by Simon Pilgrim · 7 years ago