Displaying 1 result from an estimated 1 matches for "foldphiargloadintophi".
2020 May 05
2
Missing vectorization of loop due to load late in the loop
...e
returns false.
If we just move the load before the definition of %0 in the vectorizer
input, then we instead get
LV: We can vectorize this loop!
Originally the loop had two loads, and then one of them was actually
placed early in the loop block. That was done by instcombine, by
InstCombiner::FoldPHIArgLoadIntoPHI.
The reason this doesn't happen for the load we see in the reduced
example is because after loop rotation licm hoists the start value of
the PHI, so when instcombine tries to do FoldPHIArgLoadIntoPHI, the
start value isn't placed in the direct predecessor block of the PHI,
and the folding...