The TypePromotion pass only operates on scalar types so I've fixed up
all places where we were relying upon the implicit cast from
TypeSize->uint64_t.
Details
Details
Diff Detail
Diff Detail
Event Timeline
llvm/lib/CodeGen/TypePromotion.cpp | ||
---|---|---|
985 | Any reason to prefer getScalarSizeInBits() over PromotedVT.getSizeInBits().getFixedSize()? If it's just the number of characters, should we have a "getFixedSizeInBits()`? |
Any reason to prefer getScalarSizeInBits() over PromotedVT.getSizeInBits().getFixedSize()? If it's just the number of characters, should we have a "getFixedSizeInBits()`?