This will later be used by ThinLTOBitcodeWriter to add copies of readnone
functions to the regular LTO module.
Details
Details
Diff Detail
Diff Detail
- Repository
- rL LLVM
Event Timeline
llvm/lib/Transforms/IPO/FunctionAttrs.cpp | ||
---|---|---|
55 ↗ | (On Diff #87568) | Why don't we want to check this in the callers of computeFunctionBodyMemoryAccess? Isn't it a strong guarantee? |
llvm/lib/Transforms/IPO/FunctionAttrs.cpp | ||
---|---|---|
55 ↗ | (On Diff #87568) | Yes, we could (and that would let us simplify the caller in D29803 a little). I've uploaded a new patch that does that, and also simplifies the signature of computeFunctionBodyMemoryAccess a little by removing the SCCNodeSet, as we currently have no plans to call that function from an SCC pass. |
llvm/lib/Transforms/IPO/FunctionAttrs.cpp | ||
---|---|---|
56 ↗ | (On Diff #88253) | Document options (specifically the new ThisBody needs some comment). |