llvm-project
47bca232 - [X86] Manage atomic store of fp -> int promotion in DAG

Commit

2 days ago

[X86] Manage atomic store of fp -> int promotion in DAG When lowering atomic <1 x T> vector types with floats, selection can fail since this pattern is unsupported. To support this, floats can be casted to an integer type of the same size.

References

users/jofrn/x86-promote-atomic-store-fp

#197166 - [X86] Manage atomic store of fp -> int promotion in DAG

#197618 - [SelectionDAG] Widen <2 x T> vector types for atomic store

#197619 - [X86] Remove extra MOV after widening atomic store

#197860 - [SelectionDAG] Split vector types for atomic store

#197861 - [X86] Cast atomic store in IR to support floats

#197862 - [AtomicExpand] Add bitcasts when expanding store atomic vector

Author

jofrn

Committer

jofrn

Parents

3e49a567

llvm-project 47bca232 - [X86] Manage atomic store of fp -> int promotion in DAG

llvm-project
47bca232 - [X86] Manage atomic store of fp -> int promotion in DAG