Returns the register used as static base in RWPI variants.

Return the number of bits overwritten by a store of the specified value type. If the value type is a scalable vector type, the scalable property will

If Op is a strict floating-point operation, return the result of getOperationAction for the equivalent non-strict operation.

For example, SGE -> SGT, SLE -> SLT, ULE -> ULT, UGE -> UGT. Returns the strict version of non-strict comparisons.

Grab the string out of the object.

Get the offset of string \p S in the string table. This can insert a new element or return the offset of a pre-existing one.

Returns the strongest permitted ordering on failure, given the desired ordering on success. If the comparison in a cmpxchg operation fails, there is

Retrieve the sret type.

Returns the largest legal sub-class of RC that supports the sub-register index Idx. If no such sub-class exists, return NULL. If all registers in RC a

Returns the physical register number of sub-register "Index" for physical register RegNo. Return zero if the sub-register does not exist.

Returns the current sub-register index.

Return a bitmask representing the parts of a register that are covered by SubIdx \see LaneBitmask. SubIdx == 0 is allowed, it has the lane mask ~0u.

This static method returns a VectorType with a smaller number of elements of a larger type than the input element type. For example, a <16 x i8> subdi

Virtual method implemented by subclasses that returns a reference to that target's TargetSubtargetInfo-derived member variable.

TargetSubtargetInfo getter.

Get the ID of the subtree the given DAG node belongs to. For convenience, if DFSResults have not been computed yet, give everything tree ID 0.

Get the connection level of a subtree. For bottom-up trees, the connection level is the latency depth (in cycles) of the deepest connection to anothe

Get the node cutoff before subtrees are considered significant.

Returns the success ordering constraint of this cmpxchg instruction.

Return the atomic ordering requirements for this memory operation. For cmpxchg atomic operations, return the atomic ordering requirements when store o

Returns successor index for current case successor.

This is used to interface between code that wants to operate on terminator instructions directly.

Returns a NULL-terminated list of super-classes. The classes are ordered by ID which is also a topological ordering from large to small classes. The

Returns the synchronization scope ID of this load instruction.

Returns the synchronization scope ID of this store instruction.

Returns the synchronization scope ID of this fence instruction.

Returns the synchronization scope ID of this cmpxchg instruction.

Returns the synchronization scope ID of this rmw instruction.

Returns the synchronization scope ID for this memory operation.

Returns the synchronization scope ID for this memory operation.

Local query method delegates up to T which *must* implement this!

Getter for generic TargetLowering class.

Getter for generic TargetLowering class.

Accessors. @{

Returns the target-specific default value for tail duplication. This value will be used if the tail-dup-placement-threshold argument is not provided.