Re: [PATCH v3] AArch64: Fix copysign patterns

2024-10-23 Thread Richard Sandiford
Wilco Dijkstra writes: > The current copysign pattern has a mismatch in the predicates and constraints > - > operand[2] is a register_operand but also has an alternative X which allows > any > operand. Since it is a floating point operation, having an integer > alternative > makes no sense. C

[PATCH v3] AArch64: Fix copysign patterns

2024-10-17 Thread Wilco Dijkstra
The current copysign pattern has a mismatch in the predicates and constraints - operand[2] is a register_operand but also has an alternative X which allows any operand. Since it is a floating point operation, having an integer alternative makes no sense. Change the expander to always use vector i