Displaying 1 result from an estimated 1 matches for "shaddxh_ra".
2013 Mar 26
0
[LLVMdev] Is it correct for a glue operand to be replaced with a chain?
...once it passes to scheduling, the glue
no longer exists and the nodes are just chained, but twice.
ISel output:
ISEL: Starting pattern match on root node: 0x2c689c0: i32,ch = SHADDXH
0x2c688c0:2, 0x2c686c0, 0x2c688c0:1 [ID=13]
Initial Opcode index to 6962
Morphed node: 0x2c689c0: i32,ch = SHADDXH_RA 0x2c66500, 0x2c688c0:2,
0x2c688c0:1
Final schedule output:
SU(4): 0x2c689c0: i32,ch = SHADDXH_RA 0x2c66500, 0x2c688c0:1, 0x2c688c0:1
[ID=4]
Is this correct behavior? Because InstrEmitter then complains that the nodes
do not match the tablegen file, which I would say is correct. I'm letting...