[PATCH] Force DFP function args into FP registers per the ABI

David Edelsohn dje@watson.ibm.com
Fri Mar 2 13:52:00 GMT 2007


	The patch generally looks okay.  Too bad these new modes need to
diffuse into everything.

+	  /* Must use an even/odd register pair.  */

Would you please clarify this with "Decimal Float must..." or "_Decimal128
must..."  in both places?

+    case DDmode:
     case DFmode:
     case SFmode:
       if (CONSTANT_P (operands[1])

Please list DDmode after DFmode, so the ordering is consistent
everywhere.  When browsing these functions for modes, it's confusing to
have the decimal modes jump around.

+;; Copyright (C) 2006

The file is being added to the FSF in 2007, so the copyright should list
2007. 



Okay with those changes.

Thanks, David



More information about the Gcc-patches mailing list