This is the mail archive of the egcs@egcs.cygnus.com mailing list for the EGCS project. See the EGCS home page for more information.
In message <19990310175357.A21098@cygnus.com>you write: > On Wed, Mar 10, 1999 at 11:58:46AM -0700, Jeffrey A Law wrote: > > > (define_insn "" > > > [(set (match_operand:DF 0 "register_operand" "=f") > > > (float:DF (match_operand:SI 1 "const_int_operand" "m")))] > > > > Not exactly. > > Actually, Jeff, why does this pattern exist? Under no conditions > should we be doing something like this. We should instead convert > the value to DFmode at compile time and drop _that_ into memory. It's historical. Very historical. I think it is actually derived from the sparc port. jeff