> -----Original Message----- > From: Aldy Hernandez [mailto:al...@redhat.com] > Sent: Thursday, December 12, 2013 10:26 AM > To: Iyer, Balaji V > Cc: Jakub Jelinek; 'gcc-patches@gcc.gnu.org' > Subject: Re: [PING]: [GOMP4] [PATCH] SIMD-Enabled Functions (formerly > Elemental functions) for C > > On 12/11/13 10:44, Iyer, Balaji V wrote: > > > Just out of curiosity, why can't I keep it as-is? It is giving the > > correct output/behavior and doesn't seem to interfere with anything > > else. The only extra thing I am doing is to add an extra if-statement > > while recursing through all the functions to check for cilk simd > > function and then calling the function to handle it, which the OMP > > will have to do anyway. The only extra thing I added was an extra > > if-statement. > > If Cilk clones are tagged differently then we need to special case Cilk clones > every where we handle OMP clones. If they share an attribute, no special > handling is needed.
Will it be Ok if I don’t mark them as cilk simd function but just keep it as omp declare simd from the start? That should get around this issue. Thanks, Balaji V. Iyer. > > Aldy
