Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!iuvax!rutgers!psuvax1!schwartz@shire.cs.psu.edu From: schwartz@shire.cs.psu.edu (Scott Schwartz) Newsgroups: comp.arch Subject: risc object sizes; emperical results (was Re: When is RISC not RISC?) Summary: gcc vs sparc Message-ID: <4280@psuvax1.cs.psu.edu> Date: 9 Feb 89 23:18:12 GMT References: <747@atanasoff.cs.iastate.edu> <6310013@hpcupt1.HP.COM> <21606@ames.arc.nasa.gov> Sender: news@psuvax1.cs.psu.edu Reply-To: schwartz@shire.cs.psu.edu (Scott Schwartz) Organization: Pennsylvania State University, Computer Science Lines: 14 In-reply-to: lamaster@ames.arc.nasa.gov (Hugh LaMaster) In article <21606@ames.arc.nasa.gov>, lamaster@ames (Hugh LaMaster) writes: >I notice that, recently, some results from the GNU compiler on some RISC's >has been a challenge to the object code size of VAX code. Obviously, GNU >is doing something interesting there. Does anyone know why the code from >the GNU C compiler appears to be significantly more compact than other >compilers on the RISC machines? Which RISCs did you notice this on? My experience with gcc-1.33 on a SPARC (sun4) is that Sun's compiler generates smaller and faster code. A brief look at the assembler output shows that gcc generates lots of nops where Sun's compiler generates annuled branches, although it doesn't do this in every case. -- Scott Schwartz