pjh@mccc.edu (Peter J. Holsberg) (06/28/91)
Chalk up another for AT&T's ANSI C compiler! I recompiled s5last with the K&R compiler distributed with Microport SVR3 and it works perfectly! Welllll, *almost* perfectly -- it still doesn't exclude things in the exclude list. Also, CI 5.0 produced a 46,452 byte executable, while uP produced a 28,830 byte one! Thanks to those who emailed suggestions. Pete -- Prof. Peter J. Holsberg Mercer County Community College Voice: 609-586-4800 Engineering Technology, Computers and Math FAX: 609-586-6944 1200 Old Trenton Road, Trenton, NJ 08690 Internet: pjh@mccc.edu TCF 92 - ??? ??-??, 1992
gwyn@smoke.brl.mil (Doug Gwyn) (06/29/91)
In article <1991Jun28.005902.13061@mccc.edu> pjh@mccc.edu (Peter J. Holsberg) writes: >Chalk up another for AT&T's ANSI C compiler! I recompiled s5last with >the K&R compiler distributed with Microport SVR3 and it works perfectly! It should be noted that such symptoms do not mean that the compiler had a problem, merely that it generated code differently. >Welllll, *almost* perfectly -- it still doesn't exclude things in the >exclude list. See?
pjh@mccc.edu (Pete Holsberg) (06/30/91)
In article <16578@smoke.brl.mil> gwyn@smoke.brl.mil (Doug Gwyn) writes: =In article <1991Jun28.005902.13061@mccc.edu> pjh@mccc.edu (Peter J. Holsberg) writes: =>Chalk up another for AT&T's ANSI C compiler! I recompiled s5last with =>the K&R compiler distributed with Microport SVR3 and it works perfectly! = =It should be noted that such symptoms do not mean that the compiler had =a problem, merely that it generated code differently. Could you explain how a compiler could generate code that would cause a bus error and a core dump, while a different compiler would generate code from the same source file that does not? I'm pretty ignorant about the different codes that compilers produce. =>Welllll, *almost* perfectly -- it still doesn't exclude things in the =>exclude list. = =See? And that's a puzzler, because one of the people I thanked for his help compiled it with the ISC compiler and the exclude list works. I don't understand how legitimate C code doesn't behave under this compiler or that unless the compilers are buggy. Thanks, Pete -- Prof. Peter J. Holsberg Mercer County Community College Voice: 609-586-4800 Engineering Technology, Computers and Math FAX: 609-586-6944 1200 Old Trenton Road, Trenton, NJ 08690 Internet: pjh@mccc.edu TCF 92 - ??? ??-??, 1992
gwyn@smoke.brl.mil (Doug Gwyn) (06/30/91)
In article <1991Jun30.010604.8185@mccc.edu> pjh@mccc.edu (Pete Holsberg) writes: >I don't understand how legitimate C code doesn't behave under this compiler >or that unless the compilers are buggy. Obviously not. Only a perfectly portable program could be used to test compilers like this. There are VERY few perfectly portable programs, even among those that were intended to be. The number of things that could go wrong are legion, but since you asked for an example, here's just one on the thousands that I've seen: int test(arg) int arg; {int ok; if(arg < 0) ok = 0; else ok = 1;} This function works on some systems using some compilers but not others. I leave it to you to figure out why (for both situations). Please don't post your guesses here!