tanner@cdis-1.compu.com (Dr. T. Andrews) (06/02/90)
I'll admit that I've heard that AT&T has shown limited interest in compiler bug reports. From where I sit, microsoft seems to be in a similar position. Working through our vendor, we've turned in rather a lot of bug reports on their compiler. Mostly they get ignored. Does anyone know if any of the following are fixed with SCO UNIX? Some of them have been in the microsoft compiler (reported but not fixed) for considerable time. Rumor has it that SCO is planning to phase out xenix over time, and I don't know what they'll be using for a compiler with unix. For that matter, does anyone know if they have put any effort at all into the lint? 0380101 15-Jan-87 code-gen bug (register union doesn't work) (a picky one: declare a union of pointers to save a register. compiler does not put it in registers as instructed. ref #0380101) r050120 16-Jan-87 _doprnt() fails to work, example "pfc154.c" (if you use "%.*s" or "%.ns" for n being a digit, and your string runs to the end of memory, it dumps core, even if your limit is OK. this is a run-time problem. fixed on `386, ref #r050120) 09-Dec-88 "lint" blows ternary operator type (pfc234.c) (for expression A ? B : C, "lint" thinks that the type is that of A.) 13-Mar-89 dos rename args backwards (pfc243.c) (the arguments to the rename() function must be reversed or you will get unexpected results.) s199287 11-Aug-89 that loop-opt bug & friends (pfc253.c) (there's a rather nasty bug in the optimizer. thanks, microsoft; I blew a lot of time thinking that there was a bug in MY code! ref #s199287) 58-1012 30-Nov-89 cc-386 bungles "extern const" (pfc262{a,b}.c) ("extern const char cd_txt[]" may generate garbage in object file. ref #58-1012) 58-1160 25-Jan-90 ming's "cc -Oa" bug (pfc267.c) (there's a bug in the alias detection mechanism. ref 58-1160) T245195 08-Feb-90 compiler core dump on unknown symbol (pfc266.c) (under certain circumstances, the compiler dumps core. ref #TS-245195) S256693 26-Mar-90 "maxint" error (pfc269.c) (for 16-bit machines, certain comparisons are done wrong. ref S-256693) -- uflorida!ki4pv!cdis-1!tanner {uunet dsinc}!cdin-1!cdis-1!tanner