armin@Portia.Stanford.EDU (Ronald Mayer) (05/27/89)
Over time I've come across a number of bugs in C compilers for the IBM PC. [1 in Microsoft C 3.0, a few in MicroSoft C 4.0, and a 3 in Quick C 1.0, and 1 bug from Quick C that also causes problems in Microsoft C 5.1] These problems include crashing the compiler, wrong printf formats, and errors in evaluating things like (i>0) in while loops. Do people in this news-group care, or would posting problems like this be unnecessary clutter wasting bandwidth here. Personally, I always found the best way to deal with these problems was to call up Microsoft or send them a letter. This seems to produce faster, better and more accurate results than posting here, and it bothers far fewer people, but if people here find this kind of thing interesting, I'll post them here. -Ron Mayer armin@portia.stanford.edu
bradb@ai.toronto.edu (Brad Brown) (05/30/89)
In article <2530@Portia.Stanford.EDU> armin@Portia.Stanford.EDU (Ronald Mayer) writes: > > Over time I've come across a number of bugs in C compilers for >the IBM PC. [1 in Microsoft C 3.0, a few in MicroSoft C 4.0, and a >3 in Quick C 1.0, and 1 bug from Quick C that also causes problems >in Microsoft C 5.1] > > Do people in this news-group care, or would posting problems like >this be unnecessary clutter wasting bandwidth here. Personally I would like to see bug reports. I don't mind wading through those that don't apply to the compilers I don't use if I can run into some for the compilers that I *do* care about. Perhaps if everyone who writes a bug report would start the subject line with "BUG REPORT:" and then the machine name, those who don't want to read them could add "BUG REPORT" to their kill files... (-: Brad Brown :-) bradb@ai.utoronto.ca
dmmooney@lion.waterloo.edu (Dave Mooney) (05/30/89)
In article <89May29.140126edt.10987@ephemeral.ai.toronto.edu> bradb@ai.toronto.edu (Brad Brown) writes: >In article <2530@Portia.Stanford.EDU> armin@Portia.Stanford.EDU (Ronald Mayer) writes: >> Do people in this news-group care, [about bug reports] >>or would posting problems like >>this be unnecessary clutter wasting bandwidth here. > >Perhaps if everyone who writes a bug report would start the subject line >with "BUG REPORT:" and then the machine name, those who don't want to >read them could add "BUG REPORT" to their kill files... > Good idea. There's no reason for me to scratch my head puzzling over a piece code which doesn't work when it's the compiler's fault. I have enough trouble tracking down MY bugs without having to deal with Borland/Microsoft/Symantec's bugs as well. Especially if someone else out in net-land has already figured out the problem. daveq --- Disclaimer --- This is the University that still censors rec.humor.funny. Why would they agree with anything that I say?
campbell@redsox.bsw.com (Larry Campbell) (05/30/89)
Why limit this to PC compilers? Surely you don't believe that the C compilers on minis and mainframes are bug-free! (If you do, please contact me for some investment advice.) -- Larry Campbell The Boston Software Works, Inc. campbell@bsw.com 120 Fulton Street wjh12!redsox!campbell Boston, MA 02146