=?ISO-8859-1?Q?Gisle_S=E6lensminde?= writes: > > On Tue, 17 Feb 2004, Raymond Toy wrote: > > > >>>>> "Gisle" == lensminde <Gisle> writes: > > > > Gisle> The problem is that when I get some type-errors, the condition don't > > give > > Gisle> me a description of the problem or a stack trace, but just gives > > Gisle> segmentation fault. I don't specify the safity level explicitly, > > either to > > Gisle> more or less safe than default (or the optimization for that sake), > > but > > Gisle> the code is compiled. On Linux I get a sensible error message. > > > > You should not get a segfault on solaris. What error message do you > > get on Linux? That might help in locating the bug. > > I'm currently not able to reproduce it. But in my experience, it's > typically type errors, silly things like adding a string to a number or > calling a function with the wrong number of parameters or similar stupid > mistakes you happens to do during development. Normally CMUCL nicely tells > you exactly what you did wring like "Wrong argument count, wanted 1 and > got 2", or Argument Y is not a NUMBER: "a"
That's the expected behavior. FWIW, I've never seen what you're describing. I used 18d fairly heavily on a 4-processor machine, so you might check to see if you get the same problem there. If not that would at least give some idea of when the bug got in the system...
