[racket] debugging core dump - comments appreciated

From: Matthew Flatt (mflatt at cs.utah.edu)
Date: Mon May 23 21:52:30 EDT 2011

At Mon, 23 May 2011 15:19:21 -0600, Matthew Flatt wrote:
> The crash at GC_clear_stack_segment() looks to me like the GC's idea of
> the stack range has gone bad, since the `limit' argument is reported
> out of bounds by gdb. I imagine the immediate crash is a result of
> running out of stack space while trying to grow the stack too far.

Looking more at GC_clear_stack_segment(), it doesn't allocate nearly
enough stack space to run out with only 10 or so calls, at least not
for a normal-sized stack. Is a stack limit set when running the Racket

The `limit' value shown in gdb seems to be clearly out of bounds --- in
the space used by heap allocation --- but I don't see how it could go
so wrong. Maybe gdb isn't reporting that variable's value accurately?

Were any other threads running? That is, could some thread other than
the main thread be the one that crashed?

Posted on the users mailing list.