>> > > So, where should a solution come from? Where in the >> software/hardware >> > > chain would a change be needed to make it work properly? >> > >> > For *nix: prt/crt0, there is a xorl %ebp,%ebp But there might be code >> > that relies on this (main program's parent's frame being nil) >> >> (Continuing the thread from the fpc-pascal list on fpc-devel.) >> >> But why then is a stack trace generated for the unhandled assert >> exception >> and not for the unhandled access violation (under Linux 386)? > > The first is a called procedure, the other an exception handler I guess > (Peter already hinted at that).
This is out of reach for fpc. It is the linux kernel that resets ebp register. >> Under Mac OS X (Darwin), I do get stack traces in both cases. However, >> in spite of compiling with -gl, I don't get line numbers in either case. >> What is the cause for that? > > I don't know, but FPC/Darwin's startup mechanisms are afaik a bit > different > due to the mandatory libc use on Darwin. Linux and FreeBSD are by default > still free of libc. The lineinfo unit doesn't support Darwin yet. _______________________________________________ fpc-devel maillist - [email protected] http://lists.freepascal.org/mailman/listinfo/fpc-devel
