Eric Pouech writes: > Uwe Bonnes wrote: > > > > Hallo, > > > > MingW as a crosscompiler is a nice tool for writing test programs with > > wine or even do real win32 development work. Being able to single step > > in the win32 source would be a win. However DEBUG_ParseStabs is > > commented out in msc.c. The comment tells that some suff would be "a > > bit difficult"... > > > > I remember some discussion about that with Marcus, but haven't found > > the Mail Exchange yet. Is there some work lurking on that subject in > > some private wine-tree, not yet released? Shouldn't be that "difficult" > > stuff be already implemented in the mingw adaptation of gdb? > > I didn't test the output from mingw compiler (as a matter of stabs format), > but there's been some improvement last summer on stabs handling, so it's may > be it's sufficient (I had it on my todo list to remove this #ifdef in msc.c > but never found time to do it) > I'm currently hunting another bug in the symbol handling in the debugger > and may need to tweedle the stabs reading part. So, I may as well look at > this. I assume from your mail that removing the #ifdef isn't working, is > it ? Well, I can single step, but the lines displayed are off. B.t.w, I wonder about line 451 in hash.c: 449 if (!nh) return FALSE; 450 nh->addr = *addr; 451 nh->flags &= SYM_INVALID; 452 DEBUG_FixAddress( &nh->addr, DEBUG_context.SegDs ); 453 return TRUE; 454 } is &= really meant? I would either expect |= or &=~ Bye Uwe Bonnes [EMAIL PROTECTED] Institut fuer Kernphysik Schlossgartenstrasse 9 64289 Darmstadt --------- Tel. 06151 162516 -------- Fax. 06151 164321 ----------
