[prev in list] [next in list] [prev in thread] [next in thread] 

List:       gdb
Subject:    Re: [Fwd: Re: tdep/1155: s/390 Linux: GDB can't reselect the right frame after an inferior function 
From:       Andrew Cagney <ac131313 () redhat ! com>
Date:       2003-03-30 16:57:44
Message-ID: 3E872208.2050401 () redhat ! com
[Download RAW message or body]

> 
> The original bug, as filed, said:
> 
> <quote>
> If you ask GDB to make an inferior function call when the youngest
> frame is a frameless function, and the selected frame is the
> second-to-youngest frame, then GDB will not properly re-select the
> second-to-youngest frame when the inferior call returns.
> 
> On the S/390, if a function doesn't use alloca, then the compiler just
> uses the stack pointer as the frame pointer.  This means that GDB's
> frame_info structures use the address of the low end of the frame as
> the frame base, not the high end.  (The S/390 stack grows downwards.)
> So, if the youngest function call hasn't allocated any stack space,
> then its frame base address is equal to that of its caller. This means
> that frame_find_by_id is unable to distinguish between the two.
> </quote>
> 
> In what way do you see this problem as resolved?

Just to close this, further analysis revealed that the s390's frame ID's 
stack address wasn't correct (wasn't constant, pointing at the wrong end 
of the frame).  Fixing that will fix the above.

It's also useful to note that, in the critical edge cases, GDB is 
already doing frame/func comparisons vis:
             case PRINT_UNKNOWN:
               /* FIXME: cagney/2002-12-01: Given that a frame ID does
                  (or should) carry around the function and does (or
                  should) use that when doing a frame comparison.  */
               if (stop_step
                   && frame_id_eq (step_frame_id,
                                   get_frame_id (get_current_frame ()))
                   && step_start_function == find_pc_function (stop_pc))
Moving the test to frame_id_eq() is something of a clean up.  It will 
also make it possible to flush out all those remaining edge cases though.

Andrew


[prev in list] [next in list] [prev in thread] [next in thread]