chrisb@mipon2.intel.com (Chris Benenati) (11/23/88)
I just built a VAX version of gdb+ V2.8.0 and am running under 4.3 BSD. No matter how many routines down I am, gdb only seems to know about the current stack frame: 'bt' displays only a level 0 frame, 'up' refuses to work, etc. Has anybody else run into this problem? Or did I just screw up when I built gdb? Thanks.
alverson@decwrl.dec.com (Robert Alverson) (12/13/88)
I've brought up gdb+ 2.8.0 on a vax running Ultrix 3.0.
This is a bug in m-vax.h, I think. Here are the diffs:
*** m-vax.h Mon Dec 12 17:54:59 1988
--- m-vax.h.BAK Mon Dec 12 17:24:15 1988
***************
*** 218,224
/* Define other aspects of the stack frame. */
! #define FRAME_SAVED_PC(frame, ignore) (read_memory_integer (frame + 16, 4), 0)
/* Cannot find the AP register value directly from the FP value.
Must find it saved in the frame called by this one, or in the AP register
--- 218,224 -----
/* Define other aspects of the stack frame. */
! #define FRAME_SAVED_PC(frame, ignore) (read_memory_integer (frame + 16, 4))
/* Cannot find the AP register value directly from the FP value.
Must find it saved in the frame called by this one, or in the AP register
In addition, when I compiled expread.y with yacc, I got 101 shift/reduce
conflicts, and 1 reduce/reduce conflict. I've modified the grammar
so that it compiles with only 1 shift/reduce conflict and 1 reduce/reduce
conflict. Those relate to C++ features that I haven't figured out yet.
I need to check my fixes though, so no posting yet.
Bob <alverson@decwrl.dec.com>