This is the mail archive of the mailing list for the GDB project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]


> > 1)  You don't seem to initialize the pointer fields in the static frame 
> > structure "prologue_cache".
> My patch does this allocation in _initialize_arm_tdep() whereas yours
> does it in check_prologue_cache() and save_prologue_cache().

Argh!  I even looked for this and couldn't find it :-(  I suspect because 
patch hasn't set the function signature correctly for the hunk....

@@ -2209,6 +2225,10 @@ The valid values are:\n");
   add_com ("othernames", class_obscure, arm_othernames,
           "Switch to the next set of register names.");
+  /* Allocate extra_info and saved_regs fields in the prologue cache.  */

> > 2)  Similarly you don't seem to be allocating the saved_regs for the 
> > caller_fi frame (arm_frame_chain).  You do, however, allocate the 
> > extra_info.
> > 3)  You have a cleanup for the extra_info allocated above, which I hadn't 
> > thought about.  I think one is also needed for the saved_regs.
> Are you sure?  Here are what I believe to be the relevant lines from
> the arm_frame_chain hunk:
> +      caller_fi.saved_regs = (CORE_ADDR *) xcalloc (1, SIZEOF_FRAME_SAVED_REGS);
> +      old_chain = make_cleanup (xfree, caller_fi.saved_regs);
> +      caller_fi.extra_info = xcalloc (1, sizeof (struct frame_extra_info));
> +      make_cleanup (xfree, caller_fi.extra_info);
> saved_regs is being allocated and a cleanup is being created.  Were you
> perhaps refering to some other hunk?

Doh! right again.

OK.  I withdraw my version in favour of yours.  All we need to do now is 
get someone to approve it.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]