[cagney_framefunc-20030403-branch] Half working frame ID eq

[Andrew Cagney]

Hello,

I've created cagney_framefunc-20030403-branch as a test bed for 
frame_id_eq().

Attatched is the branches first jumbo patch.  It gets things half 
working (PPC and d10v are ok but the i386 fails).

The change:

- renames of the internal frame_id fields: pc -> func_addr, base -> 
stack_addr

- adds new methods get_frame_func() and frame_func_unwind() that 
cache/return a frame's function

- eliminates the redundant frame->pc (frame->next->pc_unwind_cache) and 
frame->frame (frame->id.stack_addr)

- tweaks to d10v so that it returns the func addr

(*) adds nasty hacks to frame.c that, for legacy targets, badly fudge 
frame->id.func_base

- makes many 'set frame debug' additions

Expect to see random bits of this separated out and committed to the 
mainline.  Testing / fixing needs to continue for a bit longer.

Feel free to try it,
Andrew
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: diffs
URL: <http://sourceware.org/pipermail/gdb-patches/attachments/20030403/2101fbe7/attachment.ksh>