2004-05-08 Andrew Cagney <cagney@redhat.com>
* ns32k-tdep.c (ns32k_gdbarch_init): Do not set deprecated pc_in_call_dummy. * inferior.h (deprecated_pc_in_call_dummy_on_stack): Delete. * blockframe.c (deprecated_pc_in_call_dummy_on_stack): Delete.
This commit is contained in:
parent
527ca6bb9e
commit
0c93b7b642
4 changed files with 7 additions and 39 deletions
|
@ -519,33 +519,6 @@ block_innermost_frame (struct block *block)
|
|||
below is for infrun.c, which may give the macro a pc without that
|
||||
subtracted out. */
|
||||
|
||||
/* Is the PC in a call dummy? SP and FRAME_ADDRESS are the bottom and
|
||||
top of the stack frame which we are checking, where "bottom" and
|
||||
"top" refer to some section of memory which contains the code for
|
||||
the call dummy. Calls to this macro assume that the contents of
|
||||
SP_REGNUM and DEPRECATED_FP_REGNUM (or the saved values thereof),
|
||||
respectively, are the things to pass.
|
||||
|
||||
This won't work on the 29k, where SP_REGNUM and
|
||||
DEPRECATED_FP_REGNUM don't have that meaning, but the 29k doesn't
|
||||
use ON_STACK. This could be fixed by generalizing this scheme,
|
||||
perhaps by passing in a frame and adding a few fields, at least on
|
||||
machines which need them for DEPRECATED_PC_IN_CALL_DUMMY.
|
||||
|
||||
Something simpler, like checking for the stack segment, doesn't work,
|
||||
since various programs (threads implementations, gcc nested function
|
||||
stubs, etc) may either allocate stack frames in another segment, or
|
||||
allocate other kinds of code on the stack. */
|
||||
|
||||
int
|
||||
deprecated_pc_in_call_dummy_on_stack (CORE_ADDR pc, CORE_ADDR sp,
|
||||
CORE_ADDR frame_address)
|
||||
{
|
||||
return (INNER_THAN ((sp), (pc))
|
||||
&& (frame_address != 0)
|
||||
&& INNER_THAN ((pc), (frame_address)));
|
||||
}
|
||||
|
||||
/* Returns true for a user frame or a call_function_by_hand dummy
|
||||
frame, and false for the CRT0 start-up frame. Purpose is to
|
||||
terminate backtrace. */
|
||||
|
|
Loading…
Add table
Add a link
Reference in a new issue