From mboxrd@z Thu Jan  1 00:00:00 1970
From: Ulrich Weigand <weigand@i1.informatik.uni-erlangen.de>
To: gdb-patches@sources.redhat.com
Subject: [PATCH] Fix frame ID comparison problem on s390
Date: Thu, 20 May 2004 13:31:00 -0000
Message-id: <200405201331.PAA24924@faui1d.informatik.uni-erlangen.de>
X-SW-Source: 2004-05/msg00606.html

Hello,

even after the siglongjmp patch, there was still one failure reported
with the signull test case on s390(x).

The reason turned out to be a problem with frame ID comparison.  In
the back trace we have
  handler
  <sigtramp>
  <NULL call>
  caller

As the NULL call doesn't set up a stack frame, and because our CFA is
determined by stack pointer at function entry, this means that the 
NULL call frame and the sigtramp frame have the same 'stack_addr'
component of their respective frame IDs.

Furthermore, the NULL call frame has 0 as 'code_addr' component of the
frame ID, because the current PC is in fact 0.

Due to the way frame ID comparison works, this causes the two IDs to
compare equal: the stack_addr is equal, and a zero code_addr is
considered a wild card matching any code_addr.

This in turn causes the backtrace to abort since it encounters two
frames with the same frame ID ...

The patch below is a simple fix (cheating a bit ...) in the s390 back
end: we simply set the code_addr component to 1 if the PC is 0.  (Note
that this can never happen normally because code addresses are always
2-aligned on our platform.)

Tested on s390-ibm-linux and s390x-ibm-linux, fixes one test suite
failure.

OK?

ChangeLog:

	* s390-tdep.c (s390_stub_frame_this_id): Set code_addr part of ID
	to 1 if the PC is 0.

Index: gdb/s390-tdep.c
===================================================================
RCS file: /cvs/src/src/gdb/s390-tdep.c,v
retrieving revision 1.132
diff -c -p -r1.132 s390-tdep.c
*** gdb/s390-tdep.c	4 May 2004 18:50:26 -0000	1.132
--- gdb/s390-tdep.c	20 May 2004 13:23:44 -0000
*************** s390_stub_frame_this_id (struct frame_in
*** 2078,2086 ****
  			 void **this_prologue_cache,
  			 struct frame_id *this_id)
  {
    struct s390_stub_unwind_cache *info
      = s390_stub_frame_unwind_cache (next_frame, this_prologue_cache);
!   *this_id = frame_id_build (info->frame_base, frame_pc_unwind (next_frame));
  }
  
  static void
--- 2078,2089 ----
  			 void **this_prologue_cache,
  			 struct frame_id *this_id)
  {
+   CORE_ADDR pc = frame_pc_unwind (next_frame);
    struct s390_stub_unwind_cache *info
      = s390_stub_frame_unwind_cache (next_frame, this_prologue_cache);
!   /* For 'call via NULL function pointer' stubs we use 1 as code address
!      in order to avoid the '0 matches everything' special case.  */
!   *this_id = frame_id_build (info->frame_base, pc? pc : 1);
  }
  
  static void
-- 
  Dr. Ulrich Weigand
  weigand@informatik.uni-erlangen.de