Re: gdb-6.8 branchpoint

[Michael Snyder <msnyder@specifix.com>]

[-- Attachment #1: Type: text/plain, Size: 542 bytes --]

On Tue, 2008-02-26 at 14:30 -0800, Joel Brobecker wrote:
> > As per earlier on-list discussion, any chance of applying this simple 
> > fix to prevent random SEGV's in gdb?
> 
> Yes, this can go in the branch on the condition that it is approved
> by a Global Maintainer for HEAD first.

Sorry for joining the discussion late -- as noted on the 
original thread, I'd prefer to fix the problem at the 
reg_flush_command function as attached.  Certainly I meant
to do this when I wrote the darn thing.

Daniel?  This seem about right to you?



[-- Attachment #2: regcache.txt --]
[-- Type: text/plain, Size: 680 bytes --]

2008-02-26  Michael Snyder  <msnyder@specifix.com>

	* regcache.c (reg_flush_command): Void the frame cache as well
	as the registers cache.

Index: regcache.c
===================================================================
RCS file: /cvs/src/src/gdb/regcache.c,v
retrieving revision 1.164
diff -p -r1.164 regcache.c
*** regcache.c	18 Feb 2008 16:37:17 -0000	1.164
--- regcache.c	26 Feb 2008 22:57:31 -0000
*************** reg_flush_command (char *command, int fr
*** 873,878 ****
--- 873,879 ----
  {
    /* Force-flush the register cache.  */
    registers_changed ();
+   reinit_frame_cache ();
    if (from_tty)
      printf_filtered (_("Register cache flushed.\n"));
  }