PATCH: In dwarf2read, restore local_symbols and param_symbols

PATCH: In dwarf2read, restore local_symbols and param_symbols

[Jim Blandy]

2002-11-08  Jim Blandy  <jimb@redhat.com>

	* dwarf2read.c (read_func_scope): Restore local_symbols and
	param_symbols after we finish the function context.  (Based on a
	patch from David Edelsohn.)

Index: gdb/dwarf2read.c
===================================================================
RCS file: /cvs/src/src/gdb/dwarf2read.c,v
retrieving revision 1.72
diff -c -r1.72 dwarf2read.c
*** gdb/dwarf2read.c	25 Oct 2002 22:25:55 -0000	1.72
--- gdb/dwarf2read.c	8 Nov 2002 19:30:13 -0000
***************
*** 2013,2018 ****
--- 2013,2026 ----
    /* Make a block for the local symbols within.  */
    finish_block (new->name, &local_symbols, new->old_blocks,
  		lowpc, highpc, objfile);
+   
+   /* In C++, we can have functions nested inside functions (e.g., when
+      a function declares a class that has methods).  This means that
+      when we finish processing a function scope, we may need to go
+      back to building a containing block's symbol lists.  */
+   local_symbols = new->locals;
+   param_symbols = new->params;
+ 
    list_in_scope = &file_symbols;
  }
  

Re: PATCH: In dwarf2read, restore local_symbols and param_symbols

[Andrew Cagney]

> 2002-11-08  Jim Blandy  <jimb@redhat.com>
> 
> 	* dwarf2read.c (read_func_scope): Restore local_symbols and
> 	param_symbols after we finish the function context.  (Based on a
> 	patch from David Edelsohn.)
> 
Should this be part of 5.3?

Andrew

Re: PATCH: In dwarf2read, restore local_symbols and param_symbols

[Jim Blandy]

Andrew Cagney <ac131313@redhat.com> writes:

> > 2002-11-08  Jim Blandy  <jimb@redhat.com>
> > 	* dwarf2read.c (read_func_scope): Restore local_symbols and
> > 	param_symbols after we finish the function context.  (Based on a
> > 	patch from David Edelsohn.)
> >
> Should this be part of 5.3?

Couldn't hurt.  The macro fix I committed definitely should be.
Shall I commit those to the branch?

Re: PATCH: In dwarf2read, restore local_symbols and param_symbols

[Andrew Cagney]

> Andrew Cagney <ac131313@redhat.com> writes:
> 
> 
>> > 2002-11-08  Jim Blandy  <jimb@redhat.com>
>> > 	* dwarf2read.c (read_func_scope): Restore local_symbols and
>> > 	param_symbols after we finish the function context.  (Based on a
>> > 	patch from David Edelsohn.)
>> >
> 
>> Should this be part of 5.3?
> 
> 
> Couldn't hurt.  The macro fix I committed definitely should be.
> Shall I commit those to the branch?

(waves hands) How would I know?  They look safe :-)

Andrew

Re: PATCH: In dwarf2read, restore local_symbols and param_symbols

[Jim Blandy]

Andrew Cagney <ac131313@redhat.com> writes:

> > Andrew Cagney <ac131313@redhat.com> writes:
> >
> >> > 2002-11-08  Jim Blandy  <jimb@redhat.com>
> >> > 	* dwarf2read.c (read_func_scope): Restore local_symbols and
> >> > 	param_symbols after we finish the function context.  (Based on a
> >> > 	patch from David Edelsohn.)
> >> >
> >
> >> Should this be part of 5.3?
> > Couldn't hurt.  The macro fix I committed definitely should be.
> > Shall I commit those to the branch?
> 
> (waves hands) How would I know?  They look safe :-)

I think they're safe, too, so I've committed them to the 5.3 branch,
along with the regression test for one of them.

Re: PATCH: In dwarf2read, restore local_symbols and param_symbols

[Andrew Cagney]

>> (waves hands) How would I know?  They look safe :-)
> 
> 
> I think they're safe, too, so I've committed them to the 5.3 branch,
> along with the regression test for one of them.

Ya!  I think there is a glimmer of light for 5.3 ..

Andrew

Re: PATCH: In dwarf2read, restore local_symbols and param_symbols

[Jim Blandy]

Andrew Cagney <ac131313@redhat.com> writes:

> >> (waves hands) How would I know?  They look safe :-)
> > I think they're safe, too, so I've committed them to the 5.3 branch,
> > along with the regression test for one of them.
> 
> Ya!  I think there is a glimmer of light for 5.3 ..

:)

You will see the kfail->fail regression I mentioned on gdb@sources,
unfortunately.  But as I say, I think it's not a real regression.