Mirror of the gdb-patches mailing list
 help / color / mirror / Atom feed
* [rfa] missing frame_register
@ 2003-03-17 19:47 David Carlton
  2003-03-17 20:56 ` Keith Seitz
  0 siblings, 1 reply; 3+ messages in thread
From: David Carlton @ 2003-03-17 19:47 UTC (permalink / raw)
  To: gdb-patches; +Cc: Andrew Cagney

There's a remaining use of get_saved_register in
gdbtk/generic/gdbtk-register.c.  This patch attempts to convert it
over, but I might have gotten it wrong: I just copied some similar
conversions from Andrew's earlier patch.  All I've done is verified
that this patch allows the file in question to compile; I don't use
gdbtk or run the gdbtk tests, so I don't have any way of checking
whether or not it works.

Andrew, is this OK?  I assume it's either obviously okay to commit or
obviously wrong.

David Carlton
carlton@math.stanford.edu

2003-03-17  David Carlton  <carlton@math.stanford.edu>

	* generic/gdbtk-register.c (get_register): Replace use of
	get_saved_register by frame_register.

Index: gdbtk-register.c
===================================================================
RCS file: /cvs/src/src/gdb/gdbtk/generic/gdbtk-register.c,v
retrieving revision 1.15
diff -u -p -r1.15 gdbtk-register.c
--- gdbtk-register.c	12 Feb 2003 00:39:46 -0000	1.15
+++ gdbtk-register.c	17 Mar 2003 19:43:25 -0000
@@ -287,8 +287,11 @@ get_register (int regnum, void *arg)
       return;
     }
 
-  get_saved_register (raw_buffer, &optim, (CORE_ADDR *) NULL, deprecated_selected_frame,
-		      regnum, (enum lval_type *) NULL);
+  {
+    int realnum;
+    frame_register (deprecated_selected_frame, regnum, &optim,
+		    NULL, NULL, &realnum, raw_buffer);
+  }
   if (optim)
     {
       Tcl_ListObjAppendElement (NULL, result_ptr->obj_ptr,


^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [rfa] missing frame_register
  2003-03-17 19:47 [rfa] missing frame_register David Carlton
@ 2003-03-17 20:56 ` Keith Seitz
  2003-03-17 21:05   ` David Carlton
  0 siblings, 1 reply; 3+ messages in thread
From: Keith Seitz @ 2003-03-17 20:56 UTC (permalink / raw)
  To: David Carlton; +Cc: gdb-patches, Andrew Cagney

On Mon, 2003-03-17 at 11:47, David Carlton wrote:

I was hoping not to have to do anything about this until later (let 'em
sweat for a few hours! :v), but I don't believe the below will work: 

> +    frame_register (deprecated_selected_frame, regnum, &optim,
> +		    NULL, NULL, &realnum, raw_buffer);

In frame.c:

void
frame_register (struct frame_info *frame, int regnum,
                int *optimizedp, enum lval_type *lvalp,
                CORE_ADDR *addrp, int *realnump, void *bufferp)
{
  /* Require all but BUFFERP to be valid.  A NULL BUFFERP indicates
     that the value proper does not need to be fetched.  */
  gdb_assert (optimizedp != NULL);
  gdb_assert (lvalp != NULL);
  gdb_assert (addrp != NULL);
  gdb_assert (realnump != NULL);
  /* gdb_assert (bufferp != NULL); */

I have a patch that does work (I think), and I will check it in
momentarily, unless you want to follow through. I've done exactly the
same as you, expect I called frame_reigster with no NULL values. Six of
one, half-dozen of the other, I think.

I'll check mine in if I don't see a commit from you. I appreciate your
thinking of me, urgh, insight!

Keith





^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [rfa] missing frame_register
  2003-03-17 20:56 ` Keith Seitz
@ 2003-03-17 21:05   ` David Carlton
  0 siblings, 0 replies; 3+ messages in thread
From: David Carlton @ 2003-03-17 21:05 UTC (permalink / raw)
  To: Keith Seitz; +Cc: gdb-patches, Andrew Cagney

On 17 Mar 2003 13:04:05 -0800, Keith Seitz <keiths@redhat.com> said:
> On Mon, 2003-03-17 at 11:47, David Carlton wrote:
> I was hoping not to have to do anything about this until later (let 'em
> sweat for a few hours! :v), but I don't believe the below will work: 

>> +    frame_register (deprecated_selected_frame, regnum, &optim,
>> +		    NULL, NULL, &realnum, raw_buffer);

> In frame.c:

> void
> frame_register (struct frame_info *frame, int regnum,
>                 int *optimizedp, enum lval_type *lvalp,
>                 CORE_ADDR *addrp, int *realnump, void *bufferp)
> {
>   /* Require all but BUFFERP to be valid.  A NULL BUFFERP indicates
>      that the value proper does not need to be fetched.  */
>   gdb_assert (optimizedp != NULL);
>   gdb_assert (lvalp != NULL);
>   gdb_assert (addrp != NULL);
>   gdb_assert (realnump != NULL);
>   /* gdb_assert (bufferp != NULL); */

Oh.  Whoops.  I guess I could have done a bit more checking!  I guess
I needed a few more dummy variables, or something.

> I have a patch that does work (I think), and I will check it in
> momentarily, unless you want to follow through.

Thanks, be my guest.

> I appreciate your thinking of me, urgh, insight!

Well, I figured I owed insight one after I inadvertently broke it
twice a month or so back.

David Carlton
carlton@math.stanford.edu


^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2003-03-17 21:05 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2003-03-17 19:47 [rfa] missing frame_register David Carlton
2003-03-17 20:56 ` Keith Seitz
2003-03-17 21:05   ` David Carlton

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox