Patch: memset fixes

Patch: memset fixes

[Anthony Green]

Dave Jones recently pointed out some memset usage errors in gdb and the
ppc simulator.  Here are some patches for the HEAD of the tree.


For gdb...

2005-03-25  Anthony Green  <green@redhat.com>

	* remote.c (remote_store_registers): Fix memset usage.
	* std-regs.c (value_of_builtin_frame_reg): Ditto.
	(value_of_builtin_frame_fp_reg): Ditto.
	(value_of_builtin_frame_reg): Ditto.

For sim/ppc...

2005-03-25  Anthony Green  <green@redhat.com>

	* tree.c (parse_reg_property): Fix memset usage.



Index: gdb/remote.c
===================================================================
RCS file: /cvs/src/src/gdb/remote.c,v
retrieving revision 1.178
diff -u -p -r1.178 remote.c
--- gdb/remote.c	13 Mar 2005 22:06:10 -0000	1.178
+++ gdb/remote.c	25 Mar 2005 20:24:03 -0000
@@ -3501,7 +3501,7 @@ remote_store_registers (int regnum)
   {
     int i;
     regs = alloca (rs->sizeof_g_packet);
-    memset (regs, rs->sizeof_g_packet, 0);
+    memset (regs, 0, rs->sizeof_g_packet);
     for (i = 0; i < NUM_REGS + NUM_PSEUDO_REGS; i++)
       {
 	struct packet_reg *r = &rs->regs[i];
Index: gdb/std-regs.c
===================================================================
RCS file: /cvs/src/src/gdb/std-regs.c,v
retrieving revision 1.13
diff -u -p -r1.13 std-regs.c
--- gdb/std-regs.c	11 Feb 2005 04:06:05 -0000	1.13
+++ gdb/std-regs.c	25 Mar 2005 20:24:03 -0000
@@ -61,7 +61,7 @@ value_of_builtin_frame_reg (struct frame
   val = allocate_value (builtin_type_frame_reg);
   VALUE_LVAL (val) = not_lval;
   buf = value_contents_raw (val);
-  memset (buf, TYPE_LENGTH (value_type (val)), 0);
+  memset (buf, 0, TYPE_LENGTH (value_type (val)));
   /* frame.base.  */
   if (frame != NULL)
     ADDRESS_TO_POINTER (builtin_type_void_data_ptr, buf,
@@ -87,7 +87,7 @@ value_of_builtin_frame_fp_reg (struct fr
       struct value *val = allocate_value (builtin_type_void_data_ptr);
       char *buf = value_contents_raw (val);
       if (frame == NULL)
-	memset (buf, TYPE_LENGTH (value_type (val)), 0);
+	memset (buf, 0, TYPE_LENGTH (value_type (val)));
       else
 	ADDRESS_TO_POINTER (builtin_type_void_data_ptr, buf,
 			    get_frame_base_address (frame));
@@ -105,7 +105,7 @@ value_of_builtin_frame_pc_reg (struct fr
       struct value *val = allocate_value (builtin_type_void_data_ptr);
       char *buf = value_contents_raw (val);
       if (frame == NULL)
-	memset (buf, TYPE_LENGTH (value_type (val)), 0);
+	memset (buf, 0, TYPE_LENGTH (value_type (val)));
       else
 	ADDRESS_TO_POINTER (builtin_type_void_data_ptr, buf,
 			    get_frame_pc (frame));




Index: sim/ppc/tree.c
===================================================================
RCS file: /cvs/src/src/sim/ppc/tree.c,v
retrieving revision 1.3
diff -u -p -r1.3 tree.c
--- sim/ppc/tree.c	11 Jul 2004 23:42:07 -0000	1.3
+++ sim/ppc/tree.c	25 Mar 2005 20:24:05 -0000
@@ -539,7 +539,7 @@ parse_reg_property(device *current,
     if (device_nr_size_cells(bus) > 0)
       chp = parse_size(current, bus, chp, &regs[reg_nr].size);
     else
-      memset(&regs[reg_nr].size, sizeof (&regs[reg_nr].size), 0);
+      memset(&regs[reg_nr].size, 0, sizeof (&regs[reg_nr].size));
   }
 
   /* create it */

Re: Patch: memset fixes

[Daniel Jacobowitz]

On Fri, Mar 25, 2005 at 12:24:32PM -0800, Anthony Green wrote:
> Dave Jones recently pointed out some memset usage errors in gdb and the
> ppc simulator.  Here are some patches for the HEAD of the tree.
> 
> 
> For gdb...
> 
> 2005-03-25  Anthony Green  <green@redhat.com>
> 
> 	* remote.c (remote_store_registers): Fix memset usage.
> 	* std-regs.c (value_of_builtin_frame_reg): Ditto.
> 	(value_of_builtin_frame_fp_reg): Ditto.
> 	(value_of_builtin_frame_reg): Ditto.
> 
> For sim/ppc...
> 
> 2005-03-25  Anthony Green  <green@redhat.com>
> 
> 	* tree.c (parse_reg_property): Fix memset usage.

Ooops.....

Yes, please commit.

-- 
Daniel Jacobowitz
CodeSourcery, LLC

Re: Patch: memset fixes

[Anthony Green]

On Fri, 2005-03-25 at 15:31 -0500, Daniel Jacobowitz wrote:
> On Fri, Mar 25, 2005 at 12:24:32PM -0800, Anthony Green wrote:
> > Dave Jones recently pointed out some memset usage errors in gdb and the
> > ppc simulator.  Here are some patches for the HEAD of the tree.

[snip]

> Yes, please commit.

Done.

Thanks!

AG