PATCH: Remove Intel PNI/VMX suffix check

[H. J. Lu]

The 'i' suffix check was added to PNI_Fixup in

http://sourceware.org/ml/binutils/2004-06/msg00421.html

Alan, do you remember why it is needed? I am going to check in this
patch to remove it unless there is a testcase to show the suffix check
is needed.

H.J.
On Mon, Jul 18, 2005 at 09:19:31AM -0600, Jan Beulich wrote:
> You appearantly copied the code from PNI_Fixup, which deals with sidt.
> The 'i' is the second letter, but for you this second letter is 'g'. And
> no, I don't have a testcase, I just saw this when I glanced through your
> patch. Jan
> 
> >>> "H. J. Lu" <hjl@lucon.org> 18.07.05 16:32:21 >>>
> On Mon, Jul 18, 2005 at 02:04:55AM -0600, Jan Beulich wrote:
> > In VMX_Fixup:
> > 
> > >+      /* We might have a suffix.  */
> > >+      if (*p == 'i')
> > >+	--p;
> > 
> > I think this needs to be
> > 
> >       /* We might have a suffix.  */
> >       if (*p == 'g')
> > 	--p;
> 
> Where does it come from? Is that for Intel mode? Do you have a
> testcase?
> 
> 
> H.J.
-------------- next part --------------
2005-07-18  H.J. Lu  <hongjiu.lu@intel.com>

	* i386-dis.c (PNI_Fixup): Remove the suffix check.
	(VMX_Fixup): Likewise.

--- opcodes/i386-dis.c.fixup	2005-07-15 08:08:25.000000000 -0700
+++ opcodes/i386-dis.c	2005-07-18 09:43:30.708634946 -0700
@@ -4419,10 +4419,6 @@ PNI_Fixup (int extrachar ATTRIBUTE_UNUSE
       /* Override "sidt".  */
       char *p = obuf + strlen (obuf) - 4;
 
-      /* We might have a suffix.  */
-      if (*p == 'i')
-	--p;
-
       if (rm)
 	{
 	  /* mwait %eax,%ecx  */
@@ -4617,10 +4613,6 @@ VMX_Fixup (int extrachar ATTRIBUTE_UNUSE
       /* Override "sgdt".  */
       char *p = obuf + strlen (obuf) - 4;
 
-      /* We might have a suffix.  */
-      if (*p == 'i')
-	--p;
-
       switch (rm)
 	{
 	case 1: