Mirror of the gdb-patches mailing list
 help / color / mirror / Atom feed
From: Andrew Cagney <ac131313@redhat.com>
To: gdb-patches@sources.redhat.com
Subject: [rfa:ppc] Make empty registers empty
Date: Sat, 21 Jun 2003 01:31:00 -0000	[thread overview]
Message-ID: <3EF3B51C.8070801@redhat.com> (raw)

[-- Attachment #1: Type: text/plain, Size: 951 bytes --]

Hello,

The PPC, for some processor variants, has an empty register R0:

#define R0              { 0, 0, 0, 0, 0 }
/* UISA-level SPRs for PowerPC without floating point support.  */
#define PPC_UISA_NOFP_SPRS \
   /* 66 */ R4(cr),  R(lr), R(ctr), R4(xer), R0
/* Motorola e500.  */
static const struct reg registers_e500[] =
{
   R(pc), R(ps),
   /* cr, lr, ctr, xer, "" */
   PPC_UISA_NOFP_SPRS,
   ...
}

The problem is that register_virtual_type for those registers returns 
int32 and, when things are added up, that doesn't agree with 
register_raw_size (aka regsize), et.al.  This makes regcache.[hc] grumpy 
(try 'set architecture powerpc:e500 :-).

The attached adds an empty integer type (for want of a better choice) 
and then modifies register_virtual_type so that it is returned when the 
register is zero.

With this in place it becomes possible (well in theory) to delete all 
the deprecated register * methods.

Ok? GDB 6 branch?

Andrew

[-- Attachment #2: diffs --]
[-- Type: text/plain, Size: 2690 bytes --]

2003-06-20  Andrew Cagney  <cagney@redhat.com>

	* rs6000-tdep.c (rs6000_register_virtual_type): Add explict cases
	for 0 "int0" and 4 "int32" sized registers.
	* gdbtypes.c (builtin_type_int0): Define.
	(build_gdbtypes): Initialize builtin_type_int0.
	* gdbtypes.h (builtin_type_int0): Declare.

Index: gdbtypes.c
===================================================================
RCS file: /cvs/src/src/gdb/gdbtypes.c,v
retrieving revision 1.73
diff -u -r1.73 gdbtypes.c
--- gdbtypes.c	8 Jun 2003 18:27:13 -0000	1.73
+++ gdbtypes.c	21 Jun 2003 01:04:54 -0000
@@ -60,6 +60,7 @@
 struct type *builtin_type_complex;
 struct type *builtin_type_double_complex;
 struct type *builtin_type_string;
+struct type *builtin_type_int0;
 struct type *builtin_type_int8;
 struct type *builtin_type_uint8;
 struct type *builtin_type_int16;
@@ -3410,6 +3411,10 @@
     init_type (TYPE_CODE_STRING, TARGET_CHAR_BIT / TARGET_CHAR_BIT,
 	       0,
 	       "string", (struct objfile *) NULL);
+  builtin_type_int0 =
+    init_type (TYPE_CODE_INT, 0 / 8,
+	       0,
+	       "int0_t", (struct objfile *) NULL);
   builtin_type_int8 =
     init_type (TYPE_CODE_INT, 8 / 8,
 	       0,
Index: gdbtypes.h
===================================================================
RCS file: /cvs/src/src/gdb/gdbtypes.h,v
retrieving revision 1.47
diff -u -r1.47 gdbtypes.h
--- gdbtypes.h	14 May 2003 17:43:17 -0000	1.47
+++ gdbtypes.h	21 Jun 2003 01:04:55 -0000
@@ -956,7 +956,10 @@
    (cf MIPS). */
 extern struct type *builtin_type_bfd_vma;
 
-/* Explicit sizes - see C9X <intypes.h> for naming scheme */
+/* Explicit sizes - see C9X <intypes.h> for naming scheme.  The "int0"
+   is for when an architecture needs to describe a register that has
+   no size.  */
+extern struct type *builtin_type_int0;
 extern struct type *builtin_type_int8;
 extern struct type *builtin_type_uint8;
 extern struct type *builtin_type_int16;
Index: rs6000-tdep.c
===================================================================
RCS file: /cvs/src/src/gdb/rs6000-tdep.c,v
retrieving revision 1.145
diff -u -r1.145 rs6000-tdep.c
--- rs6000-tdep.c	16 Jun 2003 16:47:42 -0000	1.145
+++ rs6000-tdep.c	21 Jun 2003 01:04:58 -0000
@@ -1894,6 +1894,10 @@
       int size = regsize (reg, tdep->wordsize);
       switch (size)
 	{
+	case 0:
+	  return builtin_type_int0;
+	case 4:
+	  return builtin_type_int32;
 	case 8:
 	  if (tdep->ppc_ev0_regnum <= n && n <= tdep->ppc_ev31_regnum)
 	    return builtin_type_vec64;
@@ -1904,8 +1908,8 @@
 	  return builtin_type_vec128;
 	  break;
 	default:
-	  return builtin_type_int32;
-	  break;
+	  internal_error (__FILE__, __LINE__, "Register %d size %d unknown",
+			  n, size);
 	}
     }
 }

             reply	other threads:[~2003-06-21  1:31 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2003-06-21  1:31 Andrew Cagney [this message]
2003-06-23 19:50 ` Kevin Buettner
2003-06-23 21:39   ` Andrew Cagney

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=3EF3B51C.8070801@redhat.com \
    --to=ac131313@redhat.com \
    --cc=gdb-patches@sources.redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox