New i386 tests for naked register syntax, ie. "eax" instead of "%eax"

gas/testsuite/ChangeLog

@@ -1,3 +1,9 @@
+1999-08-30  Alan Modra  <alan@spri.levels.unisa.edu.au>
+
+	* gas/i386/naked.s, gas/i386/naked.d: Test-suite addition for
+	allow_naked_reg parsing.
+	* gas/i386/i386.exp: Call it.
+
 1999-08-29  Alan Modra  <alan@spri.levels.unisa.edu.au>
 
 	* gas/i386/white.s, gas/i386/white.l: Add space in FP reg tests.

gas/testsuite/gas/i386/i386.exp

@@ -21,6 +21,7 @@ if [istarget "i*86-*-*"] then {
     run_list_test "general" "-al --listing-lhs-width=2"
     run_list_test "inval" "-al"
     run_list_test "modrm" "-al --listing-lhs-width=2"
+    run_dump_test "naked"
     run_dump_test "opcode"
     run_dump_test "prefix"
     run_dump_test "amd"

gas/testsuite/gas/i386/naked.d

@@ -0,0 +1,32 @@
+#objdump: -drw
+#name: i386 naked reg
+
+.*: +file format .*i386.*
+
+Disassembly of section .text:
+
+0+000 <foo>:
+   0:	66 26 ff 23 [ 	]*jmpw   \*%es:\(%ebx\)
+   4:	8a 25 50 00 00 00 [ 	]*mov    0x50,%ah
+   a:	b2 20 [ 	]*mov    \$0x20,%dl
+   c:	bb 00 00 00 00 [ 	]*mov    \$0x0,%ebx	d: (R_386_)?32	.text
+  11:	d9 c9 [ 	]*fxch   %st\(1\)
+  13:	36 8c a4 81 d2 04 00 00 [ 	]*movl   %fs,%ss:0x4d2\(%ecx,%eax,4\)
+  1b:	8c 2c ed 00 00 00 00 [ 	]*movl   %gs,0x0\(,%ebp,8\)
+  22:	26 88 25 00 00 00 00 [ 	]*mov    %ah,%es:0x0
+  29:	2e 8b 74 14 80 [ 	]*mov    %cs:0xffffff80\(%esp,%edx,1\),%esi
+  2e:	f3 65 a5 [ 	]*repz movsl %gs:\(%esi\),%es:\(%edi\)
+  31:	ec [ 	]*in     \(%dx\),%al
+  32:	66 ef [ 	]*out    %ax,\(%dx\)
+  34:	67 d2 14 [ 	]*addr16 rclb %cl,\(%si\)
+  37:	0f 20 d0 [ 	]*mov    %cr2,%eax
+  3a:	0f 72 d0 04 [ 	]*psrld  \$0x4,%mm0
+  3e:	66 47 [ 	]*inc    %di
+  40:	66 51 [ 	]*push   %cx
+  42:	66 58 [ 	]*pop    %ax
+  44:	66 87 dd [ 	]*xchg   %bx,%bp
+  47:	6a 02 [ 	]*push   \$0x2
+  49:	00 00 [ 	]*add    %al,\(%eax\)
+  4b:	00 00 [ 	]*add    %al,\(%eax\)
+  4d:	00 00 [ 	]*add    %al,\(%eax\)
+	...

gas/testsuite/gas/i386/naked.s

@@ -0,0 +1,25 @@
+.att_syntax noprefix
+
+foo:	jmpw es:*(ebx)
+	mov (0x8*0xa),ah
+	mov $(8*4),dl
+	mov $foo,ebx
+	fxch st(1)
+	mov fs,ss:1234(ecx,eax,4)
+	mov gs,ds:(,ebp,8)
+	mov ah,es:0
+	mov cs:-128(esp,edx),esi
+	rep movsl gs:(esi),es:(edi)
+	in dx,al
+	outw (dx)
+ addr16 rclb cl,(si)
+	mov cr2,eax
+	psrld $4,mm0
+	inc di
+	push cx
+	pop ax
+	xchg bx,bp
+	pushl $2
+
+# Force a good alignment.
+.p2align	4,0