i386.c (ix86_adjust_cost): Don't increase the cost for load+operation for PROCESSOR_PENTIUMPRO...

	* i386.c (ix86_adjust_cost): Don't increase the cost for
	load+operation for PROCESSOR_PENTIUMPRO, it is already
	modelled in the DFA description

From-SVN: r82571

gcc/ChangeLog

+2004-06-02  Steven Bosscher  <stevenb@suse.de>
+
+	* i386.c (ix86_adjust_cost): Don't increase the cost for
+	load+operation for PROCESSOR_PENTIUMPRO, it is already
+	modelled in the DFA description
+
 2004-06-01  Jerry Quinn  <jlquinn@optonline.net>
 
 	* Makefile.in (EXPR_H): Add insn-config.h, function.h,
@@ -45,7 +51,7 @@
 	and rewrite for __builtin_offsetof.
 
 2004-06-01  Peter Barada  <peter@the-baradas.com>
-            Peter Jakubek  <peter@laseranimation.com>
+	    Peter Jakubek  <peter@laseranimation.com>
 
 	* config/m68k/m68k.c(m68k_output_mi_thunk):  For ColdFire, use %d0 as
 	a scratch to perform an add to memory.
@@ -72,8 +78,8 @@
 
 2004-06-01  Jeff Law  <law@redhat.com>
 
-        * stmt.c (expand_decl): Be more selective about calling
-        mark_reg_pointer.
+	* stmt.c (expand_decl): Be more selective about calling
+	mark_reg_pointer.
 
 2004-06-01  Nicola Pero  <nicola@brainstorm.co.uk>
 

gcc/config/i386/i386.c

@@ -12261,14 +12261,6 @@ ix86_adjust_cost (rtx insn, rtx link, rtx dep_insn, int cost)
 
     case PROCESSOR_PENTIUMPRO:
       memory = get_attr_memory (insn);
-      dep_memory = get_attr_memory (dep_insn);
-
-      /* Since we can't represent delayed latencies of load+operation,
-	 increase the cost here for non-imov insns.  */
-      if (dep_insn_type != TYPE_IMOV
-          && dep_insn_type != TYPE_FMOV
-          && (dep_memory == MEMORY_LOAD || dep_memory == MEMORY_BOTH))
-	cost += 1;
 
       /* INT->FP conversion is expensive.  */
       if (get_attr_fp_int_src (dep_insn))