i386.c (ix86_fpmath, [...]): New.

Fri Dec 14 12:05:14 CET 2001  Jan Hubicka  <jh@suse.cz>

	* i386.c (ix86_fpmath, ix86_fpmath_string): New.
	(override_option): Set ix86_fpmath.
	* i386.h (MASK_MIX_SSE_I387): Remove.
	(TARGET_SSE_MATH): New.
	(TARGET_MIX_SSE_I387): Use ix86_fpmath.
	(TARGET_SWITCHES): Remove "mix-sse-i387".
	(fpmath_unit): New enum.
	(ix86_fpmath, ix86_fpmath_string): Declare.
	* i386.md (swapsf): Fix condition.
	(add?f, sub?f, mul?f, div?f, sqrt?f, min?f): Use TARGET_SSE_MATH.
	(fp_?f_*_nosse): New.
	(fp_*): Use TARGET_SSE_MATH.
	* invoke.texi (-mfpmath): Document.
	(-msse2): Add.

From-SVN: r47999

gcc/ChangeLog

+Fri Dec 14 12:05:14 CET 2001  Jan Hubicka  <jh@suse.cz>
+
+	* i386.c (ix86_fpmath, ix86_fpmath_string): New.
+	(override_option): Set ix86_fpmath.
+	* i386.h (MASK_MIX_SSE_I387): Remove.
+	(TARGET_SSE_MATH): New.
+	(TARGET_MIX_SSE_I387): Use ix86_fpmath.
+	(TARGET_SWITCHES): Remove "mix-sse-i387".
+	(fpmath_unit): New enum.
+	(ix86_fpmath, ix86_fpmath_string): Declare.
+	* i386.md (swapsf): Fix condition.
+	(add?f, sub?f, mul?f, div?f, sqrt?f, min?f): Use TARGET_SSE_MATH.
+	(fp_?f_*_nosse): New.
+	(fp_*): Use TARGET_SSE_MATH.
+	* invoke.texi (-mfpmath): Document.
+	(-msse2): Add.
+
 2001-12-14  Jason Merrill  <jason@redhat.com>
 
 	* dwarf2out.c (output_die): Print the string in the comment for

gcc/config/i386/i386.c

@@ -578,12 +578,16 @@ enum cmodel ix86_cmodel;
 /* which cpu are we scheduling for */
 enum processor_type ix86_cpu;
 
+/* which unit we are generating floating point math for */
+enum fpmath_unit ix86_fpmath;
+
 /* which instruction set architecture to use.  */
 int ix86_arch;
 
 /* Strings to hold which cpu and instruction set architecture  to use.  */
 const char *ix86_cpu_string;		/* for -mcpu=<xxx> */
 const char *ix86_arch_string;		/* for -march=<xxx> */
+const char *ix86_fpmath_string;		/* for -mfpmath=<xxx> */
 
 /* # of registers to use to pass arguments.  */
 const char *ix86_regparm_string;
@@ -1066,8 +1070,45 @@ override_options ()
       if (TARGET_RTD)
 	error ("-mrtd calling convention not supported in the 64bit mode");
       /* Enable by default the SSE and MMX builtins.  */
-      target_flags |= MASK_SSE2 | MASK_SSE | MASK_MMX | MASK_128BIT_LONG_DOUBLE;
+      target_flags |= (MASK_SSE2 | MASK_SSE | MASK_MMX | MASK_128BIT_LONG_DOUBLE);
+      ix86_fpmath = FPMATH_SSE;
      }
+  else
+    ix86_fpmath = FPMATH_387;
+
+  if (ix86_fpmath_string != 0)
+    {
+      if (! strcmp (ix86_fpmath_string, "387"))
+	ix86_fpmath = FPMATH_387;
+      else if (! strcmp (ix86_fpmath_string, "sse"))
+	{
+	  if (!TARGET_SSE)
+	    {
+	      warning ("SSE instruction set disabled, using 387 arithmetics");
+	      ix86_fpmath = FPMATH_387;
+	    }
+	  else
+	    ix86_fpmath = FPMATH_SSE;
+	}
+      else if (! strcmp (ix86_fpmath_string, "387,sse")
+	       || ! strcmp (ix86_fpmath_string, "sse,387"))
+	{
+	  if (!TARGET_SSE)
+	    {
+	      warning ("SSE instruction set disabled, using 387 arithmetics");
+	      ix86_fpmath = FPMATH_387;
+	    }
+	  else if (!TARGET_80387)
+	    {
+	      warning ("387 instruction set disabled, using SSE arithmetics");
+	      ix86_fpmath = FPMATH_SSE;
+	    }
+	  else
+	    ix86_fpmath = FPMATH_SSE | FPMATH_387;
+	}
+      else 
+	error ("bad value (%s) for -mfpmath= switch", ix86_fpmath_string);
+    }
 
   /* It makes no sense to ask for just SSE builtins, so MMX is also turned
      on by -msse.  */
@@ -8117,8 +8158,8 @@ ix86_expand_fp_movcc (operands)
 
   /* For SF/DFmode conditional moves based on comparisons
      in same mode, we may want to use SSE min/max instructions.  */
-  if (((TARGET_SSE && GET_MODE (operands[0]) == SFmode)
-       || (TARGET_SSE2 && GET_MODE (operands[0]) == DFmode))
+  if (((TARGET_SSE_MATH && GET_MODE (operands[0]) == SFmode)
+       || (TARGET_SSE2 && TARGET_SSE_MATH && GET_MODE (operands[0]) == DFmode))
       && GET_MODE (ix86_compare_op0) == GET_MODE (operands[0])
       /* The SSE comparisons does not support the LTGT/UNEQ pair.  */
       && (!TARGET_IEEE_FP

gcc/config/i386/i386.h

@@ -272,7 +272,9 @@ extern const int x86_epilogue_using_move, x86_decompose_lea;
 
 #define TARGET_SSE ((target_flags & (MASK_SSE | MASK_SSE2)) != 0)
 #define TARGET_SSE2 ((target_flags & MASK_SSE2) != 0)
-#define TARGET_MIX_SSE_I387 ((target_flags & MASK_MIX_SSE_I387) != 0)
+#define TARGET_SSE_MATH ((ix86_fpmath & FPMATH_SSE) != 0)
+#define TARGET_MIX_SSE_I387 ((ix86_fpmath & FPMATH_SSE) \
+			     && (ix86_fpmath & FPMATH_387))
 #define TARGET_MMX ((target_flags & MASK_MMX) != 0)
 #define TARGET_3DNOW ((target_flags & MASK_3DNOW) != 0)
 #define TARGET_3DNOW_A ((target_flags & MASK_3DNOW_A) != 0)
@@ -365,10 +367,6 @@ extern const int x86_epilogue_using_move, x86_decompose_lea;
   { "no-sse2",			 -MASK_SSE2, N_("") },			      \
   { "no-sse2",			 MASK_SSE2_SET,				      \
     N_("Do not support MMX, SSE and SSE2 builtins and code generation") },    \
-  { "mix-sse-i387",		 MASK_MIX_SSE_I387,			      \
-    N_("Use both SSE and i387 instruction sets for floating point arithmetics") },\
-  { "no-mix-sse-i387",		-MASK_MIX_SSE_I387,			      \
-    N_("Do not use both SSE and i387 instruction sets for floating point arithmetics") },\
   { "128bit-long-double",	 MASK_128BIT_LONG_DOUBLE,		      \
     N_("sizeof(long double) is 16") },					      \
   { "96bit-long-double",	-MASK_128BIT_LONG_DOUBLE,		      \
@@ -404,8 +402,14 @@ enum processor_type
   PROCESSOR_PENTIUM4,
   PROCESSOR_max
 };
+enum fpmath_unit
+{
+  FPMATH_387 = 1,
+  FPMATH_SSE = 2
+};
 
 extern enum processor_type ix86_cpu;
+extern enum fpmath_unit ix86_fpmath;
 
 extern int ix86_arch;
 
@@ -421,6 +425,8 @@ extern int ix86_arch;
 #define TARGET_OPTIONS						\
 { { "cpu=",		&ix86_cpu_string,			\
     N_("Schedule code for given CPU")},				\
+  { "fpmath=",		&ix86_fpmath_string,			\
+    N_("Generate floating point mathematics using given instruction set")},\
   { "arch=",		&ix86_arch_string,			\
     N_("Generate code for given CPU")},				\
   { "regparm=",		&ix86_regparm_string,			\
@@ -1271,7 +1277,7 @@ enum reg_class
 #define SSE_REG_P(n) (REG_P (n) && SSE_REGNO_P (REGNO (n)))
 
 #define SSE_FLOAT_MODE_P(m) \
-  ((TARGET_SSE && (m) == SFmode) || (TARGET_SSE2 && (m) == DFmode))
+  ((TARGET_SSE_MATH && (m) == SFmode) || (TARGET_SSE2 && (m) == DFmode))
 
 #define MMX_REGNO_P(n) ((n) >= FIRST_MMX_REG && (n) <= LAST_MMX_REG)
 #define MMX_REG_P(xop) (REG_P (xop) && MMX_REGNO_P (REGNO (xop)))
@@ -3112,6 +3118,7 @@ extern enum cmodel ix86_cmodel;
 /* Variables in i386.c */
 extern const char *ix86_cpu_string;		/* for -mcpu=<xxx> */
 extern const char *ix86_arch_string;		/* for -march=<xxx> */
+extern const char *ix86_fpmath_string;		/* for -mfpmath=<xxx> */
 extern const char *ix86_regparm_string;		/* # registers to use to pass args */
 extern const char *ix86_align_loops_string;	/* power of two alignment for loops */
 extern const char *ix86_align_jumps_string;	/* power of two alignment for non-loop jumps */

gcc/doc/invoke.texi

@@ -473,12 +473,12 @@ in the following sections.
 
 @emph{i386 and x86-64 Options}
 @gccoptlist{
--mcpu=@var{cpu-type}  -march=@var{cpu-type} @gol
+-mcpu=@var{cpu-type}  -march=@var{cpu-type} -mfpmath=@var{unit} @gol
 -mintel-syntax -mieee-fp  -mno-fancy-math-387 @gol
 -mno-fp-ret-in-387  -msoft-float  -msvr3-shlib @gol
 -mno-wide-multiply  -mrtd  -malign-double @gol
 -mpreferred-stack-boundary=@var{num} @gol
--mmmx  -msse  -m3dnow @gol
+-mmmx  -msse -msse2 -msse-math -m3dnow @gol
 -mthreads  -mno-align-stringops  -minline-all-stringops @gol
 -mpush-args  -maccumulate-outgoing-args  -m128bit-long-double @gol
 -m96bit-long-double  -mregparm=@var{num}  -momit-leaf-frame-pointer @gol
@@ -7513,6 +7513,48 @@ These options are synonyms for @option{-mcpu=i386}, @option{-mcpu=i486},
 @option{-mcpu=pentium}, and @option{-mcpu=pentiumpro} respectively.
 These synonyms are deprecated.
 
+@item -mfpmath=@var{unit}
+@opindex march
+generate floating point arithmetics for selected unit @var{unit}.  the choices
+for @var{unit} are:
+
+@table @samp
+@item 387
+Use the standard 387 floating point coprocessor present majority of chips and
+emulated otherwise.  Code compiled with this option will run almost everywhere.
+The temporary results are computed in 80bit precesion instead of precision
+specified by the type resulting in slightly different results compared to most
+of other chips. See @option{-ffloat-store} for more detailed description.
+
+This is the default choice for i386 compiler.
+
+@item sse
+Use scalar floating point instructions present in the SSE instruction set.
+This instruction set is supported by Pentium3 and newer chips, in the AMD line
+by Athlon-4, Athlon-xp and Athlon-mp chips.  The earlier version of SSE
+instruction set supports only single precision arithmetics, thus the double and
+extended precision arithmetics is still done using 387.  Later version, present
+only in Pentium4 and the future AMD x86-64 chips supports double precision
+arithmetics too.
+
+For i387 you need to use @option{-march=@var{cpu-type}}, @option{-msse} or
+@option{-msse2} switches to enable SSE extensions and make this option
+effective.  For x86-64 compiler, these extensions are enabled by default.
+
+The resulting code should be considerably faster in majority of cases and avoid
+the numerical instability problems of 387 code, but may break some existing
+code that expects temporaries to be 80bit.
+
+This is the default choice for x86-64 compiler.
+
+@item sse,387
+Attempt to utilize both instruction sets at once.  This effectivly double the
+amount of available registers and on chips with separate execution units for
+387 and SSE the execution resources too.  Use this option with care, as it is
+still experimental, because gcc register allocator does not model separate
+functional units well resulting in instable performance.
+@end table
+
 @item -mintel-syntax
 @opindex mintel-syntax
 Emit assembly using Intel syntax opcodes instead of AT&T syntax.
@@ -7663,6 +7705,8 @@ preferred alignment to @option{-mpreferred-stack-boundary=2}.
 @itemx -mno-mmx
 @item -msse
 @itemx -mno-sse
+@item -msse2
+@itemx -mno-sse2
 @item -m3dnow
 @itemx -mno-3dnow
 @opindex mmmx