Add test infrastructure for narrowing libm functions.

This patch continues preparations for adding TS 18661-1 narrowing libm
functions by adding the required testsuite infrastructure to test such
functions through the libm-test infrastructure.

That infrastructure is based around testing for a single type, FLOAT.
For the narrowing functions, FLOAT, the "main" type for testing, is
the function return type; the argument type is ARG_FLOAT.  This is
consistent with how the code built once for each type,
libm-test-support.c, depends on FLOAT for such things as calculating
ulps errors in results but can already handle different argument types
(pointers, integers, long double for nexttoward).

Makefile machinery is added to handle building tests for all pairs of
types for which there are narrowing functions (as with non-narrowing
functions, aliases are tested just the same as the functions they
alias).  gen-auto-libm-tests gains a --narrow option for building
outputs for narrowing functions (so narrowing sqrt and fma will share
the same inputs as non-narrowing, but gen-auto-libm-tests will be run
with and without that option to generate different output files).  In
the narrowing case, the auto-libm-test-out-narrow-* files include
annotations for each test about what properties ARG_FLOAT must have to
be able to represent all the inputs for that test; those annotations
result in calls to the TEST_COND_arg_fmt macro.

gen-libm-test.pl has some minor updates to handle narrowing tests (for
example, arguments in such tests must be surrounded by ARG_LIT calls
instead of LIT calls).  Various new macros are added to the C test
support code (for example, sNaN initializers need to be properly
typed, so arg_snan_value is added; other such arg_* macros are added
as it seems cleanest to do so, though some are not strictly required).
Special-casing of the ibm128 format to allow for its limitations is
adjusted to handle it as the argument format as well as as the result
format; thus, the tests of the new functions allow nonzero ulps only
in the case where ibm128 is the argument format, as otherwise the
functions correspond to fully-defined IEEE operations.  The ulps in
question appear as e.g. 'Function: "add_ldouble"' in libm-test-ulps
(with 1ulp errors then listed for double and float for that function
in powerpc); no support is added to generate corresponding faddl /
daddl ulps listings in the ulps table in the manual.

For the previous patch, I noted the need to avoid spurious macro
expansions of identifiers such as "add".  A test test-narrow-macros.c
is added to verify such macro expansions are successfully avoided, and
there is also a -mlong-double-64 version of that test for ldbl-opt.
This test is set up to cover the full set of relevant identifiers from
the start rather than adding functions one at a time as each function
group is added.

Tested for x86_64 (this patch in isolation, as well as testing for
various configurations in conjunction with the actual addition of
"add" functions).

	* math/Makefile (test-type-pairs): New variable.
	(test-type-pairs-f64xf128-yes): Likewise.
	(tests): Add test-narrow-macros.
	(libm-test-funcs-narrow): New variable.
	(libm-test-c-narrow): Likewise.
	(generated): Add $(libm-test-c-narrow).
	(libm-tests-base-narrow): New variable.
	(libm-tests-narrow): Likewise.
	(libm-tests): Add $(libm-tests-narrow).
	(libm-tests-for-type): Handle $(libm-tests-narrow).
	(libm-test-c-narrow-obj): New variable.
	($(libm-test-c-narrow-obj)): New rule.
	($(foreach t,$(libm-tests-narrow),$(objpfx)$(t).c)): Likewise.
	($(foreach f,$(libm-test-funcs-narrow),$(objpfx)$(o)-$(f).o)): Use
	$(o-iterator) to set dependencies and CFLAGS.
	* math/gen-auto-libm-tests.c: Document use for narrowing
	functions.
	(output_for_one_input_case): Take argument NARROW.
	(generate_output): Likewise.  Update call to
	output_for_one_input_case.
	(main): Take --narrow option.  Update call to generate_output.
	* math/gen-libm-test.pl (_apply_lit): Take macro name as argument.
	(apply_lit): Update call to _apply_lit.
	(apply_arglit): New function.
	(parse_args): Handle "a" arguments.
	(parse_auto_input): Handle format names using ":".
	* math/README.libm-test: Document "a" parameter type.
	* math/libm-test-support.h (ARG_TYPE_MIN): New macro.
	(ARG_TYPE_TRUE_MIN): Likewise.
	(ARG_TYPE_MAX): Likwise.
	(ARG_MIN_EXP): Likewise.
	(ARG_MAX_EXP): Likewise.
	(ARG_MANT_DIG): Likewise.
	(TEST_COND_arg_ibm128): Likewise.
	(TEST_COND_ibm128_libgcc): Define conditional on [ARG_FLOAT].
	(TEST_COND_arg_fmt): New macro.
	(init_max_error): Update prototype.
	* math/libm-test-support.c (test_ibm128): New variable.
	(init_max_error): Take argument testing_ibm128 and set test_ibm128
	instead of using [TEST_COND_ibm128] conditional.
	(test_exceptions): Use test_ibm128 instead of TEST_COND_ibm128.
	* math/libm-test-driver.c (STR_ARG_FLOAT): New macro.
	[TEST_NARROW] (TEST_MSG): New definition.
	(arg_plus_zero): New macro.
	(arg_minus_zero): Likewise.
	(arg_plus_infty): Likewise.
	(arg_minus_infty): Likewise.
	(arg_qnan_value_pl): Likewise.
	(arg_qnan_value): Likewise.
	(arg_snan_value_pl): Likewise.
	(arg_snan_value): Likewise.
	(arg_max_value): Likewise.
	(arg_min_value): Likewise.
	(arg_min_subnorm_value): Likewise.
	[ARG_FLOAT] (struct test_aa_f_data): New struct type.
	(RUN_TEST_LOOP_aa_f): New macro.
	(TEST_SUFF): New macro.
	(TEST_SUFF_STR): Likewise.
	[!TEST_MATHVEC] (VEC_SUFF): Don't define.
	(TEST_COND_any_ibm128): New macro.
	(START): Use TEST_SUFF and TEST_SUFF_STR in initializer for
	this_func.  Update call to init_max_error.
	* math/test-double.h (FUNC_NARROW_PREFIX): New macro.
	* math/test-float.h (FUNC_NARROW_PREFIX): Likewise.
	* math/test-float128.h (FUNC_NARROW_PREFIX): Likewise.
	* math/test-float32.h (FUNC_NARROW_PREFIX): Likewise.
	* math/test-float32x.h (FUNC_NARROW_PREFIX): Likewise.
	* math/test-float64.h (FUNC_NARROW_PREFIX): Likewise.
	* math/test-float64x.h (FUNC_NARROW_PREFIX): Likewise.
	* math/test-math-scalar.h (TEST_NARROW): Likewise.
	* math/test-math-vector.h (TEST_NARROW): Likewise.
	* math/test-arg-double.h: New file.
	* math/test-arg-float128.h: Likewise.
	* math/test-arg-float32x.h: Likewise.
	* math/test-arg-float64.h: Likewise.
	* math/test-arg-float64x.h: Likewise.
	* math/test-arg-ldouble.h: Likewise.
	* math/test-math-narrow.h: Likewise.
	* math/test-narrow-macros.c: Likewise.
	* sysdeps/ieee754/ldbl-opt/test-narrow-macros-ldbl-64.c: Likewise.
	* sysdeps/ieee754/ldbl-opt/Makefile (tests): Add
	test-narrow-macros-ldbl-64.
	(CFLAGS-test-narrow-macros-ldbl-64.c): New variable.

1 files changed, 136 insertions, 24 deletions

diff --git a/math/gen-auto-libm-tests.c b/math/gen-auto-libm-tests.c
index 934c64bc76..442a10d5e6 100644
--- a/math/gen-auto-libm-tests.c
+++ b/math/gen-auto-libm-tests.c
@@ -25,6 +25,15 @@
 
    gen-auto-libm-tests auto-libm-test-in <func> auto-libm-test-out-<func>
 
+   to generate results for normal libm functions, or
+
+   gen-auto-libm-tests --narrow auto-libm-test-in <func> \
+     auto-libm-test-out-narrow-<func>
+
+   to generate results for a function rounding results to a narrower
+   type (in the case of fma and sqrt, both output files are generated
+   from the same test inputs).
+
    The input file auto-libm-test-in contains three kinds of lines:
 
    Lines beginning with "#" are comments, and are ignored, as are
@@ -120,7 +129,22 @@
    missing or spurious, or because the calculation of correct results
    indicated it was optional).  Conditions "before-rounding" and
    "after-rounding" indicate tests where expectations for underflow
-   exceptions depend on how the architecture detects tininess.  */
+   exceptions depend on how the architecture detects tininess.
+
+   For functions rounding their results to a narrower type, the format
+   given on an output test line is the result format followed by
+   information about the requirements on the argument format to be
+   able to represent the argument values, in the form
+   "format:arg_fmt(MAX_EXP,NUM_ONES,MIN_EXP,MAX_PREC)".  Instead of
+   separate lines for separate argument formats, an output test line
+   relates to all argument formats that can represent the values.
+   MAX_EXP is the maximum exponent of a nonzero bit in any argument,
+   or 0 if all arguments are zero; NUM_ONES is the maximum number of
+   leading bits with value 1 in an argument with exponent MAX_EXP, or
+   0 if all arguments are zero; MIN_EXP is the minimum exponent of a
+   nonzero bit in any argument, or 0 if all arguments are zero;
+   MAX_PREC is the maximum precision required to represent all
+   arguments, or 0 if all arguments are zero.  */
 
 #define _GNU_SOURCE
 
@@ -1718,12 +1742,13 @@ output_generic_value (FILE *fp, const char *filename, const generic_value *v,
     }
 }
 
-/* Generate test output to FP (name FILENAME) for test function TF,
-   input test IT, choice of input values INPUTS.  */
+/* Generate test output to FP (name FILENAME) for test function TF
+   (rounding results to a narrower type if NARROW), input test IT,
+   choice of input values INPUTS.  */
 
 static void
 output_for_one_input_case (FILE *fp, const char *filename, test_function *tf,
-			   input_test *it, generic_value *inputs)
+			   bool narrow, input_test *it, generic_value *inputs)
 {
   bool long_bits_matters = false;
   bool fits_long32 = true;
@@ -1794,24 +1819,81 @@ output_for_one_input_case (FILE *fp, const char *filename, test_function *tf,
 	  mpfr_t res[rm_num_modes];
 	  unsigned int exc_before[rm_num_modes];
 	  unsigned int exc_after[rm_num_modes];
+	  bool have_fp_arg = false;
+	  int max_exp = 0;
+	  int num_ones = 0;
+	  int min_exp = 0;
+	  int max_prec = 0;
 	  for (size_t i = 0; i < tf->num_args; i++)
 	    {
 	      if (inputs[i].type == gtype_fp)
 		{
-		  round_real (res, exc_before, exc_after, inputs[i].value.f,
-			      f);
-		  if (!mpfr_equal_p (res[rm_tonearest], inputs[i].value.f))
-		    fits = false;
-		  for (rounding_mode m = rm_first_mode; m < rm_num_modes; m++)
-		    mpfr_clear (res[m]);
-		  if (!fits)
-		    break;
+		  if (narrow)
+		    {
+		      if (mpfr_zero_p (inputs[i].value.f))
+			continue;
+		      assert (mpfr_regular_p (inputs[i].value.f));
+		      int this_exp, this_num_ones, this_min_exp, this_prec;
+		      mpz_t tmp;
+		      mpz_init (tmp);
+		      mpfr_exp_t e = mpfr_get_z_2exp (tmp, inputs[i].value.f);
+		      if (mpz_sgn (tmp) < 0)
+			mpz_neg (tmp, tmp);
+		      size_t bits = mpz_sizeinbase (tmp, 2);
+		      mp_bitcnt_t tz = mpz_scan1 (tmp, 0);
+		      this_min_exp = e + tz;
+		      this_prec = bits - tz;
+		      assert (this_prec > 0);
+		      this_exp = this_min_exp + this_prec - 1;
+		      assert (this_exp
+			      == mpfr_get_exp (inputs[i].value.f) - 1);
+		      this_num_ones = 1;
+		      while ((size_t) this_num_ones < bits
+			     && mpz_tstbit (tmp, bits - 1 - this_num_ones))
+			this_num_ones++;
+		      mpz_clear (tmp);
+		      if (have_fp_arg)
+			{
+			  if (this_exp > max_exp
+			      || (this_exp == max_exp
+				  && this_num_ones > num_ones))
+			    {
+			      max_exp = this_exp;
+			      num_ones = this_num_ones;
+			    }
+			  if (this_min_exp < min_exp)
+			    min_exp = this_min_exp;
+			  if (this_prec > max_prec)
+			    max_prec = this_prec;
+			}
+		      else
+			{
+			  max_exp = this_exp;
+			  num_ones = this_num_ones;
+			  min_exp = this_min_exp;
+			  max_prec = this_prec;
+			}
+		      have_fp_arg = true;
+		    }
+		  else
+		    {
+		      round_real (res, exc_before, exc_after,
+				  inputs[i].value.f, f);
+		      if (!mpfr_equal_p (res[rm_tonearest], inputs[i].value.f))
+			fits = false;
+		      for (rounding_mode m = rm_first_mode;
+			   m < rm_num_modes;
+			   m++)
+			mpfr_clear (res[m]);
+		      if (!fits)
+			break;
+		    }
 		}
 	    }
 	  if (!fits)
 	    continue;
-	  /* The inputs fit this type, so compute the ideal outputs
-	     and exceptions.  */
+	  /* The inputs fit this type if required to do so, so compute
+	     the ideal outputs and exceptions.  */
 	  mpfr_t all_res[MAX_NRET][rm_num_modes];
 	  unsigned int all_exc_before[MAX_NRET][rm_num_modes];
 	  unsigned int all_exc_after[MAX_NRET][rm_num_modes];
@@ -1895,10 +1977,21 @@ output_for_one_input_case (FILE *fp, const char *filename, test_function *tf,
 		  assert ((merged_exc_before[m] & (1U << exc_underflow)) != 0);
 		}
 	      unsigned int merged_exc = merged_exc_before[m];
-	      if (fprintf (fp, "= %s %s %s%s", tf->name,
-			   rounding_modes[m].name, fp_formats[f].name,
-			   long_cond) < 0)
-		error (EXIT_FAILURE, errno, "write to '%s'", filename);
+	      if (narrow)
+		{
+		  if (fprintf (fp, "= %s %s %s%s:arg_fmt(%d,%d,%d,%d)",
+			       tf->name, rounding_modes[m].name,
+			       fp_formats[f].name, long_cond, max_exp,
+			       num_ones, min_exp, max_prec) < 0)
+		    error (EXIT_FAILURE, errno, "write to '%s'", filename);
+		}
+	      else
+		{
+		  if (fprintf (fp, "= %s %s %s%s", tf->name,
+			       rounding_modes[m].name, fp_formats[f].name,
+			       long_cond) < 0)
+		    error (EXIT_FAILURE, errno, "write to '%s'", filename);
+		}
 	      /* Print inputs.  */
 	      for (size_t i = 0; i < tf->num_args; i++)
 		output_generic_value (fp, filename, &inputs[i], false,
@@ -2158,10 +2251,12 @@ output_for_one_input_case (FILE *fp, const char *filename, test_function *tf,
     generic_value_free (&generic_outputs[i]);
 }
 
-/* Generate test output data for FUNCTION to FILENAME.  */
+/* Generate test output data for FUNCTION to FILENAME.  The function
+   is interpreted as rounding its results to a narrower type if
+   NARROW.  */
 
 static void
-generate_output (const char *function, const char *filename)
+generate_output (const char *function, bool narrow, const char *filename)
 {
   FILE *fp = fopen (filename, "w");
   if (fp == NULL)
@@ -2177,7 +2272,8 @@ generate_output (const char *function, const char *filename)
 	  if (fputs (it->line, fp) < 0)
 	    error (EXIT_FAILURE, errno, "write to '%s'", filename);
 	  for (size_t k = 0; k < it->num_input_cases; k++)
-	    output_for_one_input_case (fp, filename, tf, it, it->inputs[k]);
+	    output_for_one_input_case (fp, filename, tf, narrow,
+				       it, it->inputs[k]);
 	}
     }
   if (fclose (fp) != 0)
@@ -2187,14 +2283,30 @@ generate_output (const char *function, const char *filename)
 int
 main (int argc, char **argv)
 {
-  if (argc != 4)
+  if (argc != 4
+      && !(argc == 5 && strcmp (argv[1], "--narrow") == 0))
     error (EXIT_FAILURE, 0,
-	   "usage: gen-auto-libm-tests <input> <func> <output>");
+	   "usage: gen-auto-libm-tests [--narrow] <input> <func> <output>");
+  bool narrow;
   const char *input_filename = argv[1];
   const char *function = argv[2];
   const char *output_filename = argv[3];
+  if (argc == 4)
+    {
+      narrow = false;
+      input_filename = argv[1];
+      function = argv[2];
+      output_filename = argv[3];
+    }
+  else
+    {
+      narrow = true;
+      input_filename = argv[2];
+      function = argv[3];
+      output_filename = argv[4];
+    }
   init_fp_formats ();
   read_input (input_filename);
-  generate_output (function, output_filename);
+  generate_output (function, narrow, output_filename);
   exit (EXIT_SUCCESS);
 }