use HAVE_BUILTIN___BUILTIN_MUL_OVERFLOW

We already check for __builtin_mul_overflow in configure but never
actually referred it before.  Why not call it if available, because
that should render supposedly-optimial assembly outputs.

Optionally if __builtin_mul_overflow_p is available, which is the case
for recent GCC, use that to detect fixnum overflow.  This is much
faster than the previous.  On my machine generated assembly of
numeric.c:int_pow reduces from 480 to 448 bytes, according to nm(1).
Also on my machine, following script boosts from 7.819 to 6.929 sec.

time ./miniruby -e 'i=0; while i < 30_000_000 do i += 1; 7 ** 23; end'

Signed-off-by: Urabe, Shyouhei <shyouhei@ruby-lang.org>


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57784 b2dd03c8-39d4-4d8f-98ff-823fe69b080e

1 files changed, 3 insertions, 2 deletions

diff --git a/process.c b/process.c
index d4d15a69b3..cd6e59e290 100644
--- a/process.c
+++ b/process.c
@@ -6916,11 +6916,13 @@ typedef LONG_LONG timetick_int_t;
 #define TIMETICK_INT_MIN LLONG_MIN
 #define TIMETICK_INT_MAX LLONG_MAX
 #define TIMETICK_INT2NUM(v) LL2NUM(v)
+#define MUL_OVERFLOW_TIMETICK_P(a, b) MUL_OVERFLOW_LONG_LONG_P(a, b)
 #else
 typedef long timetick_int_t;
 #define TIMETICK_INT_MIN LONG_MIN
 #define TIMETICK_INT_MAX LONG_MAX
 #define TIMETICK_INT2NUM(v) LONG2NUM(v)
+#define MUL_OVERFLOW_TIMETICK_P(a, b) MUL_OVERFLOW_LONG_P(a, b)
 #endif
 
 CONSTFUNC(static timetick_int_t gcd_timetick_int(timetick_int_t, timetick_int_t));
@@ -7036,8 +7038,7 @@ timetick2integer(struct timetick *ttp,
         timetick_int_t t = ttp->giga_count * 1000000000 + ttp->count;
         for (i = 0; i < num_numerators; i++) {
             timetick_int_t factor = numerators[i];
-            if (MUL_OVERFLOW_SIGNED_INTEGER_P(factor, t,
-                        TIMETICK_INT_MIN, TIMETICK_INT_MAX))
+            if (MUL_OVERFLOW_TIMETICK_P(factor, t))
                 goto generic;
             t *= factor;
         }