speed up add and sub operators with overflow detection

On A72, google-benchmark measure before and after the patch:
--------------------------------------------------------
Benchmark              Time             CPU   Iterations
--------------------------------------------------------
BM_add_before       13.3 ns         13.3 ns     52626058
BM_sub_before       8.72 ns         8.72 ns     80259343
BM_add_after        4.80 ns         4.80 ns    145926004
BM_sub_after        4.80 ns         4.80 ns    145936496

Before the patch:
fast_long_add_function:
	ldr	x1, [x1]
	ldr	x2, [x2]
	add	x3, x1, x2
	eor	x4, x1, x2
	tbz	x4, #63, .L5
.L2:
	mov	w1, 4
	str	x3, [x0]
	str	w1, [x0, 8]
	ret
	.p2align 2
.L5:
	eor	x4, x1, x3
	tbz	x4, #63, .L2
	scvtf	d0, x1
	scvtf	d1, x2
	mov	w1, 5
	str	w1, [x0, 8]
	fadd	d0, d0, d1
	str	d0, [x0]
	ret

With the patch:
fast_long_add_function:
	ldr    x5, [x1]
	ldr    x6, [x2]
	adds	x5, x5, x6
	bvs	.L2
	mov	w6, 4
	str	x5, [x0]
	str	w6, [x0, 8]
	ret
.L2:
	ldr	x1, [x1]
	mov	w3, 5
	ldr	x2, [x2]
	str	w3, [x0, 8]
	scvtf	d0, x1
	scvtf	d1, x2
	fadd	d0, d0, d1
	str	d0, [x0]
	ret

    php$ ./sapi/cli/php Zend/bench.php

Base:                         Patch:
simple             0.091      simple             0.091
simplecall         0.014      simplecall         0.014
simpleucall        0.041      simpleucall        0.041
simpleudcall       0.045      simpleudcall       0.045
mandel             0.193      mandel             0.193
mandel2            0.229      mandel2            0.229
ackermann(7)       0.044      ackermann(7)       0.044
ary(50000)         0.010      ary(50000)         0.010
ary2(50000)        0.008      ary2(50000)        0.008
ary3(2000)         0.096      ary3(2000)         0.095
fibo(30)           0.149      fibo(30)           0.148
hash1(50000)       0.016      hash1(50000)       0.016
hash2(500)         0.020      hash2(500)         0.020
heapsort(20000)    0.055      heapsort(20000)    0.054
matrix(20)         0.057      matrix(20)         0.057
nestedloop(12)     0.091      nestedloop(12)     0.091
sieve(30)          0.032      sieve(30)          0.032
strcat(200000)     0.010      strcat(200000)     0.010
------------------------      ------------------------
Total              1.199      Total              1.197

    php$ ./sapi/cli/php Zend/micro_bench.php

Base:                                      Patch:
empty_loop         0.051                   empty_loop         0.051
func()             0.181    0.130          func()             0.181    0.130
undef_func()       0.186    0.135          undef_func()       0.186    0.135
int_func()         0.116    0.064          int_func()         0.116    0.064
$x = self::$x      0.235    0.183          $x = self::$x      0.233    0.182
self::$x = 0       0.198    0.147          self::$x = 0       0.198    0.147
isset(self::$x)    0.229    0.178          isset(self::$x)    0.229    0.178
empty(self::$x)    0.231    0.180          empty(self::$x)    0.231    0.180
$x = Foo::$x       0.144    0.093          $x = Foo::$x       0.144    0.093
Foo::$x = 0        0.107    0.056          Foo::$x = 0        0.107    0.056
isset(Foo::$x)     0.140    0.088          isset(Foo::$x)     0.140    0.088
empty(Foo::$x)     0.148    0.097          empty(Foo::$x)     0.148    0.097
self::f()          0.238    0.187          self::f()          0.238    0.187
Foo::f()           0.209    0.158          Foo::f()           0.209    0.158
$x = $this->x      0.123    0.072          $x = $this->x      0.123    0.072
$this->x = 0       0.124    0.073          $this->x = 0       0.124    0.073
$this->x += 2      0.151    0.099          $this->x += 2      0.153    0.101
++$this->x         0.137    0.086          ++$this->x         0.138    0.086
--$this->x         0.137    0.086          --$this->x         0.138    0.086
$this->x++         0.170    0.119          $this->x++         0.172    0.121
$this->x--         0.171    0.119          $this->x--         0.172    0.121
isset($this->x)    0.170    0.119          isset($this->x)    0.170    0.119
empty($this->x)    0.179    0.128          empty($this->x)    0.179    0.128
$this->f()         0.194    0.143          $this->f()         0.194    0.143
$x = Foo::TEST     0.188    0.137          $x = Foo::TEST     0.188    0.136
new Foo()          0.482    0.431          new Foo()          0.479    0.427
$x = TEST          0.109    0.058          $x = TEST          0.109    0.058
$x = $_GET         0.190    0.138          $x = $_GET         0.190    0.139
$x = $GLOBALS['v'] 0.242    0.191          $x = $GLOBALS['v'] 0.242    0.191
$x = $hash['v']    0.196    0.145          $x = $hash['v']    0.196    0.145
$x = $str[0]       0.146    0.094          $x = $str[0]       0.145    0.094
$x = $a ?: null    0.144    0.093          $x = $a ?: null    0.144    0.093
$x = $f ?: tmp     0.174    0.123          $x = $f ?: tmp     0.174    0.123
$x = $f ? $f : $a  0.153    0.101          $x = $f ? $f : $a  0.153    0.101
$x = $f ? $f : tmp 0.148    0.097          $x = $f ? $f : tmp 0.148    0.097
------------------------                   ------------------------
Total              6.143                   Total              6.143

1 files changed, 4 insertions, 0 deletions

diff --git a/configure.ac b/configure.ac
index 199e2435a5..80628a5df1 100644
--- a/configure.ac
+++ b/configure.ac
@@ -737,7 +737,11 @@ dnl Check for asm goto support
 AC_CACHE_CHECK([for asm goto], ac_cv__asm_goto,
 [AC_RUN_IFELSE([AC_LANG_SOURCE([[
 int main(void) {
+#if defined(__x86_64__) || defined(__i386__)
     __asm__ goto("jmp %l0\n" :::: end);
+#elif defined(__aarch64__)
+    __asm__ goto("b %l0\n" :::: end);
+#endif
 end:
     return 0;
 }