extend rb_call_cache

Prior to this changeset, majority of inline cache mishits resulted
into the same method entry when rb_callable_method_entry() resolves
a method search.  Let's not call the function at the first place on
such situations.

In doing so we extend the struct rb_call_cache from 44 bytes (in
case of 64 bit machine) to 64 bytes, and fill the gap with
secondary class serial(s).  Call cache's class serials now behavies
as a LRU cache.

Calculating -------------------------------------
                           ours         2.7         2.6
vm2_poly_same_method     2.339M      1.744M      1.369M i/s - 6.000M times in 2.565086s 3.441329s 4.381386s

Comparison:
             vm2_poly_same_method
                ours:   2339103.0 i/s
                 2.7:   1743512.3 i/s - 1.34x  slower
                 2.6:   1369429.8 i/s - 1.71x  slower
This commit is contained in:
卜部昌平 2019-10-07 12:59:57 +09:00
parent 3c252651e1
commit d45a013a1a
Notes: git 2019-11-07 17:41:56 +09:00
6 changed files with 75 additions and 19 deletions

View file

@ -87,7 +87,7 @@ has_valid_method_type(CALL_CACHE cc)
{
extern bool mjit_valid_class_serial_p(rb_serial_t class_serial);
return GET_GLOBAL_METHOD_STATE() == cc->method_state
&& mjit_valid_class_serial_p(cc->class_serial) && cc->me;
&& mjit_valid_class_serial_p(cc->class_serial[0]) && cc->me;
}
// Returns true if iseq can use fastpath for setup, otherwise NULL. This becomes true in the same condition