[libc++] Optimize / partially inline basic_string copy constructor

Splits copy constructor up inlining short initialization, outlining long
initialization into __init_long() which is the externally instantiated slow
path initialization.

Subsequently changing the copy ctor to be inlined (not externally instantiated)
provides significant speed ups for short string initialization.

Generated code given:

void StringCopyCtor(void* mem, const std::string& s) {
    std::string*p = new(mem) std::string{s};
}

asm:
        cmp     byte ptr [rsi + 23], 0
        js      .LBB0_2
        mov     rax, qword ptr [rsi + 16]
        mov     qword ptr [rdi + 16], rax
        movups  xmm0, xmmword ptr [rsi]
        movups  xmmword ptr [rdi], xmm0
        ret
.LBB0_2:
        jmp     std::basic_string::__init_long # TAILCALL

Benchmark:
BM_StringCopy_Empty                                           5.19ns ± 6%             1.50ns ± 8%  -71.02%        (p=0.000 n=10+10)
BM_StringCopy_Small                                           5.14ns ± 8%             1.53ns ± 7%  -70.17%        (p=0.000 n=10+10)
BM_StringCopy_Large                                           18.9ns ± 0%             19.3ns ± 0%   +1.92%        (p=0.000 n=10+10)
BM_StringCopy_Huge                                             309ns ± 1%              316ns ± 5%     ~            (p=0.633 n=8+10)

Patch from Martijn Vels ([email protected])
Reviewed as D72160.

Author: EricWF

libcxx/include/string

@@ -1549,6 +1549,11 @@ private:
     inline
     void __init(size_type __n, value_type __c);
 
+    // Identical to __init(s, sz), except that this function is always
+    // externally instantiated and not inlined: this function is the
+    // slow path for the (inlined) copy constructor.
+    void __init_long_external(const value_type* __s, size_type __sz);
+
     template <class _InputIterator>
     inline
     _EnableIf
@@ -1797,6 +1802,18 @@ basic_string<_CharT, _Traits, _Allocator>::__init(const value_type* __s, size_ty
     traits_type::assign(__p[__sz], value_type());
 }
 
+template <class _CharT, class _Traits, class _Allocator>
+void basic_string<_CharT, _Traits, _Allocator>::__init_long_external(
+    const _CharT* __s, size_type __sz) {
+  size_type __cap = __recommend(__sz);
+  pointer __p = __alloc_traits::allocate(__alloc(), __cap + 1);
+  __set_long_pointer(__p);
+  __set_long_cap(__cap + 1);
+  __set_long_size(__sz);
+  traits_type::copy(_VSTD::__to_address(__p), __s, __sz);
+  traits_type::assign(__p[__sz], value_type());
+}
+
 template <class _CharT, class _Traits, class _Allocator>
 template <class>
 basic_string<_CharT, _Traits, _Allocator>::basic_string(const _CharT* __s, const _Allocator& __a)
@@ -1840,7 +1857,7 @@ basic_string<_CharT, _Traits, _Allocator>::basic_string(const basic_string& __st
     if (!__str.__is_long())
         __r_.first().__r = __str.__r_.first().__r;
     else
-        __init(_VSTD::__to_address(__str.__get_long_pointer()), __str.__get_long_size());
+        __init_long_external(_VSTD::__to_address(__str.__get_long_pointer()), __str.__get_long_size());
 #if _LIBCPP_DEBUG_LEVEL >= 2
     __get_db()->__insert_c(this);
 #endif
@@ -1854,7 +1871,7 @@ basic_string<_CharT, _Traits, _Allocator>::basic_string(
     if (!__str.__is_long())
         __r_.first().__r = __str.__r_.first().__r;
     else
-        __init(_VSTD::__to_address(__str.__get_long_pointer()), __str.__get_long_size());
+        __init_long_external(_VSTD::__to_address(__str.__get_long_pointer()), __str.__get_long_size());
 #if _LIBCPP_DEBUG_LEVEL >= 2
     __get_db()->__insert_c(this);
 #endif