Multiple inheritance performance hit

[Graham Banks]

Does using multiple inheritance introduce any more performance overhead than
single inheritance?

[tom_usenet]

On Wed, 30 Jul 2003 14:32:45 +0100, "Graham Banks"
<gr************ *************** *******@sol-digi.com> wrote:

Does using multiple inheritance introduce any more performance overhead than
single inheritance?

It depends on the implementation, but calling a virtual function where
virtual multiple inheritence is involved will typically be slightly
slower than a normal virtual function. The vtable for a class involved
in multiple inheritence will probably be larger too.

Tom

[Alex Vinokur]

"Graham Banks" <gr************ *************** *******@sol-digi.com> wrote in message news:bg******** **@newsg4.svr.p ol.co.uk...

Does using multiple inheritance introduce any more performance overhead than
single inheritance?

Multiple inheritance vs. single inheritance
=============== =============== =============

C/C++ Performance Tests
=============== ========
Using C/C++ Program Perfometer
http://sourceforge.net/projects/cpp-perfometer
http://alexvn.freeservers.com/s1/perfometer.html
Environment
-----------
Windows 2000 Professional
CYGWIN_NT-5.0 1.3.22(0.78/3/2)
Intel(R) Celeron(R) CPU 1.70 GHz
GNU gcc/g++ version 3.2 20020927 (prerelease)
Compilation : No optimization

=============== ====== Classes : BEGIN =============== ======

=============== =====
Hierarchy of classes
=============== =====
Foo0
|
|
Foo1
|
|
_____________|_ ____________
| |
__________|____ ______ |
| | |
| | |
Foo2_v1 Foo2_v2 Foo2
| | |
|______________ _____| |
| |
| |
Foo3_m Foo3
| |
| |
Foo4_m Foo4

=============== ======
Definition of classes
=============== ======

// --------------
class Foo0
{
public:
Foo0 () {}
};

// --------------
class Foo1 : public Foo0
{
public:
Foo1 () : Foo0 () {}
};

// --------------
class Foo2_v1 : virtual public Foo1
{
public:
Foo2_v1 () : Foo1 () {}

};

// --------------
class Foo2_v2 : virtual public Foo1
{
public:
Foo2_v2 () : Foo1 () {}
};
// --------------
class Foo2 : public Foo1
{
public:
Foo2 () : Foo1 () {}

};

// --------------
class Foo3_m : public Foo2_v1, public Foo2_v2
{
public:
Foo3_m () : Foo1 (), Foo2_v1 (), Foo2_v2 () {}

};

// --------------
class Foo3 : public Foo2
{
public:
Foo3 () : Foo2 () {}

};

// --------------
class Foo4_m : public Foo3_m
{
public:
Foo4_m () : Foo1 (), Foo3_m () {}
};

// --------------
class Foo4 : public Foo3
{
public:
Foo4 () : Foo3 () {}
};

=============== ====== Classes : END =============== ========

=============== = Performance tests : BEGIN =============== =

=============== ===
Creating instances
=============== ===
Summary test results
=============== =====
#----------------------------------------------------------
# Comparison : multiple inheritance vs. single inheritance
#----------------------------------------------------------
# Resource Name : user time used (via rusage)
# Resource Cost Unit : milliseconds (unsigned long long)
# Resource State Unit : timeval
#----------------------------------------------------------
: ---------------------------------------------
: Foo0 : Level-0 base class -> 9
:
: Foo1 : Level-1 single inheritance -> 13
:
: Foo2 : Level-2 single inheritance -> 20
: Foo2_v1 : Level-2 virtual inheritance -> 20
: Foo2_v2 : Level-2 virtual inheritance -> 20
:
: Foo3 : Level-3 single inheritance -> 25
: Foo3_m : Level-3 multiple inheritance -> 42
:
: Foo4 : Level-4 single inheritance -> 32
: Foo4_m : Level-4 multiple inheritance -> 45
: ---------------------------------------------
Raw Log : http://groups.google.com/groups?selm....uni-berlin.de

=============== = Performance tests : END =============== ===
Regards,
=============== =============== =======
Alex Vinokur
mailto:al****@c onnect.to
http://mathforum.org/library/view/10978.html
=============== =============== =======