[Boost-users] [boost.numeric] Poor Performance of numeric_cast

95 views
Skip to first unread message

Tang Jiang Jun

unread,
Oct 15, 2012, 4:16:56 AM10/15/12
to boost...@lists.boost.org
Hi,

I have run a performance testing for numeric_cast recently, and found that the result was really unexpected bad, although the document mentioned that it will be no overhead if overflows don't happen.
Could somebody please help me to verify this testing? If this is true, I doubt whether I should use numeric_cast in the production code.

Here is my testing code and result.

#include <boost/numeric/conversion/cast.hpp>
#include <boost/format.hpp>
#include <boost/cstdint.hpp>
#include <boost/chrono.hpp>
#include <iostream>

using namespace std;
using namespace boost;
using namespace boost::numeric;
using namespace boost::chrono;


int main()
{
    const static int32_t COUNT = 1000000;
    high_resolution_clock::time_point start;

    start = high_resolution_clock::now();
    for( int32_t n = 0; n < COUNT; ++n )
    {
        int32_t i32 = 123;
        int16_t i16 = i32;
    }
    cout << format("Native Integer Cast: %1%\n") % ( ( high_resolution_clock::now() - start ) / COUNT );

    start = high_resolution_clock::now();
    for( int32_t n = 0; n < COUNT; ++n )
    {
        try
        {
            int32_t i32 = 100;
            int16_t i16 = numeric_cast< int16_t >( i32 );
        }
        catch( const bad_numeric_cast& e )
        {
            cout << e.what() << endl;
        }
    }
    cout << format("Boost Integer Cast: %1%\n") % ( ( high_resolution_clock::now() - start ) / COUNT );

    start = high_resolution_clock::now();
    for( int32_t n = 0; n < COUNT; ++n )
    {
        float f = 100.0f;
        int32_t i = static_cast< int32_t >( f );
    }
    cout << format("Native Floating-Integer Cast: %1%\n") % ( ( high_resolution_clock::now() - start ) / COUNT );

    start = high_resolution_clock::now();
    for( int32_t n = 0; n < COUNT; ++n )
    {
        try
        {
            float f = 123.0f;
            int32_t i = numeric_cast< int32_t >( f );
        }
        catch( const bad_numeric_cast& e )
        {
            cout << e.what() << endl;
        }
    }
    cout << format("Boost Floating-Integer Cast: %1%\n") % ( ( high_resolution_clock::now() - start ) / COUNT );

    start = high_resolution_clock::now();
    for( int32_t n = 0; n < COUNT; ++n )
    {
        int32_t i = 132;
        float f = static_cast< float >( i );
    }
    cout << format("Native Integer-Floating Cast: %1%\n") % ( ( high_resolution_clock::now() - start ) / COUNT );

    start = high_resolution_clock::now();
    for( int32_t n = 0; n < COUNT; ++n )
    {
        try
        {
            int32_t i = 128;
            float f = numeric_cast< float >( i );
        }
        catch( const bad_numeric_cast& e )
        {
            cout << e.what() << endl;
        }
    }
    cout << format("Boost Integer-Floating Cast: %1%\n") % ( ( high_resolution_clock::now() - start ) / COUNT );

    return 0;
};



Result:
Native Integer Cast: 3 nanoseconds
Boost Integer Cast: 311 nanoseconds
Native Floating-Integer Cast: 4 nanoseconds
Boost Floating-Integer Cast: 430 nanoseconds
Native Integer-Floating Cast: 2 nanoseconds
Boost Integer-Floating Cast: 106 nanoseconds

Oswin Krause

unread,
Oct 15, 2012, 4:30:24 AM10/15/12
to boost...@lists.boost.org
Hi,

Your complete loop got optimized away in the native test cases. Because
of the try/catch block the compiler couldn't do this in the other cases.
So you are benchmarking nothing vs somthing.

Greetings,
Oswin

>     cout << format("Native Integer Cast: %1%n") % ( (


> high_resolution_clock::now() - start ) / COUNT );
>
>     start = high_resolution_clock::now();
>     for( int32_t n = 0; n < COUNT; ++n )
>     {
>         try
>         {
>             int32_t i32 = 100;
>             int16_t i16 = numeric_cast< int16_t >( i32 );
>         }
>         catch( const bad_numeric_cast& e )
>         {
>             cout << e.what() << endl;
>         }
>     }

>     cout << format("Boost Integer Cast: %1%n") % ( (


> high_resolution_clock::now() - start ) / COUNT );
>
>     start = high_resolution_clock::now();
>     for( int32_t n = 0; n < COUNT; ++n )
>     {
>         float f = 100.0f;
>         int32_t i = static_cast< int32_t >( f );
>     }

>     cout << format("Native Floating-Integer Cast: %1%n") % ( (


> high_resolution_clock::now() - start ) / COUNT );
>
>     start = high_resolution_clock::now();
>     for( int32_t n = 0; n < COUNT; ++n )
>     {
>         try
>         {
>             float f = 123.0f;
>             int32_t i = numeric_cast< int32_t >( f );
>         }
>         catch( const bad_numeric_cast& e )
>         {
>             cout << e.what() << endl;
>         }
>     }

>     cout << format("Boost Floating-Integer Cast: %1%n") % ( (


> high_resolution_clock::now() - start ) / COUNT );
>
>     start = high_resolution_clock::now();
>     for( int32_t n = 0; n < COUNT; ++n )
>     {
>         int32_t i = 132;
>         float f = static_cast< float >( i );
>     }

>     cout << format("Native Integer-Floating Cast: %1%n") % ( (


> high_resolution_clock::now() - start ) / COUNT );
>
>     start = high_resolution_clock::now();
>     for( int32_t n = 0; n < COUNT; ++n )
>     {
>         try
>         {
>             int32_t i = 128;
>             float f = numeric_cast< float >( i );
>         }
>         catch( const bad_numeric_cast& e )
>         {
>             cout << e.what() << endl;
>         }
>     }

>     cout << format("Boost Integer-Floating Cast: %1%n") % ( (


> high_resolution_clock::now() - start ) / COUNT );
>
>     return 0;
> };
>
> Result:
> Native Integer Cast: 3 nanoseconds
> Boost Integer Cast: 311 nanoseconds
> Native Floating-Integer Cast: 4 nanoseconds
> Boost Floating-Integer Cast: 430 nanoseconds
> Native Integer-Floating Cast: 2 nanoseconds
> Boost Integer-Floating Cast: 106 nanoseconds

_______________________________________________
Boost-users mailing list
Boost...@lists.boost.org
http://lists.boost.org/mailman/listinfo.cgi/boost-users

Tang Jiang Jun

unread,
Oct 15, 2012, 5:29:18 AM10/15/12
to boost...@lists.boost.org
Hi Oswin,

Sorry, I forgot to mention that I compiled it as debug configuration in order to prevent unintended optimization.
Anyway, many thanks for reminding!

Tang

Oswin Krause

unread,
Oct 15, 2012, 6:43:10 AM10/15/12
to boost...@lists.boost.org
Hi,

Never benchmark in debug mode. Moreover, never ever benchmark boost
code in debug mode.

>> Boost...@lists.boost.org [1]
>> http://lists.boost.org/mailman/listinfo.cgi/boost-users [2]
>
>
>
> Links:
> ------
> [1] mailto:Boost...@lists.boost.org
> [2] http://lists.boost.org/mailman/listinfo.cgi/boost-users
> [3] mailto:Oswin....@ruhr-uni-bochum.de

Tang Jiang Jun

unread,
Oct 15, 2012, 11:50:36 PM10/15/12
to boost...@lists.boost.org
Hi,

I modify my code to make it can run in release mode without unintended optimization, and now the performance is acceptable. However there definitely has some runtime overhead even no overflow happens, and the overhead takes extra time as much as the plain cast itself takes. I think this maybe should be mentioned in the numeric_cast document, because if the cast is the core step in an algorithms and is executed heavily, this overhead will impact the performance significantly .

The following is the benchmark code after modification and the result run in my computer.

CODE
#include <boost/numeric/conversion/cast.hpp>

#include <boost/format.hpp>
#include <boost/cstdint.hpp>
#include <boost/chrono.hpp>
#include <iostream>

using namespace std;
using namespace boost;
using namespace boost::numeric;
using namespace boost::chrono;


typedef void (*PROFILE_FUNC)( uint32_t, uint32_t& );

nanoseconds profile( PROFILE_FUNC _profileFunc, uint32_t _count )
{
    high_resolution_clock::time_point start = high_resolution_clock::now();

    uint32_t sum = 0;
    _profileFunc( _count, sum );

    nanoseconds ns = ( high_resolution_clock::now() - start ) / _count;

    cout << sum << endl;

    return ns;
}


void native_integer_cast( uint32_t _count, uint32_t& _sum )
{
    for( uint64_t n = 0; n < _count; ++n )
    {
        _sum += static_cast< uint32_t >( n );
    }
}


void boost_integer_cast( uint32_t _count, uint32_t& _sum )
{
    for( uint64_t n = 0; n < _count; ++n )
    {
        try
        {
            _sum += numeric_cast< uint32_t >( n );

        }       
        catch( const bad_numeric_cast& e )
        {
            cout << e.what() << endl;
        }
    }
}


void native_itof_cast( uint32_t _count, uint32_t& _sum )
{
    float fsum = 0.0f;

    for( uint32_t n = 0; n < _count; ++n )
    {
        fsum += static_cast< float >( n );
    }

    _sum = static_cast< uint32_t >( fsum );
}


void boost_itof_cast( uint32_t _count, uint32_t& _sum )
{
    float fsum = 0.0f;

    for( uint32_t n = 0; n < _count; ++n )
    {
        try
        {
            fsum += numeric_cast< float >( n );

        }
        catch( const bad_numeric_cast& e )
        {
            cout << e.what() << endl;
        }
    }

    _sum = numeric_cast< uint32_t >( fsum );
}


void native_ftoi_cast( uint32_t _count, uint32_t& _sum )
{
    for( float f = 0.0f; f < _count; f += 1.0f )
    {
        _sum += static_cast< uint32_t >( f );
    }
}


void boost_ftoi_cast( uint32_t _count, uint32_t& _sum )
{
    for( float f = 0.0f; f < _count; f += 1.0f )
    {
        try
        {
            _sum += numeric_cast< uint32_t >( f );

        }
        catch( const bad_numeric_cast& e )
        {
            cout << e.what() << endl;
        }
    }
}



int main()
{
    const static int32_t COUNT = 10000;
   
    nanoseconds nsNativeIntegerCast = profile( native_integer_cast, COUNT );
    nanoseconds nsBoostIntegerCast = profile( boost_integer_cast, COUNT );
    nanoseconds nsNativeItofCast = profile( native_itof_cast, COUNT );
    nanoseconds nsBoostItofCast = profile( boost_itof_cast, COUNT );
    nanoseconds nsNativeFtoiCast = profile( native_ftoi_cast, COUNT );
    nanoseconds nsBoostFtoiCast = profile( boost_ftoi_cast, COUNT );

    cout << "Native Integer Cast: " << nsNativeIntegerCast << endl;
    cout << "Boost Integer Cast: " << nsBoostIntegerCast << endl;
    cout << "Native Integer-Floating Cast: " << nsNativeItofCast << endl;
    cout << "Boost Integer-Floating Cast: " << nsBoostItofCast << endl;
    cout << "Native Floating-Integer Cast: " << nsNativeFtoiCast << endl;
    cout << "Boost Floating-Integer Cast: " << nsBoostFtoiCast << endl;

    return 0;
};


RESULT:
Native Integer Cast: 1 nanosecond
Boost Integer Cast: 4 nanoseconds
Native Integer-Floating Cast: 3 nanoseconds
Boost Integer-Floating Cast: 3 nanoseconds
Native Floating-Integer Cast: 5 nanoseconds
Boost Floating-Integer Cast: 14 nanoseconds

Regards,
Tang

[1] mailto:Boost-users@lists.boost.org
[2] http://lists.boost.org/mailman/listinfo.cgi/boost-users
[3] mailto:Oswin.Krause@ruhr-uni-bochum.de

Brian Budge

unread,
Oct 16, 2012, 12:08:18 AM10/16/12
to boost...@lists.boost.org
Unsure, but maybe you should put the try/catch outside of the inner loop?
>>> [1] mailto:Boost...@lists.boost.org
>>> [2] http://lists.boost.org/mailman/listinfo.cgi/boost-users
>>> [3] mailto:Oswin....@ruhr-uni-bochum.de

Tang Jiang Jun

unread,
Oct 16, 2012, 12:45:07 AM10/16/12
to boost...@lists.boost.org
I've already tried to remove all the try-catch blocks, but the overhead is still there.
On the other side, when casting a number, it is general to use a dedicated try-catch block to protect it.

Oswin Krause

unread,
Oct 16, 2012, 1:24:55 AM10/16/12
to boost...@lists.boost.org
Hi,

the results turnd out to have a high variance due to the low time
usage. SInce just choosing higher count numbers already lead to an
overflow, i hacked in the following loop:

nanoseconds profile( PROFILE_FUNC _profileFunc, uint32_t _count )
{
high_resolution_clock::time_point start = high_resolution_clock::now();

double summ = 0;
for(std::size_t i = 0; i != 100000; ++i){
uint32_t sum = i;
_profileFunc( _count, sum );
summ +=sum;
}

nanoseconds ns = ( high_resolution_clock::now() - start ) / _count;

cout << summ << endl;

return ns;
}


results:

Native Integer Cast: 26729 nanoseconds
Boost Integer Cast: 26449 nanoseconds
Native Integer-Floating Cast: 105479 nanoseconds
Boost Integer-Floating Cast: 105455 nanoseconds
Native Floating-Integer Cast: 168933 nanoseconds
Boost Floating-Integer Cast: 453505 nanoseconds


so no overhead in Integer-Integer or Integer-Floating. But
Floating-Integer has bad performance.

>> >>>> Boost...@lists.boost.org [4] [1]
>> >>>> http://lists.boost.org/mailman/listinfo.cgi/boost-users [5]
>> [2]
>> >>>
>> >>>
>> >>>
>> >>>
>> >>> Links:
>> >>> ------
>> >>> [1] mailto:Boost...@lists.boost.org [6]
>> >>> [2] http://lists.boost.org/mailman/listinfo.cgi/boost-users [7]
>> >>> [3] mailto:Oswin....@ruhr-uni-bochum.de [8]
>> >>
>> >>
>> >> _______________________________________________
>> >> Boost-users mailing list
>> >> Boost...@lists.boost.org [9]
>> >> http://lists.boost.org/mailman/listinfo.cgi/boost-users [10]
>> >
>> >
>> >
>> > _______________________________________________
>> > Boost-users mailing list
>> > Boost...@lists.boost.org [11]
>> > http://lists.boost.org/mailman/listinfo.cgi/boost-users [12]
>> _______________________________________________
>> Boost-users mailing list
>> Boost...@lists.boost.org [13]
>> http://lists.boost.org/mailman/listinfo.cgi/boost-users [14]
>
>
>
> Links:
> ------
> [1] mailto:tangji...@gmail.com
> [2] mailto:Oswin....@ruhr-uni-bochum.de
> [3] mailto:Oswin....@ruhr-uni-bochum.de
> [4] mailto:Boost...@lists.boost.org
> [5] http://lists.boost.org/mailman/listinfo.cgi/boost-users
> [6] mailto:Boost...@lists.boost.org
> [7] http://lists.boost.org/mailman/listinfo.cgi/boost-users
> [8] mailto:Oswin....@ruhr-uni-bochum.de
> [9] mailto:Boost...@lists.boost.org
> [10] http://lists.boost.org/mailman/listinfo.cgi/boost-users
> [11] mailto:Boost...@lists.boost.org
> [12] http://lists.boost.org/mailman/listinfo.cgi/boost-users
> [13] mailto:Boost...@lists.boost.org
> [14] http://lists.boost.org/mailman/listinfo.cgi/boost-users
> [15] mailto:brian...@gmail.com

Tang Jiang Jun

unread,
Oct 16, 2012, 3:47:55 AM10/16/12
to boost...@lists.boost.org
Hi,

I adopted your suggestion to run the inner cast for 10000 * 10000 times, and the result is same on my computer. I guess maybe the difference is caused by the architecture of our CPU, and my CPU is intel i3.

Here is the result.
Native Integer Cast: 2 nanoseconds
Boost Integer Cast: 4 nanoseconds
Native Integer-Floating Cast: 3 nanoseconds
Boost Integer-Floating Cast: 3 nanoseconds
Native Floating-Integer Cast: 5 nanoseconds
Boost Floating-Integer Cast: 15 nanoseconds



>>
>>
>> _______________________________________________
>> Boost-users mailing list
>> Boost...@lists.boost.org [9]
>> http://lists.boost.org/mailman/listinfo.cgi/boost-users [10]
>
>
>
> _______________________________________________
> Boost-users mailing list
> Boost...@lists.boost.org [11]
> http://lists.boost.org/mailman/listinfo.cgi/boost-users [12]
_______________________________________________
Boost-users mailing list
Boost...@lists.boost.org [13]
http://lists.boost.org/mailman/listinfo.cgi/boost-users [14]



Links:
------
[1] mailto:tangji...@gmail.com

Oswin Krause

unread,
Oct 16, 2012, 3:55:25 AM10/16/12
to boost...@lists.boost.org
Hi,

is this the total run time? can you show the code? Your run times are
the same as before. you should expect a runtime factor of 1000+ in the
end result - or the compiler was too smart.

On 2012-10-16 09:47, Tang Jiang Jun wrote:
> Hi,
>

>>>> >>>> Boost...@lists.boost.org [4] [4] [1]
>>>> >>>> http://lists.boost.org/mailman/listinfo.cgi/boost-users
>>>> [5] [5]
>>>> [2]
>>>> >>>
>>>> >>>
>>>> >>>
>>>> >>>
>>>> >>> Links:
>>>> >>> ------
>>>> >>> [1] mailto:Boost...@lists.boost.org [6] [6]
>>>> >>> [2] http://lists.boost.org/mailman/listinfo.cgi/boost-users
>>>> [7] [7]
>>>> >>> [3] mailto:Oswin....@ruhr-uni-bochum.de [8] [8]
>>>> >>
>>>> >>
>>>> >> _______________________________________________
>>>> >> Boost-users mailing list
>>>> >> Boost...@lists.boost.org [9] [9]
>>>> >> http://lists.boost.org/mailman/listinfo.cgi/boost-users [10]


>>>> [10]
>>>> >
>>>> >
>>>> >
>>>> > _______________________________________________
>>>> > Boost-users mailing list

>>>> > Boost...@lists.boost.org [11] [11]
>>>> > http://lists.boost.org/mailman/listinfo.cgi/boost-users [12]


>>>> [12]
>>>> _______________________________________________
>>>> Boost-users mailing list

>>>> Boost...@lists.boost.org [13] [13]
>>>> http://lists.boost.org/mailman/listinfo.cgi/boost-users [14]
>>>> [14]
>>>
>>> Links:
>>> ------
>>> [1] mailto:tangji...@gmail.com [16]
>>> [2] mailto:Oswin....@ruhr-uni-bochum.de [17]
>>> [3] mailto:Oswin....@ruhr-uni-bochum.de [18]
>>> [4] mailto:Boost...@lists.boost.org [19]
>>> [5] http://lists.boost.org/mailman/listinfo.cgi/boost-users [20]
>>> [6] mailto:Boost...@lists.boost.org [21]
>>> [7] http://lists.boost.org/mailman/listinfo.cgi/boost-users [22]
>>> [8] mailto:Oswin....@ruhr-uni-bochum.de [23]
>>> [9] mailto:Boost...@lists.boost.org [24]
>>> [10] http://lists.boost.org/mailman/listinfo.cgi/boost-users [25]
>>> [11] mailto:Boost...@lists.boost.org [26]
>>> [12] http://lists.boost.org/mailman/listinfo.cgi/boost-users [27]
>>> [13] mailto:Boost...@lists.boost.org [28]
>>> [14] http://lists.boost.org/mailman/listinfo.cgi/boost-users [29]
>>> [15] mailto:brian...@gmail.com [30]
>>
>> _______________________________________________
>> Boost-users mailing list
>> Boost...@lists.boost.org [31]
>> http://lists.boost.org/mailman/listinfo.cgi/boost-users [32]


>
>
>
> Links:
> ------
> [1] mailto:tangji...@gmail.com

> [2] mailto:Oswin....@ruhr-uni-bochum.de
> [3] mailto:Oswin....@ruhr-uni-bochum.de
> [4] mailto:Boost...@lists.boost.org

> [5] http://lists.boost.org/mailman/listinfo.cgi/boost-users
> [6] mailto:Boost...@lists.boost.org
> [7] http://lists.boost.org/mailman/listinfo.cgi/boost-users
> [8] mailto:Oswin....@ruhr-uni-bochum.de
> [9] mailto:Boost...@lists.boost.org
> [10] http://lists.boost.org/mailman/listinfo.cgi/boost-users
> [11] mailto:Boost...@lists.boost.org
> [12] http://lists.boost.org/mailman/listinfo.cgi/boost-users
> [13] mailto:Boost...@lists.boost.org


> [14] http://lists.boost.org/mailman/listinfo.cgi/boost-users
> [15] mailto:brian...@gmail.com

> [16] mailto:tangji...@gmail.com
> [17] mailto:Oswin....@ruhr-uni-bochum.de
> [18] mailto:Oswin....@ruhr-uni-bochum.de
> [19] mailto:Boost...@lists.boost.org
> [20] http://lists.boost.org/mailman/listinfo.cgi/boost-users
> [21] mailto:Boost...@lists.boost.org
> [22] http://lists.boost.org/mailman/listinfo.cgi/boost-users
> [23] mailto:Oswin....@ruhr-uni-bochum.de
> [24] mailto:Boost...@lists.boost.org
> [25] http://lists.boost.org/mailman/listinfo.cgi/boost-users
> [26] mailto:Boost...@lists.boost.org
> [27] http://lists.boost.org/mailman/listinfo.cgi/boost-users
> [28] mailto:Boost...@lists.boost.org
> [29] http://lists.boost.org/mailman/listinfo.cgi/boost-users
> [30] mailto:brian...@gmail.com
> [31] mailto:Boost...@lists.boost.org
> [32] http://lists.boost.org/mailman/listinfo.cgi/boost-users
> [33] mailto:Oswin....@ruhr-uni-bochum.de

John Maddock

unread,
Oct 16, 2012, 4:12:06 AM10/16/12
to boost...@lists.boost.org
> results:
>
> Native Integer Cast: 26729 nanoseconds
> Boost Integer Cast: 26449 nanoseconds
> Native Integer-Floating Cast: 105479 nanoseconds
> Boost Integer-Floating Cast: 105455 nanoseconds
> Native Floating-Integer Cast: 168933 nanoseconds
> Boost Floating-Integer Cast: 453505 nanoseconds
>
>
> so no overhead in Integer-Integer or Integer-Floating. But
> Floating-Integer has bad performance.

That's sort of what I would expect - think about it - if the cast is to the
same or a wider type, then there is no check, and numeric_cast and
static_cast do the same thing. However for a narrowing cast (float to
integer), then at the very least there has to be an extra if statement to
test whether the value being cast is in range - that would normally roughly
double the runtime cost. But there *may* be another hidden cost: depending
on the loop the compiler may decide not to inline the numeric_cast in order
to give a tighter loop, and the cost of the function call would add a big
chunk of time. Plus the extra code associated with the error handling if
the value is out of range adds a certain amount of code bloat to the loop,
reducing code locality. But the thing is you can't avoid this if you want
the runtime check. There's no such thing as a free lunch sadly.

John.

Tang Jiang Jun

unread,
Oct 16, 2012, 4:13:10 AM10/16/12
to boost...@lists.boost.org
No, it's the unit time for each casting. The code is as follows.

nanoseconds profile( PROFILE_FUNC _profileFunc, uint32_t _count )
{
high_resolution_clock::time_point start = high_resolution_clock::now();

uint32_t c = 0;

for( uint32_t n = 0; n < _count; ++n )
{
uint32_t sum = n;
_profileFunc( _count, sum );
if( sum > n )
{
c++;
}
}

nanoseconds ns = ( ( high_resolution_clock::now() - start ) / _count ) / _count;

cout << c << endl;

return ns;
}

I added a trivial counter to prevent the compiler to optimize out the profiling code.


>>> [1] mailto:Boost-users@lists.boost.org [6] [6]
>>> [3] mailto:Oswin.Krause@ruhr-uni-bochum.de [8] [8]

>>
>>
>> _______________________________________________
>> Boost-users mailing list
>> Boost...@lists.boost.org [9] [9]
>> http://lists.boost.org/mailman/listinfo.cgi/boost-users [10]

[10]
>
>
>
> _______________________________________________
> Boost-users mailing list
> Boost...@lists.boost.org [11] [11]
> http://lists.boost.org/mailman/listinfo.cgi/boost-users [12]

[12]
_______________________________________________
Boost-users mailing list
Boost...@lists.boost.org [13] [13]
http://lists.boost.org/mailman/listinfo.cgi/boost-users [14]
[14]

Links:
------
[1] mailto:tangji...@gmail.com [16]

_______________________________________________
Boost-users mailing list
Boost...@lists.boost.org [31]
http://lists.boost.org/mailman/listinfo.cgi/boost-users [32]



Links:
------
[1] mailto:tangji...@gmail.com

Tang Jiang Jun

unread,
Oct 16, 2012, 4:35:03 AM10/16/12
to boost...@lists.boost.org
Very incisive! I checked the assembly code and verified your hypothesis. The integer-to-floating cast of numeric_cast was really inlined by the compiler, meanwhile the other two were using function call to invoke numeric_cast. So maybe the better solution is to force inline for numeric_cast in all cases.
Reply all
Reply to author
Forward
0 new messages