This is the mail archive of the fortran@gcc.gnu.org mailing list for the GNU Fortran project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [Fwd: Re: Polyhedron benchmark on Opteron]


(Trying my fourth and last email address, all others are blocked after I
received a mailserver reply "spam score exceeded threshold" (the last
one for this email); I also cannot reach any postmaster@ and
global-allow-subscribe-<my email=address>@sourceware.org does not work
either.)


Hi Steve,

Steve Kargl wrote:
>> I don't understand it either
>>     
>    ISET = MINLOC(DTEMP)
>
> I'll need to go read the code, but I would have thought you
> would need some like
>
>    tmp = huge(dtemp)
>    do i = 1, nodes
>       if (tmp > dtemp(i)) then
>          tmp = dtemp(i)   ! Small current smallest value for next comparison.
>          iset(1) = i      ! Save the current index.
>       end if
>    end do
>   
Ups, I shouldn't code after midnight. This makes more sense:

do-loop: 14.766s (real), user 14.589s
minloc: 15.376s,  15.069s
=> ~4% gain,
which is not too bad for something simple, but whether inlining is
worth, I'm not sure.

Intel:
do-loop: 8.527s, 8.321s
minloc: 8.429s, 8.265s
=> 1% gain

Tobias


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]