Sorry to ask a daft question, you have RH9 and Intel icc 7.1 working ?
Well what command did you use to get rid of those errors that come up with those __ctype_h sort ?
I tried this code:
#include <iostream>
using namespace std;
int main()
{
cout << "Test icc \n";
return 0;
}and the command icc -o aly aly.cc and I get errors. If you have it working then please send me the command line sequence.
Surprisingly the binary size on a RH8.0 box with icc vrs g++, icc has a larger binary size. Interesting.
Cheers,
Aly.
Avijit Ghosh wrote:
On Wed, 4 Jun 2003, Avijit Ghosh wrote:
Out of curiousity are you linking code compiled w/ the intel compilers? This is the error message I got when I moved to redhat 9 for our code based on intel compiler. If you link w/ -i_dynamic, it removes
this error message. In fact this might be a plausible work around
if your code compiles/can use intel's compilers as presumably your
3rd party libs can use implicit __ctype_h that intel is incorporating
into the lib.
I forgot to add, if you are writing gpl or "open source" software the intel compilers are free . (I dont have the link handy at the moment). Somewhat surprisingly, I get a literal 100% code speed up w/ C++ code in comparison to g++ w/ numerically intensive code (rh9). (The numerically intensive bit is "c-ish" in style.) One day in a fit
of mania, I'm going to try recompiling all the kde rh9 rpm's using
intel and see what happens :)
-avi
-- Aly S.P Dharshi [EMAIL PROTECTED] Student System Administrator/Network Analyst LDAP Project Department of Computer Science and Mathematics University of Lethbridge
"A good speech is like a good dress that's short enough to be interesting and long enough to cover the subject"
-- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
