This is the mail archive of the gcc-bugs@gcc.gnu.org mailing list for the GCC 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]

[Bug libfortran/23363] gfortran 30 x slower that g77 on random I/O



------- Comment #9 from jblomqvi at cc dot hut dot fi  2005-10-09 17:41 -------
(In reply to comment #8)
> Can you do timings with today's compiler?
> 

Performance is now equivalent with g77, a factor of 500 times faster than
mainline a month ago. I don't know if any of these patches have made it into
4.0.

The actual timings: 

Original timing (with mmap): 13.6 s.

Getting rid of mmap (committed 2005-09-10 or thereabouts): 0.77 s.

transfer_array (committed 2005-09-24): 0.040 s

sread/swrite (the patch committed 2005-10-07): 0.026 s.

g77: 0.027 s

ifort 9.0: 0.153 s.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23363


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