This is the mail archive of the gcc-help@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]

Re: program executin times differ between gcc versions


Hi Paul,

And what about volatile variables?
If you have variables that might be modified by multiple threads you should declare them as volatiles in order to prevent compiler optmization and wrong execution.


Marco


Tim Prince wrote:
Joel Dice wrote:
On Thu, 1 Nov 2007, Paul Moore wrote:

Dear List,

<snip>


Using optimization flags is not possible due to the fact that large
parts of the application failed in -O test runs and would need to be
fixed/tested extensively first. Profiling with -pg/gprof has not been
produced usable results as shared libraries are not being tracked.

In my experience, OProfile (oprofile.sourceforge.net) is much better than gprof at revealing the hotspots in my programs, including time spent in shared libraries. I strongly recommend it.
When hotspots occur in libraries, gprof is satisfactory only with static linking. Library questions seem unlikely to be relevant in this thread.

OTOH, I would worry more about correctness than performance at this point if the code breaks with optimization turned on.


That's for sure. gcc is unlikely to break reliable code with optimization, unless -ffast-math is set.


---AV & Spam Filtering by M+Guardian - Risk Free Email (TM)---





-- Marco Aurélio da Cruz

Depto. de Automação
Divisão de Tecnologia
Daruma Urmet S.A.
Tel.: +55 12 3609-5105
Fax:  +55 12 3609-5156
Web:  http://www.daruma.com.br


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