how to use "private" ?

[ejd]

i will try to post little "pseudo"-code:

myclass * myvar = function(); // this function returns a pointer to a myclass-Object
...
int i;
int n = 1000;
int z = 2;

#pragma omp parallel for private(i) shared(myvar, n) firstprivate(z) default(shared)
for (i = 1; i <n; i++)
{
api_function(myvar); // its an api-function from the CAD-API ACIS which reads the myclass-variable
}

Info: if you now little about the CAD-API ACIS: i use a pointer to a BODY- and a FACE-object

it's sure that the api_function only reads the myclass-variable!!!

I assume there must be something more to this than what you are showing, because calling api_function to do the same thing "n" times doesn't really make any sense to me. In any case, if the api_function is only reading myvar, then this should work IF the api_function is "threadsafe". That is, it may be keeping global information that if accessed by multiple threads will cause a data race. You need to look at the documentation for the CAD-API and see what it says about this issue.

[roosevelt]

thanks very much. I already understand this problem.

[Guest]

hello,

thanks for your example! it helps me alot to understand the private and firstprivate-clause with pointers.

the problem that my program didnt work was a problem that a variable causes a read error
therefore i wondered because i thought that parallel reads are no problem.
now i found out that its not a problem cuased by openmp.
its one problem from the function that i call because the function belongs to an api which is not threadsafe.

thanks for ur help. This board is really a super place to talkabout openmp and to improve my openmp skills.
thanks.

[peter]

Hello,

as i see now: i forgot to write my username in my post before.
therefore i want to tell you that the post before was from me.
thanks

[boost_fan]

I have questions on OpenMP and C++ asked by many people on this thread but I notice the scarcity of any meaningful replies. This could be because not many people use this platform. Perhaps then I should continue using boost threads where there is much help.