- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello
We have a multi-threaded app written in C# end C++ managed and not managed running on Windows with OpenMp lib.
we build a custom dll for ipp functions.
my Pc target is an i7 quad core with hyperthreading (8 active core)
but at run time, when the ipp function like ippiResize_16u_C3R or ippiResize_16u_C1R is called only one core is using.
i don't understand why the ipp function not use all of my pc core.
i tried to link my custom dll static mode or dynamic mode with sub lib
but only one core is used.
have you got an answer ?
We have a multi-threaded app written in C# end C++ managed and not managed running on Windows with OpenMp lib.
we build a custom dll for ipp functions.
my Pc target is an i7 quad core with hyperthreading (8 active core)
but at run time, when the ipp function like ippiResize_16u_C3R or ippiResize_16u_C1R is called only one core is using.
i don't understand why the ipp function not use all of my pc core.
i tried to link my custom dll static mode or dynamic mode with sub lib
but only one core is used.
have you got an answer ?
Link Copied
2 Replies
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
This question would be more likely to get an answer on the IPP forum, if it's not covered in IPP docs. The performance library functions would be threaded only if a gain has been demonstrated with threading.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi a.malagouenamesys.fr,
Do you have some source code to share? Without the source code, it is gonna be difficult to find the real problem. Some snippet could help.
Cheers,
Gastn

Reply
Topic Options
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page