I work with:
1 - Windows 7 Professional - 64 bits;
2 - Cuda 5.0;
3 - cusparse and cublas;
4 - Visual Studio.
5 - GPU: GTX 690 (2 devices).
In this case, It is possible work with GPUDirect Peer To Peer Transfers And Memory Access?
I tested cudaDeviceCanAccessPeer() for 2 devices and the output was 0. Why it is occurs?
Thanks
Your system will not support P2P. You need either of the following
1: Linux OS
OR
2: Tesla cards with the TCC drivers installed. I don’t believe the GTX 690 supports the TCC drivers.
Hi,
I read that Windows XP is possible use peer to peer memory acess and memory copy, such as:
I am running Windows 7 x64 using the GeForce GTX 690 card for a CUDA project. As a dual card, deviceQuery recognizes the card as two separate GPU’s, each with 2Gb globabl memory. I would like to be able to share the memory between the two GPU’s, to...
What do I do? I have to install new program?
Thanks.
The post you linked said Windows XP x64, or for that matter any flavor of 64-bit Linux should work. So it is pointing to that you need a different operating system altogether on your PC to get P2P features working.