Adhere to open source Road AMD Microsoft Development C + + version AMP1.2

AMD today announced that they are working with Microsoft's C + + version of AMP1.2, an open-source C + + compiler and an open-spec C + + AMP that can run on Linux and Windows for the first time. It appears that AMD's goal is to support more cross-platform solutions, continue to contribute to the open source community to provide a variety of programming languages, leveraging its research and development advantages on heterogeneous platforms, to provide developers with a new heterogeneous platform, including servers, personal computers and other handheld mobile devices.

"AMD has developed a wealth of development experience, and we are proud to say that the first open source implementation of the C + + AMP1.2 version will bring greater performance and lower power consumption, and fully increase productivity to reduce friction and latency." "AMD's cross-platform support capabilities are further enhanced to provide more development solutions in the future, allowing more accelerated applications to be shared with physical memory to improve CPU and GPU efficiency while Windows and Linux are our primary platforms"

Somasegar, vice president of Microsoft Development in the United States, has a high evaluation of AMD's work:

"AMD heterogeneous programming is an excellent development tool. Grand Casino and C + + AMP provides valuable development experience and resources to the Linux open source community.

Not only that, C + + AMP version 1.2 supports C + + with broad cross-platform compatibility and almost easily supports most platforms:

In OpenCL supports AMD Cpu/apu/gpu and Intel CPUs, also includes NVIDIA GPU and Apple Mac OS X and other OpenCL-compatible platforms. and Spir, supports AMD Cpu/apu/gpu and Intel's CPUs

AMD wants to fully support AMD APU and all future HSA-compatible platforms based on the HSA Hsail.

The key feature of the 1.2 version of the Open source C + + AMP specification is that they support shared physical memory, which means greatly simplifying the flow of data between shared CPUs and GPU heterogeneous platforms. Significantly increased productivity, these heterogeneous platforms create new development advantages that help programmers save on development costs and improve development efficiency.

Recommended Reading http://www.jieshounews.cn/blog-1568-21.html
http://www.yupoo.com/photos/summo2/93917139/

Adhere to open source Road AMD Microsoft Development C + + version AMP1.2