Showing posts with label fpga. Show all posts
Showing posts with label fpga. Show all posts

1.27.2022

China allowed AMD to take over Xilinx, but imposed a number of conditions

China allowed AMD to take over Xilinx, but imposed a number of conditions

China allowed AMD to take over Xilinx, but imposed a number of conditions

China's regulator is ready to allow AMD's takeover of Xilinx to proceed, Reuters reports. This was the last barrier to AMD's takeover of Xilinx's designs and products. The deal was announced in October 2020. The asking price is $35 billion. All countries except China had previously cleared the deal. China has agreed to it only now, putting AMD on counter conditions.

As required by the regulator, within 6 years, AMD and Xilinx products must be available in China's domestic market and not create discriminatory pressure on customers if they want to choose some products and refuse others. The merged company must also ensure the flexibility and programmability of Xilinx FPGAs, as well as compatibility of development methods with Arm-based processors. In addition, the merged company must ensure that its GPUs and CPUs sold in China will be compatible with products in the Chinese market. The regulator reserves the right to inspect the merged company in any way it can to ensure compliance with the requirements. After 6 years, the merged company has the right to request a review of the terms of this agreement, but the regulator reserves the right not to change anything. If AMD and Xilinx agree to the terms put forward, the deal will be immediately approved in China. AMD's purchase of Xilinx was a kind of response to Intel's earlier takeover of FPGA developer Altera. FPGAs are expected to greatly expand the potential of central processing units, whose further development is hampered by the proximity of technological barriers to production. New technological processes can no longer increase the performance of processors, and they need new functional solutions. Progress promises the integration of FPGAs into processors or platforms.

1.04.2021

AMD patent reveals hybrid chip design with CPU and FPGA under one cover

AMD patent reveals hybrid chip design with CPU and FPGA under one cover

AMD patent reveals hybrid chip design with CPU and FPGA under one cover

An FPGA may not be as flexible and easy to program as a CPU but it can significantly speed up a specific task. In the hands of a skilled engineer, FPGAs can relieve the CPU of a wide range of workloads. Intel took up integrating Xeon and FPGA back in 2014, but that didn't lead to any publicly available products of this kind. The new patent shows that AMD is also thinking along these lines, but has chosen a different approach.

hothardware.com

On 31 December 2020, the US Patent and Trademark Office (USPTO) published AMD's application to integrate FPGAs with the CPU, which was filed back in summer 2019. AMD included 20 claims in its patent application, but the gist of it is basically that the processor can include one or more execution modules that can be programmed to process different types of custom instruction sets. This is exactly what an FPGA does. AMD has already created a furor with its chipset-based processors, but in this case we're talking about something completely different. The programmable FPGA module in the AMD patent uses common registers with the CPU's integer execution units and floating point modules. So it's about integrating the FPGA with the CPU on the same chip - otherwise the whole thing would be too slow. Such programmable blocks will be able to handle special data types like FP16, used to accelerate AI.

hothardware.com

In the case of multiple FPGAs on a chip, each module could be programmed with a different set of specialized instructions and those programmable blocks could be reconfigured by a particular AMD customer to suit their needs. Moreover, the processor will be able to reconfigure the FPGA to accelerate specific tasks for a particular software application. AMD has been working on various ways to accelerate computing in the field of AI for many years. First and foremost, there have been specialized accelerators based on Radeon graphics cards. The move to FPGAs after the Xilinx acquisition makes sense, and we'd like to see what this patent might actually turn out to be. It may be a while before we see products based on such a design - it's too early to assume that such chips will be part of the EPYC family in the foreseeable future;