Showing results for 
Search instead for 
Did you mean: 

Archives Discussions

Journeyman III

Question on internals of GPU


I am trying to use GPU for a specific purpose, and I would like to ask following questions about internals of GPU.

Is there any operating system running on the GPU – or is everything implemented in hardware?
If everything is implemented in hardware how does the device driver interacts with the GPU?

Does ATI reveal more information about their GPU architecture than NVIDIA? If yes, are there any documents available?


2 Replies
Adept III

Please download the ATI GPU information from there.


There is no operating system or equivalent running on the GPU.

There is a  DMA controller / Command Processor which reads command packets from a ring buffer in system memory (GART) and passes them to the hardware, similar to the way an intelligent disk controller works, but in general the GPU is only working on one command at a time. That command might be "draw five thousand triangles from this list of vertices", but it's still one command.

It's actually a bit more complicated than that, in the sense that the pipeline in a GPU is much longer than in a CPU, and makes heavy use of delayed-write caches to improve throughput, so you may have more than one command "in the pipe" at the same time, but if you think about it as "one operation at a time" that will make the most sense.

We post documents directly to the site at - new docs typically hit that site first then get mirrored back to periodically. If you look at section 4 of the 5xx acceleration guide and page 8 of the r6xx-r7xx-3d guide (start with page 😎 that should give you a pretty good understanding.

Both documents include lists of "PM4 packets" - those are the command packets which get read by the Command Processor and executed by the GPU.