[OFFTOPIC] Was: Re: GGI, EGCS/PGCC, Kernel source

Torgeir Veimo (torgeir.veimo@vertech.no)
Tue, 24 Feb 1998 17:40:37 +0000

Nate Tuck:
> If you can read that data out of memory and write it directly to the
> graphics hardware, this "only" requires 640 MB/s of CPU bandwidth and 320
> MB/s of
> memory bandwidth. Write it to a DMA buffer instead, and now you are at 640
> MB/s of CPU bandwidth and 960 MB/s of memory bandwidth. Add a copy from
> the server's input queue to the DMA buffer and you are at ~1.3 GB/s of CPU
> bandwidth and 1.6 GB/s of memory bandwidth.

The way the userland driver talks to the card is either through a
ping-pong buffer, through mmio, or through dma if the card can handle
it. More and more cards do handle dma every day.

I think I should go back to coding again and present my case when I can
show that it works and how it speeds up things.

Alan Cox:
> Hopefully I can spend an evening armwaving with GGI folks (Geert you going to be there ?)
> and see what I can learn

This would be very nice.


Torgeir Veimo, Vertech AS,

email: Torgeir.Veimo@vertech.no, mobile: 90673881, office: +47 55563755

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu