Messages in this thread | | | Date | Tue, 8 Jun 1999 15:28:36 +0200 (CEST) | From | Geert Uytterhoeven <> | Subject | Re: RFC: Devices, buses and hotplug |
| |
On Mon, 7 Jun 1999, Roger Gammans wrote: > In article <199906061910.MAA30421@pizda.davem.net>, David S. Miller > <davem@redhat.com> writes > > Anyway... one question. Who does the byte swapping, the host CPU > > or the peripheral? > > > > If the answer is 'host CPU', then my response is I can write clean > > code that is faster because it takes advantage of big endian > > peripheral support. > > > >Nope, the host CPU does it, but at _ZERO_ cost. You have two ways > >usually to get it done: > > > >1) Load/Store instruction attributes: > > > >2) MMU translations > > Erm. There is a third way it could be done. I don't know about any > specific implementations but the Backplane/bus bridge could implement > byte swapping merely by the way the data lines are connected. > > If I had this problem and was building a motherboard, I would seriously > consider going this route, unless there was a good reason not to. > > Judging by the rest of this thread I assume no-one does it, which is a > shame. But I suppose if endian-ness was never specified.... sigh.
There are: some Amiga video boards use little endian PCI video chips connected through a hardware byteswapping Zorro<->PCI FPGA.
Greetings,
Geert
-- Geert Uytterhoeven Geert.Uytterhoeven@cs.kuleuven.ac.be Wavelets, Linux/{m68k~Amiga,PPC~CHRP} http://www.cs.kuleuven.ac.be/~geert/ Department of Computer Science -- Katholieke Universiteit Leuven -- Belgium
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |