Re: [PATCH] New phys_addr() syscall

Inaky Perez Gonzalez (inaky@peloncho.fis.ucm.es)
Tue, 21 Jul 1998 17:54:11 +0200


>>>>> "Alan" == Alan Cox <alan@lxorguk.ukuu.org.uk> writes:

>> I was thinking of allowing the USB Stack to directly write to user
>> pages using PCI. If PCI doesn't also honour the bits we need, could
>> it be done by hand as the write is finished?

Alan> Ok firstly why do you need to write directly to user space

It is to offer a posibility to waste less resources. If I've
got this USB disk drive and I order to write 100K, if I know where
they are at use space, I just have to write from user space directly
to the drive. This way there is no intermediate buffering ... well,
the disk drive was a bad example [disk cache]; let's change it for
Audio device.

Alan> - you would normally want to buffer messages up anyway surely,

The message-type transfer are no problem, they're better using
buffers, as they are small. I'm more concerned on _long_
transfers. Alas, it is not too difficult to do on the USB code [if I'm
not doing it wrong :)].

Cheers

Linux-USB! http://peloncho.fis.ucm.es/~inaky/USB.html -
-
Inaky Perez Gonzalez -- PGP pubkey fingerprint -
inaky@peloncho.fis.ucm.es -- 8E 34 3A 62 64 99 E2 44 -
http://peloncho.fis.ucm.es/~inaky -- AD 7B 30 D9 DD FF 3E 4C -
--------------------------------- -- ----------------------- -
The loneliness of the long distance runner .....

-
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.altern.org/andrebalsa/doc/lkml-faq.html