> 2) Because of the above, it should be possible to get data straight from the
> device into user memory without any copying.
You can always mmap things. Linus also saif that in the future, if you read
a page-sized and aligned chunk Linux is likely to just remap the page
instead of copying.
> Actually I guess what is needed is not necessarily a new device, but possibly
> an extra (non-portable, but hey) flag for open (and maybe mmap?) to say `don't
> cache this, I'm not going to see it again'. The device is just a way of
> saying this without having to code it in the program.
>
That might actually be a good idea.
> Am I missing something really obvious here? (Note - even if mmap didn't trash
> cache, which I am certain it does (page cache?) it doesn't work for this - try
> mmapping something >4Gb :-( )
>
Of course mmap works for data sets > 4 GBytes. You just need a 64-bit CPU...
...I expect Digital will happily sell you one. ;-)
Besides, llseek() also exists, and you can keep the control structures and
indices mmap()ed in the lower Gbyte and read() the actual data.
--
Set the cart before the horse.
--John Heywood
--
Matthias Urlichs \ noris network GmbH / Xlink-POP Nürnberg
Schleiermacherstraße 12 \ Linux+Internet / EMail: urlichs@noris.de
90491 Nürnberg (Germany) \ Consulting+Programming+Networking+etc'ing
PGP: 1024/4F578875 1B 89 E2 1C 43 EA 80 44 15 D2 29 CF C6 C7 E0 DE
Click <A HREF="http://info.noris.de/~smurf/finger">here</A>. 42