Re: fcntl() or ioctl() for fs readahead?

Matthias Urlichs (smurf@smurf.noris.de)
Sun, 12 May 1996 06:18:02 +0100

In linux.dev.kernel, article <4n16dl$cv9@fido.asd.sgi.com>,
lm@neteng.engr.sgi.com (Larry McVoy) writes:
>=20
> Linus and I talked about this - we were thinking about adding a=20
> reada(fd, offset, len)=20
> interface which says fetch len bytes into the buffer cache at the off=
set
> on that fd.
>=20
I'd like to be able to do this, but the interface is somewhat limited.

> One limitation of this approach is that the select can only inform yo=
u
> about one of potentially many reada() requests. We didn't know what =
to do
> about that.
>=20
Well, yes, you get the first reada() back. Then you can do your work,
select() again without that fd, and get status for the next reada().

> Thoughts? This is useful to some, kernel bloat to others. I'd like =
to
> hear if people think it is useful (I know people that do but they don=
't
> run Linux - yet).

I don't know about you, but I'd rather spend the time for this on the n=
ew
POSIX.4 signals so that you can do true async I/O, including the above
hack, with the help of clone() and your favorite C library.

--=20
O God! that men should put an epigram in their mouths to steal away
their brains!
-- William Shakespeare
--=20
Matthias Urlichs \ Noris Network GmbH i.Gr/ Xlink-POP N=FCrnberg=
=20
Schleiermacherstra=DFe 12 \ Linux+Internet / EMail: urlichs@nor=
is.de
90491 N=FCrnberg (Germany) \ Consulting+Programming+Networking+etc'i=
ng
PGP: 1024/4F578875 1B 89 E2 1C 43 EA 80 44 15 D2 29 CF C6 C7 E0 DE=
=20
Click <A HREF=3D"http://info.noris.de/~smurf/finger">here</A>. =
42