> There was some talk about reordering disk access for IDE drives
> on this list. Was that incorporated into the kernel ?
> If not , are there any www pages about the project ?
>
If you are talking about reordering read/write requests, than it is
already done in the kernel (and in SCSI drives..)
I was(am) working on a project to try to write an intelligent agent that
would try to predict drive access and preload certain files (actually file
fragments). I do not plan this to be useful on loaded machines, rather
this is intended for improvement of interactive response for machines with
a lot of memory.
So far the results can be found on
http://www.math.upenn.edu/~vdergach/Linux
Right now I am trying to find out more of the workings of Linux mm and fs
subsystems, in particular how the buffered reads relate to mmaped files..
One of the things I found out is that if you first read a file
(time cat file > /dev/null) so that it is in buffers, and then mmap it
the data will not be reused. I have also found out that (at least on my
system) mmaping a file allows a faster access than just reading it with
read(2), though in both cases I am using 4096 blocks to do the I/O..
Vladimir Dergachev
>
> --
> David Balazic , student
> E-mail : 1stein@writeme.com | living in sLOVEnija
> home page: http://surf.to/stein
> Computer: Amiga 1200 + Quantum LPS-340AT
> --
>
> -
> 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
>
-
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