In linux.kernel, you wrote:
> Now, this isn't the prettiest patch in the world. But it does allow me
> to get good spin down times on my laptop hard drive. It was somewhat
> inspired by the 2.5.early version akpm did. Basically, it adds:
OK - this is a much nicer thing than noflushd (which, for a lot of
people, manages to cause anything using pthreads to not reap zombies
anymore). There is still one deficiency (well, two) though:
One is that it assumes you only have one drive (I was to use this on
my desktop as well, which doesn't have me infront of it for 16 hours a
day). So when one drive spins up and the writes are performed, the
other drive needs to spin up at the same time (one of my drives is
basically /boot and /dos, so never gets accessed once booted). Is
there a way of doing the write deffering on a per-partition basis,
instead of all at once?
Second, given that at least one of my drives (even on my laptop, which
stays powered up 24/7) is not accessed for 16 hours at a time, I
increase the bdflush max params to 86400*HZ - is this safe? Is there a
reason why this is limited to 10000*HZ by default? Yeah, I know - I do
a manual sync when I need something to be safely stored on disk.
Thanks for the good work.
-- TimC -- http://astronomy.swin.edu.au/staff/tconnors/I haven't lost my mind -- it's backed up on tape somewhere. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Sun Jun 15 2003 - 22:00:20 EST