Re: [PATCH] speed up SATA
From: Jamie Lokier
Date: Mon Mar 29 2004 - 07:57:32 EST
Jens Axboe wrote:
> > Does it make sense to allow different numbers of outstanding TCQ-reads
> > and TCQ-writes?
> Might not be a silly thing to experiment with, definitely something that
> should be tested (added to list...)
Then there's another question: when deciding whether you can issue
another TCQ-read, do you compare the number of outstanding TCQ-reads
against the TCQ-read limit, or compare the _total_ number of
outstanding TCQs against the TCQ-read limit? Similarly for writes.
There are several other logical combinations that could be used.
Each condition will have a different effect on mean and peak latencies
under different load patterns. I'm not sure which makes more sense,
or even if multiple conditions should be used together.
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/