Re: loading EHCI_HCD slows down IDE disk performance by 50%

["Alessandro Suardi" <alessandro.suardi@gmail.com>]

On 10/30/06, Lee Revell <rlrevell@joe-job.com> wrote:
> On Sun, 2006-10-29 at 20:52 +0100, Alessandro Suardi wrote:
> > On 10/29/06, Lee Revell <rlrevell@joe-job.com> wrote:
> > > On Sun, 2006-10-29 at 19:13 +0100, Alessandro Suardi wrote:
> > > > Any hints/tips about what to try with this issue will be
> > > >  of course very welcome.
> > >
> > > git bisect
> >
> > As I clarified to Lee in private email, this method doesn't
> >  seem to apply in this situation, as the problem appeared
> >  when I selected CONFIG_EHCI_HCD in my build due to the
> >  purchase of the USB2.0 disk, back in the 2.6.16-rc cycle;
> >  so if the combination of high performance IDE + EHCI_HCD
> >  ever existed, it must be earlier than 2.6.16-rc5-git8.
> >
> > Lacking any input I'll try to find out more in the next week
> >  by starting out at 2.6.10 and eventually earlier.
>
> Maybe you could close the Bugzilla ticket you posted, as most of the
> comments relate to HAL and polling for media changes which you seem to
> be saying have nothing to do with this bug?

The bugzilla entry was hijacked by people who were incorrectly
thinking their problem had to do with my report, I myself have
always stated HAL/polling had nothing to do with the issue. Not
my fault, there...

I may of course re-file the bug, but it'd be nice if I were asked
by someone looking after the Fedora bugzilla; as a matter of
fact I did ask for opinions there after my last findings.

--alessandro

"...when I get it, I _get_ it"

    (Lara Eidemiller)
-
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/