On Mon, Mar 18, 2013 at 10:31:28AM -0700, Sarah Sharp wrote: > On Mon, Mar 11, 2013 at 10:21:57PM -0400, Alan Stern wrote: > > On Mon, 11 Mar 2013, Sarah Sharp wrote: > > > > > On Mon, Mar 11, 2013 at 02:50:57PM +0800, Joshua wrote: > > > > ... > > > > > > Recently, I read the EHCI 1.1 addendum specification. I noticed there > > > > are some implementations in kernel for that spec by intel engineers, > > > > but i didn't find the "Hardware Prefetching" support . Did the kernel > > > > support that feature? > > > > I doubt there have such a EHCI that supports the EHCI 1.1. > > > > > > What would an EHCI addendum have to do with your particular bug? Or is > > > this a different question? > > > > > > Alan Stern would probably be the best person to ask this question, since > > > he's the EHCI maintainer. > > > > ehci-hcd does not support Hardware Prefetching. > > > > Since Intel played a big role in designing the EHCI 1.1 addendum, it > > seems likely that they made some EHCI controllers supporting those > > extensions. But I don't know the details, and I don't know if the > > controllers in the current chipsets support EHCI-1.1. > > I'll ask the hardware architects if any Intel EHCI host supports it. It sounds like newer Intel EHCI hardware (since Ibex Peak/Nehalem) does support the prefetching mechanism. However, Microsoft didn't initially support the addendum, and I don't know if they do today. If Linux did add support, we would be running code on hardware that may never have had that functionality verified. Basically, it doesn't sound like something we want to implement in Linux. Sarah Sharp -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html