On Thu, 14 Apr 2016, Oliver Neukum wrote: > On Wed, 2016-04-13 at 15:11 -0700, Greg KH wrote: > > On Wed, Apr 13, 2016 at 02:37:35PM -0700, Matthew Giassa wrote: > > > Thank you for the feedback Greg. This is my first attempt to submit a > > > kernel patch. > > > > > > Is there a better approach to this? The only other option at my disposal > > > is to add about 200 products to the quirks.c file, and routinely update > > > that list as new products are released. I need to implement some means > > > of disabling LPM, ideally on a per-device basis. > > > > Why are these devices say they support LPM yet they do not? > > Most likely because there are conditions under which the Windows stack > will not use it, therefore the feature hasn't been tested. > > https://blogs.msdn.microsoft.com/usbcoreblog/2013/11/15/link-power-management-lpm-in-usb-2-0/ > and related documents > > Now to us the question is moot unless we can use the answer to work > around the failure. It is unlikely that this will be the common case. Given all the caveats in that blog, maybe we should simply disable LPM for all USB-2 devices, permanently. > So we have quirk for it. The ability to trigger this quirk the hard way > would be useful for debugging. Thus I believe that this patch is a good > idea. If this is for debugging then maybe it belongs in debugfs. Doing it that way instead of as a module parameter would allow it to be bus-specific. On the other hand, attributes in debugfs can't be set at boot time whereas module parameters can. > Known quirky devices, however, should still be added and in this > particular case we ought to add the whole vendor to the lst. Agreed. Alan Stern -- 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
