At Tue, 3 Dec 2013 16:17:15 -0800, Sarah Sharp wrote: > > On Tue, Dec 03, 2013 at 01:13:15PM -0800, Greg KH wrote: > > On Tue, Dec 03, 2013 at 01:08:08PM -0800, Sarah Sharp wrote: > > > Yeah, it doesn't make much sense to me either. But I can reproduce the > > > behavior on my Lenovo x230 if I add the PCI ID of my Intel Panther Point > > > xHCI host controller to the quirk. Without the quirk, when I load the > > > driver after boot with modprobe, xhci_init takes on average 44ms to > > > complete. With the quirk, it takes 74ms on average. I haven't measured > > > boot time too many times, but it also seems to be increased by around > > > 40ms. > > > > > > The patch only touches the PCI and platform remove function, adding a > > > host controller reset and putting the host in D3. Maybe on module > > > reload, it takes longer to get the host into D0? I don't see why that > > > would impact module load times at boot though. > > > > That might be true, try testing boot times from cold boot to get the > > "real" time here. > > Ok, it looks like the 83353 usec module load time at cold boot I got > after adding the quirk for my Panther Point xHCI host was a fluke. It's > probably the D3 to D0 transition that's taking so long on module reload. > I'll have to double check with the QA testers and ask them to test the > boot init time, not the init time between module re-load. So, maybe there is a difference when you do warm boot? I talked with Oliver, and we concluded that this is an unavoidable penalty, as long as we need the quirk. In theory, we might be able to know whether it's a shutdown or a reboot beforehand, and avoid the power state change there. But, there is always a choice for user to power off during reboot, and it leads to the problem again. Takashi -- 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
