On Sat, 14 Aug 2010 10:05:43 +0800, Ming Lei wrote: > 2010/8/14 Dennis Nezic > <dennisn@xxxxxxxxxxxxxxxxxx>: > > My usb2.0 multi-card reader, 058f:6366 Alcor Micro Corp., with > > vanilla kernel 2.6.35.1, doesn't seem to work with ehci_hcb. It > > does work with ohci_hcb, so I don't think it's a hardware problem > > necessarily? > > > > Here is my syslog when using EHCI: > > http://dennisn.dyndns.org/guest/pubstuff/ehci-bug/ehci-fails.log > > > > Here is my syslog when using OHCI: > > http://dennisn.dyndns.org/guest/pubstuff/ehci-bug/ohci-works.log > > > > The main differences that I notice are many "detected XactErr"s with > > ehci, but they don't seem to be too critical, until they (seem to) > > cause: > > > > 09:14:30 kernel: ehci_hcd 0000:00:13.2: devpath 8 ep2in 3strikes > > 09:14:30 kernel: usb-storage: Status code -71; transferred > > 76288/98304 09:14:30 kernel: usb-storage: -- unknown error > > 09:14:30 kernel: usb-storage: Bulk data transfer result 0x4 > > 09:14:30 kernel: usb-storage: -- transport indicates error, > > resetting > > > > ... which seems to happen with "usb_stor_bulk_transfer_sglist: xfer" > > greater than about 512 bytes? (The transfers seem to complete with > > smaller xfers?) > > > > Ideas? Suggestions? > > You may connect your card-reader into windows machine to see if > there is the similar issue. > > I guess it may not work steadily at high speed on windows too. > Maybe windows hcd can degrade to full speed for your card reader. It seems to work fine in Window$, at usb2-ish speeds, 10+MB/sec. Please don't make me test in in there again :S. There is something wrong with the ehci_hcd code, or something quirky with this hardware :\. -- 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
