On Mon, 3 Oct 2011, andrei radulescu-banu wrote: > Dear all, > > We are encountering a problem with Linux 2.6.35 with a Smart Modular > solid state flash drive: during an ext3 file system write operation, > the eUSB drive responds with a NYET, and expects the host side to > continue a little bit later with a PING. The PING never comes from > the host side; instead after 9 seconds the host side issues a hard > reset to the eUSB. > > We have also tested Linux 2.6.39, and we see the same problem. > > Is this a firmware problem on the host side, or is it something which > can be worked around in the USB driver? It is a firmware problem in the host's USB controller. You should be able to rule out the other components by plugging the flash drive into a PC with a more common motherboard chipset, such as ICHx. > The error causes the ext3 file system layer running on the eUSB to > become corrupted and the kernel to crash. Upon reboot, however, the > ext3 filesystem recovers. Here is additional info about our setup: > > - Smart Modular 2G flash drive > - Intel motherboard T5520UR� > - ext3 file system corruption also happens with Intel motherboard > S3200SH, but we did not double-check if it's the exact same error > with a missing PING after a NYET > - but corruption does not happen if I disable USB2.0 in the BIOS - > i.e. if the eUSB is running in USB1.1 mode. You might try checking with Intel to see if anyone there knows anything about this. 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
