On Tue, May 26, 2009 at 10:04:01AM -0400, Alan Stern wrote:
> On Tue, 26 May 2009, Michael S. Tsirkin wrote:
>
> > On Mon, May 25, 2009 at 05:08:12PM -0400, Alan Stern wrote:
> > > On Mon, 25 May 2009, Michael S. Tsirkin wrote:
> > >
> > > > > So apparently this is a bug in the device; it doesn't respond correctly
> > > > > to the first READ command. But since it does respond correctly to
> > > > > later commands, everything works okay thereafter. You ought to be able
> > > > > to recover from the error by running
> > > > >
> > > > > blockdev --rereadpt /dev/sdb
> > > > >
> > > > > manually.
> > > >
> > > > Yes, this helps.
> > > > Would it make sense for kernel to retry automatically?
> > > > Why doesn't it?
> > >
> > > I don't know the details in this case. Most likely the error code
> > > (Logical Block Address Out of Range) is interpreted as a fatal
> > > non-retryable error. For other sorts of errors, the kernel does retry.
> >
> > Who would know? The scsi crowd?
>
> They would know. But it's easy enough to find out. (Looks through
> the SCSI code...) Here we go. scsi_io_completion() contains this:
>
> case ILLEGAL_REQUEST:
> /* If we had an ILLEGAL REQUEST returned, then
> * we may have performed an unsupported
> * command. The only thing this should be
> * would be a ten byte read where only a six
> * byte read was supported. Also, on a system
> * where READ CAPACITY failed, we may have
> * read past the end of the disk.
> */
> if ((cmd->device->use_10_for_rw &&
> sshdr.asc == 0x20 && sshdr.ascq == 0x00) &&
> (cmd->cmnd[0] == READ_10 ||
> cmd->cmnd[0] == WRITE_10)) {
> /* This will issue a new 6-byte command. */
> cmd->device->use_10_for_rw = 0;
> action = ACTION_REPREP;
> } else if (sshdr.asc == 0x10) /* DIX */ {
> description = "Host Data Integrity Failure";
> action = ACTION_FAIL;
> error = -EILSEQ;
> } else
> action = ACTION_FAIL;
> break;
Which kernel version is this? I see different code in 2.6.30-rc7.
> Since the Sense Key value was ILLEGAL_REQUEST and the ASC value wasn't
> 0x10 or 0x20, action gets set to ACTION_FAIL. Hence the command is not
> retried.
>
> In the end, there's a limit to how far the kernel should go in
> compensating for buggy devices. Your device may well have passed that
> limit.
>
> Alan Stern
Let's see, hope to find a workaround that isn't too ugly to be included.
--
MST
--
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
