On Fri, 19 Sep 2008, Antonio Ospite wrote: > On Fri, 5 Sep 2008 15:35:13 -0400 (EDT) > Alan Stern <stern@xxxxxxxxxxxxxxxxxxx> wrote: > > > On Wed, 27 Aug 2008, Alan Stern wrote: > > > > > This patch (as1133) fixes a bug in the interaction between > > > scsi_end_request() and scsi_io_completion(). The bug was triggered by > > > recent changes to accomodate USB drives can't access their last few > > > sectors using multi-sector transfers (the so-called last_sector_bug > > > flag). > > > > > > The bug shows up when a multi-sector I/O request has been broken up > > > into single-sector commands. If one of those commands gets an error, > > > the remainder of the request never gets completed. This is because > > > the "bytes" value passed to scsi_end_request() is completely wrong; it > > > is the transfer size of the current command, not the total transfer > > > size of the request. > > > > There hasn't been any reply to this patch submission. For reference, > > the original message is at > > > > http://marc.info/?l=linux-scsi&m=121986305201166&w=2 > > > > Has it been accepted, rejected, or did it fall through the cracks? > > > > Alan Stern > > > > Hi again, > > anything new about this patch? Alan, I'll keep reminding about it from > time to time, until it is accepted or rejected. All I can tell you is that I asked James Bottomley at the Linux Kernel Summit, and he said he has been too busy to look at it but it's got an "!" on his to-do list. Alan Stern -- To unsubscribe from this list: send the line "unsubscribe linux-scsi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
