Hi Alan, > From: Alan Stern, Sent: Friday, June 14, 2019 2:22 AM > > On Thu, 13 Jun 2019, Christoph Hellwig wrote: > > > On Thu, Jun 13, 2019 at 01:06:40PM -0400, Alan Stern wrote: > > > Hmmm. Would it be easier instead to copy the DMA mapping parameters > > > from us->pusb_dev->bus->sysdev into the SCSI host's parent before > > > calling scsi_add_host()? That way the correct values would be > > > available from the beginning, so the existing DMA setup would > > > automatically use the correct sizes. > > > > It would in theory. But at usb-storage has a special max_sectors quirk > > for tape devices, and as the device type is a per-LU property we'd > > still have to override it in ->slave_configure. > > Not just for tape devices. But that's okay; those max_sectors > overrides have been present for a long time and we can keep them. > Getting rid of the virt_boundary_mask stuff would still be a big > improvement. Thank you for the comments. So, should I wait for getting rid of the virt_boundary_mask stuff? If I revise the commit log of this patch, is it acceptable for v5.2-stable as a workaround? In other words, I worry about this issue exists on v5.2-stable. Best regards, Yoshihiro Shimoda > Alan Stern
