Eric, > Do you know if io_opt can be configured on scsi devices that do not > have io_opt provided by the controller? If so, or if it would be easy > to add, then configuring io_opt at the scsi layer is probably a better > option so subsystems besides bcache would benefit. There is currently no way to do this in the SCSI sysfs interface. However, io_opt is a queue_limits parameter. The appropriate location for modification is blk-sysfs.c. We already allow overriding other block device properties such as rotational, stable_writes, discards, max I/O size, etc. I don't see any particular reason why we couldn't also permit overriding the optimal I/O size. -- Martin K. Petersen Oracle Linux Engineering
