On Thursday, September 07, 2006 8:23 AM, James Bottomley wrote: > > Good ... I'm starting to think we might junk all the function > callbacks > and simply have even the HBA configured via SMP (i.e. the driver would > interpret SMP messages and set the HBA phys appropriately). > Id rather vote for seeing that we keep scsi lld callbacks for hba properties, and smp passthru for expander properties. Its going to be alot of work to rip out the current code, and to add code for hba that snoops smp passthru, and interprets/translate what is meant for it. I doubt all smp passthru's directly map to hba properties. Also, all these KABI breakages are causing me major headaches. Between SLES10 and RHEL5 there is a different API due to the wide port support. I think adding new callbacks in transport, doesn't necessary mean I have to implement them in the lld, which makes my oem supplied drivers and distro's drivers more compatible between kernel rev's, and upstream. Eric - 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
