On Fri, 9 Mar 2007 20:15:36 +0900, Tejun Heo wrote: > Convert sata_via and sata_promise to nwe init model. Both controllers > can have combined configuration (SATA + PATA) and used twisted > initialization method (modifying port in ->port_start) to overcome > probe_ent limitations. > > This patch converts both drivers to new init model in which such > configuration is natively supported. > > * promise: Combined pata port now uses separate port_info entry right > after the sata counterpart entry. > > * promise: Controller configuration is discerned using ap->flags. > This simplifies init path and makes it look more like other LLDs. > > * promise: SATA/PATA branches are converted into separate ops. > Remaining ones are converted to more standard ap->cbl == > ATA_CBL_SATA check. I don't object to the new init model in principle. My only reservation here is that you are folding what logically are separate changes into one big change to sata_promise, and that makes the patch more difficult to review (and regressions more difficult to trace). I would like to do the SATA/PATA splitup into distinct operations as a separate first step, with you doing the new init model adaptation in a followup step. Also, at least the error handler changes would seem to be in conflict with Alan's cable type changes. /Mikael - To unsubscribe from this list: send the line "unsubscribe linux-ide" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
