Alan Cox wrote: > On Gwe, 2006-05-12 at 00:32 +0900, Tejun Heo wrote: >> Convert ata_piix to new probing mechanism. Automatic hotplug is not >> supported due to hardware limitation (no PHY event interrupt), but >> warm plugging works. > > True for the SATA ports but don't try it on the PATA ones. The PATA side > requires the right buffers are present, and also in some situations that > IORDY is not in use before the unplug. > Yeap, also the power plug is troublesome too. It doesn't have spike protection and the machine goes out quite often if hot plugged. ata_piix PATA warm plugging is mainly for notebook users with hotswap bays and people with PATA->SATA bridges attached. In the long term, we need to establish use cases and conventions for warm/hot plugging. Even SATA hot plugging can destroy filesystems on unrelated disks if power supply cannot hold 12v steady while the hot plugged drive spins up. To be on the safe side, disks need to be flushed and held before hotplugging and released after the new device reaches stable state. Sturdy machines including most servers with decent power supplies should be okay though. So, kernel IO holding support and nice CLI/GUI programs to show the current configuration and drive plugging sequence would be nice, I think. -- tejun - : 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
