> Hmm, is there a way to get a dump of the host side registers when a > timeout occurs? Add printks and recompile ? There is almost no host side register state on the PIIX4 like most older IDE controllers, just a few timing registers. You can certainly print out the failing task file (and the drive itself will if you use the smart utils dump out the last 5 failed commands) > | What I would say is that turning off the drive side power management > | (-S0) will prevent the problem occuring - it only appears to occur when > | the drive has already been spun down. > > which I would now delete because I can't keep it in -S0 mode! ;( If the APM bios code is randomly issuing drive task files behind our back then its going to go kerboom when APM falls out with us. However that might suggest why only you see a problem and only with libata. The libata code manages the taskfile writes intelligently (as they are expensive) which old IDE doesn't. That means old IDE is much more resilient to that kind of BIOS abuse if that is what occurs. If so you can try pasting the standard tf write routine into the driver, removing the caching checks so you always write the whole taskfile (except the LBA48 check) and see if that cures it. Alan -- 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
