This message is from the T13 list server.
On Mon, 9 Sep 2002 [EMAIL PROTECTED] wrote: > This message is from the T13 list server. > > > A routine comparison of disc drives from various products and vendors has > brought a discrepancy to light. A search of ATA-7 does not offer any clear > resolution. > > Please offer an opinion on what YOU expect is 'proper' behaviour'. > > Problem: > If a host has issued a plain vanilla IDLE or STANDBY command with a > non-zero timer, AND > the timer has not expired AND > 'Advanced Power Management' is not enabled AND > 'Power-up in standby' feature is not enabled AND > a <Hard> Reset event occurs - - - The drive is still active/idle so the reset did not change the state yet. > We specify in state diagrams that (after the reset has been handled): > a) the device's power management state does not change The non volitale setting remain as they should, if they don't it is a bad device. > but we do not address these questions: > a) does the timer get reset ? Logically yes, the device was not idle, which would cause the timer to reset. > b) does the timer get disabled ? If the timer is disable, effectively it looks like an expired timer. Thus the device could/should/may drop into standby instantly. As long as we are constrained to "standby" it is not a problem from the host side. We kick the device with a command and it spins back to life. One issue of concern for a HOST would be attempting to sync a local timer against what the device may be doing. This could make the HOST driver upset. Anyways, if it used the checkpower calls proper it would know now to adjust for observed variations. Andre Hedrick LAD Storage Consulting Group It is 12:30am and I have had little sleep, and the tax deductions inc. by one last week, so I expect to be wrong above. Regardless it will add some traffic while everyone is off at IDF showing SATA 2.0 with SAS painted on it w/ a fire hose. The crystal ball show X3 storage big bang returning to a big crunch.
