TIMING QUESTIONS SCSI-2

Gerry Houlder Gerry_Houlder at notes.seagate.com
Sun Sep 6 23:24:59 PDT 2015


_Date: 03-23-94 07:44:05 AM
_To: scsireflector
_Recipient: scsi at WichitaKS.NCR.COM@internet
_From: Gerry Houlder at SEAGATE
_Subject: Re: TIMING QUESTIONS SCSI-2

These were the questions asked:

>Question 1: Is there a maximum time for a device to check if it has been 
>  selected. Above in step 2 and 3, Target 2 determined it
>  had been selected. How long should Target 4 continue to
>  check if it has been selected?
>  
>Question 2: What is the minimum time between a device asserting BSY once
>  it determines it was selected, Step 3 above, and the bus
>  returning to Bus Free, Step 5 above? 
>
Answer 1: Common sense tells me that target 4 should stop checking to see if it 
has been selected as soon as Selection phase is no longer present on the bus. 
SEL must be true and BUSY must be false for the bus to be in selection phase. 
If either BUSY becomes true or SEL becomes false the target should no longer be 
looking for selection. SCSI-2 (section 6.1.3 SELECTION phase) doesn't 
explicitly define the action for targets that don't get selected, so the 
"common sense" answer applies. Should a target really keep looking to see if it 
has been selected during BUS FREE phase, MESSAGE OUT phase, or COMMAND phase? 
Such an action would be illogical.

Answer 2: The selection phase may only last from 2 to 5 microseconds with many 
of today's initiator and target chips. A sequencer driven initiator chip 
working with a sequencer driven target chip can recognize selection, accept 
IDENTIFY message, accept command bytes, send DISCONNECT message, and go bus 
free within 15 microseconds (for all phases!). This 15 usec number is not a 
minimum, either. I'm sure a smaller minumum could be calculated from the SCSI-2 
minimum values. Since this number is already much less than the 200 usec used 
by the target to determine if it is being selected, the problem is already 
glaringly exposed.

If your initiator really wants to have slow targets on the bus, it will have to 
delay at least 200 usec between commands so the slow target has time to detect 
the bus free and to reset itself. This will slow your performance, but at least 
your system will still work!

--
Gerry Houlder (Gerry_Houlder at notes.seagate.com)
-------------------------------------------------------------------------
Seagate Technology   -   920 Disc Drive   -   Scotts Valley, CA 95066 USA
Main Phone 408-438-6550   -   Email Problems postmaster at notes.seagate.com
Technical Support: BBS 408-438-8771  Fax 408-438-8137  Voice 408-438-8222  
-------------------------------------------------------------------------




More information about the T10 mailing list