09-063r1 and invalid dword / running disparity error counts

richard.bohn at seagate.com richard.bohn at seagate.com
Mon Apr 6 10:44:47 PDT 2009


* From the T10 Reflector (t10 at t10.org), posted by:
* richard.bohn at seagate.com
*
Hello all,
I have a question regarding the SAS interface power management
specification and the invalid dword / running disparity error counts from
the PHY ERROR LOG.
Currently the invalid dword and running disparity error counters are
defined as follows:
The INVALID DWORD COUNT field indicates the number of invalid dwords (see
3.1.121) that have been received
outside of phy reset sequences (i.e., between when the SP state machine
(see 6.8) sends a Phy Layer Ready
(SAS) confirmation or Phy Layer Ready (SATA) confirmation and when it sends
a Phy Layer Not Ready
confirmation to the link layer).
Does this wording need to be updated to account for state machine behavior
when the PHY enters its low power states? The new SP state machine as
proposed in 09-063r1 never sends a "Phy Layer Not Ready" confirmation to
the link layer when the PHY is powered down (although it does send a "Phy
Layer Ready" confirmation to the link layer when transitioning back to SP15
PHY_Ready). I imagine this was intentional so as to not reset the IR state
machine during power management modes.
I think once SP15 PHY_Ready is left, the invalid dword and running
disparity error counters should cease counting until SP15 PHY_Ready is
reached again. Thus, I think the current wording that defines those
counters as counting until a "Phy Layer Not Ready" message is received
needs to be updated to account for the window between entering SP31
SAS_PS_Low_Phy_Power and re-entering SP15 PHY_Ready.
Thanks,
Richard Bohn
Seagate Tech.
*
* For T10 Reflector information, send a message with
* 'info t10' (no quotes) in the message body to majordomo at t10.org



More information about the T10 mailing list