Mux and Align rotation

Elliott, Robert (Server Storage) Elliott at hp.com
Tue Feb 27 06:58:55 PST 2007


Formatted message: <A HREF="r0702271_f.htm">HTML-formatted message</A>
Attachment #1: <A HREF="r0702271_smime-3.p7s">smime-3.p7s</A>

When the deletable primitives are only showing up 1 out of 128 dwords, it's
not very important at all.  ALIGN rotation is most important for rate
matching, where they show up as 1 out of 2 dwords or 3 out of 4 dwords.
In those cases, MUX does serve the purpose of breaking up the same pattern,
so these patterns are equally beneficial, EMI-wise:
ALIGN (0) MUX ALIGN (1) ALIGN (2) ALIGN (3) ALIGN (0) ALIGN (1) ALIGN (2)
ALIGN (3) ... 
ALIGN (0) MUX ALIGN (2) ALIGN (3) ALIGN (0) ALIGN (1) ALIGN (2) ALIGN (3)
... 
So, the standard is pretty loose: (7.2.5.10) "Substitution of a MUX for an
ALIGN may or may not affect the ALIGN rotation (i.e., the MUX may take the
place of one of the ALIGNs in the rotation through ALIGN (0), ALIGN (1),
ALIGN (2), and ALIGN (3), or it may delay the rotation)."
  _____  
From: owner-t10 at t10.org [mailto:owner-t10 at t10.org] On Behalf Of Craig Stoops
Sent: Monday, February 19, 2007 6:44 PM
To: t10 at t10.org
Subject: Mux and Align rotation
Hi Rob,
In SAS2R08 it indicates a MUX replaces an ALIGN for the occasional
transmission. What isn't clear is how this affects align rotation. 
Is a rotation still expected, or is it like a notify in that is is abiguous
as to whether the rotation is advanced?
Thanks,
Craig
ExpertIO, Inc.
"Your Storage Deisgn and Verification Expert"s
805-428-0839



More information about the T10 mailing list