[T10] Explanation for approved SPL4 changes

Tim Symons tim.symons at microsemi.com
Wed Sep 27 10:26:59 PDT 2017


Changes approved by the WG to the SPL4r12 standard were made in response to the issues identified by Toshiba in document 17-135.

1. The SPL4r12 text was found to be incomplete, which meant that the standard was technically inaccurate. The group approved changing the lists of exceptions for dword separation to include:
a) the original primitive;
b) a deletable primitive;
c) a deletable binary primitive; or
d) a deletable extended binary primitive.

This text was added in two places as suggested in proposal 17-135. It was determined that the Scrambled idle segment should not be included in this list as it is removed at the phy layer and is not processed by the link layer (the same as ALIGN primitives in dword mode).

2. The group agreed that SPL4r12 figure 131 and figure 133 were inaccurate because the Scrambled idle segment would be deleted at the phy layer (the same as ALIGN primitives in dword mode). The group decided to revise figures 131 and 133 to retain a similar structure as the original figures.


·         Figure 131: Change label for the pair of Scrambled idle segments to Idle dword segments.

·         Figure 133: Change the first Scrambled idle segment into a pair of Idle dword segments to ensure that 8 dwords separate the two events of Redundant primitive sequence A. Change the Scrambled idle segment pair into a single Scrambled idle segment as an example of a packet that may be present without disrupting the Redundant primitive sequence A.

We acknowledge that the figures proposed in 17-135r0 also corrected the errors, but decided not to use the proposed figures in favor of keeping the figure structure and length closer to the original figures.

The group thanks Toshiba for identifying these errors to ensure the accuracy and integrity of the SPL4 standard.
SPL4 is now in the INCITS letter ballot process and any future proposals should be for SPL5.

Regards
Tim Symons
(INCITS/T10 SPL editor)


Tim Symons | Storage Architect
Microsemi Corporation
8555 Baxter Place, Burnaby, BC V5A 4V7. Canada
Tel. 604 415 6000
Tim.Symons at microsemi.com<mailto:Tim.Symons at microsemi.com>

[cid:image004.png at 01D1C17C.FB0C0730]<http://intranet/>
http://www.microsemi.com/






From: Tom Friend [mailto:Tom.Friend at taec.toshiba.com]
Sent: Monday, September 25, 2017 7:59 AM
To: T10, Reflector <T10 at t10.org<mailto:T10 at t10.org>>; Tim Symons <tim.symons at microsemi.com<mailto:tim.symons at microsemi.com>>
Subject: FW: Copy of revised chapter 6 showing approved changes

EXTERNAL EMAIL
I have received the following from our engineering department concerning the SPL-4 edits discussed in our last T10 F2F meeting. Can one of our SPL experts reply to this please?

TIA,

Tom Friend

Director, Industry Standards
Toshiba America Electronic Components, Inc.
SSD Business Unit
55 Trimble Road, San Jose, CA  95134
Mobile: +1(360)860-1831



From: kenji.yoshida at toshiba.co.jp<mailto:kenji.yoshida at toshiba.co.jp> [mailto:kenji.yoshida at toshiba.co.jp]
Sent: Sunday, September 24, 2017 6:24 PM
To: Tom Friend <Tom.Friend at taec.toshiba.com<mailto:Tom.Friend at taec.toshiba.com>>; toyohide1.isshi at toshiba.co.jp<mailto:toyohide1.isshi at toshiba.co.jp>; katsuhiko.takeuchi at toshiba.co.jp<mailto:katsuhiko.takeuchi at toshiba.co.jp>
Cc: koichi.nagai at toshiba.co.jp<mailto:koichi.nagai at toshiba.co.jp>; shisei1.fujiwara at toshiba.co.jp<mailto:shisei1.fujiwara at toshiba.co.jp>; tasuku.kasebayashi at toshiba.co.jp<mailto:tasuku.kasebayashi at toshiba.co.jp>; keiichi.yorimitsu at toshiba.co.jp<mailto:keiichi.yorimitsu at toshiba.co.jp>; daisuke.mito at toshiba.co.jp<mailto:daisuke.mito at toshiba.co.jp>; masafumi1.satou at toshiba.co.jp<mailto:masafumi1.satou at toshiba.co.jp>; Patrick Hery <Patrick.Hery at taec.toshiba.com<mailto:Patrick.Hery at taec.toshiba.com>>
Subject: RE: Copy of revised chapter 6 showing approved changes


Tom-san,



You first have to correct your misunderstanding of:

>>Scrambled Idle segments are not deletable



Scrambled idle segments are discarded at SP_PS receiver. Therefore, they are discarded (deleted) at

the phy layer and not sent to the link layer.

Same thing is mentioned by Mr. Brad Besmer at

http://www.t10.org/cgi-bin/ac.pl?t=d&f=17-151r0.pdf



He describes : A Scrambled Idle Segment is used for RATE MATCHING and the transmitter may add them

and the receiver simply discards them



He has to post it because Toshiba’s r1 is wrong. In Toshiba r1, scrambled idle segments are counted as

a part of consecutive dwords. (same mistake as original spl4r11)



So the modified figure in his proposal is consistent with the text.

However, I would rather bring the figure back to Toshiba r0 because Toshiba r0 is also correct and

consistent, and it also has the example of not detecting the second instance due to insufficient dwords.

Mr. Brad Besmer’s modification and Toshiba r2 don’t have this example. So Toshiba r0 figure is better.



So the history is like this:

spl4r11’s figure 131 & 133 : wrong

Toshiba r0’s figure 131 &133  : correct

Toshiba r1’s figure 131 &133 : wrong

Toshiba r2’s figure 131 &133 : correct, but less informative





Further, as I repeatedly say,



e) a scrambled idle segment.



should be added back in the text. So simply r0 is correct and there was no need to change.

If you still insist r0 needs to be changed, please specify which portion.



Regards,



Kenji Yoshida







From: Tom Friend [mailto:tom1.friend at glb.toshiba.co.jp]
Sent: Saturday, September 23, 2017 1:25 AM
To: yoshida kenji(吉田 賢治 TDSC ○SP統□HD技○HD六); isshi toyohide(一司 豊秀 TMC ○SSDジ□SSD応○SD応企); takeuchi katsuhiko(竹内 克彦 TMC ○SSDジ□ES技○ES二)
Cc: nagai koichi(永井 宏一 TMC ○SSDジ□SSD応○SD応企); fujiwara shisei(藤原 至誠 TMC ○SSDジ□SSD応○SD応二); kasebayashi tasuku(加瀬林 祐 TDSC ○SPジ□HDD応○HD応企); yorimitsu keiichi(依光 圭一 TDSC ○SPジ□HDD応○HD応企); mito daisuke(水戸 大輔 TDSC ○CSRD□ET開○ET3); satou masafumi(佐藤 正文 TDSC ○SPジ□HDD応○HD応企); Patrick Hery
Subject: RE: Copy of revised chapter 6 showing approved changes



Hello Yoshida-san,



Attached is the final version that was approved in the plenary yesterday. It has changed from the previously sent version. Please review this so that we can use this as the basis for all ongoing conversations.



There is one important thing you need to understand. Scrambled Idle segments are not deletable and they are 4 dwords in length. This is the main reason that so many changes were made to your corrections. Once that misunderstanding was taken into account, T10 was able to understand the rest of the errors in SPL-4 and so the attached file shows how they will make those corrections.



Thank you for your help with this,



Tom Friend



Director, Industry Standards

Toshiba America Electronic Components, Inc.

SSD Business Unit

55 Trimble Road, San Jose, CA  95134

Mobile: +1(360)860-1831









From: kenji.yoshida at toshiba.co.jp<mailto:kenji.yoshida at toshiba.co.jp> [mailto:kenji.yoshida at toshiba.co.jp]
Sent: Thursday, September 21, 2017 8:54 PM
To: toyohide1.isshi at toshiba.co.jp<mailto:toyohide1.isshi at toshiba.co.jp>; katsuhiko.takeuchi at toshiba.co.jp<mailto:katsuhiko.takeuchi at toshiba.co.jp>
Cc: Tom Friend <Tom.Friend at taec.toshiba.com<mailto:Tom.Friend at taec.toshiba.com>>; koichi.nagai at toshiba.co.jp<mailto:koichi.nagai at toshiba.co.jp>;  shisei1.fujiwara at toshiba.co.jp<mailto:shisei1.fujiwara at toshiba.co.jp>;  tasuku.kasebayashi at toshiba.co.jp<mailto:tasuku.kasebayashi at toshiba.co.jp>;  keiichi.yorimitsu at toshiba.co.jp<mailto:keiichi.yorimitsu at toshiba.co.jp>;  daisuke.mito at toshiba.co.jp<mailto:daisuke.mito at toshiba.co.jp>; masafumi1.satou at toshiba.co.jp<mailto:masafumi1.satou at toshiba.co.jp>; Patrick Hery <Patrick.Hery at taec.toshiba.com<mailto:Patrick.Hery at taec.toshiba.com>>
Subject: RE: Copy of revised chapter 6 showing approved changes



(I added some more HDD members in the loop.)



Tom-san,



Instead of modifying our correction requests w.r.t. 6.2.4.6 and 6.2.4.7

from 17-135r0 to 17-135r2 as you uploaded recently,

I would rather bring it back to r0.

In the r0 request, the figure contains both

“second instance is not detected due to insufficient consecutive dwords” and

“second instance is detected because of enough consecutive dwords”.

So the r0 figure is more informative than r2.

Furthermore, please modify back to original r0 for the below portion of 6.2.4.6 and 6.2.4.7



a) the original primitive;

b) a deletable primitive;

c) a deletable binary primitive;

d) a deletable extended binary primitive; or

e) a scrambled idle segment.



As you know,  the dwords in scrambled idle segments are all discarded at SP_PS

receiver and hence it cannot be a part of “consecutive dwords” so the red portion above

is needed.

Please let us know your opinion.



Best Regards,



Kenji Yoshida





From: isshi toyohide(一司 豊秀 TMC ○SSDジ□SSD応○SD応企)
Sent: Friday, September 22, 2017 11:13 AM
To: yoshida kenji(吉田 賢治 TDSC ○SP統□HD技○HD六); takeuchi katsuhiko(竹内 克彦 TMC ○SSDジ□ES技○ES二)
Cc: friend tom(TAEC ISSPI); nagai koichi(永井 宏一 TMC ○SSDジ□SSD応○SD応企); fujiwara shisei(藤原 至誠 TMC ○SSDジ□SSD応○SD応二); kasebayashi tasuku(加瀬林 祐 TDSC ○SPジ□HDD応○HD応企)
Subject: RE: Copy of revised chapter 6 showing approved changes



吉田さん



一司です。



ありがとうございます。



大変申し訳ありませんが、私はT10の内容までカバーできておりません。

英語でFeedbackいただけないでしょうか?

日本語でやり取りしていましたが、あて先にTomが入ったままでした。

スレッドにレスする形でOKです。





From: yoshida kenji(吉田 賢治  TDSC ○SP統□HD技○HD六)
Sent: Friday, September 22, 2017 10:13 AM
To: isshi toyohide(一司  豊秀 TMC ○SSDジ□SSD応○SD応企) <toyohide1.isshi at toshiba.co.jp<mailto:toyohide1.isshi at toshiba.co.jp>>; takeuchi katsuhiko(竹内 克彦  TMC ○SSDジ□ES技○ES二) <katsuhiko.takeuchi at toshiba.co.jp<mailto:katsuhiko.takeuchi at toshiba.co.jp>>
Cc: friend tom(TAEC ISSPI) <tom1.friend at glb.toshiba.co.jp<mailto:tom1.friend at glb.toshiba.co.jp>>; nagai koichi(永井 宏一  TMC ○SSDジ□SSD応○SD応企) <koichi.nagai at toshiba.co.jp<mailto:koichi.nagai at toshiba.co.jp>>; fujiwara shisei(藤原 至誠  TMC ○SSDジ□SSD応○SD応二) <shisei1.fujiwara at toshiba.co.jp<mailto:shisei1.fujiwara at toshiba.co.jp>>; kasebayashi tasuku(加瀬林 祐  TDSC ○SPジ□HDD応○HD応企) <tasuku.kasebayashi at toshiba.co.jp<mailto:tasuku.kasebayashi at toshiba.co.jp>>
Subject: RE: Copy of revised chapter 6 showing approved changes



一司さん



変更点は、6.2.4.6(2回目検出判断の除外条件追加)と、

6.2.4.7(2回目検出判断の除外条件追加)およびFigure133

ですが、この変更は誤りです。

元の提案通りに直していただく必要があります。

このFigure133修正でダメなことは、すでにT10でBrad Besmerさんから指摘されていますが、

http://www.t10.org/cgi-bin/ac.pl?t=d&f=17-151r0.pdf

そこでまた場当たり的な訂正をTomがしているので

http://www.t10.org/cgi-bin/ac.pl?t=d&f=17-135r2.pdf



これでもだめ(不十分)です。

すべて吉田の元の修正に戻していただきたいです。

吉田の元の修正は、Packetモード時にRedundantやTrippleの2回目を検出

「しない」例と検出「できる」例が両方含まれています。Tomの上げた17-135r2のほうは

「しない」例がありません。

また、除外条件の中に、Scrambled idle segment が含まれるべき

ですので、その点でも元の吉田の提案通りにすることをお願いします。



吉田



From: isshi toyohide(一司 豊秀 TMC ○SSDジ□SSD応○SD応企)
Sent: Friday, September 22, 2017 9:00 AM
To: takeuchi katsuhiko(竹内 克彦 TMC ○SSDジ□ES技○ES二); yoshida kenji(吉田 賢治 TDSC ○SP統□HD技○HD六)
Cc: friend tom(TAEC ISSPI); nagai koichi(永井 宏一 TMC ○SSDジ□SSD応○SD応企); fujiwara shisei(藤原 至誠 TMC ○SSDジ□SSD応○SD応二); kasebayashi tasuku(加瀬林 祐 TDSC ○SPジ□HDD応○HD応企)
Subject: RE: Copy of revised chapter 6 showing approved changes



竹内さん、吉田さん



(SD応企)一司です。



お世話になっております。



TEAC Tom  から、SPL-4の修正があり、Voting が必要だから確認してくれと

依頼がありました。

どこが変更されているかもよくわからないDraftが添付されてきたのですが、

これって、何ヶ月か前のSSS F2F meeting でTom にSAS SPL-4 の変更確認を

依頼した件かと思い、転送します。



Tomには1日で確認なんてできないとは伝えてはありますが、

Voting  はあるようなので、何らかの投票はなされるかもしれません。



何はともあれ、内容確認をお願いできますでしょうか?



お手数おかけいたしますが、よろしくお願いいたします。



From: isshi toyohide(一司 豊秀  TMC ○SSDジ□SSD応○SD応企)
Sent: Thursday, September 21, 2017 4:25 PM
To: 'Tom Friend' <tom1.friend at glb.toshiba.co.jp<mailto:tom1.friend at glb.toshiba.co.jp>>; nagai koichi(永井 宏一  TMC ○SSDジ□SSD応○SD応企) <koichi.nagai at toshiba.co.jp<mailto:koichi.nagai at toshiba.co.jp>>; fujiwara shisei(藤原 至誠  TMC ○SSDジ□SSD応○SD応二) <shisei1.fujiwara at toshiba.co.jp<mailto:shisei1.fujiwara at toshiba.co.jp>>
Subject: RE: Copy of revised chapter 6 showing approved changes



+Fujiwara-san,



Hello Fujiwara-san,

Could you please check the file and provide Tom feedbacks if you have?



TIA

Toyohide Isshi



From: Tom Friend [mailto:tom1.friend at glb.toshiba.co.jp]
Sent: Thursday, September 21, 2017 1:41 PM
To: isshi toyohide(一司  豊秀 TMC ○SSDジ□SSD応○SD応企) <toyohide1.isshi at toshiba.co.jp<mailto:toyohide1.isshi at toshiba.co.jp>>; nagai koichi(永井 宏一  TMC ○SSDジ□SSD応○SD応企) <koichi.nagai at toshiba.co.jp<mailto:koichi.nagai at toshiba.co.jp>>
Subject: FW: Copy of revised chapter 6 showing approved changes



Hello Isshi-san & Nagai-san,



Here are the changes that were implemented today to SPL-4 in the T10 CAP meeting. We will be voting to approve this tomorrow during the plenary session, so it is not finalized until approved during that meeting.



Warmest Regards,



Tom Friend



Director, Industry Standards

Toshiba America Electronic Components, Inc.

SSD Business Unit

55 Trimble Road, San Jose, CA  95134

Mobile: +1(360)860-1831







From: Tim Symons [mailto:tim.symons at microsemi.com]
Sent: Wednesday, September 20, 2017 8:24 PM
To: Tom Friend <Tom.Friend at taec.toshiba.com<mailto:Tom.Friend at taec.toshiba.com>>; Patrick Hery <Patrick.Hery at taec.toshiba.com<mailto:Patrick.Hery at taec.toshiba.com>>; John Geldman <John.Geldman at taec.toshiba.com<mailto:John.Geldman at taec.toshiba.com>>
Cc: Bill Martin <bill.martin at samsung.com<mailto:bill.martin at samsung.com>>
Subject: Copy of revised chapter 6 showing approved changes



Tom, John and Patrick,



Here is a copy of SPL4 chapter 6 with the changes that were approved by the WG today.

One caveat is that this need to be ratified at the plenary vote tomorrow.





Regards

Tim.





Tim Symons | Storage Architect

Microsemi Corporation

8555 Baxter Place, Burnaby, BC V5A 4V7. Canada

Tel. 604 415 6000

Tim.Symons at microsemi.com<mailto:Tim.Symons at microsemi.com>



cid:image004.png at 01D1C17C.FB0C0730<http://intranet/>

http://www.microsemi.com/



 (See attached file: image001.png)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.t10.org/pipermail/t10/attachments/20170927/7df75bbd/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.png
Type: image/png
Size: 10250 bytes
Desc: image001.png
URL: <http://www.t10.org/pipermail/t10/attachments/20170927/7df75bbd/attachment-0001.png>


More information about the T10 mailing list