E00132r0
D1386 Letter Ballot Comments Resolution
Comments to accompany the Maxtor Corporation yes vote on the letter ballot for project 1386D.
1m. Title page, Abstract, first sentence, add a hyphen to IA64, i.e., IA-64.
a. Accepted
2m. Page after title page, patent statement, why should we see clause 2?
a. Standard boiler plate?
3m. Table of contents, tables , and figures are in a different font than the rest of the document.
a. Will change from Times New Roman to Arial
4m. Pages should be number starting with the value i on the first page of the table of contents.
a. Will make the first page number appear
5m. The table of tables has no figure 1.
a. Will fix the figure numbering
6m. Page ii, starting with the second paragraph the Forward is in a different font.
a. Will change font from times new roman to arial
7m. Page iv, Introduction, some of the sentences are not sentences and some have no period at the end.
a. Will remove the periods
b. Will change clause 4 to read “Clause 4 is the overview”
8m. Page 1, Scope, bulleted list, the fifth, sixth, and seventh items should end with a semicolon, the last item should end with a period.
a. Accepted
9m. Page 2, starting with 2.1, the x.x headings are in the wrong font.
a. Will change template to Arial
10m. Page 2, 2.1 Approved references, can technical reports be normative references?
a. Good question??
11m. Page 2, 2.2 References under development, 1394a should move up to the approved list. It is now IEEE 1394a-2000 IEEE Standard for a High Performance Serial Bus – Amendment 1.
a. Accepted
12m. Global, do “keep withs” so that headings don’t end by themselves at the bottom of a page.
a. Point taken
13m. Page 5, 3.2.13 INT 13h, Floppy, Hard Drive, should not be capitalized.
a. Accepted
14m. Page 5, 3.2.14 INT 40 should be INT 40h.
a. OK, do we want to put in the h on everything?
15m. Page 5, 3.2.17 Logical address/geometry, global, the document sometimes uses “FN”, sometimes uses “Fn”, and sometimes uses “Function”. Choose one and make it consistent.
a. Will change all references to Fn
16m. Page 5, 3.2.18 NV memory, first sentence should say “Non-Volatile memory is memory that retains content even when the power has been shut off.
a. Good point
17m. Page 6, 3.2.20 Default address/geometry is out of alphabetical order.
a. Will Fix
18m. Page 7, clause 4 Overview, Change INT 13 to INT 13h. This is a global change.
a. Will do
19m. Page 8, clause 4 Overview, second paragraph, second sentence, change to “This standard builds on NCITS TR-21 BIOS Enhanced Disk Drive Technical Report to enable other mass storage technologies, such as 1394, Fibre Channel, and USB.”
a. Will Do
20m. Page 8, clause 4 Overview, fourth paragraph, first sentence, change to “Data written on media can render the media incompatible with certain drive letters when some drive letter based operating systems are used.”
a. OK
21m. Page 8, clause 4 Overview, fifth paragraph, first sentence, change to “One of the important aspects of this standard is to allow a BIOS to describe the physical path to a device.”
a. OK
22m. Page 8, clause 4 Overview, figure 2, this is the first figure in the document.
a. Will fix
23m. Page 9, 5.1 Data structure, table 1 should be in this subclause not the next one.
a. Will move the table to the correct position
24m. Page 10, table 1, offset 4 description, change to “Address of host transfer buffer. The host buffer that Read/Write operations will use to transfer the data. This is a 32-bit host address of the form Seg:Offset. If this field is set to FFFFh:FFFFh then the address of the transfer buffer is found at offset 10h.”
a. OK
25m. Page 10, table 1, offset 10 description, change to “64 bit unsigned linear address of the host transfer buffer. This is the host buffer that Read/Write operations will use to transfer the data if the data at offset 4 is set to FFFFh:FFFFh, or the data at offset 2 is set to FFh.
a. OK
26m. Page 12, table 3, this should be table 2.
a. Interesting, my doc reads correctly
27m. Page 12, 6.3 Extended write, first sentence should say “This function transfers sectors from memory to the device.” The last sentence needs a period.
a. OK
28m. Page 15, table 5, this should be table 3.
a. My doc reads correctly
29m. Page 15, table 5, offset 0 description should say “The caller sets this value to the maximum buffer length in bytes. If the length of this buffer is less than 30 bytes, this function does not return the pointer to DPT extension. If the buffer length is 30 or greater on entry, it shall be set to 30 on exit. If the buffer length is between 26 and 29, it shall be set to 26 on exit. If the buffer length is less than 26 on entry an error is returned.
a. OK
30m. Page 15, table 5, offset 32 description, is the value 44 in fact 44h?
a. This is a decimal number, as are the offsets
31m. Page 16, table 7, this should be table 4.
a. Fixed
32m. Page 16, table 7, font is different.
a. OK
33m. Page 17, table 8, this should be table 5.
a. Fixed
34m. Page 17, table 8, font is different.
a. OK
35m. Page 18, table 9, should be table 6.
a. Fixed
36m. Page 18, 6.8.3.2, is this the base address of the set of Control Block registers or is this the address of the Device Control register?
a.
37m. Page 19, 6.8.3.7, global change multi-word to multiword.
a. OK
38m. Page 19, 6.8.3.7, change 0 to zero and 2 to two.
a. OK
39m. Pages 19 and 20, subclauses of 6.8.3.9, change to “bit set to one’ and bit cleared to zero”.
a. OK
40m. Page 20, 6.8.3.9.7, change to “If this ATA device implements the PACKET command feature set (ATAPI) as defined in ATA/ATAPI-4, this bit is set to one.”
a. OK
41m. Page 20, table 10, this should be table 7.
a. Fixed
42m. Page 21, table 12, should be table 8.
a. Fixed
43m. Page 21, 6.10, change “2” to “two”, “1” to “one”, and “0” to “zero”.
a. OK
44m. Page 22, table 13, should be table 9.
a. Fixed
45m. Page 22, table 13, font is different.
a. OK
Quantum’s comments for the letter ballot of T13/1386D, rev 4 (EDD) Some comments include more than one suggestion or question.
1q. Global, beginning on page 3 (DOCUMENT STATUS, Revision 2): The font of the document starts out just in Arial. Then other fonts are used beginning with the bulleted list under Revision 2. The font in this document should be changed to be consistent, and, I believe it should be changed to Arial.
a. Will change to Arial
2q. Second cover page (PDF page 4) at top: The temporary NCITS document number should reflect year 200X, not 199X.
a. Yeah..
3q. Second cover page (PDF page 4) Foreword: Change "BIOS provided INT 13 interface" to "INT13 programming interface provided by BIOS firmware to higher level software". This distinguishes a software interface from a hardware interface and simplifies the wording.
a. OK
4q. PDF page 5, CAUTION: Change "holder's" to "holders" to represent plural form of the noun.
a. OK
5q. Abstract (PDF page 8): Change "personal computers systems" to "personal computer systems" to correct grammar. Both words should not be plural.
a. OK
6q. page iii (PDF page 9), Foreword: At the time this document was submitted for letter ballot, Dan Colegrove had resigned from T13 and Mark Evans was the secretary.
a. OK
7q. Global (first occurrence, page 1, PDF page 11): It appears as if some of the characters did not translate properly from the Word document to the PDF document. At least some dashes and bullets appear as one or two question marks in the PDF document.
a. OK
8q. Page 1 (PDF page 11) Scope: Add a sentence to the first paragraph saying something like, "To comply with this standard, higher-level software (the “caller”) or lower-level BIOS firmware (the “callee”) shall support the functions and data structures described in this document." This is to cover the lack of "shall" in many places and to describe the entities responsible for implementing the standard. Ideally, a software interface would describe each requirement as, "the caller shall perform software interrupt ## and provide these parameters", and, "the callee shall provide this data structure", etc.
a. How about: To comply with this standard, higher level software shall call the INT functions using the data structures described herein; and system firmware shall provide the INT functions and data structures described herein
9q. Page 2 (PDF page 12), 2.2 References under development: Add document titles and parenthesize acronyms to read "AT Attachment with Packet Interface (ATA/ATAPI-5)", "SCSI Primary Commands - 2 (SPC-2)", and Multi-Media Commands - 3 (MMC-3)".
a. OK
10q. Page 2 (PDF page 12), 2.3 Other references: The WWW link for Mass Storage Overview is not fully underlined.
a. OK
11q. page 3 (PDF page 13), Definitions, abbreviations, and conventions: The title of this clause is, “Definitions, abbreviations, and conventions”. However, the subclauses are, “Keywords” and “Definitions and abbreviations”. One way or the other, this should be made consistent.
a. How about we delete the key words, I don’t think they are used
12q. page 3 (PDF page 13), Keywords: It is more typical for me to see the format of definitions in standards to be “word_to_be_defined: definition words.” Implementing this format in addition to implementing the recommendation in Quantum comment 13 might save a page in the document.
a. See 11
13q. page 3 (PDF page 13), Keywords: I searched both the .doc and .pdf files and could find no occurrence of the words “expected”, “obsolete”, “retired”, or any of their variants. Since these words are not used in the standard, I would recommend deleting them.
a. See 11
14q. Global (first occurrence, page 4, PDF page 14, 3.2.1 ATA): Italics are used in several places. First of all, I don’t think Italics are allowed in NCITS standards. Second, they are used inconsistently for this purpose. The following example is the format that I recommend should be used in this clause (please note as to how the abbreviation and its “spelled out” name are handled; only the abbreviation should be used from this point forward in the document): 3.2.1 ATA (Advanced Technology Attachment): An ATA drive, also known as an IDE drive, is a hard drive that conforms to an ATA standard.
a. Will remove the italics from the definitions
15q. Page 4 (PDF page 14, 3.2.4: The term "Boot Device" is not used in the document, so this definition is unnecessary and should be deleted.
a. OK
16q. PDF page 4, (PDF page 15) 3.2.6 CHS: I think this should be changed to something like, “CHS addressing: CHS addressing is a method of addressing the contents of a storage device using logical cylinders (C), logical heads (S), and logical sectors (S). This method of accessing allows a maximum C=16383, H=16, S=63, or 8.4GB. See LBA addressing for another addressing method.”
a. OK
17q. Global (first occurrence, page 4, PDF page 14, 3.2.6 CHS): The term "access" is so general it can refer to connecting or even turning on the power. Because of this, the document should be searched for other occurrences, and the following changes should be made where appropriate: "accessing" to "addressing", "access" to "addressing", and "accessed" to "addressed" to be more specific.
a. Is this going a bit far??
18q. Page 4 (PDF page 14), 3.2.7 Conventional vs enhanced (this is the first occurrence of what should begin a global search of the document): The “which” shall be changed to a “shall”. Please search the document and replace all “which”s accordingly.
a. I disagree, but what do I know??
19q. page 4 (PDF page 14), 3.2.8 DOS [this may be a global]: I think that a “TM” or a “REG TM” or something needs to be used with MS-DOS, et al., as appropriate.
a. I can place a TM by these…
20q. Page 4/5 (PDF page 14/15), 3.2.9 DWord: Clarify the meanings of "first" and "last" since they usually refer to time. Replace the sentence with "Byte 0 of a Dword is stored in the lowest byte address and Byte 3 is stored in the highest byte address".
a. OK
21q. Page 5 (PDF page 15), 3.2.11 IA-32 (the first of several): This is the first of several definitions in this subclause that are incomplete sentences. I recommend that this be changed to, “IA-32: IA-32 refers to the Intel Architecture 32-bit wide processor data bus.” The sentence in the subclause following this subclause and others in this clause should be made into complete sentences
a. OK. Will change IA-32 and IA-64 definitions
22q. Page 5 (PDF page 15), 3.2.16 LBA: This definition is not correct. An LBA is a logical block address. I think that this should be the definition for this term. In addition, I think there should then be a definition for “LBA addressing”. This could be: “LBA addressing is a method of addressing the contents of a storage device as a contiguous set of LBAs. This method of accessing allows a maximum LBA of 2^28 -1, or 137.4GB of data. See CHS addressing for another addressing method.”