[cgl_discussion] CGL RD 2.0 doc review

John Mehaffey mehaf at mvista.com
Tue Dec 2 12:28:46 PST 2003


Just a couple more, then I'll leave you alone -
Added items II(5) and V(4) (below).

Also, it is not clear where v1.1 ID 4.13 (fast reboot) should go -
is it an ignored requirement, or is it intended to be part of
AVL.fast.1 (already assumed to include 4.7)? My personal preference
would be to create a new AVL.fast.x for it.

Best Regards,
John Mehaffey
MontaVista Technical Marketing

John Mehaffey wrote:

> OK, I have completed most of my study of the 2.0 CGL RD (Version
> 2.0, Ratified Publication Version - 9/23/2003), comments are in
> reference to 1.1 CGL RD (Version Date 10/31/02).
> 
> This email contains all of my previously submitted comments.
> 
> Best Regards,
> John Mehaffey
> MontaVista Technical Marketing
> 
> I. Incorrect references:
>   1. Reference to wrong section
>     a) PLT.1.0 should reference 2.1.3
>     b) AVL.3.2 should reference 3.8.1
>     c) AVL.3.3 should reference 3.8.2
>     d) AVL.3.4 should reference 3.8.3
>     e) AVL.3.6 should reference 3.8.4
>     f) AVL.3.7 should reference 3.8.4
>     g) AVL.4.0 should reference 4.10
>     h) PLT.4.0 should reference 2.3
>     i) PLT.5.0 should reference 2.6.2
>     j) SVC.2.0 should reference 4.9.3
>     k) SVC.5.2 should reference 4.12
>     l) PRF.3   should reference 6.3
>     m) TLS.1.0 should reference 5.2
>     n) TLS.2.1 should reference 5.4.2
>     o) TLS.2.2 should reference 5.4.3
>     p) CFH.5.0 should reference 3.9
>     q) Section 8, ID 3.1 should be 3.1.1
>     r) CSM.cfs should reference 3.12
>     s) SVC.diags.0 should reference 4.9
> 
>   2. Reference to non-existent section
>     a) AVL.3.1
>     b) AVL.3.5
>     c) AVL.3.8
>     d) SVC.5.1
>     e) PRF.3.1
>     f) PRF.3.2
>     g) Section 8, ID 35.2 (no mention of requirement in v1.1)
>     h) CSV.dlm
>     i) CSV.ev
> 
>   3. Reference missing
>     a) PRF.4.0 should reference 6.9
>     b) PRF.5.0 should reference 6.10
>     c) PRF.8.0 should reference 6.8 (P2 in v1.1)
>     d) AVL.fast.1 should reference 4.7 (P2 in v1.1)
> 
> II. Typos
>   1. STD.2.20
>     a) Statement that "This functionality is used in POSIX Timers" is
>        incorrect.  Suggest using "POSIX Threads", or deleting comment.
>   2. Table 7 (page 94) - Requirements Carried Forward Unchanged
>     a) 2.1.1 Requirement name was "Hot Insert"
>     b) 2.1.2 Requirement name was "Hot Remove"
>     c) 3.21 should be 3.2.1
>     d) 3.21 Requirement name was "Watchdog Timer Interface Requirements"
>   3. ACC.3.0
>     a) "Buffer Overflows" is awkward, suggest "Buffer Overflow"
>   4. CON.6.0
>     a) "Secure Integrity Verification Of Binaries Before Loading Them" is
>        awkward, suggest "Verification of Binaries Before Loading"
   5. PLT.atca
     a) "Advanced TCA IPMI Support" should just be "Advanced TCA Support"
> 
> III. Missing sequence numbers
>     a) CFH.1
>     b) CON.4
>     c) CSM.1
>     d) SVC.4
> 
> IV. Ignored v1.1 Requirements
>     a) 1.3   (P1) POSIX 1003.25 Event Logging
>     b) 1.7.1 (P2) IPv6 additional RFCs
>     c) 1.7.2 (P2) IPSecv6 additional RFCs
>     d) 1.7.3 (P2) MIPv6 additional RFCs
>     e) 1.9   (P3) SAF Proposed Standards Compliance
>                   (Part of standard included, see PLT.3.0, CCM.2)
>     f) 2.1.3.1 (P1) System Device Enumeration Specification
>     g) 2.1.3.2 (P1) System Device Enumeration Framework
>     h) 2.7   (P2) Automatic Alternate Boot
>     i) 2.8   (P2) Hyperthreading of CPUs
>     j) 3.1.2 (P1) Sample Hardened Device Driver
>     k) 3.2.2 (P2) Watchdog Timer Pre-Timeout Interrupt
>     l) 4.1.1 (P1) Resource Monitor Specification
>     m) 4.1.2 (P1) Resource Monitor Framework
>     n) 4.1.3 (P1) Resource Monitor Subsystems
>     o) 4.1.4 (P1) Resource Monitor Data Persistence
>     p) 4.1.5 (P1) Resource Monitor Performance
>     q) 5.3.2 (P2) Kernel Debugger Authentication
>     r) 5.8   (P2) Multithreaded Core Dump Support for Threaded Applications
>     s) 6.4.1 (P2) Concurrent Timers Scaling Behavior and Report
>     t) 6.4.2 (P2) Concurrent Thread Scaling Behavior and Report
>     u) 6.4.3 (P2) Concurrent Process Scaling Behavior and Report
>     v) 6.5.1 (P2) Enhancements for High Concurrent Timer Scaling
>     w) 6.5.2 (P2) Enhancements for High Concurrent Thread Scaling
>     x) 6.5.3 (P2) Concurrent Process Scaling
>     y) 6.5.4 (P2) SMP and Lock Contention Scaling
>     z) 6.6   (P2) Kernel I/O Performance Analysis
> 
> V. Parallel Structure
>    1. Capitalization inconsistent for prepositions (of, and, for) in
>       Requirement Names
>       Most of doc does not capitalize prepositions, but many security
>       definitions capitalize all words (starting with CON.3.0)
>    2. Support
>       Most of the document does not use the word "support", but
>       (especially later in the document) many requirements use support,
>       support for, or support of.  Recommend dropping the word support
>       where possible, and parallel usage where not (ie just use
>       "<technology> support", rather than a mixture of "support of
>       <technology>", "support for <technology>", etc.
>       Example (page 81): CON.2.0 "Support for IKE" could just be
>       "Internet Key Exchange" or even just "IKE"
>       Example2 (same page): CON.5.0 "PF_KEY support" could be
>       "PF_KEY Management"
>    3. Priority sections
>       Many of the items in the Priority 1 Requirements part of the
>       General Systems Requirements section are actually priority 2,
>       and even priority 3.
>       (Examples: AVL.3.6, AVL.3.7, SVC.1.6)
    4. Keep Header with body
       Page 101-102, the header for PLT.atca.3 is separated from the body.
> 
> VI.  ID issue for items carried forward
>      Items carried forward (Section 7) should have IDs consistent with
>      the IDs of the rest of the 2.0 specification. For Example, category
>      platform, item 2.2 (Remote boot support) should have a PLT.x.y
>      number.
> 
> 
> _______________________________________________
> cgl_discussion mailing list
> cgl_discussion at lists.osdl.org
> http://lists.osdl.org/mailman/listinfo/cgl_discussion
> 
> 




More information about the cgl_discussion mailing list