[J3] JoR papers

[Bleikamp, Richard]

[AMD Official Use Only - Internal Distribution Only]

Hi,
Just a reminder, JoR will move 21-122r2 tonight (new IEEE min/max).

Rich

Paper status for JoR:
21-101r1 was uploaded last week, with only those edits that were voted on/passed last week.
21-115r3 and 21-117r3 have been uploaded.  They contain all the changes to the r2 versions as amended/passed last night.  Diffs follow:

rich at ORL-L10-BLEIKAM:~/files/j3$ diff 21-115r2.txt 21-115r3.txt
1c1
< To: J3                                                     J3/21-115r2
---
> To: J3                                                     J3/21-115r3
4c4
< Date: 2021-February-25
---
> Date: 2021-March-02
14c14
< intinisic from the other forms, and either just keeping the 2 forms
---
> intrinisic from the other forms, and either just keeping the 2 forms
190,191d189
<
<
rich at ORL-L10-BLEIKAM:~/files/j3$ diff 21-117r2.txt 21-117r3.txt
1c1
< To: J3                                                     J3/21-117r2
---
> To: J3                                                     J3/21-117r3
4c4
< Date: 2021-February-25
---
> Date: 2021-March-02
8c8,11
< This "r2" paper adds edits to the Introduction and Compatibility
---
> This "r3" papers rewords the NOTE that explains why using
> larger integer arguments to SYSTEM_CLOCK is a good thing.
>
> The "r2" paper added edits to the Introduction and Compatibility
20,21c23,24
< should always be consistant across all calls to system_clock(), if
< the values returned are used to compute elasped time.
---
> should always be consistent across all calls to system_clock(), if
> the values returned are used to compute elapsed time.
52,55c55,59
<      SYSTEM_CLOCK solely with integer arguments with the same kind type
<      parameter in all invocations will have the same interpretation
<      under this document as in Fortran 2018.  All other programs that
<      invoke SYSTEM_CLOCK may have a different interpretation under this
---
>      SYSTEM_CLOCK solely with integer arguments, always with the same
>      kind type parameter that has a decimal exponent range at least as
>      large as a default integer, will have the same interpretation
>      under this document as in Fortran 2018.  Other programs that invoke
>      SYSTEM_CLOCK could have a different interpretation under this
94c98,100
<      count.
---
>      count. The documentation of the Fortran processor should describe
>      the relationship between the kind type parameter of integer
>      arguments to SYSTEM_CLOCK and clock selection.
107,109c113,115
<   Using SYSTEM_CLOCK in this manner may minimize how often the
<   returned COUNT "rolls over" and the processor will typically select
<   the most accurate clock available.
---
>   Using SYSTEM_CLOCK in this manner lets the processor select the
>   most accurate clock available and minimize how often the COUNT
>   value resets to zero.
111c117
< [541:37+] add several new bulleted items (processor dependendcies)
---
> [541:37+] add several new bulleted items (processor dependencies)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.j3-fortran.org/pipermail/j3/attachments/20210302/4d0d3a19/attachment.htm>