Check for NaNs and INFs in input to shr_reprosum_calc #1703
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
In recent development NaNs and INFs are often first identified when passed to the shr_reprosum_calc routine, where they either lead to segmentation faults or to very slow performance (due to the interaction of NaNs and INFs with the reproducible sum logic). To more readily identify this error condition (and to prevent false attributions of error to shr_reprosum_calc), the input array is checked for the presence of NaNs and INFs. If found, an appropriate error message is output and the job is terminated. [BFB] *********1*********2*********3*********4*********5*********6*********7** Longer commit message body describing the commit. Can contain lists as follows: * Item 1 * Item 2 * Item 3 A good commit message should be written like an email, a subject followed by a blank line, followed by a more descriptive body. Can also contain a tag at the bottom describing what type of commit this is. [BFB] - Bit-For-Bit [FCC] - Flag Climate Changing [Non-BFB] - Non Bit-For-Bit [CC] - Climate Changing [NML] - Namelist Changing See confluence for a more detailed description about these tags.
|
Tested with -compset A_WCYCL2000 -res ne30_oEC on Titan and using PGI. Was BFB (in that the error test was not true, so everything worked as if the test was not there). I also modified the code to insert both some NaNs and some INFs, and the test reported these and then killed the job, as designed. |
amametjanov
reviewed
Aug 7, 2017
| @@ -338,8 +339,14 @@ subroutine shr_reprosum_calc (arr, arr_gsum, nsummands, dsummands, & | |||
| logical :: validate ! flag indicating need to | |||
| ! verify gmax and max_levels | |||
| ! are accurate/sufficient | |||
| integer :: nan_check, inf_check ! flag on whether there are | |||
There was a problem hiding this comment.
This was probably meant to be logical :: nan_check, inf_check.
There was a problem hiding this comment.
yes - I'll update immediately. My tests still worked though ...
There was a problem hiding this comment.
just got lucky that assigning logical output to an integer and then testing using the integer gave the same result? At least it was consistent (assignment of logical to integer and evaluating integer as a logical).
amametjanov
approved these changes
Aug 7, 2017
New logical variables nan_check and inf_check were mistakenly declared as integer.
amametjanov
added a commit
that referenced
this pull request
Aug 7, 2017
…1703) Check for NaNs and INFs in input to shr_reprosum_calc: In recent development NaNs and INFs are often first identified when passed to the shr_reprosum_calc routine, where they either lead to segmentation faults or to very slow performance (due to the interaction of NaNs and INFs with the reproducible sum logic). To more readily identify this error condition (and to prevent false attributions of error to shr_reprosum_calc), the input array is checked for the presence of NaNs and INFs. If found, an appropriate error message is output and the job is terminated. [BFB]
rljacob
approved these changes
Aug 8, 2017
amametjanov
added a commit
that referenced
this pull request
Aug 9, 2017
Check for NaNs and INFs in input to shr_reprosum_calc: In recent development NaNs and INFs are often first identified when passed to the shr_reprosum_calc routine, where they either lead to segmentation faults or to very slow performance (due to the interaction of NaNs and INFs with the reproducible sum logic). To more readily identify this error condition (and to prevent false attributions of error to shr_reprosum_calc), the input array is checked for the presence of NaNs and INFs. If found, an appropriate error message is output and the job is terminated. [BFB]
jgfouca
added a commit
that referenced
this pull request
Oct 3, 2017
…amelist-parsing-order Fix bug in case.run where arg skip-preview-namelist is out of sequence arg parsing needs to be done after adding arguments master is broken on acme mira/cetus and blues/anvil without this fix Test suite: acme_developer on anvil Test baseline: Test namelist changes: Test status: Fixes #1690 User interface changes?: Update gh-pages html (Y/N)?: Code review:
jgfouca
pushed a commit
that referenced
this pull request
Oct 25, 2017
Check for NaNs and INFs in input to shr_reprosum_calc: In recent development NaNs and INFs are often first identified when passed to the shr_reprosum_calc routine, where they either lead to segmentation faults or to very slow performance (due to the interaction of NaNs and INFs with the reproducible sum logic). To more readily identify this error condition (and to prevent false attributions of error to shr_reprosum_calc), the input array is checked for the presence of NaNs and INFs. If found, an appropriate error message is output and the job is terminated. [BFB]
jgfouca
pushed a commit
that referenced
this pull request
Feb 27, 2018
Check for NaNs and INFs in input to shr_reprosum_calc: In recent development NaNs and INFs are often first identified when passed to the shr_reprosum_calc routine, where they either lead to segmentation faults or to very slow performance (due to the interaction of NaNs and INFs with the reproducible sum logic). To more readily identify this error condition (and to prevent false attributions of error to shr_reprosum_calc), the input array is checked for the presence of NaNs and INFs. If found, an appropriate error message is output and the job is terminated. [BFB]
jgfouca
pushed a commit
that referenced
this pull request
Mar 14, 2018
Check for NaNs and INFs in input to shr_reprosum_calc: In recent development NaNs and INFs are often first identified when passed to the shr_reprosum_calc routine, where they either lead to segmentation faults or to very slow performance (due to the interaction of NaNs and INFs with the reproducible sum logic). To more readily identify this error condition (and to prevent false attributions of error to shr_reprosum_calc), the input array is checked for the presence of NaNs and INFs. If found, an appropriate error message is output and the job is terminated. [BFB]
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
In recent development NaNs and INFs are often first
identified when passed to the shr_reprosum_calc routine,
where they either lead to segmentation faults or to
very slow performance (due to the interaction of NaNs and
INFs with the reproducible sum logic). To more readily
identify this error condition (and to prevent false
attributions of error to shr_reprosum_calc), the input array
is checked for the presence of NaNs and INFs. If found,
an appropriate error message is output and the job is
terminated.
[BFB]