Merge the maint-cime5.1 branch including ACME 5.1 changes #1065
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
The env module system on melvin changed over the weekend. [BFB]
The env module system on melvin changed over the weekend. [BFB]
Update melvin env The env module system on melvin changed over the weekend. [BFB] * jgfouca/fix_melvin_env: Update melvin env
Fix HOMME cmake files based on env changes on melvin [BFB] * jgfouca/fix_melvin_homme: Fix HOMME cmake files based on env changes on melvin
In the previous PR #1077 ATM_NCPL and ROF_NCPL were not set properly for ne4_ne4 and ne11_ne11 grids. The new values for these grids allows for much faster run times. [BFB]
In the previous PR #1077 ATM_NCPL and ROF_NCPL were not set properly for ne4_ne4 and ne11_ne11 grids. The new values for these grids allows for much faster run times. [BFB]
[BFB] - Bit-For-Bit
[BFB] - Bit-For-Bit
Fix for ATM_NCPL and ROF_NCPL In the previous PR #1077 ATM_NCPL and ROF_NCPL were not set properly for ne4_ne4 and ne11_ne11 grids. The new values for these grids allows for much faster run times. Nightly testing revealed that it is non-BFB for ERS.ne11_oQU240.I20TRCLM45 test. BFB has been changed to non-BFB for the master merge. [non-BFB] * erichlf/cam/FC5AV1C04-ne4ne4: removed/added some white spaces to get rid of diffs due to white spaces Fix for ATM_NCPL and ROF_NCPL
Fix for ATM_NCPL and ROF_NCPL In the previous PR #1077 ATM_NCPL and ROF_NCPL were not set properly for ne4_ne4 and ne11_ne11 grids. The new values for these grids allows for much faster run times. Nightly testing revealed that it is non-BFB for ERS.ne11_oQU240.I20TRCLM45 test. BFB has been changed to non-BFB for the master merge. [non-BFB] * erichlf/cam/FC5AV1C04-ne4ne4: removed/added some white spaces to get rid of diffs due to white spaces Fix for ATM_NCPL and ROF_NCPL
Changed the threshold value in QNEG3 from 1.e-12 to 1.e-8 Fixes #1085 [BFB] * kaizhangpnl/atm/QNEG3: Changed the threshold value in QNEG3 from 1.e-12 to 1.e-9
in components/cam/src/physics/rrtmg/ext/rrtmg_mcica/rrtmg_lw_rtrnmc.f90
- fix the case where the diffusivity angle was going negative under very
moist conditions
- the code has been changed so the interpolation for the lw ice optics has
been factored out and is now used for for the ice and snow cloud optics.
So the mod to limit the ice optical depths is now applied to both ice and
snow clouds.
in components/cam/src/physics/rrtmg/cloud_rad_props.F90
- Limit the snow/ice optical depths fix the case where there are large masses
of snow at high altitudes
Testing results from NCAR:
http://webext.cgd.ucar.edu/FCLIMO/f.e15.F2000.f09_f09.pd.100_rrtmg_cam5_4_79/atm/f.e15.F2000.f09_f09.pd.100_rrtmg_cam5_4_79-f.e15.F2000.f09_f09.pd.100_cam5_4_79/
Comments from Rich Neale:
- from https://acme-climate.atlassian.net/wiki/display/ATM/2016-10-06+Meeting+notes+-+General+meeting
"High temperature runs in CESM1 were sporadically crashing (in RCP8.5 extensions
for example). After a lot of work this was diagnosed as erroneous calculation of
optical properties in the radiation for very high ice water comments, as you
would see in grid point storms. It manifests as a negative surface flux of
downwelling long-wave radiation."
[non-BFB]
…station_lib_ver_updates Updating modules on anlworkstation - maint-cime5.1
…r (PR #1098) Update cmake files for sandia machines for Homme [BFB] * jgfouca/update_homme_cmake_files_for_sandia: Update cmake files for sandia machines for Homme
gold2718
reviewed
Jan 29, 2017
utils/python/CIME/test_scheduler.py
Outdated
| @@ -574,13 +574,22 @@ def _run_catch_exceptions(self, test, phase, run): | |||
| return False | |||
|
|
|||
| ########################################################################### | |||
| def _get_procs_needed(self, test, phase, no_batch=False): | |||
| def _get_procs_needed(self, test, phase, threads_in_flight=None, no_batch=False): | |||
There was a problem hiding this comment.
Is the problem which caused this to be turned off resolved?
|
@mvertens ACME made this addition to namelist_defaults_drv.xml: And this addition to namelist_definition_drv.xml |
|
@rjacob - all that is needed is to add the following entry to
cime_config/config_definition_drv.xml
<entry id="gust_fac">
<type>real</type>
<category>control</category>
<group>seq_infodata_inparm</group>
<desc>wind gustiness factor</desc>
<values>
<value>0.0D0</default>
</values>
</entry>
You do not need to change cime_config/buildnml - since this with the new nmlgen this will automatically get picked up in generating the driver namelist. Let me know
if you have any problem with implementing this change.
…On Sun, Jan 29, 2017 at 8:20 PM, Robert Jacob ***@***.***> wrote:
@mvertens <https://github.com/mvertens> ACME made this addition to
namelist_defaults_drv.xml:
<flux_diurnal>.false.</flux_diurnal>
+<gust_fac>0.0D0</gust_fac>
And this addtion to namelist_definition.xml
+<entry>
+id="gust_fac"
+type="real"
+category="control"
+group="seq_infodata_inparm">
+wind gustiness factor
+default: 0.0D0
+</entry>
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#1065 (comment)>, or mute
the thread
<https://github.com/notifications/unsubscribe-auth/AHlxE5bRV8v_EBXGSExrKLrBzFDYvJM4ks5rXVcBgaJpZM4LwW3l>
.
|
Upstream merge from master now that dashboard is green again.
* master: (28 commits)
update netcdf on cheyenne
fix typo in xml
change args to match
update create_newcase --mach to --machine to match create_test
reconnect timer level and depth to xml variables
ERR XML settings need CONTINUE_RUN=FALSE
Add ESMF_TimeIntervalProdI8 routine to multiply a time interval by an 8-byte int
fix issues with lt_archive
add back the mask grid dropped in v2 but used by components
fix expect statement for directive nesting error
fix pylint issue
revert to $ENV style in config_compilers
remove incorrect comment
add tests for new syntax
fix regression tests
update config_compilers to use $ENV{} $SHELL{} and ${} syntax
Revert "Change over acme env usage to correct syntax"
updates to template and ChangeLog
fix documentation error
update changelog
...
|
Thanks. I think that is a good idea.
…On Sat, Feb 4, 2017 at 2:48 PM, Robert Jacob ***@***.***> wrote:
***@***.**** commented on this pull request.
------------------------------
In share/csm_share/shr/shr_flux_mod.F90
<#1065>:
> @@ -267,7 +277,15 @@ real(R8) :: psix2 ! stability function at ztref reference height
if (mask(n) /= 0) then
!--- compute some needed quantities ---
- vmag = max(umin, sqrt( (ubot(n)-us(n))**2 + (vbot(n)-vs(n))**2) )
+
This is getting a little long for the PR review so I'll move it to a
separate issue. #1105 <#1105>
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#1065>, or mute the thread
<https://github.com/notifications/unsubscribe-auth/AHlxE-kuRSqOgaI6fRUAIIbO2wlnJQLJks5rZPIrgaJpZM4LwW3l>
.
|
|
@rljacob unforunately, I'm pretty useless when it comes to the fortran driver/coupler code, so I'm going to hand this branch to you. I think that all the python and XML has been correctly merged. |
|
I think that for the purposes of this PR all that needs to be done is:
1) Change line 486 of seq_infodata_mod.F90
gust_fac = huge(1.0_SHR_KIND_R8) <==== current
gust_fac = 0._r8 <==== new
at line 486 of seq_infodata_mod.F90.
2) Change line 910 of seq_flux_mct.F90:
real(r8) :: gust_fac = huge(1.0_r8) !wind gust factor <====
current
real(r8) :: gust_fac <==== new
since you are obtaining gust_fac from seq_infodata in line 910 of
seq_flux_mct.F90.
call seq_infodata_GetData(infodata, &
read_restart=read_restart, &
dead_comps=dead_comps, &
atm_nx=atm_nx, atm_ny=atm_ny, &
ocn_nx=ocn_nx, ocn_ny=ocn_ny, &
ocn_prognostic=ocn_prognostic, &
flux_diurnal=flux_diurnal, &
gust_fac = gust_fac )
Moving forwards, we will be getting input from CESM scientists as to how
they want to see this science change implemented to satisfy CESM
requirements. However, that can be put into a upcoming PR.
Mariana
…On Mon, Feb 6, 2017 at 11:54 AM, James Foucar ***@***.***> wrote:
@rljacob <https://github.com/rljacob> unforunately, I'm pretty useless
when it comes to the fortran driver/coupler code, so I'm going to hand this
branch to you. I think that all the python and XML has been correctly
merged.
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#1065 (comment)>, or mute
the thread
<https://github.com/notifications/unsubscribe-auth/AHlxE8jj8ECJLl7wj--P50ZvHWOZ81_tks5rZ2xrgaJpZM4LwW3l>
.
|
Restore shr_orb_mod.F90 to ESMCI master version that includes bug fix ACME isn't using yet.
|
With the conclusion in issue #1105 , everything in this PR has been agreed to except for the changes to ESMF_TimeMod.F90 by @douglasjacobsen . What should we do about those? |
|
A test merge to master shows only 3 files with conflicts: The first 2 are pretty easy. In seq_infodata_mod.F90, git is confused by the gust_fac addition and the ESP additions. I accepted the "HEAD" version and then had to put declarations of gust_fac back in a few routines. So if we can decide on the ESMF_TimeMod changes, this is ready to go. |
|
@rjacob - I don't know enough about ESMF_TimeMod.F90 to determine if this
will effect CESM or not. Why was this change brought in? Is it really
needed in this PR?
I am really uncomfortable in bringing in an change to esmf_wrf_timemgr
without input from Tony - who is currently the main expert on this.
…On Mon, Feb 6, 2017 at 9:18 PM, Robert Jacob ***@***.***> wrote:
A test merge to master shows only 3 files with conflicts:
both modified: cime_config/acme/machines/config_compilers.xml
both modified: cime_config/acme/machines/config_machines.xml
both modified: driver_cpl/shr/seq_infodata_mod.F90
The first 2 are pretty easy. In seq_infodata_mod.F90, git is confused by
the gust_fac addition and the ESP additions. I accepted the "HEAD" version
and then had to put declarations of gust_fac back in a few routines.
So if we can decide on the ESMF_TimeMod changes, this is ready to go.
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#1065 (comment)>, or mute
the thread
<https://github.com/notifications/unsubscribe-auth/AHlxE7L-EkGBh6auVD16MOKp263pxv0xks5rZ_CSgaJpZM4LwW3l>
.
|
|
Here's the commit: fddc0d0 Doug found this while doing some testing with MPAS I think. |
|
Have we confirmed via testing Doug's assertion that "These changes
shouldn't affect intervals that previously worked, and
should only allow year and month intervals to be compared."
…On Mon, Feb 6, 2017 at 9:37 PM, Robert Jacob ***@***.***> wrote:
Here's the commit: fddc0d0
<fddc0d0>
Doug found this while doing some testing with MPAS I think.
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#1065 (comment)>, or mute
the thread
<https://github.com/notifications/unsubscribe-auth/AHlxE59KE0qnFYqjk6y-5zhYt8KRCuc8ks5rZ_USgaJpZM4LwW3l>
.
|
|
Tony has just confirmed that he thinks this change is okay. So I think this
PR is now good to go.
On Mon, Feb 6, 2017 at 9:47 PM, Mariana Vertenstein <mvertens@ucar.edu>
wrote:
… Have we confirmed via testing Doug's assertion that "These changes
shouldn't affect intervals that previously worked, and
should only allow year and month intervals to be compared."
On Mon, Feb 6, 2017 at 9:37 PM, Robert Jacob ***@***.***>
wrote:
> Here's the commit: fddc0d0
> <fddc0d0>
>
> Doug found this while doing some testing with MPAS I think.
>
> —
> You are receiving this because you were mentioned.
> Reply to this email directly, view it on GitHub
> <#1065 (comment)>, or mute
> the thread
> <https://github.com/notifications/unsubscribe-auth/AHlxE59KE0qnFYqjk6y-5zhYt8KRCuc8ks5rZ_USgaJpZM4LwW3l>
> .
>
|
rljacob
changed the title
rljacob
added a commit
that referenced
this pull request
Feb 8, 2017
Merge the maint-cime5.1 branch along with ACME's changes to 5.1 XML changes: ACME compset additions, ACME PEs changes, ACME grids ACME build system updates (Depends files, Makefiles) ACME config changes (batch, compilers, machines) Minor modifications to config_component to split out entry ids that are not the same between ACME and CESM Lots of additions to config_component_acme CPL/DRV changes: Addition of gust_fac into driver Addition of cron_script, cron equivalent of jenkins_script Python changes: Remove obsolete ACME env_mach_specific files If model is ACME, auto-add compiler to baseline name (impacts create_test, bless_test_results, and compare_test_results) Fix HOMME test, Updates to ACME performance archiving Raise exception if batch job-id fails to match pattern Bug fix for point calculation in pes.py Changes to build.py because clm is not a shared lib in ACME ACME now supports recommended test timings based on test-suite Test suite: scripts_regression_tests.py Test baseline: none Test namelist changes: Test status: bit for bit Closes #1105 Fixes #834 User interface changes?: See above Code review: Mariana Vertenstein, Jim Edwards, Steve Goldhaber, Rob Jacob Conflicts: cime_config/acme/machines/config_compilers.xml - took ACME version and made modes to add -DCPR cime_config/acme/machines/config_machines.xml - took branch version driver_cpl/shr/seq_infodata_mod.F90 - took ESMCI version and re-added declarations for gust_fac
|
This merge was undone by a force push to the previous version of master. Turns out github was right and there were to many commits/files in the original subtree split from ACME. Replaced by PR #1123 |
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.
Merge the maint-cime5.1 branch along with ACME's changes to 5.1
XML changes:
CPL/DRV changes:
Python changes:
Test suite: scripts_regression_tests.py
Test baseline: none
Test namelist changes:
Test status: bit for bit
Closes #1105
Fixes #834
User interface changes?: See above
Code review: Mariana Vertenstein, Jim Edwards, Steve Goldhaber, Rob Jacob