fix for cris-fsr memory corruption (issue #755) #767
Conversation
|
@ADCollard, @emilyhcliu , @xincjin-NOAA , or @azadeh-gh : If any of you have time, would you please review this simple one line change from @jderber-NOAA ? John fixed a bug in If none of you have time to review the change, any suggestions as to who we can ask? |
There was a problem hiding this comment.
Looked through the issues documented in Issue #755. I learned a lot by reading the issue about how to trace the stepsize problem in the minimization.
|
It seems that the number of active channels (nch_active) stored in the correlated obs file is not consistent with the number of active channels (counts) from the sat info file. GSI/src/gsi/correlated_obsmod.F90 Lines 391 to 425 in 529bb79 |
|
WCOSS2 ctests |
Does this mean that the nch_active should be equal to coun? |
|
Emily, That is correct😁. And the nch_active is less than the number of active channels from the satinfo file. This is only true for the npp fsr data. This is also the only case where the code will fail and the change is necessary.JohnSent from my Verizon, Samsung Galaxy smartphone
-------- Original message --------From: emilyhcliu ***@***.***> Date: 7/8/24 11:03 AM (GMT-05:00) To: NOAA-EMC/GSI ***@***.***> Cc: jderber-NOAA ***@***.***>, Mention ***@***.***> Subject: Re: [NOAA-EMC/GSI] fix for cris-fsr memory corruption (issue #755) (PR #767)
It seems that the number of active channels (nch_active) stored in the correlated obs file is not consistent with the number of active channels (counts) from the sat info file.
https://github.com/NOAA-EMC/GSI/blob/529bb796bea0e490f186729cd168a91c034bb12d/src/gsi/correlated_obsmod.F90#L391-L425
—Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you were mentioned.Message ID: ***@***.***>
|
I think so! |
|
I think with this change coun and nch_active do not have to be equal. If coun < nch_active the code will delete those parts of the correlation matrix that use channels not in coun. If nch_active < coun then the code will add the additional channels found in coun into the covariance matrix using the variances (no correlated error) in the satinfo file (not positive where the variances come from). The original code was set up to do this, just the one array was too small (when coun > nch_active) and created the overwriting of the memory. |
|
@jderber-NOAA and @emilyhcliu : Are we good with the changes in this PR? If "yes", I will work with the GSI Review Committee to merge this PR into |
|
I am.Sent from my Verizon, Samsung Galaxy smartphone
-------- Original message --------From: RussTreadon-NOAA ***@***.***> Date: 7/9/24 8:42 AM (GMT-05:00) To: NOAA-EMC/GSI ***@***.***> Cc: jderber-NOAA ***@***.***>, Mention ***@***.***> Subject: Re: [NOAA-EMC/GSI] fix for cris-fsr memory corruption (issue #755) (PR #767)
@jderber-NOAA and @emilyhcliu : Are we good with the changes in this PR? If "yes", I will work with the GSI Review Committee to merge this PR into develop.
—Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you were mentioned.Message ID: ***@***.***>
|
@RussTreadon-NOAA Yes!! |
|
Hera & Hercules ctests Hera The A check of the Correlate observation error for radiances is not active in the The This is not a fatal fail. The This is not a fatal fail. A rerun of the failed tests resulted in Passed for each test Hercules The This is not a fatal fail. The This is not a fatal fail. The This is not a fatal fail. The This is not a fatal fail. A rerun of the ctests on Hercules using
|
* origin/develop: Move to contrib spack-stack on Jet (NOAA-EMC#787) a quick workaround for increasing the mpi task numbers on orion for ctest :: rrfs_3denvar_rdasens (NOAA-EMC#788) Recover the capability of handling model fields from operation gfs.v16.3 (NOAA-EMC#785) fix a bug in deter_sfc_gmi (NOAA-EMC#781) add safeguard to thompson_reff (NOAA-EMC#779) Fix incorrect usage of real(i_kind) in mg_input.f90 (NOAA-EMC#760) Transition to Thompson Microphysics for Microwave All-sky Assimilation (NOAA-EMC#743) Format changes for EUMETSAT metop-sg and CADS debug fix (NOAA-EMC#773) Update global_4denvar and global_enkf ctests to reflect GFS v17 (NOAA-EMC#774) fix for cris-fsr memory corruption (NOAA-EMC#767) Gnssrwnd1.0 (NOAA-EMC#747)
one line change from @jderber-NOAA to fix memory corruption issue in correlated_obsmod.F90 for cris-fsr data found in reanalysis scout runs. Detailed description found in issue #755.
Resolves #755
Checklist