Fix additional memory leaks in GEOS-Chem

[yantosca]

Name and Institution (Required)

Name: Bob Yantosca
Institution: Harvard + GCST

Confirm you have reviewed the following documentation

• Contributing guidelines

Describe the update

This is a placeholder PR and should not yet be merged. I have made the base branch main but will change it later.

This is a companion PR to #2102. Configuring GEOS-Chem with -DSANITIZE=y has revealed the following memory leaks in GEOS-Chem and HEMCO:

=================================================================
==1495473==ERROR: LeakSanitizer: detected memory leaks

Direct leak of 107464 byte(s) in 133 object(s) allocated from:
    #0 0x14815712993f in __interceptor_malloc ../../.././libsanitizer/asan/asan_malloc_linux.cpp:145
    #1 0x1bd566d in __hco_filedata_mod_MOD_filedata_init /n/holyscratch01/jacob_lab/ryantosca/tests/cloudj/test_memory/CodeDir/src/HEMCO/src/Core/hco_filedata_mod.F90:174
    #2 0x202020202020201f  (<unknown module>)

Direct leak of 29248 byte(s) in 8 object(s) allocated from:
    #0 0x14815712993f in __interceptor_malloc ../../.././libsanitizer/asan/asan_malloc_linux.cpp:145
    #1 0x10c438e in __histcontainer_mod_MOD_histcontainer_create /n/holyscratch01/jacob_lab/ryantosca/tests/cloudj/test_memory/CodeDir/src/GEOS-Chem/History/histcontainer_mod.F90:319
    #2 0x202020202020201f  (<unknown module>)

Direct leak of 3720 byte(s) in 1 object(s) allocated from:
    #0 0x14815712993f in __interceptor_malloc ../../.././libsanitizer/asan/asan_malloc_linux.cpp:145
    #1 0x18f6ae1 in __state_chm_mod_MOD_init_state_chm /n/holyscratch01/jacob_lab/ryantosca/tests/cloudj/test_memory/CodeDir/src/GEOS-Chem/Headers/state_chm_mod.F90:803

Direct leak of 3160 byte(s) in 1 object(s) allocated from:
    #0 0x14815712993f in __interceptor_malloc ../../.././libsanitizer/asan/asan_malloc_linux.cpp:145
    #1 0x18fe162 in __state_chm_mod_MOD_init_state_chm /n/holyscratch01/jacob_lab/ryantosca/tests/cloudj/test_memory/CodeDir/src/GEOS-Chem/Headers/state_chm_mod.F90:1027

SUMMARY: AddressSanitizer: 143592 byte(s) leaked in 143 allocation(s).

Leak 1 is in HEMCO and will be addressed by a separate PR at https://github.com/geoschem/hemco. We are still investigating.

Leak 2 is in the GEOS-Chem Classic History code. We are still investigating.

Leaks 3 and 4 were resolved by adding DEALLOCATE statements (see commit 78fc204).

Tagging @msulprizio @lizziel

Expected changes

This will be a zero-diff update that will remove memory leaks. It will not change results.

Reference(s)

N/A

Related Github Issue(s)

• See [WIP] Fix remaining memory leaks in HEMCO HEMCO#255
• Closes Memory leaks in 14.3.0 development code #2102

[yantosca]

@lizziel @msulprizio: I have traced the memory leak in the History component to routine MetaHistItem_Insert, in History/metahistitem_mod.F90. Specifically to this code:

geos-chem/History/metahistitem_mod.F90

Lines 348 to 380 in 3cd802e

	!=======================================================================
	! Initialize a METAHISTORY ITEM named "Head", which will become the
	! head of the existing list. "Head" will contain a new HISTORY ITEM.
	!=======================================================================
	! Allocate the "Head" object
	ALLOCATE( Head, STAT=RC )
	IF ( RC /= GC_SUCCESS ) THEN
	ErrMsg = 'Could not allocate "Head"!'
	CALL GC_Error( ErrMsg, RC, ThisLoc )
	RETURN
	ENDIF
	! Allocate the "Head%Item" field, which will hold the HISTORY ITEM
	ALLOCATE( Head%Item, STAT=RC )
	IF ( RC /= GC_SUCCESS ) THEN
	ErrMsg = 'Could not allocate "Head%Item"!'
	CALL GC_Error( ErrMsg, RC, ThisLoc )
	RETURN
	ENDIF
	!=======================================================================
	! Insert "Head" at the start of the existing linked list
	!=======================================================================
	! Save the HISTORY ITEM argument in the "Item" field of "Head"
	Head%Item = Item
	! The "Next" field of "Head" points to the current head of the list
	Head%Next => Node
	! Set "Head" as the new head of the linked list
	Node => Head

The issue is that we allocate a local variable HEAD (to serve as the head of the linked list) but then we never deallocate it. So the memory is leaked. This might be a by-product of the original code on which I based the History linked lists (see: https://github.com/mapmeld/fortran-machine/blob/main/flibs-0.9/flibs/src/datastructures/linkedlist.f90, which has a similar issue).

It might take a bit of doing to be able to rethink solving this memory leak in the GCClassic History diagnostics. I propose to thus fix the memory leaks #3 and #4 and then return to solving #2 later.

[yantosca]

All GEOS-Chem Classic integration tests passed:

==============================================================================
GEOS-Chem Classic: Execution Test Results

GCClassic #c7aa8ee Merge pull request #48 for branch 'dev/no-diff-to-benchmark' into dev/14.3.0
GEOS-Chem #78fc20489 Fix memory leaks in State_Chm%AerMass and State_Chm%Phot
HEMCO     #0c7e10e Merge pull request #252 from branch 'origin/feature/geosit_meteorology_constants_file_year_change' into dev/3.8.0

Using 24 OpenMP threads
Number of execution tests: 26

Submitted as SLURM job: 14850037
==============================================================================

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%  All execution tests passed!  %%%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%