Repository navigation
Conversation
added 8 commits
October 16, 2026 14:47
…unified output directories
- Change out_app_flag from bool to int with three states:
0: no append, no subdirectory (overwrite)
1: append to same file (default, backward compatible)
2: write to unified subdirectories
- Add new global directory variables in System_para:
global_psi_dir, global_chg_dir, global_pot_dir,
global_elf_dir, global_dos_dir, global_dip_dir, global_mat_dir
- Initialize directories in set_global_dir():
out_app_flag==2 -> OUT.suffix/{PSI,CHG,POT,ELF,DOS,DIP,MAT}/
otherwise -> all point to global_out_dir (backward compatible)
- Map legacy WFC->PSI, matrix->MAT
- Update all out_app_flag usages from bool to int semantics
- Broadcast new directory variables via Parallel_Common::bcast_string
Issue: deepmodeling#8151
- ctrl_output_fp.cpp: route chg/tau to global_chg_dir, pot/potes to global_pot_dir, ELF to global_elf_dir, dipole to global_dip_dir - ctrl_output_pw.cpp: route wfc to global_psi_dir, dos to global_dos_dir, ldos to global_dos_dir - ctrl_scf_lcao.cpp: route dos to global_dos_dir - ctrl_runner_lcao.cpp: route ldos to global_dos_dir - read_input.cpp: create subdirectories (PSI/CHG/POT/ELF/DOS/DIP/MAT) when out_app_flag==2 - Update docs: out_app_flag semantics from bool to int (0/1/2) Issue: deepmodeling#8151
When out_app_flag defaults to 1 (not 2), all new unified directories (PSI, CHG, POT, ELF, DOS, DIP, MAT) map to global_out_dir for backward compatibility. Update io_sys_var_test.cpp expectations: - global_matrix_dir: "OUT.test/matrix/" -> "OUT.test/" - Add assertions for all 7 new directory variables
- read_input.cpp: change !out_app_flag to out_app_flag == 0 for matrix/WFC legacy directory decisions - Update three unit tests to use int out_app_flag (0/1) instead of bool (false/true), matching the new function signatures: test_cal_dos.cpp, test_write_dos_pw.cpp, write_wfc_nao_test.cpp - Update two comments to describe 0/1 semantics
Replace repeated PARAM.globalv.global_*_dir accesses with the local const variables already declared at function top, eliminating the net +4 PARAM global dependency introduced by the path routing change.
With out_app_flag changed from bool to int, the legacy WFC/ subdirectory is no longer created when out_app_flag=0 (now maps to global_out_dir). Update the multi-k wfc reference path accordingly.
- read_input.cpp: only create PSI/CHG/POT/ELF/DOS/DIP/MAT subdirectories when the corresponding output flag is enabled (out_app_flag=2) - ctrl_scf_lcao.cpp: route write_dmk and angular momentum calculator output to global_mat_dir instead of global_out_dir - hsr_legacy.cpp: route tr_nao.csr to global_matrix_dir - pos_op_writer.cpp: route rr_nao.txt to global_matrix_dir
mohanchen
added
Input&Output
Collaborator
Author
|
@ABACUS-PR-BOT please help review the PR |
ABACUS-PR-BOT
left a comment
ABACUS-PR-BOT
left a comment
Collaborator
There was a problem hiding this comment.
Static review of 00ece15 with independent verification. Five implementation issues and three non-blocking suggestions are noted inline. Whitespace checks passed; the pinned governance check reported warnings only. No build or runtime tests were run, and CI diagnostics were unavailable.
- Remove md_no_append ternary in hsr_legacy.cpp, pos_op_writer.cpp, hterm_writer.cpp, dhs_sparse_writer.cpp: use global_matrix_dir unconditionally (it already equals global_out_dir when out_app_flag != 2, so no behavior change for legacy modes) - write_dmr: add dmr_dir parameter, route DM(R) to MAT instead of hardcoded global_out_dir; drop parameter.h include - read_input.cpp: add out_dmr/out_mat_l/out_mat_dh/out_mat_ds to MAT subdirectory creation flags Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- global_stru_dir now follows out_app_flag: STRU/ for flag 2,
global_out_dir otherwise (consistent with MAT/PSI/CHG pattern)
- relax_driver: route STRU_NOW/STRU{step}/STRU_FINAL through
global_stru_dir instead of global_out_dir
- run_md: drop _MD suffix, use STRU{step} naming consistent with
relax
- global_file: remove unconditional STRU dir creation from
make_dir_out; remove global_stru_dir parameter
- read_input: create STRU dir on demand (out_stru>0 or MD)
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- run_md: use out_freq_ion (with md_restartfreq fallback) as the effective ionic-step output frequency for both STRU and restart - run_md: gate STRU output on out_stru > 0, consistent with relax - read_input: simplify STRU dir creation to out_stru > 0 - read_inp_out: update out_stru description to cover MD - read_input_item_md: mark md_restartfreq as deprecated Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- commit 3a38485 made global_stru_dir follow out_app_flag (default 1 maps it to the output dir), but the MD restart lookup still probed STRU_MD_{istep} there, so restart cases whose fixtures live in OUT.suffix/STRU/ (e.g. tests/04_FF/14_LJ_rescale_v) failed to start - search the current global_stru_dir first, then fall back to the legacy OUT.suffix/STRU/ layout; accept both the new STRU{istep+1} naming (1-based, unified with relax) and the legacy STRU_MD_{istep} - quit with a clear diagnostic when no restart structure file exists - add unit test covering both naming schemes in a user read_file_dir
- add PSI/ candidate in read_wfc_nao so RT-TDDFT restart finds mode-2 snapshots - route H(k) term outputs to MAT directory; add out_hsr_npz to MAT dir creation predicate; narrow DOS dir creation predicate to out_dos/out_ldos - pass boolean append to save_mat in write_hk_common (was int out_app_flag) - update out_app_flag/out_wfc_lcao INPUT metadata and position_matrix docs - add mode-2 directory mapping and PSI layout roundtrip unit tests
Conflict resolutions: - read_input_test.cpp: keep both added tests (OutAppFlagParsing ours, ExplicitNoccSurvivesUnresolvedElectronNumber upstream) - relax_driver.cpp: identical condition; keep ours comment (md STRU output now handled by run_md.cpp, md_restartfreq removed) Post-merge verification: make -j 30 in build_max_para_test, 100% Built target abacus_max_para, 0 errors.
In append mode (out_app_flag=1), chg/tau/pot/potes/ELF cube outputs
now reuse a fixed filename (e.g. chgs1.cube, elftot.cube) so every
step overwrites the previous snapshot instead of accumulating
g{N}-suffixed files. Modes 0 and 2 keep the per-step g{N} suffix.
Add an explicit out_app_flag parameter to ModuleIO::write_dmk. In append mode the geometry-step index is dropped from the filename (istep_file = -1) so every ionic step overwrites the previous snapshot, matching the fixed name that read_dmk already expects; the file body still records the real ionic step.
Drop the g{N} suffix from tr_nao.csr and rr_nao.txt in append mode.
The filenames were changing every ionic step, which defeated the
existing MD append branches in save_sparse and the rr assemble path;
with fixed names those branches now accumulate all steps as designed.
Replace the hardcoded out_app_flag=0 in write_wfc_pw (a 2025-05-17
TODO) with an explicit parameter wired from inp.out_app_flag. In the
default append mode the PW wavefunction files now use fixed names
(wfk1_pw.dat), which finally matches what the restart reader
read_wf2rho_pw looks for; modes 0 and 2 keep the g{N} suffix.
write_chg_init/write_pot_init now keep the fixed _ini.cube names in
append mode so each snapshot overwrites the previous one, instead of
accumulating chgs1g{N}_ini.cube / potg{N}_ini.cube files. These
initial-density cubes are write-only snapshots; no reader consumes
them, so no restart path is affected.
Clarify that mode 1 reuses fixed filenames across steps (text records append, cube snapshots overwrite), fix the remaining Boolean wording in out_hsk, and note that out_chg=2/out_pot=3 initial-density cubes omit the geometry-step index in append mode.
The fixed filename alone still truncated the file every ionic step. Add a calculation parameter and open the stream with ios::app under the same condition as open_sparse_file (calculation=="md" && out_app_flag==1 && istep>=0), so each MD step appends a new record while other calculations start fresh.
Rule 22: user-facing output files print ionic steps starting from 1,
matching the g{N} filename suffix convention; zero-based counters such
as "STEP: 0" are not allowed.
Naming (mode 1 fixed name, modes 0/2 keep g{N}):
- DOS (LCAO+PW) and PDOS honor out_app_flag for the g{N} suffix
- lx/ly/lz matrices: Angmom_op::calculate takes out_app_flag
Append (relax now appends like MD; was calculation=="md" only):
- open_sparse_file, rr/dH/dS writers, DM(k): condition relaxed to
(out_app_flag == 1) && (istep >= 0); SCF passes istep=-1, unaffected
- PDOS and lx/ly/lz append per ionic step in mode 1
- DM(k) drops the calculation parameter again (no longer used)
Step indexing (governance rule 22, one-based):
- STEP: headers of sparse text files and the binary CSR step field
now start from 1 (hs_sparse_io, pos_op_writer, dhs_sparse_writer,
hsr_writer)
Build: make -j 30 verified by developer.
Instead of numbered STRU{N} files, out_app_flag=1 now writes one
fixed-name STRU file that accumulates one structure block per written
step, for both relax and MD:
- unitcell::print_stru_file and relax_stru_io::write_stru take a new
append flag; relax writes "# Ionic Step N" headers (one-based,
governance rule 22) via the existing build_stru_header
- mdcell::print_stru_file takes frame_comment + append; MD frames are
prefixed with "# MD step N". The MPI path broadcasts a base offset
(lseek SEEK_END) and keeps absolute pwrite offsets; O_APPEND is
deliberately avoided since Linux pwrite ignores the offset with it
- CIF output keeps single-frame semantics: fixed-name STRU.cif is
overwritten each written step
- modes 0/2 keep the numbered STRU{N} files unchanged
- CLI help of out_stru/out_app_flag documents the mode-1 behavior
Build: make -j 30 verified by developer.
The DM(k) writer emitted the row newline only before each group of elements, so the file ended without a trailing newline. In append mode (out_app_flag == 1) the next " --- Ionic Step N ---" record was glued to the last value of the previous step. Add a trailing newline before closing the file and cover it with a regression assertion.
write_wfc_pw now mirrors write_wfc_nao's append_flag: the first
written step truncates the fixed-name wf*_pw.{txt,dat} files and later
ionic steps append one frame per step (all payload writes already use
append mode; only the per-step pre-clear is skipped). Modes 0/2 keep
per-step g{N} files unchanged.
Also terminate the DM(k) matrix body with a newline so appended
records start on a fresh line (developer fix), and document in
read_wf2rho_pw/out_wfc_pw help that multi-frame binary files are read
from their first frame only.
Build: make -j 30 verified by developer.
band.txt/bands1.txt/bands2.txt contained numeric rows only, so the k-point
blocks appended across ionic steps of a relax/md run were indistinguishable,
and a single-step run carried no version or timestamp. Print the STRU-style
header ("# ABACUS version", "# Written at", "# IONIC STEP n") in front of
every ionic step instead: write_bands now takes the step index, nscf_bands
takes the prepared header, and the step label is one-based with the collinear
spin-channel tag when nspin=2.
The integration comparison strips "#" lines before comparing band.txt, and
the unit test skips comment lines while parsing and asserts the header of
each appended block.
make_spin_tag pads "up" so that both collinear channels share the same width wherever they are printed next to each other in the running log. The band header instead ends up in a per-spin file (bands1.txt/bands2.txt), so the padding only leaves a trailing blank before the closing parenthesis. The header now uses " (spin up)" through a local band_spin_tag helper, while the log line keeps make_spin_tag unchanged.
…nd semantics Add focused unit tests exercising out_app_flag modes 0/1/2 across the filename_output primitive and all major output writers: - test_filename.cpp (new): filename_output three-mode matrix, iter/gamma/nspin - write_dmk_test.cpp: Mode0PerStepFiles / Mode1FixedNameAppend / Mode2SameNamingAsMode0 - test_write_pdos_text.cpp: Mode1FixedNameAppend / Mode2SameNamingAsMode0 - test_cal_dos.cpp: Mode1Appends (two ionic step headers) / Mode2Overwrites - test_print_cell.cpp: STRUAppend (two ATOMIC_SPECIES blocks accumulated) - test_mdcell_reader.cpp: STRUAppendWithFrameComment (# MD step N markers) - write_wfc_nao_test.cpp: OutAppFlagMode2SameNamingAsMode0 (completes 0/1/2) Also sync out_dos CLI description to document the three-mode semantics. Governance: net_delta=-9 (global dependency budget decreased).
- CMakeLists: MODULE_BASE_filename needs both `base` (filename.cpp) and `device` (kernel ops referenced by base) in its LIBS - test_mdcell_reader: include gmock/gmock.h for HasSubstr matcher - write_dmk_test: new three-mode tests were missing the K_Vectors construction and the precision parameter (third positional arg); sync with the WriteDMK precedent in the same file Build verified by developer (make -j 30).
Follow-up to a658239 which moved the sparse/CSR STEP field to the one-based convention (governance rule 22) without updating the tests: - hs_sparse_io / dhs_sparse_writer / hsr_writer: expect istep+1 in the text STEP header and in the native binary step field - write_dmk modes 0/1/2: resize kvec_d before calling write_dmk, which dereferences kv.kvec_d[ik] and segfaulted on the empty vector - write_dmk mode 1: match the actual " --- Ionic Step N ---" header Verified: the four test targets pass, including mpirun -np 2.
The sparse CSR writers emit std::max(istep, 0) + 1 in the STEP field since a658239 (governance rule 22), but these integration references still expected the zero-based value. Only the header line changes; all matrix elements are byte-identical. Affected cases: 02_NAO_Gamma/scf_out_hxc, 03_NAO_multik/scf_out_dh_t, 03_NAO_multik/get_s, 03_NAO_multik/nscf_out_hsr_tr_rr, 03_NAO_multik/scf_out_hsr_spin4. Verified with Autotest.sh on a clean OUT.autotest: 36 test cases pass.
…ader compare_hsr_binary.py still rejected any non-zero step in the native binary record, so scf_out_hsr_binary_spin2 aborted after the sparse writers moved to the one-based convention (a658239, rule 22). Accept step 1 and describe the merged field as one-based in the out_hsr help text and in the binary CSR format documentation. Verified: scf_out_hsr_binary_spin2 passes; full 02_NAO_Gamma (35 cases) and 03_NAO_multik (69 cases) rerun clean, with only the two DFT-D4 cases failing because this build lacks -DENABLE_DFTD4=ON.
write_dh_perI had two relic caller-side branches that duplicated the directory selection already performed by ReadInput::set_global_dir: 1. r_dir used a ternary keyed on calculation=="md" && out_app_flag!=1 2. the dense k-space dH(k) filename was hardcoded against global_out_dir instead of the r_dir variable Both are replaced with a single mat_dir = PARAM.globalv.global_mat_dir, so relax + out_app_flag=2 now lands CSR dH/dR and dense dH/dk in OUT.suffix/MAT/ together with H(R), S(R), DM(k) and the H(k) term matrices already routed in the fix commit. Modes 0 and 1 keep the exact same output path because global_mat_dir aliases global_out_dir there.
The mode-2 subdir_flags OR chain for MAT/ covered the aggregate switches (out_mat_hs2, out_mat_dh) and a handful of legacy flags (out_dmk, out_hsk, out_mat_tk, ...) but none of the per-component switches that actually write into MAT/: H(R) components: out_mat_h_t/vnl/vl/vh/vxc/exx dH components: out_mat_dh_t/vl/vnl/vh/vxc With only one of those per-component flags enabled in mode 2, MAT/ would not be created and the H(R)/dH component writers would fail to open their output files. Append all 11 component switches to the OR chain so MAT/ is created whenever any matrix output is requested.
…or out_app_flag
Two related fixes for the out_app_flag rework:
1. ctrl_output_pw.cpp — the last two PW-specific outputs that still wrote
into OUT.root are now routed into their purpose-built mode-2
subdirectories:
out_pchg -> global_chg_dir (CHG/)
out_wfc_norm / _re_im -> global_psi_dir (PSI/)
Every PW writer (wfc, DOS, chg, pot, ELF, dipole, pchg, wfc_norm)
now honors mode-2 routing — no more OUT.root leakage from PW.
2. read_inp_out.cpp — out_app_flag's availability was
"basis_type==lcao and gamma_only==0" even though the description
said the routing applies to both LCAO and PW. Empty the availability
expression so the flag is visible to every run type. The lcao
restriction was never enforced in the code paths and PW users were
only being hidden from the parameter — the PW writers above now back
that description up.
Two regression guards for the review feedback asking for focused mode-2
coverage:
test_hs_dense_io.cpp :: AppendFalseOverwritesSameIstep
save_mat must truncate on append=false even when the same istep is
rewritten. The older BinaryAppend test cross-stepped istep 0 -> 1 -> 2
and never caught the same-step overwrite case. This pairs append=true
then append=false on the same file for istep=0 and asserts exactly one
record remains.
test_hsk_writer.cpp :: Mode2OverwritesSameIstep
write_hsk was passing the integer out_app_flag directly into save_mat's
boolean append parameter before the fix, so out_app_flag=2 would append
instead of truncating. The existing tests only exercised append=false
implicit via k_out_app=0 (mode 0 behaviour). This test drives
out_app_flag=2 for istep=0, swaps the FakeHamilt payload, writes again,
and verifies the second payload replaced the first.
Governance: one new test file matches one source file, test_hsk_writer.cpp
was already in place so a new case joins the existing fixture.
out_app_flag == 1 writes one fixed-name "STRU" file, but a previous
release kept per-step STRU_MD_{step} files inside a STRU/ subdirectory
of the output directory. Restarting an MD from such output (14_LJ_-
rescale_v ships that layout) made open() hit the directory and abort
with "Unable to create MDCell restart STRU file: ... Is a directory".
Resolve the STRU output prefix once before the MD loop: when mode 1 is
requested and <out_dir>/STRU already exists as a directory, keep the
legacy numbered files inside that directory and skip accumulation.
Modes 0 and 2, and mode 1 without a legacy directory, are unchanged.
Also refresh the Mulliken output sample in the documentation: the file
now starts with " --- Ionic Step 1 ---" and uses the zeta/spin table
with sum m / sum mz / sum lmz labels.
Verified: make -j 30; 14_LJ_rescale_v 4/4; full 04_FF 78 checks;
02_NAO_Gamma md_*/relax_* (9 cases) 53 checks.
Two bugs in source/source_io/module_dipole/write_dipole.cpp caused out_dipole=1 to abort under MPI: 1. Validation check rejected rhopw->nplane == 0, but nplane is a per-rank local count (number of z-planes owned by the current process). Some ranks legitimately hold zero z-planes under z-axis MPI decomposition; hitting WARNING_QUIT there made those ranks exit with code 1 while the others waited forever. 2. The dipole data line was written to 'ofs' on every rank, but the file is only opened on rank 0. Non-root ranks wrote to an unopened stream, which is undefined behaviour and corrupts output when it doesn't crash outright. Fixes: - Remove rhopw->nplane == 0 from the guard; keep only global dimensions nx/ny/nz/nxyz, which truly must be positive. - Wrap the ofs << ... write inside an if (GlobalV::MY_RANK == 0) block, consistent with the existing file-open and file-close guards.
A rank can own zero real-space grid points (nrxx == 0) when the FFT grid is decomposed across more processes than it has z-slabs. The asserts in write_elf rejected that legitimate empty partition. Dropping the asserts alone is not enough if the FFT body is skipped: real2recip/recip2real go through PW_Basis::gatherp_scatters, which performs an MPI_Alltoallv that requires every rank in POOL_WORLD to participate, so a rank that returns early deadlocks the grid-owning ranks inside Alltoallv. Run the same FFT path unconditionally on every rank (the per-grid-point loops simply iterate zero times when nrxx == 0) and keep every rank in the write_vdata_palgrid output section for the collective MPI_Gatherv in Parallel_Grid::reduce. Add regression tests for the zero-length container and zero-iteration invariants. Verified with examples/17_relax/04_relax_with_output_lcao (out_elf 1): OMP_NUM_THREADS=4 mpirun -np 4 abacus_max_para finishes in 4 s and writes elfs1.cube, elfs2.cube and elftot.cube; MODULE_IO_write_elf_logic_test passes 14/14. Co-Authored-By: Claude Sonnet 4.6 noreply@anthropic.com>
The 'Orbital' file only describes the NAO basis orbitals of the cell (index, species, l, m, zeta, symmetry name) and is consumed by the projected-band and Mulliken outputs. Rename it to 'nao.txt' so the name states the basis type and does not collide conceptually with plane-wave outputs. Update both writers, the affected unit tests, the pband integration comparison script and its golden reference, and the Mulliken docs.
wfk*_nao.txt now carries the same three-line header as bands1.txt (version, timestamp, ionic step with spin tag) before the data block, with one header per frame in append mode. The reader skips leading '#' comment lines on the ASCII path only; the binary record layout is untouched. Downstream Python consumers of the text format filter '#' lines so they tolerate the new header. Co-Authored-By: Claude Sonnet 4.6 noreply@anthropic.com
Align test_lr_util.cpp with governance rule 22: hs_sparse_io writes STEP: max(istep,0)+1, and LR_Util::save_HR passes istep = 0, so the header is "STEP: 1". Fixes MODULE_LR_dm_regressions and MODULE_LR_sparse_rank_keys. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
With out_app_flag = 1 the multi-step writers reuse a fixed filename, so
write_cube truncated the previous snapshot and only the last written step
survived in chg/tau/pot/elf cube files. Text outputs already append records
under the same flag, and out_wfc_lcao already truncates on the first step and
appends afterwards, so cube output was the remaining inconsistency.
Pass an explicit append flag down the writer chain instead of reading
PARAM inside the writers:
- write_cube / write_vdata_palgrid take a new append parameter and open the
file with std::ios::app when it is set, otherwise truncate;
- write_elf and write_elecstat_pot forward it, and write_chg_init /
write_pot_init compute it from inp.out_app_flag;
- ctrl_output_fp uses append = (out_app_flag == 1 && istep > 0), mirroring
the out_wfc_lcao rule: the first written ionic step truncates, later
written steps append one complete frame whose first comment line is
Ionic_Step {#}, one-based. Single-point SCF/NSCF stays at istep == 0, so it
keeps overwriting rather than stacking on a stale file from a rerun.
- every other call site (LDOS, LCAO/PW wavefunction cubes, partial charge,
LR-TDDFT density, exciton density, DM2rho) passes false explicitly to keep
its existing truncate behaviour.
Update the out_app_flag and out_elf help text, the output specification, and
add CubeIOTest.WriteCubeAppendFrames covering truncate-then-append-then-
truncate frame counting and the trailing newline contract.
Co-Authored-By: Claude Sonnet 4.6 noreply@anthropic.com>
Rename vpp_local_out.dat, vpp_nonlocal_out.dat, vhartree_out.dat and kinetic_out.dat to eband_vl.txt, eband_vnl.txt, eband_hart.txt and eband_kin.txt, and prefix each file with version, timestamp, ionic step and total energy, expectation-value formula, and unit comments. Write the files on rank 0 only to avoid concurrent writes in MPI runs.
This branch has not been deployed
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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.
Summary
Resolves #8151. This PR unifies the output directory structure by extending
out_app_flagfrombooltointwith three modes, and routes all output files to consistent uppercase subdirectories when enabled.Problem
In RT-TDDFT and MD calculations, output files (
.cubecharge densities, WFC files, matrices, etc.) accumulate in theOUT.<suffix>/root directory, making it difficult to navigate and manage results.Solution
out_app_flagsemantics (backward compatible)0OUT.xxx/1(default)OUT.xxx/2(new)Unified subdirectories (only created when
out_app_flag=2and the output is requested)PSI/WFC/)CHG/chg,tau)POT/pot,potes)ELF/DOS/DIP/MAT/matrix/, includesdmk,lx/ly/lz,rr,tr)Key changes
out_app_flagchanged frombooltointwith validation{0,1,2}System_paramembers (global_psi_dir,global_chg_dir, etc.) with MPI broadcastout_app_flag=1preserves existing behavior (root directory, append mode)Verification
make -j 30passes onbuild_max_para_testPARAM/GlobalVcross-layer control)out_app_flagbehaviorFiles changed
source/source_io/module_parameter/— input parameter, read logic, global directory setupsource/source_io/module_ctrl/— output path routing for chg/pot/elf/dos/dip/wfcsource/source_io/module_dm/— dmk path routingsource/source_io/module_hs/— matrix output path routing (hsr, pos_op, dhs, vxc)source/source_base/module_out/—filename_outputsignature updatetests/— unit and integration test updatesdocs/— documentation updates