(Closes #2845) fix for inline symbol bug

[arporter]

No description provided.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 99.89%. Comparing base (10a3a6e) to head (edf4556).

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #2848      +/-   ##
==========================================
- Coverage   99.89%   99.89%   -0.01%     
==========================================
  Files         359      359              
  Lines       50995    50993       -2     
==========================================
- Hits        50940    50938       -2     
  Misses         55       55              

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[arporter]

A small change to the inlining transformation so that symbols are added to the table of the Routine of the call site rather than to the table of the local scope. This allows us to spot problems in the validate rather than crashing at the end of the appy() method. The integration tests will need to be run but I don't want to do that during the day while Glados is busy.

[LonelyCat124]

Need to check if the integration tests are ok, but the code changes all look fine, and tests/coverage/etc. are all fine.

Edit: Someone started the integration tests about 30 minutes ago - @arporter is this ok or do we need to stop them and try them later?

[arporter]

It's OK, it was me and I see NEMO v.4 failed :-(

[arporter]

Failure was:

  File "/home/gh_runner/actions-runner/_work/PSyclone-mirror/PSyclone-mirror/.runner_venv/lib/python3.13/site-packages/psyclone/psyir/symbols/symbol_table.py", line 1793, in rename_symbol
    raise ValueError(
        f"The symbol argument of rename_symbol() must belong to this "
        f"symbol_table instance, but '{symbol}' does not.")
ValueError: The symbol argument of rename_symbol() must belong to this symbol_table instance, but 'psyclone_cmp_int: DataSymbol<Scalar<BOOLEAN, UNDEFINED>, Automatic>' does not.
make: *** [Makefile:70: psycloned-openacc_kernels/dynldf.f90] Error 1

and must be because we are adding symbols to tables in nested scopes rather than the parent Routine scope.

[LonelyCat124]

Ok - I'll sent it back to you to resolve then

Integration tests failed

[arporter]

That was easier than I expected. Will wait until tonight to trigger integration tests again.

[arporter]

NEMO4 OpenACC kernels integration test failed again :-( Will investigate.

[arporter]

The failure is:

nemo4_src]$ psyclone -I ./ -s ~/PSyclone/examples/nemo/scripts/acc_kernels_trans.py nemogcm.f90

It is the inlining of nemo_init that causes the failure with a symbol that is unresolved in both the routine and, at the point of inlining, the call site:

psyclone.psyir.symbols.symbol.SymbolError: PSyclone SymbolTable error: A symbol named 'sn_ctl' is present but unresolved in one or both tables.

[arporter]

I realised I could 'simply' change validate to check nested scopes so I've done that. Transforming nemogcm::nemo_init is still failing though - I think this is because we inline the same routine (nemo_set_cfctl) more than once. The first one succeeds but then we get symbol clashes with the second one. The routine itself is simple:

   SUBROUTINE nemo_set_cfctl(sn_cfctl, setto, for_all )
      !!----------------------------------------------------------------------
      LOGICAL :: setto, for_all
      TYPE(sn_ctl) :: sn_cfctl
      !!----------------------------------------------------------------------
      IF( for_all ) THEN
         sn_cfctl%l_runstat = setto
         sn_cfctl%l_trcstat = setto
      ENDIF
      sn_cfctl%l_oceout  = setto
      sn_cfctl%l_layout  = setto
      sn_cfctl%l_mppout  = setto
      sn_cfctl%l_mpptop  = setto
   END SUBROUTINE nemo_set_cfctl

and I think it's the sn_ctl datatype that's causing the failure. There is of course no explicit import of this symbol and there are very many wildcard imports at the module level.

[arporter]

In InlineTrans.apply we take a copy of the Routine we are inlining:

but the result of this is an orphaned Routine - it is no longer connected to the parent Container which has all the wildcard imports.

[arporter]

When we do KernelModuleInlineTrans, we have _prepare_code_to_inline which brings the definition of all referenced Symbols into the subroutine. However, in this case it happens that nemo_set_cfctl is already in the module containing the call site so that hasn't happened.