Merge remote-tracking branch 'origin/master' into 2617_generic_functi…

…on_on_master

changelog

@@ -156,6 +156,16 @@
 	55) PR #2445 for #2444. Changes the kernel-extraction tooling so
 	that filenames are construction using dashes instead of colons.
 
+	56) PR #2646 for #2641. Fixes incorrect module names within the
+	PSyData API (e.g. when profiling).
+
+	57) PR #2647 for #1992. Adds MPI support to PSyKE. Each rank will write
+    its own output file(s). The generated driver has been extended so that
+    the name of the file to use can be specified on the command line.
+
+	58) PR #2637 for #2138. Add Fortran frontend support to parser array
+	declarations with expressions in their shape dimensions.
+
 release 2.5.0 14th of February 2024
 
 	1) PR #2199 for #2189. Fix bugs with missing maps in enter data

doc/user_guide/profiling.rst

@@ -462,14 +462,30 @@ The automatic name generation depends on whether you are using a
 PSyKAl DSL or only the transformation capabilities of PSyclone. If
 you are transforming existing code:
 
-* the ``module_name`` string is set to the name of the parent
-  function/subroutine/program. This name is unique as Fortran requires
+* the ``module_name`` string is set to the module which contains the
+  current code. If there is no module (e.g. a stand-alone subroutine),
+  the subroutine name is used instead. This name is unique as Fortran requires
   these names to be unique within a program.
 
-* the ``region_name`` is set to an ``r`` (standing for region) followed by
-  an integer which uniquely identifies the profile within the parent
-  function/subroutine/program (based on the profile node's position in
-  the PSyIR representation relative to any other profile nodes).
+* the ``region_name`` is set to the name of the subroutine, followed by
+  an ``r`` (standing for region) followed by an integer which uniquely identifies
+  the profile within the parent function/subroutine/program (based on the
+  profile node's position in the PSyIR representation relative to any other
+  profile nodes). If there is no module name (which means the ``module_name``
+  is already set to the subroutine name), only the ``r`` followed by an
+  integer number is specified.
+
+  Example:
+    .. code-block::
+        :caption: Profiling names used when transforming existing code.
+        :emphasize-lines: 2,4
+
+        ! If the subroutine tra_adv is contained in module tra_adv_mod:
+        CALL profile_psy_data % PreStart("tra_adv_mod", "tra_adv-r0", 0, 0)
+        ! If the subroutinetra_adv is not contained in a module:
+        CALL profile_psy_data % PreStart("tra_adv", "r0", 0, 0)
+
+
 
 For the :ref:`LFRic <lfric-api>` and
 :ref:`GOcean <gocean-api>` APIs:
@@ -486,96 +502,69 @@ For the :ref:`LFRic <lfric-api>` and
   node's position in the PSyIR representation relative to any other
   profile nodes). For example:
 
-.. code-block::
-    :caption: PSyIR with profiling nodes.
-    :emphasize-lines: 2
-
-    InvokeSchedule[invoke='invoke_0', dm=True]
-      0: Profile[]
-          Schedule[]
-              0: Profile[]
-                  Schedule[]
-                      0: HaloExchange[field='f2', type='region', depth=1,
-                                      check_dirty=True]
-                      1: HaloExchange[field='m1', type='region', depth=1,
-                                      check_dirty=True]
-                      2: HaloExchange[field='m2', type='region', depth=1,
-                                      check_dirty=True]
-              1: Profile[]
-                  Schedule[]
-                      0: Loop[type='', field_space='w1', it_space='cells',
-                              upper_bound='cell_halo(1)']
-                          Literal[value:'1', DataType.INTEGER]
-                          Literal[value:'mesh%get_last_halo_cell(1)',
-                                  DataType.INTEGER]
-                          Literal[value:'1', DataType.INTEGER]
-                          Schedule[]
-                              0: CodedKern testkern_code(a,f1,f2,m1,m2)
-                                 [module_inline=False]
-                      1: Profile[]
-                          Schedule[]
-                              0: Loop[type='', field_space='w1',
-                                      it_space='cells',
-                                      upper_bound='cell_halo(1)']
-                                  Literal[value:'1', DataType.INTEGER]
-                                  Literal[value:'mesh%get_last_halo_cell(1)',
-                                          DataType.INTEGER]
-                                  Literal[value:'1', DataType.INTEGER]
-                                  Schedule[]
-                                      0: CodedKern testkern_code(a,f1,f2,m1,m2)
-                                         [module_inline=False]
-              2: Loop[type='', field_space='w1', it_space='cells',
-                      upper_bound='cell_halo(1)']
-                  Literal[value:'1', DataType.INTEGER]
-                  Literal[value:'mesh%get_last_halo_cell(1)', DataType.INTEGER]
-                  Literal[value:'1', DataType.INTEGER]
-                  Schedule[]
-                      0: CodedKern testkern_qr_code(f1,f2,m1,a,m2,istp)
-                         [module_inline=False]
-
-This is the code created for this example:
-
-.. code-block::
-    :caption: Created Fortran source code with profiling regions.
-    :emphasize-lines: 5,6,7,18,19,24,25
-
-     MODULE container
-      CONTAINS
-      SUBROUTINE invoke_0(a, f1, f2, m1, m2, istp, qr)
-        ...
-        CALL psy_data_3%PreStart("multi_functions_multi_invokes_psy", "invoke_0-r0", &
+    .. code-block::
+        :caption: PSyIR with profiling nodes.
+        :emphasize-lines: 2,4,12
+
+        InvokeSchedule[invoke='invoke_0', dm=True]
+          0: Profile[]
+              Schedule[]
+                  0: Profile[]
+                      Schedule[]
+                          0: HaloExchange[field='f2', type='region', depth=1,
+                                          check_dirty=True]
+                          1: HaloExchange[field='m1', type='region', depth=1,
+                                          check_dirty=True]
+                          2: HaloExchange[field='m2', type='region', depth=1,
+                                          check_dirty=True]
+                  1: Profile[]
+                      Schedule[]
+                          0: Loop[type='', field_space='w1', it_space='cells',
+                                  upper_bound='cell_halo(1)']
+                              Literal[value:'1', DataType.INTEGER]
+                              Literal[value:'mesh%get_last_halo_cell(1)',
+                                      DataType.INTEGER]
+                              Literal[value:'1', DataType.INTEGER]
+                              Schedule[]
+                                  0: CodedKern testkern_code(a,f1,f2,m1,m2)
+                                     [module_inline=False]
+
+    This is the code created for this example:
+
+    .. code-block::
+        :caption: Created Fortran source code with profiling regions.
+        :emphasize-lines: 5,6,7,17,18,19,24,30
+
+         MODULE container
+          CONTAINS
+          SUBROUTINE invoke_0(a, f1, f2, m1, m2, istp, qr)
+            ...
+            CALL psy_data_2%PreStart("multi_functions_multi_invokes_psy", "invoke_0-r0", &
                                      0, 0)
-        CALL psy_data%PreStart("multi_functions_multi_invokes_psy", "invoke_0-r1", 0, 0)
-        IF (f2_proxy%is_dirty(depth=1)) THEN
-          CALL f2_proxy%halo_exchange(depth=1)
-        END IF 
-        IF (m1_proxy%is_dirty(depth=1)) THEN
-          CALL m1_proxy%halo_exchange(depth=1)
-        END IF 
-        IF (m2_proxy%is_dirty(depth=1)) THEN
-          CALL m2_proxy%halo_exchange(depth=1)
-        END IF 
-        CALL psy_data%PreEnd()
-        CALL psy_data_1%PreStart("multi_functions_multi_invokes_psy", "invoke_0-r2", &
+            CALL psy_data%PreStart("multi_functions_multi_invokes_psy", "invoke_0-r1", 0, 0)
+            IF (f2_proxy%is_dirty(depth=1)) THEN
+              CALL f2_proxy%halo_exchange(depth=1)
+            END IF 
+            IF (m1_proxy%is_dirty(depth=1)) THEN
+              CALL m1_proxy%halo_exchange(depth=1)
+            END IF 
+            IF (m2_proxy%is_dirty(depth=1)) THEN
+              CALL m2_proxy%halo_exchange(depth=1)
+            END IF 
+            CALL psy_data%PreEnd()
+            CALL psy_data_1%PreStart("multi_functions_multi_invokes_psy", "invoke_0-r2", &
                                      0, 0)
-        DO cell=1,mesh%get_last_halo_cell(1)
-          CALL testkern_code(...)
-        END DO 
-        ...
-        CALL psy_data_2%PreStart("multi_functions_multi_invokes_psy", &
-                          "invoke_0-testkern_code-r3", 0, 0)
-        DO cell=1,mesh%get_last_halo_cell(1)
-          CALL testkern_code(...)
-        END DO 
-        ...
-        CALL psy_data_2%PostEnd()
-        CALL psy_data_1%PostEnd()
-        ...
-        DO cell=1,mesh%get_last_halo_cell(1)
-          CALL testkern_qr_code(...)
-        END DO 
-        ...
-        CALL psy_data_3%PostEnd()
-        ...
-      END SUBROUTINE invoke_0
-    END MODULE container
+            DO cell=1,mesh%get_last_halo_cell(1)
+              CALL testkern_code(...)
+            END DO 
+
+            CALL psy_data_1%PostEnd()
+            ...
+            DO cell=1,mesh%get_last_halo_cell(1)
+              CALL testkern_qr_code(...)
+            END DO 
+            ...
+            CALL psy_data_2%PostEnd()
+            ...
+          END SUBROUTINE invoke_0
+        END MODULE container

doc/user_guide/psyke.rst

@@ -153,11 +153,16 @@ are used or not.
 Distributed memory
 ##################
 
-As noted in the :ref:`PSyKAl Introduction <introduction_to_psykal>` section,
-PSyKAl can support distributed memory. However, since the generated PSy-layer
-code with DM enabled contains infrastructure calls (e.g. checks for runtime
-status of field halos, halo exchanges etc.), code extraction is not allowed
-when distributed memory is enabled.
+Kernel extraction for distributed memory is supported in as much as each
+process will write its own output file by adding its rank to the output
+file name. So each kernel and each rank will produce one file. It is possible
+to extract several consecutive kernels, but there must be no halo exchange
+calls between the kernels. The extraction transformation will test for this
+and raise an exception if this should happen.
+The compiled driver program accepts the name of the extracted kernel file as
+a command line parameter. If this is not specified, it will use the default
+name (``module-region`` without a rank).
+
 
 .. _psyke-intro-restrictions-shared:
 
@@ -176,6 +181,8 @@ The ``ExtractTrans`` transformation cannot be applied to:
 
 * An inner Loop without its parent outer Loop in the GOcean API.
 
+* Kernels that have a halo exchange call between them.
+
 .. _psyke-use:
 
 Use
@@ -381,6 +388,19 @@ and in
 `lib/extract/netcdf
 <https://github.com/stfc/PSyclone/tree/master/lib/extract/netcdf>`_.
 
+All versions of the extraction libraries can be compiled with MPI
+support by setting the variable ``MPI=yes``:
+
+.. code-block:: shell
+
+  make MPI=yes ...
+
+The only difference is that the output files will now have the process
+rank in the name. The compiled driver program accepts the name of the
+extracted kernel file as a command line parameter. If this is not specified,
+it will use the default name (``module-region`` without a rank).
+
+
 .. _extraction_for_gocean:
 
 Extraction for GOcean

examples/gocean/eg5/extract/.gitignore

@@ -1,5 +1,6 @@
 alg.f90
 psy.f90
+test.x90
 extract_test.standalone
 extract_test.netcdf
 *.nc

examples/gocean/eg5/extract/Makefile

@@ -43,6 +43,17 @@
 # The dl_esm_inf extract wrapper library:
 #     export EXTRACT_DIR = ../../../../lib/extract/netcdf/dl_esm_inf
 
+MPI?=no
+
+ifeq ($(MPI), yes)
+	# -P suppresses linemarkers
+	FPP?=cpp -P -D_MPI
+	# We need to compile with mpif90 if MPI is enabled
+	F90?=mpif90
+else
+	FPP?=cpp -P
+endif
+
 PSYROOT=../../../..
 include $(PSYROOT)/examples/common.mk
 
@@ -51,7 +62,8 @@ TYPE?=standalone
 
 GENERATED_FILES += *.o *.mod $(NAME)  alg.f90 psy.f90   \
 		  $(DRIVER_INIT).$(TYPE)   $(DRIVER_INIT).f90   \
-		  $(DRIVER_UPDATE).$(TYPE) $(DRIVER_UPDATE).f90
+		  $(DRIVER_UPDATE).$(TYPE) $(DRIVER_UPDATE).f90 \
+		  main.x90
 
 # Location of the infrastucture code (which is a submodule of the
 # PSyclone git repo).
@@ -142,6 +154,9 @@ $(KERNELS):  $(INF_LIB)
 %.o: %.f90
 	$(F90) $(F90FLAGS) -c $<
 
+%.x90: %.X90
+	$(FPP) $< >$@
+
 # This target requires that the netcdf (Fortran) development package be
 # installed
 $(EXTRACT_DIR)/lib_kernel_data_netcdf.a:

examples/gocean/eg5/extract/test.x90 → examples/gocean/eg5/extract/test.X90

@@ -1,4 +1,16 @@
 Program test
+    !> This program is a simple gocean program that contains two invokes,
+    !> the first to initialise a few fields, the second to update a field.
+    !> It can be used with kernel extraction for both invokes. After
+    !> instrumenting the code using PSyclone with the extract_transform.py
+    !> script, it can be compiled, and when executed will create two kernel data
+    !> files. The instrumentation will also create two stand-alone driver
+    !> programs, which can be compiled. They will each read the corresponding
+    !> kernel data file, execute the kernel, and compare the results.
+
+#ifdef _MPI
+    use mpi
+#endif
     USE field_mod
     USE grid_mod
     use decomposition_mod, only    : decomposition_type
@@ -17,6 +29,11 @@ Program test
     ! interface does not provide a method for single precision).
     real(kind=kind(1.0d0))        :: z
     TYPE(grid_type), target       :: grid
+#ifdef _MPI
+    integer                       :: ierr
+
+    call MPI_Init(ierr)
+#endif
 
     call parallel_init()
     call extract_PSyDataInit()
@@ -42,4 +59,9 @@ Program test
 
     print *,a_fld%data(1:5, 1:5)
     call extract_PSyDataShutdown()
+
+#ifdef _MPI
+    call MPI_Finalize(ierr)
+#endif
+
 end program test

examples/gocean/eg5/profile/README.md

@@ -89,10 +89,10 @@ library, you should see:
  ...
  profile_PSyDataInit called
  ...
- PreStart called for module 'invoke_0' region 'r0'
- PostEnd called for module 'invoke_0' region 'r0'
- PreStart called for module 'invoke_1_update_field' region 'r0'
- PostEnd called for module 'invoke_1_update_field' region 'r0'
+ PreStart called for module 'psy_test' region 'invoke_0-r0'
+ PostEnd called for module 'psy_test' region 'invoke_0-r0'
+ PreStart called for module 'psy_test' region 'invoke_1_update_field-r0'
+ PostEnd called for module 'psy_test' region 'invoke_1_update_field-r0'
  ...  
  profile_PSyDataShutdown called
 ```

examples/gocean/eg5/profile/test.x90

@@ -37,10 +37,10 @@
 !> stdout. Expected output:
 !>
 !>  profile_PSyDataInit called
-!>  PreStart called for module 'psy_test' region 'invoke_0:r0'
-!>  PostEnd called for module 'psy_test' region 'invoke_0:r0'
-!>  PreStart called for module 'psy_test' region 'invoke_1_update_field:update_field_code:r0'
-!>  PostEnd called for module 'psy_test' region 'invoke_1_update_field:update_field_code:r0'
+!>  PreStart called for module 'psy_test' region 'invoke_0-r0'
+!>  PostEnd called for module 'psy_test' region 'invoke_0-r0'
+!>  PreStart called for module 'psy_test' region 'invoke_1_update_field-r0'
+!>  PostEnd called for module 'psy_test' region 'invoke_1_update_field-r0'
 
 Program test
     USE field_mod

examples/lfric/eg17/full_example_extract/.gitignore

@@ -1,5 +1,6 @@
 extract.standalone
 extract.netcdf
+main.x90
 main_alg.f90
 main_psy.f90
 main-*.nc