Merge branch 'feature/chey11' into develop

build_scripts/chey_intel_mkl_omp.sh

@@ -0,0 +1,17 @@
+#!/bin/bash
+
+cd ../fortran/build_help
+ifort -o sizes -qopenmp omp_sizes.f90
+python sub_sizes.py
+
+cd ..
+ifort ompgen.F90 -qopenmp -fpp -save-temps > /dev/null 2>&1
+mv ompgen.i90 omp.f90
+f2py *.f90 -m _wrffortran -h wrffortran.pyf --overwrite-signature
+cd ..
+
+python setup.py clean --all
+python setup.py config_fc --f90flags="-xHost -mkl -qopenmp" build_ext --rpath="/glade/u/apps/opt/intel/2017u1/compilers_and_libraries/linux/lib/intel64_lin:/glade/u/apps/opt/intel/2017u1/compilers_and_libraries/linux/mkl/lib/intel64_lin" --library-dirs="/glade/u/apps/opt/intel/2017u1/compilers_and_libraries/linux/lib/intel64_lin:/glade/u/apps/opt/intel/2017u1/compilers_and_libraries/linux/mkl/lib/intel64_lin" --libraries="mkl_intel_ilp64 mkl_intel_thread mkl_core iomp5 pthread m dl" build --compiler=intelem --fcompiler=intelem bdist_wheel
+
+#pip install .
+

chey_intel.py

@@ -0,0 +1,219 @@
+# http://developer.intel.com/software/products/compilers/flin/
+from __future__ import division, absolute_import, print_function
+
+import sys
+
+from numpy.distutils.ccompiler import simple_version_match
+from numpy.distutils.fcompiler import FCompiler, dummy_fortran_file
+
+compilers = ['IntelFCompiler', 'IntelVisualFCompiler',
+             'IntelItaniumFCompiler', 'IntelItaniumVisualFCompiler',
+             'IntelEM64VisualFCompiler', 'IntelEM64TFCompiler']
+
+
+def intel_version_match(type):
+    # Match against the important stuff in the version string
+    return simple_version_match(start=r'Intel.*?Fortran.*?(?:%s).*?Version' % (type,))
+
+
+class BaseIntelFCompiler(FCompiler):
+    def update_executables(self):
+        f = dummy_fortran_file()
+        self.executables['version_cmd'] = ['<F77>', '-FI', '-V', '-c',
+                                           f + '.f', '-o', f + '.o']
+
+    def runtime_library_dir_option(self, dir):
+        return '-Wl,-rpath="%s"' % dir
+
+
+class IntelFCompiler(BaseIntelFCompiler):
+
+    compiler_type = 'intel'
+    compiler_aliases = ('ifort',)
+    description = 'Intel Fortran Compiler for 32-bit apps'
+    version_match = intel_version_match('32-bit|IA-32')
+
+    possible_executables = ['ifort', 'ifc']
+
+    executables = {
+        'version_cmd'  : None,          # set by update_executables
+        'compiler_f77' : [None, "-72", "-w90", "-w95"],
+        'compiler_f90' : [None],
+        'compiler_fix' : [None, "-FI"],
+        'linker_so'    : ["<F90>", "-shared"],
+        'archiver'     : ["ar", "-cr"],
+        'ranlib'       : ["ranlib"]
+        }
+
+    pic_flags = ['-fPIC']
+    module_dir_switch = '-module '  # Don't remove ending space!
+    module_include_switch = '-I'
+
+    def get_flags_free(self):
+        return ['-FR']
+
+    def get_flags(self):
+        return ['-fPIC']
+
+    def get_flags_opt(self):  # Scipy test failures with -O2
+        v = self.get_version()
+        mpopt = ''
+        return ['-fp-model strict -O2']
+
+    def get_flags_arch(self):
+        return []
+
+    def get_flags_linker_so(self):
+        opt = FCompiler.get_flags_linker_so(self)
+        v = self.get_version()
+        if v and v >= '8.0':
+            opt.append('-nofor_main')
+        if sys.platform == 'darwin':
+            # Here, it's -dynamiclib
+            try:
+                idx = opt.index('-shared')
+                opt.remove('-shared')
+            except ValueError:
+                idx = 0
+            opt[idx:idx] = ['-dynamiclib', '-Wl,-undefined,dynamic_lookup']
+        return opt
+
+
+class IntelItaniumFCompiler(IntelFCompiler):
+    compiler_type = 'intele'
+    compiler_aliases = ()
+    description = 'Intel Fortran Compiler for Itanium apps'
+
+    version_match = intel_version_match('Itanium|IA-64')
+
+    possible_executables = ['ifort', 'efort', 'efc']
+
+    executables = {
+        'version_cmd'  : None,
+        'compiler_f77' : [None, "-FI", "-w90", "-w95"],
+        'compiler_fix' : [None, "-FI"],
+        'compiler_f90' : [None],
+        'linker_so'    : ['<F90>', "-shared"],
+        'archiver'     : ["ar", "-cr"],
+        'ranlib'       : ["ranlib"]
+        }
+
+
+class IntelEM64TFCompiler(IntelFCompiler):
+    compiler_type = 'intelem'
+    compiler_aliases = ()
+    description = 'Intel Fortran Compiler for 64-bit apps'
+
+    version_match = intel_version_match('EM64T-based|Intel\\(R\\) 64|64|IA-64|64-bit')
+
+    possible_executables = ['ifort', 'efort', 'efc']
+
+    executables = {
+        'version_cmd'  : None,
+        'compiler_f77' : [None, "-FI"],
+        'compiler_fix' : [None, "-FI"],
+        'compiler_f90' : [None],
+        'linker_so'    : ['<F90>', "-shared"],
+        'archiver'     : ["ar", "-cr"],
+        'ranlib'       : ["ranlib"]
+        }
+
+    def get_flags(self):
+        return ['-fPIC']
+
+    def get_flags_opt(self):  # Scipy test failures with -O2
+        v = self.get_version()
+        mpopt = ''
+        return ['-fp-model strict -O2']
+
+    def get_flags_arch(self):
+        return ['']
+
+# Is there no difference in the version string between the above compilers
+# and the Visual compilers?
+
+
+class IntelVisualFCompiler(BaseIntelFCompiler):
+    compiler_type = 'intelv'
+    description = 'Intel Visual Fortran Compiler for 32-bit apps'
+    version_match = intel_version_match('32-bit|IA-32')
+
+    def update_executables(self):
+        f = dummy_fortran_file()
+        self.executables['version_cmd'] = ['<F77>', '/FI', '/c',
+                                           f + '.f', '/o', f + '.o']
+
+    ar_exe = 'lib.exe'
+    possible_executables = ['ifort', 'ifl']
+
+    executables = {
+        'version_cmd'  : None,
+        'compiler_f77' : [None],
+        'compiler_fix' : [None],
+        'compiler_f90' : [None],
+        'linker_so'    : [None],
+        'archiver'     : [ar_exe, "/verbose", "/OUT:"],
+        'ranlib'       : None
+        }
+
+    compile_switch = '/c '
+    object_switch = '/Fo'     # No space after /Fo!
+    library_switch = '/OUT:'  # No space after /OUT:!
+    module_dir_switch = '/module:'  # No space after /module:
+    module_include_switch = '/I'
+
+    def get_flags(self):
+        opt = ['/nologo', '/MD', '/nbs', '/names:lowercase', '/assume:underscore']
+        return opt
+
+    def get_flags_free(self):
+        return []
+
+    def get_flags_debug(self):
+        return ['/4Yb', '/d2']
+
+    def get_flags_opt(self):
+        return ['/O1']  # Scipy test failures with /O2
+
+    def get_flags_arch(self):
+        return ["/arch:IA32", "/QaxSSE3"]
+
+    def runtime_library_dir_option(self, dir):
+        raise NotImplementedError
+
+
+class IntelItaniumVisualFCompiler(IntelVisualFCompiler):
+    compiler_type = 'intelev'
+    description = 'Intel Visual Fortran Compiler for Itanium apps'
+
+    version_match = intel_version_match('Itanium')
+
+    possible_executables = ['efl']  # XXX this is a wild guess
+    ar_exe = IntelVisualFCompiler.ar_exe
+
+    executables = {
+        'version_cmd'  : None,
+        'compiler_f77' : [None, "-FI", "-w90", "-w95"],
+        'compiler_fix' : [None, "-FI", "-4L72", "-w"],
+        'compiler_f90' : [None],
+        'linker_so'    : ['<F90>', "-shared"],
+        'archiver'     : [ar_exe, "/verbose", "/OUT:"],
+        'ranlib'       : None
+        }
+
+
+class IntelEM64VisualFCompiler(IntelVisualFCompiler):
+    compiler_type = 'intelvem'
+    description = 'Intel Visual Fortran Compiler for 64-bit apps'
+
+    version_match = simple_version_match(start=r'Intel\(R\).*?64,')
+
+    def get_flags_arch(self):
+        return ['']
+
+
+if __name__ == '__main__':
+    from distutils import log
+    log.set_verbosity(2)
+    from numpy.distutils import customized_fcompiler
+    print(customized_fcompiler(compiler='intel').get_version())

conda_recipe/build.sh

@@ -1,7 +1,11 @@
 #!/bin/bash
 
 cd fortran/build_help
-gfortran -o sizes -fopenmp omp_sizes.f90
+if [ `uname` == Darwin ]; then
+    gfortran -o sizes -fopenmp omp_sizes.f90 -Wl,-rpath,${CONDA_PREFIX}/lib
+else
+    gfortran -o sizes -fopenmp omp_sizes.f90
+fi
 python sub_sizes.py
 
 cd ..
@@ -16,5 +20,3 @@ fi
 $PYTHON setup.py config_fc --f90flags="-mtune=generic -fopenmp" build_ext --libraries="gomp" build
 
 $PYTHON setup.py install --single-version-externally-managed --record=record.txt
-
-

conda_recipe/meta.yaml

@@ -6,7 +6,7 @@ package:
 
 source:
   git_url: https://github.com/NCAR/wrf-python
-  git_branch: release/1.1.0
+  git_branch: develop 
   
 build:
   number: 0

doc/source/new.rst

@@ -4,6 +4,19 @@ What's New
 Releases
 -------------
 
+v1.1.1
+^^^^^^^^^^^^^^
+
+- Release 1.1.1
+- Added script for building on Cheyenne with maxed out Intel settings, which 
+  also required a patch for numpy.distutils.
+- Fixed a few unicode characters hiding in a docstring that were causing 
+  problems on Cheyenne, and also building the docs with Sphinx on Python 2.x.
+- Fix issue with np.amax not working with xarray on Cheyenne, causing an error
+  with the mdbz product.
+- Fix cape_2d private variable bug when running with multiple CPUs.
+
+
 v1.1.0
 ^^^^^^^^^^^^^^
 

fortran/rip_cape.f90

@@ -708,7 +708,7 @@ SUBROUTINE DCAPECALC2D(prs,tmk,qvp,ght,ter,sfp,cape,cin,&
     !$OMP PARALLEL DO COLLAPSE(2) PRIVATE(tlcl, ethpari, &
     !$OMP zlcl, kk, ilcl, klcl, tmklift, tvenv, tvlift, ghtlift, &
     !$OMP facden, tmkenv, qvpenv, eslift, qvplift, buoy, benamin, &
-    !$OMP benaccum, zrel, kmax, dz, elfound, &
+    !$OMP benaccum, zrel, kmax, dz, elfound, gammam, cpm, e, &
     !$OMP kel, klfc, pavg, p2, p1, totthe, totqvp, totprs, &
     !$OMP i, j, k, kpar, kpar1, kpar2, qvppari, tmkpari, p, pup, pdn, th, &
     !$OMP pp1, pp2, ethmax, eth_temp, klev) SCHEDULE(runtime)

setup.py

@@ -1,8 +1,26 @@
 import os
 import sys
 import setuptools
-import numpy.distutils.core
+import socket
 
+if not socket.gethostname().startswith("cheyenne"):
+    import numpy.distutils.core
+else:
+    import chey_intel
+    import numpy.distutils.core
+    import numpy.distutils.fcompiler.intel
+    
+    numpy.distutils.fcompiler.intel.IntelFCompiler = chey_intel.IntelFCompiler
+    numpy.distutils.fcompiler.intel.IntelVisualFCompiler = (
+        chey_intel.IntelVisualFCompiler)
+    numpy.distutils.fcompiler.intel.IntelItaniumFCompiler = (
+        chey_intel.IntelItaniumFCompiler)
+    numpy.distutils.fcompiler.intel.IntelItaniumVisualFCompiler = (
+        chey_intel.IntelItaniumVisualFCompiler)
+    numpy.distutils.fcompiler.intel.IntelEM64VisualFCompiler = (
+        chey_intel.IntelEM64VisualFCompiler)
+    numpy.distutils.fcompiler.intel.IntelEM64TFCompiler = (
+        chey_intel.IntelEM64TFCompiler)
 
 ext1 = numpy.distutils.core.Extension(
     name = "wrf._wrffortran",

src/wrf/constants.py

@@ -76,7 +76,7 @@ else:
 # Add the integers based on python 2.x or 3.x
 def default_fill(dtype=None):
     dt = np.dtype(dtype) if dtype is not None else None
-    return _DEFAULT_FILL_MAP.get(dt, 0)
+    return _DEFAULT_FILL_MAP.get(dt, Constants.DEFAULT_FILL)
     
     
     

src/wrf/extension.py

@@ -1569,10 +1569,10 @@ def omp_get_wtime():
     """Return elapsed wall clock time in seconds.
     
     The omp_get_wtime routine returns a value equal to the elapsed wall clock 
-    time in seconds since some “time in the past”. The actual 
+    time in seconds since some "time in the past". The actual 
-    “time in the past” is arbitrary, but it is guaranteed not to change during 
+    "time in the past" is arbitrary, but it is guaranteed not to change during 
     the execution of the application program. The time returned is a 
-    “per-thread time”, so it is not required to be globally consistent across 
+    "per-thread time", so it is not required to be globally consistent across 
     all threads participating in an application.
     
     Returns:

src/wrf/g_dbz.py

@@ -6,7 +6,7 @@ import numpy as np
 #from .extension import computedbz,computetk
 from .extension import _dbz, _tk
 from .constants import Constants
-from .util import extract_vars
+from .util import extract_vars, to_np
 from .metadecorators import copy_and_set_metadata
 
 
@@ -194,7 +194,7 @@ def get_max_dbz(wrfin, timeidx=0, method="cat",
         be a :class:`numpy.ndarray` object with no metadata.
     
     """
-    return np.amax(get_dbz(wrfin, timeidx, method, squeeze, cache, meta,
+    return np.amax(to_np(get_dbz(wrfin, timeidx, method, squeeze, cache, meta,
-                           _key, use_varint, use_liqskin), 
+                           _key, use_varint, use_liqskin)), 
                   axis=-3)
 

src/wrf/version.py

@@ -1,2 +1,2 @@
-__version__ = "1.1.0"
+__version__ = "1.1.1"