portelli/Grid
1
0
Fork 0
mirror of https://github.com/paboyle/Grid.git synced 2024-11-10 07:55:35 +00:00
Code Releases Activity

Merge branch 'release/v0.6.0' into feature/feynman-rules

Browse Source
This commit is contained in:
Antonin Portelli 2016-10-31 18:36:21 +00:00
commit d5e95bc350
6 changed files with 252 additions and 143 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

112
README.md
View File

@ -16,11 +16,27 @@
**Data parallel C++ mathematical object library.** **Data parallel C++ mathematical object library.**
Please send all pull requests to the `develop` branch.
License: GPL v2. License: GPL v2.
Last update 2016/08/03. Last update Nov 2016.
_Please send all pull requests to the `develop` branch._
### Bug report
_To help us tracking and solving more efficiently issues with Grid, please report problems using the issue system of GitHub rather than sending emails to Grid developers._
When you file an issue, please go though the following checklist:
1. Check that the code is pointing to the `HEAD` of `develop` or any commit in `master` which is tagged with a version number.
2. Give a description of the target platform (CPU, network, compiler).
3. Give the exact `configure` command used.
4. Attach `config.log`.
5. Attach `config.summary`.
6. Attach the output of `make V=1`.
7. Describe the issue and any previous attempt to solve it. If relevant, show how to reproduce the issue using a minimal working example.
### Description ### Description
This library provides data parallel C++ container classes with internal memory layout This library provides data parallel C++ container classes with internal memory layout
@ -42,7 +58,7 @@ optimally use MPI, OpenMP and SIMD parallelism under the hood. This is a signifi
for most programmers. for most programmers.
The layout transformations are parametrised by the SIMD vector length. This adapts according to the architecture. The layout transformations are parametrised by the SIMD vector length. This adapts according to the architecture.
Presently SSE4 (128 bit) AVX, AVX2 (256 bit) and IMCI and AVX512 (512 bit) targets are supported (ARM NEON and BG/Q QPX on the way). Presently SSE4 (128 bit) AVX, AVX2, QPX (256 bit), IMCI, and AVX512 (512 bit) targets are supported (ARM NEON on the way).
These are presented as `vRealF`, `vRealD`, `vComplexF`, and `vComplexD` internal vector data types. These may be useful in themselves for other programmers. These are presented as `vRealF`, `vRealD`, `vComplexF`, and `vComplexD` internal vector data types. These may be useful in themselves for other programmers.
The corresponding scalar types are named `RealF`, `RealD`, `ComplexF` and `ComplexD`. The corresponding scalar types are named `RealF`, `RealD`, `ComplexF` and `ComplexD`.
@ -50,7 +66,7 @@ The corresponding scalar types are named `RealF`, `RealD`, `ComplexF` and `Compl
MPI, OpenMP, and SIMD parallelism are present in the library. MPI, OpenMP, and SIMD parallelism are present in the library.
Please see https://arxiv.org/abs/1512.03487 for more detail. Please see https://arxiv.org/abs/1512.03487 for more detail.
### Installation ### Quick start
First, start by cloning the repository: First, start by cloning the repository:
``` bash ``` bash
@ -71,12 +87,10 @@ mkdir build; cd build
../configure --enable-precision=double --enable-simd=AVX --enable-comms=mpi-auto --prefix=<path> ../configure --enable-precision=double --enable-simd=AVX --enable-comms=mpi-auto --prefix=<path>
``` ```
where `--enable-precision=` set the default precision (`single` or `double`), where `--enable-precision=` set the default precision,
`--enable-simd=` set the SIMD type (see possible values below), `--enable- `--enable-simd=` set the SIMD type, `--enable-
comms=` set the protocol used for communications (`none`, `mpi`, `mpi-auto` or comms=`, and `<path>` should be replaced by the prefix path where you want to
`shmem`), and `<path>` should be replaced by the prefix path where you want to install Grid. Other options are detailed in the next section, you can also use `configure
install Grid. The `mpi-auto` communication option set `configure` to determine
automatically how to link to MPI. Other options are available, use `configure
--help` to display them. Like with any other program using GNU autotool, the --help` to display them. Like with any other program using GNU autotool, the
`CXX`, `CXXFLAGS`, `LDFLAGS`, ... environment variables can be modified to `CXX`, `CXXFLAGS`, `LDFLAGS`, ... environment variables can be modified to
customise the build. customise the build.
@ -93,24 +107,86 @@ To minimise the build time, only the tests at the root of the `tests` directory
make -C tests/<subdir> tests make -C tests/<subdir> tests
``` ```
### Build configuration options
- `--prefix=<path>`: installation prefix for Grid.
- `--with-gmp=<path>`: look for GMP in the UNIX prefix `<path>`
- `--with-mpfr=<path>`: look for MPFR in the UNIX prefix `<path>`
- `--with-fftw=<path>`: look for FFTW in the UNIX prefix `<path>`
- `--enable-lapack[=<path>]`: enable LAPACK support in Lanczos eigensolver. A UNIX prefix containing the library can be specified (optional).
- `--enable-mkl[=<path>]`: use Intel MKL for FFT (and LAPACK if enabled) routines. A UNIX prefix containing the library can be specified (optional).
- `--enable-numa`: ???
- `--enable-simd=<code>`: setup Grid for the SIMD target `<code>` (default: `GEN`). A list of possible SIMD targets is detailed in a section below.
- `--enable-precision={single|double}`: set the default precision (default: `double`).
- `--enable-precision=<comm>`: Use `<comm>` for message passing (default: `none`). A list of possible SIMD targets is detailed in a section below.
- `--enable-rng={ranlux48|mt19937}`: choose the RNG (default: `ranlux48 `).
- `--disable-timers`: disable system dependent high-resolution timers.
- `--enable-chroma`: enable Chroma regression tests.
### Possible communication interfaces
The following options can be use with the `--enable-simd=` option to target different communication interfaces:
| `<comm>` | Description |
| ------------- | -------------------------------------------- |
| `none` | no communications |
| `mpi[-auto]` | MPI communications |
| `mpi3[-auto]` | MPI communications using MPI 3 shared memory |
| `shmem ` | Cray SHMEM communications |
For `mpi` and `mpi3` the optional `-auto` suffix instructs the `configure` scripts to determine all the necessary compilation and linking flags. This is done by extracting the informations from the MPI wrapper specified in the environment variable `MPICXX` (if not specified `configure` will scan though a list of default names).
### Possible SIMD types ### Possible SIMD types
The following options can be use with the `--enable-simd=` option to target different SIMD instruction sets: The following options can be use with the `--enable-simd=` option to target different SIMD instruction sets:
| String | Description | | `<code>` | Description |
| ----------- | -------------------------------------- | | ----------- | -------------------------------------- |
| `GEN` | generic portable vector code | | `GEN` | generic portable vector code |
| `SSE4` | SSE 4.2 (128 bit) | | `SSE4` | SSE 4.2 (128 bit) |
| `AVX` | AVX (256 bit) | | `AVX` | AVX (256 bit) |
| `AVXFMA4` | AVX (256 bit) + FMA | | `AVXFMA` | AVX (256 bit) + FMA |
| `AVXFMA4` | AVX (256 bit) + FMA4 |
| `AVX2` | AVX 2 (256 bit) | | `AVX2` | AVX 2 (256 bit) |
| `AVX512` | AVX 512 bit | | `AVX512` | AVX 512 bit |
| `AVX512MIC` | AVX 512 bit for Intel MIC architecture | | `QPX` | QPX (256 bit) |
| `ICMI` | Intel ICMI instructions (512 bit) |
Alternatively, some CPU codenames can be directly used: Alternatively, some CPU codenames can be directly used:
| String | Description | | `<code>` | Description |
| ----------- | -------------------------------------- | | ----------- | -------------------------------------- |
| `KNC` | [Intel Knights Corner](http://ark.intel.com/products/codename/57721/Knights-Corner) | | `KNC` | [Intel Xeon Phi codename Knights Corner](http://ark.intel.com/products/codename/57721/Knights-Corner) |
| `KNL` | [Intel Knights Landing](http://ark.intel.com/products/codename/48999/Knights-Landing) | | `KNL` | [Intel Xeon Phi codename Knights Landing](http://ark.intel.com/products/codename/48999/Knights-Landing) |
| `BGQ` | Blue Gene/Q |
#### Notes:
- We currently support AVX512 only for the Intel compiler. Support for GCC and clang will appear in future versions.
- For BG/Q only [bgclang](http://trac.alcf.anl.gov/projects/llvm-bgq) is supported. We do not presently plan to support more compilers for this platform.
- BG/Q performances are currently rather poor. This is being investigated for future versions.
### Build setup for Intel Knights Landing platform
The following configuration is recommended for the Intel Knights Landing platform:
``` bash
../configure --enable-precision=double\
--enable-simd=KNL \
--enable-comms=mpi3-auto \
--with-gmp=<path> \
--with-mpfr=<path> \
--enable-mkl \
CXX=icpc MPICXX=mpiicpc
```
where `<path>` is the UNIX prefix where GMP and MPFR are installed. If you are working on a Cray machine that does not use the `mpiicpc` wrapper, please use:
``` bash
../configure --enable-precision=double\
--enable-simd=KNL \
--enable-comms=mpi3 \
--with-gmp=<path> \
--with-mpfr=<path> \
--enable-mkl \
CXX=CC CC=cc
```

4
VERSION
View File

@ -1,4 +1,6 @@
Version : 0.5.0 Version : 0.6.0
- AVX512, AVX2, AVX, SSE good - AVX512, AVX2, AVX, SSE good
- Clang 3.5 and above, ICPC v16 and above, GCC 4.9 and above - Clang 3.5 and above, ICPC v16 and above, GCC 4.9 and above
- MPI and MPI3
- HiRep, Smearing, Generic gauge group

236
configure.ac
View File

@ -1,5 +1,5 @@
AC_PREREQ([2.63]) AC_PREREQ([2.63])
AC_INIT([Grid], [0.5.1-dev], [https://github.com/paboyle/Grid], [Grid]) AC_INIT([Grid], [0.6.0], [https://github.com/paboyle/Grid], [Grid])
AC_CANONICAL_BUILD AC_CANONICAL_BUILD
AC_CANONICAL_HOST AC_CANONICAL_HOST
AC_CANONICAL_TARGET AC_CANONICAL_TARGET
@ -9,18 +9,29 @@ AC_CONFIG_SRCDIR([lib/Grid.h])
AC_CONFIG_HEADERS([lib/Config.h]) AC_CONFIG_HEADERS([lib/Config.h])
m4_ifdef([AM_SILENT_RULES], [AM_SILENT_RULES([yes])]) m4_ifdef([AM_SILENT_RULES], [AM_SILENT_RULES([yes])])
############### Checks for programs ############### Checks for programs
AC_LANG(C++)
CXXFLAGS="-O3 $CXXFLAGS" CXXFLAGS="-O3 $CXXFLAGS"
AC_PROG_CXX AC_PROG_CXX
AC_PROG_RANLIB AC_PROG_RANLIB
############ openmp ############### ############### Get compiler informations
AC_LANG([C++])
AX_CXX_COMPILE_STDCXX_11([noext],[mandatory])
AX_COMPILER_VENDOR
AC_DEFINE_UNQUOTED([CXX_COMP_VENDOR],["$ax_cv_cxx_compiler_vendor"],
[vendor of C++ compiler that will compile the code])
AX_GXX_VERSION
AC_DEFINE_UNQUOTED([GXX_VERSION],["$GXX_VERSION"],
[version of g++ that will compile the code])
############### Checks for typedefs, structures, and compiler characteristics
AC_TYPE_SIZE_T
AC_TYPE_UINT32_T
AC_TYPE_UINT64_T
############### OpenMP
AC_OPENMP AC_OPENMP
ac_openmp=no ac_openmp=no
if test "${OPENMP_CXXFLAGS}X" != "X"; then if test "${OPENMP_CXXFLAGS}X" != "X"; then
ac_openmp=yes ac_openmp=yes
AM_CXXFLAGS="$OPENMP_CXXFLAGS $AM_CXXFLAGS" AM_CXXFLAGS="$OPENMP_CXXFLAGS $AM_CXXFLAGS"
@ -37,12 +48,7 @@ AC_CHECK_HEADERS(execinfo.h)
AC_CHECK_DECLS([ntohll],[], [], [[#include <arpa/inet.h>]]) AC_CHECK_DECLS([ntohll],[], [], [[#include <arpa/inet.h>]])
AC_CHECK_DECLS([be64toh],[], [], [[#include <arpa/inet.h>]]) AC_CHECK_DECLS([be64toh],[], [], [[#include <arpa/inet.h>]])
############### Checks for typedefs, structures, and compiler characteristics ############### GMP and MPFR
AC_TYPE_SIZE_T
AC_TYPE_UINT32_T
AC_TYPE_UINT64_T
############### GMP and MPFR #################
AC_ARG_WITH([gmp], AC_ARG_WITH([gmp],
[AS_HELP_STRING([--with-gmp=prefix], [AS_HELP_STRING([--with-gmp=prefix],
[try this for a non-standard install prefix of the GMP library])], [try this for a non-standard install prefix of the GMP library])],
@ -54,7 +60,14 @@ AC_ARG_WITH([mpfr],
[AM_CXXFLAGS="-I$with_mpfr/include $AM_CXXFLAGS"] [AM_CXXFLAGS="-I$with_mpfr/include $AM_CXXFLAGS"]
[AM_LDFLAGS="-L$with_mpfr/lib $AM_LDFLAGS"]) [AM_LDFLAGS="-L$with_mpfr/lib $AM_LDFLAGS"])
################## lapack #################### ############### FFTW3
AC_ARG_WITH([fftw],
[AS_HELP_STRING([--with-fftw=prefix],
[try this for a non-standard install prefix of the FFTW3 library])],
[AM_CXXFLAGS="-I$with_fftw/include $AM_CXXFLAGS"]
[AM_LDFLAGS="-L$with_fftw/lib $AM_LDFLAGS"])
############### lapack
AC_ARG_ENABLE([lapack], AC_ARG_ENABLE([lapack],
[AC_HELP_STRING([--enable-lapack=yes|no|prefix], [enable LAPACK])], [AC_HELP_STRING([--enable-lapack=yes|no|prefix], [enable LAPACK])],
[ac_LAPACK=${enable_lapack}], [ac_LAPACK=no]) [ac_LAPACK=${enable_lapack}], [ac_LAPACK=no])
@ -67,10 +80,26 @@ case ${ac_LAPACK} in
*) *)
AM_CXXFLAGS="-I$ac_LAPACK/include $AM_CXXFLAGS" AM_CXXFLAGS="-I$ac_LAPACK/include $AM_CXXFLAGS"
AM_LDFLAGS="-L$ac_LAPACK/lib $AM_LDFLAGS" AM_LDFLAGS="-L$ac_LAPACK/lib $AM_LDFLAGS"
AC_DEFINE([USE_LAPACK],[1],[use LAPACK]) AC_DEFINE([USE_LAPACK],[1],[use LAPACK]);;
esac esac
################## first-touch #################### ############### MKL
AC_ARG_ENABLE([mkl],
[AC_HELP_STRING([--enable-mkl=yes|no|prefix], [enable Intel MKL for LAPACK & FFTW])],
[ac_MKL=${enable_mkl}], [ac_MKL=no])
case ${ac_MKL} in
no)
;;
yes)
AC_DEFINE([USE_MKL], [1], [Define to 1 if you use the Intel MKL]);;
*)
AM_CXXFLAGS="-I$ac_MKL/include $AM_CXXFLAGS"
AM_LDFLAGS="-L$ac_MKL/lib $AM_LDFLAGS"
AC_DEFINE([USE_MKL], [1], [Define to 1 if you use the Intel MKL]);;
esac
############### first-touch
AC_ARG_ENABLE([numa], AC_ARG_ENABLE([numa],
[AC_HELP_STRING([--enable-numa=yes|no|prefix], [enable first touch numa opt])], [AC_HELP_STRING([--enable-numa=yes|no|prefix], [enable first touch numa opt])],
[ac_NUMA=${enable_NUMA}],[ac_NUMA=no]) [ac_NUMA=${enable_NUMA}],[ac_NUMA=no])
@ -84,56 +113,44 @@ case ${ac_NUMA} in
AC_DEFINE([GRID_NUMA],[1],[First touch numa locality]);; AC_DEFINE([GRID_NUMA],[1],[First touch numa locality]);;
esac esac
################## FFTW3 ####################
AC_ARG_WITH([fftw],
[AS_HELP_STRING([--with-fftw=prefix],
[try this for a non-standard install prefix of the FFTW3 library])],
[AM_CXXFLAGS="-I$with_fftw/include $AM_CXXFLAGS"]
[AM_LDFLAGS="-L$with_fftw/lib $AM_LDFLAGS"])
################ Get compiler informations
AC_LANG([C++])
AX_CXX_COMPILE_STDCXX_11([noext],[mandatory])
AX_COMPILER_VENDOR
AC_DEFINE_UNQUOTED([CXX_COMP_VENDOR],["$ax_cv_cxx_compiler_vendor"],
[vendor of C++ compiler that will compile the code])
AX_GXX_VERSION
AC_DEFINE_UNQUOTED([GXX_VERSION],["$GXX_VERSION"],
[version of g++ that will compile the code])
############### Checks for library functions ############### Checks for library functions
CXXFLAGS_CPY=$CXXFLAGS CXXFLAGS_CPY=$CXXFLAGS
LDFLAGS_CPY=$LDFLAGS LDFLAGS_CPY=$LDFLAGS
CXXFLAGS="$AM_CXXFLAGS $CXXFLAGS" CXXFLAGS="$AM_CXXFLAGS $CXXFLAGS"
LDFLAGS="$AM_LDFLAGS $LDFLAGS" LDFLAGS="$AM_LDFLAGS $LDFLAGS"
AC_CHECK_FUNCS([gettimeofday]) AC_CHECK_FUNCS([gettimeofday])
AC_CHECK_LIB([gmp],[__gmpf_init],
[AC_CHECK_LIB([mpfr],[mpfr_init], if test "${ac_MKL}x" != "nox"; then
[AC_DEFINE([HAVE_LIBMPFR], [1], [Define to 1 if you have the `MPFR' library (-lmpfr).])] AC_SEARCH_LIBS([mkl_set_interface_layer], [mkl_rt], [],
[have_mpfr=true] [AC_MSG_ERROR("MKL enabled but library not found")])
[LIBS="$LIBS -lmpfr"], fi
[AC_MSG_ERROR([MPFR library not found])])]
[AC_DEFINE([HAVE_LIBGMP], [1], [Define to 1 if you have the `GMP' library (-lgmp).])] AC_SEARCH_LIBS([__gmpf_init], [gmp],
[have_gmp=true] [AC_SEARCH_LIBS([mpfr_init], [mpfr],
[LIBS="$LIBS -lgmp"], [AC_DEFINE([HAVE_LIBMPFR], [1],
[AC_MSG_WARN([**** GMP library not found, Grid can still compile but RHMC will not work ****])]) [Define to 1 if you have the `MPFR' library])]
[have_mpfr=true], [AC_MSG_ERROR([MPFR library not found])])]
[AC_DEFINE([HAVE_LIBGMP], [1], [Define to 1 if you have the `GMP' library])]
[have_gmp=true])
if test "${ac_LAPACK}x" != "nox"; then if test "${ac_LAPACK}x" != "nox"; then
AC_CHECK_LIB([lapack],[LAPACKE_sbdsdc],[], AC_SEARCH_LIBS([LAPACKE_sbdsdc], [lapack], [],
[AC_MSG_ERROR("LAPACK enabled but library not found")]) [AC_MSG_ERROR("LAPACK enabled but library not found")])
fi fi
AC_CHECK_LIB([fftw3],[fftw_execute],
[AC_DEFINE([HAVE_FFTW],[1],[Define to 1 if you have the `FFTW' library (-lfftw3).])] AC_SEARCH_LIBS([fftw_execute], [fftw3],
[have_fftw=true] [AC_SEARCH_LIBS([fftwf_execute], [fftw3f], [],
[LIBS="$LIBS -lfftw3 -lfftw3f"], [AC_MSG_ERROR("single precision FFTW library not found")])]
[AC_MSG_WARN([**** FFTW library not found, Grid can still compile but FFT-based routines will not work ****])]) [AC_DEFINE([HAVE_FFTW], [1], [Define to 1 if you have the `FFTW' library])]
[have_fftw=true])
CXXFLAGS=$CXXFLAGS_CPY CXXFLAGS=$CXXFLAGS_CPY
LDFLAGS=$LDFLAGS_CPY LDFLAGS=$LDFLAGS_CPY
############### SIMD instruction selection ############### SIMD instruction selection
AC_ARG_ENABLE([simd],[AC_HELP_STRING([--enable-simd=SSE4|AVX|AVXFMA4|AVXFMA|AVX2|AVX512|AVX512MIC|IMCI|KNL|KNC],\ AC_ARG_ENABLE([simd],[AC_HELP_STRING([--enable-simd=<code>],
[Select instructions to be SSE4.0, AVX 1.0, AVX 2.0+FMA, AVX 512, IMCI])],\ [select SIMD target (cf. README.md)])], [ac_SIMD=${enable_simd}], [ac_SIMD=GEN])
[ac_SIMD=${enable_simd}],[ac_SIMD=GEN])
case ${ax_cv_cxx_compiler_vendor} in case ${ax_cv_cxx_compiler_vendor} in
clang|gnu) clang|gnu)
@ -153,12 +170,15 @@ case ${ax_cv_cxx_compiler_vendor} in
AVX2) AVX2)
AC_DEFINE([AVX2],[1],[AVX2 intrinsics]) AC_DEFINE([AVX2],[1],[AVX2 intrinsics])
SIMD_FLAGS='-mavx2 -mfma';; SIMD_FLAGS='-mavx2 -mfma';;
AVX512|AVX512MIC|KNL) AVX512)
AC_DEFINE([AVX512],[1],[AVX512 intrinsics]) AC_DEFINE([AVX512],[1],[AVX512 intrinsics])
SIMD_FLAGS='-mavx512f -mavx512pf -mavx512er -mavx512cd';; SIMD_FLAGS='-mavx512f -mavx512pf -mavx512er -mavx512cd';;
IMCI|KNC) KNC)
AC_DEFINE([IMCI],[1],[IMCI intrinsics for Knights Corner]) AC_DEFINE([IMCI],[1],[IMCI intrinsics for Knights Corner])
SIMD_FLAGS='';; SIMD_FLAGS='';;
KNL)
AC_DEFINE([AVX512],[1],[AVX512 intrinsics])
SIMD_FLAGS='-march=knl';;
GEN) GEN)
AC_DEFINE([GENERIC_VEC],[1],[generic vector code]) AC_DEFINE([GENERIC_VEC],[1],[generic vector code])
SIMD_FLAGS='';; SIMD_FLAGS='';;
@ -176,9 +196,6 @@ case ${ax_cv_cxx_compiler_vendor} in
AVX) AVX)
AC_DEFINE([AVX1],[1],[AVX intrinsics]) AC_DEFINE([AVX1],[1],[AVX intrinsics])
SIMD_FLAGS='-mavx -xavx';; SIMD_FLAGS='-mavx -xavx';;
AVXFMA4)
AC_DEFINE([AVXFMA4],[1],[AVX intrinsics with FMA4])
SIMD_FLAGS='-mavx -mfma';;
AVXFMA) AVXFMA)
AC_DEFINE([AVXFMA],[1],[AVX intrinsics with FMA4]) AC_DEFINE([AVXFMA],[1],[AVX intrinsics with FMA4])
SIMD_FLAGS='-mavx -mfma';; SIMD_FLAGS='-mavx -mfma';;
@ -188,12 +205,12 @@ case ${ax_cv_cxx_compiler_vendor} in
AVX512) AVX512)
AC_DEFINE([AVX512],[1],[AVX512 intrinsics]) AC_DEFINE([AVX512],[1],[AVX512 intrinsics])
SIMD_FLAGS='-xcore-avx512';; SIMD_FLAGS='-xcore-avx512';;
AVX512MIC|KNL) KNC)
AC_DEFINE([AVX512],[1],[AVX512 intrinsics for Knights Landing])
SIMD_FLAGS='-xmic-avx512';;
IMCI|KNC)
AC_DEFINE([IMCI],[1],[IMCI Intrinsics for Knights Corner]) AC_DEFINE([IMCI],[1],[IMCI Intrinsics for Knights Corner])
SIMD_FLAGS='';; SIMD_FLAGS='';;
KNL)
AC_DEFINE([AVX512],[1],[AVX512 intrinsics for Knights Landing])
SIMD_FLAGS='-xmic-avx512';;
GEN) GEN)
AC_DEFINE([GENERIC_VEC],[1],[generic vector code]) AC_DEFINE([GENERIC_VEC],[1],[generic vector code])
SIMD_FLAGS='';; SIMD_FLAGS='';;
@ -208,14 +225,18 @@ AM_CXXFLAGS="$SIMD_FLAGS $AM_CXXFLAGS"
AM_CFLAGS="$SIMD_FLAGS $AM_CFLAGS" AM_CFLAGS="$SIMD_FLAGS $AM_CFLAGS"
case ${ac_SIMD} in case ${ac_SIMD} in
AVX512|AVX512MIC|KNL) AVX512|KNL)
AC_DEFINE([TEST_ZMM],[1],[compile ZMM test]);; AC_DEFINE([TEST_ZMM],[1],[compile ZMM test]);;
*) *)
;; ;;
esac esac
############### precision selection ############### Precision selection
AC_ARG_ENABLE([precision],[AC_HELP_STRING([--enable-precision=single|double],[Select default word size of Real])],[ac_PRECISION=${enable_precision}],[ac_PRECISION=double]) AC_ARG_ENABLE([precision],
[AC_HELP_STRING([--enable-precision=single|double],
[Select default word size of Real])],
[ac_PRECISION=${enable_precision}],[ac_PRECISION=double])
case ${ac_PRECISION} in case ${ac_PRECISION} in
single) single)
AC_DEFINE([GRID_DEFAULT_PRECISION_SINGLE],[1],[GRID_DEFAULT_PRECISION is SINGLE] ) AC_DEFINE([GRID_DEFAULT_PRECISION_SINGLE],[1],[GRID_DEFAULT_PRECISION is SINGLE] )
@ -226,25 +247,17 @@ case ${ac_PRECISION} in
esac esac
############### communication type selection ############### communication type selection
AC_ARG_ENABLE([comms],[AC_HELP_STRING([--enable-comms=none|mpi|mpi-auto|shmem],[Select communications])],[ac_COMMS=${enable_comms}],[ac_COMMS=none]) AC_ARG_ENABLE([comms],[AC_HELP_STRING([--enable-comms=none|mpi|mpi-auto|mpi3|mpi3-auto|shmem],
[Select communications])],[ac_COMMS=${enable_comms}],[ac_COMMS=none])
case ${ac_COMMS} in case ${ac_COMMS} in
none) none)
AC_DEFINE([GRID_COMMS_NONE],[1],[GRID_COMMS_NONE] ) AC_DEFINE([GRID_COMMS_NONE],[1],[GRID_COMMS_NONE] )
;; ;;
mpi-auto) mpi|mpi-auto)
AC_DEFINE([GRID_COMMS_MPI],[1],[GRID_COMMS_MPI] )
LX_FIND_MPI
if test "x$have_CXX_mpi" = 'xno'; then AC_MSG_ERROR(["MPI not found"]); fi
AM_CXXFLAGS="$MPI_CXXFLAGS $AM_CXXFLAGS"
AM_CFLAGS="$MPI_CFLAGS $AM_CFLAGS"
AM_LDFLAGS="`echo $MPI_CXXLDFLAGS | sed -E 's/-l@<:@^ @:>@+//g'` $AM_LDFLAGS"
LIBS="`echo $MPI_CXXLDFLAGS | sed -E 's/-L@<:@^ @:>@+//g'` $LIBS"
;;
mpi)
AC_DEFINE([GRID_COMMS_MPI],[1],[GRID_COMMS_MPI] ) AC_DEFINE([GRID_COMMS_MPI],[1],[GRID_COMMS_MPI] )
;; ;;
mpi3) mpi3|mpi3-auto)
AC_DEFINE([GRID_COMMS_MPI3],[1],[GRID_COMMS_MPI3] ) AC_DEFINE([GRID_COMMS_MPI3],[1],[GRID_COMMS_MPI3] )
;; ;;
shmem) shmem)
@ -254,9 +267,23 @@ case ${ac_COMMS} in
AC_MSG_ERROR([${ac_COMMS} unsupported --enable-comms option]); AC_MSG_ERROR([${ac_COMMS} unsupported --enable-comms option]);
;; ;;
esac esac
case ${ac_COMMS} in
*-auto)
LX_FIND_MPI
if test "x$have_CXX_mpi" = 'xno'; then AC_MSG_ERROR(["MPI not found"]); fi
AM_CXXFLAGS="$MPI_CXXFLAGS $AM_CXXFLAGS"
AM_CFLAGS="$MPI_CFLAGS $AM_CFLAGS"
AM_LDFLAGS="`echo $MPI_CXXLDFLAGS | sed -E 's/-l@<:@^ @:>@+//g'` $AM_LDFLAGS"
LIBS="`echo $MPI_CXXLDFLAGS | sed -E 's/-L@<:@^ @:>@+//g'` $LIBS";;
*)
;;
esac
AM_CONDITIONAL(BUILD_COMMS_SHMEM,[ test "X${ac_COMMS}X" == "XshmemX" ]) AM_CONDITIONAL(BUILD_COMMS_SHMEM,[ test "X${ac_COMMS}X" == "XshmemX" ])
AM_CONDITIONAL(BUILD_COMMS_MPI,[ test "X${ac_COMMS}X" == "XmpiX" || test "X${ac_COMMS}X" == "Xmpi-autoX" ]) AM_CONDITIONAL(BUILD_COMMS_MPI,
AM_CONDITIONAL(BUILD_COMMS_MPI3,[ test "X${ac_COMMS}X" == "Xmpi3X"] ) [ test "X${ac_COMMS}X" == "XmpiX" || test "X${ac_COMMS}X" == "Xmpi-autoX" ])
AM_CONDITIONAL(BUILD_COMMS_MPI3,
[ test "X${ac_COMMS}X" == "Xmpi3X" || test "X${ac_COMMS}X" == "Xmpi3-autoX" ])
AM_CONDITIONAL(BUILD_COMMS_NONE,[ test "X${ac_COMMS}X" == "XnoneX" ]) AM_CONDITIONAL(BUILD_COMMS_NONE,[ test "X${ac_COMMS}X" == "XnoneX" ])
############### RNG selection ############### RNG selection
@ -276,10 +303,11 @@ case ${ac_RNG} in
;; ;;
esac esac
############### timer option ############### Timer option
AC_ARG_ENABLE([timers],[AC_HELP_STRING([--enable-timers],\ AC_ARG_ENABLE([timers],[AC_HELP_STRING([--enable-timers],\
[Enable system dependent high res timers])],\ [Enable system dependent high res timers])],\
[ac_TIMERS=${enable_timers}],[ac_TIMERS=yes]) [ac_TIMERS=${enable_timers}],[ac_TIMERS=yes])
case ${ac_TIMERS} in case ${ac_TIMERS} in
yes) yes)
AC_DEFINE([TIMERS_ON],[1],[TIMERS_ON] ) AC_DEFINE([TIMERS_ON],[1],[TIMERS_ON] )
@ -293,7 +321,9 @@ case ${ac_TIMERS} in
esac esac
############### Chroma regression test ############### Chroma regression test
AC_ARG_ENABLE([chroma],[AC_HELP_STRING([--enable-chroma],[Expect chroma compiled under c++11 ])],ac_CHROMA=yes,ac_CHROMA=no) AC_ARG_ENABLE([chroma],[AC_HELP_STRING([--enable-chroma],
[Expect chroma compiled under c++11 ])],ac_CHROMA=yes,ac_CHROMA=no)
case ${ac_CHROMA} in case ${ac_CHROMA} in
yes|no) yes|no)
;; ;;
@ -301,6 +331,7 @@ case ${ac_CHROMA} in
AC_MSG_ERROR([${ac_CHROMA} unsupported --enable-chroma option]); AC_MSG_ERROR([${ac_CHROMA} unsupported --enable-chroma option]);
;; ;;
esac esac
AM_CONDITIONAL(BUILD_CHROMA_REGRESSION,[ test "X${ac_CHROMA}X" == "XyesX" ]) AM_CONDITIONAL(BUILD_CHROMA_REGRESSION,[ test "X${ac_CHROMA}X" == "XyesX" ])
############### Doxygen ############### Doxygen
@ -332,35 +363,36 @@ AC_CONFIG_FILES(tests/qdpxx/Makefile)
AC_CONFIG_FILES(benchmarks/Makefile) AC_CONFIG_FILES(benchmarks/Makefile)
AC_OUTPUT AC_OUTPUT
echo " echo "~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Summary of configuration for $PACKAGE v$VERSION Summary of configuration for $PACKAGE v$VERSION
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
----- PLATFORM ---------------------------------------- ----- PLATFORM ----------------------------------------
- architecture (build) : $build_cpu architecture (build) : $build_cpu
- os (build) : $build_os os (build) : $build_os
- architecture (target) : $target_cpu architecture (target) : $target_cpu
- os (target) : $target_os os (target) : $target_os
- compiler vendor : ${ax_cv_cxx_compiler_vendor} compiler vendor : ${ax_cv_cxx_compiler_vendor}
- compiler version : ${ax_cv_gxx_version} compiler version : ${ax_cv_gxx_version}
----- BUILD OPTIONS ----------------------------------- ----- BUILD OPTIONS -----------------------------------
- SIMD : ${ac_SIMD} SIMD : ${ac_SIMD}
- Threading : ${ac_openmp} Threading : ${ac_openmp}
- Communications type : ${ac_COMMS} Communications type : ${ac_COMMS}
- Default precision : ${ac_PRECISION} Default precision : ${ac_PRECISION}
- RNG choice : ${ac_RNG} RNG choice : ${ac_RNG}
- GMP : `if test "x$have_gmp" = xtrue; then echo yes; else echo no; fi` GMP : `if test "x$have_gmp" = xtrue; then echo yes; else echo no; fi`
- LAPACK : ${ac_LAPACK} LAPACK : ${ac_LAPACK}
- FFTW : `if test "x$have_fftw" = xtrue; then echo yes; else echo no; fi` FFTW : `if test "x$have_fftw" = xtrue; then echo yes; else echo no; fi`
- build DOXYGEN documentation : `if test "x$enable_doc" = xyes; then echo yes; else echo no; fi` build DOXYGEN documentation : `if test "x$enable_doc" = xyes; then echo yes; else echo no; fi`
- graphs and diagrams : `if test "x$enable_dot" = xyes; then echo yes; else echo no; fi` graphs and diagrams : `if test "x$enable_dot" = xyes; then echo yes; else echo no; fi`
----- BUILD FLAGS ------------------------------------- ----- BUILD FLAGS -------------------------------------
- CXXFLAGS: CXXFLAGS:
`echo ${AM_CXXFLAGS} ${CXXFLAGS} | tr ' ' '\n' | sed 's/^-/ -/g'` `echo ${AM_CXXFLAGS} ${CXXFLAGS} | tr ' ' '\n' | sed 's/^-/ -/g'`
- LDFLAGS: LDFLAGS:
`echo ${AM_LDFLAGS} ${LDFLAGS} | tr ' ' '\n' | sed 's/^-/ -/g'` `echo ${AM_LDFLAGS} ${LDFLAGS} | tr ' ' '\n' | sed 's/^-/ -/g'`
- LIBS: LIBS:
`echo ${LIBS} | tr ' ' '\n' | sed 's/^-/ -/g'` `echo ${LIBS} | tr ' ' '\n' | sed 's/^-/ -/g'`
------------------------------------------------------- -------------------------------------------------------" > config.summary
" echo ""
cat config.summary
echo ""

2
lib/FFT.h
View File

@ -30,7 +30,7 @@ Author: Peter Boyle <paboyle@ph.ed.ac.uk>
#define _GRID_FFT_H_ #define _GRID_FFT_H_
#ifdef HAVE_FFTW #ifdef HAVE_FFTW
#include <fftw3.h> #include <Grid/fftw/fftw3.h>
#endif #endif

6
lib/algorithms/iterative/ImplicitlyRestartedLanczos.h
View File

@ -31,7 +31,11 @@ Author: paboyle <paboyle@ph.ed.ac.uk>
#include <string.h> //memset #include <string.h> //memset
#ifdef USE_LAPACK #ifdef USE_LAPACK
#include <lapacke.h> void LAPACK_dstegr(char *jobz, char *range, int *n, double *d, double *e,
double *vl, double *vu, int *il, int *iu, double *abstol,
int *m, double *w, double *z, int *ldz, int *isuppz,
double *work, int *lwork, int *iwork, int *liwork,
int *info);
#endif #endif
#include "DenseMatrix.h" #include "DenseMatrix.h"
#include "EigenSort.h" #include "EigenSort.h"

5
lib/simd/Grid_avx512.h
View File

@ -382,14 +382,9 @@ namespace Optimization {
// Some Template specialization // Some Template specialization
// Hack for CLANG until mm512_reduce_add_ps etc... are implemented in GCC and Clang releases // Hack for CLANG until mm512_reduce_add_ps etc... are implemented in GCC and Clang releases
<<<<<<< HEAD
#define GNU_CLANG_COMPILER
#ifdef GNU_CLANG_COMPILER
=======
#ifndef __INTEL_COMPILER #ifndef __INTEL_COMPILER
#warning "Slow reduction due to incomplete reduce intrinsics" #warning "Slow reduction due to incomplete reduce intrinsics"
>>>>>>> develop
//Complex float Reduce //Complex float Reduce
template<> template<>
inline Grid::ComplexF Reduce<Grid::ComplexF, __m512>::operator()(__m512 in){ inline Grid::ComplexF Reduce<Grid::ComplexF, __m512>::operator()(__m512 in){