无法在Ubuntu 10.04，CUDA 5.0上编译MAGMA 1.3

Question

我试图在Ubuntu10.04上编译岩浆库，但是我被困住了。

来自自述文件：

若要安装MAGMA，请修改make.inc文件以指示在系统上安装CUDA、CPU BLAS和LAPACK的位置。示例 以make.inc.mkl、make.inc.acml、make.inc.goto和 make.inc.atlas，演示如何相应地链接到MKL，ACML， 去布拉斯或阿拉斯布拉斯。经过适当的修改后的make.inc 键入'make‘的文件将创建目录'lib’中的libmagma.a和libmagmablas.a库)，以测试目录' testing‘中的驱动程序。

因此，我安装了所需的软件包：

apt-get install libatlas-base-dev liblapack-dev libf2c2-dev gfortran

将make.inc.atlas复制到修改CUDA路径、nvcc路径和liblapack.a路径的make.inc，然后删除两个不必要的链接目录。

为了提供完整的参考，下面是原始的make.inc.atlas

#//////////////////////////////////////////////////////////////////////////////
#   -- MAGMA (version 1.3.0) --
#      Univ. of Tennessee, Knoxville
#      Univ. of California, Berkeley
#      Univ. of Colorado, Denver
#      November 2012
#//////////////////////////////////////////////////////////////////////////////

#
# GPU_TARGET specifies for which GPU you want to compile MAGMA:
#     "Tesla" (NVIDIA compute capability 1.x cards)
#     "Fermi" (NVIDIA compute capability 2.x cards)
# See http://developer.nvidia.com/cuda-gpus

GPU_TARGET = Fermi

CC        = gcc
NVCC      = nvcc
FORT      = gfortran

ARCH      = ar
ARCHFLAGS = cr
RANLIB    = ranlib

OPTS      = -O3 -DADD_
F77OPTS   = -O3 -DADD_
FOPTS     = -O3 -DADD_ -x f95-cpp-input
NVOPTS    = -O3 -DADD_ --compiler-options -fno-strict-aliasing -DUNIX
LDOPTS    = -fPIC -Xlinker -zmuldefs

LIB       = /home/tomov/LIB/lapack-lite-3.1.1/lapack_LINUX64.a \
            -lf77blas -latlas -lcblas -lf2c -lcublas -lm

CUDADIR   = /usr/local/cuda-3.0

LIBDIR    = -L/home/tomov/LIB/ATLAS/lib/Linux_UNKNOWNSSE2 \
            -L/home/tomov/LIB/CLAPACK-3.1.1/F2CLIBS/ \
            -L$(CUDADIR)/lib64
INC       = -I$(CUDADIR)/include

在这里，我修改的文件保存到make.inc中

#//////////////////////////////////////////////////////////////////////////////
#   -- MAGMA (version 1.3.0) --
#      Univ. of Tennessee, Knoxville
#      Univ. of California, Berkeley
#      Univ. of Colorado, Denver
#      November 2012
#//////////////////////////////////////////////////////////////////////////////

#
# GPU_TARGET specifies for which GPU you want to compile MAGMA:
#     "Tesla" (NVIDIA compute capability 1.x cards)
#     "Fermi" (NVIDIA compute capability 2.x cards)
# See http://developer.nvidia.com/cuda-gpus

GPU_TARGET = Fermi

CC        = gcc
NVCC      = /usr/local/cuda-5.0/bin/nvcc
FORT      = gfortran

ARCH      = ar
ARCHFLAGS = cr
RANLIB    = ranlib

OPTS      = -O3 -DADD_
F77OPTS   = -O3 -DADD_
FOPTS     = -O3 -DADD_ -x f95-cpp-input
NVOPTS    = -O3 -DADD_ --compiler-options -fno-strict-aliasing -DUNIX
LDOPTS    = -fPIC -Xlinker -zmuldefs

LIB       = /usr/lib/liblapack.a \
            -lf77blas -latlas -lcblas -lf2c -lcublas -lm

CUDADIR   = /usr/local/cuda-5.0

LIBDIR    = -L$(CUDADIR)/lib64
INC       = -I$(CUDADIR)/include

所以，我运行make，我得到了这一错误。我在此应用了所提出的解决方案，到目前为止，编译工作还在继续：

make[1]: Entering directory `/usr/src/magma-1.3.0/testing'
gcc -O3 -DADD_ -DHAVE_CUBLAS -DGPUSHMEM=200 -I/usr/local/cuda-5.0/include -I../include -I../control -c /usr/local/cuda-5.0/src/fortran.c -o fortran.o
In file included from /usr/local/cuda-5.0/src/fortran.c:67:
/usr/local/cuda-5.0/src/fortran_common.h:405:2: error: #error unsupported Fortran compiler
make[1]: *** [fortran.o] Error 1
make[1]: Leaving directory `/usr/src/magma-1.3.0/testing'
make: *** [test] Error 2

我能做什么?

Answer 1

简单的解决方案:将堡垒设置为：

FORT      = gfortran -DCUBLAS_GFORTRAN

Answer 2

这似乎是一个数据自动化系统的问题。打开fortran_common.h并定位此条件：

#if defined(CUBLAS_GFORTRAN) || defined (CUBLAS_G95)

应该满足这个条件，AFAIK。但是它没有，所以只需删除该行，并删除所有其他elsif和endif行以及完整路径。文件fortran_common.h应该如下所示：

/*
 * Copyright 1993-2012 NVIDIA Corporation.  All rights reserved.
 *
 * NOTICE TO LICENSEE:
 *
 * This source code and/or documentation ("Licensed Deliverables") are
 * subject to NVIDIA intellectual property rights under U.S. and
 * international Copyright laws.
 *
 * These Licensed Deliverables contained herein is PROPRIETARY and
 * CONFIDENTIAL to NVIDIA and is being provided under the terms and
 * conditions of a form of NVIDIA software license agreement by and
 * between NVIDIA and Licensee ("License Agreement") or electronically
 * accepted by Licensee.  Notwithstanding any terms or conditions to
 * the contrary in the License Agreement, reproduction or disclosure
 * of the Licensed Deliverables to any third party without the express
 * written consent of NVIDIA is prohibited.
 *
 * NOTWITHSTANDING ANY TERMS OR CONDITIONS TO THE CONTRARY IN THE
 * LICENSE AGREEMENT, NVIDIA MAKES NO REPRESENTATION ABOUT THE
 * SUITABILITY OF THESE LICENSED DELIVERABLES FOR ANY PURPOSE.  IT IS
 * PROVIDED "AS IS" WITHOUT EXPRESS OR IMPLIED WARRANTY OF ANY KIND.
 * NVIDIA DISCLAIMS ALL WARRANTIES WITH REGARD TO THESE LICENSED
 * DELIVERABLES, INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY,
 * NONINFRINGEMENT, AND FITNESS FOR A PARTICULAR PURPOSE.
 * NOTWITHSTANDING ANY TERMS OR CONDITIONS TO THE CONTRARY IN THE
 * LICENSE AGREEMENT, IN NO EVENT SHALL NVIDIA BE LIABLE FOR ANY
 * SPECIAL, INDIRECT, INCIDENTAL, OR CONSEQUENTIAL DAMAGES, OR ANY
 * DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS,
 * WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS
 * ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE
 * OF THESE LICENSED DELIVERABLES.
 *
 * U.S. Government End Users.  These Licensed Deliverables are a
 * "commercial item" as that term is defined at 48 C.F.R. 2.101 (OCT
 * 1995), consisting of "commercial computer software" and "commercial
 * computer software documentation" as such terms are used in 48
 * C.F.R. 12.212 (SEPT 1995) and is provided to the U.S. Government
 * only as a commercial end item.  Consistent with 48 C.F.R.12.212 and
 * 48 C.F.R. 227.7202-1 through 227.7202-4 (JUNE 1995), all
 * U.S. Government End Users acquire the Licensed Deliverables with
 * only those rights set forth herein.
 *
 * Any use of the Licensed Deliverables in individual and commercial
 * software must include, in the user documentation and internal
 * comments to the code, the above Disclaimer and U.S. Government End
 * Users Notice.
 */



/* Define CUBLAS_FORTRAN_COMPILER for Windows needed because 
   tbe compilation is done from the perl script fortran_nightly.pl which
   does not include fortran_tools.mk
   On Linux and Apple, CFLAGS is setup with fortran_tools.mk
*/
#if defined(_WIN32)
#ifndef CUBLAS_INTEL_FORTRAN
#define CUBLAS_INTEL_FORTRAN
#endif
#endif

#if defined(CUBLAS_GFORTRAN)
/* using option -ff2c make the ABI compatible with F77
   No need to define RETURN_COMPLEX, which cause problem 
   on Gfortran 4.x on 32 bit
*/
/* #define RETURN_COMPLEX 1 */
#endif


/* NOTE: Must use -fno-second-underscore when building Fortran source with g77
 *       g77 invocation may not use -fno-f2c, which forces different return 
 *       type conventions than the one used below
 */
#define CUBLAS_INIT             cublas_init_
#define CUBLAS_SHUTDOWN         cublas_shutdown_
#define CUBLAS_ALLOC            cublas_alloc_
#define CUBLAS_FREE             cublas_free_
#define CUBLAS_SET_VECTOR       cublas_set_vector_
#define CUBLAS_GET_VECTOR       cublas_get_vector_
#define CUBLAS_SET_MATRIX       cublas_set_matrix_
#define CUBLAS_GET_MATRIX       cublas_get_matrix_
#define CUBLAS_GET_ERROR        cublas_get_error_
#define CUBLAS_XERBLA           cublas_xerbla_
#define CUBLAS_ISAMAX           cublas_isamax_
#define CUBLAS_ISAMIN           cublas_isamin_
#define CUBLAS_SASUM            cublas_sasum_
#define CUBLAS_SAXPY            cublas_saxpy_
#define CUBLAS_SCOPY            cublas_scopy_
#define CUBLAS_SDOT             cublas_sdot_
#define CUBLAS_SNRM2            cublas_snrm2_
#define CUBLAS_SROT             cublas_srot_
#define CUBLAS_SROTG            cublas_srotg_
#define CUBLAS_SROTM            cublas_srotm_
#define CUBLAS_SROTMG           cublas_srotmg_
#define CUBLAS_SSCAL            cublas_sscal_
#define CUBLAS_SSWAP            cublas_sswap_
#define CUBLAS_CAXPY            cublas_caxpy_
#define CUBLAS_CCOPY            cublas_ccopy_
#define CUBLAS_CROT             cublas_crot_
#define CUBLAS_CROTG            cublas_crotg_
#define CUBLAS_CSCAL            cublas_cscal_
#define CUBLAS_CSROT            cublas_csrot_
#define CUBLAS_CSSCAL           cublas_csscal_
#define CUBLAS_CSWAP            cublas_cswap_
#define CUBLAS_CTRMV            cublas_ctrmv_
#define CUBLAS_CDOTU            cublas_cdotu_
#define CUBLAS_CDOTC            cublas_cdotc_
#define CUBLAS_ICAMAX           cublas_icamax_
#define CUBLAS_SCASUM           cublas_scasum_
#define CUBLAS_SCNRM2           cublas_scnrm2_
#define CUBLAS_SGBMV            cublas_sgbmv_
#define CUBLAS_SGEMV            cublas_sgemv_
#define CUBLAS_SGER             cublas_sger_
#define CUBLAS_SSBMV            cublas_ssbmv_
#define CUBLAS_SSPMV            cublas_sspmv_
#define CUBLAS_SSPR             cublas_sspr_
#define CUBLAS_SSPR2            cublas_sspr2_
#define CUBLAS_SSYMV            cublas_ssymv_
#define CUBLAS_SSYR             cublas_ssyr_
#define CUBLAS_SSYR2            cublas_ssyr2_
#define CUBLAS_STBMV            cublas_stbmv_
#define CUBLAS_STBSV            cublas_stbsv_
#define CUBLAS_STPMV            cublas_stpmv_
#define CUBLAS_STPSV            cublas_stpsv_
#define CUBLAS_STRMV            cublas_strmv_
#define CUBLAS_STRSV            cublas_strsv_
#define CUBLAS_SGEMM            cublas_sgemm_
#define CUBLAS_SSYMM            cublas_ssymm_
#define CUBLAS_SSYR2K           cublas_ssyr2k_
#define CUBLAS_SSYRK            cublas_ssyrk_
#define CUBLAS_STRMM            cublas_strmm_
#define CUBLAS_STRSM            cublas_strsm_
#define CUBLAS_CGEMM            cublas_cgemm_
#define CUBLAS_CHEMM            cublas_chemm_
#define CUBLAS_CSYMM            cublas_csymm_
#define CUBLAS_CTRMM            cublas_ctrmm_
#define CUBLAS_CTRSM            cublas_ctrsm_
#define CUBLAS_CHERK            cublas_cherk_
#define CUBLAS_CSYRK            cublas_csyrk_
#define CUBLAS_CHER2K           cublas_cher2k_
#define CUBLAS_CSYR2K           cublas_csyr2k_
#define CUBLAS_IDAMAX           cublas_idamax_
#define CUBLAS_IDAMIN           cublas_idamin_
#define CUBLAS_DASUM            cublas_dasum_
#define CUBLAS_DAXPY            cublas_daxpy_
#define CUBLAS_DCOPY            cublas_dcopy_
#define CUBLAS_DDOT             cublas_ddot_
#define CUBLAS_DNRM2            cublas_dnrm2_
#define CUBLAS_DROT             cublas_drot_
#define CUBLAS_DROTG            cublas_drotg_
#define CUBLAS_DROTM            cublas_drotm_
#define CUBLAS_DROTMG           cublas_drotmg_
#define CUBLAS_DSCAL            cublas_dscal_
#define CUBLAS_DSWAP            cublas_dswap_
#define CUBLAS_ZAXPY            cublas_zaxpy_
#define CUBLAS_ZCOPY            cublas_zcopy_
#define CUBLAS_ZROT             cublas_zrot_
#define CUBLAS_ZROTG            cublas_zrotg_
#define CUBLAS_ZSCAL            cublas_zscal_
#define CUBLAS_ZDROT            cublas_zdrot_
#define CUBLAS_ZDSCAL           cublas_zdscal_
#define CUBLAS_ZSWAP            cublas_zswap_
#define CUBLAS_ZDOTU            cublas_zdotu_
#define CUBLAS_ZDOTC            cublas_zdotc_
#define CUBLAS_IZAMAX           cublas_izamax_
#define CUBLAS_DZASUM           cublas_dzasum_
#define CUBLAS_DZNRM2           cublas_dznrm2_
#define CUBLAS_DGBMV            cublas_dgbmv_
#define CUBLAS_DGEMV            cublas_dgemv_
#define CUBLAS_ZGEMV            cublas_zgemv_
#define CUBLAS_DGER             cublas_dger_
#define CUBLAS_DSBMV            cublas_dsbmv_
#define CUBLAS_DSPMV            cublas_dspmv_
#define CUBLAS_DSPR             cublas_dspr_
#define CUBLAS_DSPR2            cublas_dspr2_
#define CUBLAS_DSYMV            cublas_dsymv_
#define CUBLAS_DSYR             cublas_dsyr_
#define CUBLAS_DSYR2            cublas_dsyr2_
#define CUBLAS_DTBMV            cublas_dtbmv_
#define CUBLAS_DTBSV            cublas_dtbsv_
#define CUBLAS_DTPMV            cublas_dtpmv_
#define CUBLAS_DTPSV            cublas_dtpsv_
#define CUBLAS_DTRMV            cublas_dtrmv_
#define CUBLAS_DTRSV            cublas_dtrsv_
#define CUBLAS_DGEMM            cublas_dgemm_
#define CUBLAS_DSYMM            cublas_dsymm_
#define CUBLAS_DSYR2K           cublas_dsyr2k_
#define CUBLAS_DSYRK            cublas_dsyrk_
#define CUBLAS_ZSYRK            cublas_zsyrk_
#define CUBLAS_DTRMM            cublas_dtrmm_
#define CUBLAS_DTRSM            cublas_dtrsm_
#define CUBLAS_ZGEMM            cublas_zgemm_
#define CUBLAS_ZHEMM            cublas_zhemm_
#define CUBLAS_ZSYMM            cublas_zsymm_
#define CUBLAS_ZTRMM            cublas_ztrmm_
#define CUBLAS_ZTRSM            cublas_ztrsm_
#define CUBLAS_ZHERK            cublas_zherk_
#define CUBLAS_ZSYRK            cublas_zsyrk_
#define CUBLAS_ZHER2K           cublas_zher2k_
#define CUBLAS_ZSYR2K           cublas_zsyr2k_

#define  CUBLAS_CGEMV           cublas_cgemv_
#define  CUBLAS_CGBMV           cublas_cgbmv_
#define  CUBLAS_CHEMV           cublas_chemv_
#define  CUBLAS_CHBMV           cublas_chbmv_
#define  CUBLAS_CHPMV           cublas_chpmv_
#define  CUBLAS_CTBMV           cublas_ctbmv_
#define  CUBLAS_CTPMV           cublas_ctpmv_
#define  CUBLAS_CTRSV           cublas_ctrsv_
#define  CUBLAS_CTBSV           cublas_ctbsv_
#define  CUBLAS_CTPSV           cublas_ctpsv_
#define  CUBLAS_CGERC           cublas_cgerc_
#define  CUBLAS_CGERU           cublas_cgeru_
#define  CUBLAS_CHPR            cublas_chpr_
#define  CUBLAS_CHPR2           cublas_chpr2_
#define  CUBLAS_CHER            cublas_cher_
#define  CUBLAS_CHER2           cublas_cher2_

// stubs for zblat2
#define CUBLAS_ZGBMV           cublas_zgbmv_
#define CUBLAS_ZHEMV           cublas_zhemv_
#define CUBLAS_ZHBMV           cublas_zhbmv_
#define CUBLAS_ZHPMV           cublas_zhpmv_
#define CUBLAS_ZTRMV           cublas_ztrmv_
#define CUBLAS_ZTBMV           cublas_ztbmv_
#define CUBLAS_ZTPMV           cublas_ztpmv_
#define CUBLAS_ZTRSV           cublas_ztrsv_
#define CUBLAS_ZTBSV           cublas_ztbsv_
#define CUBLAS_ZTPSV           cublas_ztpsv_
#define CUBLAS_ZGERC           cublas_zgerc_
#define CUBLAS_ZGERU           cublas_zgeru_
#define CUBLAS_ZHER            cublas_zher_
#define CUBLAS_ZHPR            cublas_zhpr_
#define CUBLAS_ZHER2           cublas_zher2_
#define CUBLAS_ZHPR2           cublas_zhpr2_

在这一变化之后，岩浆就可以编译了。