将jBLAS与NVBLAS配合使用

Question

由于配置脚本没有正确地找到这些库，我使用了一个有点老套的解决方案，用NVBLAS编译了jBLAS。我像这样手动编辑了jBLAS的configure.out文件，以包含NVBLAS库。

BUILD_TYPE=nvblas
CC=gcc
CCC=c99
CFLAGS=-fPIC -DHAS_CPUID
F77=gfortran
FOUND_JAVA=true
FOUND_NM=true
INCDIRS=-Iinclude -I/usr/lib/jvm/java-11-openjdk-amd64/include -I/usr/lib/jvm/java-11-openjdk-amd64/include/linux
JAVA_HOME=/usr/lib/jvm/java-11-openjdk-amd64
LAPACK_HOME=./lapack-lite-3.1.1
LD=gcc
LDFLAGS=-shared
LIB=lib
LINKAGE_TYPE=static
LOADLIBES=-Wl,-z,muldefs /home/linyi/jblas/lapack-lite-3.1.1/lapack_LINUX.a /usr/local/cuda-11.0/lib64/libnvblas.so.11 /home/linyi/jblas/lapack-lite-3.1.1/blas_LINUX.a -lgfortran
MAKE=make
NM=nm
OS_ARCH=amd64
OS_ARCH_WITH_FLAVOR=amd64/sse3
OS_NAME=Linux
RUBY=ruby
SO=so

然后，我将命令make clean all和mvn clean package作为文档记录的here运行。测试成功通过，但程序在退出时导致分段错误。

-------------------------------------------------------
 T E S T S
-------------------------------------------------------
Running org.jblas.TestEigen
[NVBLAS] NVBLAS_CONFIG_FILE environment variable is set to '/home/linyi/nvblas.conf'
Tests run: 2, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.569 sec
Running org.jblas.TestComplexFloat
Tests run: 5, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.001 sec
Running org.jblas.TestDecompose
Tests run: 7, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.004 sec
Running org.jblas.TestBlasDouble
Tests run: 8, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.003 sec
Running org.jblas.TestBlasDoubleComplex
Tests run: 3, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.001 sec
Running org.jblas.TestSingular
Tests run: 2, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.004 sec
Running org.jblas.TestDoubleMatrix
Tests run: 37, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.022 sec
Running org.jblas.TestSolve
Tests run: 5, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.001 sec
Running org.jblas.TestBlasFloat
Tests run: 8, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.001 sec
Running org.jblas.TestFloatMatrix
Tests run: 37, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.012 sec
Running org.jblas.SimpleBlasTest
Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0 sec
Running org.jblas.ranges.RangeTest
Tests run: 4, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.002 sec
Running org.jblas.TestGeometry
Tests run: 2, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.001 sec
Running org.jblas.ComplexDoubleMatrixTest
Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0 sec
-- org.jblas INFO Deleting /tmp/jblas4383455253907276334/libjblas.so
-- org.jblas INFO Deleting /tmp/jblas4383455253907276334/libjblas_arch_flavor.so
-- org.jblas INFO Deleting /tmp/jblas4383455253907276334
#
# A fatal error has been detected by the Java Runtime Environment:
#
#  SIGSEGV (0xb) at pc=0x00007fa1f6bb96b1, pid=8063, tid=8072
#
# JRE version: OpenJDK Runtime Environment (11.0.8+10) (build 11.0.8+10-post-Ubuntu-0ubuntu118.04.1)
# Java VM: OpenJDK 64-Bit Server VM (11.0.8+10-post-Ubuntu-0ubuntu118.04.1, mixed mode, sharing, tiered, compressed oops, g1 gc, linux-amd64)
# Problematic frame:
# C  [libcublas.so.11+0xa096b1]
#
# Core dump will be written. Default location: Core dumps may be processed with "/usr/share/apport/apport %p %s %c %d %P %E" (or dumping to /home/linyi/jblas/jblas/core.8063)
#
# An error report file with more information is saved as:
# /home/linyi/jblas/jblas/hs_err_pid8063.log
#
# If you would like to submit a bug report, please visit:
#   https://bugs.launchpad.net/ubuntu/+source/openjdk-lts
#
Aborted (core dumped)

Results :

Tests run: 122, Failures: 0, Errors: 0, Skipped: 0

我决定运行mvn clean package -DskipTests，因为看起来测试通过得很好，只是程序在终止时导致了分段错误。然而，当我在我的Java项目中使用这个库时，nvblas.log透露，尽管NVBLAS截获了对BLAS例程的调用，但它们实际上是在中央处理器而不是图形处理器上执行的。在我的程序中运行nvprof --print-gpu-summary也会得到同样的结果。

#
==7711== Profiling result:
            Type  Time(%)      Time     Calls       Avg       Min       Max  Name
 GPU activities:  100.00%  1.8240us         1  1.8240us  1.8240us  1.8240us  [CUDA memcpy HtoD]
======== Error: Application received signal 134

nvblas.log的内容如下：

[NVBLAS] Using devices :0
[NVBLAS] Config parsed
[NVBLAS] dgemm[cpu]: ta=N, tb=N, m=1, n=1, k=1
[NVBLAS] dsyr2k[cpu]: up=U, ta=N, n=24, k=28
[NVBLAS] dsyr2k[cpu]: up=U, ta=N, n=32, k=28
[NVBLAS] dsyr2k[cpu]: up=U, ta=N, n=26, k=28
[NVBLAS] dsyr2k[cpu]: up=U, ta=N, n=22, k=28
[NVBLAS] dsyr2k[cpu]: up=U, ta=N, n=20, k=28
[NVBLAS] dsyr2k[cpu]: up=U, ta=N, n=26, k=28
[NVBLAS] dtrmm[cpu]: si=R, up=U, ta=N, di=U, m=52, n=31
[NVBLAS] dtrmm[cpu]: si=R, up=U, ta=N, di=U, m=54, n=31
[NVBLAS] dtrmm[cpu]: si=R, up=L, ta=T, di=N, m=52, n=31
[NVBLAS] dtrmm[cpu]: si=R, up=L, ta=T, di=N, m=54, n=31
[NVBLAS] dtrmm[cpu]: si=R, up=U, ta=N, di=U, m=60, n=31
[NVBLAS] dtrmm[cpu]: si=R, up=U, ta=T, di=U, m=54, n=31
[NVBLAS] dtrmm[cpu]: si=R, up=U, ta=T, di=U, m=52, n=31
[NVBLAS] dtrmm[cpu]: si=R, up=L, ta=T, di=N, m=60, n=31
[NVBLAS] dsyr2k[cpu]: up=U, ta=N, n=22, k=28
[NVBLAS] dtrmm[cpu]: si=R, up=U, ta=T, di=U, m=60, n=31
[NVBLAS] dtrmm[cpu]: si=R, up=U, ta=N, di=U, m=54, n=22
[NVBLAS] dgemm[cpu]: ta=T, tb=N, m=54, n=22, k=31
[NVBLAS] dtrmm[cpu]: si=R, up=U, ta=N, di=U, m=60, n=28
[NVBLAS] dtrmm[cpu]: si=R, up=U, ta=N, di=U, m=52, n=20
[NVBLAS] dtrmm[cpu]: si=R, up=L, ta=T, di=N, m=54, n=22
[NVBLAS] dgemm[cpu]: ta=T, tb=N, m=60, n=28, k=31
[NVBLAS] dgemm[cpu]: ta=T, tb=N, m=52, n=20, k=31
[NVBLAS] dgemm[cpu]: ta=N, tb=T, m=31, n=54, k=22
. . .

我真的不知所措，我希望有人能给我一些建议，这似乎真的很少有记录。

Answer 1

我相信我已经弄明白了。

nm libjblas.so | grep -is dgemm
         U cblas_dgemm
         U dgemm_@@libnvblas.so.11

这表明库确实是正确链接的。然后我继续运行jBlas的内置基准测试，只运行java -jar jblas.jar，其中jblas.jar是编译库，显然GPU卸载只发生在大型矩阵上，因为当n=10或n=100时没有进行gpu调用，但nvblas.log在n=1000记录了GPU计算。这对我来说是非常困惑的，我希望这对其他正在与这个问题作斗争的人有所帮助。