apache / mxnet

Lightweight, Portable, Flexible Distributed/Mobile Deep Learning with Dynamic, Mutation-aware Dataflow Dep Scheduler; for Python, R, Julia, Scala, Go, Javascript and more
https://mxnet.apache.org
Apache License 2.0
20.78k stars 6.79k forks source link

Building mxnet from source with openblas causes issues with mshadow #20542

Closed Data-drone closed 3 years ago

Data-drone commented 3 years ago

Description

I am trying to build mxnet source with the version: git checkout tags/1.9.0.rc6 -b 1.9.x

I am getting No Such file or directory errors when I set it to use openblas rather than mkl this system has both openblas and mkl installed. This happens at the mshadow compile. this error goes away. Is there some error with the way that mshadow cmake is set so that it isn't compiling against openblas instead?

Error Message

When compiling I am using the command:

mkdir /opt/mxnet/cmake_binary && \
    cd /opt/mxnet/cmake_binary \
  && cmake -DCMAKE_BUILD_TYPE=Release \
    -DCMAKE_INSTALL_PREFIX=$(python -c "import sys; print(sys.prefix)") \
    -DCMAKE_FIND_ROOT_PATH=${CONDA_PREFIX} \
    -DCMAKE_FIND_ROOT_PATH_MODE_LIBRARY=BOTH \
    -DUSE_OPENCV=ON \
    -DUSE_MKLDNN=ON \
    -DUSE_LAPACK=ON \
    -DUSE_ONEDNN=ON \
    -DUSE_MKL_IF_AVAILABLE=OFF \
    -DUSE_BLAS=mkl \
    -DUSE_INTEL_PATH=${CONDA_PREFIX}/lib \
    -DUSE_CUDA=ON \
    -DUSE_CUDNN=ON \
    -DUSE_OPENMP=ON \
    -DUSE_TVM_OP=OFF \
    -DMXNET_CUDA_ARCH="7.5" \
    -DUSE_NCCL=ON \
    -DUSE_SSE=ON \
    .. \
  && cmake --build . --parallel 28

When I set -DUSE_BLAS=openblas then I get an error with:

/opt/mxnet/include/mshadow/./base.h:192:12: fatal error: mkl_blas.h: No such file or directory
  192 |   #include <mkl_blas.h>
      |            ^~~~~~~~~~~~
compilation terminated.
make[2]: *** [CMakeFiles/mxnet_static.dir/build.make:398: CMakeFiles/mxnet_static.dir/src/engine/threaded_engine_pooled.cc.o] Error 1
make[2]: *** Waiting for unfinished jobs....
In file included from /opt/mxnet/include/mshadow/tensor.h:35,
                 from /opt/mxnet/include/mxnet/base.h:33,
                 from /opt/mxnet/src/base.cc:25:
/opt/mxnet/include/mshadow/./base.h:192:12: fatal error: mkl_blas.h: No such file or directory
  192 |   #include <mkl_blas.h>
      |            ^~~~~~~~~~~~
compilation terminated.
make[2]: *** [CMakeFiles/mxnet_static.dir/build.make:132: CMakeFiles/mxnet_static.dir/src/base.cc.o] Error 1
In file included from /opt/mxnet/include/mshadow/tensor.h:35,
                 from /opt/mxnet/include/mxnet/./base.h:33,
                 from /opt/mxnet/include/mxnet/resource.h:29,
                 from /opt/mxnet/src/executor/attach_op_resource_pass.cc:26:
/opt/mxnet/include/mshadow/./base.h:192:12: fatal error: mkl_blas.h: No such file or directory
  192 |   #include <mkl_blas.h>
      |            ^~~~~~~~~~~~
compilation terminated.
make[2]: *** [CMakeFiles/mxnet_static.dir/build.make:426: CMakeFiles/mxnet_static.dir/src/executor/attach_op_resource_pass.cc.o] Error 1
In file included from /opt/mxnet/include/mshadow/tensor.h:35,
                 from /opt/mxnet/include/mxnet/base.h:33,
                 from /opt/mxnet/src/c_api/c_api_ndarray.cc:26:
/opt/mxnet/include/mshadow/./base.h:192:12: fatal error: mkl_blas.h: No such file or directory
  192 |   #include <mkl_blas.h>
      |            ^~~~~~~~~~~~
compilation terminated.
make[2]: *** [CMakeFiles/mxnet_static.dir/build.make:202: CMakeFiles/mxnet_static.dir/src/c_api/c_api_ndarray.cc.o] Error 1
In file included from /opt/mxnet/include/mshadow/tensor.h:35,
                 from /opt/mxnet/include/mxnet/base.h:33,
                 from /opt/mxnet/src/c_api/./c_api_common.h:33,
                 from /opt/mxnet/src/c_api/c_api_error.cc:26:
/opt/mxnet/include/mshadow/./base.h:192:12: fatal error: mkl_blas.h: No such file or directory
  192 |   #include <mkl_blas.h>
      |            ^~~~~~~~~~~~
compilation terminated.
make[2]: *** [CMakeFiles/mxnet_static.dir/build.make:160: CMakeFiles/mxnet_static.dir/src/c_api/c_api_error.cc.o] Error 1
In file included from /opt/mxnet/include/mshadow/tensor.h:35,
                 from /opt/mxnet/include/mxnet/base.h:33,
                 from /opt/mxnet/src/common/cuda_utils.cc:26:
/opt/mxnet/include/mshadow/./base.h:192:12: fatal error: mkl_blas.h: No such file or directory
  192 |   #include <mkl_blas.h>
      |            ^~~~~~~~~~~~
compilation terminated.
In file included from /opt/mxnet/include/mshadow/tensor.h:35,
                 from /opt/mxnet/include/mxnet/./base.h:33,
                 from /opt/mxnet/include/mxnet/rtc.h:22,
                 from /opt/mxnet/src/common/rtc.cc:20:
/opt/mxnet/include/mshadow/./base.h:192:12: fatal error: mkl_blas.h: No such file or directory
  192 |   #include <mkl_blas.h>
      |            ^~~~~~~~~~~~
compilation terminated.
In file included from /opt/mxnet/include/mshadow/tensor.h:35,
                 from /opt/mxnet/include/mxnet/base.h:33,
                 from /opt/mxnet/src/c_api/./c_api_common.h:33,
                 from /opt/mxnet/src/c_api/c_api_test.cc:27:
/opt/mxnet/include/mshadow/./base.h:192:12: fatal error: mkl_blas.h: No such file or directory
  192 |   #include <mkl_blas.h>
      |            ^~~~~~~~~~~~
compilation terminated.
In file included from /opt/mxnet/include/mshadow/tensor.h:35,
                 from /opt/mxnet/include/mxnet/base.h:33,
                 from /opt/mxnet/src/c_api/c_api_symbolic.cc:25:
/opt/mxnet/include/mshadow/./base.h:192:12: fatal error: mkl_blas.h: No such file or directory
  192 |   #include <mkl_blas.h>
      |            ^~~~~~~~~~~~
In file included from /opt/mxnet/include/mshadow/tensor.h:35,
                 from /opt/mxnet/include/mxnet/base.h:33,
                 from /opt/mxnet/src/c_api/c_api_executor.cc:25:
/opt/mxnet/include/mshadow/./base.h:192:12: fatal error: mkl_blas.h: No such file or directory
  192 |   #include <mkl_blas.h>
      |            ^~~~~~~~~~~~
compilation terminated.
compilation terminated.
In file included from /opt/mxnet/include/mshadow/tensor.h:35,
                 from /opt/mxnet/include/mxnet/./base.h:33,
                 from /opt/mxnet/include/mxnet/storage.h:29,
                 from /opt/mxnet/src/engine/./threaded_engine.h:33,
                 from /opt/mxnet/src/engine/threaded_engine.cc:32:
/opt/mxnet/include/mshadow/./base.h:192:12: fatal error: mkl_blas.h: No such file or directory
  192 |   #include <mkl_blas.h>
      |            ^~~~~~~~~~~~
make[2]: *** [CMakeFiles/mxnet_static.dir/build.make:272: CMakeFiles/mxnet_static.dir/src/common/cuda_utils.cc.o] Error 1
compilation terminated.
make[2]: *** [CMakeFiles/mxnet_static.dir/build.make:300: CMakeFiles/mxnet_static.dir/src/common/rtc.cc.o] Error 1
make[2]: *** [CMakeFiles/mxnet_static.dir/build.make:244: CMakeFiles/mxnet_static.dir/src/c_api/c_api_test.cc.o] Error 1
make[2]: *** [CMakeFiles/mxnet_static.dir/build.make:174: CMakeFiles/mxnet_static.dir/src/c_api/c_api_executor.cc.o] Error 1
make[2]: *** [CMakeFiles/mxnet_static.dir/build.make:230: CMakeFiles/mxnet_static.dir/src/c_api/c_api_symbolic.cc.o] Error 1
make[2]: *** [CMakeFiles/mxnet_static.dir/build.make:370: CMakeFiles/mxnet_static.dir/src/engine/threaded_engine.cc.o] Error 1
In file included from /opt/mxnet/include/mshadow/tensor.h:35,
                 from /opt/mxnet/include/mxnet/./base.h:33,
                 from /opt/mxnet/include/mxnet/engine.h:34,
                 from /opt/mxnet/src/engine/engine.cc:25:
/opt/mxnet/include/mshadow/./base.h:192:12: fatal error: mkl_blas.h: No such file or directory
  192 |   #include <mkl_blas.h>
      |            ^~~~~~~~~~~~
compilation terminated.
In file included from /opt/mxnet/include/mshadow/tensor.h:35,
                 from /opt/mxnet/include/mxnet/base.h:33,
                 from /opt/mxnet/src/c_api/c_api.cc:39:
/opt/mxnet/include/mshadow/./base.h:192:12: fatal error: mkl_blas.h: No such file or directory
  192 |   #include <mkl_blas.h>
      |            ^~~~~~~~~~~~
compilation terminated.
In file included from /opt/mxnet/include/mshadow/tensor.h:35,
                 from /opt/mxnet/include/mxnet/./base.h:33,
                 from /opt/mxnet/include/mxnet/executor.h:35,
                 from /opt/mxnet/src/c_api/c_predict_api.cc:28:
/opt/mxnet/include/mshadow/./base.h:192:12: fatal error: mkl_blas.h: No such file or directory
  192 |   #include <mkl_blas.h>
      |            ^~~~~~~~~~~~
compilation terminated.
In file included from /opt/mxnet/include/mshadow/tensor.h:35,
                 from /opt/mxnet/include/mxnet/./base.h:33,
                 from /opt/mxnet/include/mxnet/engine.h:34,
                 from /opt/mxnet/src/common/./utils.h:32,
                 from /opt/mxnet/src/common/utils.cc:25:
/opt/mxnet/include/mshadow/./base.h:192:12: fatal error: mkl_blas.h: No such file or directory
  192 |   #include <mkl_blas.h>
      |            ^~~~~~~~~~~~
compilation terminated.
In file included from /opt/mxnet/include/mshadow/tensor.h:35,
                 from /opt/mxnet/include/mxnet/./base.h:33,
                 from /opt/mxnet/include/mxnet/ndarray.h:43,
                 from /opt/mxnet/include/mxnet/runtime/packed_func.h:37,
                 from /opt/mxnet/src/api/operator/utils.h:29,
                 from /opt/mxnet/src/api/operator/utils.cc:24:
/opt/mxnet/include/mshadow/./base.h:192:12: fatal error: mkl_blas.h: No such file or directory
  192 |   #include <mkl_blas.h>
      |            ^~~~~~~~~~~~
compilation terminated.
In file included from /opt/mxnet/include/mshadow/tensor.h:35,
                 from /opt/mxnet/include/mxnet/base.h:33,
                 from /opt/mxnet/src/executor/graph_executor.cc:25:
/opt/mxnet/include/mshadow/./base.h:192:12: fatal error: mkl_blas.h: No such file or directory
  192 |   #include <mkl_blas.h>
      |            ^~~~~~~~~~~~
compilation terminated.
In file included from /opt/mxnet/include/mshadow/tensor.h:35,
                 from /opt/mxnet/include/mxnet/base.h:33,
                 from /opt/mxnet/src/executor/attach_op_execs_pass.cc:25:
/opt/mxnet/include/mshadow/./base.h:192:12: fatal error: mkl_blas.h: No such file or directory
  192 |   #include <mkl_blas.h>
      |            ^~~~~~~~~~~~
compilation terminated.
make[2]: *** [CMakeFiles/mxnet_static.dir/build.make:328: CMakeFiles/mxnet_static.dir/src/engine/engine.cc.o] Error 1
In file included from /opt/mxnet/include/mshadow/tensor.h:35,
                 from /opt/mxnet/include/mxnet/base.h:33,
                 from /opt/mxnet/src/executor/eliminate_common_expr_pass.cc:27:
/opt/mxnet/include/mshadow/./base.h:192:12: fatal error: mkl_blas.h: No such file or directory
  192 |   #include <mkl_blas.h>
      |            ^~~~~~~~~~~~
compilation terminated.
In file included from /opt/mxnet/include/mshadow/tensor.h:35,
                 from /opt/mxnet/include/mxnet/base.h:33,
                 from /opt/mxnet/src/c_api/c_api_function.cc:26:
/opt/mxnet/include/mshadow/./base.h:192:12: fatal error: mkl_blas.h: No such file or directory
  192 |   #include <mkl_blas.h>
      |            ^~~~~~~~~~~~
compilation terminated.
make[2]: *** [CMakeFiles/mxnet_static.dir/build.make:146: CMakeFiles/mxnet_static.dir/src/c_api/c_api.cc.o] Error 1
make[2]: *** [CMakeFiles/mxnet_static.dir/build.make:258: CMakeFiles/mxnet_static.dir/src/c_api/c_predict_api.cc.o] Error 1
make[2]: *** [CMakeFiles/mxnet_static.dir/build.make:314: CMakeFiles/mxnet_static.dir/src/common/utils.cc.o] Error 1
make[2]: *** [CMakeFiles/mxnet_static.dir/build.make:118: CMakeFiles/mxnet_static.dir/src/api/operator/utils.cc.o] Error 1
make[2]: *** [CMakeFiles/mxnet_static.dir/build.make:454: CMakeFiles/mxnet_static.dir/src/executor/graph_executor.cc.o] Error 1
In file included from /opt/mxnet/include/mshadow/tensor.h:35,
                 from /opt/mxnet/include/mxnet/./base.h:33,
                 from /opt/mxnet/include/mxnet/engine.h:34,
                 from /opt/mxnet/src/common/../common/utils.h:32,
                 from /opt/mxnet/src/common/exec_utils.h:33,
                 from /opt/mxnet/src/common/exec_utils.cc:25:
/opt/mxnet/include/mshadow/./base.h:192:12: fatal error: mkl_blas.h: No such file or directory
  192 |   #include <mkl_blas.h>
      |            ^~~~~~~~~~~~
compilation terminated.
make[2]: *** [CMakeFiles/mxnet_static.dir/build.make:412: CMakeFiles/mxnet_static.dir/src/executor/attach_op_execs_pass.cc.o] Error 1
In file included from /opt/mxnet/include/mshadow/tensor.h:35,
                 from /opt/mxnet/include/mxnet/./base.h:33,
                 from /opt/mxnet/include/mxnet/engine.h:34,
                 from /opt/mxnet/src/engine/./engine_impl.h:28,
                 from /opt/mxnet/src/engine/naive_engine.cc:30:
/opt/mxnet/include/mshadow/./base.h:192:12: fatal error: mkl_blas.h: No such file or directory
  192 |   #include <mkl_blas.h>
      |            ^~~~~~~~~~~~
compilation terminated.
make[2]: *** [CMakeFiles/mxnet_static.dir/build.make:188: CMakeFiles/mxnet_static.dir/src/c_api/c_api_function.cc.o] Error 1
make[2]: *** [CMakeFiles/mxnet_static.dir/build.make:440: CMakeFiles/mxnet_static.dir/src/executor/eliminate_common_expr_pass.cc.o] Error 1
In file included from /opt/mxnet/include/mshadow/tensor.h:35,
                 from /opt/mxnet/include/mxnet/./base.h:33,
                 from /opt/mxnet/include/mxnet/storage.h:29,
                 from /opt/mxnet/src/engine/./threaded_engine.h:33,
                 from /opt/mxnet/src/engine/threaded_engine_perdevice.cc:32:
/opt/mxnet/include/mshadow/./base.h:192:12: fatal error: mkl_blas.h: No such file or directory
  192 |   #include <mkl_blas.h>
      |            ^~~~~~~~~~~~
compilation terminated.
make[2]: *** [CMakeFiles/mxnet_static.dir/build.make:286: CMakeFiles/mxnet_static.dir/src/common/exec_utils.cc.o] Error 1
In file included from /opt/mxnet/include/mshadow/tensor.h:35,
                 from /opt/mxnet/include/mxnet/./base.h:33,
                 from /opt/mxnet/include/mxnet/ndarray.h:43,
                 from /opt/mxnet/include/mxnet/runtime/packed_func.h:37,
                 from /opt/mxnet/include/mxnet/runtime/registry.h:49,
                 from /opt/mxnet/include/mxnet/api_registry.h:31,
                 from /opt/mxnet/src/api/_api_internal/_api_internal.cc:25:
/opt/mxnet/include/mshadow/./base.h:192:12: fatal error: mkl_blas.h: No such file or directory
  192 |   #include <mkl_blas.h>
      |            ^~~~~~~~~~~~
compilation terminated.
make[2]: *** [CMakeFiles/mxnet_static.dir/build.make:342: CMakeFiles/mxnet_static.dir/src/engine/naive_engine.cc.o] Error 1
In file included from /opt/mxnet/include/mshadow/tensor.h:35,
                 from /opt/mxnet/include/mxnet/./base.h:33,
                 from /opt/mxnet/include/mxnet/ndarray.h:43,
                 from /opt/mxnet/include/mxnet/runtime/packed_func.h:37,
                 from /opt/mxnet/src/api/operator/numpy/../utils.h:29,
                 from /opt/mxnet/src/api/operator/numpy/np_init_op.cc:24:
/opt/mxnet/include/mshadow/./base.h:192:12: fatal error: mkl_blas.h: No such file or directory
  192 |   #include <mkl_blas.h>
      |            ^~~~~~~~~~~~
compilation terminated.
make[2]: *** [CMakeFiles/mxnet_static.dir/build.make:384: CMakeFiles/mxnet_static.dir/src/engine/threaded_engine_perdevice.cc.o] Error 1
make[2]: *** [CMakeFiles/mxnet_static.dir/build.make:76: CMakeFiles/mxnet_static.dir/src/api/_api_internal/_api_internal.cc.o] Error 1
make[2]: *** [CMakeFiles/mxnet_static.dir/build.make:90: CMakeFiles/mxnet_static.dir/src/api/operator/numpy/np_init_op.cc.o] Error 1
In file included from /opt/mxnet/include/mshadow/tensor.h:35,
                 from /opt/mxnet/include/mxnet/./base.h:33,
                 from /opt/mxnet/include/mxnet/ndarray.h:43,
                 from /opt/mxnet/include/mxnet/runtime/packed_func.h:37,
                 from /opt/mxnet/src/api/operator/numpy/../utils.h:29,
                 from /opt/mxnet/src/api/operator/numpy/np_tensordot_op.cc:24:
/opt/mxnet/include/mshadow/./base.h:192:12: fatal error: mkl_blas.h: No such file or directory
  192 |   #include <mkl_blas.h>
      |            ^~~~~~~~~~~~
compilation terminated.
make[2]: *** [CMakeFiles/mxnet_static.dir/build.make:104: CMakeFiles/mxnet_static.dir/src/api/operator/numpy/np_tensordot_op.cc.o] Error 1
In file included from /opt/mxnet/include/mshadow/tensor.h:35,
                 from /opt/mxnet/include/mxnet/./base.h:33,
                 from /opt/mxnet/include/mxnet/ndarray.h:43,
                 from /opt/mxnet/include/mxnet/../../src/kvstore/gradient_compression.h:32,
                 from /opt/mxnet/include/mxnet/kvstore.h:34,
                 from /opt/mxnet/src/c_api/c_api_profile.cc:32:
/opt/mxnet/include/mshadow/./base.h:192:12: fatal error: mkl_blas.h: No such file or directory
  192 |   #include <mkl_blas.h>
      |            ^~~~~~~~~~~~
compilation terminated.
make[2]: *** [CMakeFiles/mxnet_static.dir/build.make:216: CMakeFiles/mxnet_static.dir/src/c_api/c_api_profile.cc.o] Error 1
make[1]: *** [CMakeFiles/Makefile2:510: CMakeFiles/mxnet_static.dir/all] Error 2
make: *** [Makefile:146: all] Error 2

Environment

We recommend using our script for collecting the diagnostic information with the following command

----------Python Info----------
Version      : 3.8.10
Compiler     : GCC 9.3.0
Build        : ('default', 'May 11 2021 07:01:05')
Arch         : ('64bit', 'ELF')
------------Pip Info-----------
Version      : 21.2.4
Directory    : /opt/conda/envs/computer_vision/lib/python3.8/site-packages/pip
----------MXNet Info-----------
No MXNet installed.
----------System Info----------
Platform     : Linux-5.4.0-80-generic-x86_64-with-glibc2.10
system       : Linux
node         : e4941ae20901
release      : 5.4.0-80-generic
version      : #90-Ubuntu SMP Fri Jul 9 22:49:44 UTC 2021
----------Hardware Info----------
machine      : x86_64
processor    : x86_64
Architecture:                    x86_64
CPU op-mode(s):                  32-bit, 64-bit
Byte Order:                      Little Endian
Address sizes:                   48 bits physical, 48 bits virtual
CPU(s):                          32
On-line CPU(s) list:             0-31
Thread(s) per core:              2
Core(s) per socket:              16
Socket(s):                       1
NUMA node(s):                    1
Vendor ID:                       AuthenticAMD
CPU family:                      25
Model:                           33
Model name:                      AMD Ryzen 9 5950X 16-Core Processor
Stepping:                        0
Frequency boost:                 enabled
CPU MHz:                         2899.888
CPU max MHz:                     3400.0000
CPU min MHz:                     2200.0000
BogoMIPS:                        6799.59
Virtualization:                  AMD-V
L1d cache:                       512 KiB
L1i cache:                       512 KiB
L2 cache:                        8 MiB
L3 cache:                        64 MiB
NUMA node0 CPU(s):               0-31
Vulnerability Itlb multihit:     Not affected
Vulnerability L1tf:              Not affected
Vulnerability Mds:               Not affected
Vulnerability Meltdown:          Not affected
Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl and seccomp
Vulnerability Spectre v1:        Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2:        Mitigation; Full AMD retpoline, IBPB conditional, IBRS_FW, STIBP always-on, RSB filling
Vulnerability Srbds:             Not affected
Vulnerability Tsx async abort:   Not affected
Flags:                           fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good nopl nonstop_tsc cpuid extd_apicid aperfmperf 
                                 pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core p
                                 erfctr_nb bpext perfctr_llc mwaitx cpb cat_l3 cdp_l3 hw_pstate ssbd mba ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 erms invpcid cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 xsa
                                 ves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr wbnoinvd arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload v
                                 gif umip pku ospke vaes vpclmulqdq rdpid overflow_recov succor smca

github-actions[bot] commented 3 years ago

Welcome to Apache MXNet (incubating)! We are on a mission to democratize AI, and we are glad that you are contributing to it by opening this issue. Please make sure to include all the relevant context, and one of the @apache/mxnet-committers will be here shortly. If you are interested in contributing to our project, let us know! Also, be sure to check out our guide on contributing to MXNet and our development guides wiki.

leezu commented 3 years ago

Does it happen with a clean checkout / clean build folder?

bartekkuncer commented 3 years ago

@mxnet-label-bot add [mkl]

akarbown commented 3 years ago

@Data-drone have you tried '-DUSE_BLAS=open' to compile with openblas? With reference to the MKL - where do you have the MKL-specific files/libraries (path to /mkl directory that contains: /lib && include/).

Data-drone commented 3 years ago

@leezu It was with a clean build clone (I was using fresh docker script to start each time)

akarbown commented 3 years ago

@Data-drone, have you tried to compile with openblas using -DUSE_BLAS=open (instead of -DUSE_BLAS=openblas)? I'm not sure but the error you have pasted looks like it tried to compile mxnet with MKL, but couldn't find MKL files. I can try to help you, but need more data.

Data-drone commented 3 years ago

@akarbown I haven't had time yet but I will hopefully get time to try later today

Data-drone commented 3 years ago

Okay looks like -DUSE_BLAS=open sorted it. But I also moved to 1.9.0-rc7 as well.

My final compile options are:

cmake -DCMAKE_BUILD_TYPE=Release \
    -DCMAKE_INSTALL_PREFIX=$(python -c "import sys; print(sys.prefix)") \
    -DCMAKE_FIND_ROOT_PATH=${CONDA_PREFIX} \
    -DCMAKE_FIND_ROOT_PATH_MODE_LIBRARY=BOTH \
    -DUSE_OPENCV=OFF \
    -DUSE_LAPACK=ON \
    -DUSE_MKL_IF_AVAILABLE=OFF \
    -DUSE_BLAS=open \
    -DUSE_INTEL_PATH=${CONDA_PREFIX}/lib \
    -DUSE_CUDA=ON \
    -DUSE_CUDNN=ON \
    -DUSE_OPENMP=ON \
    -DUSE_TVM_OP=OFF \
    -DMXNET_CUDA_ARCH="7.5" \
    -DUSE_NCCL=ON \
    -DUSE_SSE=ON \
    .. \
  && cmake --build . --parallel 28