<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" version="2.0">
  <channel>
    <title>topic nodeperf makes SegFault in Intel® oneAPI Math Kernel Library</title>
    <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/nodeperf-makes-SegFault/m-p/904093#M11528</link>
    <description>&lt;P&gt;I'm glad I was helpful.&lt;/P&gt;
&lt;P&gt;&lt;/P&gt;
&lt;P&gt;Best regards,&lt;/P&gt;
&lt;P&gt;Alexander&lt;/P&gt;</description>
    <pubDate>Mon, 01 Feb 2010 17:12:38 GMT</pubDate>
    <dc:creator>Alex_Kosenkov</dc:creator>
    <dc:date>2010-02-01T17:12:38Z</dc:date>
    <item>
      <title>nodeperf makes SegFault</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/nodeperf-makes-SegFault/m-p/904088#M11523</link>
      <description>&lt;P&gt;Good day Everyone!&lt;/P&gt;
&lt;P&gt;&lt;/P&gt;
&lt;P&gt;I've download and install Intel C Compilers &amp;amp; MKL 11.064. Then build OpenMPI with intel compilers. I wish to run nodeperf on my new cluster, so, I compile nodeperf.c from MKL benchmark as:&lt;/P&gt;
&lt;P&gt;mpicc -o nodeperf nodeperf.c ../../lib/em64t/libmkl_solver_ilp64_sequential.a -Wl,--start-group ../../lib/em64t/libmkl_intel_ilp64.a ../../lib/em64t/libmkl_sequential.a ../../lib/em64t/libmkl_core.a ../../lib/em64t/libmkl_blacs_openmpi_ilp64.a -Wl,--end-group -lpthread&lt;/P&gt;
&lt;P&gt;&lt;/P&gt;
&lt;P&gt;But when I try to run on single node it crash with a segfault:&lt;/P&gt;
&lt;P&gt;&lt;/P&gt;
&lt;P&gt;No multi-threaded MPI detected (Hybrid may or may not work)&lt;/P&gt;
&lt;P&gt;The time/date of the run... at Mon Feb 1 13:11:57 2010&lt;/P&gt;
&lt;P&gt;This driver was compiled with:&lt;/P&gt;
&lt;P&gt;-DITER=4 -DLINUX -DNOACCUR -DPREC=double&lt;/P&gt;
&lt;P&gt;Malloc done. Used 827090096 bytes&lt;/P&gt;
&lt;P&gt;[umt:22585] *** Process received signal ***&lt;/P&gt;
&lt;P&gt;[umt:22585] Signal: Segmentation fault (11)&lt;/P&gt;
&lt;P&gt;[umt:22585] Signal code: (128)&lt;/P&gt;
&lt;P&gt;[umt:22585] Failing at address: (nil)&lt;/P&gt;
&lt;P&gt;[umt:22585] [ 0] /lib64/libpthread.so.0 [0x3c3600e930]&lt;/P&gt;
&lt;P&gt;[umt:22585] [ 1] ./nodeperf(mkl_blas_mc3_dgemm_copybn+0x1c1) [0x54d641]&lt;/P&gt;
&lt;P&gt;[umt:22585] [ 2] ./nodeperf(mkl_blas_mc3_xdgemm_par+0xf24) [0x547344]&lt;/P&gt;
&lt;P&gt;[umt:22585] [ 3] ./nodeperf(mkl_blas_mc3_xdgemm+0x1c8) [0x4211d8]&lt;/P&gt;
&lt;P&gt;[umt:22585] [ 4] ./nodeperf(mkl_blas_xdgemm+0x1c5) [0x406e55]&lt;/P&gt;
&lt;P&gt;[umt:22585] [ 5] ./nodeperf(mkl_blas_dgemm+0x51) [0x406b51]&lt;/P&gt;
&lt;P&gt;[umt:22585] [ 6] ./nodeperf(DGEMM+0x13a) [0x40685a]&lt;/P&gt;
&lt;P&gt;[umt:22585] [ 7] ./nodeperf(main+0x530) [0x4060e0]&lt;/P&gt;
&lt;P&gt;[umt:22585] [ 8] /lib64/libc.so.6(__libc_start_main+0xf4) [0x3c3541d994]&lt;/P&gt;
&lt;P&gt;[umt:22585] [ 9] ./nodeperf(realloc+0xb1) [0x405af9]&lt;/P&gt;
&lt;P&gt;[umt:22585] *** End of error message ***&lt;/P&gt;
&lt;P&gt;Segmentation fault&lt;/P&gt;
&lt;P&gt;&lt;/P&gt;
&lt;P&gt;The node config:&lt;/P&gt;
&lt;P&gt;Processors: Intel Xeon CPU E5520 @ 2.27GHz&lt;/P&gt;
&lt;P&gt;RAM: 16Gb&lt;/P&gt;
&lt;P&gt;MKLROOT and other PATH's looks fine.&lt;/P&gt;
&lt;P&gt;&lt;/P&gt;
&lt;P&gt;Can you help me ?&lt;/P&gt;</description>
      <pubDate>Mon, 01 Feb 2010 08:22:17 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/nodeperf-makes-SegFault/m-p/904088#M11523</guid>
      <dc:creator>Ilya_Zelenchuk</dc:creator>
      <dc:date>2010-02-01T08:22:17Z</dc:date>
    </item>
    <item>
      <title>nodeperf makes SegFault</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/nodeperf-makes-SegFault/m-p/904089#M11524</link>
      <description>&lt;P&gt;Ilya,&lt;/P&gt;
&lt;P&gt;Please try to use/DMKL_ILP64 compiler option while linking ilp64 libraries? will it help you?&lt;/P&gt;
&lt;P&gt;--Gennady&lt;/P&gt;</description>
      <pubDate>Mon, 01 Feb 2010 09:46:37 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/nodeperf-makes-SegFault/m-p/904089#M11524</guid>
      <dc:creator>Gennady_F_Intel</dc:creator>
      <dc:date>2010-02-01T09:46:37Z</dc:date>
    </item>
    <item>
      <title>nodeperf makes SegFault</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/nodeperf-makes-SegFault/m-p/904090#M11525</link>
      <description>&lt;P&gt;Seems not. I hope thatoption DMKL_ILP64 is using correct by me.&lt;/P&gt;
&lt;P&gt;[root@umt mp_linpack]# mpicc -o nodeperf nodeperf.c -DMKL_ILP64 ../../lib/em64t/libmkl_solver_ilp64_sequential.a -Wl,--start-group ../../lib/em64t/libmkl_intel_ilp64.a ../../lib/em64t/libmkl_sequential.a ../../lib/em64t/libmkl_core.a ../../lib/em64t/libmkl_blacs_openmpi_ilp64.a -Wl,--end-group -lpthread&lt;/P&gt;
&lt;P&gt;/opt/intel/Compiler/11.1/064/lib/intel64/libimf.so: warning: warning: feupdateenv is not implemented and will always fail&lt;/P&gt;
&lt;P&gt;[root@umt mp_linpack]# ./nodeperf&lt;/P&gt;
&lt;P&gt;--------------------------------------------------------------------------&lt;/P&gt;
&lt;P&gt;A deprecated MCA parameter value was specified in an MCA parameter&lt;/P&gt;
&lt;P&gt;file. Deprecated MCA parameters should be avoided; they may disappear&lt;/P&gt;
&lt;P&gt;in future releases.&lt;/P&gt;
&lt;P&gt;Deprecated parameter: pls_rsh_agent&lt;/P&gt;
&lt;P&gt;--------------------------------------------------------------------------&lt;/P&gt;
&lt;P&gt;No multi-threaded MPI detected (Hybrid may or may not work)&lt;/P&gt;
&lt;P&gt;The time/date of the run... at Mon Feb 1 17:26:07 2010&lt;/P&gt;
&lt;P&gt;This driver was compiled with:&lt;/P&gt;
&lt;P&gt;-DITER=4 -DLINUX -DNOACCUR -DPREC=double&lt;/P&gt;
&lt;P&gt;Malloc done. Used 827090096 bytes&lt;/P&gt;
&lt;P&gt;[umt:24196] *** Process received signal ***&lt;/P&gt;
&lt;P&gt;[umt:24196] Signal: Segmentation fault (11)&lt;/P&gt;
&lt;P&gt;[umt:24196] Signal code: (128)&lt;/P&gt;
&lt;P&gt;[umt:24196] Failing at address: (nil)&lt;/P&gt;
&lt;P&gt;[umt:24196] [ 0] /lib64/libpthread.so.0 [0x3c3600e930]&lt;/P&gt;
&lt;P&gt;[umt:24196] [ 1] ./nodeperf(mkl_blas_mc3_dgemm_copybn+0x1c1) [0x54d641]&lt;/P&gt;
&lt;P&gt;[umt:24196] [ 2] ./nodeperf(mkl_blas_mc3_xdgemm_par+0xf24) [0x547344]&lt;/P&gt;
&lt;P&gt;[umt:24196] [ 3] ./nodeperf(mkl_blas_mc3_xdgemm+0x1c8) [0x4211d8]&lt;/P&gt;
&lt;P&gt;[umt:24196] [ 4] ./nodeperf(mkl_blas_xdgemm+0x1c5) [0x406e55]&lt;/P&gt;
&lt;P&gt;[umt:24196] [ 5] ./nodeperf(mkl_blas_dgemm+0x51) [0x406b51]&lt;/P&gt;
&lt;P&gt;[umt:24196] [ 6] ./nodeperf(DGEMM+0x13a) [0x40685a]&lt;/P&gt;
&lt;P&gt;[umt:24196] [ 7] ./nodeperf(main+0x530) [0x4060e0]&lt;/P&gt;
&lt;P&gt;[umt:24196] [ 8] /lib64/libc.so.6(__libc_start_main+0xf4) [0x3c3541d994]&lt;/P&gt;
&lt;P&gt;[umt:24196] [ 9] ./nodeperf(realloc+0xb1) [0x405af9]&lt;/P&gt;
&lt;P&gt;[umt:24196] *** End of error message ***&lt;/P&gt;
&lt;P&gt;Segmentation fault&lt;/P&gt;
&lt;P&gt;&lt;/P&gt;
&lt;P&gt;&lt;/P&gt;</description>
      <pubDate>Mon, 01 Feb 2010 12:30:34 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/nodeperf-makes-SegFault/m-p/904090#M11525</guid>
      <dc:creator>Ilya_Zelenchuk</dc:creator>
      <dc:date>2010-02-01T12:30:34Z</dc:date>
    </item>
    <item>
      <title>nodeperf makes SegFault</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/nodeperf-makes-SegFault/m-p/904091#M11526</link>
      <description>&lt;P&gt;Hello, Ilya&lt;/P&gt;
&lt;P&gt;The answer to this problem is quite simple: MP LINPACK doesn't support ILP64.&lt;/P&gt;
&lt;P&gt;You can try to compile it like this:&lt;/P&gt;
&lt;P&gt;mpiicc -o nodeperf nodeperf.c -O3 -Wl,--start-group $MKL/libmkl_intel_lp64.a $MKL/libmkl_sequential.a $MKL/libmkl_core.a -Wl,--end-group -lpthread&lt;/P&gt;
&lt;P&gt;This way it doesn't crash.&lt;/P&gt;
&lt;P&gt;Also don't forget that if you are using mpicc - most likely you are using MPI wrapper for gcc. I don't see the point why not to use mpiicc which is a wrapper for icc.&lt;/P&gt;
&lt;P&gt;Hope I've been of any help.&lt;/P&gt;
&lt;P&gt;&lt;/P&gt;
&lt;P&gt;Best regards,&lt;/P&gt;
&lt;P&gt;Alexander&lt;/P&gt;</description>
      <pubDate>Mon, 01 Feb 2010 14:00:49 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/nodeperf-makes-SegFault/m-p/904091#M11526</guid>
      <dc:creator>Alex_Kosenkov</dc:creator>
      <dc:date>2010-02-01T14:00:49Z</dc:date>
    </item>
    <item>
      <title>nodeperf makes SegFault</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/nodeperf-makes-SegFault/m-p/904092#M11527</link>
      <description>&lt;P&gt;Yes, it works now!&lt;/P&gt;
&lt;P&gt;Thanks.&lt;/P&gt;
&lt;P&gt;&lt;/P&gt;
&lt;P&gt;I know mpicc is wrapper. But i'm using icc in it.&lt;/P&gt;
&lt;P&gt;&lt;/P&gt;
&lt;P&gt;[root@umt mp_linpack]# mpicc --version&lt;/P&gt;
&lt;P&gt;icc (ICC) 11.1 20091130&lt;/P&gt;
&lt;P&gt;Copyright (C) 1985-2009 Intel Corporation. All rights reserved.&lt;/P&gt;
&lt;P&gt;[root@umt mp_linpack]# mpicc -O3 -o nodeperf nodeperf.c -Wl,--start-group ../../lib/em64t/libmkl_intel_lp64.a ../../lib/em64t/libmkl_sequential.a ../../lib/em64t/libmkl_core.a -Wl,--end-group -lpthread&lt;/P&gt;
&lt;P&gt;/opt/intel/Compiler/11.1/064/lib/intel64/libimf.so: warning: warning: feupdateenv is not implemented and will always fail&lt;/P&gt;
&lt;P&gt;[root@umt mp_linpack]# taskset 08 ./nodeperf&lt;/P&gt;
&lt;P&gt;--------------------------------------------------------------------------&lt;/P&gt;
&lt;P&gt;A deprecated MCA parameter value was specified in an MCA parameter&lt;/P&gt;
&lt;P&gt;file. Deprecated MCA parameters should be avoided; they may disappear&lt;/P&gt;
&lt;P&gt;in future releases.&lt;/P&gt;
&lt;P&gt;Deprecated parameter: pls_rsh_agent&lt;/P&gt;
&lt;P&gt;--------------------------------------------------------------------------&lt;/P&gt;
&lt;P&gt;No multi-threaded MPI detected (Hybrid may or may not work)&lt;/P&gt;
&lt;P&gt;The time/date of the run... at Mon Feb 1 19:09:27 2010&lt;/P&gt;
&lt;P&gt;This driver was compiled with:&lt;/P&gt;
&lt;P&gt;-DITER=4 -DLINUX -DNOACCUR -DPREC=double&lt;/P&gt;
&lt;P&gt;Malloc done. Used 827090096 bytes&lt;/P&gt;
&lt;P&gt;(0 of 1): NN lda=10000 ldb= 168 ldc=10000 0 0 0 9342.140 x.x.x.x&lt;/P&gt;
&lt;P&gt;[root@umt mp_linpack]#&lt;/P&gt;</description>
      <pubDate>Mon, 01 Feb 2010 14:12:32 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/nodeperf-makes-SegFault/m-p/904092#M11527</guid>
      <dc:creator>Ilya_Zelenchuk</dc:creator>
      <dc:date>2010-02-01T14:12:32Z</dc:date>
    </item>
    <item>
      <title>nodeperf makes SegFault</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/nodeperf-makes-SegFault/m-p/904093#M11528</link>
      <description>&lt;P&gt;I'm glad I was helpful.&lt;/P&gt;
&lt;P&gt;&lt;/P&gt;
&lt;P&gt;Best regards,&lt;/P&gt;
&lt;P&gt;Alexander&lt;/P&gt;</description>
      <pubDate>Mon, 01 Feb 2010 17:12:38 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/nodeperf-makes-SegFault/m-p/904093#M11528</guid>
      <dc:creator>Alex_Kosenkov</dc:creator>
      <dc:date>2010-02-01T17:12:38Z</dc:date>
    </item>
  </channel>
</rss>

