https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Is-there-a-max-num-of-threads-for-mkl-dss/m-p/1055059#M21394 <P>Hello, everyone.</P> <P>I used MKL DSS to solve linear systems with openMP. When calling dss with 32 threads, i came up with a run time error. Debugging messages are as below:</P> <P>......</P> <P>[New Thread 0x7fff86b5e700 (LWP 10279)]<BR /> [New Thread 0x7fff8675d700 (LWP 10280)]<BR /> [New Thread 0x7fff8635c700 (LWP 10281)]<BR /> [New Thread 0x7fff85f5b700 (LWP 10282)]<BR /> [New Thread 0x7fff85b5a700 (LWP 10283)]<BR /> [New Thread 0x7fff85759700 (LWP 10284)]<BR /> [New Thread 0x7fff85358700 (LWP 10285)]<BR /> [New Thread 0x7fff84f57700 (LWP 10286)]<BR /> [New Thread 0x7fff84b56700 (LWP 10287)]<BR /> [New Thread 0x7fff84755700 (LWP 10288)]<BR /> [New Thread 0x7fff7fafe700 (LWP 10289)]<BR /> [New Thread 0x7fff7f6fd700 (LWP 10290)]<BR /> [New Thread 0x7fff7f2fc700 (LWP 10291)]<BR /> [New Thread 0x7fff7eefb700 (LWP 10292)]<BR /> [New Thread 0x7fff7eafa700 (LWP 10293)]<BR /> [New Thread 0x7fff7e6f9700 (LWP 10294)]<BR /> [New Thread 0x7fff7e2f8700 (LWP 10295)]<BR /> [New Thread 0x7fff7def7700 (LWP 10296)]<BR /> [New Thread 0x7fff7daf6700 (LWP 10297)]<BR /> [New Thread 0x7fff7d6f5700 (LWP 10298)]<BR /> [New Thread 0x7fff7d2f4700 (LWP 10299)]<BR /> [New Thread 0x7fff7cef3700 (LWP 10300)]<BR /> [New Thread 0x7fff7caf2700 (LWP 10301)]<BR /> [New Thread 0x7fff7c6f1700 (LWP 10302)]<BR /> TAG:3</P> <P>Program received signal SIGSEGV, Segmentation fault.<BR /> [Switching to Thread 0x7fff7cef3700 (LWP 10300)]<BR /> 0x00007ffff6da6ca2 in mkl_pds_lp64_c_blkl_omp_pardiso ()<BR /> &nbsp;&nbsp; from /home/***/intel/composer_xe_2015.0.090/mkl/lib/intel64/libmkl_intel_thread.so<BR /> Missing separate debuginfos, use: debuginfo-install glibc-2.12-1.149.el6.x86_64 libgcc-4.4.7-11.el6.x86_64 libgomp-4.4.7-11.el6.x86_64</P> <P>"Tag:3" is a label in my program after dss_reorder, thus it may be something wrong with call of dss_factor_*. However, when i run the same program with only the difference of less threads, say 16, everyting seems OK. No error emerges when calling pardiso with 32 threads also. What happened after all?</P> <P>Thanks,</P> <P>wzmumu</P> Thu, 13 Nov 2014 14:04:32 GMT _____ 2014-11-13T14:04:32Z https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Is-there-a-max-num-of-threads-for-mkl-dss/m-p/1055059#M21394 <P>Hello, everyone.</P> <P>I used MKL DSS to solve linear systems with openMP. When calling dss with 32 threads, i came up with a run time error. Debugging messages are as below:</P> <P>......</P> <P>[New Thread 0x7fff86b5e700 (LWP 10279)]<BR /> [New Thread 0x7fff8675d700 (LWP 10280)]<BR /> [New Thread 0x7fff8635c700 (LWP 10281)]<BR /> [New Thread 0x7fff85f5b700 (LWP 10282)]<BR /> [New Thread 0x7fff85b5a700 (LWP 10283)]<BR /> [New Thread 0x7fff85759700 (LWP 10284)]<BR /> [New Thread 0x7fff85358700 (LWP 10285)]<BR /> [New Thread 0x7fff84f57700 (LWP 10286)]<BR /> [New Thread 0x7fff84b56700 (LWP 10287)]<BR /> [New Thread 0x7fff84755700 (LWP 10288)]<BR /> [New Thread 0x7fff7fafe700 (LWP 10289)]<BR /> [New Thread 0x7fff7f6fd700 (LWP 10290)]<BR /> [New Thread 0x7fff7f2fc700 (LWP 10291)]<BR /> [New Thread 0x7fff7eefb700 (LWP 10292)]<BR /> [New Thread 0x7fff7eafa700 (LWP 10293)]<BR /> [New Thread 0x7fff7e6f9700 (LWP 10294)]<BR /> [New Thread 0x7fff7e2f8700 (LWP 10295)]<BR /> [New Thread 0x7fff7def7700 (LWP 10296)]<BR /> [New Thread 0x7fff7daf6700 (LWP 10297)]<BR /> [New Thread 0x7fff7d6f5700 (LWP 10298)]<BR /> [New Thread 0x7fff7d2f4700 (LWP 10299)]<BR /> [New Thread 0x7fff7cef3700 (LWP 10300)]<BR /> [New Thread 0x7fff7caf2700 (LWP 10301)]<BR /> [New Thread 0x7fff7c6f1700 (LWP 10302)]<BR /> TAG:3</P> <P>Program received signal SIGSEGV, Segmentation fault.<BR /> [Switching to Thread 0x7fff7cef3700 (LWP 10300)]<BR /> 0x00007ffff6da6ca2 in mkl_pds_lp64_c_blkl_omp_pardiso ()<BR /> &nbsp;&nbsp; from /home/***/intel/composer_xe_2015.0.090/mkl/lib/intel64/libmkl_intel_thread.so<BR /> Missing separate debuginfos, use: debuginfo-install glibc-2.12-1.149.el6.x86_64 libgcc-4.4.7-11.el6.x86_64 libgomp-4.4.7-11.el6.x86_64</P> <P>"Tag:3" is a label in my program after dss_reorder, thus it may be something wrong with call of dss_factor_*. However, when i run the same program with only the difference of less threads, say 16, everyting seems OK. No error emerges when calling pardiso with 32 threads also. What happened after all?</P> <P>Thanks,</P> <P>wzmumu</P> Thu, 13 Nov 2014 14:04:32 GMT https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Is-there-a-max-num-of-threads-for-mkl-dss/m-p/1055059#M21394 _____ 2014-11-13T14:04:32Z https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Is-there-a-max-num-of-threads-for-mkl-dss/m-p/1055060#M21395 <P>You may need to boost stack limit, if you hadn't already done so; even "unlimited" stack may have a maximum number of threads for a given test case.</P> <P>If you aren't using the entire thread stack allotment, you might try reducing it.</P> <P>You may need to know whether you are running in 32- or 64-bit mode.</P> Thu, 13 Nov 2014 19:28:00 GMT https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Is-there-a-max-num-of-threads-for-mkl-dss/m-p/1055060#M21395 TimP 2014-11-13T19:28:00Z