<?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 Hi, in Software Archive</title>
    <link>https://community.intel.com/t5/Software-Archive/offload-error-process-on-the-device-0-was-terminated-by-signal/m-p/951095#M19566</link>
    <description>&lt;P&gt;Hi,&lt;/P&gt;

&lt;P&gt;One of our experts looked at your problem.&lt;/P&gt;

&lt;P&gt;He believes that this may be a bug and has submitted a bug report.&lt;/P&gt;

&lt;P&gt;As experts are want to do, he made some recommendations as well: The threads may overwrite each other when they write back to the host; and it is a good idea to initialize your data when you allocate it. None of these will cause the issue you observe.&lt;/P&gt;

&lt;P&gt;Regards&lt;BR /&gt;
	--&lt;BR /&gt;
	Taylor&lt;BR /&gt;
	&lt;SPAN style="font-size: 1em; line-height: 1.5;"&gt;&amp;nbsp;&lt;/SPAN&gt;&lt;/P&gt;</description>
    <pubDate>Mon, 17 Mar 2014 21:41:50 GMT</pubDate>
    <dc:creator>TaylorIoTKidd</dc:creator>
    <dc:date>2014-03-17T21:41:50Z</dc:date>
    <item>
      <title>offload error: process on the device 0 was terminated by signal 11 (SIGSEGV)</title>
      <link>https://community.intel.com/t5/Software-Archive/offload-error-process-on-the-device-0-was-terminated-by-signal/m-p/951094#M19565</link>
      <description>&lt;P&gt;Hi!&lt;/P&gt;

&lt;P&gt;I have tried to use tree threads for CPU and MIC cooperation.&lt;/P&gt;

&lt;P&gt;But there is error "offload error: process on the device 0 was terminated by signal 11 (SIGSEGV)"&lt;/P&gt;

&lt;P&gt;The compile operation is "ifort mic.f90 -mkl -openmp". The codes are as follow:&amp;nbsp;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;program mic&lt;BR /&gt;
	use mic_lib&lt;BR /&gt;
	use omp_lib&lt;BR /&gt;
	implicit none&lt;BR /&gt;
	integer::mics,idx&lt;/P&gt;

&lt;P&gt;DOUBLE PRECISION,allocatable::A(:)&lt;BR /&gt;
	DOUBLE PRECISION,allocatable::B(:)&lt;BR /&gt;
	DOUBLE PRECISION,allocatable::C(:)&lt;BR /&gt;
	allocate(A(256*256))&lt;BR /&gt;
	allocate(B(256*256))&lt;BR /&gt;
	allocate(C(256*256))&lt;/P&gt;

&lt;P&gt;mics = offload_number_of_devices()&lt;/P&gt;

&lt;P&gt;!dir$ attributes offload:mic :: DGEMM&lt;/P&gt;

&lt;P&gt;!$OMP PARALLEL PRIVATE(idx) NUM_THREADS(mics+1)&lt;BR /&gt;
	!$OMP DO SCHEDULE (static)&lt;BR /&gt;
	do idx=0,mics&lt;BR /&gt;
	if(idx==mics) then&lt;BR /&gt;
	CALL DGEMM('N','N',256,256,256,1.d0,A,256,B,256,0.d0,C,256)&lt;BR /&gt;
	else&lt;BR /&gt;
	!dir$ offload target(mic:idx) in(A,B:length(256*256)) out(C:length(256*256))&lt;BR /&gt;
	CALL DGEMM('N','N',256,256,256,1.d0,A,256,B,256,0.d0,C,256)&lt;BR /&gt;
	end if&lt;BR /&gt;
	end do&lt;BR /&gt;
	!$OMP END DO&lt;BR /&gt;
	!$OMP END PARALLEL&lt;/P&gt;

&lt;P&gt;deallocate(A)&lt;BR /&gt;
	deallocate(B)&lt;BR /&gt;
	deallocate(C)&lt;BR /&gt;
	end program mic&lt;/P&gt;

&lt;P&gt;zhou&lt;/P&gt;</description>
      <pubDate>Tue, 07 Jan 2014 07:13:39 GMT</pubDate>
      <guid>https://community.intel.com/t5/Software-Archive/offload-error-process-on-the-device-0-was-terminated-by-signal/m-p/951094#M19565</guid>
      <dc:creator>zhou</dc:creator>
      <dc:date>2014-01-07T07:13:39Z</dc:date>
    </item>
    <item>
      <title>Hi,</title>
      <link>https://community.intel.com/t5/Software-Archive/offload-error-process-on-the-device-0-was-terminated-by-signal/m-p/951095#M19566</link>
      <description>&lt;P&gt;Hi,&lt;/P&gt;

&lt;P&gt;One of our experts looked at your problem.&lt;/P&gt;

&lt;P&gt;He believes that this may be a bug and has submitted a bug report.&lt;/P&gt;

&lt;P&gt;As experts are want to do, he made some recommendations as well: The threads may overwrite each other when they write back to the host; and it is a good idea to initialize your data when you allocate it. None of these will cause the issue you observe.&lt;/P&gt;

&lt;P&gt;Regards&lt;BR /&gt;
	--&lt;BR /&gt;
	Taylor&lt;BR /&gt;
	&lt;SPAN style="font-size: 1em; line-height: 1.5;"&gt;&amp;nbsp;&lt;/SPAN&gt;&lt;/P&gt;</description>
      <pubDate>Mon, 17 Mar 2014 21:41:50 GMT</pubDate>
      <guid>https://community.intel.com/t5/Software-Archive/offload-error-process-on-the-device-0-was-terminated-by-signal/m-p/951095#M19566</guid>
      <dc:creator>TaylorIoTKidd</dc:creator>
      <dc:date>2014-03-17T21:41:50Z</dc:date>
    </item>
    <item>
      <title>I have exactly the same error</title>
      <link>https://community.intel.com/t5/Software-Archive/offload-error-process-on-the-device-0-was-terminated-by-signal/m-p/951096#M19567</link>
      <description>&lt;P&gt;I have exactly the same error. How do I fix it?&lt;/P&gt;</description>
      <pubDate>Mon, 07 Apr 2014 01:29:37 GMT</pubDate>
      <guid>https://community.intel.com/t5/Software-Archive/offload-error-process-on-the-device-0-was-terminated-by-signal/m-p/951096#M19567</guid>
      <dc:creator>Dave_O_</dc:creator>
      <dc:date>2014-04-07T01:29:37Z</dc:date>
    </item>
    <item>
      <title>In the case cited in the</title>
      <link>https://community.intel.com/t5/Software-Archive/offload-error-process-on-the-device-0-was-terminated-by-signal/m-p/951097#M19568</link>
      <description>&lt;P&gt;In the case cited in the original post the compiler mishandles the constants (256, 0.d0, 1.d0) for the offload region treating them as read-write variables instead of read-only. That leads to the seg-fault when exiting from the offload region trying to write their values back to the host.&lt;BR /&gt;&lt;BR /&gt;The issue is expected to be fixed in the coming Update 3 later this month. To work around you would unfortunately need to define variables to pass in the constants, but do not declare with &lt;STRONG&gt;parameter &lt;/STRONG&gt;as those are disallowed in IN/OUT/INOUT. Something like this enables the case to run successfully:&lt;BR /&gt;&lt;BR /&gt;integer :: size=256&lt;BR /&gt;double precision:: zero_Dbl=0.d0&lt;BR /&gt;double precision:: one_Dbl=1.d0&lt;BR /&gt;&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Mon, 07 Apr 2014 17:41:04 GMT</pubDate>
      <guid>https://community.intel.com/t5/Software-Archive/offload-error-process-on-the-device-0-was-terminated-by-signal/m-p/951097#M19568</guid>
      <dc:creator>Kevin_D_Intel</dc:creator>
      <dc:date>2014-04-07T17:41:04Z</dc:date>
    </item>
    <item>
      <title>The 14.0 update 3 compiler</title>
      <link>https://community.intel.com/t5/Software-Archive/offload-error-process-on-the-device-0-was-terminated-by-signal/m-p/951098#M19569</link>
      <description>&lt;P&gt;The 14.0 update 3 compiler (14.0.3.174) containing the fix has now been posted and is available for download from the Intel registration center.&lt;/P&gt;</description>
      <pubDate>Sat, 31 May 2014 00:30:36 GMT</pubDate>
      <guid>https://community.intel.com/t5/Software-Archive/offload-error-process-on-the-device-0-was-terminated-by-signal/m-p/951098#M19569</guid>
      <dc:creator>Martyn_C_Intel</dc:creator>
      <dc:date>2014-05-31T00:30:36Z</dc:date>
    </item>
    <item>
      <title>Thanks very much for all of</title>
      <link>https://community.intel.com/t5/Software-Archive/offload-error-process-on-the-device-0-was-terminated-by-signal/m-p/951099#M19570</link>
      <description>&lt;P&gt;Thanks very much for all of your replies.&lt;BR /&gt;
	The code can implement well on our system now.&lt;BR /&gt;
	&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Tue, 31 Mar 2015 00:19:00 GMT</pubDate>
      <guid>https://community.intel.com/t5/Software-Archive/offload-error-process-on-the-device-0-was-terminated-by-signal/m-p/951099#M19570</guid>
      <dc:creator>zhou</dc:creator>
      <dc:date>2015-03-31T00:19:00Z</dc:date>
    </item>
  </channel>
</rss>

