<html><body><div style="color:#000; background-color:#fff; font-family:times new roman, new york, times, serif;font-size:16px"><div><span></span></div><div></div><div id="yui_3_16_0_1_1428954309550_5194" dir="ltr">Hi Wim! </div><div id="yui_3_16_0_1_1428954309550_5193" dir="ltr">I tried your suggestions (CC=icc and including -mmic in CFLAGS) and copied the executable to </div><div id="yui_3_16_0_1_1428954309550_14289" dir="ltr">the Phi co-processor card, but it fails due to a library dependency. </div><div id="yui_3_16_0_1_1428954309550_8188" dir="ltr">---- </div><div id="yui_3_16_0_1_1428954309550_9299" dir="ltr">$ ./example2-c bottles.b </div><div id="yui_3_16_0_1_1428954309550_4639"> /lib64/ld-linux-x86-64.so.2: No such file or directory<br></div><div id="yui_3_16_0_1_1428954309550_4638">$ ls -l /lib64/ld-linux*<br>lrwxrwxrwx 1 root root 13 Jan 1 1970 /lib64/ld-linux-k1om.so.2 -> ld-2.14.90.so</div><div id="yui_3_16_0_1_1428954309550_9296" dir="ltr">---- </div><div id="yui_3_16_0_1_1428954309550_9295" dir="ltr"><br></div><div id="yui_3_16_0_1_1428954309550_8189" dir="ltr">When I tried to run it from the Xeon host with micnativeloadex I also get an error:</div><div id="yui_3_16_0_1_1428954309550_8190" dir="ltr">---- </div><div id="yui_3_16_0_1_1428954309550_9294" dir="ltr">$ export SINK_LD_LIBRARY_PATH=/opt/intel/composer_xe_2015.1.133/lib/mic<br>$ micnativeloadex ./example2-c<br>Supplied binary does not match the Intel(R) Xeon Phi(TM)<br>coprocessor that is installed.</div><div id="yui_3_16_0_1_1428954309550_9860" dir="ltr">---- <br></div><div id="yui_3_16_0_1_1428954309550_9300" dir="ltr"><br></div><div id="yui_3_16_0_1_1428954309550_9862" dir="ltr">I couldn't find any unresolved library dependencies: </div><div id="yui_3_16_0_1_1428954309550_15395" dir="ltr">---- </div><div id="yui_3_16_0_1_1428954309550_13736" dir="ltr">$ micnativeloadex ./example2-c -l</div><div id="yui_3_16_0_1_1428954309550_14841" dir="ltr">Dependency information for ./example2-c</div><div id="yui_3_16_0_1_1428954309550_13737" dir="ltr"> Full path was resolved as <br> /tmp/usession-release-2.5.1-6/testing_1/./example2-c</div><div id="yui_3_16_0_1_1428954309550_13179" dir="ltr"> Binary was built for X86_64 architecture</div><div id="yui_3_16_0_1_1428954309550_13178" dir="ltr"> SINK_LD_LIBRARY_PATH = /opt/intel/composer_xe_2015.1.133/lib/mic</div><div id="yui_3_16_0_1_1428954309550_13177" dir="ltr"> Dependencies Found:<br> (none found)</div><div id="yui_3_16_0_1_1428954309550_11522" dir="ltr"> Dependencies Not Found Locally (but may exist already on the coprocessor):<br> librt.so.1<br> libm.so.6<br> libgcc_s.so.1<br> libpthread.so.0<br> libc.so.6<br> libdl.so.2</div><div id="yui_3_16_0_1_1428954309550_12074" dir="ltr">---- <br></div><div id="yui_3_16_0_1_1428954309550_9863" dir="ltr"><br></div><div id="yui_3_16_0_1_1428954309550_13181" dir="ltr">I noticed in the output above that the "Binary was built for X86_64 architecture" </div><div id="yui_3_16_0_1_1428954309550_15947" dir="ltr">but couldn't find any info re it. </div><div id="yui_3_16_0_1_1428954309550_19814" dir="ltr"><br></div><div id="yui_3_16_0_1_1428954309550_19260" dir="ltr">Plain ldd does show the dependency on ld-linux-x86-64.so.2 (here shown with </div><div id="yui_3_16_0_1_1428954309550_23127" dir="ltr">extra libraries libm, libgcc and libdl included with the "-mmic" flag):</div><div id="yui_3_16_0_1_1428954309550_17604" dir="ltr">---- </div><div id="yui_3_16_0_1_1428954309550_17603" dir="ltr">$ ldd example2-c<br> linux-vdso.so.1 => (0x00007fffcabc1000)<br> librt.so.1 => /lib64/librt.so.1 (0x00000037d8c00000)<br> libm.so.6 => /lib64/libm.so.6 (0x00000037d9400000)<br> libgcc_s.so.1 => /lib64/libgcc_s.so.1 (0x00000037de800000)<br> libpthread.so.0 => /lib64/libpthread.so.0 (0x00000037d8400000)<br> libc.so.6 => /lib64/libc.so.6 (0x00000037d8000000)<br> libdl.so.2 => /lib64/libdl.so.2 (0x00000037d8800000)<br> /lib64/ld-linux-x86-64.so.2 (0x00000037d7c00000)<br></div><div id="yui_3_16_0_1_1428954309550_13735" dir="ltr">---- </div><div id="yui_3_16_0_1_1428954309550_22025" dir="ltr"><br></div><div id="yui_3_16_0_1_1428954309550_24231" dir="ltr">I'd appreciate any help/pointers to the library dependency! </div><div id="yui_3_16_0_1_1428954309550_24230" dir="ltr"><br></div><div id="yui_3_16_0_1_1428954309550_5192" dir="ltr">Thanks, </div><div id="yui_3_16_0_1_1428954309550_5191" dir="ltr">Ajit</div><div id="yui_3_16_0_1_1428954309550_9298" dir="ltr">==== </div><div id="yui_3_16_0_1_1428954309550_9299" dir="ltr"><br></div><div id="yui_3_16_0_1_1428954309550_4205"><div id="yui_3_16_0_1_1428954309550_4204"></div></div> <br><div class="qtdSeparateBR"><br><br></div><div class="yahoo_quoted" style="display: block;"> <div style="font-family: times new roman, new york, times, serif; font-size: 16px;"> <div style="font-family: HelveticaNeue, Helvetica Neue, Helvetica, Arial, Lucida Grande, sans-serif; font-size: 16px;"> <div dir="ltr"> <font face="Arial" size="2"> On Sunday, April 12, 2015 9:46 PM, Ajit Dingankar <ajit.dingankar@ieee.org> wrote:<br> </font> </div> <blockquote style="padding-left: 5px; margin-top: 5px; margin-left: 5px; border-left-color: rgb(16, 16, 255); border-left-width: 2px; border-left-style: solid;"> <br><br> <div class="y_msg_container"><div id="yiv2366168267"><div><div style="color: rgb(0, 0, 0); font-family: times new roman, new york, times, serif; font-size: 16px; background-color: rgb(255, 255, 255);"><div id="yiv2366168267yui_3_16_0_1_1428899521914_5854" dir="ltr">@Wim: <br clear="none"></div><div id="yiv2366168267yui_3_16_0_1_1428899521914_6769" dir="ltr">Yes, I am trying to translate on the card since I couldn't find much info <br clear="none"></div><div id="yiv2366168267yui_3_16_0_1_1428899521914_6770" dir="ltr">re cross-compiling and someone had suggested direct translation on the <br clear="none"></div><div id="yiv2366168267yui_3_16_0_1_1428899521914_7211" dir="ltr">target platform (though not specifically for Xeon Phi). Anyway, thanks <br clear="none"></div><div id="yiv2366168267yui_3_16_0_1_1428899521914_9105" dir="ltr">a lot for the tips! I'll try them at work tomorrow. BTW, I'm more interested <br clear="none"></div><div id="yiv2366168267yui_3_16_0_1_1428899521914_11942" dir="ltr">in providing new functionality, not targeting high performance initially, so <br clear="none"></div><div id="yiv2366168267yui_3_16_0_1_1428899521914_13403" dir="ltr">multi-threading is not important for me now; I just need to get it working... <br clear="none"></div><div id="yiv2366168267yui_3_16_0_1_1428899521914_8576" dir="ltr"><br clear="none"></div><div id="yiv2366168267yui_3_16_0_1_1428899521914_9106" dir="ltr">@Armin: <br clear="none"></div><div id="yiv2366168267yui_3_16_0_1_1428899521914_9107" dir="ltr">Sorry I didn't have access to the actual error message when I posted the <br clear="none"></div><div id="yiv2366168267yui_3_16_0_1_1428899521914_9569" dir="ltr">question as a general query. Will do so if I can't make progress trying <br clear="none"></div><div id="yiv2366168267yui_3_16_0_1_1428899521914_10028" dir="ltr">Wim's suggestions from work. </div><div id="yiv2366168267yui_3_16_0_1_1428899521914_13874" dir="ltr"><br clear="none"></div><div id="yiv2366168267yui_3_16_0_1_1428899521914_11471" dir="ltr">@Yury: </div><div id="yiv2366168267yui_3_16_0_1_1428899521914_11472" dir="ltr">Thanks for the clarification re Xeon Phi generations and usage models. <br clear="none"></div><div id="yiv2366168267yui_3_16_0_1_1428899521914_14346" dir="ltr"><br clear="none"></div><div id="yiv2366168267yui_3_16_0_1_1428899521914_14347" dir="ltr">Thanks, <br clear="none"></div><div dir="ltr">Ajit</div><div dir="ltr">==== <br clear="none"></div><div id="yiv2366168267yui_3_16_0_1_1428899521914_10973" dir="ltr"><br clear="none"></div><div id="yiv2366168267yui_3_16_0_1_1428899521914_14818" dir="ltr"><br clear="none"></div><div id="yiv2366168267yui_3_16_0_1_1428899521914_10029" dir="ltr"><br clear="none"></div><div class="yiv2366168267" id="yiv2366168267yui_3_16_0_1_1428899521914_7212" dir="ltr"> </div><br clear="none"><div class="yiv2366168267qtdSeparateBR"><br clear="none"><br clear="none"></div><div class="yiv2366168267yqt1789143785" id="yiv2366168267yqt22741"><div class="yiv2366168267yahoo_quoted" style="display: block;"> <div style="font-family: times new roman, new york, times, serif; font-size: 16px;"> <div style="font-family: HelveticaNeue, Helvetica Neue, Helvetica, Arial, Lucida Grande, sans-serif; font-size: 16px;"> <div dir="ltr"> <font face="Arial" size="2"> On Sunday, April 12, 2015 2:46 AM, Yury V. Zaytsev <yury@shurup.com> wrote:<br clear="none"> </font> </div> <blockquote style="padding-left: 5px; margin-top: 5px; margin-left: 5px; border-left-color: rgb(16, 16, 255); border-left-width: 2px; border-left-style: solid;"> <br clear="none"><br clear="none"> <div class="yiv2366168267y_msg_container">On Sun, 2015-04-12 at 09:32 +0200, Armin Rigo wrote:<br clear="none">> Wim's reply is not helpful at all for me, as it is mostly given as a<br clear="none">> series of three-letter acronyms I've never heard about :-)<br clear="none"><br clear="none">He's simply referring to the different generations of MICs (MIC = Many<br clear="none">Integrated Core architecture, KNC = Knights Corner [older models], KNL =<br clear="none">Knights Landing [newer models]).<div class="yiv2366168267yqt1452461996" id="yiv2366168267yqtfd66835"><br clear="none"><br clear="none">> I don't know where Xeon Phi fits there. It seems to be an x86-64 from<br clear="none">> Wikipedia, but I'm not sure about what is special about it.</div><br clear="none"><br clear="none">I've shortly played with KNC, and put very simply in its current shape<br clear="none">it's basically a plug-in computer extension card, which can function in<br clear="none">several modes, e.g. as an accelerator which receives tasks from the host<br clear="none">system and executes them, or even as a more or less stand-alone box<br clear="none">inside the box running (for instance) a stripped down Linux system.<br clear="none"><br clear="none">In the latter mode, software just requires cross-compilation and then<br clear="none">can run on the board as if it was a stand-alone computer, in the former<br clear="none">you have to make use of special APIs to run your tasks on the MICs. It<br clear="none">looks like Wim has taken the first approach, which makes total sense to<br clear="none">get it working with minimal effort :-)<br clear="none"><br clear="none">So yes, in this approximation, assume it's x86-64 which requires a<br clear="none">special cross-compiler and has a bit different subset of supported<br clear="none">insns.<br clear="none"><br clear="none">-- <br clear="none">Sincerely yours,<br clear="none">Yury V. Zaytsev<div class="yiv2366168267yqt1452461996" id="yiv2366168267yqtfd13693"><br clear="none"><br clear="none"><br clear="none"></div><br clear="none"><br clear="none"></div> </blockquote> </div> </div> </div></div></div></div></div><br><br></div> </blockquote> </div> </div> </div></div></body></html>