<div dir="ltr"><font size="2" face="Calibri"><span style="font-size:11pt"><div>Dear Joel Sherrill, Sebastian Huber, Pavel Pisa, and enthusiastic RTEMS community,</div>
<div> </div>
<div>I am working on Texas board TMS570LC4357 in RTEMS version 4.11.</div>
<div>I have managed to make the board support package and my sample application worked perfectly on board.</div>
<div> </div>
<div>Currently, I have started working on an application project which includes <u>RTEMS Makefile as a child Makefile</u> in my project. I use Eclipse IDE.</div>
<div>I was able to link RTEMS libraries and sources required for the application into my project.</div>
<div> </div>
<div>After this, I came across an error which is quite confusing for me.</div>
<div> </div>
<div><b>“</b><b>/opt/rtems/4.11/bin/arm-rtems4.11-ld: </b><b>rpc</b><b>.o: compiled for a big-endian system and target is little-endian</b><b>”</b><b> </b>which is a strange error because the target is definitely big-endian(TMS570).</div>
<div> </div>
<div>rpc is a source file used for my application.</div>
<div> </div>
<div>Note that <b><u>I am using the RTEMS toolset to compile my project</u></b><b><u> and build binaries for the application.</u></b></div>
<div>My project does contain many other source files which should be compiled as per big-endian configuration.</div>
<div> </div>
<div>The CFLAGS script I used inside the Makefile for target configuration is as follows:</div>
<div> </div>
<div>-march=armv7-r -mthumb -mbig-endian -D__ARMEB__ -D__VFP_NOFP__ -mfpu=vfpv3-d16 -mfloat-abi=hard</div>
<div> </div>
<div>This configuration is for ARM Cortex-R5F processor, floating-point disabled, and big-endian target.</div>
<div>TMS570 series are big-endian variants and hence I have configured it in such a way. (Texas Instruments TMS570LC4357 )</div>
<div> </div>
<div>I have configured the RTEMS toolchain as per the official RTEMS documentation. I build the TMS570 BSP with the same CFLAGS script for a sample RTEMS application as mentioned above and it worked like a charm. (Here I used <u>RTEMS Makefile as the parent makefile</u>)</div>
<div> </div>
<div>While going through RTEMS mailing lists, I could not find adequate information on this issue.</div>
<div> </div>
<div>So my questions are:</div>
<div> </div>
<ol style="margin:0px;padding-left:20.25pt"><li>What is the reason for above-described error “<b>compiled for a big endian system and target is little endian</b>” ?</li></ol>
<div style="padding-left:20.25pt">Is it because the RTEMS GCC by default doesn’t support big-endian?</div>
<div style="padding-left:20.25pt"> </div>
<ol style="margin:0px;padding-left:20.25pt" start="2"><li>If so, how to make the RTEMS GCC compiler to support big-endian targets?</li></ol>
<div style="padding-left:20.25pt">Should I re-build RTEMS toolchain with big-endian supported GCC?</div>
<div style="padding-left:20.25pt"> </div>
<div style="padding-left:20.25pt">I have found a thread ( <a href="https://lists.rtems.org/pipermail/devel/2014-June/007232.html" target="_blank"><font color="blue"><u>https://lists.rtems.org/pipermail/devel/2014-June/007232.html</u></font></a> ) from Pavel Pisa regarding this.</div>
<div style="padding-left:20.25pt">Can I use the experimental GCC found at <a href="https://rtime.felk.cvut.cz/debian/pool/main/g/gcc-4.9.3/" target="_blank"><font color="blue"><u>https://rtime.felk.cvut.cz/debian/pool/main/g/gcc-4.9.3/</u></font></a> for my project or should I configure GCC by myself?</div>
<div style="padding-left:20.25pt"> </div>
<ol style="margin:0px;padding-left:20.25pt" start="3"><li>Is there any other RTEMS toolchain components need to be configured (for example <b>binutils</b><b> </b>)?</li></ol>
<div> </div>
<div>I really need some help to make any further progress.</div>
<div> </div>
<div>Thank you for your time and consideration.</div><div><br></div><div>Best regards,</div><div>Aby Tom<br></div></span></font></div>