Dynamic Load Failure on Three MIPS BSPs

[Joel Sherrill]

Well.. looking at the log, I think the answer is that rtems-syms is not
processing the "-EL" option on the MIPS. Thus the objects are always
big endian on the MIPS but three BSPs use the -EL option.

mips-rtems4.12-gcc -B../../../../../hurricane/lib/ -specs bsp_specs -qrtems
-mips3 -G0 -EL -O2 -g -fomit-frame-pointer -ffunction-sections
-fdata-sections -Wall -Wmissing-prototypes -Wimplicit-function-declaration
-Wstrict-prototypes -Wnested-externs  -Wl,--gc-sections  -mips3 -G0 -EL
-o dl01.pre.exe init.o dl-load.o dl-tar.o
mips-rtems4.12-objcopy -O srec dl01.pre.exe dl01.pre.srec1
../../../../../../tools/build/packhex < dl01.pre.srec1 > dl01.pre.ralf
rm -f dl01.pre.srec1
mips-rtems4.12-nm -g -n dl01.pre.exe > dl01.pre.num
mips-rtems4.12-size dl01.pre.exe
   text   data    bss    dec    hex filename
 167912  35392   9968 213272  34118 dl01.pre.exe
rm -f dl01.pre.ralf
mv dl01.pre.exe dl01.pre
rtems-syms -e -c "-mips3 -G0 -EL -O2 -g -fomit-frame-pointer
-ffunction-sections -fdata-sections -Wall -Wmissing-prototypes
-Wimplicit-function-declaration -Wstrict-prototypes -Wnested-externs" -o
dl-sym.o dl01.pre
mips-rtems4.12-gcc -B../../../../../hurricane/lib/ -specs bsp_specs -qrtems
-mips3 -G0 -EL -O2 -g -fomit-frame-pointer -ffunction-sections
-fdata-sections -Wall -Wmissing-prototypes -Wimplicit-function-declaration
-Wstrict-prototypes -Wnested-externs  -Wl,--gc-sections  -mips3 -G0 -EL   \
   -o dl01.exe init.o dl-load.o dl-tar.o dl-sym.o
/data/home/joel/rtems-4.11-work/tools/4.12/bin/../lib/gcc/mips-rtems4.12/6.1.1/../../../../mips-rtems4.12/bin/ld:
dl-sym.o: compiled for a big endian system and target is little endian
/data/home/joel/rtems-4.11-work/tools/4.12/bin/../lib/gcc/mips-rtems4.12/6.1.1/../../../../mips-rtems4.12/bin/ld:
dl-sym.o: endianness incompatible with that of the selected emulation
/data/home/joel/rtems-4.11-work/tools/4.12/bin/../lib/gcc/mips-rtems4.12/6.1.1/../../../../mips-rtems4.12/bin/ld:
failed to merge target specific data of file dl-sym.o
collect2: error: ld returned 1 exit status
gmake[7]: *** [dl01.exe] Error 1


On Tue, Oct 4, 2016 at 9:15 AM, Joel Sherrill <joel at rtems.org> wrote:

> Sending a build log privately.
>
> On Mon, Oct 3, 2016 at 5:02 PM, Chris Johns <chrisj at rtems.org> wrote:
>
>> On 04/10/2016 02:57, Joel Sherrill wrote:
>>
>>>
>>> Looking at failures from a build sweep.
>>>
>>> I think this one is known but am passing along. The rtbx4925, rtbx4938,
>>> and hurricane fail with this:
>>>
>>>     -o dl04.exe init.o dl-load.o dl-cpp.o dl-tar.o dl-sym.o
>>> /data/home/joel/rtems-4.11-work/tools/4.12/bin/../lib/gcc/mi
>>> ps-rtems4.12/6.1.1/../../../../mips-rtems4.12/bin/ld:
>>> dl-sym.o: compiled for a big endian system and target is little endian
>>> /data/home/joel/rtems-4.11-work/tools/4.12/bin/../lib/gcc/mi
>>> ps-rtems4.12/6.1.1/../../../../mips-rtems4.12/bin/ld:
>>> dl-sym.o: endianness incompatible with that of the selected emulation
>>> /data/home/joel/rtems-4.11-work/tools/4.12/bin/../lib/gcc/mi
>>> ps-rtems4.12/6.1.1/../../../../mips-rtems4.12/bin/ld:
>>> failed to merge target specific data of file dl-sym.o
>>>
>>> I am adding dl03-dl05 to the file
>>> testsuites/testdata/dltests-broken-on-this-bsp.tcfg
>>> but wanted to report this in case there was discussion.
>>>
>>>
>> Could you please look at the build output and see what flags are
>> different between the building of dl-sym.o and another object file?
>>
>> I wonder if a flag in the MIPS BSP is not being seen by the symbol
>> builder tool.
>>
>> Chris
>> _______________________________________________
>> devel mailing list
>> devel at rtems.org
>> http://lists.rtems.org/mailman/listinfo/devel
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.rtems.org/pipermail/devel/attachments/20161004/7fad5e9d/attachment-0002.html>