RSB 4.11.2-rc4 rtems-arm build fails on Windows

Chris Johns chrisj at rtems.org
Fri Apr 21 00:02:56 UTC 2017


On 20/04/2017 19:42, Linda Huxley wrote:
> Hi Chris,
>
> After 3 other attempts, I did a clean build of the ARM version of RTEMS
> on Windows, using the following commands:
>
> $ cd /c
> $ mkdir -p opt/rtems
> $ cd opt/rtems
> $ wget
> https://ftp.rtems.org/pub/rtems/releases/4.11/rc/4.11.2-rc4/rtems-source-builder-4.11.2-rc4.tar.xz
>
> $ tar Jxf rtems-source-builder-4.11.2-rc4.tar.xz
> $ mv rtems-source-builder-4.11.2-rc4 r4.11.2
> $ cd r4.11.2/rtems/
> $ ../source-builder/sb-set-builder --prefix=/c/opt/rtems/4.11
> 4.11/rtems-arm --jobs=none
>
> The build fails with the following trace, much like the SPARC build
> failed before applying the patch provided by Worth Burruss.

Yes I saw this. I am sorry I had no time to look into this.

>
> error: building arg4n2xwm1
> Build FAILED
>   See error report:
> rsb-report-arm-rtems4.11-gcc-4.9.3-newlib-2.2.0.20150423-x86_64-w64-mingw32-1.txt
>
> error: building arg4n2xwm1
> Build Set: Time 0:36:57.587859
> Build FAILED
>
> The salient error messages at the bottom of the error report are:
>
> In file included from ../../gcc-4.9.3/gcc/cp/except.c:1013:0:
> cfns.gperf: In function 'const char* libc_name_p(const char*, unsigned
> int)':
> cfns.gperf:101:1: error: 'const char* libc_name_p(const char*, unsigned
> int)' redeclared inline with 'gnu_inline' attribute
> cfns.gperf:26:14: note: 'const char* libc_name_p(const char*, unsigned
> int)' previously declared here
>
> Is it possible that RTEMS 4.11.2-rc4 does not apply Worth's patch when
> building the ARM version of RTEMS?
>
> When I look in:
>
> /c/opt/rtems/r4.11.2/rtems/patches
>
> I only see the following patch files:
>
> automake-1.12.6-bugzilla.redhat.com-1239379.diff
> binutils-2.26-gas-reloc.patch
> newlib-ARM-Optimize-IEEE-754-sqrt-implementation.diff
>
> I do not see Worth's patch file anywhere:
>
> gcc-4.9.3-20170404-1.patch
>
> (In the mean time, I was able to build the i386 version of RTEMS, build
> an example RTEMS program then boot it and run it using VirtualBox.)

Hmm that is weird. Thanks for the hint. I will see if this makes it 
simpler to figure out.

Chris



More information about the users mailing list