Re: RPi build issue on OSX

[Bryan Hundven <bryanhundven at gmail dot com>]

On 5/17/13 10:06 AM, "Nathan Plesnicher" <nplez1@gmail.com> wrote:

>Hi Bryan,
>
>I actually was able to get it working. It turned out that just
>switching to eglibc 2_17 (the default for the RPi sample was 2_16)
>fixed the problem when combined with my other patches to get the build
>working on OSX.
>
>For others trying to build this toolchain, the patches I had to apply
>were as follows:
>
>http://sourceware.org/ml/crossgcc/2012-05/msg00012.html
>Need libintl linked in, and this diff did the trick. Not sure if the
>ncurses and escdelay change is actually needed
>
>http://sourceware.org/ml/crossgcc/2011-05/msg00094.html
>(optional; only use if you are building gdb) Include the host expat so
>that gdb works as expected
>
>The following one actually needs to be applied to the eglibc source
>that is downloaded...
>http://sourceware.org/ml/crossgcc/2013-03/msg00109.html
>Patch rpc/types.h so that it doesn't redefine the types on OSX
>(already in sys/types.h)
>
>I hope this helps someone in the future.

Cool! Thanks Nathan!

-Bryan


>Nathan
>
>On Thu, May 16, 2013 at 2:46 PM, Bryan Hundven <bryanhundven@gmail.com>
>wrote:
>> On 5/16/13 1:55 PM, "Nathan Plesnicher" <nplez1@gmail.com> wrote:
>>
>> Hi Nathan,
>>
>>
>>>I am having some trouble getting the RPi sample to build using
>>>homebrew on Mountain Lion. I am getting the following error in the
>>>'Installing C library headers & start files' step:
>>>
>>>[CFG  ]    checking for __thread... yes
>>>[CFG  ]    checking for tls_model attribute... yes
>>>[CFG  ]    checking for compiler option to disable generation of FMA
>>>instructions... -ffp-contract=off
>>>[CFG  ]    checking for libgd... no
>>>[CFG  ]    checking for is_selinux_enabled in -lselinux... no
>>>[CFG  ]    checking size of long double... 8
>>>[CFG  ]    checking for _FORTIFY_SOURCE predefine... no
>>>[ERROR]    checking whether armv6-rpi-linux-gnueabi-gcc     implicitly
>>>enables -fstack-protector... configure: error: unexpected symbols in
>>>test:
>>>[ERROR]
>>>[ERROR]  >>
>>>[ERROR]  >>  Build failed in step 'Installing C library headers & start
>>>files'
>>>[ERROR]  >>        called in step '(top-level)'
>>>[ERROR]  >>
>>>[ERROR]  >>  Error happened in: CT_DoExecLog[scripts/functions@257]
>>>[ERROR]  >>        called from:
>>>do_libc_backend_once[scripts/build/libc/glibc-eglibc.sh-common@347]
>>>[ERROR]  >>        called from:
>>>do_libc_backend[scripts/build/libc/glibc-eglibc.sh-common@143]
>>>[ERROR]  >>        called from:
>>>do_libc_start_files[scripts/build/libc/glibc-eglibc.sh-common@60]
>>>[ERROR]  >>        called from: main[scripts/crosstool-NG.sh@632]
>>>[ERROR]  >>
>>>[ERROR]  >>  For more info on this error, look at the file: 'build.log'
>>>[ERROR]  >>  There is a list of known issues, some with workarounds, in:
>>>[ERROR]  >>      '/Volumes/CrossTool2NG/crosstool-ng-1.18.0/docs/B -
>>>Known issues.txt'
>>>[ERROR]
>>>[ERROR]  (elapsed: 6:48.00)
>>
>> Would it be possible for you to post your build.log somewhere?
>>Preferably
>> not attached, as the mailing list has limited storage.
>>
>>>Looks like some sort of script problem but I am having trouble
>>>locating the specific problem. Any ideas?
>>>
>>>Not sure it matters, but I have been able to successfully build a
>>>toolchain using the non-Linaro gcc w/ the standard glibc (instead of
>>>eglibc) after applying a few manual patches, but I was hoping I could
>>>get the RPi sample to work since I am building the toolchain for RPi
>>>development.
>>
>> Maybe the linaro source has different requirements on mac, we'll see
>>when
>> we get the build.log
>>
>> Cheers,
>>
>> -Bryan
>>
>>>--
>>>For unsubscribe information see http://sourceware.org/lists.html#faq
>>>
>>
>>



--
For unsubscribe information see http://sourceware.org/lists.html#faq