Hi,

On 17 Dec 2018, at 18:54, Joe Kozak 
<[email protected]> wrote:

> oh i used the fixfile and then a cd.. then a difffixfile > 
> ../packa...blabla/funtoy.patch  the whole works!    then went in the 
> src.xxxxxxxxxxxxxxx/directory/.....  and found the unchanged file!     :)   
> I'll figure it out soon....  probobly need sleep too....  sometime.....  
> maybe…

in t2/trunk you do not need to redirect fixfilediff output, with a filename 
supplied fixfilediff will create the file directly in the package’s $confdir:

fixfilediff hotfix.patch
fixfilediff glibc-2.18.patch

are common names.

> I also have a sun 3/50... how workable a system do you think can be built?   
> I might have to tftp boot it, but i did it 20 years ago, i can figure it out 
> again! :)
> 
> On Mon, Dec 17, 2018 at 11:48 AM René Rebe <[email protected]> wrote:
> Hey,
> 
> On 17 Dec 2018, at 18:39, Joe Kozak 
> <[email protected]> wrote:
> 
>> On Mon, Dec 17, 2018 at 11:29 AM René Rebe <[email protected]> wrote:
>> Hi,
>> 
>> keeping the list cc’ed for other’s to learn, too. Hope that is ok, ...
>> 
>> On 17 Dec 2018, at 18:23, Joe Kozak 
>> <[email protected]> wrote:
>> 
>>> i am still learning this t2 setup, so i am sure that is much of it. 
>>> I have been trying the crosscompile on the x86 from the newest iso inside a 
>>> VM,   it complained a lot about linux-headers missing PATH_MAX  but I just 
>>> #define PATH_MAX 4096  wherever it complained. (maybe 3 places)  inside the 
>>> source tar file
>> 
>> This could be a latest glibc thing, I fixed a couple of those. Often they 
>> clean up things upstream and such regressions occur.
>> 
>>> and recompressed it into the download directory.    on native (not VM)  svn 
>>> checkout build attemts, this is not the case, and
>> 
>> Do you mean you edit the downloaded source tar balls? You should NEVER do 
>> this! patches go into packages/…/$pkg/*.patch, not editing the checksumed 
>> original tar balls. ever.
>> 
>> OH I know!  I was just playing, I tried patches too, but the patches didnt 
>> seem to work whereas the src tarball edit did.  Mainly i was trying to 
>> figure out why the patches were overwritten (or the makesystem rewrote them) 
> 
> patches are automatically applied from the t2-src package/*/$pkg/*.patch
> 
> no idea where you placed them ;-)
> 
>> I am trying fresh install in vm now. 
>> yes, cc all!
>> 
>>> linux-headers compiles fine.   
>>> 
>>> i think the buildchain is somehow not correct on the iso.         Q: is 
>>> there a way to use mine to rebuild the toolchains ?  
>> 
>> When you build a whole new target, t2 bootstraps a new toolchain in the 
>> stage 0.
>> Stage 0 is tooclchain, stage 1 is cross compile, then is 2 to re-build, 
>> sanity check tooclahin, and then mostly stage 5 compiling normally.
>> 
>> If you want to rebuild your “toolchain”, you could -force Emerge gcc, 
>> binutils, … but that is really not that necessary.
>> 
>>      René
>> 
>>> On Mon, Dec 17, 2018 at 9:38 AM René Rebe <[email protected]> wrote:
>>> Hi,
>>> 
>>> On 17 Dec 2018, at 16:01, Joe Kozak 
>>> <[email protected]> wrote:
>>> 
>>>> building trunk i get this.   tried fixfile on the offending code to just 
>>>> comment it out for fun, gets overwritten, must be dynamicly made after 
>>>> patches are applied.   tried extra compiler options '-O3' to be sure, no 
>>>> effect.    ALSO ./debug  then cd  into dir then eval $MAKE $makeopt  fails
>>>> 
>>>> will keep informed
>>> 
>>> 
>>> sorry, I did not yet memorise your other emails, you want to only build for 
>>> sparc64, or do you try random things?
>>> are you building on t2 already or some Debian / Ubuntu / other flavour?
>>> 
>>> I ask because the last time I tried sparc64 built for me, on my t2 based 
>>> build server:
>>> 
>>>     https://www.youtube.com/watch?v=XYsKct4T2xk
>>>     https://www.youtube.com/watch?v=10q2OxHAzQ4
>>> 
>>> 
>>> I can re-try latest trunk just for the fun of it, but the last binary iso 
>>> tested in this videos is not yet that old:
>>> 
>>>     t2-minimal-sparc64-r46697.iso
>>> 
>>> René
>>> 
>>>> ============
>>>> ============
>>>> ============
>>>> ============
>>>> 
>>>> moo-PowerEdge-R710 
>>>> src.glibc.default.20181217.085229.19023.moo-PowerEdge-R710 # ./debug.sh 
>>>> debug-glibc:[src.glibc.default.20181217.085229.19023.moo-PowerEdge-R710]# 
>>>> ls
>>>> .   ERROR-LOG  badfiles.txt  cmd_wrapper.log  debug.hooks  files.lst       
>>>>  fl_wrapper.wlog  flist.txt.new  flist.txt.tmp  install_wrapper.log  
>>>> olist.txt.old  xsrcdir.txt
>>>> ..  archdir    build.pid     debug.buildenv   debug.sh     fl_wrapper.rlog 
>>>>  flist.txt        flist.txt.old  glibc-2.28     olist.txt.new        
>>>> untar.txt
>>>> debug-glibc:[src.glibc.default.20181217.085229.19023.moo-PowerEdge-R710]# 
>>>> cd glibc-2.28/
>>>> debug-glibc:[glibc-2.28]# eval $MAKE $makeopt
>>>> Makeconfig:42: *** missing separator.  Stop.
>>>> 
>>>> 
>>>> ============
>>>> ============
>>>> ============
>>>> ============
>>>> 
>>>> In file included from <command-line>:
>>>> ./../include/libc-symbols.h:75:3: error: #error "glibc cannot be compiled 
>>>> without optimization"
>>>>  # error "glibc cannot be compiled without optimization"
>>>>    ^~~~~
>>>> make[2]: *** [../Makerules:287: 
>>>> /home/moo/t2-trunk/src.glibc.default.20181217.085229.19023.moo-PowerEdge-R710/glibc-2.28/objdir/tcb-offsets.h]
>>>>  Error 1
>>>> make[2]: Leaving directory 
>>>> '/home/moo/t2-trunk/src.glibc.default.20181217.085229.19023.moo-PowerEdge-R710/glibc-2.28/csu'
>>>> make[1]: *** [Makefile:258: csu/subdir_lib] Error 2
>>>> make[1]: Leaving directory 
>>>> '/home/moo/t2-trunk/src.glibc.default.20181217.085229.19023.moo-PowerEdge-R710/glibc-2.28'
>>>> make: *** [Makefile:9: all] Error 2
>>>> Due to previous errors, no 1-glibc.log file!
>>>> (Try enabling xtrace in the config to track an error inside the build 
>>>> system.)
>>>> --- BUILD ERROR ---
>>>> Install SysV Init script 'nscd' (19/81):  rcX done.
>>>> Creating file list and doing final adaptions ... 
>>>> Searching for orphaned files ...
>>>> Found 11 files for this package.
>>>> Found 1 orphaned files for this package.
>>>> Clear (old) md5sums ...
>>>> Creating md5sum files ... done.
>>>> Creating package description ...
>>>> Making post-install adaptions.
>>>> == 12/17/18 08:52:39 =[1]=> Aborted building package glibc.
>>>> -> Unmounting loop mounts ...
>>> 
>>> -- 
>>>  ExactCODE GmbH, Lietzenburger Str. 42, DE-10789 Berlin
>>>  http://exactcode.com | http://exactscan.com | http://ocrkit.com | 
>>> http://t2-project.org | http://rene.rebe.de
>>> 
>> 
>> -- 
>>  ExactCODE GmbH, Lietzenburger Str. 42, DE-10789 Berlin
>>  http://exactcode.com | http://exactscan.com | http://ocrkit.com | 
>> http://t2-project.org | http://rene.rebe.de
>> 
> 
> -- 
>  ExactCODE GmbH, Lietzenburger Str. 42, DE-10789 Berlin
>  http://exactcode.com | http://exactscan.com | http://ocrkit.com | 
> http://t2-project.org | http://rene.rebe.de
> 

-- 
 ExactCODE GmbH, Lietzenburger Str. 42, DE-10789 Berlin
 DE Legal: Amtsgericht Berlin (Charlottenburg) HRB 105123B, Tax-ID#: DE251602478
 Managing Director: René Rebe
 http://exactcode.com | http://exactscan.com | http://ocrkit.com | 
http://t2-project.org | http://rene.rebe.de

----------------------------------------------------------- 
If you wish to unsubscribe from this mailing, send mail to
[email protected] with a subject of: unsubscribe t2

Reply via email to