[Trac #57] Decide on toolchain for compiling LinuxBIOS

[Zarro Boogs per Child]

#57: Decide on toolchain for compiling LinuxBIOS
-------------------------------+--------------------------------------------
 Reporter:  wmb at firmworks.com  |        Owner:  marcelo 
     Type:  task               |       Status:  assigned
 Priority:  low                |    Milestone:  final   
Component:  infrastructure     |   Resolution:          
 Keywords:                     |  
-------------------------------+--------------------------------------------
Changes (by marcelo):

  * owner:  krstic => marcelo
  * status:  new => assigned

Comment:

 Replying to [comment:7 rsmith]:
 > Replying to [comment:6 marcelo]:
 >
 > > Conclusion:
 > >
 > > From the experiments above the problem seems to be related to the
 build environment, but not the compiler/assembler/linker combination.
 > >
 > > Until the root cause is found we will continue to generate LinuxBIOS
 images on FC5 environments.
 >
 > Marcelo:  Will you please try to take the romcc that is compiled from
 FC6 and use it in the FC5 > build and see if the problem shows up?

 No, FC5 generated binaries with FC6's romcc work normally.

 Doing it the other way around (FC5 generated LB tools on FC6 box):

 [marcelo at xeon4 fallback]$ md5sum build_opt_tbl nrv2b buildrom romcc
 fabe4d33f8b5e739397e4b2aee60aeb4  build_opt_tbl
 4fbcca107bc61c8396aeca90fb94377b  nrv2b
 f448659cdb4a46070cc768f4dee25925  buildrom
 f49da33329474b5432a213fbf8abd69d  romcc

 [root at dmt fallback]# md5sum build_opt_tbl nrv2b buildrom romcc
 fabe4d33f8b5e739397e4b2aee60aeb4  build_opt_tbl
 4fbcca107bc61c8396aeca90fb94377b  nrv2b
 f448659cdb4a46070cc768f4dee25925  buildrom
 f49da33329474b5432a213fbf8abd69d  romcc

 Still exhibits the problem.

-- 
Ticket URL: <http://dev.laptop.org/ticket/57#comment:9>
One Laptop Per Child <http://laptop.org/>