eglibc_2.17.bb: Failure expanding variable SELECTED_OPTIMIZATION[:=]

eglibc_2.17.bb: Failure expanding variable SELECTED_OPTIMIZATION[:=]

[Denys Dmytriyenko]

Hi,

Has anyone seen this error before? It happens for me quite often, but not 
always, on Dylan branch during initial parsing:

Parsing recipes...
ERROR: ExpansionError during parsing .../oe-core/meta/recipes-core/eglibc/eglibc_2.17.bb:
  Failure expanding variable SELECTED_OPTIMIZATION[:=], expression was ${@get_optimization(d)}
  which triggered exception NameError: name 'get_optimization' is not defined
ERROR: Command execution failed: Exited with 1


Looking at the code, I don't see anything suspicious:

# eglibc can't be built without optimization, if someone tries to compile an
# entire image as -O0, we override it with -O2 here and give a note about it.
def get_optimization(d):
    selected_optimization = d.getVar("SELECTED_OPTIMIZATION", True)
    if base_contains("SELECTED_OPTIMIZATION", "-O0", "x", "", d) == "x":
        bb.note("eglibc can't be built with -O0, -O2 will be used instead.")
        return selected_optimization.replace("-O0", "-O2")
    return selected_optimization

SELECTED_OPTIMIZATION := "${@get_optimization(d)}"


It sounds like by the time SELECTED_OPTIMIZATION tries to make an assignment 
with immediate var expansion, get_optimization() function is not yet defined.

This happens with high BB_NUM_THREADS and bitbake 1.18. Trying bitbake from 
master on Dylan gives "AttributeError: 'NoneType' object has no attribute 
'rfind'", but that's another story.

Anyone has any clues or suggestions? Thanks.

-- 
Denys

Re: eglibc_2.17.bb: Failure expanding variable SELECTED_OPTIMIZATION[:=]

[Denys Dmytriyenko]

Any ideas or suggestions at this point would be greatly appreciated! Thanks.


On Fri, Sep 06, 2013 at 02:57:41PM -0400, Denys Dmytriyenko wrote:
> Hi,
> 
> Has anyone seen this error before? It happens for me quite often, but not 
> always, on Dylan branch during initial parsing:
> 
> Parsing recipes...
> ERROR: ExpansionError during parsing .../oe-core/meta/recipes-core/eglibc/eglibc_2.17.bb:
>   Failure expanding variable SELECTED_OPTIMIZATION[:=], expression was ${@get_optimization(d)}
>   which triggered exception NameError: name 'get_optimization' is not defined
> ERROR: Command execution failed: Exited with 1
> 
> 
> Looking at the code, I don't see anything suspicious:
> 
> # eglibc can't be built without optimization, if someone tries to compile an
> # entire image as -O0, we override it with -O2 here and give a note about it.
> def get_optimization(d):
>     selected_optimization = d.getVar("SELECTED_OPTIMIZATION", True)
>     if base_contains("SELECTED_OPTIMIZATION", "-O0", "x", "", d) == "x":
>         bb.note("eglibc can't be built with -O0, -O2 will be used instead.")
>         return selected_optimization.replace("-O0", "-O2")
>     return selected_optimization
> 
> SELECTED_OPTIMIZATION := "${@get_optimization(d)}"
> 
> 
> It sounds like by the time SELECTED_OPTIMIZATION tries to make an assignment 
> with immediate var expansion, get_optimization() function is not yet defined.
> 
> This happens with high BB_NUM_THREADS and bitbake 1.18. Trying bitbake from 
> master on Dylan gives "AttributeError: 'NoneType' object has no attribute 
> 'rfind'", but that's another story.
> 
> Anyone has any clues or suggestions? Thanks.
> 
> -- 
> Denys
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core@lists.openembedded.org
> http://lists.openembedded.org/mailman/listinfo/openembedded-core

Re: eglibc_2.17.bb: Failure expanding variable SELECTED_OPTIMIZATION[:=]

[Richard Purdie]

On Mon, 2013-09-09 at 09:51 -0400, Denys Dmytriyenko wrote:
> Any ideas or suggestions at this point would be greatly appreciated! Thanks.

I have vague recollections about this from something happening in
master. Perhaps try with:

http://git.openembedded.org/bitbake/commit/?id=4d50690489ee8dc329a9b0c7bc4ceb29b71e95e9

applied. This might things more deterministic, perhaps always breaking
or perhaps never.

Have you got two definitions of the "get_optimization" function?

Cheers,

Richard

Re: eglibc_2.17.bb: Failure expanding variable SELECTED_OPTIMIZATION[:=]

[Denys Dmytriyenko]

On Mon, Sep 09, 2013 at 04:24:23PM +0100, Richard Purdie wrote:
> On Mon, 2013-09-09 at 09:51 -0400, Denys Dmytriyenko wrote:
> > Any ideas or suggestions at this point would be greatly appreciated! Thanks.

Richard,

> I have vague recollections about this from something happening in
> master. Perhaps try with:
> 
> http://git.openembedded.org/bitbake/commit/?id=4d50690489ee8dc329a9b0c7bc4ceb29b71e95e9
> 
> applied. This might things more deterministic, perhaps always breaking
> or perhaps never.

Ah, good call! Massaging the patch onto 1.18 seems to have solved the parsing 
breakage - it passed 10 out of 10 tries so far. I'll keep testing and run a 
few real builds, but would you be willing to backport it to 1.18? Please let 
me know if you need an updated patch re-submitted. Thanks.


> Have you got two definitions of the "get_optimization" function?

I grepped through all my layers and eglibc_2.17.inc is the only place where it 
gets defined and used.

-- 
Denys