disapproval or revision 462968658bdc2a3c48a8d634f9be59d450259f49

disapproval or revision 462968658bdc2a3c48a8d634f9be59d450259f49

[Koen Kooi]

Hi,

Some days ago I fixed openssl-native so it doesn't put target flags into 
the host build anymore:

openssl-native: stop leaking target compile flags (FULL_OPTIMIZATION) 
into native build, it kills kittens.

r462968658bdc2a3c48a8d634f9be59d450259f49 reverses that fix and breaks 
my build. So it has been disapproved.

regards,

Koen

Re: disapproval or revision 462968658bdc2a3c48a8d634f9be59d450259f49

[Jeremy Lainé]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Koen,

I guess our messages missed each other..

Koen Kooi wrote:
> Some days ago I fixed openssl-native so it doesn't put target flags
>  into the host build anymore:
>
> openssl-native: stop leaking target compile flags
> (FULL_OPTIMIZATION) into native build, it kills kittens.
>
> r462968658bdc2a3c48a8d634f9be59d450259f49 reverses that fix and
> breaks my build. So it has been disapproved.
>
My commit did *not* reverse your fix, I kept your FULL_OPTIMIZATION =
"", so these do not end up in the CFLAG variable. As explained in my
other email, CFLAG mustn't be exported, it breaks things as it
forcibly replace the compilation flags.

- --
Jeremy LAINE
Bolloré telecom | 11bis, rue Scribe | F-75009 Paris
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFILchGCL1n3lVB9EYRAmkrAJ0aEZgrcN0AXsgdMZqVBizwzoecjQCgxzlO
x/cYM0HMpWrK1NVaRzSzuoE=
=Jhkw
-----END PGP SIGNATURE-----

Re: disapproval or revision 462968658bdc2a3c48a8d634f9be59d450259f49

[Koen Kooi]

Jeremy Lainé wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Hi Koen,
>
> I guess our messages missed each other..
>
> Koen Kooi wrote:
>> Some days ago I fixed openssl-native so it doesn't put target flags
>>   into the host build anymore:
>>
>> openssl-native: stop leaking target compile flags
>> (FULL_OPTIMIZATION) into native build, it kills kittens.
>>
>> r462968658bdc2a3c48a8d634f9be59d450259f49 reverses that fix and
>> breaks my build. So it has been disapproved.
>>
> My commit did *not* reverse your fix, I kept your FULL_OPTIMIZATION =
> "", so these do not end up in the CFLAG variable.

Ah, while I was trying to get rid of FULL_OPTIMIZATION, it wasn't enough 
to set it to "", but it seems I probably had a broken bitbake cache, 
since now everything works as expected (including a correct 
FULL_OPTIMIZATION[1]!).

Sorry for the mess, I committed a new revision that removes the exports 
and makes 0.9.8m the default again, since it now builds (probably due to 
having correct CFLAG this time *cough*).

regards,

Koen

[1] I kept it empty just to be complete sure. It's not like we're going 
to do heavy crypto during builds.

> As explained in my
> other email, CFLAG mustn't be exported, it breaks things as it
> forcibly replace the compilation flags.
>
> - --
> Jeremy LAINE
> Bolloré telecom | 11bis, rue Scribe | F-75009 Paris
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.6 (GNU/Linux)
> Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
>
> iD8DBQFILchGCL1n3lVB9EYRAmkrAJ0aEZgrcN0AXsgdMZqVBizwzoecjQCgxzlO
> x/cYM0HMpWrK1NVaRzSzuoE=
> =Jhkw
> -----END PGP SIGNATURE-----

Re: disapproval or revision 462968658bdc2a3c48a8d634f9be59d450259f49

[Jeremy Lainé]

>> Sorry for the mess, I committed a new revision that removes the exports
>> and makes 0.9.8m the default again, since it now builds (probably due to
>> having correct CFLAG this time *cough*).

Cool... well, nearly! openssl-native_0.9.8g.bb still contains
DEFAULT_PREFERENCE = "-1".

If it's OK with you I'd like to see 0.9.8g as the default for native
builds too, so we only download one .tar.gz by default.

Cheers,
Jeremy