[PATCH] BUILD: Link just libsss_crypto with crypto libraries

[Lukas Slebodnik]

ehlo,

Attached patch should prevent such failures as in commit
73ec8fdfddb2d4bf99977f758eec80e1b1ee8542
    BUILD: Link test_data_provider_be with -ldl

SSS_CRYPT_LIBS should not be part of SSSD_LIBS.
all necessary program and libraries should be linked with
libsss_cryto.so an do not use SSS_CRYPT_LIBS directly.

LS

[Jakub Hrozek]

On Fri, Oct 02, 2015 at 11:26:33AM +0200, Lukas Slebodnik wrote:
>  ehlo,
>  
>  Attached patch should prevent such failures as in commit
>  73ec8fdfddb2d4bf99977f758eec80e1b1ee8542
>      BUILD: Link test_data_provider_be with -ldl
>  
>  SSS_CRYPT_LIBS should not be part of SSSD_LIBS.
>  all necessary program and libraries should be linked with
>  libsss_cryto.so an do not use SSS_CRYPT_LIBS directly.
>  
>  LS 
Can you add a comment to the definition of SSS_CRYPT_LIBS saying that
they shouldn't be used directly, but libsss_cryto should be used
instead?

[Lukas Slebodnik]

On (04/10/15 21:51), Jakub Hrozek wrote:
> On Fri, Oct 02, 2015 at 11:26:33AM +0200, Lukas Slebodnik wrote:
> > ehlo,
> > 
> > Attached patch should prevent such failures as in commit
> > 73ec8fdfddb2d4bf99977f758eec80e1b1ee8542
> >     BUILD: Link test_data_provider_be with -ldl
> > 
> > SSS_CRYPT_LIBS should not be part of SSSD_LIBS.
> > all necessary program and libraries should be linked with
> > libsss_cryto.so an do not use SSS_CRYPT_LIBS directly.
> > 
> > LS
>
> Can you add a comment to the definition of SSS_CRYPT_LIBS saying that
> they shouldn't be used directly, but libsss_cryto should be used
> instead? 
Is the comment sufficient?

LS

[Jakub Hrozek]

On Thu, Oct 08, 2015 at 09:49:51AM +0200, Lukas Slebodnik wrote:
>  On (04/10/15 21:51), Jakub Hrozek wrote:
>  >On Fri, Oct 02, 2015 at 11:26:33AM +0200, Lukas Slebodnik wrote:
>  >> ehlo,
>  >> 
>  >> Attached patch should prevent such failures as in commit
>  >> 73ec8fdfddb2d4bf99977f758eec80e1b1ee8542
>  >>     BUILD: Link test_data_provider_be with -ldl
>  >> 
>  >> SSS_CRYPT_LIBS should not be part of SSSD_LIBS.
>  >> all necessary program and libraries should be linked with
>  >> libsss_cryto.so an do not use SSS_CRYPT_LIBS directly.
>  >> 
>  >> LS
>  >
>  >Can you add a comment to the definition of SSS_CRYPT_LIBS saying that
>  >they shouldn't be used directly, but libsss_cryto should be used
>  >instead?
>  Is the comment sufficient?
>  
>  LS 
Thanks, that's exactly what I wanted.

Code-wise ACK, I'm waiting for CI to finish before pushing.

[Jakub Hrozek]

On Thu, Oct 08, 2015 at 12:19:17PM +0200, Jakub Hrozek wrote:
>  On Thu, Oct 08, 2015 at 09:49:51AM +0200, Lukas Slebodnik wrote:
>  > On (04/10/15 21:51), Jakub Hrozek wrote:
>  > >On Fri, Oct 02, 2015 at 11:26:33AM +0200, Lukas Slebodnik wrote:
>  > >> ehlo,
>  > >> 
>  > >> Attached patch should prevent such failures as in commit
>  > >> 73ec8fdfddb2d4bf99977f758eec80e1b1ee8542
>  > >>     BUILD: Link test_data_provider_be with -ldl
>  > >> 
>  > >> SSS_CRYPT_LIBS should not be part of SSSD_LIBS.
>  > >> all necessary program and libraries should be linked with
>  > >> libsss_cryto.so an do not use SSS_CRYPT_LIBS directly.
>  > >> 
>  > >> LS
>  > >
>  > >Can you add a comment to the definition of SSS_CRYPT_LIBS saying that
>  > >they shouldn't be used directly, but libsss_cryto should be used
>  > >instead?
>  > Is the comment sufficient?
>  > 
>  > LS
>  
>  Thanks, that's exactly what I wanted.
>  
>  Code-wise ACK, I'm waiting for CI to finish before pushing. 
CI passed:
    http://sssd-ci.duckdns.org/logs/job/29/52/summary.html

master:
    * 69b46c32357ccf1aab9c0bd6d1afa33a8724ad77
    * 8f61739e0de45ce2ee3be436fc91ef12a1a1c4f3
    * a4ed6c3a56b68f0deb27f291df510c3ba735f836

[Jakub Hrozek]

On Thu, Oct 08, 2015 at 01:49:11PM +0200, Jakub Hrozek wrote:
>  On Thu, Oct 08, 2015 at 12:19:17PM +0200, Jakub Hrozek wrote:
>  > On Thu, Oct 08, 2015 at 09:49:51AM +0200, Lukas Slebodnik wrote:
>  > > On (04/10/15 21:51), Jakub Hrozek wrote:
>  > > >On Fri, Oct 02, 2015 at 11:26:33AM +0200, Lukas Slebodnik wrote:
>  > > >> ehlo,
>  > > >> 
>  > > >> Attached patch should prevent such failures as in commit
>  > > >> 73ec8fdfddb2d4bf99977f758eec80e1b1ee8542
>  > > >>     BUILD: Link test_data_provider_be with -ldl
>  > > >> 
>  > > >> SSS_CRYPT_LIBS should not be part of SSSD_LIBS.
>  > > >> all necessary program and libraries should be linked with
>  > > >> libsss_cryto.so an do not use SSS_CRYPT_LIBS directly.
>  > > >> 
>  > > >> LS
>  > > >
>  > > >Can you add a comment to the definition of SSS_CRYPT_LIBS saying that
>  > > >they shouldn't be used directly, but libsss_cryto should be used
>  > > >instead?
>  > > Is the comment sufficient?
>  > > 
>  > > LS
>  > 
>  > Thanks, that's exactly what I wanted.
>  > 
>  > Code-wise ACK, I'm waiting for CI to finish before pushing.
>  
>  CI passed:
>      http://sssd-ci.duckdns.org/logs/job/29/52/summary.html
>  
>  master:
>      * 69b46c32357ccf1aab9c0bd6d1afa33a8724ad77
>      * 8f61739e0de45ce2ee3be436fc91ef12a1a1c4f3
>      * a4ed6c3a56b68f0deb27f291df510c3ba735f836 
btw unless anyone is opposed, I'd also like to push these patches to
sssd-1-13 -- I'm pretty sure it would help backporting Makefile.am
patches later.

[Lukas Slebodnik]

On (08/10/15 13:53), Jakub Hrozek wrote:
> On Thu, Oct 08, 2015 at 01:49:11PM +0200, Jakub Hrozek wrote:
> > On Thu, Oct 08, 2015 at 12:19:17PM +0200, Jakub Hrozek wrote:
> > > On Thu, Oct 08, 2015 at 09:49:51AM +0200, Lukas Slebodnik wrote:
> > > > On (04/10/15 21:51), Jakub Hrozek wrote:
> > > > >On Fri, Oct 02, 2015 at 11:26:33AM +0200, Lukas Slebodnik wrote:
> > > > >> ehlo,
> > > > >> 
> > > > >> Attached patch should prevent such failures as in commit
> > > > >> 73ec8fdfddb2d4bf99977f758eec80e1b1ee8542
> > > > >>     BUILD: Link test_data_provider_be with -ldl
> > > > >> 
> > > > >> SSS_CRYPT_LIBS should not be part of SSSD_LIBS.
> > > > >> all necessary program and libraries should be linked with
> > > > >> libsss_cryto.so an do not use SSS_CRYPT_LIBS directly.
> > > > >> 
> > > > >> LS
> > > > >
> > > > >Can you add a comment to the definition of SSS_CRYPT_LIBS saying that
> > > > >they shouldn't be used directly, but libsss_cryto should be used
> > > > >instead?
> > > > Is the comment sufficient?
> > > > 
> > > > LS
> > > 
> > > Thanks, that's exactly what I wanted.
> > > 
> > > Code-wise ACK, I'm waiting for CI to finish before pushing.
> > 
> > CI passed:
> >     http://sssd-ci.duckdns.org/logs/job/29/52/summary.html
> > 
> > master:
> >     * 69b46c32357ccf1aab9c0bd6d1afa33a8724ad77
> >     * 8f61739e0de45ce2ee3be436fc91ef12a1a1c4f3
> >     * a4ed6c3a56b68f0deb27f291df510c3ba735f836
>
> btw unless anyone is opposed, I'd also like to push these patches to
> sssd-1-13 -- I'm pretty sure it would help backporting Makefile.am
> patches later. 
Go ahead.

They are mostly cosmetic changes.

LS

[Jakub Hrozek]

On Thu, Oct 08, 2015 at 01:56:38PM +0200, Lukas Slebodnik wrote:
>  On (08/10/15 13:53), Jakub Hrozek wrote:
>  >On Thu, Oct 08, 2015 at 01:49:11PM +0200, Jakub Hrozek wrote:
>  >> On Thu, Oct 08, 2015 at 12:19:17PM +0200, Jakub Hrozek wrote:
>  >> > On Thu, Oct 08, 2015 at 09:49:51AM +0200, Lukas Slebodnik wrote:
>  >> > > On (04/10/15 21:51), Jakub Hrozek wrote:
>  >> > > >On Fri, Oct 02, 2015 at 11:26:33AM +0200, Lukas Slebodnik wrote:
>  >> > > >> ehlo,
>  >> > > >> 
>  >> > > >> Attached patch should prevent such failures as in commit
>  >> > > >> 73ec8fdfddb2d4bf99977f758eec80e1b1ee8542
>  >> > > >>     BUILD: Link test_data_provider_be with -ldl
>  >> > > >> 
>  >> > > >> SSS_CRYPT_LIBS should not be part of SSSD_LIBS.
>  >> > > >> all necessary program and libraries should be linked with
>  >> > > >> libsss_cryto.so an do not use SSS_CRYPT_LIBS directly.
>  >> > > >> 
>  >> > > >> LS
>  >> > > >
>  >> > > >Can you add a comment to the definition of SSS_CRYPT_LIBS saying
> that
>  >> > > >they shouldn't be used directly, but libsss_cryto should be
> used
>  >> > > >instead?
>  >> > > Is the comment sufficient?
>  >> > > 
>  >> > > LS
>  >> > 
>  >> > Thanks, that's exactly what I wanted.
>  >> > 
>  >> > Code-wise ACK, I'm waiting for CI to finish before pushing.
>  >> 
>  >> CI passed:
>  >>     http://sssd-ci.duckdns.org/logs/job/29/52/summary.html
>  >> 
>  >> master:
>  >>     * 69b46c32357ccf1aab9c0bd6d1afa33a8724ad77
>  >>     * 8f61739e0de45ce2ee3be436fc91ef12a1a1c4f3
>  >>     * a4ed6c3a56b68f0deb27f291df510c3ba735f836
>  >
>  >btw unless anyone is opposed, I'd also like to push these patches to
>  >sssd-1-13 -- I'm pretty sure it would help backporting Makefile.am
>  >patches later.
>  Go ahead.
>  
>  They are mostly cosmetic changes. 
OK:
    * 77f801a329ada6fb15c78297190e9d9173a52e4a
    * 2dcf0c023a3bf06f53de21de1c960ac1a5aacaf9
    * a164c1fa79bd4718f29fb18476a3e21230e00f74