[Buildroot] [PATCH v1 1/2] package/batman-adv: fix compile with BR2_PACKAGE_BATMAN_ADV_BATMAN_V disabled

[Buildroot] [PATCH v1 1/2] package/batman-adv: fix compile with BR2_PACKAGE_BATMAN_ADV_BATMAN_V disabled

[Peter Seiderer]

The given 'CONFIG_BATMAN_ADV_BATMAN_V=' is enough to trigger the wrong
code compile path in net/batman-adv/bat_v.h missing the static inline
dummy implementations.

Fixes:

  ERROR: modpost: "batadv_v_mesh_free" [.../build/batman-adv-2021.0/net/batman-adv/batman-adv.ko] undefined!
  ERROR: modpost: "batadv_v_mesh_init" [.../build/batman-adv-2021.0/net/batman-adv/batman-adv.ko] undefined!
  ERROR: modpost: "batadv_v_hardif_init" [.../build/batman-adv-2021.0/net/batman-adv/batman-adv.ko] undefined!
  ERROR: modpost: "batadv_v_init" [.../build/batman-adv-2021.0/net/batman-adv/batman-adv.ko] undefined!

Signed-off-by: Peter Seiderer <ps.report@gmx.net>
---
 package/batman-adv/batman-adv.mk | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/package/batman-adv/batman-adv.mk b/package/batman-adv/batman-adv.mk
index bd648b5a6c..0b9fcaefc0 100644
--- a/package/batman-adv/batman-adv.mk
+++ b/package/batman-adv/batman-adv.mk
@@ -20,12 +20,12 @@ BATMAN_ADV_MODULE_MAKE_OPTS = \
 	INSTALL_MOD_DIR=updates/net/batman-adv \
 	NOSTDINC_FLAGS="$(BATMAN_ADV_CFLAGS)" \
 	CONFIG_BATMAN_ADV=m \
-	CONFIG_BATMAN_ADV_BATMAN_V=$(BR2_PACKAGE_BATMAN_ADV_BATMAN_V) \
+	$(if $(BR2_PACKAGE_BATMAN_ADV_BATMAN_V), CONFIG_BATMAN_ADV_BATMAN_V=y) \
 	CONFIG_BATMAN_ADV_BLA=y \
 	CONFIG_BATMAN_ADV_DAT=y \
-	CONFIG_BATMAN_ADV_DEBUG=$(BR2_PACKAGE_BATMAN_ADV_DEBUG) \
+	$(if $(BR2_PACKAGE_BATMAN_ADV_DEBUG), CONFIG_BATMAN_ADV_DEBUG=y) \
 	CONFIG_BATMAN_ADV_MCAST=y \
-	CONFIG_BATMAN_ADV_NC=$(BR2_PACKAGE_BATMAN_ADV_NC)
+	$(if $(BR2_PACKAGE_BATMAN_ADV_NC), CONFIG_BATMAN_ADV_NC=y)
 BATMAN_ADV_MODULE_SUBDIRS = net/batman-adv
 
 define BATMAN_ADV_CONFIGURE_CMDS
-- 
2.30.1

[Buildroot] [PATCH v1 2/2] package/batman-adv: enable B.A.T.M.A.N. V protocol by default

[Peter Seiderer]

As upstream enables CONFIG_BATMAN_ADV_BATMAN_V option since
2018.2 by defaullt ([1]) honor this in buildroot and do
the same (and remove the leagcy experimental comment accordingly).

[1] https://git.open-mesh.org/batman-adv.git/blob/HEAD:/CHANGELOG.rst#l174

Signed-off-by: Peter Seiderer <ps.report@gmx.net>
---
 package/batman-adv/Config.in | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/package/batman-adv/Config.in b/package/batman-adv/Config.in
index 9d3d7967b9..f3ac974296 100644
--- a/package/batman-adv/Config.in
+++ b/package/batman-adv/Config.in
@@ -25,11 +25,10 @@ config BR2_PACKAGE_BATMAN_ADV_NC
 	  https://www.open-mesh.org/projects/batman-adv/wiki/NetworkCoding
 
 config BR2_PACKAGE_BATMAN_ADV_BATMAN_V
-	bool "enable B.A.T.M.A.N. V protocol (experimental)"
+	bool "enable B.A.T.M.A.N. V protocol"
+	default y
 	help
 	  Enables B.A.T.M.A.N. V network protocol.
-	  This is the next-gen protocol and is experimental at this
-	  time.
 
 	  https://www.open-mesh.org/projects/batman-adv/wiki/BATMAN_V
 
-- 
2.30.1

[Buildroot] [PATCH v1 2/2] package/batman-adv: enable B.A.T.M.A.N. V protocol by default

[Yann E. MORIN]

Peter, All,

On 2021-03-14 23:32 +0100, Peter Seiderer spake thusly:
> As upstream enables CONFIG_BATMAN_ADV_BATMAN_V option since
> 2018.2 by defaullt ([1]) honor this in buildroot and do
> the same (and remove the leagcy experimental comment accordingly).
> 
> [1] https://git.open-mesh.org/batman-adv.git/blob/HEAD:/CHANGELOG.rst#l174
> 
> Signed-off-by: Peter Seiderer <ps.report@gmx.net>
> ---
>  package/batman-adv/Config.in | 5 ++---
>  1 file changed, 2 insertions(+), 3 deletions(-)
> 
> diff --git a/package/batman-adv/Config.in b/package/batman-adv/Config.in
> index 9d3d7967b9..f3ac974296 100644
> --- a/package/batman-adv/Config.in
> +++ b/package/batman-adv/Config.in
> @@ -25,11 +25,10 @@ config BR2_PACKAGE_BATMAN_ADV_NC
>  	  https://www.open-mesh.org/projects/batman-adv/wiki/NetworkCoding
>  
>  config BR2_PACKAGE_BATMAN_ADV_BATMAN_V
> -	bool "enable B.A.T.M.A.N. V protocol (experimental)"
> +	bool "enable B.A.T.M.A.N. V protocol"

Upstream seems to still consider this to be experimental [0].

> +	default y

What we talked about earlier [1], was to always enable it, not to change
the default.

Now, people that have a defconfig file with BATMAN-V disabled (i.e.
there would be nothing in the defconfig as 'is not set' is the default),
will have it enabled now.

So, if upstream recommends to not disable it (do you have a ref?), then
we should just drop our option, and to always enable it.

Thoughts?

[0] https://git.open-mesh.org/batman-adv.git/blob/HEAD:/Makefile#l7
[1] http://lists.busybox.net/pipermail/buildroot/2021-March/305811.html

Regards,
Yann E. MORIN.

>  	help
>  	  Enables B.A.T.M.A.N. V network protocol.
> -	  This is the next-gen protocol and is experimental at this
> -	  time.
>  
>  	  https://www.open-mesh.org/projects/batman-adv/wiki/BATMAN_V
>  
> -- 
> 2.30.1
> 
> _______________________________________________
> buildroot mailing list
> buildroot at busybox.net
> http://lists.busybox.net/mailman/listinfo/buildroot

-- 
.-----------------.--------------------.------------------.--------------------.
|  Yann E. MORIN  | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: |
| +33 662 376 056 | Software  Designer | \ / CAMPAIGN     |  ___               |
| +33 561 099 427 `------------.-------:  X  AGAINST      |  \e/  There is no  |
| http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL    |   v   conspiracy.  |
'------------------------------^-------^------------------^--------------------'

[Buildroot] [PATCH v1 2/2] package/batman-adv: enable B.A.T.M.A.N. V protocol by default

[Christian Stewart]

Yann,

On Mon, Mar 15, 2021 at 12:59 PM Yann E. MORIN <yann.morin.1998@free.fr> wrote:
> >  config BR2_PACKAGE_BATMAN_ADV_BATMAN_V
> > -     bool "enable B.A.T.M.A.N. V protocol (experimental)"
> > +     bool "enable B.A.T.M.A.N. V protocol"
>
> Upstream seems to still consider this to be experimental [0].
>
> > +     default y
>
> What we talked about earlier [1], was to always enable it, not to change
> the default.
>
> Now, people that have a defconfig file with BATMAN-V disabled (i.e.
> there would be nothing in the defconfig as 'is not set' is the default),
> will have it enabled now.
>
> So, if upstream recommends to not disable it (do you have a ref?), then
> we should just drop our option, and to always enable it.
>
> Thoughts?
>
> [0] https://git.open-mesh.org/batman-adv.git/blob/HEAD:/Makefile#l7
> [1] http://lists.busybox.net/pipermail/buildroot/2021-March/305811.html

What would the advantage be for removing the ability to disable the
CONFIG_BATMAN_ADV_BATMAN_V option from the Buildroot config?

I guess if someone doesn't need it it could still be good to remove to
simplify things.

Best,
Christian

[Buildroot] [PATCH v1 2/2] package/batman-adv: enable B.A.T.M.A.N. V protocol by default

[Peter Seiderer]

Hello Yann,

On Mon, 15 Mar 2021 20:59:00 +0100, "Yann E. MORIN" <yann.morin.1998@free.fr> wrote:

> Peter, All,
>
> On 2021-03-14 23:32 +0100, Peter Seiderer spake thusly:
> > As upstream enables CONFIG_BATMAN_ADV_BATMAN_V option since
> > 2018.2 by defaullt ([1]) honor this in buildroot and do
> > the same (and remove the leagcy experimental comment accordingly).
> >
> > [1] https://git.open-mesh.org/batman-adv.git/blob/HEAD:/CHANGELOG.rst#l174
> >
> > Signed-off-by: Peter Seiderer <ps.report@gmx.net>
> > ---
> >  package/batman-adv/Config.in | 5 ++---
> >  1 file changed, 2 insertions(+), 3 deletions(-)
> >
> > diff --git a/package/batman-adv/Config.in b/package/batman-adv/Config.in
> > index 9d3d7967b9..f3ac974296 100644
> > --- a/package/batman-adv/Config.in
> > +++ b/package/batman-adv/Config.in
> > @@ -25,11 +25,10 @@ config BR2_PACKAGE_BATMAN_ADV_NC
> >  	  https://www.open-mesh.org/projects/batman-adv/wiki/NetworkCoding
> >
> >  config BR2_PACKAGE_BATMAN_ADV_BATMAN_V
> > -	bool "enable B.A.T.M.A.N. V protocol (experimental)"
> > +	bool "enable B.A.T.M.A.N. V protocol"
>
> Upstream seems to still consider this to be experimental [0].

Would count this as a oversight/leftover...and after 3 years of
default to enabled would not count it as experimental...

>
> > +	default y
>
> What we talked about earlier [1], was to always enable it, not to change
> the default.
>
> Now, people that have a defconfig file with BATMAN-V disabled (i.e.
> there would be nothing in the defconfig as 'is not set' is the default),
> will have it enabled now.

I believe the count of people who have BATMAN-V disabled is zero (or they
are to shy to report the compile failure which should occur since the
introduction of BATMAN-V as the ifdef pattern did not change since 2016,
see [2])...

>
> So, if upstream recommends to not disable it (do you have a ref?), then
> we should just drop our option, and to always enable it.
>

Same rationale as Christian in the follow-up mail...

Regards,
Peter

[2] https://git.open-mesh.org/batman-adv.git/blobdiff/ec46535b827539e29628881fa46e5644ec84eba7..a4b88af77e28f90e650c779f673a84d5b8cb5985:/net/batman-adv/bat_algo.h

> Thoughts?
>
> [0] https://git.open-mesh.org/batman-adv.git/blob/HEAD:/Makefile#l7
> [1] http://lists.busybox.net/pipermail/buildroot/2021-March/305811.html
>
> Regards,
> Yann E. MORIN.
>
> >  	help
> >  	  Enables B.A.T.M.A.N. V network protocol.
> > -	  This is the next-gen protocol and is experimental at this
> > -	  time.
> >
> >  	  https://www.open-mesh.org/projects/batman-adv/wiki/BATMAN_V
> >
> > --
> > 2.30.1
> >
> > _______________________________________________
> > buildroot mailing list
> > buildroot at busybox.net
> > http://lists.busybox.net/mailman/listinfo/buildroot
>

Re: [Buildroot] [PATCH v1 2/2] package/batman-adv: enable B.A.T.M.A.N. V protocol by default

[Arnout Vandecappelle]

On 14/03/2021 23:32, Peter Seiderer wrote:
> As upstream enables CONFIG_BATMAN_ADV_BATMAN_V option since
> 2018.2 by defaullt ([1]) honor this in buildroot and do
> the same (and remove the leagcy experimental comment accordingly).
> 
> [1] https://git.open-mesh.org/batman-adv.git/blob/HEAD:/CHANGELOG.rst#l174
> 
> Signed-off-by: Peter Seiderer <ps.report@gmx.net>

  I finally applied to master with just a bit extension of the commit message to 
indicate that upstream's label of it being experimental is no longer really true.

  Regards,
  Arnout

> ---
>   package/batman-adv/Config.in | 5 ++---
>   1 file changed, 2 insertions(+), 3 deletions(-)
> 
> diff --git a/package/batman-adv/Config.in b/package/batman-adv/Config.in
> index 9d3d7967b9..f3ac974296 100644
> --- a/package/batman-adv/Config.in
> +++ b/package/batman-adv/Config.in
> @@ -25,11 +25,10 @@ config BR2_PACKAGE_BATMAN_ADV_NC
>   	  https://www.open-mesh.org/projects/batman-adv/wiki/NetworkCoding
>   
>   config BR2_PACKAGE_BATMAN_ADV_BATMAN_V
> -	bool "enable B.A.T.M.A.N. V protocol (experimental)"
> +	bool "enable B.A.T.M.A.N. V protocol"
> +	default y
>   	help
>   	  Enables B.A.T.M.A.N. V network protocol.
> -	  This is the next-gen protocol and is experimental at this
> -	  time.
>   
>   	  https://www.open-mesh.org/projects/batman-adv/wiki/BATMAN_V
>   
_______________________________________________
buildroot mailing list
buildroot@buildroot.org
https://lists.buildroot.org/mailman/listinfo/buildroot

[Buildroot] [PATCH v1 1/2] package/batman-adv: fix compile with BR2_PACKAGE_BATMAN_ADV_BATMAN_V disabled

[Yann E. MORIN]

Peter, All,

On 2021-03-14 23:32 +0100, Peter Seiderer spake thusly:
> The given 'CONFIG_BATMAN_ADV_BATMAN_V=' is enough to trigger the wrong
> code compile path in net/batman-adv/bat_v.h missing the static inline
> dummy implementations.
> 
> Fixes:
> 
>   ERROR: modpost: "batadv_v_mesh_free" [.../build/batman-adv-2021.0/net/batman-adv/batman-adv.ko] undefined!
>   ERROR: modpost: "batadv_v_mesh_init" [.../build/batman-adv-2021.0/net/batman-adv/batman-adv.ko] undefined!
>   ERROR: modpost: "batadv_v_hardif_init" [.../build/batman-adv-2021.0/net/batman-adv/batman-adv.ko] undefined!
>   ERROR: modpost: "batadv_v_init" [.../build/batman-adv-2021.0/net/batman-adv/batman-adv.ko] undefined!
> 
> Signed-off-by: Peter Seiderer <ps.report@gmx.net>

Applied to master, thanks.

Regards,
Yann E. MORIN.

> ---
>  package/batman-adv/batman-adv.mk | 6 +++---
>  1 file changed, 3 insertions(+), 3 deletions(-)
> 
> diff --git a/package/batman-adv/batman-adv.mk b/package/batman-adv/batman-adv.mk
> index bd648b5a6c..0b9fcaefc0 100644
> --- a/package/batman-adv/batman-adv.mk
> +++ b/package/batman-adv/batman-adv.mk
> @@ -20,12 +20,12 @@ BATMAN_ADV_MODULE_MAKE_OPTS = \
>  	INSTALL_MOD_DIR=updates/net/batman-adv \
>  	NOSTDINC_FLAGS="$(BATMAN_ADV_CFLAGS)" \
>  	CONFIG_BATMAN_ADV=m \
> -	CONFIG_BATMAN_ADV_BATMAN_V=$(BR2_PACKAGE_BATMAN_ADV_BATMAN_V) \
> +	$(if $(BR2_PACKAGE_BATMAN_ADV_BATMAN_V), CONFIG_BATMAN_ADV_BATMAN_V=y) \
>  	CONFIG_BATMAN_ADV_BLA=y \
>  	CONFIG_BATMAN_ADV_DAT=y \
> -	CONFIG_BATMAN_ADV_DEBUG=$(BR2_PACKAGE_BATMAN_ADV_DEBUG) \
> +	$(if $(BR2_PACKAGE_BATMAN_ADV_DEBUG), CONFIG_BATMAN_ADV_DEBUG=y) \
>  	CONFIG_BATMAN_ADV_MCAST=y \
> -	CONFIG_BATMAN_ADV_NC=$(BR2_PACKAGE_BATMAN_ADV_NC)
> +	$(if $(BR2_PACKAGE_BATMAN_ADV_NC), CONFIG_BATMAN_ADV_NC=y)
>  BATMAN_ADV_MODULE_SUBDIRS = net/batman-adv
>  
>  define BATMAN_ADV_CONFIGURE_CMDS
> -- 
> 2.30.1
> 
> _______________________________________________
> buildroot mailing list
> buildroot at busybox.net
> http://lists.busybox.net/mailman/listinfo/buildroot

-- 
.-----------------.--------------------.------------------.--------------------.
|  Yann E. MORIN  | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: |
| +33 662 376 056 | Software  Designer | \ / CAMPAIGN     |  ___               |
| +33 561 099 427 `------------.-------:  X  AGAINST      |  \e/  There is no  |
| http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL    |   v   conspiracy.  |
'------------------------------^-------^------------------^--------------------'

[Buildroot] [PATCH v1 1/2] package/batman-adv: fix compile with BR2_PACKAGE_BATMAN_ADV_BATMAN_V disabled

[Christian Stewart]

Peter, all,

On Sun, Mar 14, 2021 at 3:32 PM Peter Seiderer <ps.report@gmx.net> wrote:
>
> The given 'CONFIG_BATMAN_ADV_BATMAN_V=' is enough to trigger the wrong
> code compile path in net/batman-adv/bat_v.h missing the static inline
> dummy implementations.
>
> Fixes:

Unfortunately after applying this commit it still breaks.

I confirmed that the CONFIG_BATMAN_ADV_BATMAN_V is not being set now.


CONFIG_BATMAN_ADV=m
CONFIG_BATMAN_ADV_BLA=y
CONFIG_BATMAN_ADV_DAT=y
CONFIG_BATMAN_ADV_MCAST=y
M=build/batman-adv-2021.0/net/batman-adv
make modules

[snip]

  MODPOST /build/batman-adv-2021.0/net/batman-adv/Module.symvers
ERROR: modpost: "batadv_v_mesh_free" undefined!
ERROR: modpost: "batadv_v_mesh_init" undefined!
ERROR: modpost: "batadv_v_hardif_init" undefined!
ERROR: modpost: "batadv_v_init" undefined!

Thanks,
Christian Stewart

[Buildroot] [PATCH v1 1/2] package/batman-adv: fix compile with BR2_PACKAGE_BATMAN_ADV_BATMAN_V disabled

[Peter Seiderer]

Hello Christian,

On Mon, 15 Mar 2021 14:00:38 -0700, Christian Stewart <christian@paral.in> wrote:

> Peter, all,
>
> On Sun, Mar 14, 2021 at 3:32 PM Peter Seiderer <ps.report@gmx.net> wrote:
> >
> > The given 'CONFIG_BATMAN_ADV_BATMAN_V=' is enough to trigger the wrong
> > code compile path in net/batman-adv/bat_v.h missing the static inline
> > dummy implementations.
> >
> > Fixes:
>
> Unfortunately after applying this commit it still breaks.
>
> I confirmed that the CONFIG_BATMAN_ADV_BATMAN_V is not being set now.
>
>
> CONFIG_BATMAN_ADV=m
> CONFIG_BATMAN_ADV_BLA=y
> CONFIG_BATMAN_ADV_DAT=y
> CONFIG_BATMAN_ADV_MCAST=y
> M=build/batman-adv-2021.0/net/batman-adv
> make modules
>
> [snip]
>
>   MODPOST /build/batman-adv-2021.0/net/batman-adv/Module.symvers
> ERROR: modpost: "batadv_v_mesh_free" undefined!
> ERROR: modpost: "batadv_v_mesh_init" undefined!
> ERROR: modpost: "batadv_v_hardif_init" undefined!
> ERROR: modpost: "batadv_v_init" undefined!

Mind to share your defconfig (which cross-compiler)?

Did you do a complete re-build of the batman-adv-2021.0
package (rm -rf build/batman-adv-2021.0 && make batman-adv)?

Regards,
Peter

>
> Thanks,
> Christian Stewart
> _______________________________________________
> buildroot mailing list
> buildroot at busybox.net
> http://lists.busybox.net/mailman/listinfo/buildroot

[Buildroot] [PATCH v1 1/2] package/batman-adv: fix compile with BR2_PACKAGE_BATMAN_ADV_BATMAN_V disabled

[Christian Stewart]

Hi Peter,

On Mon, Mar 15, 2021 at 2:50 PM Peter Seiderer <ps.report@gmx.net> wrote:
> On Mon, 15 Mar 2021 14:00:38 -0700, Christian Stewart <christian@paral.in> wrote:
> > On Sun, Mar 14, 2021 at 3:32 PM Peter Seiderer <ps.report@gmx.net> wrote:
> > >
> > > The given 'CONFIG_BATMAN_ADV_BATMAN_V=' is enough to trigger the wrong
> > > code compile path in net/batman-adv/bat_v.h missing the static inline
> > > dummy implementations.
> > >
> > > Fixes:
> >
> > Unfortunately after applying this commit it still breaks.
> >
> > I confirmed that the CONFIG_BATMAN_ADV_BATMAN_V is not being set now.
> >
> >
> > CONFIG_BATMAN_ADV=m
> > CONFIG_BATMAN_ADV_BLA=y
> > CONFIG_BATMAN_ADV_DAT=y
> > CONFIG_BATMAN_ADV_MCAST=y
> > M=build/batman-adv-2021.0/net/batman-adv
> > make modules
> >
> > [snip]
> >
> >   MODPOST /build/batman-adv-2021.0/net/batman-adv/Module.symvers
> > ERROR: modpost: "batadv_v_mesh_free" undefined!
> > ERROR: modpost: "batadv_v_mesh_init" undefined!
> > ERROR: modpost: "batadv_v_hardif_init" undefined!
> > ERROR: modpost: "batadv_v_init" undefined!
>
> Mind to share your defconfig (which cross-compiler)?
>
> Did you do a complete re-build of the batman-adv-2021.0
> package (rm -rf build/batman-adv-2021.0 && make batman-adv)?

Buildroot toolchain,

BR2_x86_64=y
BR2_TOOLCHAIN=y
BR2_TOOLCHAIN_USES_GLIBC=y
BR2_TOOLCHAIN_BUILDROOT=y
BR2_BINUTILS_VERSION_2_35_X=y
BR2_GCC_VERSION_10_X=y
BR2_PACKAGE_BATMAN_ADV=y

Yes, deleted the build/batman-adv* dir and fully re-built, and
confirmed that it was not setting the BATMAN_ADV_V flag anymore.

Thanks,
Christian

[Buildroot] [PATCH v1 1/2] package/batman-adv: fix compile with BR2_PACKAGE_BATMAN_ADV_BATMAN_V disabled

[Peter Seiderer]

Hello Christian,

On Mon, 15 Mar 2021 18:45:20 -0700, Christian Stewart <christian@paral.in> wrote:

> Hi Peter,
>
> On Mon, Mar 15, 2021 at 2:50 PM Peter Seiderer <ps.report@gmx.net> wrote:
> > On Mon, 15 Mar 2021 14:00:38 -0700, Christian Stewart <christian@paral.in> wrote:
> > > On Sun, Mar 14, 2021 at 3:32 PM Peter Seiderer <ps.report@gmx.net> wrote:
> > > >
> > > > The given 'CONFIG_BATMAN_ADV_BATMAN_V=' is enough to trigger the wrong
> > > > code compile path in net/batman-adv/bat_v.h missing the static inline
> > > > dummy implementations.
> > > >
> > > > Fixes:
> > >
> > > Unfortunately after applying this commit it still breaks.
> > >
> > > I confirmed that the CONFIG_BATMAN_ADV_BATMAN_V is not being set now.
> > >
> > >
> > > CONFIG_BATMAN_ADV=m
> > > CONFIG_BATMAN_ADV_BLA=y
> > > CONFIG_BATMAN_ADV_DAT=y
> > > CONFIG_BATMAN_ADV_MCAST=y
> > > M=build/batman-adv-2021.0/net/batman-adv
> > > make modules
> > >
> > > [snip]
> > >
> > >   MODPOST /build/batman-adv-2021.0/net/batman-adv/Module.symvers
> > > ERROR: modpost: "batadv_v_mesh_free" undefined!
> > > ERROR: modpost: "batadv_v_mesh_init" undefined!
> > > ERROR: modpost: "batadv_v_hardif_init" undefined!
> > > ERROR: modpost: "batadv_v_init" undefined!
> >
> > Mind to share your defconfig (which cross-compiler)?
> >
> > Did you do a complete re-build of the batman-adv-2021.0
> > package (rm -rf build/batman-adv-2021.0 && make batman-adv)?
>
> Buildroot toolchain,
>
> BR2_x86_64=y
> BR2_TOOLCHAIN=y
> BR2_TOOLCHAIN_USES_GLIBC=y
> BR2_TOOLCHAIN_BUILDROOT=y
> BR2_BINUTILS_VERSION_2_35_X=y
> BR2_GCC_VERSION_10_X=y
> BR2_PACKAGE_BATMAN_ADV=y
>
> Yes, deleted the build/batman-adv* dir and fully re-built, and
> confirmed that it was not setting the BATMAN_ADV_V flag anymore.

O.k, using the following defconfig I get the same failure:

BR2_x86_64=y
BR2_GCC_VERSION_10_X=y
BR2_LINUX_KERNEL=y
BR2_LINUX_KERNEL_USE_ARCH_DEFAULT_CONFIG=y
BR2_PACKAGE_BATMAN_ADV=y

My RPi4 test case had an RPi4 linux kernel already enabling the build-in
batman-adv module inclusive batman-v...

Hopefully fixed this time for real with

https://patchwork.ozlabs.org/project/buildroot/patch/20210316181429.30804-1-ps.report at gmx.net/
https://patchwork.ozlabs.org/project/buildroot/patch/20210316181429.30804-2-ps.report at gmx.net/

and a note about the linux mainline kernel module added with

https://patchwork.ozlabs.org/project/buildroot/patch/20210316181429.30804-3-ps.report at gmx.net/

Regards,
Peter

>
> Thanks,
> Christian
> _______________________________________________
> buildroot mailing list
> buildroot at busybox.net
> http://lists.busybox.net/mailman/listinfo/buildroot

[Buildroot] [PATCH v1 1/2] package/batman-adv: fix compile with BR2_PACKAGE_BATMAN_ADV_BATMAN_V disabled

[Peter Korsgaard]

>>>>> "Peter" == Peter Seiderer <ps.report@gmx.net> writes:

 > The given 'CONFIG_BATMAN_ADV_BATMAN_V=' is enough to trigger the wrong
 > code compile path in net/batman-adv/bat_v.h missing the static inline
 > dummy implementations.

 > Fixes:

 >   ERROR: modpost: "batadv_v_mesh_free" [.../build/batman-adv-2021.0/net/batman-adv/batman-adv.ko] undefined!
 >   ERROR: modpost: "batadv_v_mesh_init" [.../build/batman-adv-2021.0/net/batman-adv/batman-adv.ko] undefined!
 >   ERROR: modpost: "batadv_v_hardif_init" [.../build/batman-adv-2021.0/net/batman-adv/batman-adv.ko] undefined!
 >   ERROR: modpost: "batadv_v_init" [.../build/batman-adv-2021.0/net/batman-adv/batman-adv.ko] undefined!

 > Signed-off-by: Peter Seiderer <ps.report@gmx.net>

Committed to 2020.02.x, 2020.11.x and 2021.02.x, thanks.

-- 
Bye, Peter Korsgaard