With NETFILTER_XT_TARGET_TEE=y and IP6_NF_IPTABLES=m, we get a link
error when referencing the NF_DUP_IPV6 module:
net/netfilter/xt_TEE.o: In function `tee_tg6':
xt_TEE.c:(.text+0x14): undefined reference to `nf_dup_ipv6'
The problem here is the 'select NF_DUP_IPV6 if IP6_NF_IPTABLES'
that forces NF_DUP_IPV6 to be =m as well rather than setting it
to =y as was intended here. Adding a soft dependency on
IP6_NF_IPTABLES avoids that broken configuration.
Fixes: 5d400a4933e8 ("netfilter: Kconfig: Change select IPv6 dependencies")
Cc: Máté Eckl <ecklm94@xxxxxxxxx>
Cc: Taehee Yoo <ap420073@xxxxxxxxx>
Link: https://patchwork.ozlabs.org/patch/999498/
Link: https://lore.kernel.org/patchwork/patch/960062/
Reported-by: Randy Dunlap <rdunlap@xxxxxxxxxxxxx>
Reported-by: Stephen Rothwell <sfr@xxxxxxxxxxxxxxxx>
Signed-off-by: Arnd Bergmann <arnd@xxxxxxxx>
---
Resending this one, I had sent another version before last July,
and then Máté Eckl did a combined version that I thought was
sufficient, but I had left my old patch in the randconfig
build tree, so I failed to notice that it was still required.
Taehee Yoo also prepared a patch to address this, but that patch
was never merged, and I suspect it did not catch all cases either,
so we still need mine.
---
net/netfilter/Kconfig | 1 +
1 file changed, 1 insertion(+)
diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig
index 964bc317ad0a..ca3ba7548998 100644
--- a/net/netfilter/Kconfig
+++ b/net/netfilter/Kconfig
@@ -1008,6 +1008,7 @@ config NETFILTER_XT_TARGET_TEE
depends on NETFILTER_ADVANCED
depends on IPV6 || IPV6=n
depends on !NF_CONNTRACK || NF_CONNTRACK
+ depends on IP6_NF_IPTABLES || !IP6_NF_IPTABLES
select NF_DUP_IPV4
select NF_DUP_IPV6 if IP6_NF_IPTABLES
---help---
--
2.20.0
