On Tue, 2022-04-05 at 11:18 -0400, Sean Anderson via lists.openembedded.org wrote: > > On 4/5/22 11:02 AM, Mittal, Anuj wrote: > > [You don't often get email from [email protected]. Learn why this is > > important at http://aka.ms/LearnAboutSenderIdentification.] > > > > On Thu, 2022-02-10 at 15:49 -0500, Sean Anderson via > > lists.openembedded.org wrote: > > > libpcap looks for DPDK in the same way it looks for DAG. This can > > > poison > > > the build even if it doesn't find anything. Explicitly disable dpdk. > > > > > > Signed-off-by: Sean Anderson <[email protected]> > > > --- > > > > > > meta/recipes-connectivity/libpcap/libpcap_1.10.1.bb | 3 ++- > > > 1 file changed, 2 insertions(+), 1 deletion(-) > > > > > > diff --git a/meta/recipes-connectivity/libpcap/libpcap_1.10.1.bb > > > b/meta/recipes-connectivity/libpcap/libpcap_1.10.1.bb > > > index 9a8c46e0ef..dbe2fd8157 100644 > > > --- a/meta/recipes-connectivity/libpcap/libpcap_1.10.1.bb > > > +++ b/meta/recipes-connectivity/libpcap/libpcap_1.10.1.bb > > > @@ -19,10 +19,11 @@ BINCONFIG = "${bindir}/pcap-config" > > > > > > # Explicitly disable dag support. We don't have recipe for it and if > > > enabled here, > > > # configure script poisons the include dirs with /usr/local/include > > > even when the > > > -# support hasn't been detected. > > > +# support hasn't been detected. Do the same thing for DPDK. > > > EXTRA_OECONF = " \ > > > --with-pcap=linux \ > > > --without-dag \ > > > + --without-dpdk \ > > > " > > > > Can this be converted to a PACKAGECONFIG instead? There is a DPDK > > recipe in a separate layer so it might be useful for someone including > > that layer. > > There's no distro feature for dpdk. How would we detect if it is enabled? > Additionally, the configure script has hard-coded paths to > /usr/local/include. I think this really needs to be addressed upstream > before we could actually support dpdk.
As you mention, the change merged. We don't need it to be a DISTRO_FEATURE in order for it to be a PACKAGECONFIG so that is a patch we could consider. It would need to be as a follow up patch as the original has merged now (it was a valid/correct change in it's own right). Cheers, Richard
-=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#164048): https://lists.openembedded.org/g/openembedded-core/message/164048 Mute This Topic: https://lists.openembedded.org/mt/89055414/21656 Group Owner: [email protected] Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [[email protected]] -=-=-=-=-=-=-=-=-=-=-=-
