Discussion:
[libseccomp-discuss] [QEMU] Build of v2.1.0 fails on armv7l due to undeclared __NR_select
Eduardo Otubo
2014-10-21 09:47:07 UTC
Permalink
Hello all,

Qemu is failing to build on armv7l arch according to this launchpad
entry[0]. I'm not quite sure if this is the case of adapting
libseccomp instead of creating ifdef's at qemu level. How hard would
it be to add new specific arch (arm 32 bits) into libseccomp now?

Regards,

[0] https://bugs.launchpad.net/qemu/+bug/1363641
--
Eduardo Otubo
ProfitBricks
Paul Moore
2014-10-21 20:13:00 UTC
Permalink
Post by Eduardo Otubo
Hello all,
Qemu is failing to build on armv7l arch according to this launchpad
entry[0]. I'm not quite sure if this is the case of adapting
libseccomp instead of creating ifdef's at qemu level. How hard would
it be to add new specific arch (arm 32 bits) into libseccomp now?
Regards,
[0] https://bugs.launchpad.net/qemu/+bug/1363641
We have support for 32-bit ARM in the libseccomp v2.1.0 release, although
there are some bugs in that release as you've seen. The solution would be to
either backport the required patches or wait for a new release. How
significant is this problem for you?
--
paul moore
security and virtualization @ redhat
Paul Moore
2014-10-21 22:26:02 UTC
Permalink
-- otubo
Post by Paul Moore
Post by Eduardo Otubo
Hello all,
Qemu is failing to build on armv7l arch according to this launchpad
entry[0]. I'm not quite sure if this is the case of adapting
libseccomp instead of creating ifdef's at qemu level. How hard would
it be to add new specific arch (arm 32 bits) into libseccomp now?
Regards,
[0] https://bugs.launchpad.net/qemu/+bug/1363641
We have support for 32-bit ARM in the libseccomp v2.1.0 release, although
there are some bugs in that release as you've seen. The solution would be
to either backport the required patches or wait for a new release. How
significant is this problem for you?
Not exactly top priority, so I can wait for the next release. When is it
going to be?
Good question ...
In the meanwhile I'll just leave a comment at launchpad saying the problem
has already been fixed in upstream. Is that ok?
Yep, thanks.
--
paul moore
security and virtualization @ redhat
Loading...