Todd Zullinger writes:
Jerry James wrote:
> On Mon, Aug 3, 2020 at 4:25 PM Sam Varshavchik <mrsam(a)courier-mta.com>
wrote:
>> I'm running an F32 aarch64 guest qemu VM on an x86_64 F32 host, and, in
that
>> VM, I attempted to build something fairly simple via mock. This is my
reward:
>>
>> configure:3590: gcc -O2 -g -pipe -Wall -Werror=format-security -Wp,-
>> D_FORTIFY_SOURCE=2 -Wp,-D_GLIBCXX_ASSERTIONS -fexceptions -fstack-
protector-
>> strong -grecord-gcc-switches -specs=/usr/lib/rpm/redhat/redhat-hardened-
cc1 -
>> specs=/usr/lib/rpm/redhat/redhat-annobin-cc1 -fasynchronous-unwind-tables
-
>> fstack-clash-protection -Wl,-z,relro -Wl,--as-needed -Wl,-z,now -
>> specs=/usr/lib/rpm/redhat/redhat-hardened-ld conftest.c >&5
>>
>> annobin: conftest.c: AArch64: The annobin plugin is out of date with
respect
>> to gcc
>> *** WARNING *** there are active plugins, do not report this as a bug
unless
>> you can reproduce it without enabling any plugins.
>> Event | Plugins
>> PLUGIN_FINISH_UNIT | annobin: Generate final annotations
>> PLUGIN_START_UNIT | annobin: Generate global annotations
>> PLUGIN_ALL_PASSES_START | annobin: Generate per-function
annotations
>> PLUGIN_ALL_PASSES_END | annobin: Register per-function end
symbol
>> conftest.c:1: internal compiler error: Segmentation fault
>>
>> First time I've seen something like that, any idea what to do here, for
this?
>> This if from a first build in mock, which presumably pulled in the most
>> recent versions of all rpms into the chroot.
>
> Are you building for a rawhide target? If so, mock is pulling in rpms
> from the most recent rawhide compose. Add "--enablerepo=local" to get
> the very latest rpms from koji.
>
> There have been a fewl instances recently of the annobin plugin not
> matching gcc. As far as I know, the issue has been resolved, so
> hopefully the extra mock flag will work for you.
I think the same gcc version from rawhide was pushed to f32,
while the corresponding annonbin release (9.27-1.fc32) is
still in updates-testing¹. So it's possible this comes up
even for an f32 target.
Yup, this is in stable:
root@arm mock]# ls -al default.cfg
lrwxrwxrwx. 1 root root 21 Jul 28 13:51 default.cfg → fedora-32-aarch64.cfg
In that case, enabling either the
updates-testing or the 'local' repo for the mock build
target would probably help.
Karma for the annobin release would likely be appreciated by
the maintainer(s) and anyone else who's running into this
problem, if you are able to test and confirm it fixes the
issue(s). :)
¹
https://bodhi.fedoraproject.org/updates/FEDORA-2020-db7719f642
https://bugzilla.redhat.com/1862297
Yup, that update fixes my mock builds, and is on its way to stable.