Hi Adrian,
I'd just like to kindly ask whether you have had the time to look into
this? See below:
[...]
>>> I think the clashing symbols might be emitted indirectly by the Rust
>>> compiler in libsuricata.a (as libsuricata.a is just the artifact for the
>>> part of Suricata written in Rust) and will only cause a clash when
>>> trying to link the Rust code with the rest of the Suricata C code (as
>>> libgcc provides these as well).
>>
>> I do remember that we explicitly undefined the atomic functions for armel
>> in the rustc package because we were running into the same problem.
>
> I see.
>
>> I have to look up what we did back then. It must be in the rustc source
>> package.
I did some searching in git and found #891902 [1] which was addressed in
1.24.1+dfsg1-1~exp1:
https://salsa.debian.org/rust-team/rust/commit/f25948bb2cb2e66efd4898ba1ec9ac0637ed4252#0558e67dd6ff049e04d97512848b405fea2d54e7
However, this was removed again later with a reference to [2]:
rustc (1.29.0+dfsg1-1) unstable; urgency=medium
...
* Drop d-armel-disable-kernel-helpers.patch as a necessary part of the
fix to #906520, so it is actually fixed.
...
-- Ximin Luo <infinity0@debian.org> Sun, 23 Sep 2018 10:16:53 -0700
So I guess this is the reason why we're seeing these build failures [3] now?
Id be happy to hear any hints or ideas? I'm not too familiar with this
codebase but would be reluctant to just drop armel from the list of
supported archs for suricata...
Thanks!
Cheers
Sascha
[1] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=891902
[2] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=906520
[3]
https://buildd.debian.org/status/fetch.php?pkg=suricata&arch=armel&ver=1%3A5.0.0-1%7Eexp2&stamp=1572536070&raw=0
Attachment:
signature.asc
Description: OpenPGP digital signature