Hi Nilesh,
thanks! > On 12/28/21 5:36 PM, Martin Uecker wrote: > > Hi all, > > this is caused by some compiler bug in GCC 11. Any advice how > to proceed? > > - use GCC 11 (I meant GCC 10) > If this really is a bug in the _compiler itself_, maybe simply > file a bug report against gcc-11 itself? > Please do this regardless of the solution ^^ Yes, I already did this. (Although not too useful without access to the archs because I can't provide a minimized test case). > - exclude affected architectures? > > It looks unlikely for someone to use bart-view on these archs, so > this is an acceptable solution as well, i.e. filing a removal bug on specific > archs. > It however takes time since FTP masters will manually process it. > - try to work around by patching? (would be easy > only with access to the affected architectures) > > This would be best. You could try with qemu to emulate these archs. I guess this is what I will try. > If it does not work, you can apply for guest access on these machines as > given here[1] Last time I tried this I never got a useful response. > - talk to the release team? > > They can't do much here. Maybe the could simply not remove bart-view if the failure to update on some archs is caused by the compiler? Martin > [1]: https://dsa.debian.org/doc/guest-account/ > > Regards, > Nilesh
