On 2021-10-14 18:30, Drew Parsons wrote:
On 2021-10-14 17:52, Julien Lamy wrote:
I don't think it's purely related to gtest: the test programs
(benchmark and Mandelbrot set) don't use it, and also fail to compile.
Regarding the arch limitation, Drew Parsons suggested earlier [2] that
build failures on a never-uploaded package wasn't really a problem: we
could keep the failing architectures so that, when the situation
improves with a future upstream version, they would be packaged
automatically.
We'll see in a few days if migration to testing is that simple :p
Evidently not that simple. Migration seems to be upset about the
failing builds. The case of permitted migration I was thinking of must
only apply if successful arches are already in testing. Not sure
entirely.
Looks like xsimd 8 has made changes which might be relevant. Hopefully
it will enable more arches to pass.
Drew
