Simon McVittie wrote:
> On 26/08/15 00:52, Simon McVittie wrote:
> > If it's any help, this is reproducible on minkus.debian.org (mips, Octeon),
> > but is not reproducible with DEB_BUILD_OPTIONS=noopt on the same machine.
> > So it might be possible to coax it into working (well enough to get
> > through this transition) by reducing or omitting optimization on mips.
>
> The same thing happened on eder: a build with noopt passes its tests.
>
> To keep the g++-5 transition moving, I'm tempted to NMU a version of
> protobuf that builds with -O0 on mips{,el}, and drop this bug down to
> important severity; and then the mips porters and the protobuf
> maintainers can work out what is actually going on and do a more
> targeted fix, without stalling various transitions in the meantime. Does
> that seem a reasonable plan?
Hi,
This sounds very vaguely similar to #572923, for which the workaround
was to disable optimization of a single function:
https://sources.debian.net/src/protobuf/2.3.0-4/debian/patches/arm_optimization.diff/
I'm OK with a NMU that disables optimization of the entire build on
mips{,el} for now, though, if it will keep the GCC-5 transition moving
along. Please feel free to upload to incoming rather than DELAYED.
Thanks for looking into this!
--
Robert Edmonds
edmo...@debian.org
