On Tue, Jun 10, 2014 at 07:54:26AM +0100, Richard W.M. Jones wrote:
On Mon, Jun 09, 2014 at 10:20:46PM +0100, Matthew Garrett wrote:
> On Mon, Jun 09, 2014 at 08:43:07PM +0100, Richard W.M. Jones wrote:
>
> > Can we excludearch %{arm} for this one?
>
> Why? It's a bug that it doesn't build on ARM. Refusing to build it
> doesn't fix the bug, and then someone else will crash into the same
> issue when they dare to build something that needs llvm.
It seems the alternative is hfsplus-tools doesn't work at all for
anyone.
Eh. We're constrained by our own policies here, not by anything
fundamental - LLVM being broken on ARM ought to mean that our ARM
product is worse, not that everything else is dragged down to the same
level.
BTW this package is also affected by the -fstack-protector-strong
LLVM
bug, but that is simple to work around.
Yeah that complaint is completely fair.
--
Matthew Garrett | mjg59(a)srcf.ucam.org