On Fri, Feb 13, 2015 at 11:07:27PM +0100, Bjørn Forsman wrote: > On 13 February 2015 at 22:20, Lluís Batlle i Rossell <[email protected]> wrote: > > Thank you! Is the plan for failing libsigsegv to wait for > > kernel.org fixes? > > What's the connection between libsigsegv and kernel.org? I noticed > libsigsegv test failures the other day (I was playing with armv7 > bootstrap), and I bisected the error to this commit: > > 88fd7f452703f3280a3747e97b938c8e50ae962c (Re-Revert merge #5505: > util-linux: use /run/current-system/sw/bin/*) > > I was planning to make a gihub issue or post on nix-dev ML, but didn't > get to it before now :-)
I think that linus changed the signal of stack overflow from SIGSEGV to SIGBUS. At some point, people realized it was a bad idea, and they took the change back in recent kernels. But if you bisected it... no idea. :) That change you found is quite weird. _______________________________________________ nix-dev mailing list [email protected] http://lists.science.uu.nl/mailman/listinfo/nix-dev
