@veprbl, thanks for your PR! By analyzing the history of the files in this pull request, we identified @Ericson2314, @peti and @edolstra to be potential reviewers.

Could you use camelCase? Otherwise looks fine to me.

This is not a mass rebuild, only few old compilers affected. Can we merge?

It would be good to use this for gccCrossStageStatic, which I think cannot do some of this stuff. That would just be adding some stdenv.lib.optionals crossStageStage [ .. ] to the GCC derivations. @bnikolic what do you think, since you just hit this in #33167?

@Ericson2314 agreed, would be useful for various cross compiling situations.

Ideally the compiler toolchains should be inteligently ignoring the hardening flags but they don't seem to do that. So hardening should fall into a category of non-functional build attributes where the builders should try to do as many of that as possible but ignore any which can not be achieved.

@bnikolic Well, I wouldn't want it to be too seemless less we accidentally end up with less clean machine code. Making the opt-out static and determinable from Nix, not just bash/builders themselves, seems to me like a sweet spot.

Do you agree that no SPP seems to be a stage-static thing in generally, or mingw stage static thing in particular?

@Ericson2314 I'm not sure. I think it depends on whether SSP is in the glibc or as separate library but haven't look in a lot of detail -- will keep this in mind.