| 1 |
On Mon, 2022-06-27 at 14:56 -0400, Mike Gilbert wrote: |
| 2 |
> On Mon, Jun 27, 2022 at 2:35 PM Kenton Groombridge |
| 3 |
> <concord@g.o> wrote: |
| 4 |
> > > so looks like we need to combine both methods and do the |
| 5 |
> > > following: |
| 6 |
> > > - if signing requested without compression - sign in |
| 7 |
> > > pkg_preinst. |
| 8 |
> > > - if signing requested with compression - sign in src_install |
| 9 |
> > > |
| 10 |
> > |
| 11 |
> > Why can't we do both in pkg_preinst? I am thinking it would be best |
| 12 |
> > if |
| 13 |
> > we drop the current compression implementation and rework your old |
| 14 |
> > code |
| 15 |
> > to handle both compression and signing since the signing code is |
| 16 |
> > more or |
| 17 |
> > less already complete. |
| 18 |
> |
| 19 |
> Signing modules in pkg_preinst seems like a bad idea to me. That |
| 20 |
> means |
| 21 |
> you need to copy your private keys around to every host where the |
| 22 |
> package might be installed. |
| 23 |
> |
| 24 |
> If you sign in src_compile or src_install, you only need private keys |
| 25 |
> on the system building your binpkg. |
| 26 |
> |
| 27 |
|
| 28 |
unfortunately portage will unconditionally strip .ko objects, rendering |
| 29 |
modules unloadable by stripping signature, unless we do dostrip -x |
| 30 |
(requires EAPI7+, which should not be a problem nowadays, but was a |
| 31 |
problem back in 2018), which can be quite unfortunate on debug enabled |
| 32 |
kernels. |
Archives