linux-kernel - Re: linux-next: build failure after merge of the keys tree

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date:   Thu, 14 May 2020 17:25:13 -0600
From:   "Jason A. Donenfeld" <Jason@...c4.com>
To:     Masahiro Yamada <masahiroy@...nel.org>
Cc:     David Howells <dhowells@...hat.com>,
        Stephen Rothwell <sfr@...b.auug.org.au>,
        Linux Next Mailing List <linux-next@...r.kernel.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: linux-next: build failure after merge of the keys tree

On Thu, May 14, 2020 at 6:35 AM Masahiro Yamada <masahiroy@...nel.org> wrote:
>
> On Thu, May 14, 2020 at 9:11 PM David Howells <dhowells@...hat.com> wrote:
> >
> > Jason A. Donenfeld <Jason@...c4.com> wrote:
> >
> > > Your touch might be helpful here. CRYPTO_LIB_CHACHA20POLY1305 is a
> > > tristate and depends on as well as selects other things that are
> > > tristates.
> > >
> > > Meanwhile BIG_KEYS is a bool, which needs to select
> > > CRYPTO_LIB_CHACHA20POLY1305. However, it gets antsy if the the symbol
> > > its selecting has =m items in its hierarchy.
> > >
> > > Any suggestions? The ideal thing to happen would be that the select of
> > > CRYPTO_LIB_CHACHA20POLY1305 in BIG_KEYS causes all of the descendants
> > > to become =y too.
> >
> > I think that select is broken in its behaviour - it doesn't propagate the
> > selection enforcement up the tree.  You could try changing it to a depends on
> > or you could put in a select for every dependency.
>
> I agree.
> 'depends on' will be cleaner.

That's fine, but also makes it more annoying for people to select
big_keys, and I don't know how David feels in that regard.

Seems like it'd be useful to have something that means "select X and
all the things X needs to not be broken", though satisfiability
problems like that can get really complicated quite fast.