linux-kernel - Re: [PATCH] init/main.c: Initialize early LSMs after arch code

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

Message-ID: <CACYkzJ5J8K2D8xqT+CCrbvp57P=GbCB+XYXkAaKXojsFhuaWEw@mail.gmail.com>
Date: Tue, 13 Aug 2024 00:02:42 +0200
From: KP Singh <kpsingh@...nel.org>
To: Paul Moore <paul@...l-moore.com>
Cc: Guenter Roeck <linux@...ck-us.net>, Nathan Chancellor <nathan@...nel.org>, linux-kernel@...r.kernel.org, 
	linux-security-module@...r.kernel.org, bp@...en8.de, sfr@...b.auug.org.au, 
	peterz@...radead.org, ink@...assic.park.msu.ru, richard.henderson@...aro.org
Subject: Re: [PATCH] init/main.c: Initialize early LSMs after arch code

On Mon, Aug 12, 2024 at 11:33 PM Paul Moore <paul@...l-moore.com> wrote:
>
> On Mon, Aug 12, 2024 at 5:14 PM KP Singh <kpsingh@...nel.org> wrote:
> > On Mon, Aug 12, 2024 at 9:33 PM Paul Moore <paul@...l-moore.com> wrote:
> > > On Mon, Aug 12, 2024 at 1:12 PM KP Singh <kpsingh@...nel.org> wrote:
> > > >
> > > > JFYI, I synced with Guenter and all arch seem to pass and alpha does
> > > > not work due to a reason that I am unable to debug. I will try doing
> > > > more debugging but I will need more alpha help here (Added the
> > > > maintainers to this thread).
> > >
> > > Thanks for the update; I was hoping that we might have a resolution
> > > for the Alpha failure by now but it doesn't look like we're that
> > > lucky.  Hopefully the Alpha devs will be able to help resolve this
> > > without too much trouble.
> > >
> > > Unfortunately, this does mean that I'm going to drop the static call
> > > patches from the lsm/dev branch so that we can continue merging other
> > > things.  Of course this doesn't mean the static call patches can't
> > > come back in later during this dev cycle once everything is solved if
> > > there is still time, and worst case there is always the next dev
> > > cycle.
> > >
> >
> > Do we really want to drop them for alpha? I would rather disable
> > CONFIG_SECURITY for alpha and if people really care for alpha we can
> > enable it. Alpha folks, what do you think?
>
> Seriously?  I realize Alpha is an older, lesser used arch, but it is
> still a supported arch and we are not going to cause a regression for
> the sake of a new feature.  As I mentioned earlier, once the problem
> is resolved we can bring the patchset back into lsm/dev; if it gets
> resolved soon enough we can even do it during this dev cycle.
>

Okay, more data for the alpha folks, when I moved trap_init() before
early_security_init() everything seemed to work, I think we might need
to call trap_init() from setup_arch and this would fix the issue. As
to why? I don't know :)

Would alpha folks be okay with this patch:

kpsingh@...ingh:~/projects/linux$ git diff
diff --git a/arch/alpha/kernel/setup.c b/arch/alpha/kernel/setup.c
index bebdffafaee8..53909c1be4cf 100644
--- a/arch/alpha/kernel/setup.c
+++ b/arch/alpha/kernel/setup.c
@@ -657,6 +657,7 @@ setup_arch(char **cmdline_p)
        setup_smp();
 #endif
        paging_init();
+       trap_init();
 }


and provide me some reason as to why this works, it would be great for
a patch description

- KP





> --
> paul-moore.com