linux-kernel - Re: [GIT PULL] non-MM updates for 6.6-rc1

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: <20230830225335.kmdj5er253fr7rk4@treble>
Date:   Wed, 30 Aug 2023 15:53:35 -0700
From:   Josh Poimboeuf <jpoimboe@...nel.org>
To:     Linus Torvalds <torvalds@...ux-foundation.org>
Cc:     Andrew Morton <akpm@...ux-foundation.org>, linux-mm@...ck.org,
        mm-commits@...r.kernel.org, linux-kernel@...r.kernel.org,
        x86@...nel.org
Subject: Re: [GIT PULL] non-MM updates for 6.6-rc1

On Tue, Aug 29, 2023 at 03:20:37PM -0700, Linus Torvalds wrote:
> On Mon, 28 Aug 2023 at 22:54, Andrew Morton <akpm@...ux-foundation.org> wrote:
> >
> > Linus, please pull mm.git's non-MM updates for this cycle.
> 
> Hmm. For some non-obvious (to me) reason this triggers
> 
>   vmlinux.o: warning: objtool: ibt_selftest+0x18: sibling call from
> callable instruction with modified stack frame
> 
> which didn't happen before, even though the pull doesn't actually
> change that file at all.
> 
> I see that Josh has a patch to fix this by making it be real assembly
> code rather than a hacky inline asm in a C file, but it's a bit
> surprising, and I don't see why it's triggered now when it wasn't
> before.
> 
> At a guess it's related to the Kconfig changes, but I have nothing to
> back that guess up with.
> 
> Slightly annoying, but I guess I'll live with the warning for now.

As far as I can tell this warning only shows up with the rare
combination of CONFIG_X86_KERNEL_IBT and CONFIG_UNWINDER_FRAME_POINTER
(the latter being pretty much deprecated at this point).

Did you manage to enable frame pointers somehow?

The fix (to rewrite the IBT selftest) was merged into tip/x86/shstk.
Presumably it'll be in a pull request soon-ish.

-- 
Josh