linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Sami Tolvanen <samitolvanen@google.com>
To: Kees Cook <keescook@chromium.org>
Cc: Nathan Chancellor <nathan@kernel.org>,
	Nick Desaulniers <ndesaulniers@google.com>,
	Masahiro Yamada <masahiroy@kernel.org>,
	Will Deacon <will@kernel.org>, Jessica Yu <jeyu@kernel.org>,
	Arnd Bergmann <arnd@arndb.de>, Tejun Heo <tj@kernel.org>,
	bpf@vger.kernel.org, linux-hardening@vger.kernel.org,
	linux-arch <linux-arch@vger.kernel.org>,
	linux-arm-kernel <linux-arm-kernel@lists.infradead.org>,
	linux-kbuild <linux-kbuild@vger.kernel.org>,
	PCI <linux-pci@vger.kernel.org>,
	LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 04/17] module: cfi: ensure __cfi_check alignment
Date: Tue, 16 Mar 2021 13:03:45 -0700	[thread overview]
Message-ID: <CABCJKucW8dwXzEK-NZEHyQzN1wu-sGOiOqXFfR=C-PHvsn1+Dg@mail.gmail.com> (raw)
In-Reply-To: <202103111837.813997B4@keescook>

On Thu, Mar 11, 2021 at 6:39 PM Kees Cook <keescook@chromium.org> wrote:
>
> On Thu, Mar 11, 2021 at 04:49:06PM -0800, Sami Tolvanen wrote:
> > CONFIG_CFI_CLANG_SHADOW assumes the __cfi_check() function is page
> > aligned and at the beginning of the .text section. While Clang would
> > normally align the function correctly, it fails to do so for modules
> > with no executable code.
> >
> > This change ensures the correct __cfi_check() location and
> > alignment. It also discards the .eh_frame section, which Clang can
> > generate with certain sanitizers, such as CFI.
> >
> > Link: https://bugs.llvm.org/show_bug.cgi?id=46293
> > Signed-off-by: Sami Tolvanen <samitolvanen@google.com>
> > ---
> >  scripts/module.lds.S | 14 +++++++++++++-
> >  1 file changed, 13 insertions(+), 1 deletion(-)
> >
> > diff --git a/scripts/module.lds.S b/scripts/module.lds.S
> > index 168cd27e6122..552ddb084f76 100644
> > --- a/scripts/module.lds.S
> > +++ b/scripts/module.lds.S
> > @@ -3,10 +3,13 @@
> >   * Archs are free to supply their own linker scripts.  ld will
> >   * combine them automatically.
> >   */
> > +#include <asm/page.h>
> > +
> >  SECTIONS {
> >       /DISCARD/ : {
> >               *(.discard)
> >               *(.discard.*)
> > +             *(.eh_frame)
> >       }
> >
> >       __ksymtab               0 : { *(SORT(___ksymtab+*)) }
> > @@ -40,7 +43,16 @@ SECTIONS {
> >               *(.rodata..L*)
> >       }
> >
> > -     .text : { *(.text .text.[0-9a-zA-Z_]*) }
> > +#ifdef CONFIG_CFI_CLANG
> > +     /*
> > +      * With CFI_CLANG, ensure __cfi_check is at the beginning of the
> > +      * .text section, and that the section is aligned to page size.
> > +      */
> > +     .text : ALIGN(PAGE_SIZE) {
> > +             *(.text.__cfi_check)
> > +             *(.text .text.[0-9a-zA-Z_]* .text..L.cfi*)
> > +     }
> > +#endif
>
> Whoops, I think this reverts to the default .text declaration when
> CONFIG_CFI_CLANG is unset.

Oops, thanks for pointing this out. I'll fix it in v2.

> I think the only thing that needs the ifdef is the ALIGN, yes? Perhaps
> something like this?
>
> #ifdef CONFIG_CFI_CLANG
> # define ALIGN_CFI ALIGN(PAGE_SIZE)
> #else
> # define ALIGN_CFI
> #endif
>
>         .text : ALIGN_CFI {
>                 *(.text.__cfi_check)
>                 *(.text .text.[0-9a-zA-Z_]* .text..L.cfi*)
>         }

Agreed, that looks better.

Sami

  reply	other threads:[~2021-03-16 20:04 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-03-12  0:49 [PATCH 00/17] Add support for Clang CFI Sami Tolvanen
2021-03-12  0:49 ` [PATCH 01/17] add " Sami Tolvanen
2021-03-12  2:39   ` Kees Cook
2021-03-12  0:49 ` [PATCH 02/17] cfi: add __cficanonical Sami Tolvanen
2021-03-12  2:40   ` Kees Cook
2021-03-12 20:28   ` Bjorn Helgaas
2021-03-12  0:49 ` [PATCH 03/17] mm: add generic __va_function and __pa_function macros Sami Tolvanen
2021-03-12  2:40   ` Kees Cook
2021-03-12  0:49 ` [PATCH 04/17] module: cfi: ensure __cfi_check alignment Sami Tolvanen
2021-03-12  2:39   ` Kees Cook
2021-03-16 20:03     ` Sami Tolvanen [this message]
2021-03-12  0:49 ` [PATCH 05/17] workqueue: cfi: disable callback pointer check with modules Sami Tolvanen
2021-03-12  2:43   ` Kees Cook
2021-03-12  0:49 ` [PATCH 06/17] kthread: " Sami Tolvanen
2021-03-12  2:43   ` Kees Cook
2021-03-12  6:13   ` Christoph Hellwig
2021-03-17 16:05     ` Sami Tolvanen
2021-03-12  0:49 ` [PATCH 07/17] kallsyms: cfi: strip hashes from static functions Sami Tolvanen
2021-03-12  2:45   ` Kees Cook
2021-03-16 20:33     ` Sami Tolvanen
2021-03-12  0:49 ` [PATCH 08/17] bpf: disable CFI in dispatcher functions Sami Tolvanen
2021-03-12  2:45   ` Kees Cook
2021-03-12  0:49 ` [PATCH 09/17] lib/list_sort: fix function type mismatches Sami Tolvanen
2021-03-12  2:45   ` Kees Cook
2021-03-12  0:49 ` [PATCH 10/17] lkdtm: use __va_function Sami Tolvanen
2021-03-12  2:45   ` Kees Cook
2021-03-12  0:49 ` [PATCH 11/17] psci: use __pa_function for cpu_resume Sami Tolvanen
2021-03-12  2:45   ` Kees Cook
2021-03-12  0:49 ` [PATCH 12/17] arm64: implement __va_function Sami Tolvanen
2021-03-12  2:46   ` Kees Cook
2021-03-12  0:49 ` [PATCH 13/17] arm64: use __pa_function Sami Tolvanen
2021-03-12  2:46   ` Kees Cook
2021-03-12  0:49 ` [PATCH 14/17] arm64: add __nocfi to functions that jump to a physical address Sami Tolvanen
2021-03-12  2:47   ` Kees Cook
2021-03-12  0:49 ` [PATCH 15/17] arm64: add __nocfi to __apply_alternatives Sami Tolvanen
2021-03-12  2:50   ` Kees Cook
2021-03-12  0:49 ` [PATCH 16/17] KVM: arm64: Disable CFI for nVHE Sami Tolvanen
2021-03-12  2:50   ` Kees Cook
2021-03-12  0:49 ` [PATCH 17/17] arm64: allow CONFIG_CFI_CLANG to be selected Sami Tolvanen
2021-03-12  2:51   ` Kees Cook
2021-03-16 20:44     ` Sami Tolvanen
2021-03-16 23:02       ` Kees Cook

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='CABCJKucW8dwXzEK-NZEHyQzN1wu-sGOiOqXFfR=C-PHvsn1+Dg@mail.gmail.com' \
    --to=samitolvanen@google.com \
    --cc=arnd@arndb.de \
    --cc=bpf@vger.kernel.org \
    --cc=jeyu@kernel.org \
    --cc=keescook@chromium.org \
    --cc=linux-arch@vger.kernel.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-hardening@vger.kernel.org \
    --cc=linux-kbuild@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pci@vger.kernel.org \
    --cc=masahiroy@kernel.org \
    --cc=nathan@kernel.org \
    --cc=ndesaulniers@google.com \
    --cc=tj@kernel.org \
    --cc=will@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).