From: Masahiro Yamada <masahiroy@kernel.org>
To: Mauricio Faria de Oliveira <mfo@canonical.com>
Cc: Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
linux-modules <linux-modules@vger.kernel.org>,
Linux Kbuild mailing list <linux-kbuild@vger.kernel.org>,
Linux FS-devel Mailing List <linux-fsdevel@vger.kernel.org>,
Michal Marek <michal.lkml@markovi.net>,
Nick Desaulniers <ndesaulniers@google.com>,
Luis Chamberlain <mcgrof@kernel.org>,
Kees Cook <keescook@chromium.org>,
Iurii Zaikin <yzaikin@google.com>
Subject: Re: [RFC PATCH 3/6] sysctl, mod_devicetable: shadow struct ctl_table.procname for file2alias
Date: Tue, 26 Jul 2022 18:25:51 +0900 [thread overview]
Message-ID: <CAK7LNARvJEhEOwg_PHe3WKT9BkSchnGOmeiUaB+7E__NS9qrVw@mail.gmail.com> (raw)
In-Reply-To: <20220722022416.137548-4-mfo@canonical.com>
On Fri, Jul 22, 2022 at 11:24 AM Mauricio Faria de Oliveira
<mfo@canonical.com> wrote:
>
> In order to expose a sysctl entry to modpost (file2alias.c, precisely)
> we have to shadow 'struct ctl_table' in mod_devicetable.h, as scripts
> should not access kernel headers or its types (see file2alias.c).
>
> The required field is '.procname' (basename of '/proc/sys/.../entry').
>
> Since 'struct ctl_table' is annotated for structure randomization and
> we need a known offset for '.procname' (remember, no kernel headers),
> take it out of the randomized portion (as in, eg, 'struct task_struct').
>
> Of course, add build-time checks for struct size and .procname offset
> between both structs. (This has to be done on kernel side; for headers.)
>
> With that in place, use the regular macros in devicetable-offsets.c to
> define SIZE_... and OFF_... macros for the shadow struct and the field
> of interest.
>
> Signed-off-by: Mauricio Faria de Oliveira <mfo@canonical.com>
> ---
> fs/proc/proc_sysctl.c | 19 +++++++++++++++++++
> include/linux/mod_devicetable.h | 25 +++++++++++++++++++++++++
> include/linux/sysctl.h | 11 ++++++++++-
> kernel/sysctl.c | 1 +
> scripts/mod/devicetable-offsets.c | 3 +++
> 5 files changed, 58 insertions(+), 1 deletion(-)
>
> diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c
> index 021e83fe831f..ebbf8702387e 100644
> --- a/fs/proc/proc_sysctl.c
> +++ b/fs/proc/proc_sysctl.c
> @@ -19,6 +19,24 @@
> #include <linux/kmemleak.h>
> #include "internal.h"
>
> +#ifdef CONFIG_MODULES
> +#include <linux/mod_devicetable.h>
> +
> +static void check_struct_sysctl_device_id(void)
> +{
> + /*
> + * The shadow struct sysctl_device_id for file2alias.c needs
> + * the same size of struct ctl_table and offset for procname.
> + */
> + BUILD_BUG_ON(sizeof(struct sysctl_device_id)
> + != sizeof(struct ctl_table));
> + BUILD_BUG_ON(offsetof(struct sysctl_device_id, procname)
> + != offsetof(struct ctl_table, procname));
Nit:
If you use static_assert(), you can remove
check_struct_sysctl_device_id().
You can write static_assert() out of a function.
> diff --git a/kernel/sysctl.c b/kernel/sysctl.c
> index 223376959d29..15073621cfa8 100644
> --- a/kernel/sysctl.c
> +++ b/kernel/sysctl.c
> @@ -2487,6 +2487,7 @@ int __init sysctl_init_bases(void)
>
> return 0;
> }
> +
Noise.
> #endif /* CONFIG_SYSCTL */
> /*
> * No sense putting this after each symbol definition, twice,
> diff --git a/scripts/mod/devicetable-offsets.c b/scripts/mod/devicetable-offsets.c
> index c0d3bcb99138..43b2549940d2 100644
> --- a/scripts/mod/devicetable-offsets.c
> +++ b/scripts/mod/devicetable-offsets.c
> @@ -262,5 +262,8 @@ int main(void)
> DEVID(ishtp_device_id);
> DEVID_FIELD(ishtp_device_id, guid);
>
> + DEVID(sysctl_device_id);
> + DEVID_FIELD(sysctl_device_id, procname);
> +
> return 0;
> }
> --
> 2.25.1
>
--
Best Regards
Masahiro Yamada
next prev parent reply other threads:[~2022-07-26 9:27 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-07-22 2:24 [RFC PATCH 0/6] Introduce "sysctl:" module aliases Mauricio Faria de Oliveira
2022-07-22 2:24 ` [RFC PATCH 1/6] modpost: factor out elf/arch-specific code from section_rel[a]() Mauricio Faria de Oliveira
2022-07-22 2:24 ` [RFC PATCH 2/6] modpost: deduplicate section_rel[a]() Mauricio Faria de Oliveira
2022-07-26 9:19 ` Masahiro Yamada
2022-07-27 17:10 ` Mauricio Faria de Oliveira
2022-07-22 2:24 ` [RFC PATCH 3/6] sysctl, mod_devicetable: shadow struct ctl_table.procname for file2alias Mauricio Faria de Oliveira
2022-07-26 9:25 ` Masahiro Yamada [this message]
2022-07-27 17:11 ` Mauricio Faria de Oliveira
2022-07-22 2:24 ` [RFC PATCH 4/6] module, modpost: introduce support for MODULE_SYSCTL_TABLE Mauricio Faria de Oliveira
2022-07-22 2:24 ` [RFC PATCH 5/6] netfilter: conntrack: use MODULE_SYSCTL_TABLE Mauricio Faria de Oliveira
2022-07-22 2:24 ` [RFC PATCH 6/6] sysctl: introduce /proc/sys/kernel/modprobe_sysctl_alias Mauricio Faria de Oliveira
2022-07-26 9:22 ` Masahiro Yamada
2022-07-27 17:11 ` Mauricio Faria de Oliveira
2022-07-26 9:02 ` [RFC PATCH 0/6] Introduce "sysctl:" module aliases Masahiro Yamada
2022-07-27 17:09 ` Mauricio Faria de Oliveira
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=CAK7LNARvJEhEOwg_PHe3WKT9BkSchnGOmeiUaB+7E__NS9qrVw@mail.gmail.com \
--to=masahiroy@kernel.org \
--cc=keescook@chromium.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kbuild@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-modules@vger.kernel.org \
--cc=mcgrof@kernel.org \
--cc=mfo@canonical.com \
--cc=michal.lkml@markovi.net \
--cc=ndesaulniers@google.com \
--cc=yzaikin@google.com \
/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).