linux-raid.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH] md/md-linear: Annotate struct linear_conf with __counted_by
@ 2023-09-15 20:03 Kees Cook
  2023-09-15 20:28 ` Gustavo A. R. Silva
  2023-09-29 19:21 ` Kees Cook
  0 siblings, 2 replies; 6+ messages in thread
From: Kees Cook @ 2023-09-15 20:03 UTC (permalink / raw)
  To: Song Liu
  Cc: Kees Cook, linux-raid, Nathan Chancellor, Nick Desaulniers,
	Tom Rix, linux-kernel, llvm, linux-hardening

Prepare for the coming implementation by GCC and Clang of the __counted_by
attribute. Flexible array members annotated with __counted_by can have
their accesses bounds-checked at run-time checking via CONFIG_UBSAN_BOUNDS
(for array indexing) and CONFIG_FORTIFY_SOURCE (for strcpy/memcpy-family
functions).

As found with Coccinelle[1], add __counted_by for struct linear_conf.
Additionally, since the element count member must be set before accessing
the annotated flexible array member, move its initialization earlier.

[1] https://github.com/kees/kernel-tools/blob/trunk/coccinelle/examples/counted_by.cocci

Cc: Song Liu <song@kernel.org>
Cc: linux-raid@vger.kernel.org
Signed-off-by: Kees Cook <keescook@chromium.org>
---
 drivers/md/md-linear.c | 26 +++++++++++++-------------
 drivers/md/md-linear.h |  2 +-
 2 files changed, 14 insertions(+), 14 deletions(-)

diff --git a/drivers/md/md-linear.c b/drivers/md/md-linear.c
index 71ac99646827..ae2826e9645b 100644
--- a/drivers/md/md-linear.c
+++ b/drivers/md/md-linear.c
@@ -69,6 +69,19 @@ static struct linear_conf *linear_conf(struct mddev *mddev, int raid_disks)
 	if (!conf)
 		return NULL;
 
+	/*
+	 * conf->raid_disks is copy of mddev->raid_disks. The reason to
+	 * keep a copy of mddev->raid_disks in struct linear_conf is,
+	 * mddev->raid_disks may not be consistent with pointers number of
+	 * conf->disks[] when it is updated in linear_add() and used to
+	 * iterate old conf->disks[] earray in linear_congested().
+	 * Here conf->raid_disks is always consitent with number of
+	 * pointers in conf->disks[] array, and mddev->private is updated
+	 * with rcu_assign_pointer() in linear_addr(), such race can be
+	 * avoided.
+	 */
+	conf->raid_disks = raid_disks;
+
 	cnt = 0;
 	conf->array_sectors = 0;
 
@@ -112,19 +125,6 @@ static struct linear_conf *linear_conf(struct mddev *mddev, int raid_disks)
 			conf->disks[i-1].end_sector +
 			conf->disks[i].rdev->sectors;
 
-	/*
-	 * conf->raid_disks is copy of mddev->raid_disks. The reason to
-	 * keep a copy of mddev->raid_disks in struct linear_conf is,
-	 * mddev->raid_disks may not be consistent with pointers number of
-	 * conf->disks[] when it is updated in linear_add() and used to
-	 * iterate old conf->disks[] earray in linear_congested().
-	 * Here conf->raid_disks is always consitent with number of
-	 * pointers in conf->disks[] array, and mddev->private is updated
-	 * with rcu_assign_pointer() in linear_addr(), such race can be
-	 * avoided.
-	 */
-	conf->raid_disks = raid_disks;
-
 	return conf;
 
 out:
diff --git a/drivers/md/md-linear.h b/drivers/md/md-linear.h
index 24e97db50ebb..5587eeedb882 100644
--- a/drivers/md/md-linear.h
+++ b/drivers/md/md-linear.h
@@ -12,6 +12,6 @@ struct linear_conf
 	struct rcu_head		rcu;
 	sector_t		array_sectors;
 	int			raid_disks; /* a copy of mddev->raid_disks */
-	struct dev_info		disks[];
+	struct dev_info		disks[] __counted_by(raid_disks);
 };
 #endif
-- 
2.34.1


^ permalink raw reply related	[flat|nested] 6+ messages in thread

* Re: [PATCH] md/md-linear: Annotate struct linear_conf with __counted_by
  2023-09-15 20:03 [PATCH] md/md-linear: Annotate struct linear_conf with __counted_by Kees Cook
@ 2023-09-15 20:28 ` Gustavo A. R. Silva
  2023-09-22 18:27   ` Song Liu
  2023-09-29 19:21 ` Kees Cook
  1 sibling, 1 reply; 6+ messages in thread
From: Gustavo A. R. Silva @ 2023-09-15 20:28 UTC (permalink / raw)
  To: Kees Cook, Song Liu
  Cc: linux-raid, Nathan Chancellor, Nick Desaulniers, Tom Rix,
	linux-kernel, llvm, linux-hardening



On 9/15/23 14:03, Kees Cook wrote:
> Prepare for the coming implementation by GCC and Clang of the __counted_by
> attribute. Flexible array members annotated with __counted_by can have
> their accesses bounds-checked at run-time checking via CONFIG_UBSAN_BOUNDS
> (for array indexing) and CONFIG_FORTIFY_SOURCE (for strcpy/memcpy-family
> functions).
> 
> As found with Coccinelle[1], add __counted_by for struct linear_conf.
> Additionally, since the element count member must be set before accessing
> the annotated flexible array member, move its initialization earlier.
> 
> [1] https://github.com/kees/kernel-tools/blob/trunk/coccinelle/examples/counted_by.cocci
> 
> Cc: Song Liu <song@kernel.org>
> Cc: linux-raid@vger.kernel.org
> Signed-off-by: Kees Cook <keescook@chromium.org>

Reviewed-by: Gustavo A. R. Silva <gustavoars@kernel.org>

Thanks
-- 
Gustavo

> ---
>   drivers/md/md-linear.c | 26 +++++++++++++-------------
>   drivers/md/md-linear.h |  2 +-
>   2 files changed, 14 insertions(+), 14 deletions(-)
> 
> diff --git a/drivers/md/md-linear.c b/drivers/md/md-linear.c
> index 71ac99646827..ae2826e9645b 100644
> --- a/drivers/md/md-linear.c
> +++ b/drivers/md/md-linear.c
> @@ -69,6 +69,19 @@ static struct linear_conf *linear_conf(struct mddev *mddev, int raid_disks)
>   	if (!conf)
>   		return NULL;
>   
> +	/*
> +	 * conf->raid_disks is copy of mddev->raid_disks. The reason to
> +	 * keep a copy of mddev->raid_disks in struct linear_conf is,
> +	 * mddev->raid_disks may not be consistent with pointers number of
> +	 * conf->disks[] when it is updated in linear_add() and used to
> +	 * iterate old conf->disks[] earray in linear_congested().
> +	 * Here conf->raid_disks is always consitent with number of
> +	 * pointers in conf->disks[] array, and mddev->private is updated
> +	 * with rcu_assign_pointer() in linear_addr(), such race can be
> +	 * avoided.
> +	 */
> +	conf->raid_disks = raid_disks;
> +
>   	cnt = 0;
>   	conf->array_sectors = 0;
>   
> @@ -112,19 +125,6 @@ static struct linear_conf *linear_conf(struct mddev *mddev, int raid_disks)
>   			conf->disks[i-1].end_sector +
>   			conf->disks[i].rdev->sectors;
>   
> -	/*
> -	 * conf->raid_disks is copy of mddev->raid_disks. The reason to
> -	 * keep a copy of mddev->raid_disks in struct linear_conf is,
> -	 * mddev->raid_disks may not be consistent with pointers number of
> -	 * conf->disks[] when it is updated in linear_add() and used to
> -	 * iterate old conf->disks[] earray in linear_congested().
> -	 * Here conf->raid_disks is always consitent with number of
> -	 * pointers in conf->disks[] array, and mddev->private is updated
> -	 * with rcu_assign_pointer() in linear_addr(), such race can be
> -	 * avoided.
> -	 */
> -	conf->raid_disks = raid_disks;
> -
>   	return conf;
>   
>   out:
> diff --git a/drivers/md/md-linear.h b/drivers/md/md-linear.h
> index 24e97db50ebb..5587eeedb882 100644
> --- a/drivers/md/md-linear.h
> +++ b/drivers/md/md-linear.h
> @@ -12,6 +12,6 @@ struct linear_conf
>   	struct rcu_head		rcu;
>   	sector_t		array_sectors;
>   	int			raid_disks; /* a copy of mddev->raid_disks */
> -	struct dev_info		disks[];
> +	struct dev_info		disks[] __counted_by(raid_disks);
>   };
>   #endif

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH] md/md-linear: Annotate struct linear_conf with __counted_by
  2023-09-15 20:28 ` Gustavo A. R. Silva
@ 2023-09-22 18:27   ` Song Liu
  0 siblings, 0 replies; 6+ messages in thread
From: Song Liu @ 2023-09-22 18:27 UTC (permalink / raw)
  To: Gustavo A. R. Silva
  Cc: Kees Cook, linux-raid, Nathan Chancellor, Nick Desaulniers,
	Tom Rix, linux-kernel, llvm, linux-hardening

On Fri, Sep 15, 2023 at 1:27 PM Gustavo A. R. Silva
<gustavo@embeddedor.com> wrote:
>
>
>
> On 9/15/23 14:03, Kees Cook wrote:
> > Prepare for the coming implementation by GCC and Clang of the __counted_by
> > attribute. Flexible array members annotated with __counted_by can have
> > their accesses bounds-checked at run-time checking via CONFIG_UBSAN_BOUNDS
> > (for array indexing) and CONFIG_FORTIFY_SOURCE (for strcpy/memcpy-family
> > functions).
> >
> > As found with Coccinelle[1], add __counted_by for struct linear_conf.
> > Additionally, since the element count member must be set before accessing
> > the annotated flexible array member, move its initialization earlier.
> >
> > [1] https://github.com/kees/kernel-tools/blob/trunk/coccinelle/examples/counted_by.cocci
> >
> > Cc: Song Liu <song@kernel.org>
> > Cc: linux-raid@vger.kernel.org
> > Signed-off-by: Kees Cook <keescook@chromium.org>
>
> Reviewed-by: Gustavo A. R. Silva <gustavoars@kernel.org>

Applied to md-next. Thanks!
Song

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH] md/md-linear: Annotate struct linear_conf with __counted_by
  2023-09-15 20:03 [PATCH] md/md-linear: Annotate struct linear_conf with __counted_by Kees Cook
  2023-09-15 20:28 ` Gustavo A. R. Silva
@ 2023-09-29 19:21 ` Kees Cook
  2023-09-29 23:40   ` Song Liu
  1 sibling, 1 reply; 6+ messages in thread
From: Kees Cook @ 2023-09-29 19:21 UTC (permalink / raw)
  To: Song Liu, Kees Cook
  Cc: linux-raid, Nathan Chancellor, Nick Desaulniers, Tom Rix,
	linux-kernel, llvm, linux-hardening

On Fri, 15 Sep 2023 13:03:28 -0700, Kees Cook wrote:
> Prepare for the coming implementation by GCC and Clang of the __counted_by
> attribute. Flexible array members annotated with __counted_by can have
> their accesses bounds-checked at run-time checking via CONFIG_UBSAN_BOUNDS
> (for array indexing) and CONFIG_FORTIFY_SOURCE (for strcpy/memcpy-family
> functions).
> 
> As found with Coccinelle[1], add __counted_by for struct linear_conf.
> Additionally, since the element count member must be set before accessing
> the annotated flexible array member, move its initialization earlier.
> 
> [...]

Applied to for-next/hardening, thanks!

[1/1] md/md-linear: Annotate struct linear_conf with __counted_by
      https://git.kernel.org/kees/c/9add7681e09b

Take care,

-- 
Kees Cook


^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH] md/md-linear: Annotate struct linear_conf with __counted_by
  2023-09-29 19:21 ` Kees Cook
@ 2023-09-29 23:40   ` Song Liu
  2023-09-29 23:43     ` Kees Cook
  0 siblings, 1 reply; 6+ messages in thread
From: Song Liu @ 2023-09-29 23:40 UTC (permalink / raw)
  To: Kees Cook, Jens Axboe
  Cc: linux-raid, Nathan Chancellor, Nick Desaulniers, Tom Rix,
	linux-kernel, llvm, linux-hardening

On Fri, Sep 29, 2023 at 12:21 PM Kees Cook <keescook@chromium.org> wrote:
>
> On Fri, 15 Sep 2023 13:03:28 -0700, Kees Cook wrote:
> > Prepare for the coming implementation by GCC and Clang of the __counted_by
> > attribute. Flexible array members annotated with __counted_by can have
> > their accesses bounds-checked at run-time checking via CONFIG_UBSAN_BOUNDS
> > (for array indexing) and CONFIG_FORTIFY_SOURCE (for strcpy/memcpy-family
> > functions).
> >
> > As found with Coccinelle[1], add __counted_by for struct linear_conf.
> > Additionally, since the element count member must be set before accessing
> > the annotated flexible array member, move its initialization earlier.
> >
> > [...]
>
> Applied to for-next/hardening, thanks!
>
> [1/1] md/md-linear: Annotate struct linear_conf with __counted_by
>       https://git.kernel.org/kees/c/9add7681e09b

Hmm..

Jens pulled this into his for-next branch and for-6.7/block branch
earlier today:

https://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux-block.git/commit/?h=for-next&id=e887544d7620f1d3cef017e45df7bc625182caff

Do we need to do anything about this (drop one of them)?

Thanks,
Song

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH] md/md-linear: Annotate struct linear_conf with __counted_by
  2023-09-29 23:40   ` Song Liu
@ 2023-09-29 23:43     ` Kees Cook
  0 siblings, 0 replies; 6+ messages in thread
From: Kees Cook @ 2023-09-29 23:43 UTC (permalink / raw)
  To: Song Liu
  Cc: Jens Axboe, linux-raid, Nathan Chancellor, Nick Desaulniers,
	Tom Rix, linux-kernel, llvm, linux-hardening

On Fri, Sep 29, 2023 at 04:40:13PM -0700, Song Liu wrote:
> On Fri, Sep 29, 2023 at 12:21 PM Kees Cook <keescook@chromium.org> wrote:
> >
> > On Fri, 15 Sep 2023 13:03:28 -0700, Kees Cook wrote:
> > > Prepare for the coming implementation by GCC and Clang of the __counted_by
> > > attribute. Flexible array members annotated with __counted_by can have
> > > their accesses bounds-checked at run-time checking via CONFIG_UBSAN_BOUNDS
> > > (for array indexing) and CONFIG_FORTIFY_SOURCE (for strcpy/memcpy-family
> > > functions).
> > >
> > > As found with Coccinelle[1], add __counted_by for struct linear_conf.
> > > Additionally, since the element count member must be set before accessing
> > > the annotated flexible array member, move its initialization earlier.
> > >
> > > [...]
> >
> > Applied to for-next/hardening, thanks!
> >
> > [1/1] md/md-linear: Annotate struct linear_conf with __counted_by
> >       https://git.kernel.org/kees/c/9add7681e09b
> 
> Hmm..
> 
> Jens pulled this into his for-next branch and for-6.7/block branch
> earlier today:
> 
> https://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux-block.git/commit/?h=for-next&id=e887544d7620f1d3cef017e45df7bc625182caff
> 
> Do we need to do anything about this (drop one of them)?

Whoops! Sorry, I hadn't seen it picked up. I'll drop it from my tree.
Thanks!

-Kees

-- 
Kees Cook

^ permalink raw reply	[flat|nested] 6+ messages in thread

end of thread, other threads:[~2023-09-29 23:43 UTC | newest]

Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2023-09-15 20:03 [PATCH] md/md-linear: Annotate struct linear_conf with __counted_by Kees Cook
2023-09-15 20:28 ` Gustavo A. R. Silva
2023-09-22 18:27   ` Song Liu
2023-09-29 19:21 ` Kees Cook
2023-09-29 23:40   ` Song Liu
2023-09-29 23:43     ` Kees Cook

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).