public inbox for linux-bcachefs@vger.kernel.org
 help / color / mirror / Atom feed
* [PATCH] Revert "bcachefs: Annotate struct bch_xattr with __counted_by()"
@ 2025-05-01 18:41 Alan Huang
  2025-05-01 19:09 ` Jan Hendrik Farr
  2025-05-01 19:16 ` Kent Overstreet
  0 siblings, 2 replies; 4+ messages in thread
From: Alan Huang @ 2025-05-01 18:41 UTC (permalink / raw)
  To: kent.overstreet, kees, gustavoars, thorsten.blum
  Cc: linux-bcachefs, linux-hardening, Alan Huang

This reverts commit 86e92eeeb23741a072fe7532db663250ff2e726a.

After the x_name, there is a value. According to the disscussion[1],
__counted_by assumes that the flexible array member contains exactly
the amount of elements that are specified. Now there are users came across
buffer overflow caused by the __counted_by here[2], so revert that.

[1] https://lore.kernel.org/lkml/Zv8VDKWN1GzLRT-_@archlinux/T/#m0ce9541c5070146320efd4f928cc1ff8de69e9b2
[2] https://privatebin.net/?a0d4e97d590d71e1#9bLmp2Kb5NU6X6cZEucchDcu88HzUQwHUah8okKPReEt

Signed-off-by: Alan Huang <mmpgouride@gmail.com>
---
 fs/bcachefs/xattr_format.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/bcachefs/xattr_format.h b/fs/bcachefs/xattr_format.h
index c7916011ef34..e9f810539552 100644
--- a/fs/bcachefs/xattr_format.h
+++ b/fs/bcachefs/xattr_format.h
@@ -13,7 +13,7 @@ struct bch_xattr {
 	__u8			x_type;
 	__u8			x_name_len;
 	__le16			x_val_len;
-	__u8			x_name[] __counted_by(x_name_len);
+	__u8			x_name[];
 } __packed __aligned(8);
 
 #endif /* _BCACHEFS_XATTR_FORMAT_H */
-- 
2.48.1


^ permalink raw reply related	[flat|nested] 4+ messages in thread

* Re: [PATCH] Revert "bcachefs: Annotate struct bch_xattr with __counted_by()"
  2025-05-01 18:41 [PATCH] Revert "bcachefs: Annotate struct bch_xattr with __counted_by()" Alan Huang
@ 2025-05-01 19:09 ` Jan Hendrik Farr
  2025-05-01 19:16 ` Kent Overstreet
  1 sibling, 0 replies; 4+ messages in thread
From: Jan Hendrik Farr @ 2025-05-01 19:09 UTC (permalink / raw)
  To: Alan Huang
  Cc: kent.overstreet, kees, gustavoars, thorsten.blum, linux-bcachefs,
	linux-hardening

On 02 02:41:50, Alan Huang wrote:
> This reverts commit 86e92eeeb23741a072fe7532db663250ff2e726a.
> 
> After the x_name, there is a value. According to the disscussion[1],
> __counted_by assumes that the flexible array member contains exactly
> the amount of elements that are specified. Now there are users came across
> buffer overflow caused by the __counted_by here[2], so revert that.

Nit: It's not causing a buffer overflow. It's causing a false positive
detection of an out of bounds write.

> 
> [1] https://lore.kernel.org/lkml/Zv8VDKWN1GzLRT-_@archlinux/T/#m0ce9541c5070146320efd4f928cc1ff8de69e9b2
> [2] https://privatebin.net/?a0d4e97d590d71e1#9bLmp2Kb5NU6X6cZEucchDcu88HzUQwHUah8okKPReEt
> 
> Signed-off-by: Alan Huang <mmpgouride@gmail.com>
> ---
>  fs/bcachefs/xattr_format.h | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/fs/bcachefs/xattr_format.h b/fs/bcachefs/xattr_format.h
> index c7916011ef34..e9f810539552 100644
> --- a/fs/bcachefs/xattr_format.h
> +++ b/fs/bcachefs/xattr_format.h
> @@ -13,7 +13,7 @@ struct bch_xattr {
>  	__u8			x_type;
>  	__u8			x_name_len;
>  	__le16			x_val_len;
> -	__u8			x_name[] __counted_by(x_name_len);
> +	__u8			x_name[];
>  } __packed __aligned(8);
>  
>  #endif /* _BCACHEFS_XATTR_FORMAT_H */
> -- 
> 2.48.1
> 

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH] Revert "bcachefs: Annotate struct bch_xattr with __counted_by()"
  2025-05-01 18:41 [PATCH] Revert "bcachefs: Annotate struct bch_xattr with __counted_by()" Alan Huang
  2025-05-01 19:09 ` Jan Hendrik Farr
@ 2025-05-01 19:16 ` Kent Overstreet
  2025-05-01 19:22   ` Jan Hendrik Farr
  1 sibling, 1 reply; 4+ messages in thread
From: Kent Overstreet @ 2025-05-01 19:16 UTC (permalink / raw)
  To: Alan Huang
  Cc: kees, gustavoars, thorsten.blum, linux-bcachefs, linux-hardening

On Fri, May 02, 2025 at 02:41:50AM +0800, Alan Huang wrote:
> This reverts commit 86e92eeeb23741a072fe7532db663250ff2e726a.
> 
> After the x_name, there is a value. According to the disscussion[1],
> __counted_by assumes that the flexible array member contains exactly
> the amount of elements that are specified. Now there are users came across
> buffer overflow caused by the __counted_by here[2], so revert that.
> 
> [1] https://lore.kernel.org/lkml/Zv8VDKWN1GzLRT-_@archlinux/T/#m0ce9541c5070146320efd4f928cc1ff8de69e9b2
> [2] https://privatebin.net/?a0d4e97d590d71e1#9bLmp2Kb5NU6X6cZEucchDcu88HzUQwHUah8okKPReEt
> 
> Signed-off-by: Alan Huang <mmpgouride@gmail.com>
> ---
>  fs/bcachefs/xattr_format.h | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/fs/bcachefs/xattr_format.h b/fs/bcachefs/xattr_format.h
> index c7916011ef34..e9f810539552 100644
> --- a/fs/bcachefs/xattr_format.h
> +++ b/fs/bcachefs/xattr_format.h
> @@ -13,7 +13,7 @@ struct bch_xattr {
>  	__u8			x_type;
>  	__u8			x_name_len;
>  	__le16			x_val_len;
> -	__u8			x_name[] __counted_by(x_name_len);
> +	__u8			x_name[];

this needs a comment, so we don't make the same mistake again

>  } __packed __aligned(8);
>  
>  #endif /* _BCACHEFS_XATTR_FORMAT_H */
> -- 
> 2.48.1
> 

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH] Revert "bcachefs: Annotate struct bch_xattr with __counted_by()"
  2025-05-01 19:16 ` Kent Overstreet
@ 2025-05-01 19:22   ` Jan Hendrik Farr
  0 siblings, 0 replies; 4+ messages in thread
From: Jan Hendrik Farr @ 2025-05-01 19:22 UTC (permalink / raw)
  To: Kent Overstreet
  Cc: Alan Huang, kees, gustavoars, thorsten.blum, linux-bcachefs,
	linux-hardening

On 01 15:16:55, Kent Overstreet wrote:
> On Fri, May 02, 2025 at 02:41:50AM +0800, Alan Huang wrote:
> > This reverts commit 86e92eeeb23741a072fe7532db663250ff2e726a.
> > 
> > After the x_name, there is a value. According to the disscussion[1],
> > __counted_by assumes that the flexible array member contains exactly
> > the amount of elements that are specified. Now there are users came across
> > buffer overflow caused by the __counted_by here[2], so revert that.
> > 
> > [1] https://lore.kernel.org/lkml/Zv8VDKWN1GzLRT-_@archlinux/T/#m0ce9541c5070146320efd4f928cc1ff8de69e9b2
> > [2] https://privatebin.net/?a0d4e97d590d71e1#9bLmp2Kb5NU6X6cZEucchDcu88HzUQwHUah8okKPReEt
> > 
> > Signed-off-by: Alan Huang <mmpgouride@gmail.com>
> > ---
> >  fs/bcachefs/xattr_format.h | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> > 
> > diff --git a/fs/bcachefs/xattr_format.h b/fs/bcachefs/xattr_format.h
> > index c7916011ef34..e9f810539552 100644
> > --- a/fs/bcachefs/xattr_format.h
> > +++ b/fs/bcachefs/xattr_format.h
> > @@ -13,7 +13,7 @@ struct bch_xattr {
> >  	__u8			x_type;
> >  	__u8			x_name_len;
> >  	__le16			x_val_len;
> > -	__u8			x_name[] __counted_by(x_name_len);
> > +	__u8			x_name[];
> 
> this needs a comment, so we don't make the same mistake again

Better yet: change x_name to x_name_and_value

> 
> >  } __packed __aligned(8);
> >  
> >  #endif /* _BCACHEFS_XATTR_FORMAT_H */
> > -- 
> > 2.48.1
> > 

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2025-05-01 19:22 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-05-01 18:41 [PATCH] Revert "bcachefs: Annotate struct bch_xattr with __counted_by()" Alan Huang
2025-05-01 19:09 ` Jan Hendrik Farr
2025-05-01 19:16 ` Kent Overstreet
2025-05-01 19:22   ` Jan Hendrik Farr

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox