[0/3] net: Cleanups for FORTIFY_SOURCE

Message

Kees Cook Aug. 19, 2021, 8:28 p.m. UTC

Hi,

In preparation for FORTIFY_SOURCE performing compile-time and run-time
field bounds checking for memcpy(), memmove(), and memset(), avoid
intentionally writing across neighboring fields.

These three changes have been living in my memcpy() series[1], but have
no external dependencies. It's probably better to have these go via
netdev.

Thanks!

-Kees

[1] https://lore.kernel.org/lkml/[e-mail]/

Kees Cook (3):
  ipw2x00: Avoid field-overflowing memcpy()
  net/mlx5e: Avoid field-overflowing memcpy()
  pcmcia: ray_cs: Split memcpy() to avoid bounds check warning

 drivers/net/ethernet/mellanox/mlx5/core/en.h  |  4 +-
 .../net/ethernet/mellanox/mlx5/core/en/xdp.c  |  4 +-
 .../net/wireless/intel/ipw2x00/libipw_rx.c    | 56 ++++++-------------
 drivers/net/wireless/ray_cs.c                 |  4 +-
 4 files changed, 25 insertions(+), 43 deletions(-)

Comments

Jakub Kicinski Aug. 20, 2021, 5:01 p.m. UTC | #1

On Thu, 19 Aug 2021 13:28:22 -0700 Kees Cook wrote:
> Hi,
> 
> In preparation for FORTIFY_SOURCE performing compile-time and run-time
> field bounds checking for memcpy(), memmove(), and memset(), avoid
> intentionally writing across neighboring fields.
> 
> These three changes have been living in my memcpy() series[1], but have
> no external dependencies. It's probably better to have these go via
> netdev.

Thanks.

Kalle, Saeed - would you like to take the relevant changes? Presumably
they would get into net-next anyway by the time the merge window opens.

Kalle Valo Aug. 21, 2021, 10:11 a.m. UTC | #2

Jakub Kicinski <[e-mail]> writes:

> On Thu, 19 Aug 2021 13:28:22 -0700 Kees Cook wrote:
>> Hi,
>> 
>> In preparation for FORTIFY_SOURCE performing compile-time and run-time
>> field bounds checking for memcpy(), memmove(), and memset(), avoid
>> intentionally writing across neighboring fields.
>> 
>> These three changes have been living in my memcpy() series[1], but have
>> no external dependencies. It's probably better to have these go via
>> netdev.
>
> Thanks.
>
> Kalle, Saeed - would you like to take the relevant changes? Presumably
> they would get into net-next anyway by the time the merge window opens.

Ok, I'll take patch 1 to wireless-drivers-next.

Kalle Valo Aug. 21, 2021, 10:13 a.m. UTC | #3

Kalle Valo <[e-mail]> writes:

> Jakub Kicinski <[e-mail]> writes:
>
>> On Thu, 19 Aug 2021 13:28:22 -0700 Kees Cook wrote:
>>> Hi,
>>> 
>>> In preparation for FORTIFY_SOURCE performing compile-time and run-time
>>> field bounds checking for memcpy(), memmove(), and memset(), avoid
>>> intentionally writing across neighboring fields.
>>> 
>>> These three changes have been living in my memcpy() series[1], but have
>>> no external dependencies. It's probably better to have these go via
>>> netdev.
>>
>> Thanks.
>>
>> Kalle, Saeed - would you like to take the relevant changes? Presumably
>> they would get into net-next anyway by the time the merge window opens.
>
> Ok, I'll take patch 1 to wireless-drivers-next.

Correction: I'll take patches 1 and 3 to wireless-drivers-next.

Kees Cook Aug. 22, 2021, 5:16 a.m. UTC | #4

On Sat, Aug 21, 2021 at 01:13:37PM +0300, Kalle Valo wrote:
> Kalle Valo <[e-mail]> writes:
> 
> > Jakub Kicinski <[e-mail]> writes:
> >
> >> On Thu, 19 Aug 2021 13:28:22 -0700 Kees Cook wrote:
> >>> Hi,
> >>> 
> >>> In preparation for FORTIFY_SOURCE performing compile-time and run-time
> >>> field bounds checking for memcpy(), memmove(), and memset(), avoid
> >>> intentionally writing across neighboring fields.
> >>> 
> >>> These three changes have been living in my memcpy() series[1], but have
> >>> no external dependencies. It's probably better to have these go via
> >>> netdev.
> >>
> >> Thanks.
> >>
> >> Kalle, Saeed - would you like to take the relevant changes? Presumably
> >> they would get into net-next anyway by the time the merge window opens.
> >
> > Ok, I'll take patch 1 to wireless-drivers-next.
> 
> Correction: I'll take patches 1 and 3 to wireless-drivers-next.

Great; thanks!