[GIT PULL] hardening fix for v7.3-rc2

[GIT PULL] hardening fix for v7.3-rc2

[Kees Cook]

Hi Linus,

Please pull this small hardening fix for v7.3-rc2.

Thanks!

-Kees

The following changes since commit cee9395acd8043be0644b25c34bfa86623f2b935:

  Linux 7.3-rc1 (2026-08-30 13:34:40 -0700)

are available in the Git repository at:

  https://git.kernel.org/pub/scm/linux/kernel/git/kees/linux.git tags/hardening-v7.3-rc2

for you to fetch changes up to 2625480a1bf79c62ffb09aafdf61778e682da492:

  hardening: Default randstruct off with rust for better allmodconfig support (2026-09-01 16:01:35 -0700)

----------------------------------------------------------------
hardening fix for v7.3-rc2

- Default randstruct off with rust for better allmodconfig coverage
  (Mark Brown)

----------------------------------------------------------------
Mark Brown (1):
      hardening: Default randstruct off with rust for better allmodconfig support

 security/Kconfig.hardening | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

-- 
Kees Cook

Re: [GIT PULL] hardening fix for v7.3-rc2

[Linus Torvalds]

On Wed, 2 Sept 2026 at 09:31, Kees Cook <kees@kernel.org> wrote:
>
> - Default randstruct off with rust for better allmodconfig coverage
>   (Mark Brown)

Should we just default randstruct off entirely?

It defaults to off for any *real* build for good reasons. I'm not
convinced it should be on for build testing either. Maybe the active
compiler bugs are gone, but this whole commit is just an example of
how RANDSTRUCT_FULL really isn't a good default - not for real life,
and not for build test coverage.

                  Linus

Re: [GIT PULL] hardening fix for v7.3-rc2

[Kees Cook]

On Wed, Sep 02, 2026 at 04:07:46PM -0700, Linus Torvalds wrote:
> On Wed, 2 Sept 2026 at 09:31, Kees Cook <kees@kernel.org> wrote:
> >
> > - Default randstruct off with rust for better allmodconfig coverage
> >   (Mark Brown)
> 
> Should we just default randstruct off entirely?
> 
> It defaults to off for any *real* build for good reasons. I'm not
> convinced it should be on for build testing either. Maybe the active
> compiler bugs are gone, but this whole commit is just an example of
> how RANDSTRUCT_FULL really isn't a good default - not for real life,
> and not for build test coverage.

I have a fix for it with Rust, and it'll be in -next after -rc2 is out.
We still have real-world users that depend on randstruct, so I'd like to
continue to make breakage visible via all*config builds, etc.

-- 
Kees Cook

Re: [GIT PULL] hardening fix for v7.3-rc2

[Miguel Ojeda]

On Thu, Sep 3, 2026 at 1:49 AM Kees Cook <kees@kernel.org> wrote:
>
> I have a fix for it with Rust, and it'll be in -next after -rc2 is out.
> We still have real-world users that depend on randstruct, so I'd like to
> continue to make breakage visible via all*config builds, etc.

Do you mean the ones we discussed in the past like e.g. my

  https://lore.kernel.org/rust-for-linux/20260323130224.165738-1-ojeda@kernel.org/

or your

  https://lore.kernel.org/rust-for-linux/202512092319.ECBB8613@keescook/

or something else?

Cheers,
Miguel

Re: [GIT PULL] hardening fix for v7.3-rc2

[Kees Cook]

On September 3, 2026 4:02:57 PM PDT, Miguel Ojeda <miguel.ojeda.sandonis@gmail.com> wrote:
>On Thu, Sep 3, 2026 at 1:49 AM Kees Cook <kees@kernel.org> wrote:
>>
>> I have a fix for it with Rust, and it'll be in -next after -rc2 is out.
>> We still have real-world users that depend on randstruct, so I'd like to
>> continue to make breakage visible via all*config builds, etc.
>
>Do you mean the ones we discussed in the past like e.g. my
>
>  https://lore.kernel.org/rust-for-linux/20260323130224.165738-1-ojeda@kernel.org/
>
>or your
>
>  https://lore.kernel.org/rust-for-linux/202512092319.ECBB8613@keescook/

Yes, this latter one. I was going to send it after rc2 and we could let it bake for the whole dev cycle.

-Kees

-- 
Kees Cook

Re: [GIT PULL] hardening fix for v7.3-rc2

[Miguel Ojeda]

On Fri, Sep 4, 2026 at 3:38 AM Kees Cook <kees@kernel.org> wrote:
>
> Yes, this latter one. I was going to send it after rc2 and we could let it bake for the whole dev cycle.

If you prefer a single patch doing both, then I can integrate it into
mine if that helps; but if you want to drive it, that sounds good too,
of course.

Otherwise, I suggested back then to do 2 patches, e.g. one simplifying
the C side first, and then the one adding the Rust side, because that
sounded cleaner:

  https://lore.kernel.org/rust-for-linux/CANiq72n=hgH4bqJjp8MsMHAaxaAo75GSBcHGTvFT3NTSaVPGWg@mail.gmail.com/

(We could do it the other way around to keep the Tested-by from
Andreas, but a lot of time has passed, so it would be best to ask for
a re-test -- Cc'ing Andreas).

Cheers,
Miguel

Re: [GIT PULL] hardening fix for v7.3-rc2

[pr-tracker-bot]

The pull request you sent on Wed, 2 Sep 2026 09:31:50 -0700:

> https://git.kernel.org/pub/scm/linux/kernel/git/kees/linux.git tags/hardening-v7.3-rc2

has been merged into torvalds/linux.git:
https://git.kernel.org/torvalds/c/940de590b839f71d6dc846160534bf202401b8b7

Thank you!

-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/prtracker.html