backport mseal and mseal_test to 6.10

backport mseal and mseal_test to 6.10

[Jeff Xu]

Hi Greg,

How are you?

What is the process to backport Pedro's recent mseal fixes to 6.10 ?

Specifically those 5 commits:

67203f3f2a63d429272f0c80451e5fcc469fdb46
    selftests/mm: add mseal test for no-discard madvise

4d1b3416659be70a2251b494e85e25978de06519
    mm: move can_modify_vma to mm/vma.h

 4a2dd02b09160ee43f96c759fafa7b56dfc33816
  mm/mprotect: replace can_modify_mm with can_modify_vma

23c57d1fa2b9530e38f7964b4e457fed5a7a0ae8
      mseal: replace can_modify_mm_madv with a vma variant

f28bdd1b17ec187eaa34845814afaaff99832762
   selftests/mm: add more mseal traversal tests

There will be merge conflicts, I  can backport them to 5.10 and test
to help the backporting process.

Those 5 fixes are needed for two reasons: maintain the consistency of
mseal's semantics across releases, and for ease of backporting future
fixes.

PS: There are also three other commits for munmap and remap (see below),
 they have dependency on Michael Ellerman's  arch_unmap() patch [1] and maybe
uprobe change [2]. If Michael and Oleg are OK with backporting their
patches, then great !
Otherwise, since those commits below don't change mseal's semantics, I
think it is OK to just backport above 5 patches.

df2a7df9a9aa32c3df227de346693e6e802c8591
     mm/munmap: replace can_modify_mm with can_modify_vma
38075679b5f157eeacd46c900e9cfc684bdbc167
   mm/mremap: replace can_modify_mm with can_modify_vma
5b3db2b812a1f86dfab587324d198a5d10c7a5cf
   mm: remove can_modify_mm()

[1] https://lore.kernel.org/all/20240812082605.743814-1-mpe@ellerman.id.au/
[2] https://lore.kernel.org/all/20240911131320.GA3448@redhat.com/

Thanks!
-Jeff

Re: backport mseal and mseal_test to 6.10

[Greg KH]

On Sun, Oct 13, 2024 at 10:17:48PM -0700, Jeff Xu wrote:
> Hi Greg,
> 
> How are you?
> 
> What is the process to backport Pedro's recent mseal fixes to 6.10 ?

Please read:
    https://www.kernel.org/doc/html/latest/process/stable-kernel-rules.html
for how all of this works :)

> Specifically those 5 commits:
> 
> 67203f3f2a63d429272f0c80451e5fcc469fdb46
>     selftests/mm: add mseal test for no-discard madvise
> 
> 4d1b3416659be70a2251b494e85e25978de06519
>     mm: move can_modify_vma to mm/vma.h
> 
>  4a2dd02b09160ee43f96c759fafa7b56dfc33816
>   mm/mprotect: replace can_modify_mm with can_modify_vma
> 
> 23c57d1fa2b9530e38f7964b4e457fed5a7a0ae8
>       mseal: replace can_modify_mm_madv with a vma variant
> 
> f28bdd1b17ec187eaa34845814afaaff99832762
>    selftests/mm: add more mseal traversal tests
> 
> There will be merge conflicts, I  can backport them to 5.10 and test
> to help the backporting process.

5.10 or 6.10?

And why 6.10?  If you look at the front page of kernel.org you will see
that 6.10 is now end-of-life, so why does that kernel matter to you
anymore?

> Those 5 fixes are needed for two reasons: maintain the consistency of
> mseal's semantics across releases, and for ease of backporting future
> fixes.

Backporting more to 6.10?  Again, it's end-of-life, who would be
backporting anything else?

confused,

greg k-h

Re: backport mseal and mseal_test to 6.10

[Jeff Xu]

On Sun, Oct 13, 2024 at 10:54 PM Greg KH <gregkh@linuxfoundation.org> wrote:
>
> On Sun, Oct 13, 2024 at 10:17:48PM -0700, Jeff Xu wrote:
> > Hi Greg,
> >
> > How are you?
> >
> > What is the process to backport Pedro's recent mseal fixes to 6.10 ?
>
> Please read:
>     https://www.kernel.org/doc/html/latest/process/stable-kernel-rules.html
> for how all of this works :)
>
> > Specifically those 5 commits:
> >
> > 67203f3f2a63d429272f0c80451e5fcc469fdb46
> >     selftests/mm: add mseal test for no-discard madvise
> >
> > 4d1b3416659be70a2251b494e85e25978de06519
> >     mm: move can_modify_vma to mm/vma.h
> >
> >  4a2dd02b09160ee43f96c759fafa7b56dfc33816
> >   mm/mprotect: replace can_modify_mm with can_modify_vma
> >
> > 23c57d1fa2b9530e38f7964b4e457fed5a7a0ae8
> >       mseal: replace can_modify_mm_madv with a vma variant
> >
> > f28bdd1b17ec187eaa34845814afaaff99832762
> >    selftests/mm: add more mseal traversal tests
> >
> > There will be merge conflicts, I  can backport them to 5.10 and test
> > to help the backporting process.
>
> 5.10 or 6.10?
>
6.10.

> And why 6.10?  If you look at the front page of kernel.org you will see
> that 6.10 is now end-of-life, so why does that kernel matter to you
> anymore?
>
OK, I didn't know that. Less work is nice :-)

Thanks!
-Jeff

> > Those 5 fixes are needed for two reasons: maintain the consistency of
> > mseal's semantics across releases, and for ease of backporting future
> > fixes.
>
> Backporting more to 6.10?  Again, it's end-of-life, who would be
> backporting anything else?
>
> confused,
>
> greg k-h

Re: backport mseal and mseal_test to 6.10

[Greg KH]

On Mon, Oct 14, 2024 at 08:27:29AM -0700, Jeff Xu wrote:
> On Sun, Oct 13, 2024 at 10:54 PM Greg KH <gregkh@linuxfoundation.org> wrote:
> >
> > On Sun, Oct 13, 2024 at 10:17:48PM -0700, Jeff Xu wrote:
> > > Hi Greg,
> > >
> > > How are you?
> > >
> > > What is the process to backport Pedro's recent mseal fixes to 6.10 ?
> >
> > Please read:
> >     https://www.kernel.org/doc/html/latest/process/stable-kernel-rules.html
> > for how all of this works :)
> >
> > > Specifically those 5 commits:
> > >
> > > 67203f3f2a63d429272f0c80451e5fcc469fdb46
> > >     selftests/mm: add mseal test for no-discard madvise
> > >
> > > 4d1b3416659be70a2251b494e85e25978de06519
> > >     mm: move can_modify_vma to mm/vma.h
> > >
> > >  4a2dd02b09160ee43f96c759fafa7b56dfc33816
> > >   mm/mprotect: replace can_modify_mm with can_modify_vma
> > >
> > > 23c57d1fa2b9530e38f7964b4e457fed5a7a0ae8
> > >       mseal: replace can_modify_mm_madv with a vma variant
> > >
> > > f28bdd1b17ec187eaa34845814afaaff99832762
> > >    selftests/mm: add more mseal traversal tests
> > >
> > > There will be merge conflicts, I  can backport them to 5.10 and test
> > > to help the backporting process.
> >
> > 5.10 or 6.10?
> >
> 6.10.
> 
> > And why 6.10?  If you look at the front page of kernel.org you will see
> > that 6.10 is now end-of-life, so why does that kernel matter to you
> > anymore?
> >
> OK, I didn't know that. Less work is nice :-)

So, now that you don't care about 6.10.y, what about 6.11.y?  Are any of
these actually bugfixes that people need?

thanks,

greg k-h

Re: backport mseal and mseal_test to 6.10

[Jeff Xu]

On Mon, Oct 14, 2024 at 9:12 AM Greg KH <gregkh@linuxfoundation.org> wrote:
>
> On Mon, Oct 14, 2024 at 08:27:29AM -0700, Jeff Xu wrote:
> > On Sun, Oct 13, 2024 at 10:54 PM Greg KH <gregkh@linuxfoundation.org> wrote:
> > >
> > > On Sun, Oct 13, 2024 at 10:17:48PM -0700, Jeff Xu wrote:
> > > > Hi Greg,
> > > >
> > > > How are you?
> > > >
> > > > What is the process to backport Pedro's recent mseal fixes to 6.10 ?
> > >
> > > Please read:
> > >     https://www.kernel.org/doc/html/latest/process/stable-kernel-rules.html
> > > for how all of this works :)
> > >
> > > > Specifically those 5 commits:
> > > >
> > > > 67203f3f2a63d429272f0c80451e5fcc469fdb46
> > > >     selftests/mm: add mseal test for no-discard madvise
> > > >
> > > > 4d1b3416659be70a2251b494e85e25978de06519
> > > >     mm: move can_modify_vma to mm/vma.h
> > > >
> > > >  4a2dd02b09160ee43f96c759fafa7b56dfc33816
> > > >   mm/mprotect: replace can_modify_mm with can_modify_vma
> > > >
> > > > 23c57d1fa2b9530e38f7964b4e457fed5a7a0ae8
> > > >       mseal: replace can_modify_mm_madv with a vma variant
> > > >
> > > > f28bdd1b17ec187eaa34845814afaaff99832762
> > > >    selftests/mm: add more mseal traversal tests
> > > >
> > > > There will be merge conflicts, I  can backport them to 5.10 and test
> > > > to help the backporting process.
> > >
> > > 5.10 or 6.10?
> > >
> > 6.10.
> >
> > > And why 6.10?  If you look at the front page of kernel.org you will see
> > > that 6.10 is now end-of-life, so why does that kernel matter to you
> > > anymore?
> > >
> > OK, I didn't know that. Less work is nice :-)
>
> So, now that you don't care about 6.10.y, what about 6.11.y?  Are any of
> these actually bugfixes that people need?
>
Oh, yes. It would be great to backport those 5 mentioned to 6.11.y.
I don't know what will be the lifetime of 6.11.y, but keeping mseal's
semantics consistent across releases is important.

Thanks
-Jeff

> thanks,
>
> greg k-h

Re: backport mseal and mseal_test to 6.10

[Greg KH]

On Mon, Oct 14, 2024 at 09:19:55AM -0700, Jeff Xu wrote:
> On Mon, Oct 14, 2024 at 9:12 AM Greg KH <gregkh@linuxfoundation.org> wrote:
> >
> > On Mon, Oct 14, 2024 at 08:27:29AM -0700, Jeff Xu wrote:
> > > On Sun, Oct 13, 2024 at 10:54 PM Greg KH <gregkh@linuxfoundation.org> wrote:
> > > >
> > > > On Sun, Oct 13, 2024 at 10:17:48PM -0700, Jeff Xu wrote:
> > > > > Hi Greg,
> > > > >
> > > > > How are you?
> > > > >
> > > > > What is the process to backport Pedro's recent mseal fixes to 6.10 ?
> > > >
> > > > Please read:
> > > >     https://www.kernel.org/doc/html/latest/process/stable-kernel-rules.html
> > > > for how all of this works :)
> > > >
> > > > > Specifically those 5 commits:
> > > > >
> > > > > 67203f3f2a63d429272f0c80451e5fcc469fdb46
> > > > >     selftests/mm: add mseal test for no-discard madvise
> > > > >
> > > > > 4d1b3416659be70a2251b494e85e25978de06519
> > > > >     mm: move can_modify_vma to mm/vma.h
> > > > >
> > > > >  4a2dd02b09160ee43f96c759fafa7b56dfc33816
> > > > >   mm/mprotect: replace can_modify_mm with can_modify_vma
> > > > >
> > > > > 23c57d1fa2b9530e38f7964b4e457fed5a7a0ae8
> > > > >       mseal: replace can_modify_mm_madv with a vma variant
> > > > >
> > > > > f28bdd1b17ec187eaa34845814afaaff99832762
> > > > >    selftests/mm: add more mseal traversal tests
> > > > >
> > > > > There will be merge conflicts, I  can backport them to 5.10 and test
> > > > > to help the backporting process.
> > > >
> > > > 5.10 or 6.10?
> > > >
> > > 6.10.
> > >
> > > > And why 6.10?  If you look at the front page of kernel.org you will see
> > > > that 6.10 is now end-of-life, so why does that kernel matter to you
> > > > anymore?
> > > >
> > > OK, I didn't know that. Less work is nice :-)
> >
> > So, now that you don't care about 6.10.y, what about 6.11.y?  Are any of
> > these actually bugfixes that people need?
> >
> Oh, yes. It would be great to backport those 5 mentioned to 6.11.y.

Why, are they bugfixes?

> I don't know what will be the lifetime of 6.11.y, but keeping mseal's
> semantics consistent across releases is important.

Stable kernels last until the next release happens, like has been
happening for 15+ years now, nothing new here :)

If you wish to have patches backported to stable kernels, please read
    https://www.kernel.org/doc/html/latest/process/stable-kernel-rules.html
for how to do this properly.

thanks,

greg k-h

Re: backport mseal and mseal_test to 6.10

[Jeff Xu]

On Mon, Oct 14, 2024 at 9:23 AM Greg KH <gregkh@linuxfoundation.org> wrote:
>
> On Mon, Oct 14, 2024 at 09:19:55AM -0700, Jeff Xu wrote:
> > On Mon, Oct 14, 2024 at 9:12 AM Greg KH <gregkh@linuxfoundation.org> wrote:
> > >
> > > On Mon, Oct 14, 2024 at 08:27:29AM -0700, Jeff Xu wrote:
> > > > On Sun, Oct 13, 2024 at 10:54 PM Greg KH <gregkh@linuxfoundation.org> wrote:
> > > > >
> > > > > On Sun, Oct 13, 2024 at 10:17:48PM -0700, Jeff Xu wrote:
> > > > > > Hi Greg,
> > > > > >
> > > > > > How are you?
> > > > > >
> > > > > > What is the process to backport Pedro's recent mseal fixes to 6.10 ?
> > > > >
> > > > > Please read:
> > > > >     https://www.kernel.org/doc/html/latest/process/stable-kernel-rules.html
> > > > > for how all of this works :)
> > > > >
> > > > > > Specifically those 5 commits:
> > > > > >
> > > > > > 67203f3f2a63d429272f0c80451e5fcc469fdb46
> > > > > >     selftests/mm: add mseal test for no-discard madvise
> > > > > >
> > > > > > 4d1b3416659be70a2251b494e85e25978de06519
> > > > > >     mm: move can_modify_vma to mm/vma.h
> > > > > >
> > > > > >  4a2dd02b09160ee43f96c759fafa7b56dfc33816
> > > > > >   mm/mprotect: replace can_modify_mm with can_modify_vma
> > > > > >
> > > > > > 23c57d1fa2b9530e38f7964b4e457fed5a7a0ae8
> > > > > >       mseal: replace can_modify_mm_madv with a vma variant
> > > > > >
> > > > > > f28bdd1b17ec187eaa34845814afaaff99832762
> > > > > >    selftests/mm: add more mseal traversal tests
> > > > > >
> > > > > > There will be merge conflicts, I  can backport them to 5.10 and test
> > > > > > to help the backporting process.
> > > > >
> > > > > 5.10 or 6.10?
> > > > >
> > > > 6.10.
> > > >
> > > > > And why 6.10?  If you look at the front page of kernel.org you will see
> > > > > that 6.10 is now end-of-life, so why does that kernel matter to you
> > > > > anymore?
> > > > >
> > > > OK, I didn't know that. Less work is nice :-)
> > >
> > > So, now that you don't care about 6.10.y, what about 6.11.y?  Are any of
> > > these actually bugfixes that people need?
> > >
> > Oh, yes. It would be great to backport those 5 mentioned to 6.11.y.
>
> Why, are they bugfixes?
>
Yes. For performance, there are 5% impact with mprotect/madvise.

> > I don't know what will be the lifetime of 6.11.y, but keeping mseal's
> > semantics consistent across releases is important.
>
> Stable kernels last until the next release happens, like has been
> happening for 15+ years now, nothing new here :)
>
Does it mean that with 6.12, 6.11.y will be EOL soon ? say in the next
few months?
(Sorry that I didn't know much about linux release cycle. )

> If you wish to have patches backported to stable kernels, please read
>     https://www.kernel.org/doc/html/latest/process/stable-kernel-rules.html
> for how to do this properly.
>
> thanks,
>
> greg k-h

Re: backport mseal and mseal_test to 6.10

[Pedro Falcato]

On Mon, Oct 14, 2024 at 6:26 PM Jeff Xu <jeffxu@chromium.org> wrote:
>
> On Mon, Oct 14, 2024 at 9:23 AM Greg KH <gregkh@linuxfoundation.org> wrote:
> >
> > On Mon, Oct 14, 2024 at 09:19:55AM -0700, Jeff Xu wrote:
> > > On Mon, Oct 14, 2024 at 9:12 AM Greg KH <gregkh@linuxfoundation.org> wrote:
> > > >
> > > > On Mon, Oct 14, 2024 at 08:27:29AM -0700, Jeff Xu wrote:
> > > > > On Sun, Oct 13, 2024 at 10:54 PM Greg KH <gregkh@linuxfoundation.org> wrote:
> > > > > >
> > > > > > On Sun, Oct 13, 2024 at 10:17:48PM -0700, Jeff Xu wrote:
> > > > > > > Hi Greg,
> > > > > > >
> > > > > > > How are you?
> > > > > > >
> > > > > > > What is the process to backport Pedro's recent mseal fixes to 6.10 ?
> > > > > >
> > > > > > Please read:
> > > > > >     https://www.kernel.org/doc/html/latest/process/stable-kernel-rules.html
> > > > > > for how all of this works :)
> > > > > >
> > > > > > > Specifically those 5 commits:
> > > > > > >
> > > > > > > 67203f3f2a63d429272f0c80451e5fcc469fdb46
> > > > > > >     selftests/mm: add mseal test for no-discard madvise
> > > > > > >
> > > > > > > 4d1b3416659be70a2251b494e85e25978de06519
> > > > > > >     mm: move can_modify_vma to mm/vma.h
> > > > > > >
> > > > > > >  4a2dd02b09160ee43f96c759fafa7b56dfc33816
> > > > > > >   mm/mprotect: replace can_modify_mm with can_modify_vma
> > > > > > >
> > > > > > > 23c57d1fa2b9530e38f7964b4e457fed5a7a0ae8
> > > > > > >       mseal: replace can_modify_mm_madv with a vma variant
> > > > > > >
> > > > > > > f28bdd1b17ec187eaa34845814afaaff99832762
> > > > > > >    selftests/mm: add more mseal traversal tests
> > > > > > >
> > > > > > > There will be merge conflicts, I  can backport them to 5.10 and test
> > > > > > > to help the backporting process.
> > > > > >
> > > > > > 5.10 or 6.10?
> > > > > >
> > > > > 6.10.
> > > > >
> > > > > > And why 6.10?  If you look at the front page of kernel.org you will see
> > > > > > that 6.10 is now end-of-life, so why does that kernel matter to you
> > > > > > anymore?
> > > > > >
> > > > > OK, I didn't know that. Less work is nice :-)
> > > >
> > > > So, now that you don't care about 6.10.y, what about 6.11.y?  Are any of
> > > > these actually bugfixes that people need?
> > > >
> > > Oh, yes. It would be great to backport those 5 mentioned to 6.11.y.
> >
> > Why, are they bugfixes?
> >
> Yes. For performance, there are 5% impact with mprotect/madvise.

They're not bugfixes, they fix a performance regression. As far as I'm
aware, they do not fit the criteria for -stable inclusion.

But ofc Greg might know better :)

-- 
Pedro

Re: backport mseal and mseal_test to 6.10

[Greg KH]

On Mon, Oct 14, 2024 at 10:25:54AM -0700, Jeff Xu wrote:
> On Mon, Oct 14, 2024 at 9:23 AM Greg KH <gregkh@linuxfoundation.org> wrote:
> >
> > On Mon, Oct 14, 2024 at 09:19:55AM -0700, Jeff Xu wrote:
> > > On Mon, Oct 14, 2024 at 9:12 AM Greg KH <gregkh@linuxfoundation.org> wrote:
> > > >
> > > > On Mon, Oct 14, 2024 at 08:27:29AM -0700, Jeff Xu wrote:
> > > > > On Sun, Oct 13, 2024 at 10:54 PM Greg KH <gregkh@linuxfoundation.org> wrote:
> > > > > >
> > > > > > On Sun, Oct 13, 2024 at 10:17:48PM -0700, Jeff Xu wrote:
> > > > > > > Hi Greg,
> > > > > > >
> > > > > > > How are you?
> > > > > > >
> > > > > > > What is the process to backport Pedro's recent mseal fixes to 6.10 ?
> > > > > >
> > > > > > Please read:
> > > > > >     https://www.kernel.org/doc/html/latest/process/stable-kernel-rules.html
> > > > > > for how all of this works :)
> > > > > >
> > > > > > > Specifically those 5 commits:
> > > > > > >
> > > > > > > 67203f3f2a63d429272f0c80451e5fcc469fdb46
> > > > > > >     selftests/mm: add mseal test for no-discard madvise
> > > > > > >
> > > > > > > 4d1b3416659be70a2251b494e85e25978de06519
> > > > > > >     mm: move can_modify_vma to mm/vma.h
> > > > > > >
> > > > > > >  4a2dd02b09160ee43f96c759fafa7b56dfc33816
> > > > > > >   mm/mprotect: replace can_modify_mm with can_modify_vma
> > > > > > >
> > > > > > > 23c57d1fa2b9530e38f7964b4e457fed5a7a0ae8
> > > > > > >       mseal: replace can_modify_mm_madv with a vma variant
> > > > > > >
> > > > > > > f28bdd1b17ec187eaa34845814afaaff99832762
> > > > > > >    selftests/mm: add more mseal traversal tests
> > > > > > >
> > > > > > > There will be merge conflicts, I  can backport them to 5.10 and test
> > > > > > > to help the backporting process.
> > > > > >
> > > > > > 5.10 or 6.10?
> > > > > >
> > > > > 6.10.
> > > > >
> > > > > > And why 6.10?  If you look at the front page of kernel.org you will see
> > > > > > that 6.10 is now end-of-life, so why does that kernel matter to you
> > > > > > anymore?
> > > > > >
> > > > > OK, I didn't know that. Less work is nice :-)
> > > >
> > > > So, now that you don't care about 6.10.y, what about 6.11.y?  Are any of
> > > > these actually bugfixes that people need?
> > > >
> > > Oh, yes. It would be great to backport those 5 mentioned to 6.11.y.
> >
> > Why, are they bugfixes?
> >
> Yes. For performance, there are 5% impact with mprotect/madvise.

That's not a bugfix, but we do sometimes take performance improvements
if it's really needed and the maintainer is willing to do the backport
for us.

> > > I don't know what will be the lifetime of 6.11.y, but keeping mseal's
> > > semantics consistent across releases is important.
> >
> > Stable kernels last until the next release happens, like has been
> > happening for 15+ years now, nothing new here :)
> >
> Does it mean that with 6.12, 6.11.y will be EOL soon ?

Yes.

> say in the next few months?

Yes.

> (Sorry that I didn't know much about linux release cycle. )

It's well documented, please see the Documentation/process/2.Process.rst
file for details.  If you have questions after that, please let us know.

thanks,

greg k-h