mbox series

[SRU,Jammy,0/1] Fix for LP:1972899 generic kernel

Message ID 20220511155405.250107-1-joseph.salisbury@canonical.com
Headers show
Series Fix for LP:1972899 generic kernel | expand

Message

Joseph Salisbury May 11, 2022, 3:54 p.m. UTC
BugLink: https://bugs.launchpad.net/bugs/1972899

[SRU Justification]
A new commit in the Jammy 5.15.0-29.30 kernel is causing a build failure
for the real-time kernel and may affect other derivatives.

After a bisect, I found this commit is causing the real-time kernel build
failure:
dc66c1b8a2be1 ("x86/pkru: Remove useless include")

Commit dc66c1b8a2be1 was added in 5.16-rc1 and not cc'd to stable. It was
picked into Jammy to enable AMX support. See bug LP:1967750.

The following commit resolves this regression by fixing up the includes:
35fa745286ac4 ("x86/mm: Include spinlock_t definition in pgtable")

Commit 35fa745286ac4 was merged into mainline 5.17-rc1.

This commit was already requested for the real-time kernel.  However,
we should also include it in the generic kernel to prevent other 
codebases from hitting this issue.


[Fix]
35fa745286ac4 ("x86/mm: Include spinlock_t definition in pgtable")

[Test case]
Real-time kernel builds are now working after picking this commit.

[Regression potential]
Low. The new commit is just including a header file.

Sebastian Andrzej Siewior (1):
  x86/mm: Include spinlock_t definition in pgtable.

 arch/x86/include/asm/pgtable.h | 1 +
 1 file changed, 1 insertion(+)

--
2.34.1

Comments

Tim Gardner May 11, 2022, 4:22 p.m. UTC | #1
Acked-by: Tim Gardner <tim.gardner@canonical.com>

On 5/11/22 09:54, Joseph Salisbury wrote:
> BugLink: https://bugs.launchpad.net/bugs/1972899
> 
> [SRU Justification]
> A new commit in the Jammy 5.15.0-29.30 kernel is causing a build failure
> for the real-time kernel and may affect other derivatives.
> 
> After a bisect, I found this commit is causing the real-time kernel build
> failure:
> dc66c1b8a2be1 ("x86/pkru: Remove useless include")
> 
> Commit dc66c1b8a2be1 was added in 5.16-rc1 and not cc'd to stable. It was
> picked into Jammy to enable AMX support. See bug LP:1967750.
> 
> The following commit resolves this regression by fixing up the includes:
> 35fa745286ac4 ("x86/mm: Include spinlock_t definition in pgtable")
> 
> Commit 35fa745286ac4 was merged into mainline 5.17-rc1.
> 
> This commit was already requested for the real-time kernel.  However,
> we should also include it in the generic kernel to prevent other
> codebases from hitting this issue.
> 
> 
> [Fix]
> 35fa745286ac4 ("x86/mm: Include spinlock_t definition in pgtable")
> 
> [Test case]
> Real-time kernel builds are now working after picking this commit.
> 
> [Regression potential]
> Low. The new commit is just including a header file.
> 
> Sebastian Andrzej Siewior (1):
>    x86/mm: Include spinlock_t definition in pgtable.
> 
>   arch/x86/include/asm/pgtable.h | 1 +
>   1 file changed, 1 insertion(+)
> 
> --
> 2.34.1
> 
>
Philip Cox May 11, 2022, 6:45 p.m. UTC | #2
Acked-By: Philip Cox <philip.cox@canonical.com>

On 2022-05-11 11:54 a.m., Joseph Salisbury wrote:
> BugLink: https://bugs.launchpad.net/bugs/1972899
>
> [SRU Justification]
> A new commit in the Jammy 5.15.0-29.30 kernel is causing a build failure
> for the real-time kernel and may affect other derivatives.
>
> After a bisect, I found this commit is causing the real-time kernel build
> failure:
> dc66c1b8a2be1 ("x86/pkru: Remove useless include")
>
> Commit dc66c1b8a2be1 was added in 5.16-rc1 and not cc'd to stable. It was
> picked into Jammy to enable AMX support. See bug LP:1967750.
>
> The following commit resolves this regression by fixing up the includes:
> 35fa745286ac4 ("x86/mm: Include spinlock_t definition in pgtable")
>
> Commit 35fa745286ac4 was merged into mainline 5.17-rc1.
>
> This commit was already requested for the real-time kernel.  However,
> we should also include it in the generic kernel to prevent other
> codebases from hitting this issue.
>
>
> [Fix]
> 35fa745286ac4 ("x86/mm: Include spinlock_t definition in pgtable")
>
> [Test case]
> Real-time kernel builds are now working after picking this commit.
>
> [Regression potential]
> Low. The new commit is just including a header file.
>
> Sebastian Andrzej Siewior (1):
>    x86/mm: Include spinlock_t definition in pgtable.
>
>   arch/x86/include/asm/pgtable.h | 1 +
>   1 file changed, 1 insertion(+)
>
> --
> 2.34.1
>
>
Kleber Souza May 26, 2022, 4:34 p.m. UTC | #3
On 11.05.22 17:54, Joseph Salisbury wrote:
> BugLink: https://bugs.launchpad.net/bugs/1972899
> 
> [SRU Justification]
> A new commit in the Jammy 5.15.0-29.30 kernel is causing a build failure
> for the real-time kernel and may affect other derivatives.
> 
> After a bisect, I found this commit is causing the real-time kernel build
> failure:
> dc66c1b8a2be1 ("x86/pkru: Remove useless include")
> 
> Commit dc66c1b8a2be1 was added in 5.16-rc1 and not cc'd to stable. It was
> picked into Jammy to enable AMX support. See bug LP:1967750.
> 
> The following commit resolves this regression by fixing up the includes:
> 35fa745286ac4 ("x86/mm: Include spinlock_t definition in pgtable")
> 
> Commit 35fa745286ac4 was merged into mainline 5.17-rc1.
> 
> This commit was already requested for the real-time kernel.  However,
> we should also include it in the generic kernel to prevent other
> codebases from hitting this issue.
> 
> 
> [Fix]
> 35fa745286ac4 ("x86/mm: Include spinlock_t definition in pgtable")
> 
> [Test case]
> Real-time kernel builds are now working after picking this commit.
> 
> [Regression potential]
> Low. The new commit is just including a header file.
> 
> Sebastian Andrzej Siewior (1):
>    x86/mm: Include spinlock_t definition in pgtable.

Applied to jammy:linux.

Thanks,
Kleber

> 
>   arch/x86/include/asm/pgtable.h | 1 +
>   1 file changed, 1 insertion(+)
> 
> --
> 2.34.1
> 
>