Rephrase the "For MDS" section in core-scheduling.rst for the purpose of
making it clearer what is meant by "kernel memory is still considered
untrusted".
Suggested-by: Vineeth Pillai <[email protected]>
Signed-off-by: Fabio M. De Francesco <[email protected]>
---
Documentation/admin-guide/hw-vuln/core-scheduling.rst | 9 +++++----
1 file changed, 5 insertions(+), 4 deletions(-)
diff --git a/Documentation/admin-guide/hw-vuln/core-scheduling.rst b/Documentation/admin-guide/hw-vuln/core-scheduling.rst
index 7b410aef9c5c..e6b5ceb219ec 100644
--- a/Documentation/admin-guide/hw-vuln/core-scheduling.rst
+++ b/Documentation/admin-guide/hw-vuln/core-scheduling.rst
@@ -181,10 +181,11 @@ Open cross-HT issues that core scheduling does not solve
--------------------------------------------------------
1. For MDS
~~~~~~~~~~
-Core scheduling cannot protect against MDS attacks between an HT running in
-user mode and another running in kernel mode. Even though both HTs run tasks
-which trust each other, kernel memory is still considered untrusted. Such
-attacks are possible for any combination of sibling CPU modes (host or guest mode).
+Core scheduling cannot protect against MDS attacks between the siblings running in
+user mode and the others running in kernel mode. Even though all siblings run tasks
+which trust each other, when the kernel is executing code on behalf of a task, it
+cannot trust the code running in the sibling. Such attacks are possible for any
+combination of sibling CPU modes (host or guest mode).
2. For L1TF
~~~~~~~~~~~
--
2.32.0
On Wed, Jul 21, 2021 at 3:02 PM Fabio M. De Francesco
<[email protected]> wrote:
>
> Rephrase the "For MDS" section in core-scheduling.rst for the purpose of
> making it clearer what is meant by "kernel memory is still considered
> untrusted".
>
> Suggested-by: Vineeth Pillai <[email protected]>
> Signed-off-by: Fabio M. De Francesco <[email protected]>
Reviewed-by: Joel Fernandes (Google) <[email protected]>
thanks,
- Joel
> ---
> Documentation/admin-guide/hw-vuln/core-scheduling.rst | 9 +++++----
> 1 file changed, 5 insertions(+), 4 deletions(-)
>
> diff --git a/Documentation/admin-guide/hw-vuln/core-scheduling.rst b/Documentation/admin-guide/hw-vuln/core-scheduling.rst
> index 7b410aef9c5c..e6b5ceb219ec 100644
> --- a/Documentation/admin-guide/hw-vuln/core-scheduling.rst
> +++ b/Documentation/admin-guide/hw-vuln/core-scheduling.rst
> @@ -181,10 +181,11 @@ Open cross-HT issues that core scheduling does not solve
> --------------------------------------------------------
> 1. For MDS
> ~~~~~~~~~~
> -Core scheduling cannot protect against MDS attacks between an HT running in
> -user mode and another running in kernel mode. Even though both HTs run tasks
> -which trust each other, kernel memory is still considered untrusted. Such
> -attacks are possible for any combination of sibling CPU modes (host or guest mode).
> +Core scheduling cannot protect against MDS attacks between the siblings running in
> +user mode and the others running in kernel mode. Even though all siblings run tasks
> +which trust each other, when the kernel is executing code on behalf of a task, it
> +cannot trust the code running in the sibling. Such attacks are possible for any
> +combination of sibling CPU modes (host or guest mode).
>
> 2. For L1TF
> ~~~~~~~~~~~
> --
> 2.32.0
>
"Fabio M. De Francesco" <[email protected]> writes:
> Rephrase the "For MDS" section in core-scheduling.rst for the purpose of
> making it clearer what is meant by "kernel memory is still considered
> untrusted".
>
> Suggested-by: Vineeth Pillai <[email protected]>
> Signed-off-by: Fabio M. De Francesco <[email protected]>
> ---
> Documentation/admin-guide/hw-vuln/core-scheduling.rst | 9 +++++----
> 1 file changed, 5 insertions(+), 4 deletions(-)
>
> diff --git a/Documentation/admin-guide/hw-vuln/core-scheduling.rst b/Documentation/admin-guide/hw-vuln/core-scheduling.rst
> index 7b410aef9c5c..e6b5ceb219ec 100644
> --- a/Documentation/admin-guide/hw-vuln/core-scheduling.rst
> +++ b/Documentation/admin-guide/hw-vuln/core-scheduling.rst
> @@ -181,10 +181,11 @@ Open cross-HT issues that core scheduling does not solve
> --------------------------------------------------------
> 1. For MDS
> ~~~~~~~~~~
> -Core scheduling cannot protect against MDS attacks between an HT running in
> -user mode and another running in kernel mode. Even though both HTs run tasks
> -which trust each other, kernel memory is still considered untrusted. Such
> -attacks are possible for any combination of sibling CPU modes (host or guest mode).
> +Core scheduling cannot protect against MDS attacks between the siblings running in
> +user mode and the others running in kernel mode. Even though all siblings run tasks
> +which trust each other, when the kernel is executing code on behalf of a task, it
> +cannot trust the code running in the sibling. Such attacks are possible for any
> +combination of sibling CPU modes (host or guest mode).
Applied, thanks. I took the liberty of reflowing that paragraph to keep
the line lengths reasonable...
jon