[PATCH 0/2] add notes about license implications of bundled initramfs

[PATCH 0/2] add notes about license implications of bundled initramfs

[Daniel Wagenknecht]

As discussed on IRC the license implications of using
INITRAMFS_IMAGE_BUNDLE are not necessarily obvious. This patchseries
adds notes about them in the relevant places.

Source of this interpretation of the license implications is the second
paragraph of
https://www.kernel.org/doc/html/latest/filesystems/ramfs-rootfs-initramfs.html#external-initramfs-images

I placed them where I saw fit, although this leads to multiple `note`
blocks in direct succession. If this should be avoided a more thorough
rewrite of those sections might be necessary.

Daniel Wagenknecht (2):
  common-tasks: add note about license implications of bundled initramfs
  ref-manual: add note about license implications of bundled initramfs

 documentation/dev-manual/common-tasks.rst | 6 ++++++
 documentation/ref-manual/variables.rst    | 6 ++++++
 2 files changed, 12 insertions(+)

-- 
2.25.1

[PATCH 1/2] common-tasks: add note about license implications of bundled initramfs

[Daniel Wagenknecht]

Ensure users are aware that usage of INITRAMFS_IMAGE_BUNDLE is only
appropiate for use with GPLv2 compatible software.

Signed-off-by: Daniel Wagenknecht <dwagenknecht@emlix.com>
---
 documentation/dev-manual/common-tasks.rst | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/documentation/dev-manual/common-tasks.rst b/documentation/dev-manual/common-tasks.rst
index bd8bbecb3..b81f51bf8 100644
--- a/documentation/dev-manual/common-tasks.rst
+++ b/documentation/dev-manual/common-tasks.rst
@@ -3886,6 +3886,12 @@ Follow these steps to create an initramfs image:
    variable, allowing the initramfs image to be built into the kernel
    normally.
 
+   .. note::
+
+      Bundling the initramfs with the kernel conflates the code in the initramfs
+      with the GPLv2 licensed Linux kernel binary. Thus only GPLv2 compatible
+      software may be part of a bundled initramfs.
+
    .. note::
 
       If you choose to not bundle the initramfs image with the kernel
-- 
2.25.1

[PATCH 2/2] ref-manual: add note about license implications of bundled initramfs

[Daniel Wagenknecht]

Ensure users are aware that usage of INITRAMFS_IMAGE_BUNDLE is only
appropiate for use with GPLv2 compatible software.

Signed-off-by: Daniel Wagenknecht <dwagenknecht@emlix.com>
---
 documentation/ref-manual/variables.rst | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/documentation/ref-manual/variables.rst b/documentation/ref-manual/variables.rst
index 28eb5f233..4496e2f46 100644
--- a/documentation/ref-manual/variables.rst
+++ b/documentation/ref-manual/variables.rst
@@ -3587,6 +3587,12 @@ system and gives an overview of their function and contents.
       :term:`CONFIG_INITRAMFS_SOURCE` kernel
       feature.
 
+      .. note::
+
+         Bundling the initramfs with the kernel conflates the code in the
+         initramfs with the GPLv2 licensed Linux kernel binary. Thus only GPLv2
+         compatible software may be part of a bundled initramfs.
+
       .. note::
 
          Using an extra compilation pass to bundle the initramfs avoids a
-- 
2.25.1

Re: [docs] [PATCH 0/2] add notes about license implications of bundled initramfs

[Michael Opdenacker]

Hi Daniel,

Many thanks for the patches!

On 9/23/21 10:34 AM, Daniel Wagenknecht wrote:
> As discussed on IRC the license implications of using
> INITRAMFS_IMAGE_BUNDLE are not necessarily obvious. This patchseries
> adds notes about them in the relevant places.
>
> Source of this interpretation of the license implications is the second
> paragraph of
> https://www.kernel.org/doc/html/latest/filesystems/ramfs-rootfs-initramfs.html#external-initramfs-images

I agree. Good to point this out.
>
> I placed them where I saw fit, although this leads to multiple `note`
> blocks in direct succession. If this should be avoided a more thorough
> rewrite of those sections might be necessary.


Right. We might find a way to improve this later, but merging your
changes is important too.

Reviewed-by: Michael Opdenacker <michael.opdenacker@bootlin.com>
Merged into master-next.

Thanks again,

Cheers,
Michael
-- 

Michael Opdenacker, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com