Add documentation for SBOM#676
Conversation
|
New warnings found with rstcheck: |
1 similar comment
|
New warnings found with rstcheck: |
|
New warnings found with rstcheck: |
16bcbfe to
5732a1e
Compare
|
New warnings found with rstcheck: |
Add release artefacts SBOM information for AM64X,AM62X,AM62PX and AM62LX devices. Signed-off-by: Yogesh Hegde <y-hegde@ti.com>
9338c00 to
b9ad5de
Compare
| @@ -0,0 +1,178 @@ | |||
| .. _how-to-work-with-sbom: | |||
There was a problem hiding this comment.
This newly added section doesn't seem to follow the guidelines of section headers
https://github.com/TexasInstruments/processor-sdk-doc?tab=contributing-ov-file#headings--sections
There was a problem hiding this comment.
Fixed.
But few questions,
- Why do we follow Python Developer's Guide for documenting , what benefits do we get by following this guide just for heading ?
- rst does not put any limitations on the section headers, the doc says whatever we encounter 1st is h1 and so on.
- Since for h1 and h2 the characters underlining are different why do we need an overline just for h1 and h2 ? Since it is just for aesthetic purposes why not overline for all or no overline for all? Why not simplify and use one format for headings?
- I see alot of
.rstfiles not following the format, is there a plan to fix those documents?
There was a problem hiding this comment.
- That's the one highlighted by sphinx in their documentation - https://www.sphinx-doc.org/en/master/usage/restructuredtext/basics.html#sections
- Yeah, that's a known issue. Infact that inconsistency led to encorfing this under Contribution guidelines - https://github.com/TexasInstruments/processor-sdk-doc?tab=contributing-ov-file#headings--sections. There were internal jira tickets created for few of those inconsistent rst files but later most of them were fixed by @StaticRocket, @praneethbajjuri and myself in past. Feel free to submit a patch incase you happen to see any rst's still having inconsistent usage of section headers.
| SPDX 3.0 is generated by default when building |__SDK_FULL_NAME__| Yocto, no extra steps required. | ||
| If you require additional vulnerability information, follow these steps: | ||
|
|
||
| 1. Add the following line to your ``local.conf``: |
There was a problem hiding this comment.
role file can be used to highlight the same
| 1. Add the following line to your ``local.conf``: | |
| 1. Add the following line to your :file:`local.conf`: |
There was a problem hiding this comment.
What is the difference between
``local.conf``
&
:file:`local.conf`
How is it rendered different to the user?
There was a problem hiding this comment.
In Sphinx (& reStructuredText), these 2 syntaxes serve different semantic purposes, even if they sometimes look similar in the final output. Think of it as the difference between "just some code / variable" and "specifically a file."
:file: role specifically for filenames and paths. It tells Sphinx (& the reader) exactly what the data is.
In terms of rendering, both usually result in a monospaced (fixed-width) font but the way they appear can vary based on the theme you are using
| 1. Start with the build instructions in :ref:`Processor SDK - Building the SDK with Yocto <building-the-sdk-with-yocto>` | ||
| 2. After cloning ``oe-layersetup``, uncomment the ``meta-cyclonedx`` line in | ||
| the layer configuration file, for example: | ||
|
|
There was a problem hiding this comment.
Any reasons we aren't pushing a new oe-config file to oe-layersetup dedicated for SBOMs?
We can avoid the following manual local changes & improve the user experience
There was a problem hiding this comment.
Here are few reasons,
- CycloneDX SBOM is not default format for SBOM, but there might be some users who want to use the format.
- Duplication of oe-config, there are 4 oe-config for 12.00.00 release, creating new oe-config just for cyclonedx would take that number to 8 since we have to give each oe-config with SPDX and CycloneDX SBOM generation.
- 8 oe-config will lead to more user confusion instead of improving user experience because for users who just want sane defaults they will be confused which oe-config to use.
- oe-layersetup does not allow us to
include files / config fragments, while kas (yaml) and repo (xml) both have include files feature where we can create fragments and user can include/exclude fragments enabling/disabling features. This would simplify both maintenance and improve user experience.
With these constraints, I believe this is the best course, since it is just uncommenting a line for the user and they also see less oe-configs.
There was a problem hiding this comment.
I agree with this approach for the 12.0 release. Since we are still in the evaluation phase for SBOMs, it makes sense to avoid config bloat and keep the user experience simple for now.
However, SBOMs are the way to go in the future. Moving forward, we should look into how we can incorporate them into the default SDK build flow...
Add How to guide for working with SBOM's with sections * Generating SBOM in SPDX and CycloneDX format * Tools and references for Working with SBOM i.e visualizing, merging, modifying SBOMs Signed-off-by: Yogesh Hegde <y-hegde@ti.com>
Add link to working with SBOM in release specific section Signed-off-by: Yogesh Hegde <y-hegde@ti.com>
Add release artefacts SBOM information to release notes for AM62DX device. Signed-off-by: Yogesh Hegde <y-hegde@ti.com>
There was a problem hiding this comment.
Could you use auto-numbered lists for this instead of manual numbering?
feat(linux): Add how to guide for working with SBOMs
Add How to guide for working with SBOM's with sections
modifying SBOMs
feat(linux): Add SBOM section to release notes
Add release artefacts SBOM information for AM64X,AM62X,AM62PX and AM62LX
devices.
Doc Link - https://yogeshhegde.github.io/processor-sdk-doc/processor-sdk-linux-AM62LX/esd/docs/master/linux/How_to_Guides/FAQ/How_to_work_with_SBOM.html