[PATCH v2 0/6] docs: verify/bisect: fine tuning, testing fixes, build host, order

Thorsten Leemhuis posted 6 patches 1 year, 10 months ago
.../verify-bugs-and-bisect-regressions.rst    | 597 ++++++++++++------
1 file changed, 417 insertions(+), 180 deletions(-)
[PATCH v2 0/6] docs: verify/bisect: fine tuning, testing fixes, build host, order
Posted by Thorsten Leemhuis 1 year, 10 months ago
A quick series with various improvements for
Documentation/admin-guide/verify-bugs-and-bisect-regressions.rst after
receiving feedback on the text or seeing people struggle with certain
tasks.

* The first two patches contains various fixes and some fine tuning.

* The third is mainly a layout improvement that results in a somewhat
  large patch.

* The fourth adds instructions for later testing of reverts, fixes, or
  newer versions. This was already hinted at; but during some early
  work on Documentation/admin-guide/reporting-issues.rst to better
  reconcile the two it seemed wise to cover this properly here.

* The fifth patch briefly outlines how to build kernels on a different
  host; it came into being after a second user within one week asked for
  this. That also allowed the text to briefly cover cross-compilation.

* The sixth makes users that face a regression within a stable series
  test that series first before they are told to test mainline.
---

Hi! While at it let me mention one more thing unrelated to the changes
that came up where I'm unsure if the current approach by the text was
a wise choice:

* Should the document tell users to avoid mainline during merge windows?

Some input from the community on this would be splendid.

Ciao, Thorsten

P.S.: Not totally sure, but I think this should go into 6.9 as this is
a new document; if not I guess it would be wise to pick the first two,
as they fix bugs.

Thorsten Leemhuis (6):
  docs: verify/bisect: use git switch, tag kernel, and various fixes
  docs: verify/bisect: add and fetch stable branches ahead of time
  docs: verify/bisect: proper headlines and more spacing
  docs: verify/bisect: explain testing reverts, patches and newer code
  docs: verify/bisect: describe how to use a build host
  docs: verify/bisect: stable regressions: first stable, then mainline

 .../verify-bugs-and-bisect-regressions.rst    | 597 ++++++++++++------
 1 file changed, 417 insertions(+), 180 deletions(-)


base-commit: b8cfda5c9065cd619a97c17da081cbfab3b1e756
-- 
2.44.0
Re: [PATCH v2 0/6] docs: verify/bisect: fine tuning, testing fixes, build host, order
Posted by Jonathan Corbet 1 year, 9 months ago
Thorsten Leemhuis <linux@leemhuis.info> writes:

> A quick series with various improvements for
> Documentation/admin-guide/verify-bugs-and-bisect-regressions.rst after
> receiving feedback on the text or seeing people struggle with certain
> tasks.
>
> * The first two patches contains various fixes and some fine tuning.
>
> * The third is mainly a layout improvement that results in a somewhat
>   large patch.
>
> * The fourth adds instructions for later testing of reverts, fixes, or
>   newer versions. This was already hinted at; but during some early
>   work on Documentation/admin-guide/reporting-issues.rst to better
>   reconcile the two it seemed wise to cover this properly here.
>
> * The fifth patch briefly outlines how to build kernels on a different
>   host; it came into being after a second user within one week asked for
>   this. That also allowed the text to briefly cover cross-compilation.
>
> * The sixth makes users that face a regression within a stable series
>   test that series first before they are told to test mainline.
> ---
>
> Hi! While at it let me mention one more thing unrelated to the changes
> that came up where I'm unsure if the current approach by the text was
> a wise choice:
>
> * Should the document tell users to avoid mainline during merge windows?
>
> Some input from the community on this would be splendid.
>
> Ciao, Thorsten
>
> P.S.: Not totally sure, but I think this should go into 6.9 as this is
> a new document; if not I guess it would be wise to pick the first two,
> as they fix bugs.

Nobody seems to be complaining, so I've gone ahead and applied the set
to docs-fixes.

Thanks,

jon