Re: More detailed text about bisecting Linux kernel regression -- request for comments and help

[Bagas Sanjaya <bagasdotme@gmail.com>]

[-- Attachment #1: Type: text/plain, Size: 1330 bytes --]

On Sun, Jan 28, 2024 at 06:28:52AM +0100, Thorsten Leemhuis wrote:
> On 28.01.24 03:18, Bagas Sanjaya wrote:
> > On Wed, Jan 24, 2024 at 01:19:16PM +0100, Thorsten Leemhuis wrote:
> >> .. _bisectstart_bissbs:
> >>
> >> * Start the bisection and tell Git about the versions earlier
> >> established as 'good' and 'bad'::
> >>
> >>     cd ~/linux/
> >>     git bisect start
> >>     git bisect good v6.0
> >>     git bisect bad v6.1.5
> > 
> > If stable release tag is supplied instead as "good" version instead (e.g.
> > v6.0.1), as in many regression cases, git will ask to test the merge base
> > instead, which is corresponding mainline release (in this case v6.0).
> 
> That should not happen if people follow the guide, as this is avoided by
> an earlier step:
> 
> "'"
> .. _rangecheck_bissbs:
> 
> * Determine the kernel versions considered 'good' and 'bad' throughout
> this guide:
> 
> [...]
> 
>   * Some function stopped working when updating from 6.0.11 to 6.1.4?
> Then for the time being consider 'v6.0' as the last 'good' version and
> 'v6.1.4' as the 'bad' one. Note, it is at this point an assumption that
> 6.0 is fine that will be checked later.
> 
> "'"
> 

Oops, I didn't see that context above. Thanks anyway.

-- 
An old man doll... just what I always wanted! - Clara

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 228 bytes --]