Re: [PATCH] Revert "x86/tsc: Consolidate init code"

[Ville Syrjälä]

On Mon, Sep 10, 2018 at 06:23:49PM +0200, Thomas Gleixner wrote:
> On Mon, 10 Sep 2018, Ville Syrjälä wrote:
> > On Mon, Sep 10, 2018 at 02:48:45PM +0200, Thomas Gleixner wrote:
> > > On Mon, 10 Sep 2018, Ville Syrjala wrote:
> > > I asked for that before and I really do not understand why you do not even
> > > make an attempt to report an issue first and allow the developers to work
> > > with you to figure out what exactly is the problem. All you do is to send
> > > an revert patch with a changelog which describes symptoms and probably
> > > breaks more than it cures. Not really helpful, really.
> > 
> > You're reading way too much into this. The revert is just a point to
> > start the conversion. I've found that it's the best way to get the
> > attention of the relevant developers. Other kind of regression
> > reports have an unfortunate habit of disappearing into /dev/null.
> 
> 1) My workflow makes things tagged as BUG and REGRESSION urgent
>    automatically while [PATCH] just is queued to the normal pile of
>    backlog, i.e. at the end. It just sprang into my eyes by chance, but in
>    general you might just get the contrary of what you are looking for.

Ah. Might be nice to document that somewhere. I might have to type up
that git-regression tool for myself, because I'm lazy.

> 
> 2) A proper bug report with proper information (it's documented what should
>    be provided), is way more worth than a patch with a mostly useless
>    change log, which forces me to ask for the proper information instead of
>    having it right away.

I do agree that not having to ask for more information would be nice,
but hard to generalize because every subsystem needs different things.

In this case you asked for the dmesg, which isn't even mentioned in
Documentation/admin-guide/reporting-bugs.rst as far as I can see.
So I'm not quite sure which documentation you're referring to here.

-- 
Ville Syrjälä
Intel