Re: [PATCHv2 0/4] tools/memory-model: Define more of LKMM in tools/memory-model

[Jonas Oberhauser]

Am 6/8/2024 um 3:00 AM schrieb Alan Stern:
> On Wed, Jun 05, 2024 at 09:58:42PM +0200, Jonas Oberhauser wrote:
> > Am 6/4/2024 um 7:56 PM schrieb Alan Stern:
> > > Just to clarify: Your first step encompasses patches 1 - 3, and the
> > > second step is patch 4.  The first three patches can be applied now, but
> > > the last one needs to wait until herd7 has been updated.  Is this all
> > > correct?
> >
> > Exactly.
>
> With regard to patch 4, how much thought have you and Hernan given to
> backward compatibility?  Once herd7 is changed, old memory model files
> will no longer work correctly.

Yes, Akira pointed this out too.

My thought back then was to update herd now, and wait with the fourth 
patch for a while until most people who run the LKMM would have 
upgraded. However...

> To avoid being so disruptive, perhaps the changes to herd7 should be
> under control of a new command-line or config-file switch.  If the
> switch is enabled, the new simplified code gets used; otherwise herd7
> would continue to use its old built-in rules for special tags
... I like that idea a lot better actually. I think it needs to be 
placed into the files, or people will get strange, silent results.

But then I think we need to make sure that there's no internal behaviors 
left. I don't want to introduce more flags in the future to turn off 
other internal behavior.

jonas