Re: [RFC PATCH] tools/memory-model: Remove (dep ; rfi) from ppo

[Andrea Parri]

> What I do object to is a model that's weaker than any possible sane
> hardware.

Not the first time I hear you calling this out.  And inevitably, every
time, other slogans come to my mind:  "C is not an assembly language",
"No features (ordering) without users", ...

For the record, I won't try to push this patch further; I also have no
plans to touch herd7 internals in order to add the ad-hoc flag for the
(dep ; rfi) thing.  (Maybe others will/can step in here.)

In the meantime, the hope (admittedly, probably vain) is that this RFC
could serve as a further warning or as a reference to those developers
who are quivering to use (dep ; rfi): enjoy it, be careful.

  Andrea