Messages in this thread | | | Date | Tue, 11 Sep 2018 15:31:51 +0200 | From | Andrea Parri <> | Subject | Re: [PATCH v6 03/14] PM: Introduce an Energy Model management framework |
| |
On Tue, Sep 11, 2018 at 01:32:50PM +0100, Quentin Perret wrote: > Hi Andrea, > > On Tuesday 11 Sep 2018 at 11:34:56 (+0200), Andrea Parri wrote: > > FYI, the directory "tools/memory-model/" provides an "automated > > memory-barriers.txt": in short, you encode your "memory ordering > > questions" into "litmus tests" to be passed to the tool/simulator; > > the tool will then answer with "Yes/No" (plus other information). > > > > Some preparation is required to set up and learn how to use the > > LKMM tools, but once there, I expect them to be more "efficient" > > than reading memory-barriers.txt... ;-) > > Thanks for pointing this out, I'll have a look. > > > Please don't hesitate > > to contact me/the LKMM maintainers if you need help with this. > > And thanks for that too. > > > You'd need some info in order to write down a _well-formed litmus > > test, e.g., matching barrier/synchronization and interested memory > > accesses on the reader side (IAC, the replacement "store-release > > -> store-once+smp_wmb" discussed above is suspicious...). > > Regarding the disccusion above, I was actually planning on removing the > smp_wmb entirely and rely on WRITE_ONCE + mutex_{un}lock here. Do you > see something obviously wrong with that ?
As said in IRC: nothing I can currently see.
> > I guess the LKMM tools should give me the yes/no answer I want, but if > that's a no, I'd also like to understand why ... :-)
That answer would be a little bit more involved ... ;-) The file Documentation/explanation.txt within the above mentioned directory could be a good starting point; skimming through litmus-tests/ and Documentation/recipes.txt could also provide some hints.
Andrea
> > Thanks, > Quentin
| |