On Tue, Sep 12, 2023 at 03:52:13PM +0200, Peter Zijlstra wrote:
> On Tue, Sep 12, 2023 at 01:28:13PM +0100, Matthew Wilcox wrote:
> > On Tue, Sep 12, 2023 at 11:03:42AM +0200, Peter Zijlstra wrote:
> > > If not, then sure we can do this; it's not like I managed to get rid of
> > > muteX_is_locked() -- and I actually tried at some point :/
> > >
> > > And just now I grepped for it, and look what I find:
> > >
> > > drivers/hid/hid-nintendo.c: if (unlikely(mutex_is_locked(&ctlr->output_mutex))) {
> > > drivers/nvdimm/btt.c: if (mutex_is_locked(&arena->err_lock)
> > >
> > > And there's more :-(
> >
> > Are these actually abuse? I looked at these two, and they both seem to
> > be asking "Does somebody else currently have this mutex?" rather than
> > "Do I have this mutex?".
>
> It's effectively a random number generator in that capacity. Someone
> might have it or might have had it when you looked and no longer have
> it, or might have it now but not when you asked.
Also, there's more fun; the 'is_locked' store from spin_lock() (or
mutex, or whatever) is not ordered vs any other write inside the
critical section.
So something like:
bar = 0;
CPU0 CPU1
spin_lock(&foo)
bar = 1; x = READ_ONCE(bar)
y = spin_is_locked(&foo);
spin_unlock(&foo);
can have x==1 && y==0, even though CPU0 is currently inside the critical
section.
Normally that doesn't matter, and for the program-order case where you
ask 'am I holding the lock' this obviously cannot go wrong. But the
moment you ask: 'is someone else holding the lock' it all goes sideways
real fast.
We've been there, done that, got a t-shirt etc..
