[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]
Re: mutexes, locks and so on...
Similar questions were raised a few times.. let's go through this again.
Johnny Billquist <bqt%softjar.se@localhost> wrote:
> With mutex_spin, you instead store the original spl at the first
> mutex_spin_enter, and later calls to mutex_spin_enter can only possibly
> raise the ipl further. At mutex_spin_exit, we do not lower the spl
> again, until the final mutex_spin_exit, which resets the spl to the
> value as before any mutex was held.
This is correct.
> The cause a slightly different behaviour, as the spl will continue to
> possibly be very high, even though you are only holding a low ipl mutex.
> While it obviously don't cause a system to fail, it can introduce delays
> which might not be neccesary, and could in theory cause interrupts to be
> dropped that needn't be.
In theory, it might. But see below.
> Is this a conscious design? Do we not expect/enforce mutexes to be
> released in the reverse order they were acquired?
Very conscious, made about 4 years ago. There are very few (would say a
tiny fraction, in all tree) where we have the following locking cases:
mutex_spin_exit(a); <- e.g. where a has IPL_HIGH
<some code> <- some code still runs at IPL_HIGH
mutex_spin_exit(b); <- e.g. where b has IPL_VM
One such case is in tty subsystem, which needs proper locking (and
perhaps generally revamped.. anyone?). Generally, such cases should
Also, AFAIK effect was was actually measured during newlock2 branch
development. Supporting such micro-optimisation would mean quite a bit
of complexity in critical mutex(9) interface with practically no gain.
> Also, there are a few places in the code in the kernel, where atomic_cas
> is called explicitly. Wouldn't it be better if such code were abstracted
> out, so we didn't depend on specific cpu instructions in the MI code?
atomic_cas_*() routines are mandatory part of atomic_ops(3/9) interface.
I believe almost any modern operating system should have it these days.
We model algorithms using them. I believe MI code, design-wise, should
look to the future, rather than the past. :)
Main Index |
Thread Index |