Hi Henrik,
On Tue, 2010-11-09 at 14:32 -0500, Henrik Rydberg wrote:
> Hi Guenter,
>
> >> +/*
>
> >> + * applesmc_init_smcreg_try - Try to initialize register cache. Idempotent.
> >> + */
> >> +static int applesmc_init_smcreg_try(void)
> >> +{
> >> + struct applesmc_registers *s = &smcreg;
> >> + int ret;
> >> +
> >> + if (s->init_complete)
> >> + return 0;
> >> +
> >> + mutex_init(&s->mutex);
> >> +
> > I am a bit concerned that mutex_init() can be called multiple times. Are
> > you sure this is safe ?
>
>
> mutex_destroy() is defined as a nop, so I guess the question is whether anything
> could be holding the lock when entering a second init. There are no sysfs files
> created at that point, so I would say no. The mutex could be put back with a
> static initializer, if this is not satisfactory. The real reason to move it to
> the smcreg struct was to force a rename of the mutex itself.
>
Alternatively, you could move the mutex initialization to the beginning
of applesmc_init_smcreg() and make it
mutex_init(&smcreg.mutex);
> >
> >> + ret = read_register_count(&s->key_count);
> >> + if (ret)
> >> + return ret;
> >> +
> >> + if (!s->cache)
> >> + s->cache = kcalloc(s->key_count, sizeof(*s->cache), GFP_KERNEL);
> >> + if (!s->cache)
> >> + return -ENOMEM;
> >> +
> >> + s->init_complete = true;
> >> +
> >> + pr_info("key=%d\n", s->key_count);
> >> +
> > Hope that means more to macbook users than it does to me ;).
>
>
> It means a lot from a diagnostic point of view - a normal user does not really
> care about the dmesg output anyways. :-)
>
Ok, guess I am not a normal user ;).
> >> +static int applesmc_init_smcreg(void)
> >> +{
> >> + int ms, ret;
> >> +
> >> + for (ms = 0; ms < INIT_TIMEOUT_MSECS; ms += INIT_WAIT_MSECS) {
> >> + ret = applesmc_init_smcreg_try();
> >> + if (!ret)
> >> + return 0;
> >> + pr_warn("slow init, retrying\n");
> >
> > INIT_WAIT_MS is 50ms, so you issue this warning every 50ms for up to
> > five seconds. Pretty noisy... sure that is what you want ? Also, does it
> > really make sense to retry if the error is ENOMEM ?
>
>
> With the empirical failure rate, it is extremely unlikely to get more than a
> couple of failures in a row - information which in itself could be very useful.
You would have alternative options, though, with less noise. For
example, something along the line of
for (...) {
...
if (!ret) {
if (ms)
pr_info("smcreg initialization took %d ms\n", ms);
return 0;
}
...
}
pr_err("smcreg initialization failed\n");
> A direct escape on ENOMEM makes sense, though.
>
> Changing the place of the mutex will ripple through all patches, so I will
> resend from this one onwards. I suppose you have more comments on the following
> patches?
Maybe it won't be that bad if you initialize it as I suggested above.
Regarding additional comments - I don't know yet. I didn't have time to
look into the other patches yet. I'll try to do that by tonight.
Guenter
_______________________________________________
lm-sensors mailing list
lm-sensors@xxxxxxxxxxxxxx
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
