Messages in this thread Patch in this message | | | Subject | Re: [patch 16/28] posix-timers: Convert timer_settime() to clockid_to_kclock() | From | John Stultz <> | Date | Tue, 01 Feb 2011 13:22:24 -0800 |
| |
On Tue, 2011-02-01 at 13:52 +0000, Thomas Gleixner wrote: > plain text document attachment (posix-timers-convert-timer-set.patch) > Set the common function for CLOCK_MONOTONIC and CLOCK_REALTIME kclocks > and use the new decoding function. No need to check for the return > value of it. If we have data corruption in the timer, we explode > somewhere else anyway. Also all kclocks which implement timer_create() > need to provide timer_settime() as well. > > Signed-off-by: Thomas Gleixner <tglx@linutronix.de> > Cc: John Stultz <john.stultz@linaro.org> > Cc: Richard Cochran <richard.cochran@omicron.at> [snip] > @@ -818,8 +821,8 @@ retry: > if (!timr) > return -EINVAL; > > - error = CLOCK_DISPATCH(timr->it_clock, timer_set, > - (timr, flags, &new_spec, rtn)); > + kc = clockid_to_kclock(timr->it_clock); > + error = kc->timer_set(timr, flags, &new_spec, rtn);
Again, me being paraniod, would probably want a null check on kc here. Also, as you suggested on irc, a WARN_ON_ONCE(), since it means kernel data has been munged.
thanks -john
| |