On Fri, 2008-08-15 at 14:49 +0200, Ingo Molnar wrote: > * Huang Ying <ying.huang at intel.com> wrote: > > > +/* Ftrace disable/restore without lock. Some synchronization mechanism > > + * must be used to prevent ftrace_enabled to be changed between > > + * disable/restore. */ > > use the proper comment style please: > > /* > * > */ OK. I will change it. > > +static inline int __ftrace_enabled_save(void) > > +{ > > +#ifdef CONFIG_FTRACE > > + int saved_ftrace_enabled = ftrace_enabled; > > + ftrace_enabled = 0; > > + return saved_ftrace_enabled; > > +#else > > + return 0; > > +#endif > > +} > > + > > +static inline void __ftrace_enabled_restore(int enabled) > > +{ > > +#ifdef CONFIG_FTRACE > > + ftrace_enabled = enabled; > > +#endif > > +} > > hm, what is this used for? > > also, instead of such an ugly inline, why not create a proper > kernel/trace/* function for this. That would also give it access to all > the proper locking mechanisms - instead of relying on some extral > mechanism. This function is used for kexec jump in machine_kexec(). Where all non-boot CPUs and IRQ are disabled, system is going to kexec, and it is not allowed to schedule to other process in this circumstance, so a non-lock version is needed. A locked version has been implemented by Steven Rostedt, I think it can be used for other circumstance. Best Regards, Huang Ying