Re: [RFC PATCH v13] sys_membarrier(): system/process-wide memory barrier (x86)

[Mathieu Desnoyers <mathieu.desnoyers@efficios.com>]

----- Original Message -----
> On Tue, Mar 17, 2015 at 01:22:02PM -0400, Mathieu Desnoyers wrote:
> > --- a/init/Kconfig
> > +++ b/init/Kconfig
> > @@ -1568,6 +1568,15 @@ config PCI_QUIRKS
> >  	  bugs/quirks. Disable this only if your target machine is
> >  	  unaffected by PCI quirks.
> >  
> > +config MEMBARRIER
> > +	bool "Enable membarrier() system call" if EXPERT
> > +	default y
> > +	help
> > +	  Enable the membarrier() system call that allows issuing
> > +	  memory barriers across cores.
> 
> Nit: CPUs, not cores.  Nothing here is specific to
> packages/cores/threads.

Fixed.

> 
> > --- a/kernel/rcu/update.c
> > +++ b/kernel/rcu/update.c
> > @@ -49,6 +49,8 @@
> >  #include <linux/module.h>
> >  #include <linux/kthread.h>
> >  #include <linux/tick.h>
> > +#include <linux/syscalls.h>
> > +#include <linux/membarrier.h>
> >  
> >  #define CREATE_TRACE_POINTS
> >  
> > @@ -775,3 +777,73 @@ late_initcall(rcu_verify_early_boot_tests);
> >  #else
> >  void rcu_early_boot_tests(void) {}
> >  #endif /* CONFIG_PROVE_RCU */
> > +
> > +#ifdef CONFIG_MEMBARRIER
> 
> As mentioned in my previous mail, please consider moving this to its own
> file, if it doesn't use any of the internals of kernel/rcu/update.c
> (which I don't think it does).  That then means you can enable/disable
> the entire file in the Makefile rather than using an #ifdef in the .c
> file.

Will do, thanks!

Mathieu

> 
> - Josh Triplett
> 

-- 
Mathieu Desnoyers
EfficiOS Inc.
http://www.efficios.com