[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-devel] [PATCH] Low mem virq incremental adjustments



> From: Andres Lagar-Cavilla [mailto:andres@xxxxxxxxxxxxxxxx]
> Sent: Wednesday, March 07, 2012 11:12 AM
> To: Jan Beulich
> Cc: Dan Magenheimer; ian.campbell@xxxxxxxxxx; ian.jackson@xxxxxxxxxx; 
> adin@xxxxxxxxxxxxxx;
> andres@xxxxxxxxxxxxxx; xen-devel; tim@xxxxxxx
> Subject: Re: [PATCH] Low mem virq incremental adjustments
> 
> >  >>> On 07.03.12 at 17:15, Andres Lagar-Cavilla <andres@xxxxxxxxxxxxxxxx>
> > wrote:
> >> --- a/xen/common/page_alloc.c
> >> +++ b/xen/common/page_alloc.c
> >> @@ -377,7 +377,10 @@ static void __init setup_low_mem_virq(vo
> >>
> >>  static void check_low_mem_virq(void)
> >>  {
> >> -    if ( unlikely(total_avail_pages <= low_mem_virq_th) )
> >> +    unsigned long avail_pages = total_avail_pages +
> >> +                (opt_tmem) ? tmem_freeable_pages(): 0;
> >
> > Can tmem_freeable_pages() return anything other than zero when
> > opt_tmem is zero? (I.e. is the [improperly parenthesized!] conditional
> > expression necessary at all?)
> 
> I'm not sure. I'll let Dan take it from here, as he surely knows the right
> way. He acked it the way it is.
> Andres

Both would be correct (other than the parentheses).

I was also going to make the same comment about tmem_freeable_pages()
but decided the way Andres coded it is clearer because it doesn't
assume anything about tmem; if tmem is enabled, it uses an
abstract interface this code doesn't need to know anything about.

Anyway, either way is fine with me.

Dan

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel


 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.