> -----Original Message----- > From: David Vrabel [mailto:david.vra...@citrix.com] > Sent: Thursday, March 10, 2016 2:02 AM > To: George Dunlap <george.dun...@citrix.com>; Jan Beulich > <jbeul...@suse.com> > Cc: Andrew Cooper <andrew.coop...@citrix.com>; Dario Faggioli > <dario.faggi...@citrix.com>; George Dunlap <george.dun...@eu.citrix.com>; > Wu, Feng <feng...@intel.com>; Tian, Kevin <kevin.t...@intel.com>; xen- > de...@lists.xen.org; Konrad Rzeszutek Wilk <konrad.w...@oracle.com>; Ian > Jackson <ian.jack...@eu.citrix.com>; Lars Kurth <lars.ku...@citrix.com> > Subject: Re: On setting clear criteria for declaring a feature acceptable (was > "vmx: VT-d posted-interrupt core logic handling") > > On 09/03/16 16:23, George Dunlap wrote: > > > > I don't know why this is controversial -- this seems obvious to me. > > What do other committers / maintainers think? > > I started on a reply to this but then I went back and read the original > thread... > > + /* > + * XXX: The length of the list depends on how many vCPU is current > + * blocked on this specific pCPU. This may hurt the interrupt > + * latency if the list grows to too many entries. > + */ > > Even the original author knows that there's a problem here, so in this > case George, I think you are unfairly criticizing Jan.
This is the potential issue Jan pointed out, and adding the comments is according Jan's comments then. But as George pointed out, it is not very clear to how to reproduce this scenario in real world and what is the criteria of "the list is too long", so here we are discussing whether it is reasonable to make this feature default off just because of this theoretically existing issue, and hence the " criteria for declaring a feature acceptable ". Thanks, Feng > > David _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel
