Re: [xen-unstable test] 164996: regressions - FAIL

[Ian Jackson <iwj@xxxxxxxxxxxxxx>]

Jan Beulich writes ("Re: [xen-unstable test] 164996: regressions - FAIL"):
> Is the Dom0 kernel used here a distro one or our own build of one of
> the upstream trees? In the latter case I'd expect propagation to be
> quite a bit faster than in the former case.

It's our own build.

> >  But I'm loathe to make this change
> > because it seems to me that it would be simply masking the bug.
> > 
> > Notably, when this goes wrong, it seems to happen after the guest has
> > been started once successfully already.  So there *is* enough
> > memory...
> 
> Well, there is enough memory, sure, but (transiently as it seems) not
> enough contiguous chunks. The likelihood of higher order allocations
> failing increases with smaller overall memory amounts (in Dom0 in this
> case), afaict, unless there's (aggressive) de-fragmentation.

Indeed.

I'm not sure, though, that I fully understand the design principles
behind non-order-0 allocations, and memory sizing, and so on.  Your
earlier mail suggeted there may not be a design principle, and that
anything relying on non-order-0 atomic allocations is only working by
luck (or an embarassing excess of ram).

Ian.