Re: [Xen-devel] [PATCH 2/4] x86/hypervisor: pass flags to hypervisor_flush_tlb

[Wei Liu <wl@xxxxxxx>]

On Thu, Feb 13, 2020 at 08:48:39AM +0000, Durrant, Paul wrote:
> > -----Original Message-----
> > From: Roger Pau Monné <roger.pau@xxxxxxxxxx>
> > Sent: 12 February 2020 18:01
> > To: Wei Liu <wl@xxxxxxx>
> > Cc: Xen Development List <xen-devel@xxxxxxxxxxxxxxxxxxxx>; Durrant, Paul
> > <pdurrant@xxxxxxxxxxxx>; Michael Kelley <mikelley@xxxxxxxxxxxxx>; Wei Liu
> > <liuwe@xxxxxxxxxxxxx>; Jan Beulich <jbeulich@xxxxxxxx>; Andrew Cooper
> > <andrew.cooper3@xxxxxxxxxx>
> > Subject: Re: [PATCH 2/4] x86/hypervisor: pass flags to
> > hypervisor_flush_tlb
> > 
> > On Wed, Feb 12, 2020 at 04:09:16PM +0000, Wei Liu wrote:
> > > Hyper-V's L0 assisted flush has fine-grained control over what gets
> > > flushed. We need all the flags available to make the best decisions
> > > possible.
> > >
> > > No functional change because Xen's implementation doesn't care about
> > > what is passed to it.
> > 
> > While it's certainly fine to pass a flags field with more information,
> > the flush flags for Xen can also contain FLUSH_CACHE, FLUSH_VCPU_STATE
> > or FLUSH_ROOT_PGTBL, can you add an assert that those never get passed
> > to the flush hook?
> > 
> > IMO we should define a mask with FLUSH_TLB, FLUSH_TLB_GLOBAL,
> > FLUSH_VA_VALID and FLUSH_ORDER_MASK and assert that those are the only
> > valid flags to be used for the hypervisor assisted flush hook.
> >
> 
> Agreed that this should be abstracted; we certainly don't want to have
> various bits of Xen needing to know what hypervisor it is running on
> top of.

OK. I can introduce a FLUSH_TLB_FLAGS for all things pertaining to TLB
-- the four things mentioned in Roger's reply.

Wei.

> 
>   Paul
> 
>  
> > Thanks, Roger.

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxxx
https://lists.xenproject.org/mailman/listinfo/xen-devel