Re: [Xen-devel] [PATCH 2/4] xen/arm: Implement a dummy Performance Monitor for ARM32

[Ian Campbell <Ian.Campbell@xxxxxxxxxx>]

On Fri, 2014-05-02 at 13:41 +0100, Julien Grall wrote:
> On 05/02/2014 12:01 PM, Ian Campbell wrote:
> > On Thu, 2014-04-24 at 23:45 +0100, Julien Grall wrote:
> >> XSA-93 (commit 0b18220 "xen/arm: Don't let guess access to Debug and 
> >> Performance
> >> Monitor registers") disable Performance Monitor.
> >>
> >> When CONFIG_PERF_EVENTS is enabled in the Linux Kernel, regardless the
> >> ID_DFR0 (which tell if Perfomance Monitors Extension is implemented) the
> >> kernel will try to access to PMCR.
> >>
> >> Therefore we tell the guest we have 0 counters. Unfortunately we must 
> >> always
> >> support PMCCNTR (the cycle counter): we just RAZ/WI for all PM register,
> >> which doesn't crash the kernel at least.
> > 
> > How often does this trap occur in practice? Once at start of day? Only
> > if you run perf? Or on every guest context switch? (obviously the last
> > one would be bad...)
> 
> There is few calls to the perf registers during the boot (when Linux is
> compiled with CONFIG_PERF_EVENTS=y).
> 
> I didn't see any usage during guest context switch. I haven't try perf.

Great. Acked-by: Ian Campbell <ian.campbell@xxxxxxxxxx>




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