[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20230714222839.GD3273303@hirez.programming.kicks-ass.net>
Date: Sat, 15 Jul 2023 00:28:39 +0200
From: Peter Zijlstra <peterz@...radead.org>
To: Arjan van de Ven <arjan@...ux.intel.com>
Cc: "Wysocki, Rafael J" <rafael.j.wysocki@...el.com>,
Bruno Goncalves <bgoncalv@...hat.com>, x86@...nel.org,
LKML <linux-kernel@...r.kernel.org>
Subject: Re: [6.5.0-rc1] unchecked MSR access error: RDMSR from 0xe2 at rIP:
0xffffffff87090227 (native_read_msr+0x7/0x40) (intel_idle_init_cstates_icpu)
On Fri, Jul 14, 2023 at 02:45:21PM -0700, Arjan van de Ven wrote:
> On 7/14/2023 2:10 PM, Peter Zijlstra wrote:
> > On Tue, Jul 11, 2023 at 08:37:51PM +0200, Wysocki, Rafael J wrote:
> >
> > > > Rafael, can we please just pull these patches and try again later?
> > >
> > > I think you mean "revert"?
> >
> > Yes, revert. The whole premise with the unparsable changelog that
> > babbles about TLB invalidates without any performance data-so-ever
> > should've been a red-flag.
> >
> > That whole TLB angle is nonsense. We have paravirt TLB invalidation for
> > this.
>
> for kvm and hyperv. not for vmware etc.
And for everybody that does play ball, this will be a net negative by
causing extra TLB invalidations :/
Powered by blists - more mailing lists