[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20140618185939.GF1695@ERROL.INI.CMU.EDU>
Date: Wed, 18 Jun 2014 14:59:39 -0400
From: "Gabriel L. Somlo" <gsomlo@...il.com>
To: Eric Northup <digitaleric@...gle.com>
Cc: Nadav Amit <nadav.amit@...il.com>,
Nadav Amit <namit@...technion.ac.il>,
Paolo Bonzini <pbonzini@...hat.com>, gleb@...nel.org,
Thomas Gleixner <tglx@...utronix.de>,
Ingo Molnar <mingo@...hat.com>,
"H. Peter Anvin" <hpa@...or.com>,
the arch/x86 maintainers <x86@...nel.org>,
Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
KVM <kvm@...r.kernel.org>, joro@...tes.org, mst@...hat.com,
agraf@...e.de
Subject: Re: [PATCH 3/3] KVM: x86: correct mwait and monitor emulation
On Wed, Jun 18, 2014 at 11:30:07AM -0700, Eric Northup wrote:
> Quoting Gabriel's post http://www.spinics.net/lists/kvm/msg103792.html :
>
> [...]
>
> > E.g., OS X 10.5 *does* check CPUID, and panics if it doesn't find it.
> > It needs the MONITOR cpuid flag to be on, *and* the actual
> > instructions to work.
That was an argument in favor of finding a mechanism to allow (qemu)
users to enable an otherwise default-off monitor cpuid flag.
We definitely don't want to advertise monitor/mwait availability to
guests which would otherwise sanely fail back to a hlt-based idle loop
when cpuid tells them monitor/mwait are not available :)
However, check my earlier proposal of backing out of monitor/mwait
entirely (as it turns out, there's a kernel command line to tell OS X
not to use monitor/mwait, which is IMHO vastly preferable to creating
"undocumented" KVM hacks :)
Thanks much,
--Gabriel
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists