From: Keir Fraser Subject: Re: [PATCH] MSR related clean up Date: Wed, 24 Jun 2009 10:03:56 +0100 Message-ID: References: <200906241650.22488.sheng@linux.intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset="US-ASCII" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <200906241650.22488.sheng@linux.intel.com> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: Sheng Yang , "xen-devel@lists.xensource.com" Cc: Eddie Dong List-Id: xen-devel@lists.xenproject.org On 24/06/2009 09:50, "Sheng Yang" wrote: > On Thursday 18 June 2009 17:57:06 Sheng Yang wrote: >> rdmsr_safe() is used to access MSR unknown to Xen, and is not safe... I >> think it is legacy. >> >> Also let msr_write be parity with msr_read on MSR_IA32_MISC_ENABLE. >> >> CC: Eddie Dong >> Signed-off-by: Sheng Yang > > Keir? Looks pretty dangerous to me. So I'm not sure. There are various MSRs that are detected via CPU family/model (which we pass through) which would then #GP on access. Also this doesn't change the AMD default. Overall, what we have now does seem to work so I'm reluctant to mess with it. -- Keir