From mboxrd@z Thu Jan  1 00:00:00 1970
From: Eduardo Habkost <ehabkost@redhat.com>
Subject: Re: [kvm-unit-tests PATCH] x86/memory: pass host clwb and clflushopt
 support information
Date: Mon, 18 Dec 2017 16:36:35 -0200
Message-ID: <20171218183635.GD5209@localhost.localdomain>
References: <20171212083524.3765-1-haozhong.zhang@intel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Cc: kvm@vger.kernel.org, Paolo Bonzini <pbonzini@redhat.com>
To: Haozhong Zhang <haozhong.zhang@intel.com>
Return-path: <kvm-owner@vger.kernel.org>
Received: from mx1.redhat.com ([209.132.183.28]:1786 "EHLO mx1.redhat.com"
        rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
        id S1758641AbdLRSgh (ORCPT <rfc822;kvm@vger.kernel.org>);
        Mon, 18 Dec 2017 13:36:37 -0500
Content-Disposition: inline
In-Reply-To: <20171212083524.3765-1-haozhong.zhang@intel.com>
Sender: kvm-owner@vger.kernel.org
List-ID: <kvm.vger.kernel.org>

On Tue, Dec 12, 2017 at 04:35:24PM +0800, Haozhong Zhang wrote:
> Intel VMX cannot intercept guest clwb and clflushopt. When clwb and
> clflushopt are not exposed in guest cpuid, clwb and clflushopt
> instructions in this test case can still succeed without #UD on the
> host CPU which has clwb and clflushopt support, though failures with
> UD are expected.
> 
> In order to avoid false alarms in such cases, introduce the following
> two arguments "has_clwb" and "has_clflushopt" to allow users to
> specify whether clwb and clflushopt are supported on the host CPU.
> 
> Signed-off-by: Haozhong Zhang <haozhong.zhang@intel.com>
> ---
>  x86/memory.c | 29 +++++++++++++++++++++++++++--
>  1 file changed, 27 insertions(+), 2 deletions(-)
> 
> diff --git a/x86/memory.c b/x86/memory.c
> index cd1eb46..03ff7d3 100644
> --- a/x86/memory.c
> +++ b/x86/memory.c
> @@ -23,10 +23,29 @@ static void handle_ud(struct ex_regs *regs)
>  	regs->rip += isize;
>  }
>  
> +/*
> + * Intel VMX cannot intercept guest clwb and clflushopt. When clwb and
> + * clflushopt are not exposed in guest cpuid, clwb and clflushopt
> + * instructions in this test case can still succeed without #UD on
> + * the host CPU which has clwb and clflushopt support. In order to avoid
> + * false alarms in such cases, introduce the following two arguments
> + * to allow users to specify whether clwb and clflushopt are supported on
> + * the host CPU:
> + * - has_clwb:       indicates clwb is supported on the host CPU
> + * - has_clflushopt: indicates clflushopt is supported on the host CPU
> + */

Why not simply use "-cpu host" to make sure the guest CPUID flags
match host CPUID?


>  int main(int ac, char **av)
>  {
>  	struct cpuid cpuid7, cpuid1;
>  	int xfail;
> +	int host_has_clwb = 0, host_has_clflushopt = 0; /* 0: unknown */
> +	int i;
> +
> +	for (i = 1; i < ac; i++)
> +		if (!strcmp(av[i], "has_clwb"))
> +			host_has_clwb = 1;
> +		else if (!strcmp(av[i], "has_clflushopt"))
> +			host_has_clflushopt = 1;
>  
>  	setup_idt();
>  	handle_exception(UD_VECTOR, handle_ud);
> @@ -63,13 +82,19 @@ int main(int ac, char **av)
>  	ud = 0;
>  	/* clflushopt (%rbx): */
>  	asm volatile(".byte 0x66, 0x0f, 0xae, 0x3b" : : "b" (&target));
> -	report_xfail("clflushopt", xfail, ud == 0);
> +	if (host_has_clflushopt)
> +		report("clflushopt", ud == 0);
> +	else
> +		report_xfail("clflushopt", xfail, ud == 0);
>  
>  	xfail = !(cpuid7.b & (1U << 24)); /* CLWB */
>  	ud = 0;
>  	/* clwb (%rbx): */
>  	asm volatile(".byte 0x66, 0x0f, 0xae, 0x33" : : "b" (&target));
> -	report_xfail("clwb", xfail, ud == 0);
> +	if (host_has_clwb)
> +		report("clwb", ud == 0);
> +	else
> +		report_xfail("clwb", xfail, ud == 0);
>  
>  	ud = 0;
>  	/* clwb requires a memory operand, the following is NOT a valid
> -- 
> 2.14.1
> 

-- 
Eduardo