From mboxrd@z Thu Jan  1 00:00:00 1970
From: Suraj Jitindar Singh <sjitindarsingh@gmail.com>
Subject: Re: [kvm-unit-tests PATCH v3] powerpc: Check whether TM is
 available before running other tests
Date: Wed, 05 Oct 2016 18:25:08 +1100
Message-ID: <1475652308.2070.2.camel@gmail.com>
References: <1475572050-13164-1-git-send-email-thuth@redhat.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: 8bit
Cc: Radim =?UTF-8?Q?Kr=C4=8Dm=C3=A1=C5=99?= <rkrcmar@redhat.com>,
        kvm-ppc@vger.kernel.org, Laurent Vivier <lvivier@redhat.com>,
        Drew Jones <drjones@redhat.com>
To: Thomas Huth <thuth@redhat.com>, kvm@vger.kernel.org
Return-path: <kvm-owner@vger.kernel.org>
Received: from mail-pa0-f68.google.com ([209.85.220.68]:33059 "EHLO
        mail-pa0-f68.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
        with ESMTP id S1752372AbcJEHZP (ORCPT <rfc822;kvm@vger.kernel.org>);
        Wed, 5 Oct 2016 03:25:15 -0400
In-Reply-To: <1475572050-13164-1-git-send-email-thuth@redhat.com>
Sender: kvm-owner@vger.kernel.org
List-ID: <kvm.vger.kernel.org>

On Tue, 2016-10-04 at 11:07 +0200, Thomas Huth wrote:
> Transactional memory is currently only supported on KVM-HV, and
> not yet on KVM-PR. So it's better to check the device tree first
> and fail gracefully if it is not available.
> 
> Reviewed-by: Laurent Vivier <lvivier@redhat.com>
> Reviewed-by: Andrew Jones <drjones@redhat.com> 
> Signed-off-by: Thomas Huth <thuth@redhat.com>
Thanks for sending a v3.

Reviewed-by: Suraj Jitindar Singh <sjitindarsingh@gmail.com>
> ---
>  v3:
>  - Removed unnecessary check for "plen >= 26" in cpu_has_tm()
> 
>  v2:
>  - Reworked the check for the "ibm,pa-features" and added a comment
>  - Use a dedicated variable "has_tm" instead of "i" in main()
> 
>  powerpc/tm.c | 43 ++++++++++++++++++++++++++++++++++++++++++-
>  1 file changed, 42 insertions(+), 1 deletion(-)
> 
> diff --git a/powerpc/tm.c b/powerpc/tm.c
> index 6ce750a..b611061 100644
> --- a/powerpc/tm.c
> +++ b/powerpc/tm.c
> @@ -10,6 +10,41 @@
>  #include <asm/processor.h>
>  #include <asm/handlers.h>
>  #include <asm/smp.h>
> +#include <asm/setup.h>
> +#include <devicetree.h>
> +
> +/* Check "ibm,pa-features" property of a CPU node for the TM flag */
> +static void cpu_has_tm(int fdtnode, u32 regval __unused, void *ptr)
> +{
> +	const struct fdt_property *prop;
> +	int plen;
> +
> +	prop = fdt_get_property(dt_fdt(), fdtnode, "ibm,pa-
> features", &plen);
> +	if (!prop)	/* No features means TM is also not
> available */
> +		return;
> +	/* Sanity check for the property layout (first two bytes are
> header) */
> +	assert(plen >= 8 && prop->data[1] == 0 && prop->data[0] <=
> plen - 2);
> +
> +	/*
> +	 * The "Transactional Memory Category Support" flags are at
> byte
> +	 * offset 22 and 23 of the attribute type 0, so when adding
> the
> +	 * two bytes for the header, we've got to look at offset 24
> for
> +	 * the TM support bit.
> +	 */
> +	if (prop->data[0] >= 24 && (prop->data[24] & 0x80) != 0)
> +		*(int *)ptr += 1;
> +}
> +
> +/* Check whether all CPU nodes have the TM flag */
> +static bool all_cpus_have_tm(void)
> +{
> +	int ret;
> +	int available = 0;
> +
> +	ret = dt_for_each_cpu_node(cpu_has_tm, &available);
> +
> +	return ret == 0 && available == nr_cpus;
> +}
>  
>  static int h_cede(void)
>  {
> @@ -101,11 +136,17 @@ struct {
>  
>  int main(int argc, char **argv)
>  {
> -	bool all;
> +	bool all, has_tm;
>  	int i;
>  
>  	report_prefix_push("tm");
>  
> +	has_tm = all_cpus_have_tm();
> +	report_xfail("TM available in 'ibm,pa-features' property",
> +		     !has_tm, has_tm);
> +	if (!has_tm)
> +		return report_summary();
> +
>  	all = argc == 1 || !strcmp(argv[1], "all");
>  
>  	for (i = 0; hctests[i].name != NULL; i++) {