From: Jeremy Fitzhardinge <jeremy@goop.org>
To: Sheng Yang <sheng@linux.intel.com>
Cc: xen-devel <xen-devel@lists.xensource.com>,
Ian Campbell <Ian.Campbell@citrix.com>,
linux-kernel@vger.kernel.org, Ian Pratt <Ian.Pratt@eu.citrix.com>,
Keir Fraser <keir.fraser@eu.citrix.com>
Subject: Re: [Xen-devel] [PATCH][v9 6/6] xen: Enable PV clocksource for HVM
Date: Fri, 12 Mar 2010 12:37:55 -0800 [thread overview]
Message-ID: <4B9AA623.6050601@goop.org> (raw)
In-Reply-To: <1268362647-5317-7-git-send-email-sheng@linux.intel.com>
On 03/11/2010 06:57 PM, Sheng Yang wrote:
> PV clocksource can provide a reliable clocksource for HVM running on Xen.
>
> To enable it, put following line to the HVM configure file:
>
> cpuid = [ '0x40000002:edx=0x3' ]
>
> It would set bit 0 and bit 1 in 0x40000002:edx, which enable PV extension
> framework and PV clocksource.
>
I guess my previous comment would be better made here. I don't see any
reason to make the Xen pv clocksource conditional at runtime. If the
user wants to disable/override it, then they can either boot with a
clocksource= parameter on the kernel command line, or change it on the
fly via /sys.
J
> Signed-off-by: Sheng Yang<sheng@linux.intel.com>
> ---
> arch/x86/xen/hvmpv.c | 24 ++++++++++++++++++++++++
> arch/x86/xen/time.c | 12 +++++++++++-
> arch/x86/xen/xen-ops.h | 1 +
> 3 files changed, 36 insertions(+), 1 deletions(-)
>
> diff --git a/arch/x86/xen/hvmpv.c b/arch/x86/xen/hvmpv.c
> index a24a1e9..1961cf0 100644
> --- a/arch/x86/xen/hvmpv.c
> +++ b/arch/x86/xen/hvmpv.c
> @@ -42,6 +42,8 @@ static void __init xen_hvm_pv_banner(void)
> pv_info.name);
> printk(KERN_INFO "Xen version: %d.%d%s\n",
> version>> 16, version& 0xffff, extra.extraversion);
> + if (xen_hvm_pv_clock_enabled())
> + printk(KERN_INFO "PV feature: PV clocksource enabled\n");
> }
>
> static int __init xen_para_available(void)
> @@ -80,6 +82,9 @@ static int __init init_hvm_pv_info(void)
> if (!(edx& XEN_CPUID_FEAT2_HVM_PV))
> return -ENODEV;
>
> + if (edx& XEN_CPUID_FEAT2_HVM_PV_CLOCK)
> + xen_hvm_pv_features |= XEN_HVM_PV_CLOCK_ENABLED;
> +
> if (pages< 1)
> return -ENODEV;
>
> @@ -110,6 +115,23 @@ static int __init init_shared_info(void)
> return 0;
> }
>
> +static void __init init_pv_clocksource(void)
> +{
> + if (!xen_hvm_pv_clock_enabled())
> + return;
> +
> + if (enable_hvm_pv(HVM_PV_CLOCK))
> + BUG();
> +
> + pv_time_ops.sched_clock = xen_sched_clock;
> +
> + x86_platform.calibrate_tsc = xen_tsc_khz;
> + x86_platform.get_wallclock = xen_get_wallclock;
> + x86_platform.set_wallclock = xen_set_wallclock;
> +
> + xen_register_clocksource();
> +}
> +
> void __init xen_guest_init(void)
> {
> int r;
> @@ -132,4 +154,6 @@ void __init xen_guest_init(void)
> pv_info = xen_hvm_pv_info;
>
> xen_domain_type = XEN_HVM_DOMAIN;
> +
> + init_pv_clocksource();
> }
> diff --git a/arch/x86/xen/time.c b/arch/x86/xen/time.c
> index 0a5aa44..d1c1c50 100644
> --- a/arch/x86/xen/time.c
> +++ b/arch/x86/xen/time.c
> @@ -474,11 +474,21 @@ void xen_timer_resume(void)
> }
> }
>
> +static bool xen_clocksource_enabled;
> +
> +void xen_register_clocksource(void)
> +{
> + if (!xen_clocksource_enabled) {
> + clocksource_register(&xen_clocksource);
> + xen_clocksource_enabled = 1;
> + }
> +}
> +
> __init void xen_time_init(void)
> {
> int cpu = smp_processor_id();
>
> - clocksource_register(&xen_clocksource);
> + xen_register_clocksource();
>
> if (HYPERVISOR_vcpu_op(VCPUOP_stop_periodic_timer, cpu, NULL) == 0) {
> /* Successfully turned off 100Hz tick, so we have the
> diff --git a/arch/x86/xen/xen-ops.h b/arch/x86/xen/xen-ops.h
> index 355fa6b..2aeaf51 100644
> --- a/arch/x86/xen/xen-ops.h
> +++ b/arch/x86/xen/xen-ops.h
> @@ -46,6 +46,7 @@ cycle_t xen_clocksource_read(void);
> void xen_setup_cpu_clockevents(void);
> unsigned long xen_tsc_khz(void);
> void __init xen_time_init(void);
> +void xen_register_clocksource(void);
> unsigned long xen_get_wallclock(void);
> int xen_set_wallclock(unsigned long time);
> unsigned long long xen_sched_clock(void);
>
prev parent reply other threads:[~2010-03-12 20:38 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-03-12 2:57 [PATCH][v9 0/6] PV extension of HVM (Hybrid) for Xen Sheng Yang
2010-03-12 2:57 ` [PATCH][v9 1/6] xen: add support for hvm_op Sheng Yang
2010-03-12 2:57 ` [PATCH][v9 2/6] xen: Import cpuid.h from Xen Sheng Yang
2010-03-12 2:57 ` [PATCH][v9 3/6] xen: Make pv drivers only work with xen_pv_domain() Sheng Yang
2010-03-12 2:57 ` [PATCH][v9 4/6] xen/hvm: Xen PV extension of HVM initialization Sheng Yang
2010-03-12 20:35 ` [Xen-devel] " Jeremy Fitzhardinge
2010-03-15 1:45 ` Sheng Yang
2010-03-15 12:04 ` Stefano Stabellini
2010-03-15 22:59 ` Jeremy Fitzhardinge
2010-03-16 1:51 ` Sheng Yang
2010-03-16 17:12 ` Jeremy Fitzhardinge
2010-03-16 17:20 ` Ian Campbell
2010-03-16 17:29 ` Jeremy Fitzhardinge
2010-03-12 2:57 ` [PATCH][v9 5/6] x86/xen: The entrance for PV extension of HVM Sheng Yang
2010-03-12 2:57 ` [PATCH][v9 6/6] xen: Enable PV clocksource for HVM Sheng Yang
2010-03-12 20:37 ` Jeremy Fitzhardinge [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=4B9AA623.6050601@goop.org \
--to=jeremy@goop.org \
--cc=Ian.Campbell@citrix.com \
--cc=Ian.Pratt@eu.citrix.com \
--cc=keir.fraser@eu.citrix.com \
--cc=linux-kernel@vger.kernel.org \
--cc=sheng@linux.intel.com \
--cc=xen-devel@lists.xensource.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox