Re: [Xen-devel] [PATCH v4 for-4.9] x86/vpmu: add cpu hot unplug notifier for vpmu

[Julien Grall]

On 31/05/17 10:46, Jan Beulich wrote:

On 31.05.17 at 11:23,  wrote:

On 05/31/2017 07:54 AM, Jan Beulich wrote:

On 26.05.17 at 07:20,  wrote:

Hi Julien,
Follow your advice, I change the tag to [PATCH v4 for-4.9] because of we
hope this patch can be merged in Xen 4.9.


Ping?


Sorry for the late reply. I was waiting for staging to re-open before
going through the potential list of patches to go in Xen 4.9.

Am I right that vPMU is disabled by default? I.e a user would have to
pass "vpmu" on the command line to use it? ([1] seems to confirm that)


Yes. And enabling it renders the entire configuration unsupported
(XSA-163).


On that basis:

Release-acked-by: Julien Grall 

Cheers,

--
Julien Grall

___
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

Re: [Xen-devel] [PATCH v4 for-4.9] x86/vpmu: add cpu hot unplug notifier for vpmu

[Jan Beulich]

>>> On 31.05.17 at 11:23,  wrote:
> On 05/31/2017 07:54 AM, Jan Beulich wrote:
> On 26.05.17 at 07:20,  wrote:
>>> Hi Julien,
>>> Follow your advice, I change the tag to [PATCH v4 for-4.9] because of we
>>> hope this patch can be merged in Xen 4.9.
>>
>> Ping?
> 
> Sorry for the late reply. I was waiting for staging to re-open before 
> going through the potential list of patches to go in Xen 4.9.
> 
> Am I right that vPMU is disabled by default? I.e a user would have to 
> pass "vpmu" on the command line to use it? ([1] seems to confirm that)

Yes. And enabling it renders the entire configuration unsupported
(XSA-163).

Jan


___
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

Re: [Xen-devel] [PATCH v4 for-4.9] x86/vpmu: add cpu hot unplug notifier for vpmu

[Julien Grall]

Hi,

On 05/31/2017 07:54 AM, Jan Beulich wrote:

On 26.05.17 at 07:20,  wrote:

Hi Julien,
Follow your advice, I change the tag to [PATCH v4 for-4.9] because of we
hope this patch can be merged in Xen 4.9.


Ping?


Sorry for the late reply. I was waiting for staging to re-open before 
going through the potential list of patches to go in Xen 4.9.


Am I right that vPMU is disabled by default? I.e a user would have to 
pass "vpmu" on the command line to use it? ([1] seems to confirm that)


Cheers,

[1] https://patchwork.kernel.org/patch/9254313/

--
Julien Grall

___
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

Re: [Xen-devel] [PATCH v4 for-4.9] x86/vpmu: add cpu hot unplug notifier for vpmu

[Jan Beulich]

>>> On 26.05.17 at 07:20,  wrote:
> Hi Julien,
> Follow your advice, I change the tag to [PATCH v4 for-4.9] because of we  
> hope this patch can be merged in Xen 4.9.

Ping?

Jan


___
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

Re: [Xen-devel] [PATCH v4 for-4.9] x86/vpmu: add cpu hot unplug notifier for vpmu

[Jan Beulich]

>>> On 26.05.17 at 07:20,  wrote:
> Hi Jan,
> If there have anything need to change in this patch?

I didn't get to look at v4 yet - too busy with other stuff.

Jan


___
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

Re: [Xen-devel] [PATCH v4 for-4.9] x86/vpmu: add cpu hot unplug notifier for vpmu

[Kang, Luwei]

Hi Julien,
Follow your advice, I change the tag to [PATCH v4 for-4.9] because of we  
hope this patch can be merged in Xen 4.9.

Hi Jan,
If there have anything need to change in this patch?

Thanks,
Luwei Kang

> -Original Message-
> From: Kang, Luwei
> Sent: Tuesday, May 23, 2017 4:46 AM
> To: xen-devel@lists.xen.org
> Cc: jbeul...@suse.com; andrew.coop...@citrix.com; boris.ostrov...@oracle.com; 
> Kang, Luwei 
> Subject: [PATCH v4] x86/vpmu: add cpu hot unplug notifier for vpmu
> 
> Currently, Hot unplug a physical CPU with vpmu enabled may cause system hang 
> due to send a remote call to an offlined pCPU. This
> patch add a cpu hot unplug notifer to save vpmu context before cpu offline.
> 
> Consider one scenario, hot unplug pCPU N with vpmu enabled.
> The vcpu which running on this pCPU will be switch to other online cpu. A 
> remote call will be send to pCPU N to save the vpmu
> context before loading the vpmu context on this pCPU.
> System will hang in function on_select_cpus() because of that pCPU is 
> offlined and can not do any respond.
> 
> The purpose of add a VPMU_CONTEXT_LOADED check in vpmu_arch_destroy() before 
> send a remote call to save vpmu contex is:
> a. when a vpmu context has been loaded in a remote pCPU, make a
>remote call to save the vpmu contex and stop counters is necessary.
> b. VPMU_CONTEXT_LOADED flag will be reset if a pCPU is offlined.
>this check will prevent send a remote call to an offlined pCPU.
> 
> Signed-off-by: Luwei Kang 
> ---
> v4:
>  1.remove cpu_online() check in vpm_load();  2.remove "vpmu_" prefix;  3.fix 
> a coding style;  4.add some commit message about
> VPMU_CONTEXT_LOADED in vpmu_arch_destroy();
> v3:
>  1.add cpu_online() check in vpm_load() and vpmu_arch_destroy();  2.add vpmu_ 
> prefix. rename cpu_callback() to
> vpmu_cpu_callback();
> v2:
>  1.fix some typo and coding style;
>  2.change "swith" to "if" in cpu_callback() because of there just have one 
> case;  3.add VPMU_CONTEX_LOADED check before send
> remote call in vpmu_arch_destroy();
> ---
>  xen/arch/x86/cpu/vpmu.c | 45 +
>  1 file changed, 41 insertions(+), 4 deletions(-)
> 
> diff --git a/xen/arch/x86/cpu/vpmu.c b/xen/arch/x86/cpu/vpmu.c index 
> 03401fd..1f7830b 100644
> --- a/xen/arch/x86/cpu/vpmu.c
> +++ b/xen/arch/x86/cpu/vpmu.c
> @@ -21,6 +21,7 @@
>  #include 
>  #include 
>  #include 
> +#include 
>  #include 
>  #include 
>  #include 
> @@ -575,15 +576,21 @@ static void vpmu_arch_destroy(struct vcpu *v)
>   * We will test it again in vpmu_clear_last() with interrupts
>   * disabled to make sure we don't clear someone else.
>   */
> -if ( per_cpu(last_vcpu, vpmu->last_pcpu) == v )
> +if ( cpu_online(vpmu->last_pcpu) &&
> + per_cpu(last_vcpu, vpmu->last_pcpu) == v )
>  on_selected_cpus(cpumask_of(vpmu->last_pcpu),
>   vpmu_clear_last, v, 1);
> 
>  if ( vpmu->arch_vpmu_ops && vpmu->arch_vpmu_ops->arch_vpmu_destroy )
>  {
> -/* Unload VPMU first. This will stop counters */
> -on_selected_cpus(cpumask_of(vcpu_vpmu(v)->last_pcpu),
> - vpmu_save_force, v, 1);
> +/*
> + * Unload VPMU first if VPMU_CONTEXT_LOADED being set.
> + * This will stop counters.
> + */
> +if ( vpmu_is_set(vpmu, VPMU_CONTEXT_LOADED) )
> +on_selected_cpus(cpumask_of(vcpu_vpmu(v)->last_pcpu),
> + vpmu_save_force, v, 1);
> +
>   vpmu->arch_vpmu_ops->arch_vpmu_destroy(v);
>  }
>  }
> @@ -835,6 +842,33 @@ long do_xenpmu_op(unsigned int op, 
> XEN_GUEST_HANDLE_PARAM(xen_pmu_params_t) arg)
>  return ret;
>  }
> 
> +static int cpu_callback(
> +struct notifier_block *nfb, unsigned long action, void *hcpu) {
> +unsigned int cpu = (unsigned long)hcpu;
> +struct vcpu *vcpu = per_cpu(last_vcpu, cpu);
> +struct vpmu_struct *vpmu;
> +
> +if ( !vcpu )
> +return NOTIFY_DONE;
> +
> +vpmu = vcpu_vpmu(vcpu);
> +if ( !vpmu_is_set(vpmu, VPMU_CONTEXT_ALLOCATED) )
> +return NOTIFY_DONE;
> +
> +if ( action == CPU_DYING )
> +{
> +vpmu_save_force(vcpu);
> +vpmu_reset(vpmu, VPMU_CONTEXT_LOADED);
> +}
> +
> +return NOTIFY_DONE;
> +}
> +
> +static struct notifier_block cpu_nfb = {
> +.notifier_call = cpu_callback
> +};
> +
>  static int __init vpmu_init(void)
>  {
>  int vendor = current_cpu_data.x86_vendor; @@ -872,8 +906,11 @@ static 
> int __init vpmu_init(void)
>  }
> 
>  if ( vpmu_mode != XENPMU_MODE_OFF )
> +{
> +register_cpu_notifier(_nfb);
>  printk(XENLOG_INFO "VPMU: version " __stringify(XENPMU_VER_MAJ) "."
> __stringify(XENPMU_VER_MIN) "\n");
> +}
>  else
>  opt_vpmu_enabled = 0;
> 
> --
> 1.8.3.1


___
Xen-devel mailing list