From: Srivatsa S. Bhat (VMware) <[email protected]>
VMware ESXi allows enabling a passthru mwait CPU-idle state in the
guest using the following VMX option:
monitor_control.mwait_in_guest = "TRUE"
This lets a vCPU in mwait to remain in guest context (instead of
yielding to the hypervisor via a VMEXIT), which helps speed up
wakeups from idle.
However, this runs into problems with CPU hotplug, because the Linux
CPU offline path prefers to put the vCPU-to-be-offlined in mwait
state, whenever mwait is available. As a result, since a vCPU in mwait
remains in guest context and does not yield to the hypervisor, an
offline vCPU *appears* to be 100% busy as viewed from ESXi, which
prevents the hypervisor from running other vCPUs or workloads on the
corresponding pCPU (particularly when vCPU - pCPU mappings are
statically defined by the user). [ Note that such a vCPU is not
actually busy spinning though; it remains in mwait idle state in the
guest ].
Fix this by overriding the CPU offline play_dead() callback for VMware
hypervisor, by putting the CPU in halt state (which actually yields to
the hypervisor), even if mwait support is available.
Signed-off-by: Srivatsa S. Bhat (VMware) <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Borislav Petkov <[email protected]>
Cc: Dave Hansen <[email protected]>
Cc: "H. Peter Anvin" <[email protected]>
Cc: Alexey Makhalov <[email protected]>
Cc: Juergen Gross <[email protected]>
Cc: [email protected]
Cc: VMware PV-Drivers Reviewers <[email protected]>
---
arch/x86/kernel/cpu/vmware.c | 17 +++++++++++++++++
1 file changed, 17 insertions(+)
diff --git a/arch/x86/kernel/cpu/vmware.c b/arch/x86/kernel/cpu/vmware.c
index c04b933f48d3..420e359ed9bb 100644
--- a/arch/x86/kernel/cpu/vmware.c
+++ b/arch/x86/kernel/cpu/vmware.c
@@ -27,6 +27,7 @@
#include <linux/clocksource.h>
#include <linux/cpu.h>
#include <linux/reboot.h>
+#include <linux/tboot.h>
#include <linux/static_call.h>
#include <asm/div64.h>
#include <asm/x86_init.h>
@@ -312,6 +313,21 @@ static int vmware_cpu_down_prepare(unsigned int cpu)
local_irq_enable();
return 0;
}
+
+static void vmware_play_dead(void)
+{
+ play_dead_common();
+ tboot_shutdown(TB_SHUTDOWN_WFS);
+
+ /*
+ * Put the vCPU going offline in halt instead of mwait (even
+ * if mwait support is available), to make sure that the
+ * offline vCPU yields to the hypervisor (which may not happen
+ * with mwait, for example, if the guest's VMX is configured
+ * to retain the vCPU in guest context upon mwait).
+ */
+ hlt_play_dead();
+}
#endif
static __init int activate_jump_labels(void)
@@ -349,6 +365,7 @@ static void __init vmware_paravirt_ops_setup(void)
#ifdef CONFIG_SMP
smp_ops.smp_prepare_boot_cpu =
vmware_smp_prepare_boot_cpu;
+ smp_ops.play_dead = vmware_play_dead;
if (cpuhp_setup_state_nocalls(CPUHP_AP_ONLINE_DYN,
"x86/vmware:online",
vmware_cpu_online,
Hi Boris, Thomas, Peter,
On 7/21/22 1:44 PM, Srivatsa S. Bhat wrote:
> From: Srivatsa S. Bhat (VMware) <[email protected]>
>
> VMware ESXi allows enabling a passthru mwait CPU-idle state in the
> guest using the following VMX option:
>
> monitor_control.mwait_in_guest = "TRUE"
>
> This lets a vCPU in mwait to remain in guest context (instead of
> yielding to the hypervisor via a VMEXIT), which helps speed up
> wakeups from idle.
>
> However, this runs into problems with CPU hotplug, because the Linux
> CPU offline path prefers to put the vCPU-to-be-offlined in mwait
> state, whenever mwait is available. As a result, since a vCPU in mwait
> remains in guest context and does not yield to the hypervisor, an
> offline vCPU *appears* to be 100% busy as viewed from ESXi, which
> prevents the hypervisor from running other vCPUs or workloads on the
> corresponding pCPU (particularly when vCPU - pCPU mappings are
> statically defined by the user). [ Note that such a vCPU is not
> actually busy spinning though; it remains in mwait idle state in the
> guest ].
>
> Fix this by overriding the CPU offline play_dead() callback for VMware
> hypervisor, by putting the CPU in halt state (which actually yields to
> the hypervisor), even if mwait support is available.
>
> Signed-off-by: Srivatsa S. Bhat (VMware) <[email protected]>
> Cc: Thomas Gleixner <[email protected]>
> Cc: Peter Zijlstra <[email protected]>
> Cc: Ingo Molnar <[email protected]>
> Cc: Borislav Petkov <[email protected]>
> Cc: Dave Hansen <[email protected]>
> Cc: "H. Peter Anvin" <[email protected]>
> Cc: Alexey Makhalov <[email protected]>
> Cc: Juergen Gross <[email protected]>
> Cc: [email protected]
> Cc: VMware PV-Drivers Reviewers <[email protected]>
> ---
Could you share your thoughts on this patch when you get a chance,
please? I verified that this patch still applies cleanly on current
mainline (6.0-rc6). I'm happy to resend the patch though, if it helps.
Thank you!
Regards,
Srivatsa
>
> arch/x86/kernel/cpu/vmware.c | 17 +++++++++++++++++
> 1 file changed, 17 insertions(+)
>
> diff --git a/arch/x86/kernel/cpu/vmware.c b/arch/x86/kernel/cpu/vmware.c
> index c04b933f48d3..420e359ed9bb 100644
> --- a/arch/x86/kernel/cpu/vmware.c
> +++ b/arch/x86/kernel/cpu/vmware.c
> @@ -27,6 +27,7 @@
> #include <linux/clocksource.h>
> #include <linux/cpu.h>
> #include <linux/reboot.h>
> +#include <linux/tboot.h>
> #include <linux/static_call.h>
> #include <asm/div64.h>
> #include <asm/x86_init.h>
> @@ -312,6 +313,21 @@ static int vmware_cpu_down_prepare(unsigned int cpu)
> local_irq_enable();
> return 0;
> }
> +
> +static void vmware_play_dead(void)
> +{
> + play_dead_common();
> + tboot_shutdown(TB_SHUTDOWN_WFS);
> +
> + /*
> + * Put the vCPU going offline in halt instead of mwait (even
> + * if mwait support is available), to make sure that the
> + * offline vCPU yields to the hypervisor (which may not happen
> + * with mwait, for example, if the guest's VMX is configured
> + * to retain the vCPU in guest context upon mwait).
> + */
> + hlt_play_dead();
> +}
> #endif
>
> static __init int activate_jump_labels(void)
> @@ -349,6 +365,7 @@ static void __init vmware_paravirt_ops_setup(void)
> #ifdef CONFIG_SMP
> smp_ops.smp_prepare_boot_cpu =
> vmware_smp_prepare_boot_cpu;
> + smp_ops.play_dead = vmware_play_dead;
> if (cpuhp_setup_state_nocalls(CPUHP_AP_ONLINE_DYN,
> "x86/vmware:online",
> vmware_cpu_online,
>
>
On Thu, Jul 21, 2022 at 01:44:33PM -0700, Srivatsa S. Bhat wrote:
> From: Srivatsa S. Bhat (VMware) <[email protected]>
>
> VMware ESXi allows enabling a passthru mwait CPU-idle state in the
> guest using the following VMX option:
>
> monitor_control.mwait_in_guest = "TRUE"
>
> This lets a vCPU in mwait to remain in guest context (instead of
> yielding to the hypervisor via a VMEXIT), which helps speed up
> wakeups from idle.
>
> However, this runs into problems with CPU hotplug, because the Linux
> CPU offline path prefers to put the vCPU-to-be-offlined in mwait
> state, whenever mwait is available. As a result, since a vCPU in mwait
> remains in guest context and does not yield to the hypervisor, an
> offline vCPU *appears* to be 100% busy as viewed from ESXi, which
> prevents the hypervisor from running other vCPUs or workloads on the
> corresponding pCPU (particularly when vCPU - pCPU mappings are
> statically defined by the user).
I would hope vCPU pinning is a mandatory thing when MWAIT passthrough it
set?
> [ Note that such a vCPU is not
> actually busy spinning though; it remains in mwait idle state in the
> guest ].
>
> Fix this by overriding the CPU offline play_dead() callback for VMware
> hypervisor, by putting the CPU in halt state (which actually yields to
> the hypervisor), even if mwait support is available.
>
> Signed-off-by: Srivatsa S. Bhat (VMware) <[email protected]>
> ---
> +static void vmware_play_dead(void)
> +{
> + play_dead_common();
> + tboot_shutdown(TB_SHUTDOWN_WFS);
> +
> + /*
> + * Put the vCPU going offline in halt instead of mwait (even
> + * if mwait support is available), to make sure that the
> + * offline vCPU yields to the hypervisor (which may not happen
> + * with mwait, for example, if the guest's VMX is configured
> + * to retain the vCPU in guest context upon mwait).
> + */
> + hlt_play_dead();
> +}
> #endif
>
> static __init int activate_jump_labels(void)
> @@ -349,6 +365,7 @@ static void __init vmware_paravirt_ops_setup(void)
> #ifdef CONFIG_SMP
> smp_ops.smp_prepare_boot_cpu =
> vmware_smp_prepare_boot_cpu;
> + smp_ops.play_dead = vmware_play_dead;
> if (cpuhp_setup_state_nocalls(CPUHP_AP_ONLINE_DYN,
> "x86/vmware:online",
> vmware_cpu_online,
No real objection here; but would not something like the below fix the
problem more generally? I'm thinking MWAIT passthrough for *any*
hypervisor doesn't want play_dead to use it.
diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index f24227bc3220..166cb3aaca8a 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -1759,6 +1759,8 @@ static inline void mwait_play_dead(void)
return;
if (!this_cpu_has(X86_FEATURE_CLFLUSH))
return;
+ if (this_cpu_has(X86_FEATURE_HYPERVISOR))
+ return;
if (__this_cpu_read(cpu_info.cpuid_level) < CPUID_MWAIT_LEAF)
return;
On 23.09.22 09:05, Peter Zijlstra wrote:
> On Thu, Jul 21, 2022 at 01:44:33PM -0700, Srivatsa S. Bhat wrote:
>> From: Srivatsa S. Bhat (VMware) <[email protected]>
>>
>> VMware ESXi allows enabling a passthru mwait CPU-idle state in the
>> guest using the following VMX option:
>>
>> monitor_control.mwait_in_guest = "TRUE"
>>
>> This lets a vCPU in mwait to remain in guest context (instead of
>> yielding to the hypervisor via a VMEXIT), which helps speed up
>> wakeups from idle.
>>
>> However, this runs into problems with CPU hotplug, because the Linux
>> CPU offline path prefers to put the vCPU-to-be-offlined in mwait
>> state, whenever mwait is available. As a result, since a vCPU in mwait
>> remains in guest context and does not yield to the hypervisor, an
>> offline vCPU *appears* to be 100% busy as viewed from ESXi, which
>> prevents the hypervisor from running other vCPUs or workloads on the
>> corresponding pCPU (particularly when vCPU - pCPU mappings are
>> statically defined by the user).
>
> I would hope vCPU pinning is a mandatory thing when MWAIT passthrough it
> set?
>
>> [ Note that such a vCPU is not
>> actually busy spinning though; it remains in mwait idle state in the
>> guest ].
>>
>> Fix this by overriding the CPU offline play_dead() callback for VMware
>> hypervisor, by putting the CPU in halt state (which actually yields to
>> the hypervisor), even if mwait support is available.
>>
>> Signed-off-by: Srivatsa S. Bhat (VMware) <[email protected]>
>> ---
>
>> +static void vmware_play_dead(void)
>> +{
>> + play_dead_common();
>> + tboot_shutdown(TB_SHUTDOWN_WFS);
>> +
>> + /*
>> + * Put the vCPU going offline in halt instead of mwait (even
>> + * if mwait support is available), to make sure that the
>> + * offline vCPU yields to the hypervisor (which may not happen
>> + * with mwait, for example, if the guest's VMX is configured
>> + * to retain the vCPU in guest context upon mwait).
>> + */
>> + hlt_play_dead();
>> +}
>> #endif
>>
>> static __init int activate_jump_labels(void)
>> @@ -349,6 +365,7 @@ static void __init vmware_paravirt_ops_setup(void)
>> #ifdef CONFIG_SMP
>> smp_ops.smp_prepare_boot_cpu =
>> vmware_smp_prepare_boot_cpu;
>> + smp_ops.play_dead = vmware_play_dead;
>> if (cpuhp_setup_state_nocalls(CPUHP_AP_ONLINE_DYN,
>> "x86/vmware:online",
>> vmware_cpu_online,
>
> No real objection here; but would not something like the below fix the
> problem more generally? I'm thinking MWAIT passthrough for *any*
> hypervisor doesn't want play_dead to use it.
>
> diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
> index f24227bc3220..166cb3aaca8a 100644
> --- a/arch/x86/kernel/smpboot.c
> +++ b/arch/x86/kernel/smpboot.c
> @@ -1759,6 +1759,8 @@ static inline void mwait_play_dead(void)
> return;
> if (!this_cpu_has(X86_FEATURE_CLFLUSH))
> return;
> + if (this_cpu_has(X86_FEATURE_HYPERVISOR))
> + return;
> if (__this_cpu_read(cpu_info.cpuid_level) < CPUID_MWAIT_LEAF)
> return;
>
With my Xen hat on I agree with this approach.
Juergen
+ kvm ML and leaving the whole mail quoted in for them.
On Fri, Sep 23, 2022 at 09:05:26AM +0200, Peter Zijlstra wrote:
> On Thu, Jul 21, 2022 at 01:44:33PM -0700, Srivatsa S. Bhat wrote:
> > From: Srivatsa S. Bhat (VMware) <[email protected]>
> >
> > VMware ESXi allows enabling a passthru mwait CPU-idle state in the
> > guest using the following VMX option:
> >
> > monitor_control.mwait_in_guest = "TRUE"
> >
> > This lets a vCPU in mwait to remain in guest context (instead of
> > yielding to the hypervisor via a VMEXIT), which helps speed up
> > wakeups from idle.
> >
> > However, this runs into problems with CPU hotplug, because the Linux
> > CPU offline path prefers to put the vCPU-to-be-offlined in mwait
> > state, whenever mwait is available. As a result, since a vCPU in mwait
> > remains in guest context and does not yield to the hypervisor, an
> > offline vCPU *appears* to be 100% busy as viewed from ESXi, which
> > prevents the hypervisor from running other vCPUs or workloads on the
> > corresponding pCPU (particularly when vCPU - pCPU mappings are
> > statically defined by the user).
>
> I would hope vCPU pinning is a mandatory thing when MWAIT passthrough it
> set?
>
> > [ Note that such a vCPU is not
> > actually busy spinning though; it remains in mwait idle state in the
> > guest ].
> >
> > Fix this by overriding the CPU offline play_dead() callback for VMware
> > hypervisor, by putting the CPU in halt state (which actually yields to
> > the hypervisor), even if mwait support is available.
> >
> > Signed-off-by: Srivatsa S. Bhat (VMware) <[email protected]>
> > ---
>
> > +static void vmware_play_dead(void)
> > +{
> > + play_dead_common();
> > + tboot_shutdown(TB_SHUTDOWN_WFS);
> > +
> > + /*
> > + * Put the vCPU going offline in halt instead of mwait (even
> > + * if mwait support is available), to make sure that the
> > + * offline vCPU yields to the hypervisor (which may not happen
> > + * with mwait, for example, if the guest's VMX is configured
> > + * to retain the vCPU in guest context upon mwait).
> > + */
> > + hlt_play_dead();
> > +}
> > #endif
> >
> > static __init int activate_jump_labels(void)
> > @@ -349,6 +365,7 @@ static void __init vmware_paravirt_ops_setup(void)
> > #ifdef CONFIG_SMP
> > smp_ops.smp_prepare_boot_cpu =
> > vmware_smp_prepare_boot_cpu;
> > + smp_ops.play_dead = vmware_play_dead;
> > if (cpuhp_setup_state_nocalls(CPUHP_AP_ONLINE_DYN,
> > "x86/vmware:online",
> > vmware_cpu_online,
>
> No real objection here; but would not something like the below fix the
> problem more generally? I'm thinking MWAIT passthrough for *any*
> hypervisor doesn't want play_dead to use it.
>
> diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
> index f24227bc3220..166cb3aaca8a 100644
> --- a/arch/x86/kernel/smpboot.c
> +++ b/arch/x86/kernel/smpboot.c
> @@ -1759,6 +1759,8 @@ static inline void mwait_play_dead(void)
> return;
> if (!this_cpu_has(X86_FEATURE_CLFLUSH))
> return;
> + if (this_cpu_has(X86_FEATURE_HYPERVISOR))
> + return;
> if (__this_cpu_read(cpu_info.cpuid_level) < CPUID_MWAIT_LEAF)
> return;
Yeah, it would be nice if we could get a consensus here from all
relevant HVs.
--
Regards/Gruss,
Boris.
https://people.kernel.org/tglx/notes-about-netiquette
On 9/23/22 3:45 AM, Borislav Petkov wrote:
> + kvm ML and leaving the whole mail quoted in for them.
>
> On Fri, Sep 23, 2022 at 09:05:26AM +0200, Peter Zijlstra wrote:
>> On Thu, Jul 21, 2022 at 01:44:33PM -0700, Srivatsa S. Bhat wrote:
>>> From: Srivatsa S. Bhat (VMware) <[email protected]>
>>>
>>> VMware ESXi allows enabling a passthru mwait CPU-idle state in the
>>> guest using the following VMX option:
>>>
>>> monitor_control.mwait_in_guest = "TRUE"
>>>
>>> This lets a vCPU in mwait to remain in guest context (instead of
>>> yielding to the hypervisor via a VMEXIT), which helps speed up
>>> wakeups from idle.
>>>
>>> However, this runs into problems with CPU hotplug, because the Linux
>>> CPU offline path prefers to put the vCPU-to-be-offlined in mwait
>>> state, whenever mwait is available. As a result, since a vCPU in mwait
>>> remains in guest context and does not yield to the hypervisor, an
>>> offline vCPU *appears* to be 100% busy as viewed from ESXi, which
>>> prevents the hypervisor from running other vCPUs or workloads on the
>>> corresponding pCPU (particularly when vCPU - pCPU mappings are
>>> statically defined by the user).
>>
>> I would hope vCPU pinning is a mandatory thing when MWAIT passthrough it
>> set?
>>
>>> [ Note that such a vCPU is not
>>> actually busy spinning though; it remains in mwait idle state in the
>>> guest ].
>>>
>>> Fix this by overriding the CPU offline play_dead() callback for VMware
>>> hypervisor, by putting the CPU in halt state (which actually yields to
>>> the hypervisor), even if mwait support is available.
>>>
>>> Signed-off-by: Srivatsa S. Bhat (VMware) <[email protected]>
>>> ---
>>
>>> +static void vmware_play_dead(void)
>>> +{
>>> + play_dead_common();
>>> + tboot_shutdown(TB_SHUTDOWN_WFS);
>>> +
>>> + /*
>>> + * Put the vCPU going offline in halt instead of mwait (even
>>> + * if mwait support is available), to make sure that the
>>> + * offline vCPU yields to the hypervisor (which may not happen
>>> + * with mwait, for example, if the guest's VMX is configured
>>> + * to retain the vCPU in guest context upon mwait).
>>> + */
>>> + hlt_play_dead();
>>> +}
>>> #endif
>>>
>>> static __init int activate_jump_labels(void)
>>> @@ -349,6 +365,7 @@ static void __init vmware_paravirt_ops_setup(void)
>>> #ifdef CONFIG_SMP
>>> smp_ops.smp_prepare_boot_cpu =
>>> vmware_smp_prepare_boot_cpu;
>>> + smp_ops.play_dead = vmware_play_dead;
>>> if (cpuhp_setup_state_nocalls(CPUHP_AP_ONLINE_DYN,
>>> "x86/vmware:online",
>>> vmware_cpu_online,
>>
>> No real objection here; but would not something like the below fix the
>> problem more generally? I'm thinking MWAIT passthrough for *any*
>> hypervisor doesn't want play_dead to use it.
>>
That would be better indeed, thank you!
>> diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
>> index f24227bc3220..166cb3aaca8a 100644
>> --- a/arch/x86/kernel/smpboot.c
>> +++ b/arch/x86/kernel/smpboot.c
>> @@ -1759,6 +1759,8 @@ static inline void mwait_play_dead(void)
>> return;
>> if (!this_cpu_has(X86_FEATURE_CLFLUSH))
>> return;
>> + if (this_cpu_has(X86_FEATURE_HYPERVISOR))
>> + return;
>> if (__this_cpu_read(cpu_info.cpuid_level) < CPUID_MWAIT_LEAF)
>> return;
>
> Yeah, it would be nice if we could get a consensus here from all
> relevant HVs.
>
I'll send out a v2 after trying out these changes.
Thank you!
Regards,
Srivatsa
VMware Photon OS