2018-04-27 00:57:17

by Wanpeng Li

Subject: [PATCH v3] KVM: Extend MAX_IRQ_ROUTES to 4096 for all archs

From: Wanpeng Li <[email protected]>

Our virtual machines make use of device assignment by configuring
12 NVMe disks for high I/O performance. Each NVMe device has 129
MSI-X Table entries:
Capabilities: [50] MSI-X: Enable+ Count=129 Masked-Vector table: BAR=0 offset=00002000
The windows virtual machines fail to boot since they will map the number of
MSI-table entries that the NVMe hardware reported to the bus to msi routing
table, this will exceed the 1024. This patch extends MAX_IRQ_ROUTES to 4096
for all archs, in the future this might be extended again if needed.

Reviewed-by: Cornelia Huck <[email protected]>
Cc: Paolo Bonzini <[email protected]>
Cc: Radim Krčmář <[email protected]>
Cc: Cornelia Huck <[email protected]>
Cc: Christian Borntraeger <[email protected]>
Signed-off-by: Wanpeng Li <[email protected]>
Signed-off-by: Tonny Lu <[email protected]>
---
v2 -> v3:
* update comments in the code
v1 -> v2:
* extend MAX_IRQ_ROUTES to 4096 for all archs

include/linux/kvm_host.h | 8 +-------
1 file changed, 1 insertion(+), 7 deletions(-)

diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index 6930c63..6d6e79c 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -1045,13 +1045,7 @@ static inline int mmu_notifier_retry(struct kvm *kvm, unsigned long mmu_seq)

#ifdef CONFIG_HAVE_KVM_IRQ_ROUTING

-#ifdef CONFIG_S390
-#define KVM_MAX_IRQ_ROUTES 4096 //FIXME: we can have more than that...
-#elif defined(CONFIG_ARM64)
-#define KVM_MAX_IRQ_ROUTES 4096
-#else
-#define KVM_MAX_IRQ_ROUTES 1024
-#endif
+#define KVM_MAX_IRQ_ROUTES 4096 /* might need extension/rework in the future */

bool kvm_arch_can_set_irq_routing(struct kvm *kvm);
int kvm_set_irq_routing(struct kvm *kvm,
--
2.7.4

2018-04-27 06:45:34

by Christian Borntraeger

Subject: Re: [PATCH v3] KVM: Extend MAX_IRQ_ROUTES to 4096 for all archs

On 04/27/2018 02:55 AM, Wanpeng Li wrote:
> From: Wanpeng Li <[email protected]>
>
> Our virtual machines make use of device assignment by configuring
> 12 NVMe disks for high I/O performance. Each NVMe device has 129
> MSI-X Table entries:
> Capabilities: [50] MSI-X: Enable+ Count=129 Masked-Vector table: BAR=0 offset=00002000
> The windows virtual machines fail to boot since they will map the number of
> MSI-table entries that the NVMe hardware reported to the bus to msi routing
> table, this will exceed the 1024. This patch extends MAX_IRQ_ROUTES to 4096
> for all archs, in the future this might be extended again if needed.
>
> Reviewed-by: Cornelia Huck <[email protected]>
> Cc: Paolo Bonzini <[email protected]>
> Cc: Radim Krčmář <[email protected]>
> Cc: Cornelia Huck <[email protected]>
> Cc: Christian Borntraeger <[email protected]>
> Signed-off-by: Wanpeng Li <[email protected]>
> Signed-off-by: Tonny Lu <[email protected]>

Acked-by: Christian Borntraeger <[email protected]>

> ---
> v2 -> v3:
> * update comments in the code
> v1 -> v2:
> * extend MAX_IRQ_ROUTES to 4096 for all archs
>
> include/linux/kvm_host.h | 8 +-------
> 1 file changed, 1 insertion(+), 7 deletions(-)
>
> diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
> index 6930c63..6d6e79c 100644
> --- a/include/linux/kvm_host.h
> +++ b/include/linux/kvm_host.h
> @@ -1045,13 +1045,7 @@ static inline int mmu_notifier_retry(struct kvm *kvm, unsigned long mmu_seq)
>
> #ifdef CONFIG_HAVE_KVM_IRQ_ROUTING
>
> -#ifdef CONFIG_S390
> -#define KVM_MAX_IRQ_ROUTES 4096 //FIXME: we can have more than that...
> -#elif defined(CONFIG_ARM64)
> -#define KVM_MAX_IRQ_ROUTES 4096
> -#else
> -#define KVM_MAX_IRQ_ROUTES 1024
> -#endif
> +#define KVM_MAX_IRQ_ROUTES 4096 /* might need extension/rework in the future */
>
> bool kvm_arch_can_set_irq_routing(struct kvm *kvm);
> int kvm_set_irq_routing(struct kvm *kvm,
>

2018-05-04 12:30:29

by Wanpeng Li

Subject: Re: [PATCH v3] KVM: Extend MAX_IRQ_ROUTES to 4096 for all archs

ping,
2018-04-27 8:55 GMT+08:00 Wanpeng Li <[email protected]>:
> From: Wanpeng Li <[email protected]>
>
> Our virtual machines make use of device assignment by configuring
> 12 NVMe disks for high I/O performance. Each NVMe device has 129
> MSI-X Table entries:
> Capabilities: [50] MSI-X: Enable+ Count=129 Masked-Vector table: BAR=0 offset=00002000
> The windows virtual machines fail to boot since they will map the number of
> MSI-table entries that the NVMe hardware reported to the bus to msi routing
> table, this will exceed the 1024. This patch extends MAX_IRQ_ROUTES to 4096
> for all archs, in the future this might be extended again if needed.
>
> Reviewed-by: Cornelia Huck <[email protected]>
> Cc: Paolo Bonzini <[email protected]>
> Cc: Radim Krčmář <[email protected]>
> Cc: Cornelia Huck <[email protected]>
> Cc: Christian Borntraeger <[email protected]>
> Signed-off-by: Wanpeng Li <[email protected]>
> Signed-off-by: Tonny Lu <[email protected]>
> ---
> v2 -> v3:
> * update comments in the code
> v1 -> v2:
> * extend MAX_IRQ_ROUTES to 4096 for all archs
>
> include/linux/kvm_host.h | 8 +-------
> 1 file changed, 1 insertion(+), 7 deletions(-)
>
> diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
> index 6930c63..6d6e79c 100644
> --- a/include/linux/kvm_host.h
> +++ b/include/linux/kvm_host.h
> @@ -1045,13 +1045,7 @@ static inline int mmu_notifier_retry(struct kvm *kvm, unsigned long mmu_seq)
>
> #ifdef CONFIG_HAVE_KVM_IRQ_ROUTING
>
> -#ifdef CONFIG_S390
> -#define KVM_MAX_IRQ_ROUTES 4096 //FIXME: we can have more than that...
> -#elif defined(CONFIG_ARM64)
> -#define KVM_MAX_IRQ_ROUTES 4096
> -#else
> -#define KVM_MAX_IRQ_ROUTES 1024
> -#endif
> +#define KVM_MAX_IRQ_ROUTES 4096 /* might need extension/rework in the future */
>
> bool kvm_arch_can_set_irq_routing(struct kvm *kvm);
> int kvm_set_irq_routing(struct kvm *kvm,
> --
> 2.7.4
>

2018-05-07 16:26:59

by Paolo Bonzini

Subject: Re: [PATCH v3] KVM: Extend MAX_IRQ_ROUTES to 4096 for all archs

On 27/04/2018 02:55, Wanpeng Li wrote:
> From: Wanpeng Li <[email protected]>
>
> Our virtual machines make use of device assignment by configuring
> 12 NVMe disks for high I/O performance. Each NVMe device has 129
> MSI-X Table entries:
> Capabilities: [50] MSI-X: Enable+ Count=129 Masked-Vector table: BAR=0 offset=00002000
> The windows virtual machines fail to boot since they will map the number of
> MSI-table entries that the NVMe hardware reported to the bus to msi routing
> table, this will exceed the 1024. This patch extends MAX_IRQ_ROUTES to 4096
> for all archs, in the future this might be extended again if needed.
>
> Reviewed-by: Cornelia Huck <[email protected]>
> Cc: Paolo Bonzini <[email protected]>
> Cc: Radim Krčmář <[email protected]>
> Cc: Cornelia Huck <[email protected]>
> Cc: Christian Borntraeger <[email protected]>
> Signed-off-by: Wanpeng Li <[email protected]>
> Signed-off-by: Tonny Lu <[email protected]>
> ---
> v2 -> v3:
> * update comments in the code
> v1 -> v2:
> * extend MAX_IRQ_ROUTES to 4096 for all archs
>
> include/linux/kvm_host.h | 8 +-------
> 1 file changed, 1 insertion(+), 7 deletions(-)
>
> diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
> index 6930c63..6d6e79c 100644
> --- a/include/linux/kvm_host.h
> +++ b/include/linux/kvm_host.h
> @@ -1045,13 +1045,7 @@ static inline int mmu_notifier_retry(struct kvm *kvm, unsigned long mmu_seq)
>
> #ifdef CONFIG_HAVE_KVM_IRQ_ROUTING
>
> -#ifdef CONFIG_S390
> -#define KVM_MAX_IRQ_ROUTES 4096 //FIXME: we can have more than that...
> -#elif defined(CONFIG_ARM64)
> -#define KVM_MAX_IRQ_ROUTES 4096
> -#else
> -#define KVM_MAX_IRQ_ROUTES 1024
> -#endif
> +#define KVM_MAX_IRQ_ROUTES 4096 /* might need extension/rework in the future */
>
> bool kvm_arch_can_set_irq_routing(struct kvm *kvm);
> int kvm_set_irq_routing(struct kvm *kvm,
>

Queued, thanks.

Paolo