[PATCH V2 0/2] KVM: arm64: Increase number of user memslots

[PATCH V2 0/2] KVM: arm64: Increase number of user memslots

[linucherian at gmail.com]

From: Linu Cherian <linu.cherian@cavium.com>

v1 -> v2:
- Enabling KVM_CAP_NR_MEMSLOTS for arm/arm64 moved to separate patch.
- Updated commit message so that what is reported to userspace is explicit. 

Linu Cherian (2):
  KVM: arm/arm64: Enable KVM_CAP_NR_MEMSLOTS on arm/arm64
  KVM: arm64: Increase number of user memslots to 508

 arch/arm/kvm/arm.c                | 3 +++
 arch/arm64/include/asm/kvm_host.h | 2 +-
 2 files changed, 4 insertions(+), 1 deletion(-)

-- 
1.9.1

[PATCH V2 1/2] KVM: arm/arm64: Enable KVM_CAP_NR_MEMSLOTS on arm/arm64

[linucherian at gmail.com]

From: Linu Cherian <linu.cherian@cavium.com>

Return KVM_USER_MEM_SLOTS for userspace capability query on
NR_MEMSLOTS.

Signed-off-by: Linu Cherian <linu.cherian@cavium.com>
---
 arch/arm/kvm/arm.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/arch/arm/kvm/arm.c b/arch/arm/kvm/arm.c
index 9d74464..b64b47f 100644
--- a/arch/arm/kvm/arm.c
+++ b/arch/arm/kvm/arm.c
@@ -221,6 +221,9 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
 	case KVM_CAP_MAX_VCPUS:
 		r = KVM_MAX_VCPUS;
 		break;
+	case KVM_CAP_NR_MEMSLOTS:
+		r = KVM_USER_MEM_SLOTS;
+		break;
 	case KVM_CAP_MSI_DEVID:
 		if (!kvm)
 			r = -EINVAL;
-- 
1.9.1

[PATCH V2 1/2] KVM: arm/arm64: Enable KVM_CAP_NR_MEMSLOTS on arm/arm64

[Christoffer Dall]

On Wed, Feb 08, 2017 at 12:45:40PM +0530, linucherian at gmail.com wrote:
> From: Linu Cherian <linu.cherian@cavium.com>
> 
> Return KVM_USER_MEM_SLOTS for userspace capability query on
> NR_MEMSLOTS.

This feature seems to be undocumented (I cannot find it in
Documentation/virtual/kvm/api.txt).

Can we add documentation for it?

What does user space do currently without this capability?  Does it just
rely on some default?

Thanks,
-Christoffer

> 
> Signed-off-by: Linu Cherian <linu.cherian@cavium.com>
> ---
>  arch/arm/kvm/arm.c | 3 +++
>  1 file changed, 3 insertions(+)
> 
> diff --git a/arch/arm/kvm/arm.c b/arch/arm/kvm/arm.c
> index 9d74464..b64b47f 100644
> --- a/arch/arm/kvm/arm.c
> +++ b/arch/arm/kvm/arm.c
> @@ -221,6 +221,9 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
>  	case KVM_CAP_MAX_VCPUS:
>  		r = KVM_MAX_VCPUS;
>  		break;
> +	case KVM_CAP_NR_MEMSLOTS:
> +		r = KVM_USER_MEM_SLOTS;
> +		break;
>  	case KVM_CAP_MSI_DEVID:
>  		if (!kvm)
>  			r = -EINVAL;
> -- 
> 1.9.1
> 

[PATCH V2 1/2] KVM: arm/arm64: Enable KVM_CAP_NR_MEMSLOTS on arm/arm64

[Linu Cherian]

On Thu, Feb 23, 2017 at 3:40 PM, Christoffer Dall <cdall@linaro.org> wrote:
> On Wed, Feb 08, 2017 at 12:45:40PM +0530, linucherian at gmail.com wrote:
>> From: Linu Cherian <linu.cherian@cavium.com>
>>
>> Return KVM_USER_MEM_SLOTS for userspace capability query on
>> NR_MEMSLOTS.
>
> This feature seems to be undocumented (I cannot find it in
> Documentation/virtual/kvm/api.txt).
>
> Can we add documentation for it?
>
> What does user space do currently without this capability?  Does it just
> rely on some default?
>

Yes, without this capability,  qemu userspace take the default value of 32.
Will give it a try on the documentation part as a separate patch.

[PATCH V2 2/2] KVM: arm64: Increase number of user memslots to 508

[linucherian at gmail.com]

From: Linu Cherian <linu.cherian@cavium.com>

Having only 32 memslots is a real constraint for the maximum
number of PCI devices that can be assigned to a single guest.
Assuming each PCI device/virtual function having two memory BAR
regions, we could assign only 15 devices/virtual functions to a
guest.

Hence increase KVM_USER_MEM_SLOTS to 508, so that KVM_MEM_SLOTS_NUM is
512 as done in other archs like x86 and powerpc.

Signed-off-by: Linu Cherian <linu.cherian@cavium.com>
---
 arch/arm64/include/asm/kvm_host.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h
index e505038..88f017d 100644
--- a/arch/arm64/include/asm/kvm_host.h
+++ b/arch/arm64/include/asm/kvm_host.h
@@ -30,7 +30,7 @@
 
 #define __KVM_HAVE_ARCH_INTC_INITIALIZED
 
-#define KVM_USER_MEM_SLOTS 32
+#define KVM_USER_MEM_SLOTS 508
 #define KVM_PRIVATE_MEM_SLOTS 4
 #define KVM_COALESCED_MMIO_PAGE_OFFSET 1
 #define KVM_HALT_POLL_NS_DEFAULT 500000
-- 
1.9.1

[PATCH V2 2/2] KVM: arm64: Increase number of user memslots to 508

[Christoffer Dall]

On Wed, Feb 08, 2017 at 12:45:41PM +0530, linucherian at gmail.com wrote:
> From: Linu Cherian <linu.cherian@cavium.com>
> 
> Having only 32 memslots is a real constraint for the maximum
> number of PCI devices that can be assigned to a single guest.
> Assuming each PCI device/virtual function having two memory BAR
> regions, we could assign only 15 devices/virtual functions to a
> guest.
> 
> Hence increase KVM_USER_MEM_SLOTS to 508, so that KVM_MEM_SLOTS_NUM is
> 512 as done in other archs like x86 and powerpc.

Actually on powerpc they just define KVM_USER_MEM_SLOTS as 512 directly.

On x86 they define them as 509 and have 3 private mem slots.

I don't understand the difference and as far as I can tell we don't have
any private memslots on arm/arm64, so this is just weird to me.

Thanks,
-Christoffer

> 
> Signed-off-by: Linu Cherian <linu.cherian@cavium.com>
> ---
>  arch/arm64/include/asm/kvm_host.h | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h
> index e505038..88f017d 100644
> --- a/arch/arm64/include/asm/kvm_host.h
> +++ b/arch/arm64/include/asm/kvm_host.h
> @@ -30,7 +30,7 @@
>  
>  #define __KVM_HAVE_ARCH_INTC_INITIALIZED
>  
> -#define KVM_USER_MEM_SLOTS 32
> +#define KVM_USER_MEM_SLOTS 508
>  #define KVM_PRIVATE_MEM_SLOTS 4
>  #define KVM_COALESCED_MMIO_PAGE_OFFSET 1
>  #define KVM_HALT_POLL_NS_DEFAULT 500000
> -- 
> 1.9.1
> 

[PATCH V2 2/2] KVM: arm64: Increase number of user memslots to 508

[Linu Cherian]

On Thu, Feb 23, 2017 at 3:44 PM, Christoffer Dall <cdall@linaro.org> wrote:
> On Wed, Feb 08, 2017 at 12:45:41PM +0530, linucherian at gmail.com wrote:
>> From: Linu Cherian <linu.cherian@cavium.com>
>>
>> Having only 32 memslots is a real constraint for the maximum
>> number of PCI devices that can be assigned to a single guest.
>> Assuming each PCI device/virtual function having two memory BAR
>> regions, we could assign only 15 devices/virtual functions to a
>> guest.
>>
>> Hence increase KVM_USER_MEM_SLOTS to 508, so that KVM_MEM_SLOTS_NUM is
>> 512 as done in other archs like x86 and powerpc.
>
> Actually on powerpc they just define KVM_USER_MEM_SLOTS as 512 directly.
>
> On x86 they define them as 509 and have 3 private mem slots.
>
> I don't understand the difference and as far as I can tell we don't have
> any private memslots on arm/arm64, so this is just weird to me.
>

Since the KVM_PRIVATE_MEM_SLOTS was already defined as 4, just kept
it untouched. Should we remove this and keep KVM_USER_MEM_SLOTS as 512 ?

[PATCH V2 2/2] KVM: arm64: Increase number of user memslots to 508

[Christoffer Dall]

On Sun, Feb 26, 2017 at 12:36:58PM +0530, Linu Cherian wrote:
> On Thu, Feb 23, 2017 at 3:44 PM, Christoffer Dall <cdall@linaro.org> wrote:
> > On Wed, Feb 08, 2017 at 12:45:41PM +0530, linucherian at gmail.com wrote:
> >> From: Linu Cherian <linu.cherian@cavium.com>
> >>
> >> Having only 32 memslots is a real constraint for the maximum
> >> number of PCI devices that can be assigned to a single guest.
> >> Assuming each PCI device/virtual function having two memory BAR
> >> regions, we could assign only 15 devices/virtual functions to a
> >> guest.
> >>
> >> Hence increase KVM_USER_MEM_SLOTS to 508, so that KVM_MEM_SLOTS_NUM is
> >> 512 as done in other archs like x86 and powerpc.
> >
> > Actually on powerpc they just define KVM_USER_MEM_SLOTS as 512 directly.
> >
> > On x86 they define them as 509 and have 3 private mem slots.
> >
> > I don't understand the difference and as far as I can tell we don't have
> > any private memslots on arm/arm64, so this is just weird to me.
> >
> 
> Since the KVM_PRIVATE_MEM_SLOTS was already defined as 4, just kept
> it untouched. Should we remove this and keep KVM_USER_MEM_SLOTS as 512 ?

Do we have any use for KVM_PRIVATE_MEM_SLOTSA being 4?  I can't seem to
see this.

I think we should either get rid of the definition or at least make sure
we understand why we're doing things this way.

Thanks,
-Christoffer