[PATCH] KVM: arm64: Mark some VM-scoped allocations as __GFP_ACCOUNT

[PATCH] KVM: arm64: Mark some VM-scoped allocations as __GFP_ACCOUNT

[Oliver Upton]

Generally speaking, any memory allocations that can be associated with a
particular VM should be charged to the cgroup of its process.
Nonetheless, there are a couple spots in KVM/arm64 that aren't currently
accounted:

 - the ccsidr array containing the virtualized cache hierarchy

 - the cpumask of supported cpus, for use of the vPMU on heterogeneous
   systems

Go ahead and set __GFP_ACCOUNT for these allocations.

Signed-off-by: Oliver Upton <oliver.upton@linux.dev>
---

 Applies to github.com/oupton/linux kvm-arm64/virtual-cache-geometry

 arch/arm64/kvm/arm.c      | 2 +-
 arch/arm64/kvm/sys_regs.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/arm64/kvm/arm.c b/arch/arm64/kvm/arm.c
index 9c5573bc4614..3619f1e472f5 100644
--- a/arch/arm64/kvm/arm.c
+++ b/arch/arm64/kvm/arm.c
@@ -146,7 +146,7 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 	if (ret)
 		goto err_unshare_kvm;
 
-	if (!zalloc_cpumask_var(&kvm->arch.supported_cpus, GFP_KERNEL)) {
+	if (!zalloc_cpumask_var(&kvm->arch.supported_cpus, GFP_KERNEL_ACCOUNT)) {
 		ret = -ENOMEM;
 		goto err_unshare_kvm;
 	}
diff --git a/arch/arm64/kvm/sys_regs.c b/arch/arm64/kvm/sys_regs.c
index 6a742af8010f..5dc13d02721e 100644
--- a/arch/arm64/kvm/sys_regs.c
+++ b/arch/arm64/kvm/sys_regs.c
@@ -160,7 +160,7 @@ static int set_ccsidr(struct kvm_vcpu *vcpu, u32 csselr, u32 val)
 		if (val == get_ccsidr(vcpu, csselr))
 			return 0;
 
-		ccsidr = kmalloc_array(CSSELR_MAX, sizeof(u32), GFP_KERNEL);
+		ccsidr = kmalloc_array(CSSELR_MAX, sizeof(u32), GFP_KERNEL_ACCOUNT);
 		if (!ccsidr)
 			return -ENOMEM;
 
-- 
2.39.1.519.gcb327c4b5f-goog

Re: [PATCH] KVM: arm64: Mark some VM-scoped allocations as __GFP_ACCOUNT

[Akihiko Odaki]

On 2023/02/07 8:52, Oliver Upton wrote:
> Generally speaking, any memory allocations that can be associated with a
> particular VM should be charged to the cgroup of its process.
> Nonetheless, there are a couple spots in KVM/arm64 that aren't currently
> accounted:
> 
>   - the ccsidr array containing the virtualized cache hierarchy
> 
>   - the cpumask of supported cpus, for use of the vPMU on heterogeneous
>     systems
> 
> Go ahead and set __GFP_ACCOUNT for these allocations.
> 
> Signed-off-by: Oliver Upton <oliver.upton@linux.dev>
> ---
> 
>   Applies to github.com/oupton/linux kvm-arm64/virtual-cache-geometry
> 
>   arch/arm64/kvm/arm.c      | 2 +-
>   arch/arm64/kvm/sys_regs.c | 2 +-
>   2 files changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/arch/arm64/kvm/arm.c b/arch/arm64/kvm/arm.c
> index 9c5573bc4614..3619f1e472f5 100644
> --- a/arch/arm64/kvm/arm.c
> +++ b/arch/arm64/kvm/arm.c
> @@ -146,7 +146,7 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
>   	if (ret)
>   		goto err_unshare_kvm;
>   
> -	if (!zalloc_cpumask_var(&kvm->arch.supported_cpus, GFP_KERNEL)) {
> +	if (!zalloc_cpumask_var(&kvm->arch.supported_cpus, GFP_KERNEL_ACCOUNT)) {
>   		ret = -ENOMEM;
>   		goto err_unshare_kvm;
>   	}
> diff --git a/arch/arm64/kvm/sys_regs.c b/arch/arm64/kvm/sys_regs.c
> index 6a742af8010f..5dc13d02721e 100644
> --- a/arch/arm64/kvm/sys_regs.c
> +++ b/arch/arm64/kvm/sys_regs.c
> @@ -160,7 +160,7 @@ static int set_ccsidr(struct kvm_vcpu *vcpu, u32 csselr, u32 val)
>   		if (val == get_ccsidr(vcpu, csselr))
>   			return 0;
>   
> -		ccsidr = kmalloc_array(CSSELR_MAX, sizeof(u32), GFP_KERNEL);
> +		ccsidr = kmalloc_array(CSSELR_MAX, sizeof(u32), GFP_KERNEL_ACCOUNT);
>   		if (!ccsidr)
>   			return -ENOMEM;
>   

Reviewed-by: Akihiko Odaki <akihiko.odaki@daynix.com>

Re: [PATCH] KVM: arm64: Mark some VM-scoped allocations as __GFP_ACCOUNT

[Gavin Shan]

On 2/7/23 10:52 AM, Oliver Upton wrote:
> Generally speaking, any memory allocations that can be associated with a
> particular VM should be charged to the cgroup of its process.
> Nonetheless, there are a couple spots in KVM/arm64 that aren't currently
> accounted:
> 
>   - the ccsidr array containing the virtualized cache hierarchy
> 
>   - the cpumask of supported cpus, for use of the vPMU on heterogeneous
>     systems
> 
> Go ahead and set __GFP_ACCOUNT for these allocations.
> 
> Signed-off-by: Oliver Upton <oliver.upton@linux.dev>
> ---
> 
>   Applies to github.com/oupton/linux kvm-arm64/virtual-cache-geometry
> 
>   arch/arm64/kvm/arm.c      | 2 +-
>   arch/arm64/kvm/sys_regs.c | 2 +-
>   2 files changed, 2 insertions(+), 2 deletions(-)
> 

Reviewed-by: Gavin Shan <gshan@redhat.com>

> diff --git a/arch/arm64/kvm/arm.c b/arch/arm64/kvm/arm.c
> index 9c5573bc4614..3619f1e472f5 100644
> --- a/arch/arm64/kvm/arm.c
> +++ b/arch/arm64/kvm/arm.c
> @@ -146,7 +146,7 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
>   	if (ret)
>   		goto err_unshare_kvm;
>   
> -	if (!zalloc_cpumask_var(&kvm->arch.supported_cpus, GFP_KERNEL)) {
> +	if (!zalloc_cpumask_var(&kvm->arch.supported_cpus, GFP_KERNEL_ACCOUNT)) {
>   		ret = -ENOMEM;
>   		goto err_unshare_kvm;
>   	}
> diff --git a/arch/arm64/kvm/sys_regs.c b/arch/arm64/kvm/sys_regs.c
> index 6a742af8010f..5dc13d02721e 100644
> --- a/arch/arm64/kvm/sys_regs.c
> +++ b/arch/arm64/kvm/sys_regs.c
> @@ -160,7 +160,7 @@ static int set_ccsidr(struct kvm_vcpu *vcpu, u32 csselr, u32 val)
>   		if (val == get_ccsidr(vcpu, csselr))
>   			return 0;
>   
> -		ccsidr = kmalloc_array(CSSELR_MAX, sizeof(u32), GFP_KERNEL);
> +		ccsidr = kmalloc_array(CSSELR_MAX, sizeof(u32), GFP_KERNEL_ACCOUNT);
>   		if (!ccsidr)
>   			return -ENOMEM;
>   
> 

Re: [PATCH] KVM: arm64: Mark some VM-scoped allocations as __GFP_ACCOUNT

[Marc Zyngier]

On Mon, 06 Feb 2023 23:52:29 +0000,
Oliver Upton <oliver.upton@linux.dev> wrote:
> 
> Generally speaking, any memory allocations that can be associated with a
> particular VM should be charged to the cgroup of its process.
> Nonetheless, there are a couple spots in KVM/arm64 that aren't currently
> accounted:
> 
>  - the ccsidr array containing the virtualized cache hierarchy
> 
>  - the cpumask of supported cpus, for use of the vPMU on heterogeneous
>    systems
> 
> Go ahead and set __GFP_ACCOUNT for these allocations.
> 
> Signed-off-by: Oliver Upton <oliver.upton@linux.dev>

Reviewed-by: Marc Zyngier <maz@kernel.org>

	M.

-- 
Without deviation from the norm, progress is not possible.

Re: [PATCH] KVM: arm64: Mark some VM-scoped allocations as __GFP_ACCOUNT

[Oliver Upton]

On Mon, 6 Feb 2023 23:52:29 +0000, Oliver Upton wrote:
> Generally speaking, any memory allocations that can be associated with a
> particular VM should be charged to the cgroup of its process.
> Nonetheless, there are a couple spots in KVM/arm64 that aren't currently
> accounted:
> 
>  - the ccsidr array containing the virtualized cache hierarchy
> 
> [...]

Applied to kvmarm/next, thanks!

[1/1] KVM: arm64: Mark some VM-scoped allocations as __GFP_ACCOUNT
      https://git.kernel.org/kvmarm/kvmarm/c/5f623a598d12

--
Best,
Oliver