RISC-V: KVM: Account VM-scoped allocations to the VM cgroup
Subsystems (3)
PATCH 1/1 RISC-V: KVM: Account VM-scoped allocations to the VM cgroup
From: "Yuhang.chen" <[email protected]>
Charge the per-VM and per-vCPU allocations (stage-2 PGD, APLIC state, IMSIC context, vector context, FWFT config, PMU snapshot) to the allocating process's memory cgroup via GFP_KERNEL_ACCOUNT / __GFP_ACCOUNT. Per-CPU module-init allocations and transient scratch buffers are left unaccounted.
Measured results (KVM guest run in a memory cgroup; VM cgroup memory.current after guest boot):
VM cgroup memory.current 1044480 bytes
--- arch/riscv/kvm/aia_aplic.c | 3 ++- arch/riscv/kvm/aia_imsic.c | 4 ++-- arch/riscv/kvm/mmu.c | 4 ++-- arch/riscv/kvm/vcpu_pmu.c | 2 +- arch/riscv/kvm/vcpu_sbi_fwft.c | 2 +- arch/riscv/kvm/vcpu_vector.c | 4 ++-- 6 files changed, 10 insertions(+), 9 deletions(-)
arch/riscv/kvm/aia_aplic.c +2 -1
@@ -581,7 +581,8 @@ int kvm_riscv_aia_aplic_init(struct kvm *kvm)return 0 ;/* Allocate APLIC global state */ - aplic =kzalloc_flex (*aplic, irqs, kvm->arch.aia.nr_sources +1 ); + aplic =kzalloc_flex (*aplic, irqs, kvm->arch.aia.nr_sources +1 , +GFP_KERNEL_ACCOUNT );if (!aplic)return -ENOMEM ; kvm->arch.aia.aplic_state = aplic;
arch/riscv/kvm/aia_imsic.c +2 -2
@@ -1108,7 +1108,7 @@ int kvm_riscv_vcpu_aia_imsic_init(struct kvm_vcpu *vcpu)return -EINVAL ;/* Allocate IMSIC context */ - imsic =kzalloc_obj (*imsic); + imsic =kzalloc_obj (*imsic,GFP_KERNEL_ACCOUNT );if (!imsic)return -ENOMEM ; vcpu->arch.aia_context.imsic_state = imsic;
@@ -1121,7 +1121,7 @@ int kvm_riscv_vcpu_aia_imsic_init(struct kvm_vcpu *vcpu) imsic->vsfile_hgei = imsic->vsfile_cpu =-1 ;/* Setup IMSIC SW-file */ - swfile_page =alloc_pages (GFP_KERNEL | __GFP_ZERO, + swfile_page =alloc_pages (GFP_KERNEL_ACCOUNT | __GFP_ZERO,get_order (sizeof (*imsic->swfile)));if (!swfile_page) { ret = -ENOMEM ;
arch/riscv/kvm/mmu.c +2 -2
@@ -669,8 +669,8 @@ int kvm_riscv_mmu_alloc_pgd(struct kvm *kvm)return -EINVAL ; } - pgd_page =alloc_pages (GFP_KERNEL | __GFP_ZERO, -get_order (kvm_riscv_gstage_pgd_size)); + pgd_page =alloc_pages (GFP_KERNEL_ACCOUNT | __GFP_ZERO, +get_order (kvm_riscv_gstage_pgd_size));if (!pgd_page)return -ENOMEM ; kvm->arch.pgd =page_to_virt (pgd_page);
arch/riscv/kvm/vcpu_pmu.c +1 -1
@@ -452,7 +452,7 @@ int kvm_riscv_vcpu_pmu_snapshot_set_shmem(struct kvm_vcpu *vcpu, unsigned long s } } - kvpmu->sdata =kzalloc (snapshot_area_size,GFP_ATOMIC ); + kvpmu->sdata =kzalloc (snapshot_area_size,GFP_ATOMIC | __GFP_ACCOUNT);if (!kvpmu->sdata) { sbiret =SBI_ERR_FAILURE ; goto out;
arch/riscv/kvm/vcpu_sbi_fwft.c +1 -1
@@ -377,7 +377,7 @@ static int kvm_sbi_ext_fwft_init(struct kvm_vcpu *vcpu)int i; fwft->configs = kzalloc_objs(struct kvm_sbi_fwft_config, -ARRAY_SIZE (features)); +ARRAY_SIZE (features),GFP_KERNEL_ACCOUNT );if (!fwft->configs)return -ENOMEM ;
arch/riscv/kvm/vcpu_vector.c +2 -2
@@ -76,11 +76,11 @@ void kvm_riscv_vcpu_host_vector_restore(struct kvm_cpu_context *cntx)int kvm_riscv_vcpu_alloc_vector_context (struct kvm_vcpu *vcpu) { - vcpu->arch.guest_context.vector.datap =kzalloc (riscv_v_vsize,GFP_KERNEL ); + vcpu->arch.guest_context.vector.datap =kzalloc (riscv_v_vsize,GFP_KERNEL_ACCOUNT );if (!vcpu->arch.guest_context.vector.datap)return -ENOMEM ; - vcpu->arch.host_context.vector.datap =kzalloc (riscv_v_vsize,GFP_KERNEL ); + vcpu->arch.host_context.vector.datap =kzalloc (riscv_v_vsize,GFP_KERNEL_ACCOUNT );if (!vcpu->arch.host_context.vector.datap) {kfree (vcpu->arch.guest_context.vector.datap); vcpu->arch.guest_context.vector.datap =NULL ;
Signature · 1 line
2.34.1
[email protected]
AI reviewThank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:
Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:
New issues:
- [Medium] The kvpmu->sdata allocation uses GFP_ATOMIC | __GFP_ACCOUNT in a context that can safely sleep, draining atomic reserves and causing spurious memcg allocation failures.
Pre-existing issues:
- [High] Unconditional memory allocation for the PMU snapshot area leads to a memory leak if the guest repeatedly configures the snapshot. --
--- Patch [1]: [PATCH] RISC-V: KVM: Account VM-scoped allocations to the VM cgroup --- commit 7f2fdf55e4a9227b8c7e4853c498204f0c3f9d48
Author: Yuhang.chen <[email protected]>
RISC-V: KVM: Account VM-scoped allocations to the VM cgroup
Charge the per-VM and per-vCPU allocations (stage-2 PGD, APLIC state, IMSIC context, vector context, FWFT config, PMU snapshot) to the allocating process's memory cgroup via GFP_KERNEL_ACCOUNT / __GFP_ACCOUNT.
arch/riscv/kvm/vcpu_pmu.c int kvm_riscv_vcpu_pmu_snapshot_set_shmem(struct kvm_vcpu *vcpu, unsigned long s · line 455
Quoted 10 lines
arch/riscv/kvm/vcpu_pmu.c +1
-1@@ -452,7 +452,7 @@ int kvm_riscv_vcpu_pmu_snapshot_set_shmem(struct kvm_vcpu *vcpu, unsigned long s } } - kvpmu->sdata =kzalloc (snapshot_area_size,GFP_ATOMIC ); + kvpmu->sdata =kzalloc (snapshot_area_size,GFP_ATOMIC | __GFP_ACCOUNT);
[Severity: High]
This isn't a bug introduced by this patch, but does this leak memory if the guest repeatedly configures the snapshot?
If a guest makes multiple SBI_EXT_PMU_SNAPSHOT_SET_SHMEM calls, it looks like this unconditionally allocates and overwrites kvpmu->sdata without freeing the previous allocation.
[Severity: Medium]
Is it necessary to use GFP_ATOMIC here?
Since the immediate next call is kvm_vcpu_write_guest(), which can sleep:
/* No need to check writable slot explicitly as kvm_vcpu_write_guest does it internally */
if (kvm_vcpu_write_guest(vcpu, saddr, kvpmu->sdata, snapshot_area_size)) {
kfree(kvpmu->sdata);
kvpmu->sdata = NULL;
sbiret = SBI_ERR_INVALID_ADDRESS;
goto out;
}it seems this allocation could safely use GFP_KERNEL_ACCOUNT instead.
Combining __GFP_ACCOUNT with GFP_ATOMIC prevents the allocation from sleeping to perform direct reclaim, which might cause spurious memcg allocation failures when the cgroup is under pressure.
Quoted 3 lines
arch/riscv/kvm/vcpu_pmu.c +0
-0if (!kvpmu->sdata) { sbiret =SBI_ERR_FAILURE ; goto out;
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=1
Quoted 12 lines · On Thu, Jul 16, 2026 at 12:11 PM <[email protected]> wrote:
From: "Yuhang.chen" <[email protected]>
Charge the per-VM and per-vCPU allocations (stage-2 PGD, APLIC state, IMSIC context, vector context, FWFT config, PMU snapshot) to the allocating process's memory cgroup via
GFP_KERNEL_ACCOUNT/__GFP_ACCOUNT. Per-CPU module-init allocations and transient scratch buffers are left unaccounted.Measured results (KVM guest run in a memory cgroup; VM cgroup memory.current after guest boot):
Quoted 6 lines
VM cgroup memory.current 1044480 bytes
Assisted-by YuanSheng:deepseek-v4-proCo-developed-by Quan Zhou <[email protected]>Signed-off-by Quan Zhou <[email protected]>Signed-off-by Yuhang.chen <[email protected]>
LGTM.
Queued this patch for Linux-7.3
Thanks,
Anup
Quoted 107 lines
--- arch/riscv/kvm/aia_aplic.c | 3 ++- arch/riscv/kvm/aia_imsic.c | 4 ++-- arch/riscv/kvm/mmu.c | 4 ++-- arch/riscv/kvm/vcpu_pmu.c | 2 +- arch/riscv/kvm/vcpu_sbi_fwft.c | 2 +- arch/riscv/kvm/vcpu_vector.c | 4 ++-- 6 files changed, 10 insertions(+), 9 deletions(-)arch/riscv/kvm/aia_aplic.c +2
-1@@ -581,7 +581,8 @@ int kvm_riscv_aia_aplic_init(struct kvm *kvm)return 0 ;/* Allocate APLIC global state */ - aplic =kzalloc_flex (*aplic, irqs, kvm->arch.aia.nr_sources +1 ); + aplic =kzalloc_flex (*aplic, irqs, kvm->arch.aia.nr_sources +1 , +GFP_KERNEL_ACCOUNT );if (!aplic)return -ENOMEM ; kvm->arch.aia.aplic_state = aplic;arch/riscv/kvm/aia_imsic.c +2
-2@@ -1108,7 +1108,7 @@ int kvm_riscv_vcpu_aia_imsic_init(struct kvm_vcpu *vcpu)return -EINVAL ;/* Allocate IMSIC context */ - imsic =kzalloc_obj (*imsic); + imsic =kzalloc_obj (*imsic,GFP_KERNEL_ACCOUNT );if (!imsic)return -ENOMEM ; vcpu->arch.aia_context.imsic_state = imsic;@@ -1121,7 +1121,7 @@ int kvm_riscv_vcpu_aia_imsic_init(struct kvm_vcpu *vcpu) imsic->vsfile_hgei = imsic->vsfile_cpu =-1 ;/* Setup IMSIC SW-file */ - swfile_page =alloc_pages (GFP_KERNEL | __GFP_ZERO, + swfile_page =alloc_pages (GFP_KERNEL_ACCOUNT | __GFP_ZERO,get_order (sizeof (*imsic->swfile)));if (!swfile_page) { ret = -ENOMEM ;arch/riscv/kvm/mmu.c +2
-2@@ -669,8 +669,8 @@ int kvm_riscv_mmu_alloc_pgd(struct kvm *kvm)return -EINVAL ; } - pgd_page =alloc_pages (GFP_KERNEL | __GFP_ZERO, -get_order (kvm_riscv_gstage_pgd_size)); + pgd_page =alloc_pages (GFP_KERNEL_ACCOUNT | __GFP_ZERO, +get_order (kvm_riscv_gstage_pgd_size));if (!pgd_page)return -ENOMEM ; kvm->arch.pgd =page_to_virt (pgd_page);arch/riscv/kvm/vcpu_pmu.c +1
-1@@ -452,7 +452,7 @@ int kvm_riscv_vcpu_pmu_snapshot_set_shmem(struct kvm_vcpu *vcpu, unsigned long s } } - kvpmu->sdata =kzalloc (snapshot_area_size,GFP_ATOMIC ); + kvpmu->sdata =kzalloc (snapshot_area_size,GFP_ATOMIC | __GFP_ACCOUNT);if (!kvpmu->sdata) { sbiret =SBI_ERR_FAILURE ; goto out;arch/riscv/kvm/vcpu_sbi_fwft.c +1
-1@@ -377,7 +377,7 @@ static int kvm_sbi_ext_fwft_init(struct kvm_vcpu *vcpu)int i; fwft->configs = kzalloc_objs(struct kvm_sbi_fwft_config, -ARRAY_SIZE (features)); +ARRAY_SIZE (features),GFP_KERNEL_ACCOUNT );if (!fwft->configs)return -ENOMEM ;arch/riscv/kvm/vcpu_vector.c +2
-3@@ -76,11 +76,11 @@ void kvm_riscv_vcpu_host_vector_restore(struct kvm_cpu_context *cntx)int kvm_riscv_vcpu_alloc_vector_context (struct kvm_vcpu *vcpu) { - vcpu->arch.guest_context.vector.datap =kzalloc (riscv_v_vsize,GFP_KERNEL ); + vcpu->arch.guest_context.vector.datap =kzalloc (riscv_v_vsize,GFP_KERNEL_ACCOUNT );if (!vcpu->arch.guest_context.vector.datap)return -ENOMEM ; - vcpu->arch.host_context.vector.datap =kzalloc (riscv_v_vsize,GFP_KERNEL ); + vcpu->arch.host_context.vector.datap =kzalloc (riscv_v_vsize,GFP_KERNEL_ACCOUNT );if (!vcpu->arch.host_context.vector.datap) {kfree (vcpu->arch.guest_context.vector.datap); vcpu->arch.guest_context.vector.datap =NULL ; --2.34.1