KVM: x86: Drop arbitraty KVM_SOFT_MAX_VCPUS
RFC PATCH 1/1 KVM: x86: Drop arbitraty KVM_SOFT_MAX_VCPUS
KVM_CAP_NR_VCPUS is used to get the "recommended" maximum number of VCPUs and arm64/mips/riscv report num_online_cpus(). Powerpc reports either num_online_cpus() or num_present_cpus(), s390 has multiple constants depending on hardware features. On x86, KVM reports an arbitrary value of '710' which is supposed to be the maximum tested value but it's possible to test all KVM_MAX_VCPUS even when there are less physical CPUs available.
Drop the arbitrary '710' value and return num_online_cpus() on x86 as well. The recommendation will match other architectures and will mean 'no CPU overcommit'.
For reference, QEMU only queries KVM_CAP_NR_VCPUS to print a warning when the requested vCPU number exceeds it. The static limit of '710' is quite weird as smaller systems with just a few physical CPUs should certainly "recommend" less.
--- arch/x86/include/asm/kvm_host.h | 1 - arch/x86/kvm/x86.c | 2 +- 2 files changed, 1 insertion(+), 2 deletions(-)
arch/x86/include/asm/kvm_host.h +0 -1
@@ -38,7 +38,6 @@#define __KVM_HAVE_ARCH_VCPU_DEBUGFS#define KVM_MAX_VCPUS 1024 -#define KVM_SOFT_MAX_VCPUS 710/* * In x86, the VCPU ID corresponds to the APIC ID, and APIC IDs
arch/x86/kvm/x86.c +1 -1
@@ -4137,7 +4137,7 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext) r = !static_call (kvm_x86_cpu_has_accelerated_tpr)();break ;case KVM_CAP_NR_VCPUS : - r =KVM_SOFT_MAX_VCPUS ; + r =num_online_cpus ();break ;case KVM_CAP_MAX_VCPUS : r =KVM_MAX_VCPUS ;
Signature · 1 line
2.33.1
Quoted 12 lines · Vitaly Kuznetsov
KVM_CAP_NR_VCPUSis used to get the "recommended" maximum number of VCPUs and arm64/mips/riscv reportnum_online_cpus(). Powerpc reports eithernum_online_cpus()ornum_present_cpus(), s390 has multiple constants depending on hardware features. On x86, KVM reports an arbitrary value of '710' which is supposed to be the maximum tested value but it's possible to test allKVM_MAX_VCPUSeven when there are less physical CPUs available.Drop the arbitrary '710' value and return
num_online_cpus()on x86 as well. The recommendation will match other architectures and will mean 'no CPU overcommit'.
Quoted 7 lines
For reference, QEMU only queries
KVM_CAP_NR_VCPUSto print a warning when the requested vCPU number exceeds it. The static limit of '710' is quite weird as smaller systems with just a few physical CPUs should certainly "recommend" less.Suggested-by Eduardo Habkost <[email protected]>Signed-off-by Vitaly Kuznetsov <[email protected]>
Yes, this is a good idea. We cannot move it entirely to common code due to POWER's handling of secondary threads in hypervisors; still, this is as close as we can get to a common idea of what KVM_CAP_NR_VCPUS means.
Queued, thanks.
Paolo
Quoted 31 lines
--- arch/x86/include/asm/kvm_host.h | 1 - arch/x86/kvm/x86.c | 2 +- 2 files changed, 1 insertion(+), 2 deletions(-)arch/x86/include/asm/kvm_host.h +0
-1@@ -38,7 +38,6 @@#define __KVM_HAVE_ARCH_VCPU_DEBUGFS#define KVM_MAX_VCPUS 1024 -#define KVM_SOFT_MAX_VCPUS 710/* * In x86, the VCPU ID corresponds to the APIC ID, and APIC IDs arch/x86/kvm/x86.c +1
-1@@ -4137,7 +4137,7 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext) r = !static_call (kvm_x86_cpu_has_accelerated_tpr)();break ;case KVM_CAP_NR_VCPUS : - r =KVM_SOFT_MAX_VCPUS ; + r =num_online_cpus ();break ;case KVM_CAP_MAX_VCPUS : r =KVM_MAX_VCPUS ;
Sean Christopherson
maintainerQuoted 11 lines · Paolo Bonzini
arch/x86/kvm/x86.c int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext) · line 4140
Quoted 10 lines · Vitaly Kuznetsov
arch/x86/kvm/x86.c +1
-1@@ -4137,7 +4137,7 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext) r = !static_call (kvm_x86_cpu_has_accelerated_tpr)();break ;case KVM_CAP_NR_VCPUS : - r =KVM_SOFT_MAX_VCPUS ; + r =num_online_cpus ();
I doubt it matters much in practice, but this really should be
r = min(num_online_cpus(), KVM_MAX_VCPUS);
Quoted 5 lines
Quoted 4 lines
arch/x86/kvm/x86.c +0
-0break ;case KVM_CAP_MAX_VCPUS : r =KVM_MAX_VCPUS ;
Quoted 15 lines · Sean Christopherson
Quoted 11 lines · Paolo Bonzini
arch/x86/kvm/x86.c int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext) · line 4140
Quoted 10 lines · Vitaly Kuznetsov
arch/x86/kvm/x86.c +1
-1@@ -4137,7 +4137,7 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext) r = !static_call (kvm_x86_cpu_has_accelerated_tpr)();break ;case KVM_CAP_NR_VCPUS : - r =KVM_SOFT_MAX_VCPUS ; + r =num_online_cpus ();I doubt it matters much in practice, but this really should be
Quoted 2 lines
r = min(num_online_cpus(), KVM_MAX_VCPUS);
Nice catch, actually! It makes no sense to recommend > KVM_MAX_VCPUS. We should fix this across all arches though, I'll take that as an action item.
Signature · 1 line
Vitaly
Quoted 18 lines · Paolo Bonzini
Quoted 17 lines · Vitaly Kuznetsov
KVM_CAP_NR_VCPUSis used to get the "recommended" maximum number of VCPUs and arm64/mips/riscv reportnum_online_cpus(). Powerpc reports eithernum_online_cpus()ornum_present_cpus(), s390 has multiple constants depending on hardware features. On x86, KVM reports an arbitrary value of '710' which is supposed to be the maximum tested value but it's possible to test allKVM_MAX_VCPUSeven when there are less physical CPUs available.Drop the arbitrary '710' value and return
num_online_cpus()on x86 as well. The recommendation will match other architectures and will mean 'no CPU overcommit'.For reference, QEMU only queries
KVM_CAP_NR_VCPUSto print a warning when the requested vCPU number exceeds it. The static limit of '710' is quite weird as smaller systems with just a few physical CPUs should certainly "recommend" less.
Quoted 7 lines
Quoted 2 lines
Suggested-by Eduardo Habkost <[email protected]>Signed-off-by Vitaly Kuznetsov <[email protected]>Yes, this is a good idea. We cannot move it entirely to common code due to POWER's handling of secondary threads in hypervisors; still, this is as close as we can get to a common idea of what
KVM_CAP_NR_VCPUSmeans.
S390's idea is also different and while I don't understand at all all these hardware features, KVM_CAP_NR_VCPUS == KVM_CAP_MAX_VCPUS (afaict). This was the first reason to keep KVM_CAP_NR_VCPUS handling in arch specific code.
Signature · 1 line
Vitaly