Skip to contentSkip to conversation

KVM: ARM: enable Cortex A7 hosts

PATCH RESEND NAK Ulrich Hecht 3 participants 3 messages Sep 26, 2013

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

PATCH RESEND 1/1 KVM: ARM: enable Cortex A7 hosts

KVM runs fine on Cortex A7 cores, so they should be enabled. Tested on an APE6EVM board (r8a73a4 SoC).

---
 arch/arm/kvm/guest.c | 2 ++
 1 file changed, 2 insertions(+)
arch/arm/kvm/guest.c +2 -0
index 152d036..05c62d5 100644
@@ -192,6 +192,8 @@ int __attribute_const__ kvm_target_cpu(void)
 	switch (part_number) {
 	case ARM_CPU_PART_CORTEX_A15:
 		return KVM_ARM_TARGET_CORTEX_A15;
+	case ARM_CPU_PART_CORTEX_A7:
+		return KVM_ARM_TARGET_CORTEX_A15;
 	default:
 		return -EINVAL;
 	}
Signature · 1 line
1.8.3.1
#1 lore raw
Quoted 24 lines · Ulrich Hecht

KVM runs fine on Cortex A7 cores, so they should be enabled. Tested on an APE6EVM board (r8a73a4 SoC).

---
 arch/arm/kvm/guest.c | 2 ++
 1 file changed, 2 insertions(+)
arch/arm/kvm/guest.c +2 -0
index 152d036..05c62d5 100644
@@ -192,6 +192,8 @@ int __attribute_const__ kvm_target_cpu(void)
 	switch (part_number) {
 	case ARM_CPU_PART_CORTEX_A15:
 		return KVM_ARM_TARGET_CORTEX_A15;
+	case ARM_CPU_PART_CORTEX_A7:
+		return KVM_ARM_TARGET_CORTEX_A15;
 	default:
 		return -EINVAL;
 	}
Signature · 1 line
1.8.3.1

nack

we need to have support and implementation for A7 cores and not try to shoehorn A7 support by pretending that it's an A15.

The fact that you have tested this and it happens to work with the workload you ran does not mean it is the right solution. At the very least, you need to document and think about the different system register implementation and deal with them correctly.

-Christoffer

#2 lore raw
Quoted 28 lines · Christoffer Dall
Quoted 24 lines · Ulrich Hecht

KVM runs fine on Cortex A7 cores, so they should be enabled. Tested on an APE6EVM board (r8a73a4 SoC).

---
 arch/arm/kvm/guest.c | 2 ++
 1 file changed, 2 insertions(+)
arch/arm/kvm/guest.c +2 -0
index 152d036..05c62d5 100644
@@ -192,6 +192,8 @@ int __attribute_const__ kvm_target_cpu(void)
 	switch (part_number) {
 	case ARM_CPU_PART_CORTEX_A15:
 		return KVM_ARM_TARGET_CORTEX_A15;
+	case ARM_CPU_PART_CORTEX_A7:
+		return KVM_ARM_TARGET_CORTEX_A15;
 	default:
 		return -EINVAL;
 	}
Signature · 1 line
1.8.3.1

nack

Quoted 7 lines

we need to have support and implementation for A7 cores and not try to shoehorn A7 support by pretending that it's an A15.

The fact that you have tested this and it happens to work with the workload you ran does not mean it is the right solution. At the very least, you need to document and think about the different system register implementation and deal with them correctly.

Not to mention that supporting Cortex-A7 requires some generic KVM/ARM fixes to work properly.

I believe Jonny is going to post these patches shortly, along with a more compelling set of changes to support Cortex-A7.

Cheers,

	M.
Signature · 1 line
Jazz is not dead. It just smells funny...