drivers: firmware: psci: add basic v1.0 support

According to the PSCI specification and the SMC/HVC calling convention, PSCI function_ids that are not implemented must return NOT_SUPPORTED as return value.

Current KVM implementation takes an unhandled PSCI function_id as an error and injects an undefined instruction into the guest if PSCI implementation is called with a function_id that is not handled by the resident PSCI version (ie it is not implemented), which is not the behaviour expected by a guest when calling a PSCI function_id that is not implemented.

This patch fixes this issue by returning NOT_SUPPORTED whenever the kvm PSCI call is executed for a function_id that is not implemented by the PSCI kvm layer.

Signed-off-by Lorenzo Pieralisi <[email protected]>

Reported-by Sudeep Holla <[email protected]>

Cc Christoffer Dall <[email protected]>

Cc Anup Patel <[email protected]>

Cc Marc Zyngier <[email protected]>

---
  arch/arm/kvm/psci.c | 6 +++---
  1 file changed, 3 insertions(+), 3 deletions(-)

arch/arm/kvm/psci.c +0 -0

index 7e9398c..ec5943b 100644

According to the PSCI specification and the SMC/HVC calling convention, PSCI function_ids that are not implemented must return NOT_SUPPORTED as return value.

Current KVM implementation takes an unhandled PSCI function_id as an error and injects an undefined instruction into the guest if PSCI implementation is called with a function_id that is not handled by the resident PSCI version (ie it is not implemented), which is not the behaviour expected by a guest when calling a PSCI function_id that is not implemented.

This patch fixes this issue by returning NOT_SUPPORTED whenever the kvm PSCI call is executed for a function_id that is not implemented by the PSCI kvm layer.

Signed-off-by Lorenzo Pieralisi <[email protected]>

Reported-by Sudeep Holla <[email protected]>

Cc Christoffer Dall <[email protected]>

Cc Anup Patel <[email protected]>

Cc Marc Zyngier <[email protected]>

---
 arch/arm/kvm/psci.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

arch/arm/kvm/psci.c +2 -1

index 7e9398c..ec5943b 100644
@@ -273,7 +273,8 @@ static int kvm_psci_0_2_call(struct kvm_vcpu *vcpu)
 		ret = 0;
 		break;
 	default:
-		return -EINVAL;
+		val = PSCI_RET_NOT_SUPPORTED;
+		break;
 	}
 
 	*vcpu_reg(vcpu, 0) = val;

@@ -295,10 +296,9 @@ static int kvm_psci_0_1_call(struct kvm_vcpu *vcpu)
 		break;

Hi Lorenzo,

Quoted 49 lines · Lorenzo Pieralisi

According to the PSCI specification and the SMC/HVC calling convention, PSCI function_ids that are not implemented must return NOT_SUPPORTED as return value.

Current KVM implementation takes an unhandled PSCI function_id as an error and injects an undefined instruction into the guest if PSCI implementation is called with a function_id that is not handled by the resident PSCI version (ie it is not implemented), which is not the behaviour expected by a guest when calling a PSCI function_id that is not implemented.

This patch fixes this issue by returning NOT_SUPPORTED whenever the kvm PSCI call is executed for a function_id that is not implemented by the PSCI kvm layer.

Signed-off-by Lorenzo Pieralisi <[email protected]>

Reported-by Sudeep Holla <[email protected]>

Cc Christoffer Dall <[email protected]>

Cc Anup Patel <[email protected]>

Cc Marc Zyngier <[email protected]>

---
 arch/arm/kvm/psci.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

arch/arm/kvm/psci.c +3 -3

index 7e9398c..ec5943b 100644
@@ -273,7 +273,8 @@ static int kvm_psci_0_2_call(struct kvm_vcpu *vcpu)
 		ret = 0;
 		break;
 	default:
-		return -EINVAL;
+		val = PSCI_RET_NOT_SUPPORTED;
+		break;
 	}
 
 	*vcpu_reg(vcpu, 0) = val;

@@ -295,10 +296,9 @@ static int kvm_psci_0_1_call(struct kvm_vcpu *vcpu)
 		break;
 	case KVM_PSCI_FN_CPU_SUSPEND:
 	case KVM_PSCI_FN_MIGRATE:
+	default:
 		val = PSCI_RET_NOT_SUPPORTED;
 		break;
-	default:
-		return -EINVAL;
 	}
 

arch/arm/kvm/psci.c static int kvm_psci_0_1_call(struct kvm_vcpu *vcpu) · line 304

Quoted 2 lines

arch/arm/kvm/psci.c +0 -0

 	*vcpu_reg(vcpu, 0) = val;

Looks good to me. How do you want to proceed with this one? can I take it independently from the rest of the series? Or would you prefer it being kept as a whole?

Functions implemented on arm64 to check if a power_state parameter is valid and if the power_state implies context loss are not arm64 specific and should be moved to generic code so that they can be reused on arm systems too.

This patch moves the functions handling the power_state parameter to generic PSCI firmware layer code.

Signed-off-by Lorenzo Pieralisi <[email protected]>

Acked-by Sudeep Holla <[email protected]>

Cc Will Deacon <[email protected]>

Cc Catalin Marinas <[email protected]>

Cc Mark Rutland <[email protected]>

---
 arch/arm64/kernel/psci.c | 14 --------------
 drivers/firmware/psci.c  | 15 +++++++++++++++
 include/linux/psci.h     |  2 ++
 3 files changed, 17 insertions(+), 14 deletions(-)

PSCI v1.0 augmented the power_state parameter format specification (extended stateid) and introduced a way to probe it through the PSCI_FEATURES interface.

This patch implements code that detects the power_state format at run-time through the PSCI_FEATURES interface, so that the power_state argument can be properly detected and validated in the kernel according to the information provided through firmware.

Signed-off-by Lorenzo Pieralisi <[email protected]>

Cc Mark Rutland <[email protected]>

---
 drivers/firmware/psci.c   | 34 ++++++++++++++++++++++++++++++++--
 include/uapi/linux/psci.h | 12 ++++++++++++
 2 files changed, 44 insertions(+), 2 deletions(-)

drivers/firmware/psci.c +32 -2

index 2330099..4063784 100644
@@ -74,14 +74,34 @@ static u32 psci_function_id[PSCI_FN_MAX];
                                PSCI_0_2_POWER_STATE_TYPE_MASK | \
                                PSCI_0_2_POWER_STATE_AFFL_MASK)
 
+#define PSCI_EXT_POWER_STATE_MASK              \
+                               (PSCI_EXT_POWER_STATE_ID_MASK | \
+                               PSCI_EXT_POWER_STATE_TYPE_MASK)
+
+static u32 psci_cpu_suspend_feature;
+
+static inline bool psci_has_ext_power_state(void)
+{
+       return psci_cpu_suspend_feature &
+                               PSCI_FEATURES_CPU_SUSPEND_PF_MASK;
+}
+
 bool psci_power_state_loses_context(u32 state)
 {
-       return state & PSCI_0_2_POWER_STATE_TYPE_MASK;
+       const u32 mask = psci_has_ext_power_state() ?
+                                       PSCI_EXT_POWER_STATE_TYPE_MASK :
+                                       PSCI_0_2_POWER_STATE_TYPE_MASK;
+
+       return state & mask;
 }
 
 bool psci_power_state_is_valid(u32 state)
 {
-       return !(state & ~PSCI_0_2_POWER_STATE_MASK);
+       const u32 valid_mask = psci_has_ext_power_state() ?
+                              PSCI_EXT_POWER_STATE_MASK :
+                              PSCI_0_2_POWER_STATE_MASK;
+
+       return !(state & ~valid_mask);
 }
 
 static int psci_to_linux_errno(int errno)

@@ -202,6 +222,14 @@ static int __init psci_features(u32 psci_func_id)
                              psci_func_id, 0, 0);
 }
 
+static void __init psci_init_cpu_suspend(void)
+{
+       int feature = psci_features(psci_function_id[PSCI_FN_CPU_SUSPEND]);
+
+       if (feature != PSCI_RET_NOT_SUPPORTED)
+               psci_cpu_suspend_feature = feature;
+}
+
 /*
  * Detect the presence of a resident Trusted OS which may cause CPU_OFF to
  * return DENIED (which would be fatal).

@@ -280,6 +308,8 @@ static int __init psci_probe(void)
 
        psci_init_migrate();
 
+       psci_init_cpu_suspend();
+
        return 0;
 }
 

include/uapi/linux/psci.h +0 -0

index 187b828d..2598d7c 100644

include/uapi/linux/psci.h +0 -1

 /* PSCI return values (inclusive of all PSCI versions) */
 #define PSCI_RET_SUCCESS                       0
 #define PSCI_RET_NOT_SUPPORTED                 -1
--

2.2.1

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel at lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

On 29 May 2015 at 08:16, Lorenzo Pieralisi <[email protected]> wrote:

Quoted 8 lines

include/uapi/linux/psci.h +0 -0

index 187b828d..2598d7c 100644
@@ -58,6 +58,13 @@
 #define PSCI_0_2_POWER_STATE_AFFL_MASK         \
                                (0x3 << PSCI_0_2_POWER_STATE_AFFL_SHIFT)
 

include/uapi/linux/psci.h line 67

Quoted 7 lines

include/uapi/linux/psci.h +7 -0

+/* PSCI extended power state encoding for CPU_SUSPEND function */
+#define PSCI_EXT_POWER_STATE_ID_MASK           0xfffffff
+#define PSCI_EXT_POWER_STATE_ID_SHIFT          0
+#define PSCI_EXT_POWER_STATE_TYPE_SHIFT                30
+#define PSCI_EXT_POWER_STATE_TYPE_MASK         \
+                               (0x1 << PSCI_EXT_POWER_STATE_TYPE_SHIFT)
+

For consistency, dont you need version numbers here, like we have for v0.2?

Quoted 7 lines

include/uapi/linux/psci.h +0 -0

 /* PSCI v0.2 affinity level state returned by AFFINITY_INFO */
 #define PSCI_0_2_AFFINITY_LEVEL_ON             0
 #define PSCI_0_2_AFFINITY_LEVEL_OFF            1

@@ -78,6 +85,11 @@
 #define PSCI_VERSION_MINOR(ver)                        \
                ((ver) & PSCI_VERSION_MINOR_MASK)
 

include/uapi/linux/psci.h line 92

Quoted 5 lines

include/uapi/linux/psci.h +5 -0

+/* PSCI features decoding (>=1.0) */
+#define PSCI_FEATURES_CPU_SUSPEND_PF_SHIFT     1
+#define PSCI_FEATURES_CPU_SUSPEND_PF_MASK      \
+                       (0x1 << PSCI_FEATURES_CPU_SUSPEND_PF_SHIFT)
+

Likewise.