Skip to contentSkip to conversation

arm64: Expose AIDR_EL1 via sysfs

PATCH Applied Oliver Upton 4 participants 4 messages Apr 3, 2025 – Apr 29, 2025

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

PATCH 1/1 arm64: Expose AIDR_EL1 via sysfs

The KVM PV ABI recently added a feature that allows the VM to discover the set of physical CPU implementations, identified by a tuple of {MIDR_EL1, REVIDR_EL1, AIDR_EL1}. Unlike other KVM PV features, the expectation is that the VMM implements the hypercall instead of KVM as it has the authoritative view of where the VM gets scheduled.

To do this the VMM needs to know the values of these registers on any CPU in the system. While MIDR_EL1 and REVIDR_EL1 are already exposed, AIDR_EL1 is not. Provide it in sysfs along with the other identification registers.

---
 Documentation/ABI/testing/sysfs-devices-system-cpu |  1 +
 Documentation/arch/arm64/cpu-feature-registers.rst | 13 +++++++------
 arch/arm64/include/asm/cpu.h                       |  1 +
 arch/arm64/kernel/cpuinfo.c                        |  3 +++
 4 files changed, 12 insertions(+), 6 deletions(-)
Documentation/ABI/testing/sysfs-devices-system-cpu +1 -0
index 206079d3bd5b..9bbf4c27c237 100644
@@ -485,6 +485,7 @@ What:		/sys/devices/system/cpu/cpuX/regs/
 		/sys/devices/system/cpu/cpuX/regs/identification/
 		/sys/devices/system/cpu/cpuX/regs/identification/midr_el1
 		/sys/devices/system/cpu/cpuX/regs/identification/revidr_el1
+		/sys/devices/system/cpu/cpuX/regs/identification/aidr_el1
 		/sys/devices/system/cpu/cpuX/regs/identification/smidr_el1
 Date:		June 2016
 Contact:	Linux ARM Kernel Mailing list <[email protected]>
Documentation/arch/arm64/cpu-feature-registers.rst +7 -6
index 253e9743de2f..add66afc7b03 100644
@@ -72,14 +72,15 @@ there are some issues with their usage.
     process could be migrated to another CPU by the time it uses the
     register value, unless the CPU affinity is set. Hence, there is no
     guarantee that the value reflects the processor that it is
-    currently executing on. The REVIDR is not exposed due to this
-    constraint, as REVIDR makes sense only in conjunction with the
-    MIDR. Alternately, MIDR_EL1 and REVIDR_EL1 are exposed via sysfs
-    at::
+    currently executing on. REVIDR and AIDR are not exposed due to this
+    constraint, as these registers only make sense in conjunction with
+    the MIDR. Alternately, MIDR_EL1, REVIDR_EL1, and AIDR_EL1 are exposed
+    via sysfs at::
 
 	/sys/devices/system/cpu/cpu$ID/regs/identification/
-	                                              \- midr
-	                                              \- revidr
+	                                              \- midr_el1
+	                                              \- revidr_el1
+	                                              \- aidr_el1
 
 3. Implementation
 --------------------
arch/arm64/include/asm/cpu.h +1 -0
index 81e4157f92b7..71493b760b83 100644
@@ -44,6 +44,7 @@ struct cpuinfo_arm64 {
 	u64		reg_dczid;
 	u64		reg_midr;
 	u64		reg_revidr;
+	u64		reg_aidr;
 	u64		reg_gmid;
 	u64		reg_smidr;
 	u64		reg_mpamidr;
arch/arm64/kernel/cpuinfo.c +3 -0
index 285d7d538342..621218d2a991 100644
@@ -328,11 +328,13 @@ static const struct kobj_type cpuregs_kobj_type = {
 
 CPUREGS_ATTR_RO(midr_el1, midr);
 CPUREGS_ATTR_RO(revidr_el1, revidr);
+CPUREGS_ATTR_RO(aidr_el1, aidr);
 CPUREGS_ATTR_RO(smidr_el1, smidr);
 
 static struct attribute *cpuregs_id_attrs[] = {
 	&cpuregs_attr_midr_el1.attr,
 	&cpuregs_attr_revidr_el1.attr,
+	&cpuregs_attr_aidr_el1.attr,
 	NULL
 };
 
@@ -469,6 +471,7 @@ static void __cpuinfo_store_cpu(struct cpuinfo_arm64 *info)
 	info->reg_dczid = read_cpuid(DCZID_EL0);
 	info->reg_midr = read_cpuid_id();
 	info->reg_revidr = read_cpuid(REVIDR_EL1);
+	info->reg_aidr = read_cpuid(AIDR_EL1);
 
 	info->reg_id_aa64dfr0 = read_cpuid(ID_AA64DFR0_EL1);
 	info->reg_id_aa64dfr1 = read_cpuid(ID_AA64DFR1_EL1);

base-commit: e8b471285262d1561feb2eb266aab6ebe7094124

Signature · 1 line
2.39.5
Reviewed-by on v1 1/1
#1 lore raw
Quoted 44 lines · Oliver Upton

The KVM PV ABI recently added a feature that allows the VM to discover the set of physical CPU implementations, identified by a tuple of {MIDR_EL1, REVIDR_EL1, AIDR_EL1}. Unlike other KVM PV features, the expectation is that the VMM implements the hypercall instead of KVM as it has the authoritative view of where the VM gets scheduled.

To do this the VMM needs to know the values of these registers on any CPU in the system. While MIDR_EL1 and REVIDR_EL1 are already exposed, AIDR_EL1 is not. Provide it in sysfs along with the other identification registers.

---
 Documentation/ABI/testing/sysfs-devices-system-cpu |  1 +
 Documentation/arch/arm64/cpu-feature-registers.rst | 13 +++++++------
 arch/arm64/include/asm/cpu.h                       |  1 +
 arch/arm64/kernel/cpuinfo.c                        |  3 +++
 4 files changed, 12 insertions(+), 6 deletions(-)
Documentation/ABI/testing/sysfs-devices-system-cpu +1 -0
index 206079d3bd5b..9bbf4c27c237 100644
@@ -485,6 +485,7 @@ What:		/sys/devices/system/cpu/cpuX/regs/
 		/sys/devices/system/cpu/cpuX/regs/identification/
 		/sys/devices/system/cpu/cpuX/regs/identification/midr_el1
 		/sys/devices/system/cpu/cpuX/regs/identification/revidr_el1
+		/sys/devices/system/cpu/cpuX/regs/identification/aidr_el1
 		/sys/devices/system/cpu/cpuX/regs/identification/smidr_el1
 Date:		June 2016
 Contact:	Linux ARM Kernel Mailing list <[email protected]>
Documentation/arch/arm64/cpu-feature-registers.rst +1 -4
index 253e9743de2f..add66afc7b03 100644
@@ -72,14 +72,15 @@ there are some issues with their usage.
     process could be migrated to another CPU by the time it uses the
     register value, unless the CPU affinity is set. Hence, there is no
     guarantee that the value reflects the processor that it is
-    currently executing on. The REVIDR is not exposed due to this
-    constraint, as REVIDR makes sense only in conjunction with the
-    MIDR. Alternately, MIDR_EL1 and REVIDR_EL1 are exposed via sysfs
-    at::
+    currently executing on. REVIDR and AIDR are not exposed due to this

Documentation/arch/arm64/cpu-feature-registers.rst there are some issues with their usage. · line 83

Quoted 10 lines
Documentation/arch/arm64/cpu-feature-registers.rst +6 -2
+    constraint, as these registers only make sense in conjunction with
+    the MIDR. Alternately, MIDR_EL1, REVIDR_EL1, and AIDR_EL1 are exposed
+    via sysfs at::
 
 	/sys/devices/system/cpu/cpu$ID/regs/identification/
-	                                              \- midr
-	                                              \- revidr
+	                                              \- midr_el1
+	                                              \- revidr_el1
+	                                              \- aidr_el1

Right, the existing ones are also marked with _el1 suffix as well.

Quoted 42 lines
Documentation/arch/arm64/cpu-feature-registers.rst +0 -0
 3. Implementation
 --------------------
arch/arm64/include/asm/cpu.h +1 -0
index 81e4157f92b7..71493b760b83 100644
@@ -44,6 +44,7 @@ struct cpuinfo_arm64 {
 	u64		reg_dczid;
 	u64		reg_midr;
 	u64		reg_revidr;
+	u64		reg_aidr;
 	u64		reg_gmid;
 	u64		reg_smidr;
 	u64		reg_mpamidr;
arch/arm64/kernel/cpuinfo.c +3 -0
index 285d7d538342..621218d2a991 100644
@@ -328,11 +328,13 @@ static const struct kobj_type cpuregs_kobj_type = {
 
 CPUREGS_ATTR_RO(midr_el1, midr);
 CPUREGS_ATTR_RO(revidr_el1, revidr);
+CPUREGS_ATTR_RO(aidr_el1, aidr);
 CPUREGS_ATTR_RO(smidr_el1, smidr);
 
 static struct attribute *cpuregs_id_attrs[] = {
 	&cpuregs_attr_midr_el1.attr,
 	&cpuregs_attr_revidr_el1.attr,
+	&cpuregs_attr_aidr_el1.attr,
 	NULL
 };
 
@@ -469,6 +471,7 @@ static void __cpuinfo_store_cpu(struct cpuinfo_arm64 *info)
 	info->reg_dczid = read_cpuid(DCZID_EL0);
 	info->reg_midr = read_cpuid_id();
 	info->reg_revidr = read_cpuid(REVIDR_EL1);
+	info->reg_aidr = read_cpuid(AIDR_EL1);
 
 	info->reg_id_aa64dfr0 = read_cpuid(ID_AA64DFR0_EL1);
 	info->reg_id_aa64dfr1 = read_cpuid(ID_AA64DFR1_EL1);
Quoted 1 line

base-commit: e8b471285262d1561feb2eb266aab6ebe7094124

LGTM

Reviewed-by on v1 1/1
#2 lore raw
In reply to Oliver Upton
Quoted 11 lines · Oliver Upton

The KVM PV ABI recently added a feature that allows the VM to discover the set of physical CPU implementations, identified by a tuple of {MIDR_EL1, REVIDR_EL1, AIDR_EL1}. Unlike other KVM PV features, the expectation is that the VMM implements the hypercall instead of KVM as it has the authoritative view of where the VM gets scheduled.

To do this the VMM needs to know the values of these registers on any CPU in the system. While MIDR_EL1 and REVIDR_EL1 are already exposed, AIDR_EL1 is not. Provide it in sysfs along with the other identification registers.

Quoted 7 lines
---
 Documentation/ABI/testing/sysfs-devices-system-cpu |  1 +
 Documentation/arch/arm64/cpu-feature-registers.rst | 13 +++++++------
 arch/arm64/include/asm/cpu.h                       |  1 +
 arch/arm64/kernel/cpuinfo.c                        |  3 +++
 4 files changed, 12 insertions(+), 6 deletions(-)

Indeed, AIDR_EL1 should be available alongside the others.

maintainer Applied
#3 lore raw
In reply to Oliver Upton
Quoted 6 lines · Oliver Upton

The KVM PV ABI recently added a feature that allows the VM to discover the set of physical CPU implementations, identified by a tuple of {MIDR_EL1, REVIDR_EL1, AIDR_EL1}. Unlike other KVM PV features, the expectation is that the VMM implements the hypercall instead of KVM as it has the authoritative view of where the VM gets scheduled.

Quoted 6 lines

To do this the VMM needs to know the values of these registers on any CPU in the system. While MIDR_EL1 and REVIDR_EL1 are already exposed, AIDR_EL1 is not. Provide it in sysfs along with the other identification registers.

[...]

Applied to arm64 (for-next/cpufeature), thanks!

[1/1] arm64: Expose AIDR_EL1 via sysfs

      https://git.kernel.org/arm64/c/17efc1acee62

Cheers,

Signature · 5 lines
Will

https://fixes.arm64.dev
https://next.arm64.dev
https://will.arm64.dev