[PATCH v2 00/11] Add cache information to Amlogic SoC

[PATCH v2 00/11] Add cache information to Amlogic SoC

[Anand Moon]

Most publicly available Amlogic datasheets mention that the CPU employs
a architecture, quad-core ARM Cortex-A53 and ARM Cortex A55 and
Cortex-A73 and Cortex-A53 cluster, sharing a unified L2 cache to enhance
overall system performance.

However, these documents typically omit details regarding the sizes of the
L1 data cache, L1 instruction cache, and L2 cache.

The patches in question align with the cache specifications provided by
ARM TRM for the respective CPU cores.

ARM Cortex-A53
L1: 32KB instruction + 32KB data cache
L2: Unified 512KB cache
L1 cache details, L2 cache details

[1] https://developer.arm.com/documentation/ddi0500/j/Level-1-Memory-System/About-the-L1-memory-system?lang=en
[2] https://developer.arm.com/documentation/ddi0500/j/Level-2-Memory-System/About-the-L2-memory-system?lang=en

ARM Cortex-A55
Cache sizes are implementation-dependent; refer to ARM documentation for configuration options.

[3] https://developer.arm.com/documentation/100442/0200/Functional-description/Introduction-to-the-Cortex-A55-core/Implementation-options

ARM Cortex-A73 (as used in Amlogic S922X and T7)
L1: Configurable, typically 64KB instruction + 64KB data
L2: Unified cache, configurable up to 1MB or more
L2 cache details,
4× Cortex-A73 cores (up to 1.8GHz) with 1MB shared L2 cache
2× Cortex-A53 cores with 256KB shared L2 cache

[4] https://developer.arm.com/documentation/100048/0100/level-1-memory-system/about-the-l1-memory-system?lang=enL2
[5] https://developer.arm.com/documentation/100048/0100/level-2-memory-system/about-the-l2-memory-system?lang=en
[6] https://androidpctv.com/comparative-amlogic-s922x/

Changes:
v2: Modified the commit message and added cache information few more SoC.

v1: https://lists.infradead.org/pipermail/linux-arm-kernel/2024-February/901497.html

Thanks
-Anand

Anand Moon (11):
  arm64: dts: amlogic: Add cache information to the Amlogic GXBB and GXL
    SoC
  arm64: dts: amlogic: Add cache information to the Amlogic SM1 SoC
  arm64: dts: amlogic: Add cache information to the Amlogic G12A SoCS
  arm64: dts: amlogic: Add cache information to the Amlogic AXG SoCS
  arm64: dts: amlogic: Add cache information to the Amlogic GXM SoCS
  arm64: dts: amlogic: Add cache information to the Amlogic A1 SoC
  arm64: dts: amlogic: Add cache information to the Amlogic A4 SoC
  arm64: dts: amlogic: Add cache information to the Amlogic C3 SoC
  arm64: dts: amlogic: Add cache information to the Amlogic S7 SoC
  arm64: dts: amlogic: Add cache information to the Amlogic S922X SoC
  arm64: dts: amlogic: Add cache information to the Amlogic T7 SoC

 arch/arm64/boot/dts/amlogic/amlogic-a4.dtsi | 37 +++++++++++
 arch/arm64/boot/dts/amlogic/amlogic-c3.dtsi | 23 +++++++
 arch/arm64/boot/dts/amlogic/amlogic-s7.dtsi | 36 ++++++++++
 arch/arm64/boot/dts/amlogic/amlogic-t7.dtsi | 74 +++++++++++++++++++++
 arch/arm64/boot/dts/amlogic/meson-a1.dtsi   | 15 +++++
 arch/arm64/boot/dts/amlogic/meson-axg.dtsi  | 21 ++++++
 arch/arm64/boot/dts/amlogic/meson-g12a.dtsi | 27 ++++++++
 arch/arm64/boot/dts/amlogic/meson-g12b.dtsi | 62 +++++++++++++++--
 arch/arm64/boot/dts/amlogic/meson-gx.dtsi   | 27 ++++++++
 arch/arm64/boot/dts/amlogic/meson-gxm.dtsi  | 24 +++++++
 arch/arm64/boot/dts/amlogic/meson-sm1.dtsi  | 27 ++++++++
 11 files changed, 366 insertions(+), 7 deletions(-)


base-commit: 1b237f190eb3d36f52dffe07a40b5eb210280e00
-- 
2.50.1

[PATCH v2 01/11] arm64: dts: amlogic: Add cache information to the Amlogic GXBB and GXL SoC

[Anand Moon]

As per S905 and S905X datasheet add missing cache information to
the Amlogic GXBB and GXL SoC.

- Each Cortex-A53 core has 32KB of L1 instruction cache available and
	32KB of L1 data cache available.
- Along with 512KB Unified L2 cache.

Cache memory significantly reduces the time it takes for the CPU
to access data and instructions, leading to faster program execution
and overall system responsiveness.

Signed-off-by: Anand Moon <linux.amoon@gmail.com>
---
 arch/arm64/boot/dts/amlogic/meson-gx.dtsi | 27 +++++++++++++++++++++++
 1 file changed, 27 insertions(+)

diff --git a/arch/arm64/boot/dts/amlogic/meson-gx.dtsi b/arch/arm64/boot/dts/amlogic/meson-gx.dtsi
index 7d99ca44e660..c1d8e81d95cb 100644
--- a/arch/arm64/boot/dts/amlogic/meson-gx.dtsi
+++ b/arch/arm64/boot/dts/amlogic/meson-gx.dtsi
@@ -95,6 +95,12 @@ cpu0: cpu@0 {
 			compatible = "arm,cortex-a53";
 			reg = <0x0 0x0>;
 			enable-method = "psci";
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
 			next-level-cache = <&l2>;
 			clocks = <&scpi_dvfs 0>;
 			#cooling-cells = <2>;
@@ -105,6 +111,12 @@ cpu1: cpu@1 {
 			compatible = "arm,cortex-a53";
 			reg = <0x0 0x1>;
 			enable-method = "psci";
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
 			next-level-cache = <&l2>;
 			clocks = <&scpi_dvfs 0>;
 			#cooling-cells = <2>;
@@ -115,6 +127,12 @@ cpu2: cpu@2 {
 			compatible = "arm,cortex-a53";
 			reg = <0x0 0x2>;
 			enable-method = "psci";
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
 			next-level-cache = <&l2>;
 			clocks = <&scpi_dvfs 0>;
 			#cooling-cells = <2>;
@@ -125,6 +143,12 @@ cpu3: cpu@3 {
 			compatible = "arm,cortex-a53";
 			reg = <0x0 0x3>;
 			enable-method = "psci";
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
 			next-level-cache = <&l2>;
 			clocks = <&scpi_dvfs 0>;
 			#cooling-cells = <2>;
@@ -134,6 +158,9 @@ l2: l2-cache0 {
 			compatible = "cache";
 			cache-level = <2>;
 			cache-unified;
+			cache-size = <0x80000>; /* L2. 512 KB */
+			cache-line-size = <64>;
+			cache-sets = <512>;
 		};
 	};
 
-- 
2.50.1

[PATCH v2 02/11] arm64: dts: amlogic: Add cache information to the Amlogic SM1 SoC

[Anand Moon]

As per S905X3 datasheet add missing cache information to the Amlogic
SM1 SoC. ARM Cortex-A55 CPU uses unified L3 cache instead of L2 cache.

- Each Cortex-A55 core has 32KB of L1 instruction cache available and
	32KB of L1 data cache available.
- Along with 256KB Unified L2 cache.

Cache memory significantly reduces the time it takes for the CPU
to access data and instructions, leading to faster program execution
and overall system responsiveness.

Signed-off-by: Anand Moon <linux.amoon@gmail.com>
---
 arch/arm64/boot/dts/amlogic/meson-sm1.dtsi | 27 ++++++++++++++++++++++
 1 file changed, 27 insertions(+)

diff --git a/arch/arm64/boot/dts/amlogic/meson-sm1.dtsi b/arch/arm64/boot/dts/amlogic/meson-sm1.dtsi
index 966ebb19cc55..e5db8ce94062 100644
--- a/arch/arm64/boot/dts/amlogic/meson-sm1.dtsi
+++ b/arch/arm64/boot/dts/amlogic/meson-sm1.dtsi
@@ -55,6 +55,12 @@ cpu0: cpu@0 {
 			compatible = "arm,cortex-a55";
 			reg = <0x0 0x0>;
 			enable-method = "psci";
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
 			next-level-cache = <&l2>;
 			#cooling-cells = <2>;
 		};
@@ -64,6 +70,12 @@ cpu1: cpu@1 {
 			compatible = "arm,cortex-a55";
 			reg = <0x0 0x1>;
 			enable-method = "psci";
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
 			next-level-cache = <&l2>;
 			#cooling-cells = <2>;
 		};
@@ -73,6 +85,12 @@ cpu2: cpu@2 {
 			compatible = "arm,cortex-a55";
 			reg = <0x0 0x2>;
 			enable-method = "psci";
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
 			next-level-cache = <&l2>;
 			#cooling-cells = <2>;
 		};
@@ -82,6 +100,12 @@ cpu3: cpu@3 {
 			compatible = "arm,cortex-a55";
 			reg = <0x0 0x3>;
 			enable-method = "psci";
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
 			next-level-cache = <&l2>;
 			#cooling-cells = <2>;
 		};
@@ -90,6 +114,9 @@ l2: l2-cache0 {
 			compatible = "cache";
 			cache-level = <2>;
 			cache-unified;
+			cache-size = <0x40000>; /* L2. 256 KB */
+			cache-line-size = <64>;
+			cache-sets = <256>;
 		};
 	};
 
-- 
2.50.1

[PATCH v2 03/11] arm64: dts: amlogic: Add cache information to the Amlogic G12A SoCS

[Anand Moon]

As per the S905X2 datasheet add missing cache information to the Amlogic
G12A SoC.

- Each Cortex-A53 core has 32KB of L1 instruction cache available and
	32KB of L1 data cache available.
- Along with 512KB Unified L2 cache.

Cache memory significantly reduces the time it takes for the CPU
to access data and instructions, leading to faster program execution
and overall system responsiveness.

Signed-off-by: Anand Moon <linux.amoon@gmail.com>
---
 arch/arm64/boot/dts/amlogic/meson-g12a.dtsi | 27 +++++++++++++++++++++
 1 file changed, 27 insertions(+)

diff --git a/arch/arm64/boot/dts/amlogic/meson-g12a.dtsi b/arch/arm64/boot/dts/amlogic/meson-g12a.dtsi
index deee61dbe074..1321ad95923d 100644
--- a/arch/arm64/boot/dts/amlogic/meson-g12a.dtsi
+++ b/arch/arm64/boot/dts/amlogic/meson-g12a.dtsi
@@ -17,6 +17,12 @@ cpu0: cpu@0 {
 			compatible = "arm,cortex-a53";
 			reg = <0x0 0x0>;
 			enable-method = "psci";
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
 			next-level-cache = <&l2>;
 			#cooling-cells = <2>;
 		};
@@ -26,6 +32,12 @@ cpu1: cpu@1 {
 			compatible = "arm,cortex-a53";
 			reg = <0x0 0x1>;
 			enable-method = "psci";
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
 			next-level-cache = <&l2>;
 			#cooling-cells = <2>;
 		};
@@ -35,6 +47,12 @@ cpu2: cpu@2 {
 			compatible = "arm,cortex-a53";
 			reg = <0x0 0x2>;
 			enable-method = "psci";
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
 			next-level-cache = <&l2>;
 			#cooling-cells = <2>;
 		};
@@ -44,6 +62,12 @@ cpu3: cpu@3 {
 			compatible = "arm,cortex-a53";
 			reg = <0x0 0x3>;
 			enable-method = "psci";
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
 			next-level-cache = <&l2>;
 			#cooling-cells = <2>;
 		};
@@ -52,6 +76,9 @@ l2: l2-cache0 {
 			compatible = "cache";
 			cache-level = <2>;
 			cache-unified;
+			cache-size = <0x80000>; /* L2. 512 KB */
+			cache-line-size = <64>;
+			cache-sets = <512>;
 		};
 	};
 
-- 
2.50.1

[PATCH v2 04/11] arm64: dts: amlogic: Add cache information to the Amlogic AXG SoCS

[Anand Moon]

As per the AXG datasheet add missing cache information to the Amlogic AXG
SoC.

- Each Cortex-A53 core has 32KB of L1 instruction cache available and
	32KB of L1 data cache available.
- Along with 512KB Unified L2 cache.

Cache memory significantly reduces the time it takes for the CPU
to access data and instructions, leading to faster program execution
and overall system responsiveness.

Signed-off-by: Anand Moon <linux.amoon@gmail.com>
---
 arch/arm64/boot/dts/amlogic/meson-axg.dtsi | 21 +++++++++++++++++++++
 1 file changed, 21 insertions(+)

diff --git a/arch/arm64/boot/dts/amlogic/meson-axg.dtsi b/arch/arm64/boot/dts/amlogic/meson-axg.dtsi
index 2df143aa77ce..04fb130ac7c6 100644
--- a/arch/arm64/boot/dts/amlogic/meson-axg.dtsi
+++ b/arch/arm64/boot/dts/amlogic/meson-axg.dtsi
@@ -83,6 +83,12 @@ cpu1: cpu@1 {
 			compatible = "arm,cortex-a53";
 			reg = <0x0 0x1>;
 			enable-method = "psci";
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
 			next-level-cache = <&l2>;
 			clocks = <&scpi_dvfs 0>;
 			dynamic-power-coefficient = <140>;
@@ -94,6 +100,12 @@ cpu2: cpu@2 {
 			compatible = "arm,cortex-a53";
 			reg = <0x0 0x2>;
 			enable-method = "psci";
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
 			next-level-cache = <&l2>;
 			clocks = <&scpi_dvfs 0>;
 			dynamic-power-coefficient = <140>;
@@ -105,6 +117,12 @@ cpu3: cpu@3 {
 			compatible = "arm,cortex-a53";
 			reg = <0x0 0x3>;
 			enable-method = "psci";
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
 			next-level-cache = <&l2>;
 			clocks = <&scpi_dvfs 0>;
 			dynamic-power-coefficient = <140>;
@@ -115,6 +133,9 @@ l2: l2-cache0 {
 			compatible = "cache";
 			cache-level = <2>;
 			cache-unified;
+			cache-size = <0x80000>; /* L2. 512 KB */
+			cache-line-size = <64>;
+			cache-sets = <512>;
 		};
 	};
 
-- 
2.50.1

[PATCH v2 05/11] arm64: dts: amlogic: Add cache information to the Amlogic GXM SoCS

[Anand Moon]

As per the GXM datasheet add missing cache information to the Amlogic GXM
SoC.

- Each Cortex-A53 core has 32KB of L1 instruction cache available and
	32KB of L1 data cache available.
- Along with 512KB Unified L2 cache.

Cache memory significantly reduces the time it takes for the CPU
to access data and instructions, leading to faster program execution
and overall system responsiveness.

Signed-off-by: Anand Moon <linux.amoon@gmail.com>
---
 arch/arm64/boot/dts/amlogic/meson-gxm.dtsi | 24 ++++++++++++++++++++++
 1 file changed, 24 insertions(+)

diff --git a/arch/arm64/boot/dts/amlogic/meson-gxm.dtsi b/arch/arm64/boot/dts/amlogic/meson-gxm.dtsi
index 411cc312fc62..514c9bea6423 100644
--- a/arch/arm64/boot/dts/amlogic/meson-gxm.dtsi
+++ b/arch/arm64/boot/dts/amlogic/meson-gxm.dtsi
@@ -64,6 +64,12 @@ cpu4: cpu@100 {
 			reg = <0x0 0x100>;
 			enable-method = "psci";
 			capacity-dmips-mhz = <1024>;
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
 			next-level-cache = <&l2>;
 			clocks = <&scpi_dvfs 1>;
 			#cooling-cells = <2>;
@@ -75,6 +81,12 @@ cpu5: cpu@101 {
 			reg = <0x0 0x101>;
 			enable-method = "psci";
 			capacity-dmips-mhz = <1024>;
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
 			next-level-cache = <&l2>;
 			clocks = <&scpi_dvfs 1>;
 			#cooling-cells = <2>;
@@ -86,6 +98,12 @@ cpu6: cpu@102 {
 			reg = <0x0 0x102>;
 			enable-method = "psci";
 			capacity-dmips-mhz = <1024>;
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
 			next-level-cache = <&l2>;
 			clocks = <&scpi_dvfs 1>;
 			#cooling-cells = <2>;
@@ -97,6 +115,12 @@ cpu7: cpu@103 {
 			reg = <0x0 0x103>;
 			enable-method = "psci";
 			capacity-dmips-mhz = <1024>;
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
 			next-level-cache = <&l2>;
 			clocks = <&scpi_dvfs 1>;
 			#cooling-cells = <2>;
-- 
2.50.1

[PATCH v2 06/11] arm64: dts: amlogic: Add cache information to the Amlogic A1 SoC

[Anand Moon]

As per the A1 datasheet add missing cache information to the Amlogic A1
SoC.

- Each Cortex-A53 core has 32KB of L1 instruction cache available and
	32KB of L1 data cache available.
- Along with 512KB Unified L2 cache.

Cache memory significantly reduces the time it takes for the CPU
to access data and instructions, leading to faster program execution
and overall system responsiveness.

Signed-off-by: Anand Moon <linux.amoon@gmail.com>
---
 arch/arm64/boot/dts/amlogic/meson-a1.dtsi | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/arch/arm64/boot/dts/amlogic/meson-a1.dtsi b/arch/arm64/boot/dts/amlogic/meson-a1.dtsi
index f7f25a10f409..27b68ed85c4c 100644
--- a/arch/arm64/boot/dts/amlogic/meson-a1.dtsi
+++ b/arch/arm64/boot/dts/amlogic/meson-a1.dtsi
@@ -27,6 +27,12 @@ cpu0: cpu@0 {
 			compatible = "arm,cortex-a35";
 			reg = <0x0 0x0>;
 			enable-method = "psci";
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
 			next-level-cache = <&l2>;
 			#cooling-cells = <2>;
 		};
@@ -36,6 +42,12 @@ cpu1: cpu@1 {
 			compatible = "arm,cortex-a35";
 			reg = <0x0 0x1>;
 			enable-method = "psci";
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
 			next-level-cache = <&l2>;
 			#cooling-cells = <2>;
 		};
@@ -44,6 +56,9 @@ l2: l2-cache0 {
 			compatible = "cache";
 			cache-level = <2>;
 			cache-unified;
+			cache-size = <0x80000>; /* L2. 512 KB */
+			cache-line-size = <64>;
+			cache-sets = <512>;
 		};
 	};
 
-- 
2.50.1

[PATCH v2 07/11] arm64: dts: amlogic: Add cache information to the Amlogic A4 SoC

[Anand Moon]

As per A4 datasheet add missing cache information to the Amlogic A4 SoC.

- Each Cortex-A53 core has 32KB of L1 instruction cache available and
	32KB of L1 data cache available.
- Along with 512KB Unified L2 cache.

Cache memory significantly reduces the time it takes for the CPU
to access data and instructions, leading to faster program execution
and overall system responsiveness.

Signed-off-by: Anand Moon <linux.amoon@gmail.com>
---
 arch/arm64/boot/dts/amlogic/amlogic-a4.dtsi | 37 +++++++++++++++++++++
 1 file changed, 37 insertions(+)

diff --git a/arch/arm64/boot/dts/amlogic/amlogic-a4.dtsi b/arch/arm64/boot/dts/amlogic/amlogic-a4.dtsi
index 563bc2e662fa..fce45933fa28 100644
--- a/arch/arm64/boot/dts/amlogic/amlogic-a4.dtsi
+++ b/arch/arm64/boot/dts/amlogic/amlogic-a4.dtsi
@@ -17,6 +17,13 @@ cpu0: cpu@0 {
 			compatible = "arm,cortex-a53";
 			reg = <0x0 0x0>;
 			enable-method = "psci";
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
+			next-level-cache = <&l2>;
 		};
 
 		cpu1: cpu@1 {
@@ -24,6 +31,13 @@ cpu1: cpu@1 {
 			compatible = "arm,cortex-a53";
 			reg = <0x0 0x1>;
 			enable-method = "psci";
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
+			next-level-cache = <&l2>;
 		};
 
 		cpu2: cpu@2 {
@@ -31,6 +45,13 @@ cpu2: cpu@2 {
 			compatible = "arm,cortex-a53";
 			reg = <0x0 0x2>;
 			enable-method = "psci";
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
+			next-level-cache = <&l2>;
 		};
 
 		cpu3: cpu@3 {
@@ -38,6 +59,22 @@ cpu3: cpu@3 {
 			compatible = "arm,cortex-a53";
 			reg = <0x0 0x3>;
 			enable-method = "psci";
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
+			next-level-cache = <&l2>;
+		};
+
+		l2: l2-cache0 {
+			compatible = "cache";
+			cache-level = <2>;
+			cache-unified;
+			cache-size = <0x80000>; /* L2. 512 KB */
+			cache-line-size = <64>;
+			cache-sets = <512>;
 		};
 	};
 
-- 
2.50.1

[PATCH v2 08/11] arm64: dts: amlogic: Add cache information to the Amlogic C3 SoC

[Anand Moon]

As per C3 datasheet add missing cache information to the Amlogic C3 SoC.

- Each Cortex-A53 core has 32KB of L1 instruction cache available and
	32KB of L1 data cache available.
- Along with 512KB Unified L2 cache.

Cache memory significantly reduces the time it takes for the CPU
to access data and instructions, leading to faster program execution
and overall system responsiveness.

Signed-off-by: Anand Moon <linux.amoon@gmail.com>
---
 arch/arm64/boot/dts/amlogic/amlogic-c3.dtsi | 23 +++++++++++++++++++++
 1 file changed, 23 insertions(+)

diff --git a/arch/arm64/boot/dts/amlogic/amlogic-c3.dtsi b/arch/arm64/boot/dts/amlogic/amlogic-c3.dtsi
index cb9ea3ca6ee0..ee0b1ffa947e 100644
--- a/arch/arm64/boot/dts/amlogic/amlogic-c3.dtsi
+++ b/arch/arm64/boot/dts/amlogic/amlogic-c3.dtsi
@@ -23,6 +23,13 @@ cpu0: cpu@0 {
 			compatible = "arm,cortex-a35";
 			reg = <0x0 0x0>;
 			enable-method = "psci";
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
+			next-level-cache = <&l2>;
 		};
 
 		cpu1: cpu@1 {
@@ -30,6 +37,22 @@ cpu1: cpu@1 {
 			compatible = "arm,cortex-a35";
 			reg = <0x0 0x1>;
 			enable-method = "psci";
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
+			next-level-cache = <&l2>;
+		};
+
+		l2: l2-cache0 {
+			compatible = "cache";
+			cache-level = <2>;
+			cache-unified;
+			cache-size = <0x7d000>; /* L2. 512 KB */
+			cache-line-size = <64>;
+			cache-sets = <512>;
 		};
 	};
 
-- 
2.50.1

[PATCH v2 09/11] arm64: dts: amlogic: Add cache information to the Amlogic S7 SoC

[Anand Moon]

As per S7 datasheet add missing cache information to the Amlogic S7 SoC.

 ARM Cortex-A55 CPU uses unified L2 cache.

- Each Cortex-A55 core has 32KB of L1 instruction cache available and
	32KB of L1 data cache available.
- Along with 256KB Unified L2 cache.

Cache memory significantly reduces the time it takes for the CPU
to access data and instructions, leading to faster program execution
and overall system responsiveness.

Signed-off-by: Anand Moon <linux.amoon@gmail.com>
---
 arch/arm64/boot/dts/amlogic/amlogic-s7.dtsi | 36 +++++++++++++++++++++
 1 file changed, 36 insertions(+)

diff --git a/arch/arm64/boot/dts/amlogic/amlogic-s7.dtsi b/arch/arm64/boot/dts/amlogic/amlogic-s7.dtsi
index 260918b37b9a..d262c0b66e4b 100644
--- a/arch/arm64/boot/dts/amlogic/amlogic-s7.dtsi
+++ b/arch/arm64/boot/dts/amlogic/amlogic-s7.dtsi
@@ -18,6 +18,13 @@ cpu0: cpu@0 {
 			compatible = "arm,cortex-a55";
 			reg = <0x0 0x0>;
 			enable-method = "psci";
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
+			next-level-cache = <&l2>;
 		};
 
 		cpu1: cpu@100 {
@@ -25,6 +32,13 @@ cpu1: cpu@100 {
 			compatible = "arm,cortex-a55";
 			reg = <0x0 0x100>;
 			enable-method = "psci";
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
+			next-level-cache = <&l2>;
 		};
 
 		cpu2: cpu@200 {
@@ -32,6 +46,13 @@ cpu2: cpu@200 {
 			compatible = "arm,cortex-a55";
 			reg = <0x0 0x200>;
 			enable-method = "psci";
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
+			next-level-cache = <&l2>;
 		};
 
 		cpu3: cpu@300 {
@@ -39,8 +60,23 @@ cpu3: cpu@300 {
 			compatible = "arm,cortex-a55";
 			reg = <0x0 0x300>;
 			enable-method = "psci";
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
+			next-level-cache = <&l2>;
 		};
 
+		l2: l2-cache0 {
+			compatible = "cache";
+			cache-level = <2>;
+			cache-unified;
+			cache-size = <0x40000>; /* L2. 256 KB */
+			cache-line-size = <64>;
+			cache-sets = <512>;
+		};
 	};
 
 	timer {
-- 
2.50.1

[PATCH v2 10/11] arm64: dts: amlogic: Add cache information to the Amlogic S922X SoC

[Anand Moon]

As per S922X datasheet add missing cache information to the Amlogic
S922X SoC.

- Each Cortex-A53 core has 32 KB of instruction cache and
	32 KB of L1 data cache available.
- Each Cortex-A73 core has 64 KB of L1 instruction cache and
	64 KB of L1 data cache available.
- The little (A53) cluster has 256 KB of unified L2 cache available.
- The big (A73) cluster has 1 MB of unified L2 cache available.

Cache memory significantly reduces the time it takes for the CPU
to access data and instructions, leading to faster program execution
and overall system responsiveness.

Signed-off-by: Anand Moon <linux.amoon@gmail.com>
---
 arch/arm64/boot/dts/amlogic/meson-g12b.dtsi | 62 ++++++++++++++++++---
 1 file changed, 55 insertions(+), 7 deletions(-)

diff --git a/arch/arm64/boot/dts/amlogic/meson-g12b.dtsi b/arch/arm64/boot/dts/amlogic/meson-g12b.dtsi
index 86e6ceb31d5e..f04efa828256 100644
--- a/arch/arm64/boot/dts/amlogic/meson-g12b.dtsi
+++ b/arch/arm64/boot/dts/amlogic/meson-g12b.dtsi
@@ -49,7 +49,13 @@ cpu0: cpu@0 {
 			reg = <0x0 0x0>;
 			enable-method = "psci";
 			capacity-dmips-mhz = <592>;
-			next-level-cache = <&l2>;
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
+			next-level-cache = <&l2_cache_l>;
 			#cooling-cells = <2>;
 		};
 
@@ -59,7 +65,13 @@ cpu1: cpu@1 {
 			reg = <0x0 0x1>;
 			enable-method = "psci";
 			capacity-dmips-mhz = <592>;
-			next-level-cache = <&l2>;
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
+			next-level-cache = <&l2_cache_l>;
 			#cooling-cells = <2>;
 		};
 
@@ -69,7 +81,13 @@ cpu100: cpu@100 {
 			reg = <0x0 0x100>;
 			enable-method = "psci";
 			capacity-dmips-mhz = <1024>;
-			next-level-cache = <&l2>;
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
+			next-level-cache = <&l2_cache_l>;
 			#cooling-cells = <2>;
 		};
 
@@ -79,7 +97,13 @@ cpu101: cpu@101 {
 			reg = <0x0 0x101>;
 			enable-method = "psci";
 			capacity-dmips-mhz = <1024>;
-			next-level-cache = <&l2>;
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
+			next-level-cache = <&l2_cache_l>;
 			#cooling-cells = <2>;
 		};
 
@@ -89,7 +113,13 @@ cpu102: cpu@102 {
 			reg = <0x0 0x102>;
 			enable-method = "psci";
 			capacity-dmips-mhz = <1024>;
-			next-level-cache = <&l2>;
+			d-cache-line-size = <64>;
+			d-cache-size = <0x10000>;
+			d-cache-sets = <64>;
+			i-cache-line-size = <64>;
+			i-cache-size = <0x10000>;
+			i-cache-sets = <64>;
+			next-level-cache = <&l2_cache_b>;
 			#cooling-cells = <2>;
 		};
 
@@ -99,14 +129,32 @@ cpu103: cpu@103 {
 			reg = <0x0 0x103>;
 			enable-method = "psci";
 			capacity-dmips-mhz = <1024>;
-			next-level-cache = <&l2>;
+			d-cache-line-size = <64>;
+			d-cache-size = <0x10000>;
+			d-cache-sets = <64>;
+			i-cache-line-size = <64>;
+			i-cache-size = <0x10000>;
+			i-cache-sets = <64>;
+			next-level-cache = <&l2_cache_b>;
 			#cooling-cells = <2>;
 		};
 
-		l2: l2-cache0 {
+		l2_cache_l: l2-cache-cluster0 {
 			compatible = "cache";
 			cache-level = <2>;
 			cache-unified;
+			cache-size = <0x40000>;  /* L2. 256 KB */
+			cache-line-size = <64>;
+			cache-sets = <512>;
+		};
+
+		l2_cache_b: l2-cache-cluster1 {
+			compatible = "cache";
+			cache-level = <2>;
+			cache-unified;
+			cache-size = <0x100000>; /* L2. 1MB */
+			cache-line-size = <64>;
+			cache-sets = <512>;
 		};
 	};
 };
-- 
2.50.1

[PATCH v2 11/11] arm64: dts: amlogic: Add cache information to the Amlogic T7 SoC

[Anand Moon]

As per T7 datasheet add missing cache information to the Amlogic T7 SoC.

- Each Cortex-A53 core has 32 KB of instruction cache and
	32 KB of L1 data cache available.
- Each Cortex-A73 core has 64 KB of L1 instruction cache and
	64 KB of L1 data cache available.
- The little (A53) cluster has 256 KB of unified L2 cache available.
- The big (A73) cluster has 1 MB of unified L2 cache available.

Cache memory significantly reduces the time it takes for the CPU
to access data and instructions, leading to faster program execution
and overall system responsiveness.

Signed-off-by: Anand Moon <linux.amoon@gmail.com>
---
 arch/arm64/boot/dts/amlogic/amlogic-t7.dtsi | 74 +++++++++++++++++++++
 1 file changed, 74 insertions(+)

diff --git a/arch/arm64/boot/dts/amlogic/amlogic-t7.dtsi b/arch/arm64/boot/dts/amlogic/amlogic-t7.dtsi
index ec743cad57db..6510068bcff9 100644
--- a/arch/arm64/boot/dts/amlogic/amlogic-t7.dtsi
+++ b/arch/arm64/boot/dts/amlogic/amlogic-t7.dtsi
@@ -53,6 +53,13 @@ cpu100: cpu@100 {
 			compatible = "arm,cortex-a53";
 			reg = <0x0 0x100>;
 			enable-method = "psci";
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
+			next-level-cache = <&l2_cache_l>;
 		};
 
 		cpu101: cpu@101 {
@@ -60,6 +67,13 @@ cpu101: cpu@101 {
 			compatible = "arm,cortex-a53";
 			reg = <0x0 0x101>;
 			enable-method = "psci";
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
+			next-level-cache = <&l2_cache_l>;
 		};
 
 		cpu102: cpu@102 {
@@ -67,6 +81,13 @@ cpu102: cpu@102 {
 			compatible = "arm,cortex-a53";
 			reg = <0x0 0x102>;
 			enable-method = "psci";
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
+			next-level-cache = <&l2_cache_l>;
 		};
 
 		cpu103: cpu@103 {
@@ -74,6 +95,13 @@ cpu103: cpu@103 {
 			compatible = "arm,cortex-a53";
 			reg = <0x0 0x103>;
 			enable-method = "psci";
+			d-cache-line-size = <32>;
+			d-cache-size = <0x8000>;
+			d-cache-sets = <32>;
+			i-cache-line-size = <32>;
+			i-cache-size = <0x8000>;
+			i-cache-sets = <32>;
+			next-level-cache = <&l2_cache_l>;
 		};
 
 		cpu0: cpu@0 {
@@ -81,6 +109,13 @@ cpu0: cpu@0 {
 			compatible = "arm,cortex-a73";
 			reg = <0x0 0x0>;
 			enable-method = "psci";
+			d-cache-line-size = <64>;
+			d-cache-size = <0x10000>;
+			d-cache-sets = <64>;
+			i-cache-line-size = <64>;
+			i-cache-size = <0x10000>;
+			i-cache-sets = <64>;
+			next-level-cache = <&l2_cache_b>;
 		};
 
 		cpu1: cpu@1 {
@@ -88,6 +123,13 @@ cpu1: cpu@1 {
 			compatible = "arm,cortex-a73";
 			reg = <0x0 0x1>;
 			enable-method = "psci";
+			d-cache-line-size = <64>;
+			d-cache-size = <0x10000>;
+			d-cache-sets = <64>;
+			i-cache-line-size = <64>;
+			i-cache-size = <0x10000>;
+			i-cache-sets = <64>;
+			next-level-cache = <&l2_cache_b>;
 		};
 
 		cpu2: cpu@2 {
@@ -95,6 +137,13 @@ cpu2: cpu@2 {
 			compatible = "arm,cortex-a73";
 			reg = <0x0 0x2>;
 			enable-method = "psci";
+			d-cache-line-size = <64>;
+			d-cache-size = <0x10000>;
+			d-cache-sets = <64>;
+			i-cache-line-size = <64>;
+			i-cache-size = <0x10000>;
+			i-cache-sets = <64>;
+			next-level-cache = <&l2_cache_b>;
 		};
 
 		cpu3: cpu@3 {
@@ -102,6 +151,31 @@ cpu3: cpu@3 {
 			compatible = "arm,cortex-a73";
 			reg = <0x0 0x3>;
 			enable-method = "psci";
+			d-cache-line-size = <64>;
+			d-cache-size = <0x10000>;
+			d-cache-sets = <64>;
+			i-cache-line-size = <64>;
+			i-cache-size = <0x10000>;
+			i-cache-sets = <64>;
+			next-level-cache = <&l2_cache_b>;
+		};
+
+		l2_cache_l: l2-cache-cluster0 {
+			compatible = "cache";
+			cache-level = <2>;
+			cache-unified;
+			cache-size = <0x40000>;  /* L2. 256 KB */
+			cache-line-size = <64>;
+			cache-sets = <512>;
+		};
+
+		l2_cache_b: l2-cache-cluster1 {
+			compatible = "cache";
+			cache-level = <2>;
+			cache-unified;
+			cache-size = <0x100000>; /* L2. 1 Mb */
+			cache-line-size = <64>;
+			cache-sets = <512>;
 		};
 	};
 
-- 
2.50.1

Re: [PATCH v2 01/11] arm64: dts: amlogic: Add cache information to the Amlogic GXBB and GXL SoC

[Christian Hewitt]

> On 25 Aug 2025, at 10:51 am, Anand Moon <linux.amoon@gmail.com> wrote:
> 
> As per S905 and S905X datasheet add missing cache information to
> the Amlogic GXBB and GXL SoC.
> 
> - Each Cortex-A53 core has 32KB of L1 instruction cache available and
> 32KB of L1 data cache available.
> - Along with 512KB Unified L2 cache.
> 
> Cache memory significantly reduces the time it takes for the CPU
> to access data and instructions, leading to faster program execution
> and overall system responsiveness.

Hello Anand,

I’m wondering if we are “enabling caching” in these patches (could be
a significant gain, as per text) or we are “optimising caching” meaning
the kernel currently assumes generic/safe defaults so having accurate
descriptions in dt allows better efficiency (marginal gain)?

Stats are also subjective to the workload used, but do you have any
kind of before/after benchmarks? (for any of the SoCs in the patchset)

Christian

> Signed-off-by: Anand Moon <linux.amoon@gmail.com>
> ---
> arch/arm64/boot/dts/amlogic/meson-gx.dtsi | 27 +++++++++++++++++++++++
> 1 file changed, 27 insertions(+)
> 
> diff --git a/arch/arm64/boot/dts/amlogic/meson-gx.dtsi b/arch/arm64/boot/dts/amlogic/meson-gx.dtsi
> index 7d99ca44e660..c1d8e81d95cb 100644
> --- a/arch/arm64/boot/dts/amlogic/meson-gx.dtsi
> +++ b/arch/arm64/boot/dts/amlogic/meson-gx.dtsi
> @@ -95,6 +95,12 @@ cpu0: cpu@0 {
> compatible = "arm,cortex-a53";
> reg = <0x0 0x0>;
> enable-method = "psci";
> + d-cache-line-size = <32>;
> + d-cache-size = <0x8000>;
> + d-cache-sets = <32>;
> + i-cache-line-size = <32>;
> + i-cache-size = <0x8000>;
> + i-cache-sets = <32>;
> next-level-cache = <&l2>;
> clocks = <&scpi_dvfs 0>;
> #cooling-cells = <2>;
> @@ -105,6 +111,12 @@ cpu1: cpu@1 {
> compatible = "arm,cortex-a53";
> reg = <0x0 0x1>;
> enable-method = "psci";
> + d-cache-line-size = <32>;
> + d-cache-size = <0x8000>;
> + d-cache-sets = <32>;
> + i-cache-line-size = <32>;
> + i-cache-size = <0x8000>;
> + i-cache-sets = <32>;
> next-level-cache = <&l2>;
> clocks = <&scpi_dvfs 0>;
> #cooling-cells = <2>;
> @@ -115,6 +127,12 @@ cpu2: cpu@2 {
> compatible = "arm,cortex-a53";
> reg = <0x0 0x2>;
> enable-method = "psci";
> + d-cache-line-size = <32>;
> + d-cache-size = <0x8000>;
> + d-cache-sets = <32>;
> + i-cache-line-size = <32>;
> + i-cache-size = <0x8000>;
> + i-cache-sets = <32>;
> next-level-cache = <&l2>;
> clocks = <&scpi_dvfs 0>;
> #cooling-cells = <2>;
> @@ -125,6 +143,12 @@ cpu3: cpu@3 {
> compatible = "arm,cortex-a53";
> reg = <0x0 0x3>;
> enable-method = "psci";
> + d-cache-line-size = <32>;
> + d-cache-size = <0x8000>;
> + d-cache-sets = <32>;
> + i-cache-line-size = <32>;
> + i-cache-size = <0x8000>;
> + i-cache-sets = <32>;
> next-level-cache = <&l2>;
> clocks = <&scpi_dvfs 0>;
> #cooling-cells = <2>;
> @@ -134,6 +158,9 @@ l2: l2-cache0 {
> compatible = "cache";
> cache-level = <2>;
> cache-unified;
> + cache-size = <0x80000>; /* L2. 512 KB */
> + cache-line-size = <64>;
> + cache-sets = <512>;
> };
> };
> 
> -- 
> 2.50.1
> 
> 
> _______________________________________________
> linux-amlogic mailing list
> linux-amlogic@lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-amlogic

Re: [PATCH v2 01/11] arm64: dts: amlogic: Add cache information to the Amlogic GXBB and GXL SoC

[Anand Moon]

Hi Christian,

On Mon, 25 Aug 2025 at 13:29, Christian Hewitt
<christian@hewittfamily.org.uk> wrote:
>
> > On 25 Aug 2025, at 10:51 am, Anand Moon <linux.amoon@gmail.com> wrote:
> >
> > As per S905 and S905X datasheet add missing cache information to
> > the Amlogic GXBB and GXL SoC.
> >
> > - Each Cortex-A53 core has 32KB of L1 instruction cache available and
> > 32KB of L1 data cache available.
> > - Along with 512KB Unified L2 cache.
> >
> > Cache memory significantly reduces the time it takes for the CPU
> > to access data and instructions, leading to faster program execution
> > and overall system responsiveness.
>
> Hello Anand,
>
> I’m wondering if we are “enabling caching” in these patches (could be
> a significant gain, as per text) or we are “optimising caching” meaning
> the kernel currently assumes generic/safe defaults so having accurate
> descriptions in dt allows better efficiency (marginal gain)?
>
> Stats are also subjective to the workload used, but do you have any
> kind of before/after benchmarks? (for any of the SoCs in the patchset)
>

This is a fundamental feature of Arm64 CPUs that tracks active instructions
and data within cache-mapped memory pages.
Enabling it can significantly enhance overall system performance.

We can configure more l2 cache memory which is confribable as per the
Arm TRM document.
Arm Cortex - A53  - Configurable L2 cache size of 128KB, 256KB, 512KB,
1MB and 2MB.
Arm Cortex - A55  - Configurable L2 cache size of 64KB, 128KB, or 256KB
Arm Cortex - A73 -  Configurable L2 cache size of 256KB, 512KB, 1MB,
2MB, 4MB, or 8MB.

Here's an article that provides detailed insights into the cache feature.
[0] http://jake.dothome.co.kr/cache4/

I tested with a small benchmark to test factorial.

Before:>
alarm@archl-librecm:~$ sudo perf stat -e cache-references,cache-misses ./test
Simulated Cache Miss Time (avg): 589 ns
Factorial(10) = 3628800

 Performance counter stats for './test':

           3017286      cache-references
             45414      cache-misses                     #    1.51% of
all cache refs

       0.054512394 seconds time elapsed

       0.004209000 seconds user
       0.041866000 seconds sys

After:>
 # sudo perf stat -e cache-references,cache-misses ./test
Simulated Cache Miss Time (avg): 426 ns
Factorial(10) = 3628800

 Performance counter stats for './test':

           2814633      cache-references
             27054      cache-misses                     #    0.96% of
all cache refs

       0.041041585 seconds time elapsed

       0.007976000 seconds user
       0.032009000 seconds sys

> Christian

Thanks
-Anand