arm-trusted-firmware/fdts/tc-base.dtsi
Boyan Karatotev 2590e819eb perf(mpmm): greatly simplify MPMM enablement
MPMM is a core-specific microarchitectural feature. It has been present
in every Arm core since the Cortex-A510 and has been implemented in
exactly the same way. Despite that, it is enabled more like an
architectural feature with a top level enable flag. This utilised the
identical implementation.

This duality has left MPMM in an awkward place, where its enablement
should be generic, like an architectural feature, but since it is not,
it should also be core-specific if it ever changes. One choice to do
this has been through the device tree.

This has worked just fine so far, however, recent implementations expose
a weakness in that this is rather slow - the device tree has to be read,
there's a long call stack of functions with many branches, and system
registers are read. In the hot path of PSCI CPU powerdown, this has a
significant and measurable impact. Besides it being a rather large
amount of code that is difficult to understand.

Since MPMM is a microarchitectural feature, its correct placement is in
the reset function. The essence of the current enablement is to write
CPUPPMCR_EL3.MPMM_EN if CPUPPMCR_EL3.MPMMPINCTL == 0. Replacing the C
enablement with an assembly macro in each CPU's reset function achieves
the same effect with just a single close branch and a grand total of 6
instructions (versus the old 2 branches and 32 instructions).

Having done this, the device tree entry becomes redundant. Should a core
that doesn't support MPMM arise, this can cleanly be handled in the
reset function. As such, the whole ENABLE_MPMM_FCONF and platform hooks
mechanisms become obsolete and are removed.

Change-Id: I1d0475b21a1625bb3519f513ba109284f973ffdf
Signed-off-by: Boyan Karatotev <boyan.karatotev@arm.com>
2025-02-25 08:50:45 +00:00

676 lines
14 KiB
Text

/*
* Copyright (c) 2020-2024, Arm Limited. All rights reserved.
*
* SPDX-License-Identifier: BSD-3-Clause
*/
/* If SCMI power domain control is enabled */
#if TC_SCMI_PD_CTRL_EN
#define GPU_SCMI_PD_IDX (PLAT_MAX_CPUS_PER_CLUSTER + 1)
#define DPU_SCMI_PD_IDX (PLAT_MAX_CPUS_PER_CLUSTER + 2)
#endif /* TC_SCMI_PD_CTRL_EN */
/* Use SCMI controlled clocks */
#if TC_DPU_USE_SCMI_CLK
#define DPU_CLK_ATTR1 \
clocks = <&scmi_clk 0>; \
clock-names = "aclk"
#define DPU_CLK_ATTR2 \
clocks = <&scmi_clk 1>; \
clock-names = "pxclk"
#define DPU_CLK_ATTR3 \
clocks = <&scmi_clk 2>; \
clock-names = "pxclk" \
/* Use fixed clocks */
#else /* !TC_DPU_USE_SCMI_CLK */
#define DPU_CLK_ATTR1 \
clocks = <&dpu_aclk>; \
clock-names = "aclk"
#define DPU_CLK_ATTR2 \
clocks = <&dpu_pixel_clk>, <&dpu_aclk>; \
clock-names = "pxclk", "aclk"
#define DPU_CLK_ATTR3 DPU_CLK_ATTR2
#endif /* !TC_DPU_USE_SCMI_CLK */
/ {
compatible = "arm,tc";
interrupt-parent = <&gic>;
#address-cells = <2>;
#size-cells = <2>;
aliases {
serial0 = &os_uart;
};
cpus {
#address-cells = <1>;
#size-cells = <0>;
cpu-map {
cluster0 {
core0 {
cpu = <&CPU0>;
};
core1 {
cpu = <&CPU1>;
};
core2 {
cpu = <&CPU2>;
};
core3 {
cpu = <&CPU3>;
};
core4 {
cpu = <&CPU4>;
};
core5 {
cpu = <&CPU5>;
};
core6 {
cpu = <&CPU6>;
};
core7 {
cpu = <&CPU7>;
};
};
};
/*
* The timings below are just to demonstrate working cpuidle.
* These values may be inaccurate.
*/
idle-states {
entry-method = "psci";
CPU_SLEEP_0: cpu-sleep-0 {
compatible = "arm,idle-state";
arm,psci-suspend-param = <0x0010000>;
local-timer-stop;
entry-latency-us = <300>;
exit-latency-us = <1200>;
min-residency-us = <2000>;
};
CLUSTER_SLEEP_0: cluster-sleep-0 {
compatible = "arm,idle-state";
arm,psci-suspend-param = <0x1010000>;
local-timer-stop;
entry-latency-us = <400>;
exit-latency-us = <1200>;
min-residency-us = <2500>;
};
};
amus {
amu: amu-0 {
#address-cells = <1>;
#size-cells = <0>;
mpmm_gear0: counter@0 {
reg = <0>;
enable-at-el3;
};
mpmm_gear1: counter@1 {
reg = <1>;
enable-at-el3;
};
mpmm_gear2: counter@2 {
reg = <2>;
enable-at-el3;
};
};
};
CPU0:cpu@0 {
device_type = "cpu";
compatible = "arm,armv8";
reg = <0x0>;
enable-method = "psci";
clocks = <&scmi_dvfs 0>;
cpu-idle-states = <&CPU_SLEEP_0 &CLUSTER_SLEEP_0>;
capacity-dmips-mhz = <LIT_CAPACITY>;
amu = <&amu>;
};
CPU1:cpu@100 {
device_type = "cpu";
compatible = "arm,armv8";
reg = <0x100>;
enable-method = "psci";
clocks = <&scmi_dvfs 0>;
cpu-idle-states = <&CPU_SLEEP_0 &CLUSTER_SLEEP_0>;
capacity-dmips-mhz = <LIT_CAPACITY>;
amu = <&amu>;
};
CPU2:cpu@200 {
device_type = "cpu";
compatible = "arm,armv8";
reg = <0x200>;
enable-method = "psci";
cpu-idle-states = <&CPU_SLEEP_0 &CLUSTER_SLEEP_0>;
amu = <&amu>;
};
CPU3:cpu@300 {
device_type = "cpu";
compatible = "arm,armv8";
reg = <0x300>;
enable-method = "psci";
cpu-idle-states = <&CPU_SLEEP_0 &CLUSTER_SLEEP_0>;
amu = <&amu>;
};
CPU4:cpu@400 {
device_type = "cpu";
compatible = "arm,armv8";
reg = <0x400>;
enable-method = "psci";
clocks = <&scmi_dvfs 1>;
cpu-idle-states = <&CPU_SLEEP_0 &CLUSTER_SLEEP_0>;
capacity-dmips-mhz = <MID_CAPACITY>;
amu = <&amu>;
};
CPU5:cpu@500 {
device_type = "cpu";
compatible = "arm,armv8";
reg = <0x500>;
enable-method = "psci";
clocks = <&scmi_dvfs 1>;
cpu-idle-states = <&CPU_SLEEP_0 &CLUSTER_SLEEP_0>;
capacity-dmips-mhz = <MID_CAPACITY>;
amu = <&amu>;
};
CPU6:cpu@600 {
device_type = "cpu";
compatible = "arm,armv8";
reg = <0x600>;
enable-method = "psci";
cpu-idle-states = <&CPU_SLEEP_0 &CLUSTER_SLEEP_0>;
amu = <&amu>;
};
CPU7:cpu@700 {
device_type = "cpu";
compatible = "arm,armv8";
reg = <0x700>;
enable-method = "psci";
cpu-idle-states = <&CPU_SLEEP_0 &CLUSTER_SLEEP_0>;
amu = <&amu>;
};
};
reserved-memory {
#address-cells = <2>;
#size-cells = <2>;
ranges;
linux,cma {
compatible = "shared-dma-pool";
reusable;
size = <0x0 0x8000000>;
linux,cma-default;
};
optee {
compatible = "restricted-dma-pool";
reg = <0x0 TC_NS_OPTEE_BASE 0x0 TC_NS_OPTEE_SIZE>;
};
};
memory {
device_type = "memory";
reg = <0x0 TC_NS_DRAM1_BASE 0x0 TC_NS_DRAM1_SIZE>,
<HI(PLAT_ARM_DRAM2_BASE) LO(PLAT_ARM_DRAM2_BASE)
HI(TC_NS_DRAM2_SIZE) LO(TC_NS_DRAM2_SIZE)>;
};
psci {
compatible = "arm,psci-1.0", "arm,psci-0.2";
method = "smc";
};
cpu-pmu-little {
compatible = LIT_CPU_PMU_COMPATIBLE;
interrupts = <GIC_PPI 7 IRQ_TYPE_LEVEL_HIGH &ppi_partition_little>;
status = "okay";
};
cpu-pmu-mid {
compatible = MID_CPU_PMU_COMPATIBLE;
interrupts = <GIC_PPI 7 IRQ_TYPE_LEVEL_HIGH &ppi_partition_mid>;
status = "okay";
};
cpu-pmu-big {
compatible = BIG_CPU_PMU_COMPATIBLE;
interrupts = <GIC_PPI 7 IRQ_TYPE_LEVEL_HIGH &ppi_partition_big>;
status = "okay";
};
sram: sram@6000000 {
compatible = "mmio-sram";
reg = <0x0 PLAT_ARM_NSRAM_BASE 0x0 PLAT_ARM_NSRAM_SIZE>;
#address-cells = <1>;
#size-cells = <1>;
ranges = <0 0x0 PLAT_ARM_NSRAM_BASE PLAT_ARM_NSRAM_SIZE>;
cpu_scp_scmi_a2p: scp-shmem@0 {
compatible = "arm,scmi-shmem";
reg = <0x0 0x80>;
};
};
mbox_db_rx: mhu@MHU_RX_ADDR {
compatible = MHU_RX_COMPAT;
reg = <0x0 ADDRESSIFY(MHU_RX_ADDR) 0x0 MHU_OFFSET>;
clocks = <&soc_refclk>;
clock-names = "apb_pclk";
#mbox-cells = <MHU_MBOX_CELLS>;
interrupts = <GIC_SPI MHU_RX_INT_NUM IRQ_TYPE_LEVEL_HIGH 0>;
interrupt-names = MHU_RX_INT_NAME;
};
mbox_db_tx: mhu@MHU_TX_ADDR {
compatible = MHU_TX_COMPAT;
reg = <0x0 ADDRESSIFY(MHU_TX_ADDR) 0x0 MHU_OFFSET>;
clocks = <&soc_refclk>;
clock-names = "apb_pclk";
#mbox-cells = <MHU_MBOX_CELLS>;
interrupt-names = MHU_TX_INT_NAME;
};
firmware {
scmi {
compatible = "arm,scmi";
mbox-names = "tx", "rx";
#address-cells = <1>;
#size-cells = <0>;
#if TC_SCMI_PD_CTRL_EN
scmi_devpd: protocol@11 {
reg = <0x11>;
#power-domain-cells = <1>;
};
#endif /* TC_SCMI_PD_CTRL_EN */
scmi_dvfs: protocol@13 {
reg = <0x13>;
#clock-cells = <1>;
};
scmi_clk: protocol@14 {
reg = <0x14>;
#clock-cells = <1>;
};
};
};
gic: interrupt-controller@GIC_CTRL_ADDR {
compatible = "arm,gic-v3";
#address-cells = <2>;
#interrupt-cells = <4>;
#size-cells = <2>;
ranges;
interrupt-controller;
reg = <0x0 0x30000000 0 0x10000>, /* GICD */
<0x0 0x30080000 0 GIC_GICR_OFFSET>; /* GICR */
interrupts = <GIC_PPI 0x9 IRQ_TYPE_LEVEL_LOW 0>;
};
timer {
compatible = "arm,armv8-timer";
interrupts = <GIC_PPI 13 IRQ_TYPE_LEVEL_LOW 0>,
<GIC_PPI 14 IRQ_TYPE_LEVEL_LOW 0>,
<GIC_PPI 11 IRQ_TYPE_LEVEL_LOW 0>,
<GIC_PPI 10 IRQ_TYPE_LEVEL_LOW 0>;
};
spe-pmu-mid {
compatible = "arm,statistical-profiling-extension-v1";
interrupts = <GIC_PPI 1 IRQ_TYPE_LEVEL_HIGH &ppi_partition_mid>;
status = "disabled";
};
spe-pmu-big {
compatible = "arm,statistical-profiling-extension-v1";
interrupts = <GIC_PPI 1 IRQ_TYPE_LEVEL_HIGH &ppi_partition_big>;
status = "disabled";
};
soc_refclk: refclk {
compatible = "fixed-clock";
#clock-cells = <0>;
clock-frequency = <1000000000>;
clock-output-names = "apb_pclk";
};
soc_refclk60mhz: refclk60mhz {
compatible = "fixed-clock";
#clock-cells = <0>;
clock-frequency = <60000000>;
clock-output-names = "iofpga_clk";
};
soc_uartclk: uartclk {
compatible = "fixed-clock";
#clock-cells = <0>;
clock-frequency = <TC_UARTCLK>;
clock-output-names = "uartclk";
};
/* soc_uart0 on FPGA, ap_ns_uart on FVP */
os_uart: serial@2a400000 {
compatible = "arm,pl011", "arm,primecell";
reg = <0x0 0x2A400000 0x0 UART_OFFSET>;
interrupts = <GIC_SPI 63 IRQ_TYPE_LEVEL_HIGH 0>;
clocks = <&soc_uartclk>, <&soc_refclk>;
clock-names = "uartclk", "apb_pclk";
status = "okay";
};
#if !TC_DPU_USE_SCMI_CLK
dpu_aclk: dpu_aclk {
compatible = "fixed-clock";
#clock-cells = <0>;
clock-frequency = <LCD_TIMING_CLK>;
clock-output-names = "fpga:dpu_aclk";
};
dpu_pixel_clk: dpu-pixel-clk {
compatible = "fixed-clock";
#clock-cells = <0>;
clock-frequency = <LCD_TIMING_CLK>;
clock-output-names = "pxclk";
};
#endif /* !TC_DPU_USE_SCMI_CLK */
#if TC_DPU_USE_SIMPLE_PANEL
vpanel {
compatible = "panel-dpi";
post-init-providers = <&pl0>;
port {
lcd_in: endpoint {
remote-endpoint = <&dp_pl0_out0>;
};
};
panel-timing {
LCD_TIMING;
};
};
#else
vencoder {
compatible = "drm,virtual-encoder";
port {
lcd_in: endpoint {
remote-endpoint = <&dp_pl0_out0>;
};
};
display-timings {
timing-panel {
LCD_TIMING;
};
};
};
#endif
ethernet: ethernet@ETHERNET_ADDR {
reg = <0x0 ADDRESSIFY(ETHERNET_ADDR) 0x0 0x10000>;
interrupts = <GIC_SPI ETHERNET_INT IRQ_TYPE_LEVEL_HIGH 0>;
reg-io-width = <2>;
smsc,irq-push-pull;
};
bp_clock24mhz: clock24mhz {
compatible = "fixed-clock";
#clock-cells = <0>;
clock-frequency = <24000000>;
clock-output-names = "bp:clock24mhz";
};
sysreg: sysreg@SYS_REGS_ADDR {
compatible = "arm,vexpress-sysreg";
reg = <0x0 ADDRESSIFY(SYS_REGS_ADDR) 0x0 0x1000>;
gpio-controller;
#gpio-cells = <2>;
};
fixed_3v3: v2m-3v3 {
compatible = "regulator-fixed";
regulator-name = "3V3";
regulator-min-microvolt = <3300000>;
regulator-max-microvolt = <3300000>;
regulator-always-on;
};
mmci: mmci@MMC_ADDR {
compatible = "arm,pl180", "arm,primecell";
reg = <0x0 ADDRESSIFY(MMC_ADDR) 0x0 0x1000>;
interrupts = <GIC_SPI MMC_INT_0 IRQ_TYPE_LEVEL_HIGH 0>,
<GIC_SPI MMC_INT_1 IRQ_TYPE_LEVEL_HIGH 0>;
wp-gpios = <&sysreg 1 0>;
bus-width = <4>;
max-frequency = <25000000>;
vmmc-supply = <&fixed_3v3>;
clocks = <&bp_clock24mhz>, <&bp_clock24mhz>;
clock-names = "mclk", "apb_pclk";
};
gpu_clk: gpu_clk {
compatible = "fixed-clock";
#clock-cells = <0>;
clock-frequency = <1000000000>;
};
gpu_core_clk: gpu_core_clk {
compatible = "fixed-clock";
#clock-cells = <0>;
clock-frequency = <1000000000>;
};
gpu: gpu@2d000000 {
compatible = "arm,mali-midgard";
reg = <0x0 0x2d000000 0x0 0x200000>;
clocks = <&gpu_core_clk>;
clock-names = "shadercores";
#if TC_SCMI_PD_CTRL_EN
power-domains = <&scmi_devpd GPU_SCMI_PD_IDX>;
scmi-perf-domain = <3>;
#endif /* TC_SCMI_PD_CTRL_EN */
pbha {
int-id-override = <0 0x22>, <2 0x23>, <4 0x23>, <7 0x22>,
<8 0x22>, <9 0x22>, <10 0x22>, <11 0x22>,
<12 0x22>, <13 0x22>, <16 0x22>, <17 0x32>,
<18 0x32>, <19 0x22>, <20 0x22>, <21 0x32>,
<22 0x32>, <24 0x22>, <28 0x32>;
propagate-bits = <0x0f>;
};
};
power_model_simple {
/*
* Numbers used are irrelevant to Titan,
* it helps suppressing the kernel warnings.
*/
compatible = "arm,mali-simple-power-model";
static-coefficient = <2427750>;
dynamic-coefficient = <4687>;
ts = <20000 2000 (-20) 2>;
thermal-zone = "";
};
smmu_600: smmu@2ce00000 {
compatible = "arm,smmu-v3";
reg = <0 0x2ce00000 0 0x20000>;
interrupts = <GIC_SPI 75 IRQ_TYPE_EDGE_RISING 0>,
<GIC_SPI 74 IRQ_TYPE_EDGE_RISING 0>,
<GIC_SPI 76 IRQ_TYPE_EDGE_RISING 0>,
<GIC_SPI 77 IRQ_TYPE_EDGE_RISING 0>;
interrupt-names = "eventq", "priq", "cmdq-sync", "gerror";
#iommu-cells = <1>;
status = "disabled";
};
smmu_700: iommu@3f000000 {
#iommu-cells = <1>;
compatible = "arm,smmu-v3";
reg = <0x0 0x3f000000 0x0 0x5000000>;
interrupts = <GIC_SPI 228 IRQ_TYPE_EDGE_RISING 0>,
<GIC_SPI 229 IRQ_TYPE_EDGE_RISING 0>,
<GIC_SPI 230 IRQ_TYPE_EDGE_RISING 0>;
interrupt-names = "eventq", "cmdq-sync", "gerror";
dma-coherent;
status = "disabled";
};
smmu_700_dpu: iommu@4002a00000 {
#iommu-cells = <1>;
compatible = "arm,smmu-v3";
reg = <HI(0x4002a00000) LO(0x4002a00000) 0x0 0x5000000>;
interrupts = <GIC_SPI 481 IRQ_TYPE_EDGE_RISING 0>,
<GIC_SPI 482 IRQ_TYPE_EDGE_RISING 0>,
<GIC_SPI 483 IRQ_TYPE_EDGE_RISING 0>;
interrupt-names = "eventq", "cmdq-sync", "gerror";
dma-coherent;
status = "disabled";
};
dp0: display@DPU_ADDR {
#address-cells = <1>;
#size-cells = <0>;
compatible = "arm,mali-d71";
reg = <HI(ADDRESSIFY(DPU_ADDR)) LO(ADDRESSIFY(DPU_ADDR)) 0 0x20000>;
interrupts = <GIC_SPI DPU_IRQ IRQ_TYPE_LEVEL_HIGH 0>;
interrupt-names = "DPU";
DPU_CLK_ATTR1;
pl0: pipeline@0 {
reg = <0>;
DPU_CLK_ATTR2;
pl_id = <0>;
ports {
#address-cells = <1>;
#size-cells = <0>;
port@0 {
reg = <0>;
dp_pl0_out0: endpoint {
remote-endpoint = <&lcd_in>;
};
};
};
};
pl1: pipeline@1 {
reg = <1>;
DPU_CLK_ATTR3;
pl_id = <1>;
ports {
#address-cells = <1>;
#size-cells = <0>;
port@0 {
reg = <0>;
};
};
};
};
/*
* L3 cache in the DSU is the Memory System Component (MSC)
* The MPAM registers are accessed through utility bus in the DSU
*/
dsu-msc0 {
compatible = "arm,mpam-msc";
reg = <DSU_MPAM_ADDR 0x0 0x2000>;
};
ete0 {
compatible = "arm,embedded-trace-extension";
cpu = <&CPU0>;
};
ete1 {
compatible = "arm,embedded-trace-extension";
cpu = <&CPU1>;
};
ete2 {
compatible = "arm,embedded-trace-extension";
cpu = <&CPU2>;
};
ete3 {
compatible = "arm,embedded-trace-extension";
cpu = <&CPU3>;
};
ete4 {
compatible = "arm,embedded-trace-extension";
cpu = <&CPU4>;
};
ete5 {
compatible = "arm,embedded-trace-extension";
cpu = <&CPU5>;
};
ete6 {
compatible = "arm,embedded-trace-extension";
cpu = <&CPU6>;
};
ete7 {
compatible = "arm,embedded-trace-extension";
cpu = <&CPU7>;
};
trbe {
compatible = "arm,trace-buffer-extension";
interrupts = <GIC_PPI 2 IRQ_TYPE_LEVEL_LOW 0>;
};
trusty {
#size-cells = <0x02>;
#address-cells = <0x02>;
ranges = <0x00>;
compatible = "android,trusty-v1";
virtio {
compatible = "android,trusty-virtio-v1";
};
test {
compatible = "android,trusty-test-v1";
};
log {
compatible = "android,trusty-log-v1";
};
irq {
ipi-range = <0x08 0x0f 0x08>;
interrupt-ranges = <0x00 0x0f 0x00 0x10 0x1f 0x01 0x20 0x3f 0x02>;
interrupt-templates = <0x01 0x00 0x8001 0x01 0x01 0x04 0x8001 0x01 0x00 0x04>;
compatible = "android,trusty-irq-v1";
};
};
/* used in U-boot, Linux doesn't care */
arm_ffa {
compatible = "arm,ffa";
method = "smc";
};
};