What's the frequency of the CPUs in xavier?

Autonomous Machines Jetson & Embedded Systems Jetson AGX Xavier
1

I have looked through the internet and find there is no information about the frequency of CPUs in Xavier.

Thanks.
Daniel.

2

Hi Daniel, please refer to the Jetson Xavier Module Data Sheet, the maximum CPU frequency is 2265MHz.

3

That doesn’t seem right to me. The various 30W nvpmodel profiles set the core max frequencies to 1200 MHz.
The only case that seems to try to clock it up is mode 6, which turns off six of eight cores, and then sets the max clock of core 0/1 to 2100 MHz (which is still not 2265)

Can I run all eight cores at 2 GHz each, safely?

4

Based on the output of jetson_clocks.sh, 4 cores could be jacked up to 2265M HZ.

$ sudo ./jetson_clocks.sh --show
SOC family:tegra194 Machine:jetson-xavier
Online CPUs: 0-3
CPU Cluster Switching: Disabled
cpu0: Gonvernor=schedutil MinFreq=1190400 MaxFreq=1190400 CurrentFreq=1190400
cpu1: Gonvernor=schedutil MinFreq=1190400 MaxFreq=1190400 CurrentFreq=1190400
cpu2: Gonvernor=schedutil MinFreq=1190400 MaxFreq=1190400 CurrentFreq=1190400
cpu3: Gonvernor=schedutil MinFreq=1190400 MaxFreq=1190400 CurrentFreq=1190400
cpu4: Gonvernor=schedutil MinFreq=115200 MaxFreq=2265600 CurrentFreq=652800
cpu5: Gonvernor=schedutil MinFreq=115200 MaxFreq=2265600 CurrentFreq=652800
cpu6: Gonvernor=schedutil MinFreq=115200 MaxFreq=2265600 CurrentFreq=1036800
cpu7: Gonvernor=schedutil MinFreq=115200 MaxFreq=2265600 CurrentFreq=1036800
GPU MinFreq=114750000 MaxFreq=675750000 CurrentFreq=114750000
EMC MinFreq=204000000 MaxFreq=1331200000 CurrentFreq=665600000 FreqOverride=0
Fan: speed=0

5

I manually enabled all cores (0-7), and the last 4 cores were running at the highest frequency.

./jetson_clocks.sh --show

SOC family:tegra194 Machine:jetson-xavier
Online CPUs: 0-7
CPU Cluster Switching: Disabled
cpu0: Gonvernor=schedutil MinFreq=1190400 MaxFreq=1190400 CurrentFreq=1190400
cpu1: Gonvernor=schedutil MinFreq=1190400 MaxFreq=1190400 CurrentFreq=1190400
cpu2: Gonvernor=schedutil MinFreq=1190400 MaxFreq=1190400 CurrentFreq=1190400
cpu3: Gonvernor=schedutil MinFreq=1190400 MaxFreq=1190400 CurrentFreq=1190400
cpu4: Gonvernor=schedutil MinFreq=2265600 MaxFreq=2265600 CurrentFreq=2265600
cpu5: Gonvernor=schedutil MinFreq=2265600 MaxFreq=2265600 CurrentFreq=2265600
cpu6: Gonvernor=schedutil MinFreq=2265600 MaxFreq=2265600 CurrentFreq=2265600
cpu7: Gonvernor=schedutil MinFreq=2265600 MaxFreq=2265600 CurrentFreq=2265600
GPU MinFreq=675750000 MaxFreq=675750000 CurrentFreq=675750000
EMC MinFreq=204000000 MaxFreq=1331200000 CurrentFreq=1331200000 FreqOverride=1
Fan: speed=255

6

Yes, but this is not one that’s supported by NVIDIA.
NVIDIA defines the power models in the file /etc/nvpmodel.conf

There, we find models defined such as:

< POWER_MODEL ID=6 NAME=MODE_30W_2CORE >
CPU_ONLINE CORE_0 1
CPU_ONLINE CORE_1 1
CPU_ONLINE CORE_2 0
CPU_ONLINE CORE_3 0
CPU_ONLINE CORE_4 0
CPU_ONLINE CORE_5 0
CPU_ONLINE CORE_6 0
CPU_ONLINE CORE_7 0
TPC_POWER_GATING TPC_PG_MASK 0
GPU_POWER_CONTROL_ENABLE GPU_PWR_CNTL_EN on
CPU_DENVER_0 MIN_FREQ 1200000
CPU_DENVER_0 MAX_FREQ 2100000
GPU MIN_FREQ 0
GPU MAX_FREQ 900000000
GPU_POWER_CONTROL_DISABLE GPU_PWR_CNTL_DIS auto
EMC MAX_FREQ 1600000000
DLA_CORE MAX_FREQ 1050000000
DLA_FALCON MAX_FREQ 630000000
PVA_VPS MAX_FREQ 760000000
PVA_CORE MAX_FREQ 532000000

and

< POWER_MODEL ID=3 NAME=MODE_30W_ALL >
CPU_ONLINE CORE_0 1
CPU_ONLINE CORE_1 1
CPU_ONLINE CORE_2 1
CPU_ONLINE CORE_3 1
CPU_ONLINE CORE_4 1
CPU_ONLINE CORE_5 1
CPU_ONLINE CORE_6 1
CPU_ONLINE CORE_7 1
TPC_POWER_GATING TPC_PG_MASK 0
GPU_POWER_CONTROL_ENABLE GPU_PWR_CNTL_EN on
CPU_DENVER_0 MIN_FREQ 1200000
CPU_DENVER_0 MAX_FREQ 1200000
CPU_DENVER_1 MIN_FREQ 1200000
CPU_DENVER_1 MAX_FREQ 1200000
CPU_DENVER_2 MIN_FREQ 1200000
CPU_DENVER_2 MAX_FREQ 1200000
CPU_DENVER_3 MIN_FREQ 1200000
CPU_DENVER_3 MAX_FREQ 1200000
GPU MIN_FREQ 0
GPU MAX_FREQ 900000000
GPU_POWER_CONTROL_DISABLE GPU_PWR_CNTL_DIS auto
EMC MAX_FREQ 1600000000
DLA_CORE MAX_FREQ 1050000000
DLA_FALCON MAX_FREQ 630000000
PVA_VPS MAX_FREQ 760000000
PVA_CORE MAX_FREQ 532000000

and

# MAXN is the NONE power model to release all constraints
< POWER_MODEL ID=0 NAME=MAXN >
CPU_ONLINE CORE_0 1
CPU_ONLINE CORE_1 1
CPU_ONLINE CORE_2 1
CPU_ONLINE CORE_3 1
CPU_ONLINE CORE_4 1
CPU_ONLINE CORE_5 1
CPU_ONLINE CORE_6 1
CPU_ONLINE CORE_7 1
TPC_POWER_GATING TPC_PG_MASK 0
GPU_POWER_CONTROL_ENABLE GPU_PWR_CNTL_EN on
CPU_DENVER_0 MIN_FREQ 1200000
CPU_DENVER_0 MAX_FREQ -1
CPU_DENVER_1 MIN_FREQ 1200000
CPU_DENVER_1 MAX_FREQ -1
CPU_DENVER_2 MIN_FREQ 1200000
CPU_DENVER_2 MAX_FREQ -1
CPU_DENVER_3 MIN_FREQ 1200000
CPU_DENVER_3 MAX_FREQ -1
GPU MIN_FREQ 0
GPU MAX_FREQ -1
GPU_POWER_CONTROL_DISABLE GPU_PWR_CNTL_DIS auto
EMC MAX_FREQ 0
DLA_CORE MAX_FREQ -1
DLA_FALCON MAX_FREQ -1
PVA_VPS MAX_FREQ -1
PVA_CORE MAX_FREQ -1

Given that the models that are designated at 30W don’t have the cores running fast, I wonder if it’s actually safe to run the CPU and GPU at max speed at model 0, or whether it will self-throttle at some point.

7

MAX-N mode (mode 0) sets all cores to max frequency. The devkit shouldn’t throttle in nominal ambient temperatures. If you are deploying the module into extended temperatures, you may need to tweak the profile parameters or chose one of the 10/15/30W modes per your cooling solution.

8

That’s great to know; thanks!