Generated banner for Jetson Optimize Memory
Generated logo for Jetson Optimize Memory

Jetson Optimize Memory

NVIDIAOfficialCreated by Jetson TeamDistributed by NVIDIA

  • Skill
  • Workflow
  • Cloud and DevOps
  • Apache-2.0
  • Medium risk

From the author

Reclaim DRAM by disabling unused subsystems across MB1 BCT, MB2 BCT, kernel reserved-memory, and SWIOTLB. Use for headless or no-camera Jetson deployments; not for CPU/GPU frequency tuning.
— the description by Jetson Team

AutoGineer summary

Written by AutoGineer from the skill’s own files, not by the author

This skill provides a structured procedure to optimize memory usage on NVIDIA Jetson platforms by disabling unused hardware subsystems. By modifying MB1 BCT, MB2 BCT, kernel reserved-memory, and SWIOTLB configurations, users can reclaim DRAM carveouts that are typically reserved for features like display or camera pipelines. It is intended for headless or specialized deployments where these peripherals are not required, ensuring that memory is freed for application use while maintaining system stability.

What it does

  • Provides recipes to disable display-related carveouts for headless deployments.
  • Provides recipes to disable camera-related carveouts for no-camera deployments.
  • Guides the user through modifying MB1 and MB2 BCT files to zero out carveouts.
  • Instructs on disabling kernel reserved-memory nodes for specific hardware.
  • Provides instructions for adjusting the SWIOTLB DMA bounce pool size.
  • Includes mandatory verification steps to ensure changes are correctly applied and stable.

When to use it

  • When deploying a Jetson module in a headless environment without a display.
  • When deploying a Jetson module that does not require camera or ISP functionality.
  • When you need to maximize available DRAM for compute-intensive applications.
  • When you need to adjust the SWIOTLB pool size for specific peripheral DMA requirements.

Advantages

  • Increases available system memory by removing unnecessary hardware reservations.
  • Reduces the memory footprint of the OS for specialized embedded deployments.
  • Provides a systematic, multi-layer approach to ensure boot-time stability.
  • Includes verification procedures to prevent boot failures and DMA errors.

Usage examples

“I am deploying an Orin Nano in a headless configuration. How do I reclaim the memory reserved for display?”
The agent will guide you through the headless recipe, including disabling DCE-family carveouts in MB1/MB2 BCT and disabling the display node in the kernel DTS.
“How can I reduce the memory footprint of my Jetson Thor deployment by disabling camera subsystems?”
The agent will provide the no-camera recipe for T264, detailing which RCE and VI/ISP carveouts to zero out and which kernel nodes to disable.

Install

  1. Download the skill as a ZIP:

    https://mp-staging.autogineer.com/agent-skills/nvidia/jetson-optimize-memory/download
  2. Unzip it into the cross-client skills folder of your project:

    mkdir -p .agents/skills && unzip jetson-optimize-memory.zip -d .agents/skills

Documentation by Jetson Team

jetson-optimize-memory

Memory is reserved across four layers ordered by boot chronology (higher row = earlier in boot, closer to hardware):

LayerContentKey files
MB1 BCTfirmware carveoutsper-module misc DTS
MB2 BCTfirmware loading + AST controlsper-module misc DTS
Kernel DTSreserved-memory and driver bindingper-module DTS
SWIOTLBDMA bounce pool size<module>.conf.common (CMDLINE_ADD)

Critical rules:

  • Only the scenarios in Scenario recipes are validated. Refuse any request to disable a carveout/cluster/node not in that table.

  • Zeroing a carveout requires both: disabling the cluster's loading controls AND removing the AST that references it.

  • Emit explicit overrides for every cluster in the recipe, regardless of how the source #ifdef/#else looks. Verify the merged binary.

  • GPU SW stack must match the chip: T234 -> nvgpu, T264 and later -> OpenRM. Follow the shared derived-platform rule in target-platform-contract.md; stop on mismatches instead of guessing.

  • Do not set SWIOTLB to 0 — some peripherals can't use the IOMMU.


Scenario recipes

KeywordMB1 BCT carveoutsMB2 BCTKernel DTS
headlessDCE-family (see chip table)DCE auxp_controls + DCE AST(s)display@<addr> (and dce@<addr> if exposed) → disabled
no-cameraRCE/VI/ISP-familyRCE auxp_controls (each instance) + RCE AST(s)recommended: VI/ISP/NVCSI → disabled
Chip-specific carveouts
ScenarioT234 (Orin)T264 (Thor)
headlessCARVEOUT_BPMP_DCE, CARVEOUT_DCE, CARVEOUT_DCE_TSEC, CARVEOUT_TSEC_DCE, CARVEOUT_DISP_EARLY_BOOT_FBCARVEOUT_DCE, CARVEOUT_TSEC_DCE, CARVEOUT_HPSE_DCE, CARVEOUT_DISP_EARLY_BOOT_FB
no-cameraCARVEOUT_RCE, CARVEOUT_CAMERA_TASKLISTCARVEOUT_RCE, CARVEOUT_RCE1, CARVEOUT_RCE_RW, CARVEOUT_VI_TASKLIST, CARVEOUT_VI1_TASKLIST, CARVEOUT_ISP_TASKLIST, CARVEOUT_ISP1_TASKLIST

Post-boot: headless → sudo systemctl set-default multi-user.target.


MB1 BCT carveout overrides

File: Linux_for_Tegra/bootloader/generic/BCT/tegra<chip>-mb1-bct-misc-<module>.dts (e.g. tegra234-mb1-bct-misc-p3767-0000.dts for Orin Nano, tegra264-mb1-bct-misc-p3834-0008-p4071-0000.dts for Thor).

For each carveout, add inside the existing carveout node:

aux_info@<CARVEOUT_NAME> {
    pref_base = <0x0 0x0>;
    size      = <0x0 0x0>;
    alignment = <0x0 0x0>;
};

MB2 BCT cluster + AST overrides

File: Linux_for_Tegra/bootloader/generic/BCT/tegra<chip>-mb2-bct-misc-<module>.dts (includes tegra<chip>-mb2-bct-common.dtsi).

For each target cluster:

  1. Override auxp_controls@<index>:

    auxp_controls@<index> {
        enable_init    = <0>;
        enable_fw_load = <0>;
        enable_unhalt  = <0>;
    };
  2. /delete-node/ auxp_ast_config@<idx>;

Look up indices in common.dtsi: auxp_controls@N carries a comment naming its cluster; auxp_ast_config@N has ast_region children whose carveout = <CARVEOUT_…>; lines identify the owner.


Kernel DT reserved-memory

DTB=Linux_for_Tegra/kernel/dtb/<platform-dtb-name>.dtb
dtc -I dtb -O dts -o /tmp/platform.dts $DTB
# edit: status = "disabled" on target nodes
dtc -I dts -O dtb -o $DTB /tmp/platform.dts

Display — disable display@<addr>, plus dce@<addr> if exposed as a separate kernel node.

Camera — under host1x@<addr>, disable whichever of vi* / isp* / nvcsi exist on the BSP (only emit present nodes):

Locate the display controller node in the decompiled DTS and disable it. The node's unit address is chip-specific — find it by compatible string (e.g. nvidia,tegra234-display) rather than hard-coding the address.

host1x@<addr> {
    vi0@<addr>   { status = "disabled"; };
    vi1@<addr>   { status = "disabled"; };
    isp@<addr>   { status = "disabled"; };
    isp1@<addr>  { status = "disabled"; };
    nvcsi@<addr> { status = "disabled"; };
};

SWIOTLB DMA bounce pool

The NVIDIA IOMMU covers peripheral DMA, so SWIOTLB is rarely used. Edit CMDLINE_ADD (never CMDLINE) in Linux_for_Tegra/<module>.conf.common:

# Total bytes = swiotlb_value × 2048; 4 MiB pool:
CMDLINE_ADD="... swiotlb=2048"

Override verification (mandatory)

After every patched MB1/MB2 BCT .dts, reproduce the BSP's compile + decompile using the same -D… flags from bct_flags.append(...) in bootloader/tegraflash_impl_t<chip>.py:

gcc -E -nostdinc -x assembler-with-cpp \
    -DENABLE_<FLAG_1> -DENABLE_<FLAG_2> \
    -I bootloader -I bootloader/generic/BCT \
    -o /tmp/cpp.dts <patched-bct.dts>
dtc -q -I dts -O dtb -o /tmp/cpp.dtb /tmp/cpp.dts
dtc -q -I dtb -O dts /tmp/cpp.dtb | less

Confirm in the merged output:

  • Each zeroed aux_info@<NAME> (or aux_info@<id>U post macro expansion) has size = <0x0 0x0> and pref_base = <0x0 0x0>.

  • Each disabled auxp_controls@<idx> has all three enable_* fields <0>.

  • Each /delete-node/'d auxp_ast_config@<idx> is absent.


Verification (on booted target)

sudo cat /proc/iomem | grep -iE 'nv-reserved|cma|fb|carveout'
ls /proc/device-tree/reserved-memory/
dmesg | grep -iE 'firmware|carveout|bpmp|reserved|fail|error' | head -20
free -m
ScenarioSysfsdmesg grep
Display offls /sys/class/drm/ (empty)tegra-drm|nvdisplay|dce|host1x|fb0
Camera offls /dev/video* 2>/dev/null (none)rce|nvcsi|tegra-camera|vi0|vi1|isp
SWIOTLB shrinkcat /sys/kernel/debug/swiotlb/io_tlb_nslabs matches cmdlineswiotlb

For SWIOTLB: /proc/cmdline must contain swiotlb=<value>, and watch -n5 cat /sys/kernel/debug/swiotlb/io_tlb_used must stay under io_tlb_nslabs during full workload — if exceeded, restore original CMDLINE_ADD and re-flash kernel-dtb.

Purpose

Cut the unused DRAM carveouts that ship enabled in the reference BSP when a Jetson deployment skips display, camera, or other peripherals, freeing the freed bytes for the application. Always edits the four layers in boot order so an early-stage carveout never outranks a later-stage shrink.

Prerequisites

  • Active target profile resolved per ../../context/target-platform-contract.md.

  • BSP image extracted and source tree initialized (/jetson-init-image, /jetson-init-source complete).

  • For headless / no-camera recipes: confirm the workload truly does not need display or camera.

Limitations

  • Only the validated recipes (headless, no-camera, swiotlb) are exposed; ad-hoc subsystem disables outside the recipe set are refused.

  • SWIOTLB shrink is bounded by peak in-flight DMA — exceeding the new io_tlb_nslabs requires reverting the change.

  • BPMP-DTB edits land in the overlay tracker only after Customize + Build + Deploy run; this skill does not flash on its own.

Troubleshooting

  • Boot fails after MB1 BCT carveout disable — restore the pristine misc DTS and re-flash; the missing carveout is mandatory for the active SoC.

  • io_tlb_used exceeds io_tlb_nslabs — revert swiotlb= in CMDLINE_ADD and re-flash the kernel DTB partition.

  • Reclaimed delta smaller than expected — verify the recipe truly matched the deployment (e.g. display still attached); use the dmesg | grep -iE 'firmware|carveout' check in this file to confirm.

  • Validation dmesg shows the disabled subsystem still probing — the change probably did not promote through to bsp_image; re-run /jetson-promote-image.


Files

skill-card.md

Description: <br>

Reclaim DRAM by disabling unused subsystems across MB1 BCT, MB2 BCT, kernel reserved-memory, and SWIOTLB for headless or no-camera Jetson deployments. <br>

This skill is ready for commercial/non-commercial use. <br>

Owner

NVIDIA <br>

License/Terms of Use: <br>

CC-BY-4.0 AND Apache 2.0 <br>

Use Case: <br>

Developers and engineers deploying Jetson devices in headless or camera-less configurations use this skill to free reserved DRAM carveouts that are unused in their deployment scenario. <br>

Deployment Geography for Use: <br>

Global <br>

Known Risks and Mitigations: <br>

Risk: Review before execution as proposals could introduce incorrect or misleading guidance into skills. <br> Mitigation: Review and scan skill before deployment. <br>

Reference(s): <br>

Skill Output: <br>

Output Type(s): [Shell commands, Configuration instructions, Code] <br> Output Format: [Markdown with inline DTS and bash code blocks] <br> Output Parameters: [1D] <br> Other Properties Related to Output: [None] <br>

Evaluation Agents Used: <br>

  • Claude Code (claude-code) <br>

  • Codex (codex) <br>

Evaluation Tasks: <br>

Evaluated against 2 internal evaluation tasks in the astra-sandbox environment using the external NVSkills-Eval profile. <br>

Evaluation Metrics Used: <br>

Reported benchmark dimensions: <br>

  • Security: Checks whether skill-assisted execution avoids unsafe behavior such as secret leakage, destructive commands, or unauthorized access. <br>

  • Correctness: Checks whether the agent follows the expected workflow and produces the correct final output. <br>

  • Discoverability: Checks whether the agent loads the skill when relevant and avoids using it when irrelevant. <br>

  • Effectiveness: Checks whether the agent performs measurably better with the skill than without it. <br>

  • Efficiency: Checks whether the agent uses fewer tokens and avoids redundant work. <br>

Underlying evaluation signals used in this run: <br>

  • security: Checks for unsafe operations, secret leakage, and unauthorized access. <br>

  • skill_execution: Verifies that the agent loaded the expected skill and workflow. <br>

  • skill_efficiency: Checks routing quality, decoy avoidance, and redundant tool usage. <br>

  • accuracy: Grades final-answer correctness against the reference answer. <br>

  • goal_accuracy: Checks whether the overall user task completed successfully. <br>

  • behavior_check: Verifies expected behavior steps, including safety expectations. <br>

  • token_efficiency: Compares token usage with and without the skill. <br>

Evaluation Results: <br>

DimensionNumclaude-codecodex
Security2100% (+0%)100% (+0%)
Correctness2100% (+68%)91% (+30%)
Discoverability2100% (+88%)72% (+21%)
Effectiveness273% (+53%)66% (+39%)
Efficiency294% (+69%)66% (+23%)

Skill Version(s): <br>

0.0.1 (source: frontmatter) <br>

Ethical Considerations: <br>

NVIDIA believes Trustworthy AI is a shared responsibility and we have established policies and practices to enable development for a wide array of AI applications. When downloaded or used in accordance with our terms of service, developers should work with their internal team to ensure this skill meets requirements for the relevant industry and use case and addresses unforeseen product misuse. <br>

(For Release on NVIDIA Platforms Only) <br> Please report quality, risk, security vulnerabilities or NVIDIA AI Concerns here. <br>

Requirements and compatibility

Written for: Claude Code

Works with: Claude Code, Claude apps, OpenAI Codex, GitHub Copilot, Cursor, Gemini CLI, Kiro, OpenCode, Goose, Amp, Roo Code, JetBrains Junie, Factory Droid, Devin (ex-Windsurf), Mistral Vibe, OpenHands, Letta

“Works with” means the tool reads the open Agent Skills format; a skill written for one tool may use features others lack.

Needs: Claude Code frontmatter, Needs Node.js

Security

What our scanner noticed· Medium risk

  • Large base64-encoded blobs

A static scan is not a guarantee. Skills are third-party content: review a skill before you let an agent use it.

Attribution and licence

Created by
Jetson Team
Publisher
NVIDIA (Official)
Distributed by
NVIDIA
Source
NVIDIA/skills · skills/jetson-optimize-memory · commit fce3cca
Licence
Apache License 2.0 (redistribution allowed)
  • LICENSE (added by AutoGineer from LICENSE-APACHE)
  • ATTRIBUTION.md (written by AutoGineer)
Synced

AutoGineer is not affiliated with or endorsed by Jetson Team or NVIDIA.

Download ZIP

The ZIP holds the skill’s files, its licence files and an ATTRIBUTION.md. Download jetson-optimize-memory.zip