You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
description: Configure the Arm Toolchains Repository package archive on supported Linux AArch64 distributions to enable installing Arm Toolchains products using the system package manager.
26
21
author: Gary Carroll
27
22
28
-
### PAGE SETUP
29
23
weight: 1# Defines page ordering. Must be 1 for first (or only) page.
30
24
tool_install: true # Set to true to be listed in main selection page, else false
31
25
multi_install: false # Set to true if first page of multi-page article, else false
32
26
multitool_install_part: false # Set to true if a sub-page of a multi-page article, else false
33
27
layout: installtoolsall # DO NOT MODIFY. Always true for tool install articles
34
28
---
35
29
36
-
# Arm Toolchains Repository
37
-
38
30
The Arm Toolchains Linux repositories provide a native and convenient way to install selected Arm software products on supported Linux hosts using the host operating system package manager.
39
31
40
32
The repositories are published at [https://developer.arm.com/packages/arm-toolchains/](https://developer.arm.com/packages/arm-toolchains/).
@@ -46,7 +38,7 @@ They are used to distribute Linux packages for Arm Toolchains products, includin
46
38
47
39
Using the Arm Toolchains Linux repositories allows you to install, upgrade, and manage these products using standard Linux package-management tools such as `apt`, `dnf`, `yum`, and `zypper`.
48
40
49
-
Note: root permissions are required to install system packages.
41
+
Root or `sudo` permissions are required to install system packages.
50
42
51
43
{{% notice Repository signing key update July 2026 %}}
52
44
Arm has recently reviewed and strengthened the repository management and publication processes used for the Arm Toolchains Linux repositories.
@@ -60,17 +52,27 @@ Existing installed copies of Arm Toolchain for Linux and Arm Performance Librari
60
52
61
53
## Before you begin
62
54
63
-
You need a supported Linux host with `aarch64` architecture.
55
+
Confirm you are using an Arm Linux system by running:
56
+
57
+
```bash
58
+
uname -m
59
+
```
64
60
65
-
You also need the package manager for your Linux distribution:
61
+
The output should be:
62
+
63
+
```output
64
+
aarch64
65
+
```
66
+
67
+
The commands in this guide use the native package manager for your distribution:
66
68
67
69
-`apt` for Ubuntu LTS releases from 22.04
68
70
-`dnf` or `yum` for RHEL 8 or later and Amazon Linux 2023
69
71
-`zypper` for SLES 15 or 16
70
72
71
73
## What is the Arm Toolchains repository package?
72
74
73
-
Arm provides an `arm-toolchains-repository`package for each supported Linux distribution.
75
+
Arm provides an `arm-toolchains-repository` for each supported Linux distribution.
74
76
75
77
This package configures the Arm Toolchains Linux package repository on your host system.
76
78
@@ -84,27 +86,34 @@ Install this repository package before you install Arm Toolchain for Linux or Ar
84
86
85
87
## Why the signing key matters
86
88
87
-
Linux package managers use signing keys to verify that repository metadata and packages come from the expected publisher.
89
+
Linux package managers use cryptographic signing keys to verify that repository metadata and packages are authentic and come directly from Arm.
90
+
91
+
If you are setting up a new system, installing the `arm-toolchains-repository` package automatically configures the current repository signing key for you. No manual key installation or fingerprint verification is required.
88
92
89
-
For the Arm Toolchains Linux repositories, the signing key is part of the trust model for package installation and upgrades. If your package manager does not have the current Arm Toolchains repository signing key installed, it might refuse to refresh repository metadata or install packages from the repository.
93
+
However, if you are upgrading a system that previously configured the Arm Toolchains Linux repository before July 2026, you must update the repository configuration to use the new key. Otherwise, your package manager may refuse to refresh repository metadata or install packages.
90
94
91
-
The previous Arm Toolchains repository signing key was replaced in July 2026.
95
+
### Repository signing key fingerprints
92
96
93
-
The previous key fingerprint was:
97
+
For reference or manual verification, the fingerprints of the previous and current Arm Toolchains signing keys are:
98
+
99
+
Previous fingerprint:
94
100
95
101
```output
96
102
EE37 7ACD D5AD 4AB6 BD79 9B8A B83D 741E 7A05 AF82
97
103
```
98
104
99
-
The current key fingerprint is:
105
+
Current fingerprint:
100
106
101
107
```output
102
108
A406 5BCE 9386 DD1E 62FD E03B 8144 CA16 11A0 BD71
103
109
```
104
110
105
-
You can display the fingerprint from a downloaded key file with:
111
+
### (Optional) Manually verifying the signing key
112
+
113
+
If you want to manually verify the signing key, ensure the `gpg` utility (part of GnuPG) is installed on your system, then download the key file and display its fingerprint:
Copy file name to clipboardExpand all lines: content/learning-paths/cross-platform/ai-agent-cpu-orchestration/2-setup.md
+11-17Lines changed: 11 additions & 17 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -13,17 +13,6 @@ Prepare the following before running the agent: a Serper API key for web search,
13
13
14
14
These steps are identical on an Apple silicon MacBook, an Arm Linux machine, and an NVIDIA DGX Spark.
15
15
16
-
### Get a Serper API key
17
-
18
-
The agent searches the web through [Serper](https://serper.dev/), a Google Search API. The free tier is enough for you to complete the Learning Path.
19
-
20
-
To get an API key:
21
-
22
-
1. Go to [serper.dev](https://serper.dev/) and create a free account.
23
-
2. Open your dashboard and copy your API key.
24
-
25
-
You'll set this key as an environment variable later in this section, so the agent can read it without the key being written into the code.
26
-
27
16
### Create a project directory and virtual environment
28
17
29
18
A virtual environment keeps the agent's dependencies isolated from your system Python, so you always run against the right package versions.
@@ -54,16 +43,21 @@ Install the two packages:
54
43
pip install requests beautifulsoup4
55
44
```
56
45
57
-
###Set your Serper API key
46
+
## Set your Serper API key
58
47
59
-
Export your API key as an environment variable in the same terminal session:
48
+
The agent searches the web through [Serper](https://serper.dev/), a Google Search API. The free tier is enough to complete this Learning Path.
60
49
61
-
```bash
62
-
export SERPER_API_KEY="your-serper-api-key"
63
-
```
50
+
1. Go to [serper.dev](https://serper.dev/) and create a free account.
51
+
2. After logging in, you'll see the dashboard. On the left, select **API Keys**.
52
+
3. Copy the Default key.
53
+
4. Export your API key as an environment variable in the same terminal session as your virtual environment:
54
+
55
+
```bash
56
+
export SERPER_API_KEY="your-serper-api-key"
57
+
```
64
58
65
59
{{% notice Tip %}}
66
-
This variable lasts only for the current terminal session. To keep the variable across sessions, add the same line to your shell profile, for example `~/.zshrc` on macOS or `~/.bashrc` on Linux.
60
+
This variable only lasts for the current terminal session. To keep it across sessions, add the same line to your shell profile, for example `~/.zshrc` on macOS or `~/.bashrc` on Linux.
Copy file name to clipboardExpand all lines: content/learning-paths/cross-platform/ai-agent-cpu-orchestration/3-ollama.md
+16-9Lines changed: 16 additions & 9 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -21,26 +21,33 @@ Running locally has three benefits that matter for an agent:
21
21
22
22
## Install and start Ollama
23
23
24
-
Install [Ollama](https://ollama.com/) and start its server using one of the following options. The server exposes a local API at `http://localhost:11434` that the agent connects to.
24
+
Install [Ollama](https://ollama.com/) and start its server. The server exposes a local API at `http://localhost:11434` that the agent connects to.
25
+
26
+
1. Install Ollama using one of the following options:
With [Homebrew](https://brew.sh/), `brew services` runs Ollama in the background, so you can use the same terminal throughout. With the install script, `ollama serve` runs in the foreground, so keep that terminal open and use a second terminal for the remaining commands.
48
+
With [Homebrew](https://brew.sh/), `brew services` runs Ollama in the background, so you can use the same terminal throughout. To stop it later, run `brew services stop ollama`.
49
+
50
+
With the install script, `ollama serve` runs in the foreground. Keep that terminal open and use a second terminal for the remaining commands. To stop it later, press **Ctrl+C** in the terminal running `ollama serve`.
Copy file name to clipboardExpand all lines: content/learning-paths/servers-and-cloud-computing/minecraft-on-oci/_index.md
+1-1Lines changed: 1 addition & 1 deletion
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -3,7 +3,7 @@ title: Run a Minecraft server on an Arm-based Oracle Cloud Infrastructure instan
3
3
4
4
description: Deploy a Minecraft Java Edition server on an Arm-based OCI A1 instance, open port `25565`, and connect from the Minecraft client.
5
5
6
-
minutes_to_complete: 10
6
+
minutes_to_complete: 30
7
7
8
8
who_is_this_for: This is an introductory topic for developers who are new to Oracle Cloud Infrastructure (OCI) and want to provision an arm64 instance and run a persistent Minecraft server on it.
title: Profile GPT-2 instruction mix with Arm Performix
3
+
4
+
description: Learn how to profile GPT-2 inference on Arm Neoverse with the Arm Performix Instruction Mix recipe, identify scalar versus vector execution patterns, and improve throughput with NEON, SVE, and KleidiAI kernels.
5
+
6
+
minutes_to_complete: 45
7
+
8
+
who_is_this_for: This is an introductory topic for developers who want to get started using the instruction mix recipe in Arm Performix through a practical example.
9
+
10
+
learning_objectives:
11
+
- Explain how the Instruction Mix recipe combines static disassembly with runtime sampling to show execution behavior
12
+
- Build and run the GPT-2 inference example on an Arm Linux server
13
+
- Identify why matrix multiplication dominates runtime and how vectorization changes the instruction mix
14
+
- Compare throughput and instruction mix across scalar, NEON, SVE, and KleidiAI implementations
15
+
16
+
prerequisites:
17
+
- Access to Arm Performix configured with a remote Arm Linux target. For setup, see the [Arm Performix install guide](/install-guides/performix/)
18
+
- Basic understanding of C++ and compiler optimization
19
+
- Basic understanding of matrix multiplication
20
+
- Basic understanding of writing SIMD code with Neon and/or SVE.
0 commit comments