Skip to content

Commit ddad89d

Browse files
PMUV3 Learning path changes
1 parent 5458323 commit ddad89d

File tree

4 files changed

+12
-16
lines changed

4 files changed

+12
-16
lines changed

content/learning-paths/servers-and-cloud-computing/PMUv3_plugin_learning_path/_index.md

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -1,14 +1,14 @@
11
---
2-
title: Performance analysis at a code level using PMUv3_plugin
2+
title: Code level Performance Analysis using PMUv3_plugin
33

44
minutes_to_complete: 60
55

66
who_is_this_for: Engineers working on a workload/codebase to carry out performance analysis at code block level.
77

88
learning_objectives:
9-
- Get a fine-grained and precise measurement of functions handling a specific task.
9+
- Get a fine-grained and precise measurement of functions or chunk of code handling a specific task.
1010
- Instrument at a code/function level by choosing one of the 4 scenarios detailed in this learning path.
11-
- Run and collect the CSV of all the 15 bundles in one shot.
11+
- Run and collect the performance metrics as well as raw event values of 15 groups/bundles of PMUs in one shot.
1212
- Use the backend tool to plot raw PMU event values along with the KPI/metric values like MPKI, stalls, IPC and many more, thus helping in performance analysis tasks.
1313

1414
prerequisites:

content/learning-paths/servers-and-cloud-computing/PMUv3_plugin_learning_path/_next-steps.md

Lines changed: 8 additions & 9 deletions
Original file line numberDiff line numberDiff line change
@@ -5,18 +5,17 @@ recommended_path: /learning-paths/servers-and-cloud-computing/pmuv3_plugin_learn
55

66
further_reading:
77
- resource:
8-
title: PLACEHOLDER MANUAL
9-
link: PLACEHOLDER MANUAL LINK
8+
title: Arm Neoverse N2 PMU Guide
9+
link: https://developer.arm.com/documentation/PJDOC-466751330-590448/2-0/?lang=en
1010
type: documentation
1111
- resource:
12-
title: PLACEHOLDER BLOG
13-
link: PLACEHOLDER BLOG LINK
14-
type: blog
12+
title: Arm CPU Telemetry Solution Topdown Methodology Specification
13+
link: https://developer.arm.com/documentation/109542/0100/?lang=en
14+
type: documentation
1515
- resource:
16-
title: PLACEHOLDER GENERAL WEBSITE
17-
link: PLACEHOLDER GENERAL WEBSITE LINK
18-
type: website
19-
16+
title: Arm Neoverse N2 Core Telemetry Specification
17+
link: https://developer.arm.com/documentation/109215/0200/?lang=en
18+
type: documentation
2019

2120
# ================================================================================
2221
# FIXED, DO NOT MODIFY

content/learning-paths/servers-and-cloud-computing/PMUv3_plugin_learning_path/instrumentation.md

Lines changed: 0 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -118,8 +118,6 @@ shutdown_resources();
118118
process_data(cur_bundle_no);
119119
```
120120

121-
---
122-
123121
## Scenario II – Instrumentation Around Single Code block in C++ codebase
124122

125123
For Pmuv3_Bundles Instrumentation "SINGLE CHUNK OF CODE" in C++ CODEBASE, refer the below steps.
@@ -228,7 +226,6 @@ process_data(cur_bundle_no);
228226
```yaml
229227
shutdown_resources();
230228
```
231-
---
232229

233230
## Scenario IV – Instrumentation around single code blocks in C codebase
234231
For Pmuv3_Bundles Instrumentation in a C Codebase around a Single Chunk Of Code, refer below steps.

content/learning-paths/servers-and-cloud-computing/PMUv3_plugin_learning_path/intro.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -5,7 +5,7 @@ weight: 2
55
### FIXED, DO NOT MODIFY
66
layout: learningpathall
77
---
8-
## Introduction to the PMUv3 Plugin
8+
## Why should one use PMUv3 Plugin?
99

1010
There is a way to measure profiling of the target application as a whole box, but sometimes we need to add the instrumentation into the code itself to get more fine-grained and precise measurement of the functions handling a specific task we are interested to know its performance. For that intention, we developed the PMUv3 Plugin to allow the users to do so. This performance monitoring plugin helps to do performance analysis based on the Hardware events available in PMUV3.
1111

0 commit comments

Comments
 (0)