We're hiring!
*

Quick hack: Performance debugging Linux graphics on Mesa

Robert Foss avatar

Robert Foss
June 29, 2017

Share this post:

Debugging graphics performance in a simple and high-level manner is possible for all Gallium based Mesa drivers.

GALLIUM_HUD

GALLIUM_HUD is a feature that adds performance graphs to applications that describe various aspects like FPS, CPU usage, etc in realtime.

It is enabled using an environment variable, GALLIUM_HUD, that can be set for GL/EGL/etc applications. It only works for Mesa drivers that are Gallium based, which means that the most drivers (with the notable exception of some Intel drivers) support GALLIUM_HUD.

See GALLIUM_HUD options:

export GALLIUM_HUD=help
glxgears

Android

If you're building Android, you can supply system-wide environment values by doing an export in the init.rc file of the device you are using, like this.

# Go to android source code checkout
cd android

# Add export to init.rc (linaro/generic is the device I use)
nano device/linaro/generic/init.rc
export GALLIUM_HUD cpu,cpu0+cpu1+cpu2+cpu3;pixels-rendered,fps,primitives-generated

Linux

If you're using one of the usual Linux distros, GALLIUM_HUD can be enabled by setting the environtment variable in a place that it loaded early.

# Add export to /etc/environment
nano /etc/environment 
export GALLIUM_HUD cpu,cpu0+cpu1+cpu2+cpu3;pixels-rendered,fps,primitives-generated

Thanks

This post has been a part of work undertaken by my employer Collabora.

 

Original post

Comments (0)


Add a Comment






Allowed tags: <b><i><br>Add a new comment:


Search the newsroom

Latest Blog Posts

Quick hack: Patching kernel modules using DKMS

05/05/2021

DKMS is a framework that is mostly used to build and install external kernel modules. It can also be used to install a specific patch to…

Build your own application with GTK 4 as a Meson subproject!

29/04/2021

Building GTK 4 as a Meson subproject for your own application is not only useful for Windows builds, but also for many Linux distributions…

Profiling virtualized GPU acceleration with Perfetto

22/04/2021

Recently, we have been using Perfetto to successfully profile Apitrace traces in crosvm through VirGL renderer. We have now added perfetto…

Continuous 3D Hand Pose Tracking using Machine Learning & Monado OpenXR

20/04/2021

As part of a project backed by INVEST-AI, a program managed by IVADO Labs, we have developed a multi-stage neural network-based solution…

An easy to use MTP implementation for your next embedded Linux project

13/04/2021

Did you know you could run a permissively-licensed MTP implementation with minimal dependencies on an embedded device? Here's a step-by-step…

Rust: integrating LLVM source-base code coverage with GitLab

24/03/2021

Earlier this year, the Rust compiler gained support for LLVM source-base code coverage. In this post we'll explain how to setup a CI job…

Open Since 2005 logo

We use cookies on this website to ensure that you get the best experience. By continuing to use this website you are consenting to the use of these cookies. To find out more please follow this link.

Collabora Ltd © 2005-2021. All rights reserved. Privacy Notice. Sitemap.

Collabora Limited is registered in England and Wales. Company Registration number: 5513718. Registered office: The Platinum Building, St John's Innovation Park, Cambridge, CB4 0DS, United Kingdom. VAT number: 874 1630 19.