[Merged by Bors] - Add trace_tracy feature for Tracy profiling #2832
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
DJMcNab
approved these changes
Sep 16, 2021
mockersf
requested changes
Sep 16, 2021
There was a problem hiding this comment.
Could you also document the new feature in https://github.com/bevyengine/bevy/blob/main/docs/cargo_features.md
|
also some sort of doc on tracing could be useful but can be done in another pr |
Done.
Once I have used it a bit more I will write something as a quick start guide, but Tracy has a pretty comprehensive pdf documentation that ships with each release, so I would probably defer to that. |
DJMcNab
reviewed
Sep 16, 2021
DJMcNab
approved these changes
Sep 16, 2021
IceSentry
approved these changes
Sep 16, 2021
There was a problem hiding this comment.
Maybe consider updating https://github.com/bevyengine/bevy/blob/main/docs/profiling.md
|
If people are interested to see how I've been using this, you can check out the Visibility culling thread in #rendering-dev on Discord. Basically looking at mean system execution times to understand how much of an improvement frustum culling was producing, and in which systems. It's great for getting more of that sort of engineering feel for how the engine 'sounds'. :) It lead to me learning some stuff about how vsync seems to misbehave on macOS too. |
This could use an overhaul in general. Imo we shouldn't recommend thread_profiler anymore when we have |
|
Imo we should do that separately. |
|
bors r+ |
bors bot
pushed a commit
that referenced
this pull request
Sep 16, 2021
# Objective [Tracy](https://github.com/wolfpld/tracy) is: > A real time, nanosecond resolution, remote telemetry, hybrid frame and sampling profiler for games and other applications. With the `trace_tracy` feature enabled, you run your bevy app and either a headless server (`capture`) or a live, interactive profiler UI (`Tracy`), and connect that to your bevy application to then stream the metric data and events, and save it or inspect it live/offline. Previously when I implemented the spans across systems and stages and I was trying out different profiling tools, Tracy was too unstable on macOS to use. But now, quite some months later, it is working stably with Tracy 0.7.8. You can see timelines, aggregate statistics of mean system/stage execution times, and much more. It's very useful!  ## Solution - Use the `tracing-tracy` crate which supports our tracing spans - Expose via the non-default feature `trace_tracy` for consistency with other `trace*` features
|
Build failed (retrying...): |
bors bot
pushed a commit
that referenced
this pull request
Sep 16, 2021
# Objective [Tracy](https://github.com/wolfpld/tracy) is: > A real time, nanosecond resolution, remote telemetry, hybrid frame and sampling profiler for games and other applications. With the `trace_tracy` feature enabled, you run your bevy app and either a headless server (`capture`) or a live, interactive profiler UI (`Tracy`), and connect that to your bevy application to then stream the metric data and events, and save it or inspect it live/offline. Previously when I implemented the spans across systems and stages and I was trying out different profiling tools, Tracy was too unstable on macOS to use. But now, quite some months later, it is working stably with Tracy 0.7.8. You can see timelines, aggregate statistics of mean system/stage execution times, and much more. It's very useful!  ## Solution - Use the `tracing-tracy` crate which supports our tracing spans - Expose via the non-default feature `trace_tracy` for consistency with other `trace*` features
bors
bot
changed the title
superdump
added a commit
to superdump/bevy
that referenced
this pull request
Sep 17, 2021
[Tracy](https://github.com/wolfpld/tracy) is: > A real time, nanosecond resolution, remote telemetry, hybrid frame and sampling profiler for games and other applications. With the `trace_tracy` feature enabled, you run your bevy app and either a headless server (`capture`) or a live, interactive profiler UI (`Tracy`), and connect that to your bevy application to then stream the metric data and events, and save it or inspect it live/offline. Previously when I implemented the spans across systems and stages and I was trying out different profiling tools, Tracy was too unstable on macOS to use. But now, quite some months later, it is working stably with Tracy 0.7.8. You can see timelines, aggregate statistics of mean system/stage execution times, and much more. It's very useful!  - Use the `tracing-tracy` crate which supports our tracing spans - Expose via the non-default feature `trace_tracy` for consistency with other `trace*` features
superdump
added a commit
to superdump/bevy
that referenced
this pull request
Oct 3, 2021
[Tracy](https://github.com/wolfpld/tracy) is: > A real time, nanosecond resolution, remote telemetry, hybrid frame and sampling profiler for games and other applications. With the `trace_tracy` feature enabled, you run your bevy app and either a headless server (`capture`) or a live, interactive profiler UI (`Tracy`), and connect that to your bevy application to then stream the metric data and events, and save it or inspect it live/offline. Previously when I implemented the spans across systems and stages and I was trying out different profiling tools, Tracy was too unstable on macOS to use. But now, quite some months later, it is working stably with Tracy 0.7.8. You can see timelines, aggregate statistics of mean system/stage execution times, and much more. It's very useful!  - Use the `tracing-tracy` crate which supports our tracing spans - Expose via the non-default feature `trace_tracy` for consistency with other `trace*` features
superdump
added a commit
to superdump/bevy
that referenced
this pull request
Oct 8, 2021
[Tracy](https://github.com/wolfpld/tracy) is: > A real time, nanosecond resolution, remote telemetry, hybrid frame and sampling profiler for games and other applications. With the `trace_tracy` feature enabled, you run your bevy app and either a headless server (`capture`) or a live, interactive profiler UI (`Tracy`), and connect that to your bevy application to then stream the metric data and events, and save it or inspect it live/offline. Previously when I implemented the spans across systems and stages and I was trying out different profiling tools, Tracy was too unstable on macOS to use. But now, quite some months later, it is working stably with Tracy 0.7.8. You can see timelines, aggregate statistics of mean system/stage execution times, and much more. It's very useful!  - Use the `tracing-tracy` crate which supports our tracing spans - Expose via the non-default feature `trace_tracy` for consistency with other `trace*` features
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Objective
Tracy is:
With the
trace_tracyfeature enabled, you run your bevy app and either a headless server (capture) or a live, interactive profiler UI (Tracy), and connect that to your bevy application to then stream the metric data and events, and save it or inspect it live/offline.Previously when I implemented the spans across systems and stages and I was trying out different profiling tools, Tracy was too unstable on macOS to use. But now, quite some months later, it is working stably with Tracy 0.7.8. You can see timelines, aggregate statistics of mean system/stage execution times, and much more. It's very useful!
Solution
tracing-tracycrate which supports our tracing spanstrace_tracyfor consistency with othertrace*features