-
Notifications
You must be signed in to change notification settings - Fork 114
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add support for --reverse option #99
Conversation
Codecov Report
@@ Coverage Diff @@
## master #99 +/- ##
==========================================
+ Coverage 88.9% 89.13% +0.22%
==========================================
Files 12 12
Lines 1379 1408 +29
==========================================
+ Hits 1226 1255 +29
Misses 153 153
Continue to review full report at Codecov.
|
Currently the performance when using the |
I reduced allocations, but didn't see a huge speedup. It looks like the main reason the performance can be much worse with the The big slowdown is because not many frames get merged when the stack order is reversed. It makes sense and is really obvious when you look at the resulting flame graph. As an example, when creating a flame graph using Just to make sure that's what was making it slower, I created a version of |
It'd be interesting to see why we're so much slower when producing many frames. Surely it isn't just from having to write more bytes to the output? If so, maybe we need a |
Yes, it would be worth investigating why producing more frames is so much slower. It would be nice to improve the performance overall. I was just initially surprised when I benchmarked with the |
I tried using a Also, just to be clear,
|
This is benchmarking |
That was benchmarking Here is this branch against master (not using
|
@jasonrhansen fwiw, I opened #102. |
Fixes #21