vsavchenko added a comment.

Introduction
------------

I have some test results with execution times.  Because the execution time 
varies from one execution to another, I've measured performance on master twice 
(I know that it's not very conclusive) to show that equivalence tracking 
doesn't affect performance of the analyzer.  So, please, let's treat it more 
like a smoke test that everything stayed the same with a new functionality 
rather than a full-blown performance benchmarking.

Raw results
-----------

| project    | master run#1 | master run#2 | patch run#1 |
| ---------- | ------------ | ------------ | ----------- |
| box2d      | 137.57s      | 156.36s      | 142.46s     |
| cxxopts    | 26.74s       | 29.31s       | 27.01s      |
| drogon     | 257.43s      | 265.88s      | 263.00s     |
| duckdb     | 2251.08s     | 2297.60s     | 2258.06s    |
| fmt        | 266.96s      | 293.06s      | 279.26s     |
| libsoundio | 33.89s       | 35.37s       | 36.48s      |
| oatpp      | 171.58s      | 175.25s      | 172.92s     |
| re2        | 84.12s       | 133.46s      | 87.24s      |
| simbody    | 1405.25s     | 1374.38s     | 1365.58s    |
| symengine  | 455.94s      | 487.50s      | 466.28s     |
| termbox    | 20.91s       | 20.28s       | 20.88s      |
| tinyexpr   | 11.91s       | 13.53s       | 12.58s      |
| tinyspline | 48.61s       | 49.01s       | 48.90s      |
| tinyvm     | 27.77s       | 29.32s       | 28.70s      |
| zstd       | 150.28s      | 152.86s      | 154.21s     |
|



Processed results
-----------------

| project    | min delta | max delta | within range                         |
| ---------- | --------- | --------- | ------------------------------------ |
| box2d      | -8.9%     | 3.6%      | {icon check-circle color=green}      |
| cxxopts    | -7.8%     | 1.0%      | {icon check-circle color=green}      |
| drogon     | -1.1%     | 2.2%      | {icon check-circle color=green}      |
| duckdb     | -1.7%     | 0.3%      | {icon check-circle color=green}      |
| fmt        | -4.7%     | 4.6%      | {icon check-circle color=green}      |
| libsoundio | 3.1%      | 7.6%      | {icon arrow-circle-up color=red}     |
| oatpp      | -1.3%     | 0.8%      | {icon check-circle color=green}      |
| re2        | -34.6%    | 3.7%      | {icon check-circle color=green}      |
| simbody    | -2.8%     | -0.6%     | {icon arrow-circle-down color=green} |
| symengine  | -4.4%     | 2.3%      | {icon check-circle color=green}      |
| termbox    | -0.1%     | 3.0%      | {icon check-circle color=green}      |
| tinyexpr   | -7.0%     | 5.6%      | {icon check-circle color=green}      |
| tinyspline | -0.2%     | 0.6%      | {icon check-circle color=green}      |
| tinyvm     | -2.1%     | 3.3%      | {icon check-circle color=green}      |
| zstd       | 0.9%      | 2.6%      | {icon arrow-circle-up color=red}     |
|


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D83286/new/

https://reviews.llvm.org/D83286



_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

Reply via email to