fix: downgrade mimalloc_rust version because of performance issue.

[fukusuket]

I don't know if it's a mimalloc_rust problem or a mimalloc problem, the speed is degraded as shown in the benchmark results below.(On the other hand, memory usage seems to have improved a little? ... 🤔)

What Changed

• Downgrade mimalloc_rust/libmimalloc-sys versions.
  • because latest mimallc_rust(v0.1.34) seems to have some performance issue.

Evidence

Environment

• OS: Windows 10 Home edition
• Hard: Memory 16GB , Core 8, SSD, laptop

Benchmark

I ran a benchmark using this procedure(6.1GB evtx) and the results were as follows.

Version	Elapsed time	Memory(peak)	Events with hits / Total events	Output file size(bytes)
2.1.0	00:13:51.232	5.0 GiB	1,593,660 / 4,817,181	575217451
This PR	00:13:33.355	5.4 GiB	1,593,660 / 4,817,181	575217451

Console output

2.1.0

Results Summary:

Events with hits / Total events: 1,593,660 / 4,817,181 (Data reduction: 3,223,521 events (66.92%))

Total | Unique detections: 1,627,378 | 148
Total | Unique critical detections: 0 (0.00%) | 0 (0.00%)
Total | Unique high detections: 12,044 (0.74%) | 20 (13.51%)
Total | Unique medium detections: 11,267 (0.69%) | 38 (25.68%)
Total | Unique low detections: 1,053,568 (64.74%) | 40 (27.03%)
Total | Unique informational detections: 550,499 (33.83%) | 50 (33.78%)
...
Saved file: 2.csv (575.2 MB)
Elapsed time: 00:13:51.232
Rule Parse Processing Time: 00:00:20.664
Analysis Processing Time: 00:13:09.427
Output Processing Time: 00:00:21.139

Memory usage stats:
heap stats:    peak      total      freed    current       unit      count
  reserved:    5.0 GiB    5.0 GiB   83.0 MiB    4.9 GiB
 committed:    4.6 GiB   67.4 GiB   63.0 GiB    4.4 GiB

This PR

Results Summary:

Events with hits / Total events: 1,593,660 / 4,817,181 (Data reduction: 3,223,521 events (66.92%))

Total | Unique detections: 1,627,378 | 148
Total | Unique critical detections: 0 (0.00%) | 0 (0.00%)
Total | Unique high detections: 12,044 (0.74%) | 20 (13.51%)
Total | Unique medium detections: 11,267 (0.69%) | 38 (25.68%)
Total | Unique low detections: 1,053,568 (64.74%) | 40 (27.03%)
Total | Unique informational detections: 550,499 (33.83%) | 50 (33.78%)
...
Saved file: 1.csv (575.2 MB)
Elapsed time: 00:13:33.355
Rule Parse Processing Time: 00:00:19.576
Analysis Processing Time: 00:12:52.228
Output Processing Time: 00:00:21.548

Memory usage stats:
heap stats:    peak      total      freed    current       unit      count
  reserved:    5.4 GiB    5.4 GiB   56.0 MiB    5.3 GiB                        not all freed!
 committed:    4.7 GiB    6.2 GiB    1.5 GiB    4.6 GiB                        not all freed!

[YamatoSecurity]

Thank you! I will check this on my end as well.

[YamatoSecurity]

I did some benchmarks on my end.
It seems it is 2% slower but uses 2% less memory.
I think I am going to keep it at 0.1.34 for now as there apparently were some bug fixes and 2% is not a huge difference.
I submitted an issue here for now:
purpleprotocol/mimalloc_rust#89
Let's see what the say and if they can fix it.
We might just have to trust upstream MS that whatever improvements they are making are worth the degradation in performance.

[fukusuket]

Thank you for the benchmark and issue report 🙇
Yes, I'll trust upstream improvements :) Close this PR