You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: README.md
+1-1Lines changed: 1 addition & 1 deletion
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -497,7 +497,7 @@ For reporting bugs, please include `-ddd` output in the issue.
497
497
498
498
## Performance
499
499
500
-
`mpytool` uses optimized chunked transfer with automatic compression, which allows copying files very quickly. See [README_BENCH.md](README_BENCH.md) for detailed benchmarks.
500
+
`mpytool` uses optimized chunked transfer with automatic compression, which allows copying files very quickly. See [README_bench.md](README_bench.md) for detailed benchmarks.
Copy file name to clipboardExpand all lines: README_bench.md
+22Lines changed: 22 additions & 0 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -66,3 +66,25 @@ Higher speed achieved by automatic compression of text-based files during transf
66
66
-**Skip unchanged: 1.5x - 2.9x faster** than mpremote
67
67
68
68
Many more test scenarios could be designed (different file types, sizes, mixed workloads), but in most cases mpytool would be **at least 2x faster** than mpremote.
69
+
70
+
## Mount VFS Benchmark
71
+
72
+
Reading files from mounted VFS (PC filesystem accessible on device).
73
+
74
+
### RP2040 - USB-CDC - MacOS
75
+
76
+
| Test | mpytool | mpremote | Speedup |
77
+
|------|---------|----------|---------|
78
+
| Read 50 x 2KB files |**1.5s**| 3.9s |**2.5x**|
79
+
| Read 1 x 100KB file |**1.0s**| 1.1s |**1.1x**|
80
+
81
+
### ESP32-C6 - USB-CDC - MacOS
82
+
83
+
| Test | mpytool | mpremote | Speedup |
84
+
|------|---------|----------|---------|
85
+
| Read 50 x 2KB files |**0.6s**| 3.2s |**5.6x**|
86
+
| Read 1 x 100KB file |**0.3s**| 0.5s |**1.8x**|
87
+
88
+
### Summary
89
+
90
+
Batch LISTDIR (1 RTT) vs iterative ilistdir (N+1 RTT) makes significant difference when opening many files. The speedup is more pronounced on faster CPUs (ESP32-C6 160MHz vs RP2040 133MHz).
0 commit comments