Added perfbd.py and block device performance sampling in bench-runner

Browse Source

Based loosely on Linux's perf tool, perfbd.py uses trace output with
backtraces to aggregate and show the block device usage of all functions
in a program, propagating block devices operation cost up the backtrace
for each operation.

This combined with --trace-period and --trace-freq for
sampling/filtering trace events allow the bench-runner to very
efficiently record the general cost of block device operations with very
little overhead.

Adopted this as the default side-effect of make bench, replacing
cycle-based performance measurements which are less important for
littlefs.

...

This commit is contained in:

Christopher Haster

2022-10-13 11:09:26 -05:00

parent 29cbafeb67

commit 3a33c3795b

20 changed files with 2026 additions and 610 deletions

Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL

Download Patch File Download Diff File Expand all files Collapse all files

1252

scripts/perfbd.py Executable file

File diff suppressed because it is too large Load Diff