Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.debian.kernel > #90823
| From | Max Mueggler <maxmueggler@gmail.com> |
|---|---|
| Newsgroups | linux.debian.bugs.dist, linux.debian.kernel |
| Subject | Bug#1125667: Major `perf record` performance regression - why is it trying to run addr2line?? |
| Date | 2026-01-15 18:50 +0100 |
| Message-ID | <MdzDP-aMcy-5@gated-at.bofh.it> (permalink) |
| Organization | linux.* mail to news gateway |
Cross-posted to 2 groups.
Package: linux-perf Version: 6.17.13 In previous versions, I was able to run e.g. `perf record --call-graph dwarf,16384 -m 64M -p <PID>` and it would exit fairly quickly after pressing crtl+C. Now, the same command has been running for over an hour after profiling for about 10 seconds! Every few seconds it prints an error: addr2line /path/to/symbols/file.debug: could not read first record The way perf invokes addr2line is known to be very slow, but I have previously only encountered that when running `perf script`. `perf record --call-graph dwarf` used to only record stack dumps, not try and symbolicate them, which other tools, e.g. hotspot or samply, are much faster at doing. I believe `perf record` *shouldn't* be trying to do this on its own, at least without being explicitly requested to, and the manpage makes no mention of it.
Back to linux.debian.kernel | Previous | Next | Find similar | Unroll thread
Bug#1125667: Major `perf record` performance regression - why is it trying to run addr2line?? Max Mueggler <maxmueggler@gmail.com> - 2026-01-15 18:50 +0100
csiph-web