hotspot VS profiler

perf with Hotspot 👌

If you’re able to use perf on Linux, I would recommend hotspot for visualizing the results.

I can recommend you using the https://github.com/KDAB/hotspot/ tool whenever you want to study performance.

Every Linux C/C++/Rust developer should know about https://github.com/KDAB/hotspot. It's convenient and fast. I use it for Rust all the time, and it provides all of these features on the back of regular old `perf`.

Flamegraphs alone aren't a full picture of what your application is doing, but it can give you hints as to where to look. Another tool I often use is Hotspot which can open the perf.data file and provide more options for filtering and digging into the gathered data beyond the single flamegraph.

Ignore the command, it's just a placeholder to get meaningful values. The -d flag adds basic cache events, by adding another -d you also get load and load miss events for the dTLB, iTLB and L1i cache.

But as mentioned, you can instrument any event supported by your system. Including very obscure events such as uops_executed.cycles_ge_2_uops_exec (Cycles where at least 2 uops were executed per-thread) or frontend_retired.latency_ge_2_bubbles_ge_2 (Retired instructions that are fetched after an interval where the front-end had at least 2 bubble-slots for a period of 2 cycles which was not interrupted by a back-end stall).

You can also record data using perf-record(1) and inspect them using perf-report(1) or - my personal favorite - the Hotspot tool (https://github.com/KDAB/hotspot).

Sorry for hijacking the discussion a little, but I think perf is an awesome little tool and not as widely known as it should be. IMO, when using it as a profiler (perf-record), it is vastly superior to any language-specific built-in profiler. Unfortunately some languages (such as Python or Haskell) are not a good fit for profiling using perf instrumentation as their stack frame model does not quite map to the C model.

> [...] how Perf compares to vendor tools like vTune [...] ?

Regarding the hardware events that Perf can capture on x86, it has pretty much all of them. So it should be equivalent to vTune for all practical purposes.

The big difference is in the UI -- or absence thereof. Perf is a low-level tool and its output is mostly text files. There is a curses-based TUI for perf-report (and even gtk version, but it is essentially the same as the TUI, just using GTK2 widgets), but that's about it.

By contrast, vTune comes with a heavy (electron-based?) GUI and is quite helpful in guiding beginners, with many graphs and explanations.

Of course, one can (and is expected to) complement Perf with an assortment of tools that process its output for visualization. For example, the flamegraph [1] and heat map [2] tools described in the article. But also KDAB hotspot [3] or HPerf for a vTune-style perf-report.

[1] https://github.com/brendangregg/FlameGraph

[2] https://github.com/brendangregg/HeatMap

[3] https://github.com/KDAB/hotspot

[4] https://www.poirrier.ca/hperf/

https://github.com/KDAB/hotspot might contain parsing code you could use as an example (other than perf script). It always accepts raw perf.data, and there doesn't seem to be a way to feed it the output of perf script, so it might be parsing it directly instead of calling perf script.

Probably your best bet is to use the Firefox Profiler once it starts hogging the CPU to try to see what it's doing. I'd be happy to take a look at the result although I don't have a lot of experience at diagnosing performance problems; hopefully the profile will make it obvious what's going on.

I tried using the https://profiler.firefox.com/ and running with:

You can try to reproduce while taking a profile using the Firefox profiler, then share it in the #perf:mozilla.org room on Mozilla's Matrix instance.

Run a profiling session for ~30 sec when you notice something is running slower than you expect, see https://profiler.firefox.com/. Share results here if you want help interpreting them.

Next time, send a SIGABRT to the main process (kill -6 $(pidof firefox)) and use the resulting backtrace (you will find a link to the crash report in about:crashes) to file a bug. A performance profile could be useful, but you'll need to know first what to capture. This seems like a graphics issue, so try the Graphics preset. Share a link to your report.

https://profiler.firefox.com will be useful if you can capture and share it to developers.

If you can reproduce it in safe mode and a clean profile, capture a performance profile using the clean profile while trying to load a problematic webpage, don't remove any information from it, and share it here (although it's typically not easy for users to analyse them), or file a bug with it attached.

Next time, capture a performance profile using the Graphics preset, and kill the browser by entering kill -6 $(pidof firefox) in a shell. This will create a bug report (unless debian disables the crash reporter) that would indicate where Firefox is stuck at. You might need to use an official build to get useful results.