forked from coq/coq
coq dev doc profiling.md
3.2 kB · 100 lines
Markdown

How to profile Rocq? #

I (Pierre-Marie Pédrot) mainly use two OCaml branches to profile Rocq, whether I want to profile time or memory consumption. AFAIK, this only works for Linux.

Time #

In Rocq source folder:

opam switch 4.14.0+trunk+fp
make world
perf record -g _build/install/default/bin/coqc file.v
perf report -g fractal,callee --no-children

To profile only part of a file, first load it using

bin/coqtop -l file.v

and plug into the process

perf record -g -p PID

Per-component flame graphs #

I (Andres Erbsen) have found it useful to look at library-wide flame graphs of rocq time consumption. As the Ltac interpreter stack is reflected in the OCaml stack, calls to the same primitive can appear on top of multiple essentially equivalent stacks. To make the profiles more readable, one could either try to edit the stack trace to merge "equivalent" frames, or simply look at the aggregate profile on a component-by-component basis. Here is how to do the second for the standard library (example output).

#!/usr/bin/env bash
make clean
make states
perf record -F99  `# ~1GB of data` --call-graph=dwarf -- make world
perf script --time '0%-100%'  |
        stackcollapse-perf.pl |
        grep Coqtop__compile |
        sed -rf <(cat <<'EOF'
                s/;caml/;/g
                s/_[0-9]*;/;/g
                s/Logic_monad__fun;//g
                s/_apply[0-9];//g
                s/;System/@&@/
                s/;Hashcons/@&@/
                s/;Grammar/@&@/
                s/;Declaremods/@&@/
                s/;Tactics/@&@/
                s/;Pretyping/@&@/
                s/;Typeops/@&@/
                s/;Reduction/@&@/
                s/;Unification/@&@/
                s/;Evarutil/@&@/
                s/;Evd/@&@/
                s/;EConstr/@&@/
                s/;Constr/@&@/
                s/;Univ/@&@/
                s/;Ugraph/@&@/
                s/;UState/@&@/
                s/;Micromega/@&@/
                s/;Omega/@&@/
                s/;Auto/@&@/
                s/;Ltac_plugin__Tacinterp/@&@/
                s/;Ltac_plugin__Rewrite/@&@/
                s/[^@]*@;([^@]*)@/\1;\1/
                s/@//g
                :a; s/;([^;]+);\1;/;\1;/g;ta
EOF
        ) |
        flamegraph.pl

Memory (memtrace) #

memtrace is a client library for OCaml's Memprof statistical memory profiler.

See this blog post for more details: https://blog.janestreet.com/finding-memory-leaks-with-memtrace/

To profile a file, you need to install the memtrace library, then recompile Rocq. We also recommend you make a copy of the .v file (if working on the stdlib to avoid issues with artifacts.

The following command sequence will do all that:

opam install memtrace
make world # to build deps
cp theories/Corelib/Classes/RelationClasses.v /tmp
MEMTRACE=trace-byte.tcr dune exec -- rocq c /tmp/RelationClasses.v
memtrace-viewer trace-byte.tcr