[PATCH 0/4] perf script: Bounded and lazy symbol loading

Alireza Haghdoost via B4 Relay posted 4 patches 1 week, 2 days ago
There is a newer version of this series
tools/perf/Documentation/perf-script.txt           |  24 +
tools/perf/builtin-script.c                        |  33 ++
tools/perf/tests/shell/script_lazy_load_symbols.sh | 120 +++++
tools/perf/util/dso.c                              |  10 +
tools/perf/util/dso.h                              |  29 ++
tools/perf/util/map.c                              |   9 +-
tools/perf/util/symbol-elf.c                       | 561 +++++++++++++++++++++
tools/perf/util/symbol-minimal.c                   |  11 +
tools/perf/util/symbol.c                           |  87 +++-
tools/perf/util/symbol.h                           |  18 +-
tools/perf/util/symbol_conf.h                      |   2 +
11 files changed, 875 insertions(+), 29 deletions(-)
[PATCH 0/4] perf script: Bounded and lazy symbol loading
Posted by Alireza Haghdoost via B4 Relay 1 week, 2 days ago
perf script loads the entire ELF symbol table of every DSO that appears
in a sample, allocating each symbol into an rb-tree held until process
exit.  Therefore, a large enough profile turns symbol loading into an OOM kill.
This is not scalable for profiling a large cgroup with a lot of large
binaries on a production system with limited free memory.

This series adds two independent, opt-in mechanisms and a leading
regression fix they build on:

  [1/4] Fix a broken "#ifdef ELF_C_READ_MMAP" guard so perf actually
        mmaps ELF files instead of malloc'ing section data. This is a
        standalone regression fix introduced by 22dd1ac91a77.

  [2/4] --max-symbol-bytes <size>: a byte budget on struct symbol
        allocations (and the lazy index) enforced at the ELF symbol
        loader, degrading to [unknown] with a warning past the cap.
        An unbounded profile doesn't just risk OOM-killing itself.  It
        also forces memory pressure on the whole host, pushing the kernel
        to reclaim from co-located latency-sensitive processes.  Capping
        it lets the user bound that footprint up front and choose the
        trade-off explicitly.

  [3/4] --lazy-load-symbols: build a compact per-DSO sorted index and
        resolve only the sampled addresses, reading names via pread()
        from the file's string table at lookup time. On the production fixture,
        peak RssAnon drops from 265 MiB to 39 MiB (6.8x) and wall time
        from 3.1 s to 1.85 s (1.7x). Memory optimizations usually cost
        time; this one does not because lazy loading skips a lot of calloc and
        demangle calls. The output is byte-identical to the default loader.

  [4/4] Documentation and a shell test.

Signed-off-by: Alireza Haghdoost <haghdoost@uber.com>
---
Alireza Haghdoost (4):
      perf symbols: Fix broken ELF_C_READ_MMAP fallback guard
      perf script: Add --max-symbol-bytes to bound ELF symbol memory
      perf script: Add --lazy-load-symbols for lazy symbol loading
      perf script: Document and test --lazy-load-symbols and --max-symbol-bytes

 tools/perf/Documentation/perf-script.txt           |  24 +
 tools/perf/builtin-script.c                        |  33 ++
 tools/perf/tests/shell/script_lazy_load_symbols.sh | 120 +++++
 tools/perf/util/dso.c                              |  10 +
 tools/perf/util/dso.h                              |  29 ++
 tools/perf/util/map.c                              |   9 +-
 tools/perf/util/symbol-elf.c                       | 561 +++++++++++++++++++++
 tools/perf/util/symbol-minimal.c                   |  11 +
 tools/perf/util/symbol.c                           |  87 +++-
 tools/perf/util/symbol.h                           |  18 +-
 tools/perf/util/symbol_conf.h                      |   2 +
 11 files changed, 875 insertions(+), 29 deletions(-)
---
base-commit: aa18964dd64511305de0711fed912054da6f5d18
change-id: 20260915-perf-symbol-memory-send-e7cfca1ac3d9

Best regards,
--  
Alireza Haghdoost <haghdoost@uber.com>