nixos/nix
1
1
Fork 0
mirror of https://github.com/NixOS/nix.git synced 2025-11-14 22:42:41 +01:00
Code Activity

libstore: Get rid of allocations in printString, allocate 2K bytes on the stack

Browse source 
Looking at perf:

   0.21 │       push   %rbp
   0.99 │       mov    %rsp,%rbp
        │       push   %r15
   0.25 │       push   %r14
        │       push   %r13
   0.49 │       push   %r12
   0.66 │       push   %rbx
   1.23 │       lea    -0x10000(%rsp),%r11
   0.23 │ 15:   sub    $0x1000,%rsp
   1.01 │       orq    $0x0,(%rsp)
  59.12 │       cmp    %r11,%rsp
   0.27 │     ↑ jne    15

Seems like 64K is too much to have on the stack for each invocation, considering
that only a minuscule number of allocations are actually larger than 4K.

There's actually no good reason this function should use so much stack space. Or
use small_string at all. Everything can be done in small chunks that don't require
any memory allocations and use up 2K bytes on the stack.

This patch also adds a microbenchmark for tracking the unparsing performance. Here
are the results for this change:

(Before)

BM_UnparseRealDerivationFile/hello         7275 ns         7247 ns        96093 bytes_per_second=232.136Mi/s
BM_UnparseRealDerivationFile/firefox      40538 ns        40376 ns        17327 bytes_per_second=378.534Mi/s

(After)

BM_UnparseRealDerivationFile/hello         3228 ns         3218 ns       215671 bytes_per_second=522.775Mi/s
BM_UnparseRealDerivationFile/firefox      39724 ns        39584 ns        17617 bytes_per_second=386.101Mi/s

This translates into nice evaluation performance improvements (compared to 18c3d2348f):

Benchmark 1: GC_INITIAL_HEAP_SIZE=8G old-nix/bin/nix-instantiate ../nixpkgs -A nixosTests.gnome --readonly-mode
  Time (mean ± σ):      3.111 s ±  0.021 s    [User: 2.513 s, System: 0.580 s]
  Range (min … max):    3.083 s …  3.143 s    10 runs

Benchmark 2: GC_INITIAL_HEAP_SIZE=8G result/bin/nix-instantiate ../nixpkgs -A nixosTests.gnome --readonly-mode
  Time (mean ± σ):      3.037 s ±  0.038 s    [User: 2.461 s, System: 0.558 s]
  Range (min … max):    2.960 s …  3.086 s    10 runs
This commit is contained in:
Sergei Zimmerman 2025-08-31 00:48:37 +03:00
parent 18c3d2348f
commit e1c9bc0ef6
No known key found for this signature in database
2 changed files with 56 additions and 22 deletions
Download patch file Download diff file Expand all files Collapse all files

29
src/libstore-tests/derivation-parser-bench.cc
View file

@ -28,6 +28,27 @@ static void BM_ParseRealDerivationFile(benchmark::State & state, const std::stri
state.SetBytesProcessed(state.iterations() * content.size());
}
// Benchmark unparsing real derivation files
static void BM_UnparseRealDerivationFile(benchmark::State & state, const std::string & filename)
{
// Read the file once
std::ifstream file(filename);
std::stringstream buffer;
buffer << file.rdbuf();
std::string content = buffer.str();
auto store = openStore("dummy://");
ExperimentalFeatureSettings xpSettings;
auto drv = parseDerivation(*store, std::string(content), "test", xpSettings);
for (auto _ : state) {
auto unparsed = drv.unparse(*store, /*maskOutputs=*/false);
benchmark::DoNotOptimize(unparsed);
assert(unparsed.size() == content.size());
}
state.SetBytesProcessed(state.iterations() * content.size());
}
// Register benchmarks for actual test derivation files if they exist
BENCHMARK_CAPTURE(
BM_ParseRealDerivationFile,
@ -37,3 +58,11 @@ BENCHMARK_CAPTURE(
BM_ParseRealDerivationFile,
firefox,
getEnvNonEmpty("_NIX_TEST_UNIT_DATA").value_or(NIX_UNIT_TEST_DATA) + "/derivation/firefox.drv");
BENCHMARK_CAPTURE(
BM_UnparseRealDerivationFile,
hello,
getEnvNonEmpty("_NIX_TEST_UNIT_DATA").value_or(NIX_UNIT_TEST_DATA) + "/derivation/hello.drv");
BENCHMARK_CAPTURE(
BM_UnparseRealDerivationFile,
firefox,
getEnvNonEmpty("_NIX_TEST_UNIT_DATA").value_or(NIX_UNIT_TEST_DATA) + "/derivation/firefox.drv");