perf(loki.source.file): Reduce allocations during file reading

[kalleep]

Pull Request Details

There is a lot of allocations being performed while reading files. I can clearly see this being a hot spot in our internal clusters.

Two of the biggest issues are:

1. We allocate a new pending buffer as soon as we have completed a line
2. We use ReadBytes api causing new allocation for every call that we just move to pending

Instead we can use ReadSlice, this api will not allocate but return a slice pointing to the internal buffer while still advancing read position. We copy it into pending so this is fine and we remove the intermediate allocation done by ReadBytes.

We can also reset length of pending and this buffer. This is fine because we never read past newline boundary.

Issue(s) fixed by this Pull Request

Notes to the Reviewer

PR Checklist

• Documentation added
• Tests updated
• Config converters updated

[kalleep]

Benchmark diff against main

goos: linux
goarch: amd64
pkg: github.com/grafana/alloy/internal/component/loki/source/file/internal/tail
cpu: AMD Ryzen 7 8845HS w/ Radeon 780M Graphics
        │ bench_old.txt │            bench_new.txt             │
        │    sec/op     │    sec/op     vs base                │
File-16   1287.0µ ± 19%   410.5µ ± 57%  -68.10% (p=0.000 n=10)

        │ bench_old.txt │            bench_new.txt             │
        │     B/op      │     B/op      vs base                │
File-16   5057.0Ki ± 0%   557.0Ki ± 0%  -88.99% (p=0.000 n=10)

        │ bench_old.txt │            bench_new.txt            │
        │   allocs/op   │  allocs/op   vs base                │
File-16     4.022k ± 0%   2.022k ± 0%  -49.73% (p=0.000 n=10)

[Copilot]

Pull request overview

Reduces allocation pressure in the Loki file tailer’s line-reading path by reusing buffers and avoiding per-read allocations from bufio.Reader.

Changes:

• Switch line reading from ReadBytes to ReadSlice and adjust looping/error handling to reduce intermediate allocations.
• Reuse the pending buffer by resetting length instead of allocating new slices on each consumed line/flush.
• Update test helpers to accept testing.TB and add a benchmark to track allocations.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 3 comments.

File	Description
internal/component/loki/source/file/internal/tail/reader.go	Uses ReadSlice and buffer resets to reduce allocations while reading/consuming lines; fixes an error-return bug in reset.
internal/component/loki/source/file/internal/tail/file_test.go	Refactors helper signatures for TB compatibility and adds a benchmark for allocation tracking.

[grafana-alloybot]

⚠️ Automatic backport to release/v1.13 failed

The automatic backport for this PR failed, likely due to merge conflicts. Perform the backport manually:

git fetch origin main release/v1.13
git checkout release/v1.13
git pull
git checkout -b backport/pr-5405-to-v1.13
git cherry-pick -x 899467aefc97088471f79e03bcfc7a226e420cdd
# If conflicts exist, fix them, then:
git add .
git commit
# If no [more] conflicts, then:
git push -u origin backport/pr-5405-to-v1.13

Then create a PR from backport/pr-5405-to-v1.13 to release/v1.13 with the title:

perf(loki.source.file): Reduce allocations during file reading [backport]