perf: round 12 — auth write-path narrowing, fused quota gate, moka blob-cache index, media single-read, sized listing JSON

Benchmark-gated round (benches/ROUND12.md; every change ships with a
BEFORE/AFTER harness + equivalence gates, one candidate rejected by its
own bench):

DB / query shapes (bench_round12_queries):
- NC sharee search: username-only projection instead of the 21-column row
  (incl. the <=512 KiB avatar) per match, + gin_trgm_ops indexes on
  auth.users for the leading-wildcard ILIKE (4.98x; 54.7x with index).
- Password login: delete the redundant full-row update_user — create_session
  already stamps last_login_at in its own txn (4.45x per login).
- Email-verified stamp: narrow conditional UPDATE (8.9x); OIDC repeat login
  now compares profile state in memory and issues ZERO queries when nothing
  changed (was: full 17-column rewrite per login).
- Refresh rotation: revoke+insert+stamp fused into one transaction via new
  rotate_session port method (1.18x).
- WOPI CheckFileInfo / authorize_wopi_access: require(Read) + get_file +
  check(Update) overlapped with tokio::join!, original result precedence
  (cold 1.34x).
- Upload quota gate: user-envelope + drive-cap checks fused into ONE
  round-trip (check_upload_quotas) — the NC chunked PUT pays this per
  chunk (1.81x, 2 -> 1 queries/chunk); shared verdict evaluators keep
  error shapes byte-identical.

CPU / allocs (bench_round12_micro):
- sized_json: pre-sized listing serialization replacing axum Json's 128 B
  seed + doubling-realloc chain on files/folder-resources/photos/search
  responses (1.40x, 13 -> 2 allocs per 500-row page; byte-identical).
- Security headers: 4 SetResponseHeaderLayer folded into the CSP middleware
  pass (5 layers -> 1; 1.43x per request, -26 allocs; header set gated
  byte-identical incl. 304s).
- Media capture-metadata: single-read extraction — nom-exif now parses the
  buffer kamadak already read (zero-copy Bytes) and videos open once with a
  kind() dispatch; per-image opens 2-3 -> 1 (1.44x warm geomean, 1.6-3.2x
  cold cache; extraction outputs gated identical incl. the MIME-mislabel
  track fallback).
- Chunked-upload session ops: owner gate folded into the operation's own
  DashMap lookup + stack-encoded uuid compare (5 -> 3 lookups, -2 allocs,
  1.28x per chunk).

Blob cache (bench_blob_cache_index + round-3 regression guard):
- CachedBlobBackend index: tokio::sync::Mutex<LruCache> -> moka::sync::Cache
  with byte weigher. The mutex serialized every cached chunk read and scaled
  NEGATIVELY (2.08 -> 1.07 Mops/s from 1 -> 2 readers); moka probes are
  lock-free (2.17x at K=2). Byte budget now enforced by moka (manual
  current_size + collect_evictions machinery deleted); eviction listener
  unlinks size-evicted files only (Replaced entries keep their file —
  gated). Single-flight miss gate unchanged (16 concurrent misses -> 1
  fetch re-verified via the round-3 harness).
- put_blob now populates the cache BEFORE the inner backend consumes the
  source file (the old order failed 100% of the time — local renames,
  S3/Azure delete the source — so the first read after a whole-file put
  re-downloaded from the remote); inner-put failure invalidates the entry.

Frontend (vitest gates):
- List-view thumbnails request the 150px icon rendition instead of 400px
  preview into a 40px slot (~7.1x fewer pixels, ~4-5x fewer bytes per
  thumbnail across list views); grid keeps preview.

Rejected by its own bench (kept as evidence in bench_round12_micro §2):
- Single-pass compression predicate: the monomorphized And-chain already
  costs ~4.6 ns / 0 allocs total; the fused node measured within noise.

New migration: 20260719000000_users_search_trgm.sql (trgm indexes).
Deferred with prepared design: grouped file/grid view virtualization
(single-VirtualRows flatten, the photos pattern) — next round's headline.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BfidAJD5AHw23jtvBUNamB
This commit is contained in:
Claude
2026-07-19 01:32:00 +00:00
parent a793cd62eb
commit 50eca0627f
33 changed files with 3989 additions and 410 deletions
+54
View File
@@ -0,0 +1,54 @@
//! Pre-sized JSON responses for listing endpoints.
//!
//! `axum::Json` serializes into a `BytesMut::with_capacity(128)` — a 500-row
//! listing grows that seed through ~11 doubling reallocations, memcpy-ing
//! ~1.3× the payload on every hot listing response (files, folder
//! resources, photos timeline, search). `sized_json` serializes into one
//! right-sized `Vec` instead: 2 allocations total and no copy chain
//! (benches/ROUND12.md §M1, 1.40x / −11 allocs on a 500-row page).
//!
//! The per-row estimates are calibrated against the serialized DTOs (a
//! realistic `FileDto` row measures ~380 B). Underestimates cost one extra
//! doubling — still far better than the 128-byte seed; overestimates waste
//! transient capacity only (the buffer is freed after the response).
use axum::http::{HeaderValue, StatusCode, header};
use axum::response::{IntoResponse, Response};
use bytes::Bytes;
use serde::Serialize;
/// Serialized size estimate for one file/folder row (FileDto ≈ 380 B).
pub const EST_ROW_BYTES: usize = 384;
/// Serialized size estimate for one wrapped resource row (PhotoDto /
/// FolderResourcesDto items carry a FileDto plus wrapper fields).
pub const EST_WRAPPED_ROW_BYTES: usize = 448;
/// Serialize `value` into a single pre-sized buffer and wrap it as an
/// `application/json` response — drop-in for `Json(value).into_response()`
/// (byte-identical body, gated in `bench_round12_micro` §1), minus the
/// doubling-realloc chain.
pub fn sized_json<T: Serialize>(estimated_bytes: usize, value: &T) -> Response {
let mut buf = Vec::with_capacity(estimated_bytes.max(128));
match serde_json::to_writer(&mut buf, value) {
Ok(()) => (
StatusCode::OK,
[(
header::CONTENT_TYPE,
HeaderValue::from_static("application/json"),
)],
Bytes::from(buf),
)
.into_response(),
// Mirror axum's Json error arm: 500 + plain-text serializer error.
Err(err) => (
StatusCode::INTERNAL_SERVER_ERROR,
[(
header::CONTENT_TYPE,
HeaderValue::from_static("text/plain; charset=utf-8"),
)],
err.to_string(),
)
.into_response(),
}
}