marius/Open_Ultra_SAST
1
0
Fork 0
mirror of https://github.com/norandom/OpenUltraSAST.git synced 2026-10-23 06:59:28 +00:00
Code Issues Packages Projects Releases Wiki Activity

spec(pre-push-safety-net): record the M2 direction, do not rebuild a graph per push

Browse Source 
Both attacks on engine startup are measured and neither reaches the gate. The
session is equivalent for every shipped query kind and helps a long transaction,
326 s to 243 s, but its own 9 to 13 second startup is a third of a 30-second budget
so it usually declines to start. Removing JavaScript's second engine start is real,
11.0 s against 21.0 s on identical graphs, and was reverted for an asymmetric
census gate that made 12 transactions pay two starts.

What the profile names is not startup. A changed-code transaction reaches its
deadline inside the build: head build 25.2 s of 30 s, query time 0.0 s. The cost is
building a fresh whole-repository graph for every pushed revision.

Direction accepted: reuse the graph across revisions rather than optimise the
rebuild. M2 stays unmet, the gates stay unchanged, and the work is not scheduled
here. Group 12 returns to security capability, which is where the remaining release
risk sits.

Co-Authored-By: Claude Opus 5 (1M context) <[email protected]>
Claude-Session: https://claude.ai/code/session_01EmM1j46qgKMh3zqD1QMwEo
This commit is contained in:
Marius Ciepluch
2026-09-17 09:53:35 +02:00
co-authored by Claude Opus 5
parent ed0f12b593
commit 6653b6aca8
2 changed files with 20 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files
.kiro/specs/pre-push-safety-net/release-milestones.md
+15 -1
View File
@@ -101,7 +101,21 @@ comparison, admission, the recorded-veto flag and reporting. Requirements: 1.1
## M2 — The same useful analysis fits the push budget
Status: **PENDING M1**.
Status: **MEASURED AND UNMET, 2026-09-17.** Two attacks on engine startup were measured and
neither reaches the gate. A transaction-owned session is implemented and proven equivalent for
every shipped query kind, and it amortizes startup only across a long transaction: 326 s to 243 s
under a development budget, while under the 30-second deadline its own 9 to 13 second startup
means it usually declines to start. Removing JavaScript's redundant second engine start is a real
11.0 s against 21.0 s on identical graphs, but the attempt was reverted for an asymmetric census
gate that made 12 profile transactions pay two engine starts.
The profile names the actual cost: a changed-code transaction reaches its deadline inside the
build, head build 25.2 s of 30 s with query time 0.0 s. It is building a fresh whole-repository
graph for every pushed revision.
**Direction accepted 2026-09-17: do not rebuild a whole graph per push. Reuse the graph across
revisions instead of further optimising the rebuild.** The 30-second and 2-second gates are
unchanged; this milestone is not met, and the work is not scheduled here.
- Use the successful M1 case as a correctness baseline for a transaction-owned Joern
session within the existing backend. The measured census probe supports investigating