Bench log: first GPU proof on the RTX 5090 and the devnet v4 cut-over; ledger P20

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

docs/bench-log.md

@@ -574,3 +574,40 @@ The long-heal run is the measurement the 3 October harness could not make: the o
 The inclusive comparison at exactly two thirds is pinned by the integer test `3 x signed >= 2 x total` and the unit test (4 of 6, 2 of 3 lock; the 3 October S6B run had already measured the active test passing at exactly 2/3 with 4 of 6 equal voters); the devnet's 4 side sat at 67.9% rather than 66.67% because block counts are Poisson, and locked at the first checkpoint after the cut.
 
 Notes. (1) The v4 node logged "PoW rejected ... by igneum-lottery-v1-bound" about five times a second per node (2,952 lines on n0 over 6A) although the override carries `skip_proof_of_work` and the six miners saw every submission accepted; the window held exactly 1,800 blocks of weight on every node and each side's DAA advanced at 3 per second as planned, so the lines did not move the measurement, but what the v4 pipeline is re-checking there is a question for the consensus engineer before the v4 cut-over (30 of a sample of 200 rejected hashes were later accepted on the same node). (2) The repo harness `tools/finality-attacks/run.mjs` s6 criterion text and the s5 "below the 56.7% floor" pass test are now stale and should read two thirds. (3) Not done: the two-hour presence window and a 30-day window at mainnet length; the first-month gate under the new floor is unchanged (`min_daa` = window). (4) The harness's 60-s heal window (6A, 6B) is shorter than the connection manager's redial backoff for a `--connect` peer after a cut, so a healed proxy does not mean a reconnected n0 inside it; the long-heal run used 300 s and n0 redialled at 59 s after the gate reopened. (5) Stop everything: every node, miner and proxy of the three runs was stopped by the harness at the end of each run; ports 29200 to 29299 were free afterwards (`lsof` 0 listeners).
+
+## 4 October 2026, proving v0 on the RTX 5090: first GPU proof of an Igneum block (WSL2, SP1 6.8.1 cuda)
+
+Machine: the project lead's Windows 11 PC, RTX 5090 (32,607 MiB, driver 617.14), 16 cores and 45 GB visible to WSL2 Ubuntu 24.04, mining
+paused. Package `proving/windows-wsl2` (SETUP-PROVER then PROVE-BLOCK), host `igneum-prove-host` built with the `cuda` feature,
+`SP1_PROVER=cuda`, sp1-gpu-server 6.8.1 on device 0. Fixture `block-78-increment` (chain 4463, 2 transactions, 10 accounts).
+Run id `prove-<pc>-20261004-084838`, log intake id 10154.
+
+| Stage | RTX 5090 | Apple M5 Max CPU (3 October, loaded) |
+|---|---|---|
+| native re-execution | 0.0003 s, state root MATCHES the fixture | 0.0004 s, matches |
+| setup (one per program id) | 23.23 s | 20.55 s on the PC's CPU; Mac not timed separately |
+| execute | 626,876 cycles, prover gas 844,704, 0.19 s, 14 cycles per EVM gas | 626,246 cycles, 0.15 s on the PC's CPU |
+| core proof | prove 1.4 s, 7,317,217 bytes, verify 0.221 s, VERIFIED | prove 22.0 s, 7.3 MB, verify 0.16 s |
+| compressed proof | prove 2.7 s, 1,272,769 bytes, verify 0.038 s, VERIFIED | prove 55.7 s, 1.27 MB, verify 0.03 s |
+
+Reading: 15.7x on core and 20.6x on compressed against a loaded laptop CPU. The block is far below one SP1 shard, so these are
+the fixed per-proof overheads of the proof system on this card; the throughput number needs the larger fixtures (proving e2e
+standard). Post-state root and receipts root identical to the node's on every stage. Two defects, neither in the proof: the
+host aborted (exit 134) AFTER writing and uploading the results, in `sp1-cuda`'s destructor outside a Tokio runtime; and the
+host was silent for ten minutes between the core and compressed stages with the card idle. Ledger P20. Setup on the PC
+needed three package fixes found only by running it on Windows: `protobuf-compiler` in the apt list, the WSL distro check
+(UTF-16 output), the elevated window closing; and WSL2 itself needed `bcdedit /set hypervisorlaunchtype auto` on a PC whose
+BIOS already had SVM on.
+
+## 4 October 2026, devnet v4 cut-over: generator v2, 2/3 floor, three nodes and a seed on a fresh chain
+
+Sequence (BST): seed re-staged from `6457ca95` (build 55 min on the 2-vCPU VM, 08:55 to 09:54); node 1 stopped and the v3
+database moved aside at 10:05:30, `igneumd` v4 up at 10:05:31 with the execution layer; observer peer and `observer.mjs`
+restarted on fresh appdirs at 10:06:10; seed switched with `switch-v4.sh` at 10:06:28 and synced at 10:07:38 (`blocks=2`,
+`peers=1`); Mac Metal miner (generator v2 port, `prepare 1`) first accepted block at 10:07:35; the PC joined at 10:10:55
+(protocol version 12) and its first blocks followed within the minute. Block rate went from about 1 per second (Mac alone,
+difficulty easing 9.1% a block from the 134M genesis value) to about 2 per second with the 5090; the live page followed from
+block 0 with 9 identities after five minutes. First NVIDIA card to mine a generator v2 program; CPU re-check passed on every
+Mac share. One launcher defect found only on Windows: `"$machine:$vendorName"` in `igneum-common.ps1` is a drive-qualified
+variable to PowerShell, so the file failed to parse (fixed, braces). The first finality lock is due at DAA 7,200, two hours
+after the v4 genesis; the first hourly swap at DAA 3,600.

docs/fud-ledger.md

@@ -1460,6 +1460,15 @@ Evidence: `docs/bench-log.md`, 4 October 2026 "execution layer attack fixes" (be
 
 ## Difficulty attack findings (4 October 2026): fixed
 
+### P20. The SP1 GPU client panics on shutdown and the compressed stage waited ten minutes
+"Your first GPU proof run aborted with a core dump. What else aborts?"
+
+Status: Open, found by the team (4 October 2026, morning, first RTX 5090 run through WSL2, SP1 6.8.1).
+
+Answer: Two separate things, neither in the proof. (1) After every proof was written, verified and uploaded, `sp1-cuda`'s client dropped its session key outside a Tokio runtime and panicked in its destructor (`sp1-cuda-6.8.1/src/pk.rs:63`, `client.rs:221`), so the host exited 134 with the results already on disk. Fix in our host: hold a runtime for the client's lifetime or drop the proof system inside one. (2) Between the core proof (08:49:10 UTC) and "Proving with mode: Compressed" (08:59:02 UTC) the host was silent for ten minutes while the card was idle; the compressed mode's recursion setup on first use is the suspect, and the second run must time it. Both go on the proving e2e benchmark standard as fixed overheads to measure, not hide.
+
+Evidence: `docs/bench-log.md` "4 October 2026, proving v0 on the RTX 5090"; results file `block-78-increment-cuda-20261004-084838.json` on the PC; log intake id 10154. Experiment: O-7.4 (re-run with `--mode compressed` alone and a timestamp per stage).
+
 ### M23. Forge timestamps inside the rules and the controller mines you a 10x difficulty for free
 "Your fast controller clamps every solvetime to 20 s both ways and says the next honest block cancels a forged one. Good: I stamp every block of mine at the earliest the past median allows, the honest block after me gets clamped to +20 s, the pair sums to zero, and your lanes measure 1 - 2a(1 - a) of real time at my share a. With half the hashrate your chain runs at a fifth of its rate and 9.9x the difficulty, on no extra hash. Kaspa's window only drifts 5 to 11%. And while I am at it, 85 blocks a second of PoW-less input drives your target below 2^64 and `calc_work` panics the node."
 

site/build.mjs

@@ -151,6 +151,7 @@ if (existsSync(join(docs, 'bench-log.md'))) {
     ['Igneum-node devnet v2', 'Finality rule v2 live on a four-miner test network'],
     ['Execution layer devnet v3', 'EVM execution layer: identical state on three nodes'],
     ['Weak-program census', 'Census of 400,000 programs: redundant loads found'],
+    ['Proving v0 on the RTX 5090', 'First GPU proof of an Igneum block: 1.4 s on an RTX 5090'],
     ['Proving v0', 'First SP1 proof of an Igneum block, on a laptop CPU'],
     ['Windows node package', 'Windows node package: cross-compiled, two-peer sync'],
     ['R3.26 / M15', 'Cache-build attack closed: 10.6 s of rebuilds to 14 ms'],
@@ -170,6 +171,7 @@ if (existsSync(join(docs, 'bench-log.md'))) {
     ['First prototype of the random-program', 'First prototype of the lottery hash on Apple silicon'],
     ['Igneum named', 'Igneum named. Litepaper, brand and wallet designs published'],
     ['Finality floor', 'Finality floor raised to two thirds of all weight'],
+    ['Devnet v4 cut-over', 'Devnet v4 live: generator v2, two-thirds floor, fresh chain'],
   ];
   function shortTitle(t) {
     for (const [k, v] of SHORT) if (t.startsWith(k)) return v;
@@ -189,7 +191,7 @@ if (existsSync(join(docs, 'bench-log.md'))) {
   const j = JSON.parse(readFileSync(jp, 'utf8'));
   const seen = new Set(j.log.map(e => e.date + '|' + e.text));
   for (const e of entries.reverse()) { const k = e.date + '|' + e.text; if (!seen.has(k)) { j.log.unshift(e); seen.add(k); } }
-  for (const e of j.log) if (!e.short) e.short = shortTitle(e.text);
+  for (const e of j.log) e.short = shortTitle(e.text);
   j.log.sort((a, b) => (a.date < b.date ? 1 : a.date > b.date ? -1 : 0));
   j.log = j.log.slice(0, 40);
   j.updated = j.log[0] ? j.log[0].date : j.updated;

site/journey.json

@@ -50,6 +50,16 @@
     }
   ],
   "log": [
+    {
+      "date": "2026-10-04",
+      "text": "Proving v0 on the RTX 5090: first GPU proof of an Igneum block",
+      "short": "First GPU proof of an Igneum block: 1.4 s on an RTX 5090"
+    },
+    {
+      "date": "2026-10-04",
+      "text": "Devnet v4 cut-over: generator v2, 2/3 floor, three nodes and a seed on a fresh chain",
+      "short": "Devnet v4 live: generator v2, two-thirds floor, fresh chain"
+    },
     {
       "date": "2026-10-04",
       "text": "Generator version 2 adopted: exact load count, fresh-source loads, program acceptance; every vector re-cut, three workers re-checked, 20,000-program census, devnet-v4 binaries rebuilt",