Alexander Korolev
|
dd86b8a715
dsv41-flash build fixes: cutlass extraction, build tools, NVRTC, no-GPU gate
|
1 hónapja |
Alexander Korolev
|
4eb845dec7
dsv41-flash image tree: tonyd2wild overlay chain as clone-at-pin build
|
1 hónapja |
Alexander Korolev
|
37e4be4a9f
r31 campaign: 4x A/B win (159/176, +10% decode), 2x below baseline, eugr recipe ready
|
1 hónapja |
Alexander Korolev
|
df77baae75
r31 version profile + r31/eugr recipes + night shift 09-10 plan
|
1 hónapja |
Alexander Korolev
|
f4abb53cbf
recipe [env] refs: values can resolve from recipe fields ({ from = "port" })
|
1 hónapja |
Alexander Korolev
|
95dfdf3001
history: note the PORT injection revert
|
1 hónapja |
Alexander Korolev
|
bd82f6f9ae
revert generic PORT injection; PORT is an explicit recipe [env] again
|
1 hónapja |
Alexander Korolev
|
74ad0dbd4b
repo language: history + spec sections to English (per operator)
|
1 hónapja |
Alexander Korolev
|
1c53f3c737
sparkring 4x results + recipe cleanup per operator
|
1 hónapja |
Alexander Korolev
|
87340498fb
sparkring-shared image dir: thin FROM wrapper over the pinned digest
|
1 hónapja |
Alexander Korolev
|
2c8328d8a9
Add sparkring shared-image TP2 recipe (2x profile, digest-pinned)
|
1 hónapja |
Alexander Korolev
|
0b41469beb
report page: update bell — SSE diff of rows snapshots, badge count, dropdown panel with per-update links, click-through to the row; node probe + playwright click-through tests
|
1 hónapja |
Alexander Korolev
|
476b5a5cde
report serve: event-driven refresh — notify watcher (FSEvents/inotify) + 500 ms debounce replaces the 5 s poll; 30 s keepalive as safety net; events filtered to reports/ + recipe.toml
|
1 hónapja |
Alexander Korolev
|
c0409a5a0c
report --serve: live updates over SSE — 5 s refresher re-scans reports/, pushes rows JSON on change; page re-renders keeping view state
|
1 hónapja |
Alexander Korolev
|
709a8ea749
history: production switched to full r28 4x; sparkring recon (DFlash2 retired, native MTP is the path)
|
1 hónapja |
Alexander Korolev
|
0d7c9f334e
upstream issue descriptions: b12x eh_proj loader + DFlash2 dtype (retired in sparkring, native MTP is the path); 386B capacity note
|
1 hónapja |
Alexander Korolev
|
73ac661b42
bench report: 2x r28 on 2nodes-dac — 158/176, DAC == switch (±2%); history: DAC A/B, pty fix, 386B OOM, DFlash2 blocked
|
1 hónapja |
Alexander Korolev
|
87ad741ec8
bench: run tool-eval-bench under a pty (script -q) — rich Live deadlocked on plain stdout pipes after the perf phase (2x lane, 3/3 repro; 4x got lucky)
|
1 hónapja |
Alexander Korolev
|
9d9c038ede
history: 386B OOM-loop required HA power cycle (respawning workers); cluster recovered clean
|
1 hónapja |
Alexander Korolev
|
06bf0e84da
history: 386B Spark model — 4x OOM-proven (101.5 GiB/node weights + double load); needs 5-6+ nodes
|
1 hónapja |
Alexander Korolev
|
0238fe8169
recipes: dflash2-2x BLOCKED on r28 too — DFlash kernel BF16-KV-only vs fp8_ds_mla engine KV (fork-side)
|
1 hónapja |
Alexander Korolev
|
9dcb240b67
bench+history: split-2048/256 env vars — boot clean, 156/176 neutral (KV +2.6%); DFlash2 prereq
|
1 hónapja |
Alexander Korolev
|
7d2bc49082
recipes: test variants — split2048 (Discord env vars, booted, bench running), spark-nvfp4-4x (386B boot test), r28-dflash2-2x (re-test), r28-spark (2x lane)
|
1 hónapja |
Alexander Korolev
|
8bf398fb74
history: b12x loader test blocked on eh_proj routing bug (fork-side)
|
1 hónapja |
Alexander Korolev
|
e6fc89ee5a
history: b12x loader test result — works but eh_proj routing bug blocks it (upstream)
|
1 hónapja |
Alexander Korolev
|
d1f10dd1e7
bench: r28 full rerun 159/176 — 155 vs 160 was run-to-run variance (8 flips), not a regression
|
1 hónapja |
Alexander Korolev
|
ff1f727ae3
recipes: GMU 0.88 in r27/r28 4x+2x (startup gate 107.0 vs 109.5 GiB; KV pinned, capacity unchanged)
|
1 hónapja |
Alexander Korolev
|
3bb263d1af
history: r28 full-stack eval, build_profile feature + TOML trap, memory-gate flake, bench 155
|
1 hónapja |
Alexander Korolev
|
01a63d9cc5
bench: full r28 4x DCP4 — 155/176 (vs r27 160, mixed 159); 8 hard-scenario flips under full-stack noise band
|
1 hónapja |
Alexander Korolev
|
3669194b80
skills: document build_profile + TOML bare-key-after-[[table]] pitfall (r28 half-pin incident)
|
1 hónapja |