frank-primary: drop --expert-precise-memory (boot crash loop with bf16)

NRestarts climbed to 3, each attempt ~6.7 min then 'Deactivated successfully'
(API server drains after an EngineCore fatal error; Restart=always turned the
silent death into a visible loop). The arena is device-resident, so the budget
decides it: weights 6.28 + pool 6.80 (bf16 2 x 262144) + graphs 0.55 + arena
16.65 (precise overhead 5.25) + non-torch 1.45 = 31.73 of 31.86 GiB -> ~0.13
GiB free at boot, and graph capture is what allocates transiently.

Dropping the flag restores the 7.60 GiB default overhead: arena 14.31 GiB,
total ~29.4 GiB, ~2.4 GiB free. bf16 2 x 262144 (Frank's KV change) stays.
The serve script keeps the opt-in flag; the seam now asserts it is ABSENT.
This commit is contained in:
Frank committed 2026-10-01 14:16:47 -07:00
1 parent db54fa3af1
commit d3d180fd3f
3 files changed
+48 -9

No files matched your search

+4 -2
View File
@@ -110,8 +110,11 @@ DOWNLOAD="$REPO_ROOT/scripts/download-frank-primary.sh"
grep -q '^FRANK_PRIMARY_UTIL=0.98$' "$ENV_FILE"
grep -q '^FRANK_PRIMARY_KV_BYTES=7300000000$' "$ENV_FILE"
grep -q '^FRANK_PRIMARY_MNBT=2048$' "$ENV_FILE"
grep -q '^FRANK_PRIMARY_PRECISE_MEMORY=1$' "$ENV_FILE"
grep -q '^FRANK_PRIMARY_KV_DTYPE=auto$' "$ENV_FILE"
# --expert-precise-memory is NOT set: with bf16 2 x 262144 it put the boot
# 0.13 GiB from the 31.86 GiB ceiling (arena 16.65 device-resident) and
# crash-looped at graph capture. The serve script keeps the opt-in flag.
! grep -q '^FRANK_PRIMARY_PRECISE_MEMORY=' "$ENV_FILE"
grep -q '^FRANK_PRIMARY_GATHER_CHUNKS=32$' "$ENV_FILE"
grep -q '^FRANK_PRIMARY_GATHER_LANES=64$' "$ENV_FILE"
! grep -q '^FRANK_PRIMARY_SPEC=' "$ENV_FILE"
@@ -139,7 +142,6 @@ DOWNLOAD="$REPO_ROOT/scripts/download-frank-primary.sh"
# moves the plan's `room` term and therefore the expert arena: ~1,025
# more slots at 2.344 MiB each.
grep -q -- '${FRANK_PRIMARY_PRECISE_MEMORY:+--expert-precise-memory}' "$SERVE"
grep -q '^FRANK_PRIMARY_PRECISE_MEMORY=1$' "$ENV_FILE"
}
@test "tools and reasoning are wired — the client contract is served" {