enricobuehlerandClaude Fable 5 ae67315804 perf(encode/nvenc-linux): LN3 — pipelined-retrieve escalation replaces the depth-1 async foot-gun ...

PUNKTFUNK_NVENC_ASYNC gains a real tri-state: 1 = always (as before, now
documented as ~+1 tick at depth-1), 0 = never (vetoes escalation), unset =
ADAPTIVE — off until the session loop's cadence-overrun detector escalates.

The host loop's adaptive-depth leaky bucket grows a second stage: once the
capturer's depth is maxed (Linux portal is permanently depth-1), it asks
the encoder for pipelined retrieve via the new Encoder::set_pipelined hook
(asked exactly once; default impl declines, Windows untouched).

nvenc_cuda engages at a safe point via a clean session rebuild WITHOUT the
IO-stream binding: with input==output stream bound, later stream work
waits on prior encode completions and would serialize a pipelined session
— stream-ordered submit and two-thread retrieve are mutually exclusive.
The ordered gate now also requires async_rt absence (belt-and-braces for
the runtime switch). Re-open's first frame is the standard session IDR.

On-hardware test: escalate mid-session → retrieve thread live, binding
gone, all AUs deliver, first post-escalation AU is the IDR.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

2026-07-20 20:10:24 +02:00

..

src

perf(encode/nvenc-linux): LN3 — pipelined-retrieve escalation replaces the depth-1 async foot-gun

2026-07-20 20:10:24 +02:00

Cargo.toml

perf(pyrowave): elevated GPU scheduling + global-priority encode queue

2026-07-19 01:02:01 +02:00