pub const GDN_DECODE_STRIDED_SMEM_CAP_BYTES: u32 = _; // 32_768u32Expand description
2026-09-29: Dynamic shared memory each gdn_decode_f32_strided CTA reserves and does
not use, from GDN_DECODE_STRIDED_SMEM_CAP_MIN_BATCH sequences: it caps the CTAs resident
per SM (3 on GB10’s 100 KB), so a CTA’s 64 KB head state is still in L2 when its update
pass re-reads it; uncapped, the second pass of a wide launch misses L2. Measured
2026-09-29 on dgx2 (gdn_decode_strided_microtest, per layer, uncapped / capped /
per-sequence launches), 32 value heads: B=16 365 / 362 / 428 us, B=64 1904 / 1362 / 1740,
B=128 3857 / 2641 / 3541; 48 value heads: B=16 672 / 507 / 589, B=128 5737 / 3902 / 4766.
The output and state bits do not depend on it.