Let NX2NAVI_GPA_TWIN_MAX_KB keep the device address table's second writes to small pools, to measure what the twins cost.

A shader's store or atomic through the table goes to a pool's VRAM mirror and again to its guest copy, which on a
discrete GPU crosses PCIe. On PCIe 3.0 (RTX 4090, Kokiri Forest, DLSS Quality) turning the twins off took the GPU from
16 to 13.1 ms a frame; the writes that went wrong without them were foliage counts and records in pools of 20 and
44 KiB. gpa_twin_max_kb = n gives a twin only to mirrors of pools up to n KiB. A larger mirror's record names the mirror
twice, so its second write goes to the table's scratch page, its buffer fills stay in VRAM (vk_mem_guest_twin), and it
is never evicted, since guest memory lacks what the shaders wrote into it. The mirror report counts the mirrors made
without a twin. Default 0: every mirror has one, as before.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
chrg6
2026-11-09 12:42:39 -05:00
parent c29d1b8dbd
commit a720ca879c
3 changed files with 51 additions and 10 deletions

View File

@@ -565,3 +565,8 @@
# page the game wrote; 0 (default) copies only the written pages there, which saved 47-57 MiB of copies a frame in
# Kokiri Forest.
# flush_upload = 0
# For measuring: only mirrors of the game's memory pools up to this many KiB send what shaders write through the
# device address table to game memory as well; larger mirrors keep it in video memory and are not evicted. On PCIe 3.0
# those second writes cost about 3 ms of GPU time a frame. 0 (default): every mirror sends them. A pool that needs them
# and is left without shows as foliage stretched into columns or flickering.
# gpa_twin_max_kb = 0