Repository navigation
Conversation
- Size default memory watchdog limit dynamically from hw.memsize on Mac (48 GB on 64 GB, 28-40 on 36-48 GB, 12 on 16 GB; DM_MEM_CAP_GB overrides) - Raise texture view cache associativity from 4 to 16 ways (VC_WAYS = 16), preventing view thrashing in games with high view turnover - Group direct and staging texture uploads in batches of 32 when outside render command buffers, avoiding per-texture synchronous flush freezes
…hdog, 16-way view cache, and texture upload batching Co-Authored-By: Claude Sonnet 5.5 <[email protected]>
…batched uploads The PR sized the memory watchdog's default from host RAM (12 GB on a 16 GB Mac, down from 24). The watchdog exits the game, and Ocarina of Time is resident at 10.9-11.4 GB on a 16 GB Mac (game note), so the new cap would end runs that used to continue. DM_MEM_CAP_GB still overrides either way. Co-Authored-By: Claude Sonnet 5.5 <[email protected]>
seedds#1901 seedds#1459 seedds#1928 seedds#1942 seedds#1940 seedds#1952 seedds#1611 seedds#2065 seedds#2083 seedds#2008 seedds#2079 seedds#2022 seedds#2075 seedds#2077 seedds#2020 seedds#2021 onto main 04811a6 Co-Authored-By: Claude Fable 5.2 <[email protected]> Claude-Session: https://claude.ai/code/session_014eD437mxrMnW2UwLZhpwc7 Upstream pull requests applied in this commit (their authors' work, brought onto this tree with a three-way merge; the description is in each PR): - seedds#1459 by PromethOps (open): DM_UPSCALER=metalfx takes agl's agl_ndsr_vr.pck networks too, and a variant with no curve and no power is drawn as one colour space: Ocarina of Time's upscaler is MetalFX's - seedds#1611 by Esteban-V (open): An argument block whose only change is a uniform buffer is patched instead of encoded again - seedds#1774 by uncraftablexd (closed): Compute threads that exit before a barrier keep arriving at it, and every shader loop has a cap (GPU hang defences), chain change - seedds#1824 by cocoman93 (open): Clamp texel fetches past a sampled 2D texture's edge: no hard edge in Ocarina of Time's mist - seedds#1901 by FifteenPen (open): The game's compute dispatches with no game barrier between them run side by side (profile key dispatch_overlap, on for Ocarina of Time): Kokiri Forest GPU p50 15.63 → 14.76 ms - seedds#1928 by FeReuss (open): Ocarina of Time: enc_thread on, the replay encodes on a thread of its own - seedds#1929 by FeReuss (open): Unused border corrections are compiled out of pipelines that bind no sampler with a twin (function constant DMSH_BORDER_FC) - seedds#1930 by FeReuss (open): A float add to storage memory with no result read is one native atomic add instead of a compare-and-swap loop - seedds#1940 by uncraftablexd (open): A prewarm list's lines after its #@after_first_frame line are built after the first frame, in the background, instead of in the prepare step (issue seedds#1747) - seedds#1942 by uncraftablexd (open): Frame step: late frames with the GPU idle do not step up, and a longer hold the GPU does not explain is probed - seedds#1952 by AdyaGMD (open): Compute moves ahead of scene passes past occlusion queries and conditional rendering; clears wait across dispatches - seedds#2008 by bquilty25 (open): nvn: reactive memory watchdog, 16-way view cache, and texture upload batching - seedds#2020 by padsiwono (open): Ocarina of Time v131072: Output Native draws the full output size instead of the game's 80 % frame (`DM_OUTPUT`) - seedds#2021 by padsiwono (open): Ocarina of Time v131072 renders at the launcher's Resolution (`DM_RENDER_SIZE`), docked - seedds#2022 by sleepyeyedjoe (closed): Ocarina of Time: large models no longer go black for one frame twice, 4–5 frames apart, when the game streams new textures into them (a texture ID a shader has read keeps its read mark when it gets another texture) - seedds#2030 by humanperzeus (open): A NaN colour at alpha 0 is returned as 0, as NVIDIA's blender takes a zero factor: Ocarina of Time's white clouds over the ferns are gone - seedds#2035 by Know-It-All-Brother (open): The shader build runs as worker processes, and no shader waits for a batch (Ocarina of Time's cold Mac map: 3,002 s to 1,298 s) - seedds#2065 by Raymon82777 (closed): A vertex library with layer entry points of its own gets bases for its main0 pairs: Ocarina of Time's 230 ms camera-turn hitch in Hyrule Field goes to 39 ms - seedds#2075 by sleepyeyedjoe (closed): Ocarina of Time: a patch of scenery no longer draws dark for one frame when the game gives it a sharper texture (a pool texture the walk does not make holds a stand-in of its own colours until its first read) - seedds#2077 by roandegraaf (closed): Ocarina of Time: fix the "plaid" terrain / missing texture tiles (GPU-drawn textures now reach the storage the game copies) - seedds#2079 by uncraftablexd (open): Render pipelines that share a vertex function are taken from the archive or compiled one at a time: the AGX driver crashes in createVertexProgramVariant's completion block - seedds#2083 by FifteenPen (closed): Ocarina of Time's prewarm list gets 41 more task / mesh pipelines from later play
|
Two parts of this overlap open PRs. They merge cleanly with #2260, but they should be measured together:
The memory watchdog part doesn't overlap anything I know of. |
- Size default memory watchdog limit dynamically from hw.memsize on Mac (48 GB on 64 GB, 28-40 on 36-48 GB, 12 on 16 GB; DM_MEM_CAP_GB overrides) The memory watchdog part of PR seedds#2008 only (its dm/gfx/nvn/hle_nvn.c change and its DM_MEM_CAP_GB row). Its 16-way view cache and its upload batching are left out: main's view cache is keyed by the Metal texture now (87e258b, PR seedds#2542). A 48 GB Mac's limit goes from 24 to 40 GB; DM_MEM_CAP_GB=24 gives the old one. (cherry picked from commit 7167feb, in part)
|
The On the streaming-stall side: we saw 17 frames spike to 1-2.6 s over a ~117-minute session, but in our case the big ones (checked in detail) were shader pipeline compiles blocking the submit thread ( |
What and why
Three performance and stability improvements to NVN memory handling and caching:
Reactive Host Memory Watchdog (
hle_nvn.c):DM_MEM_CAP_DEFAULT_GBwas hardcoded to 24 GB for macOS hosts. On a 16 GB Mac, resident memory climbing towards 24 GB leads to kernel paging/pressure before the watchdog fires; on a 64 GB or 128 GB Mac, it arbitrarily kills processes with plenty of physical memory headroom.hw.memsize: leaving 16 GB headroom on 64+ GB Macs (48 GB limit), 8 GB on 36–48 GB Macs (28–40 GB limit), and 4 GB on 16 GB Macs (12 GB limit).DM_MEM_CAP_GBstill overrides.16-Way Texture View Cache (
VC_WAYS = 16innvn_mtl.m):VC_WAYS = 4suffers from severe thrashing in games with high texture view churn. In The Legend of Zelda: Ocarina of Time (0400486024DEC000), a 7-minute run created over 901,800 texture views on-the-fly (~1,800/sec).VC_WAYSto 16 reduced view creations from 901,825 to 4 (99.8% cache hit rate), eliminating view allocation overhead on the submit thread.Batched Texture Uploads Outside Render Command Buffers (
nvn_mtl.m):!g_cb, direct and staging texture uploads (pkg_upload_) calledup_flush()for every single texture. When streaming an area with 100+ textures (e.g. Kokiri Forest transition), this queued dozens of individual command buffer commits in series, causing submit stalls of 200+ ms.g_direct_uploads % 32 == 0,g_up_texs.count >= 32) cuts streaming spikes by over 80%. All pending uploads are still guaranteed to flush before submissions commit or begin render passes (mtl_sub_begin,res_commit).Checked
make -C dm/host GAME=0400486024DEC000 dmhostclean.python3 -m unittest dm/tools/tests/test_docs.py,python3 -m unittest dm/tools/tests/test_env_switches.py,python3 dm/tools/docs_symptoms.py --checkall pass.[dm] [mem] reactive memory watchdog limit: 48 GB.193 views reused, 4 made in total(down from ~900k views created).