Game: Intel Arc iGPU (Meteor Lake) on Linux, Mesa 26.2.5: 9-10 fps in Kokiri Forest; native float add, off by default here, more than halves the GPU time #1897

Open
opened 2026-11-20 19:36:36 +00:00 by hula4139 · 2 comments

Longshot version

v3.0.0 (source bd6a0cdd)

Channel

Release (the v3.0.0 Linux AppImage's builder, --cli; I start bin/longshot from a script, not through the launcher)

What kind of problem

Low fps or stutter

What happened

On a Meteor Lake Arc iGPU, v3.0.0 runs stable and draws Kokiri Forest correctly, but stays at 9-10 fps, GPU-bound at about 95-100 ms a frame. Halving the render resolution barely changes it. It is the same on every pass through the same spot and on every start. And that is only with RX_VK_NATIVE_FADD=1, which this device doesn't get by default (below).

Suggested change: native float add on by default for Mesa ANV on Meteor Lake. The policy line here says native_float_add=0, and the log says float atomic add on storage buffers: compare-exchange loop. Same save, same spot, Minimum preset, the same 100-420 s of each run:

RX_VK_NATIVE_FADD fps (mean) GPU time a frame (median)
0 (this device's default: the compare-exchange loop) 4.0 241 ms
1 (VK_EXT_shader_atomic_float) 8.7 103 ms

The picture looks the same to me with both (ladder-native-fadd.jpg and ladder-cas-loop.jpg: the same spot, 200 s into each run). 7 runs with it on (up to 15 minutes each) had no device loss and no kernel GPU hang.

native_fadd_listed (gpu/vk/device_policy.h) turns it on "where the card has it and a run has shown it right": Intel's Windows driver on Meteor Lake (0x7Dxx, from 101.9175), but no Mesa driver. So maybe these runs are enough to add VK_DRIVER_ID_INTEL_OPEN_SOURCE_MESA on 0x7Dxx (Mesa 26.2.5, driver version 0x6802005, here). I've only checked the picture by eye, in Kokiri Forest, so you may want more than that first.

Some good news for the Intel work. On this machine, rc.3 and rc.6 (with Mesa 25.3.6) lost the device a few minutes into play (i915 GPU HANG: ecode 12:1:85def5fb: rc.3 in 4 of 4 runs, rc.6 in 3 of 4, 5-7 minutes into play), and rc.6 drew grass and railings blown-out white. v3.0.0 on Mesa 26.2.5 ran 5 times, 3 of them for about 11½ minutes of play each, with no device loss, no kernel GPU hang and a correct picture. Mesa changed at the same time, so I can't say which of the two ended the hang.

Measurements (Kokiri Forest, Link's balcony and house, warm caches, from the attached logs):

Settings fps GPU time a frame (median of the log's lines in play)
Defaults (Output resolution 1920x1080, no Quality preset) 7-10 100 ms
Quality preset Auto (this PC) = Minimum (Render resolution 50 %) about 10 94 ms

Auto's own line: Auto (this PC), measured in play: settled (8 fps of 60, the GPU at 116.6 ms a frame of 16.7): Render resolution 50% kept. A frame is about 1070 draws (0 skipped), 121 dispatches, 162 passes, 187 barriers and 6 submits.

RX_VK_PASS_TIMING=1 (Minimum preset, three 600-present windows in play): the named spans stay at 48-51 ms a frame (dispatches 33.7-35.1 ms, passes 10.6-11.7 ms, barriers 3.4-3.9 ms), while the time between spans ("glue the spans do not name") is 45, 70 and 77 ms, 49-61 % of the first-to-last time, over 13.8-15.6 recordings' command buffers a frame. The heaviest single spans are compute programs 27ee9883e2a20323 (6.8 ms), ee7764a6b1ef459c (5.0 ms), 26fe0c70ab9fc547 (2.7 ms) and the indirect 7650519f88e59b19 (2.65 ms). The full blocks are in longshot_log-minimum-pass-timing.txt.

Also tried (each against the same Minimum baseline, 8.7 fps and 103 ms):

  • RX_VK_BATCH=1 (the log says batched submits on): 8.5 fps, 113 ms. No change.
  • RX_VK_SPLIT_BEFORE_DRAW=0 (split_before_draw=0*): 8.5 fps, 115 ms. No change. So neither batching the submits nor fewer split parts touched the time between the spans.
  • RX_VK_MESH_SHADERS=native: entering the world froze for about 11 minutes (black, stalled), then Kokiri Forest drew correctly at 4-6 fps. So I stay on the compute path.
  • First start with the shader preparation off: the 3D picture was all white about a minute into play, all black a minute later, and right by about five minutes (three screenshots), while shaders compiled in play. The later starts were right from the first screenshot (about 10 s into play).

Where in the game

Kokiri Forest: Link's house, its balcony, doorway and ladder, loaded from a save; standing, walking a little and pressing A.

How often

Every time

Graphics card and driver version

Intel Arc Graphics (MTL) iGPU, 8086:7d55; Intel open-source Mesa driver 26.2.5 (ANV), Vulkan 1.4; i915

CPU, RAM and operating system

Intel Core Ultra 7 155H (22 threads), 31 GB; Batocera 44 (Linux 7.2.8), X11; screen 1920x1080 at 120 Hz

Changes from the defaults

  • Environment: RX_VK_MESH_SHADERS=compute; RX_VK_NATIVE_FADD=1 (see the suggested change above); RX_VK_PRECOMPILE=0 (with rc.3 on Mesa 25.3.6 the up-front preparation made 220 of 38,809 programs in 14 minutes here, and without the 32-lane pin it ran toward out-of-memory); HOS_SDL_VERSION=2, HOS_DISCORD=0, HOS_CRASH_WATCH=0; HOS_CONTROL, the control bridge, which my test script used to press buttons and read fps and state.
  • Options: fullscreen=1, fps_counter=1; the Minimum runs also have quality_preset=6 (Auto → Minimum).
  • Built with LONGSHOT_THIRD_PARTY_OBJECTS=0 and --no-prepare-shaders. My dump has 3 damaged files (Fairy Fountain floor, Ganon stage and Water Temple gate textures; the builder stood in others), none of them in Kokiri Forest.
  • No overlays, ReShade, shader-cache tools or frame generation.

longshot_log.txt

Screenshots or video

Native float add (8 fps) and the compare-exchange loop (4 fps), Minimum, the same spot 200 s into each run:

ladder-native-fadd.jpg
ladder-cas-loop.jpg

Defaults, about 10 s into play, and Minimum, about 6 minutes into play:

kokiri-defaults.jpg
kokiri-minimum.jpg

Before you send it

  • The problem is still there on the newest release or nightly. (v3.0.0; I haven't tried a nightly.)
  • I searched the open issues for it.
  • I'm not attaching game files, keys, the game's executable, its .bin caches or saves.
### Longshot version v3.0.0 (source bd6a0cdd) ### Channel Release (the v3.0.0 Linux AppImage's builder, `--cli`; I start `bin/longshot` from a script, not through the launcher) ### What kind of problem Low fps or stutter ### What happened On a Meteor Lake Arc iGPU, v3.0.0 runs stable and draws Kokiri Forest correctly, but stays at 9-10 fps, GPU-bound at about 95-100 ms a frame. Halving the render resolution barely changes it. It is the same on every pass through the same spot and on every start. And that is only with `RX_VK_NATIVE_FADD=1`, which this device doesn't get by default (below). **Suggested change: native float add on by default for Mesa ANV on Meteor Lake.** The policy line here says `native_float_add=0`, and the log says `float atomic add on storage buffers: compare-exchange loop`. Same save, same spot, Minimum preset, the same 100-420 s of each run: | `RX_VK_NATIVE_FADD` | fps (mean) | GPU time a frame (median) | |---|---|---| | `0` (this device's default: the compare-exchange loop) | 4.0 | 241 ms | | `1` (`VK_EXT_shader_atomic_float`) | 8.7 | 103 ms | The picture looks the same to me with both (`ladder-native-fadd.jpg` and `ladder-cas-loop.jpg`: the same spot, 200 s into each run). 7 runs with it on (up to 15 minutes each) had no device loss and no kernel GPU hang. `native_fadd_listed` (`gpu/vk/device_policy.h`) turns it on "where the card has it and a run has shown it right": Intel's Windows driver on Meteor Lake (`0x7Dxx`, from 101.9175), but no Mesa driver. So maybe these runs are enough to add `VK_DRIVER_ID_INTEL_OPEN_SOURCE_MESA` on `0x7Dxx` (Mesa 26.2.5, driver version `0x6802005`, here). I've only checked the picture by eye, in Kokiri Forest, so you may want more than that first. **Some good news for the Intel work.** On this machine, rc.3 and rc.6 (with Mesa 25.3.6) lost the device a few minutes into play (i915 `GPU HANG: ecode 12:1:85def5fb`: rc.3 in 4 of 4 runs, rc.6 in 3 of 4, 5-7 minutes into play), and rc.6 drew grass and railings blown-out white. v3.0.0 on Mesa 26.2.5 ran 5 times, 3 of them for about 11½ minutes of play each, with no device loss, no kernel GPU hang and a correct picture. Mesa changed at the same time, so I can't say which of the two ended the hang. **Measurements** (Kokiri Forest, Link's balcony and house, warm caches, from the attached logs): | Settings | fps | GPU time a frame (median of the log's lines in play) | |---|---|---| | Defaults (Output resolution 1920x1080, no Quality preset) | 7-10 | 100 ms | | Quality preset Auto (this PC) = Minimum (Render resolution 50 %) | about 10 | 94 ms | Auto's own line: `Auto (this PC), measured in play: settled (8 fps of 60, the GPU at 116.6 ms a frame of 16.7): Render resolution 50% kept`. A frame is about 1070 draws (0 skipped), 121 dispatches, 162 passes, 187 barriers and 6 submits. **`RX_VK_PASS_TIMING=1`** (Minimum preset, three 600-present windows in play): the named spans stay at 48-51 ms a frame (dispatches 33.7-35.1 ms, passes 10.6-11.7 ms, barriers 3.4-3.9 ms), while the time between spans ("glue the spans do not name") is 45, 70 and 77 ms, 49-61 % of the first-to-last time, over 13.8-15.6 recordings' command buffers a frame. The heaviest single spans are compute programs `27ee9883e2a20323` (6.8 ms), `ee7764a6b1ef459c` (5.0 ms), `26fe0c70ab9fc547` (2.7 ms) and the indirect `7650519f88e59b19` (2.65 ms). The full blocks are in `longshot_log-minimum-pass-timing.txt`. **Also tried** (each against the same Minimum baseline, 8.7 fps and 103 ms): - `RX_VK_BATCH=1` (the log says `batched submits on`): 8.5 fps, 113 ms. No change. - `RX_VK_SPLIT_BEFORE_DRAW=0` (`split_before_draw=0*`): 8.5 fps, 115 ms. No change. So neither batching the submits nor fewer split parts touched the time between the spans. - `RX_VK_MESH_SHADERS=native`: entering the world froze for about 11 minutes (black, `stalled`), then Kokiri Forest drew correctly at 4-6 fps. So I stay on the compute path. - First start with the shader preparation off: the 3D picture was all white about a minute into play, all black a minute later, and right by about five minutes (three screenshots), while shaders compiled in play. The later starts were right from the first screenshot (about 10 s into play). ### Where in the game Kokiri Forest: Link's house, its balcony, doorway and ladder, loaded from a save; standing, walking a little and pressing A. ### How often Every time ### Graphics card and driver version Intel Arc Graphics (MTL) iGPU, 8086:7d55; Intel open-source Mesa driver 26.2.5 (ANV), Vulkan 1.4; i915 ### CPU, RAM and operating system Intel Core Ultra 7 155H (22 threads), 31 GB; Batocera 44 (Linux 7.2.8), X11; screen 1920x1080 at 120 Hz ### Changes from the defaults - Environment: `RX_VK_MESH_SHADERS=compute`; `RX_VK_NATIVE_FADD=1` (see the suggested change above); `RX_VK_PRECOMPILE=0` (with rc.3 on Mesa 25.3.6 the up-front preparation made 220 of 38,809 programs in 14 minutes here, and without the 32-lane pin it ran toward out-of-memory); `HOS_SDL_VERSION=2`, `HOS_DISCORD=0`, `HOS_CRASH_WATCH=0`; `HOS_CONTROL`, the control bridge, which my test script used to press buttons and read fps and state. - Options: `fullscreen=1`, `fps_counter=1`; the Minimum runs also have `quality_preset=6` (Auto → Minimum). - Built with `LONGSHOT_THIRD_PARTY_OBJECTS=0` and `--no-prepare-shaders`. My dump has 3 damaged files (Fairy Fountain floor, Ganon stage and Water Temple gate textures; the builder stood in others), none of them in Kokiri Forest. - No overlays, ReShade, shader-cache tools or frame generation. ### longshot_log.txt - [longshot_log-defaults.txt](https://git.magicbean.dev/attachments/f2485e0a-f736-4adb-a316-129bde7e0731): defaults, warm caches, 13 min - [longshot_log-minimum-pass-timing.txt](https://git.magicbean.dev/attachments/d04c5860-e4ac-4a52-bb12-9323c48d58a1): Minimum + `RX_VK_PASS_TIMING=1`, 5½ min - [longshot_log-minimum-cas-loop.txt](https://git.magicbean.dev/attachments/06eb90e9-1c74-4993-a03e-3e353fa5c913): Minimum + `RX_VK_NATIVE_FADD=0`, 7 min ### Screenshots or video Native float add (8 fps) and the compare-exchange loop (4 fps), Minimum, the same spot 200 s into each run: ![ladder-native-fadd.jpg](https://git.magicbean.dev/attachments/8c5c6c7f-f1e7-4cfc-a523-281f99e56823) ![ladder-cas-loop.jpg](https://git.magicbean.dev/attachments/61260562-639c-4ad2-83b4-9b15f3fa6948) Defaults, about 10 s into play, and Minimum, about 6 minutes into play: ![kokiri-defaults.jpg](https://git.magicbean.dev/attachments/9a182aa2-d93c-45fb-9236-be308f1bd0d8) ![kokiri-minimum.jpg](https://git.magicbean.dev/attachments/eaff0209-be8d-4b24-899f-f51b233630e6) ### Before you send it - [x] The problem is still there on the newest release or nightly. (v3.0.0; I haven't tried a nightly.) - [x] I searched the open issues for it. - [x] I'm not attaching game files, keys, the game's executable, its .bin caches or saves.
Member

Possibly related, for context:

  • PR #1793 WIP: fix(vk): float adds native on Lunar Lake iGPUs from driver 101.8991 — Open Mesa-excluded native float-add default change, adjacent Intel device-policy work
  • PR #191 fix(vk): Intel's Windows driver no longer loses the device on float adds to storage buffers: they are the GPU's own atomic — Earlier float-add policy change; tested Meteor Lake on Intel Windows, not Mesa
  • PR #1586 perf(vk): defaults of its own for Mesa's Intel driver, 8.3 to 18.2 fps on an Iris Xe — Merged Mesa Intel performance defaults on Iris Xe; broader driver-performance context, different measured bottleneck
Possibly related, for context: - PR #1793 WIP: fix(vk): float adds native on Lunar Lake iGPUs from driver 101.8991 — Open Mesa-excluded native float-add default change, adjacent Intel device-policy work - PR #191 fix(vk): Intel's Windows driver no longer loses the device on float adds to storage buffers: they are the GPU's own atomic — Earlier float-add policy change; tested Meteor Lake on Intel Windows, not Mesa - PR #1586 perf(vk): defaults of its own for Mesa's Intel driver, 8.3 to 18.2 fps on an Iris Xe — Merged Mesa Intel performance defaults on Iris Xe; broader driver-performance context, different measured bottleneck
Author

Follow-up with nightly-625 (a104182b, which has #1586), same machine, settings, save and spot (Minimum, compute mesh path, RX_VK_NATIVE_FADD=1):

fps (mean, 100-420 s) GPU time a frame (median)
v3.0.0 8.7 103 ms
nightly-625 (3 runs) 19.1-19.4 about 50 ms

No device loss in any of 7 runs (up to 13 minutes), and the first start was drawn right from the start (no white or black phase).

But it draws less. 41 draws a frame are skipped (v3.0.0: 0), all at vk_draw.cpp:8722 (fallback_record() refuses the mesh draw), for the whole run, with 121 render passes a frame instead of 162 and 114 dispatches instead of 121. On screen, the 3D moss and grass blades, the bushes and ivy on the stumps and the grass patches on the forest floor are gone; flat textures show instead (attached, v3.0.0 left, nightly-625 right, the same view as the save loads).

It isn't #1586's three defaults. With RX_VK_MESH_SHARED_WARP=0, RX_VK_MESH_GLOBAL_ATTRS=0 and RX_VK_MESH_PAGES=0 one at a time, and all three together (the start lines confirm each), the 41 skips stay, at 16-19.5 fps. So something else between v3.0.0 and nightly-625 leaves these mesh draws out on this device, and I'd guess part of the speed-up is those missing draws. I haven't bisected further. The log of a 13-minute nightly run is attached.

v3.0.0 (left) and nightly-625 (right), the same view as the save loads:

compare-v300-vs-nightly625-mound.jpg
compare-v300-vs-nightly625-forest.jpg

Log: longshot_log-nightly-625.txt (warm caches, 13 min)

Follow-up with **nightly-625** (`a104182b`, which has #1586), same machine, settings, save and spot (Minimum, compute mesh path, `RX_VK_NATIVE_FADD=1`): | | fps (mean, 100-420 s) | GPU time a frame (median) | |---|---|---| | v3.0.0 | 8.7 | 103 ms | | nightly-625 (3 runs) | 19.1-19.4 | about 50 ms | No device loss in any of 7 runs (up to 13 minutes), and the first start was drawn right from the start (no white or black phase). **But it draws less.** 41 draws a frame are skipped (v3.0.0: 0), all at `vk_draw.cpp:8722` (`fallback_record()` refuses the mesh draw), for the whole run, with 121 render passes a frame instead of 162 and 114 dispatches instead of 121. On screen, the 3D moss and grass blades, the bushes and ivy on the stumps and the grass patches on the forest floor are gone; flat textures show instead (attached, v3.0.0 left, nightly-625 right, the same view as the save loads). **It isn't #1586's three defaults.** With `RX_VK_MESH_SHARED_WARP=0`, `RX_VK_MESH_GLOBAL_ATTRS=0` and `RX_VK_MESH_PAGES=0` one at a time, and all three together (the start lines confirm each), the 41 skips stay, at 16-19.5 fps. So something else between v3.0.0 and nightly-625 leaves these mesh draws out on this device, and I'd guess part of the speed-up is those missing draws. I haven't bisected further. The log of a 13-minute nightly run is attached. v3.0.0 (left) and nightly-625 (right), the same view as the save loads: ![compare-v300-vs-nightly625-mound.jpg](https://git.magicbean.dev/attachments/926180dc-1faa-48fd-be93-e119faa8f0f6) ![compare-v300-vs-nightly625-forest.jpg](https://git.magicbean.dev/attachments/7960d45f-38be-41dc-8927-366d1939ad0d) Log: [longshot_log-nightly-625.txt](https://git.magicbean.dev/attachments/c64e0d4c-4c38-4d3b-b84c-b26ff75aa613) (warm caches, 13 min)
Sign in to join this conversation.
No milestone
No project
No assignees
2 participants
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set

Reference
MagicBeanDev/Longshot#1897
No description provided.