feat(agc): merge a draw into more than one unclamped normalized color target by Tehshortbus · Pull Request #12916 · boykopovar/AnyPS5

Tehshortbus · 2026-10-29T14:16:07Z

What

A draw that blends into more than one color target with BLEND_CLAMP clear stopped with a draw merges into one color target at most, so blending unclamped normalized color cannot join another merged color target. PPSA26344 (02.002.000) stops on such a draw while it streams its first world cell (#12796). The title's log names MRT1 (CB_COLOR_INFO 0x0000000c, CB_BLEND1_CONTROL 0x40010504). In #675, MRT0 was reported with the same values before #11684. This follows the shape @oneandonlydean suggested on the issue: a list of merge entries, one interlock region, and a two-target case in the unclamped-blend execution test.

Request. ShaderPixelStageInfo::targetMerge becomes targetMerges, one entry per merged target. ExportTargetMerges returns an entry for every target with a partial write mask. The cache key, the draw lookup hash, the draw register key and the recompiler's variant hash cover every entry.
Serialization. The first entry keeps the version 46 layout. Bits 17..19 of its mask word count the entries that follow it, the way bit 16 already flags a blend, so the version stays 49. Requests with no merge or one merge serialize to the same bytes as before. A request with more than eight entries is refused when written, and so is an entry with no mask, other than the lone default, when read.
SPIR-V. Each entry gets its own storage image, its own merged value and pending flag, and its own blend state. Every MRT export merges into the entry for its slot. FlushColorTargetMerge stores each pending entry inside the draw's one pixel interlock region.
Driver.
- Several unclamped blends can merge in one draw. A partial write of a packed (4_4_4_4) target still merges alone, and an unclamped blend still cannot join it; both refusals say so. Merged targets whose memory overlaps are refused.
- A merged target's blend attachment is reset to the empty state rather than only having its enable and write mask cleared. Without independentBlend, two merged targets with different blend equations otherwise gave different pAttachments (VUID-VkPipelineColorBlendStateCreateInfo-pAttachments-00605).
Debt. The unclamped normalized targets entry now lists a packed target's partial write in the same draw as refused, in place of the second merged target. A new 4_4_4_4 entry records that a partial write merges alone, and that merged targets sharing memory throw.

24 files, +371/−212. The tests are +222/−84: the existing tests that read targetMerge move to the list, and the rest are the new cases below.

Tested

CachyOS Linux (kernel 7.3.0-4-cachyos), GCC 16.2.1, Release, NVIDIA GeForce RTX 4070 SUPER (615.78.08).

Head c037722384 on main 1b75a2de11. The tests below ran on the same change on main b0465557fd (47bdfe07a2). None of the 26 commits between the two mains touch this PR's files.

New cases:
- agc_driver_graphics:
  - Two slots with CB_COLOR_INFO 0x0000000c and CB_BLEND0_CONTROL/CB_BLEND1_CONTROL 0x40010504 at different addresses decode into two merge entries for MRT0 and MRT1. Each has its own descriptor and SRC_ALPHA, ONE_MINUS_SRC_ALPHA, ADD for color and alpha.
  - With CB_BLEND1_CONTROL set to ONE, ONE, each entry keeps its own factors.
  - Two merged targets at the same address are refused.
  - A 4_4_4_4 partial write in slot 1 after the blend, and in slot 0 before it, are both refused with the new messages.
- agc_driver_unclamped_normalized_blend: one draw exports MRT0 and MRT1 and blends both, checked per pixel against the CPU reference, on the stored channels:
  - two 8_8 UNORM targets with SRC_ALPHA, ONE_MINUS_SRC_ALPHA and out-of-range sources;
  - an 8_8_8_8 UNORM MRT0 with ONE, ONE and an 8_8 UNORM MRT1 with SUBTRACT, so the entries carry different blend equations.
  Every draw in the test, including the 20 existing single-target ones, also checks that its pixel shader has one interlock begin and stores every merged target inside it. The test builds the draw state itself, so it reaches the request, the SPIR-V and the pipeline but not register decode; decode is covered by agc_driver_graphics.
- agc_shader_memory:
  - A request with two merge entries round-trips, and its second entry follows the first with the count in the first mask word.
  - A second entry with no mask is refused when read, and nine entries are refused when written.
  - The static shader identity changes with a second entry, and with its slot or blend, but not with its descriptor.
Focused run. ANYPS5_REQUIRE_VULKAN=1 ANYPS5_NO_SHADER_CACHE=1 VK_INSTANCE_LAYERS=VK_LAYER_KHRONOS_validation ctest --test-dir build-tests -V -R '^(request_serializer|agc_shader_memory|agc_driver_graphics|agc_driver_packed_unorm4444_color_target|agc_driver_packed_unorm4444_multisample_partial_write|agc_driver_packed_unorm4444_partial_write|agc_driver_unclamped_normalized_blend)$'

main b0465557fd this change

tests 7/7 pass 7/7 pass

VUID- lines 0 0

was not found but was requested by env var VK_INSTANCE_LAYERS 0 0
Full suite. 2905 of 2906 pass under the same environment, with no VUID- line. guest_freebsd_clocks fails because the host zone is not UTC (Tests: guest_freebsd_clocks aborts where the host zone is not UTC #12736) and passes with TZ=UTC.

Negative controls. Each control breaks one piece of the change; each is caught:

Change made	Result
`main`'s refusal of any second merged target	`agc_driver_graphics` fails with the title's message
only the first entry is stored	`agc_driver_unclamped_normalized_blend` fails: not every merged target is stored inside the interlock region
every entry blends with the first entry's blend state	`agc_driver_unclamped_normalized_blend` fails on MRT1's pixels
merged blend attachments keep their factors	`agc_driver_unclamped_normalized_blend` passes, but the validation layer reports `VUID-…-00605` twice
the serializer does not count further entries	`agc_shader_memory` fails

PPSA26344 (02.002.000). One run each from launch, same files and settings, with anyps5-entitlements.ini listing the add-on label the title needs:

Build	Result	Presents
`main` `5454cd2e45`	aborts at 303 s with the refusal above (exit 134)	397
this change on `5454cd2e45` (`3cbd74244b`)	passes that frame and keeps loading, showing the title's loading spinner; closed by hand at 1966 s with no driver failure	550
this change, head `c037722384`	passes that frame, plays the title's audio and controller rumble, then slows to under 0.1 fps with the texture caches unable to free memory (5118 MiB of 9038 MiB in sampled textures); closed by hand at 793 s with no driver failure	560

These runs show the draw no longer stops the title. They don't show that the merged targets look right, since main never got past this point to compare against; the execution test checks that.

python3 tools/check_conventions.py --base origin/main and git diff --check pass.

Not checked:

Windows, AMD and Intel GPUs.
Registration modules with two merges.
The command-buffer draw path with two merges (DrawDecode, DrawLookup, DrawRegisterKey), other than in the PPSA26344 run.
Two merged targets of a multisampled draw: unclamped blends need single-sample targets, and packed partial writes merge alone.
Gameplay in the title: the runs stayed in its loading stretch.

A merged target beside a target that blends normally still gives that target's attachment state next to an empty one. Without independentBlend this is the same pAttachments difference as on main before this change; the focused and full runs report no VUID- line for it.

Overlap. tools/pr_overlap.py (main 609422fefe) reports no overlap for the pull requests it could compare. For the 77 it could not, I merged each with this branch and with main and compared the conflicts in shared files:

Same merge path (3): feat(agc): blend into 11_11_10 float color targets in the pixel shader #9020, feat(agc): draw alpha-tested packed-target merges that write depth as a depth pass and an EQUAL merge pass #9335 and feat(agc): blend into 10_11_11 unorm color targets in the pixel shader #9748 extend the single TargetMerge. This branch adds conflicts in CacheKey.hpp, DrawLookup.cpp, Recompiler.cpp, ShaderStageInputInfo.hpp and ShaderInputInfoBuilder.cpp, and its lines fall inside their existing conflicts in RequestSerializer.cpp, SpirvModuleEmitter.cpp, State.cpp and ShaderMemory.cpp.
Adjacent lines (3):
- feat(agc): draw partial writes of multisampled 4_4_4_4 targets into a temporary target #11557: the packed refusal in State.cpp and the multisampled 4_4_4_4 tests.
- feat(agc): test the depth bounds in the pixel program when the device lacks the depthBounds feature #11551: DrawRegisterKey.cpp and ShaderInfoCollector.cpp.
- fix(shader): preserve raw fragment exports for integer attachments #6235: its existing conflicts in State.cpp, CacheKey.hpp, Recompiler.hpp and SpirvModuleEmitter.cpp.
This branch's line inside a conflict the PR already has with main (7): fix(agc): grow the storage cache budget while it evicts under pressure #6194, perf(agc): reuse image-table elements whose descriptors repeat #6223 and fix(agc): take a host-coherent memory type for a host import (#6512) #6548 (libSceAgcDriver.md); perf(shader): let spanned waves run at any host subgroup size where it can be narrower than 32 #8004, feat(agc): run mesh-path draws through a compute pass on devices without mesh shaders #9356 and feat(agc): emulate mesh-fallback barycentrics through tessellation #9387 (RequestSerializerTests.cpp); feat(agc): apply a T#'s MIN_LOD clamp in the shader without VK_EXT_image_view_min_lod #12662 (ShaderMemory.cpp).
The other 64: the same conflicts with this branch as with main.

All 13 are older, so this lands after them, recorded in Depends on. I'll rebase onto whichever merges first.

Checklist

🤖 Generated with Claude Code

… target A draw that blends into several targets with BLEND_CLAMP clear now carries one merge entry per target. Each entry gets its own storage image and pixel-shader blend, and every pending entry is stored inside the draw's one pixel interlock region. Partial writes of packed targets still merge alone. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

github-actions · 2026-10-29T15:10:10Z

Other open pull requests touch the same code. Merging one will break the other, or both implement the same thing. If this one builds on them, list them in Depends on.

Pull request	Conflicting files (lines on `main`)	Same exports	Same new files
PR 12968	`docs/dev/debt/libSceAgcDriver.md` L288

Some comparisons could not be completed because a pull request conflicts with the base branch. Review these shared files manually before concluding that there is no overlap.

Pull request	Shared files not compared
PR 1368	`core/libs/prx/libSceAgcDriver/Graphics/include/State.hpp`
PR 1395	`core/libs/prx/libSceAgcDriver/Graphics/include/State.hpp` `core/libs/prx/libSceAgcDriver/Graphics/src/State.cpp` `core/libs/prx/libSceAgcDriver/tests/Graphics.cpp`
PR 2434	`docs/dev/debt/libSceAgcDriver.md`
PR 4040	`core/libs/prx/libSceAgcDriver/tests/Graphics.cpp`
PR 4448	`docs/dev/debt/libSceAgcDriver.md`
PR 4449	`docs/dev/debt/libSceAgcDriver.md`
PR 5021	`core/shader/recompiler/Recompiler.cpp` `core/shader/recompiler/SpirvBackend/include/SpirvBackend/SpirvEmitterState.hpp`
PR 5371	`core/libs/prx/libSceAgcDriver/tests/ShaderMemory.cpp`
PR 5372	`docs/dev/debt/libSceAgcDriver.md`
PR 5609	`core/libs/prx/libSceAgcDriver/Execution/src/Driver/Draw/DrawLookup.cpp` `core/libs/prx/libSceAgcDriver/Execution/src/Driver/Draw/DrawRegisterKey.cpp` `core/libs/prx/libSceAgcDriver/Graphics/src/Pipeline.cpp` `core/shader/recompiler/Recompiler.cpp` `core/shader/recompiler/Recompiler.hpp`
PR 5925	`core/libs/prx/libSceAgcDriver/Graphics/src/Pipeline.cpp` `core/libs/prx/libSceAgcDriver/tests/Graphics.cpp` `core/shader/recompiler/Recompiler.hpp` `docs/dev/debt/libSceAgcDriver.md`
PR 6185	`docs/dev/debt/libSceAgcDriver.md`
PR 6186	`docs/dev/debt/libSceAgcDriver.md`
PR 6188	`core/libs/prx/libSceAgcDriver/Execution/src/Driver/Draw/DrawDecode.cpp` `core/libs/prx/libSceAgcDriver/Execution/src/Driver/Draw/DrawLookup.cpp` `core/libs/prx/libSceAgcDriver/Graphics/src/State.cpp` `core/shader/recompiler/Recompiler.hpp`
PR 6273	`docs/dev/debt/libSceAgcDriver.md`
PR 6280	`core/libs/prx/libSceAgcDriver/tests/Graphics.cpp` `docs/dev/debt/libSceAgcDriver.md`
PR 6300	`docs/dev/debt/libSceAgcDriver.md`
PR 6412	`core/libs/prx/libSceAgcDriver/tests/ShaderMemory.cpp` `core/shader/recompiler/Recompiler.cpp` `core/shader/recompiler/Recompiler.hpp` `docs/dev/debt/libSceAgcDriver.md`
PR 6416	`core/shader/recompiler/SpirvBackend/include/SpirvBackend/SpirvEmitterState.hpp`
PR 6481	`core/libs/prx/libSceAgcDriver/tests/Graphics.cpp` `docs/dev/debt/libSceAgcDriver.md`
PR 6488	`core/libs/prx/libSceAgcDriver/tests/ShaderMemory.cpp` `core/shader/recompiler/Recompiler.cpp`
PR 6549	`docs/dev/debt/libSceAgcDriver.md`
PR 6751	`core/shader/recompiler/SpirvBackend/include/SpirvBackend/SpirvEmitterState.hpp`
PR 6804	`core/libs/prx/libSceAgcDriver/Graphics/include/State.hpp` `core/libs/prx/libSceAgcDriver/Graphics/src/State.cpp` `core/libs/prx/libSceAgcDriver/tests/Graphics.cpp` `docs/dev/debt/libSceAgcDriver.md`
PR 6808	`core/libs/prx/libSceAgcDriver/tests/Graphics.cpp` `docs/dev/debt/libSceAgcDriver.md`
PR 6924	`core/shader/recompiler/Recompiler.cpp`
PR 6959	`core/libs/prx/libSceAgcDriver/Graphics/include/State.hpp` `core/libs/prx/libSceAgcDriver/Graphics/src/Pipeline.cpp` `core/libs/prx/libSceAgcDriver/tests/Graphics.cpp` `core/libs/prx/libSceAgcDriver/tests/ShaderMemory.cpp` `core/shader/recompiler/Recompiler.cpp` `core/shader/recompiler/Recompiler.hpp` `core/shader/recompiler/SpirvBackend/src/SpirvModuleEmitter.cpp` `docs/dev/debt/libSceAgcDriver.md`
PR 6970	`core/libs/prx/libSceAgcDriver/Graphics/src/Pipeline.cpp` `core/libs/prx/libSceAgcDriver/tests/Graphics.cpp` `core/shader/recompiler/Recompiler.hpp`
PR 6977	`core/libs/prx/libSceAgcDriver/Execution/src/Driver/Draw/DrawDecode.cpp` `core/libs/prx/libSceAgcDriver/Execution/src/Driver/Draw/DrawRegisterKey.cpp` `core/libs/prx/libSceAgcDriver/Graphics/include/State.hpp` `core/libs/prx/libSceAgcDriver/Graphics/src/State.cpp` `docs/dev/debt/libSceAgcDriver.md`
PR 6987	`docs/dev/debt/libSceAgcDriver.md`
PR 6989	`docs/dev/debt/libSceAgcDriver.md`
PR 6995	`docs/dev/debt/libSceAgcDriver.md`
PR 7002	`docs/dev/debt/libSceAgcDriver.md`
PR 7048	`docs/dev/debt/libSceAgcDriver.md`
PR 7256	`core/libs/prx/libSceAgcDriver/Graphics/include/State.hpp` `core/libs/prx/libSceAgcDriver/Graphics/src/State.cpp` `core/libs/prx/libSceAgcDriver/tests/Graphics.cpp` `docs/dev/debt/libSceAgcDriver.md`
PR 7275	`core/libs/prx/libSceAgcDriver/Graphics/src/Pipeline.cpp`
PR 7392	`core/shader/recompiler/SpirvBackend/include/SpirvBackend/SpirvEmitterState.hpp`
PR 7449	`docs/dev/debt/libSceAgcDriver.md`
PR 7604	`docs/dev/debt/libSceAgcDriver.md`
PR 7642	`core/libs/prx/libSceAgcDriver/Execution/src/Driver/Draw/DrawLookup.cpp` `core/libs/prx/libSceAgcDriver/Graphics/src/Pipeline.cpp` `core/libs/prx/libSceAgcDriver/tests/Graphics.cpp` `core/libs/prx/libSceAgcDriver/tests/ShaderMemory.cpp` `core/shader/recompiler/CacheKey.hpp` `core/shader/recompiler/Recompiler.cpp` `core/shader/recompiler/Recompiler.hpp` `core/shader/recompiler/SpirvBackend/include/SpirvBackend/SpirvEmitterState.hpp` `core/shader/recompiler/SpirvBackend/src/SpirvModuleEmitter.cpp` `docs/dev/debt/libSceAgcDriver.md`
PR 7684	`core/libs/prx/libSceAgcDriver/tests/ShaderMemory.cpp`
PR 7689	`core/shader/recompiler/Recompiler.hpp`
PR 7844	`core/libs/prx/libSceAgcDriver/tests/Graphics.cpp` `docs/dev/debt/libSceAgcDriver.md`
PR 7978	`docs/dev/debt/libSceAgcDriver.md`
PR 8014	`docs/dev/debt/libSceAgcDriver.md`
PR 8223	`docs/dev/debt/libSceAgcDriver.md`
PR 8246	`core/libs/prx/libSceAgcDriver/tests/Graphics.cpp`
PR 8319	`docs/dev/debt/libSceAgcDriver.md`
PR 8542	`docs/dev/debt/libSceAgcDriver.md`
PR 8592	`docs/dev/debt/libSceAgcDriver.md`
PR 8874	`core/libs/prx/libSceAgcDriver/tests/Graphics.cpp` `docs/dev/debt/libSceAgcDriver.md`
PR 9432	`core/libs/prx/libSceAgcDriver/Execution/src/Driver/Draw/DrawDecode.cpp` `core/libs/prx/libSceAgcDriver/Execution/src/Driver/Draw/DrawRegisterKey.cpp` `core/libs/prx/libSceAgcDriver/Graphics/include/State.hpp` `core/libs/prx/libSceAgcDriver/Graphics/src/State.cpp` `docs/dev/debt/libSceAgcDriver.md`
PR 9489	`core/libs/prx/libSceAgcDriver/Execution/src/Driver/Draw/DrawRegisterKey.cpp` `core/libs/prx/libSceAgcDriver/Graphics/include/State.hpp` `core/libs/prx/libSceAgcDriver/Graphics/src/Pipeline.cpp` `core/libs/prx/libSceAgcDriver/Graphics/src/State.cpp` `core/libs/prx/libSceAgcDriver/tests/Graphics.cpp`
PR 9592	`docs/dev/debt/libSceAgcDriver.md`
PR 9700	`core/libs/prx/libSceAgcDriver/Graphics/include/State.hpp` `core/libs/prx/libSceAgcDriver/Graphics/src/State.cpp` `core/libs/prx/libSceAgcDriver/tests/Graphics.cpp` `docs/dev/debt/libSceAgcDriver.md`
PR 9712	`core/libs/prx/libSceAgcDriver/tests/Graphics.cpp`
PR 9889	`core/libs/prx/libSceAgcDriver/tests/ShaderMemory.cpp`
PR 9891	`core/libs/prx/libSceAgcDriver/Execution/src/Driver/Draw/DrawDecode.cpp` `core/libs/prx/libSceAgcDriver/Execution/src/Driver/Draw/DrawRegisterKey.cpp` `core/libs/prx/libSceAgcDriver/Graphics/include/State.hpp` `core/libs/prx/libSceAgcDriver/Graphics/src/State.cpp` `docs/dev/debt/libSceAgcDriver.md`
PR 9893	`core/libs/prx/libSceAgcDriver/Execution/src/Driver/Draw/DrawDecode.cpp` `core/libs/prx/libSceAgcDriver/Execution/src/Driver/Draw/DrawRegisterKey.cpp` `core/libs/prx/libSceAgcDriver/Graphics/include/State.hpp` `core/libs/prx/libSceAgcDriver/Graphics/src/State.cpp` `docs/dev/debt/libSceAgcDriver.md`
PR 9895	`core/libs/prx/libSceAgcDriver/Execution/src/Driver/Draw/DrawDecode.cpp` `core/libs/prx/libSceAgcDriver/Execution/src/Driver/Draw/DrawRegisterKey.cpp` `core/libs/prx/libSceAgcDriver/Graphics/include/State.hpp` `core/libs/prx/libSceAgcDriver/Graphics/src/State.cpp` `docs/dev/debt/libSceAgcDriver.md`
PR 9897	`core/libs/prx/libSceAgcDriver/tests/Graphics.cpp` `core/libs/prx/libSceAgcDriver/tests/ShaderMemory.cpp` `core/shader/recompiler/Recompiler.hpp`
PR 9898	`docs/dev/debt/libSceAgcDriver.md`
PR 9915	`core/libs/prx/libSceAgcDriver/tests/Graphics.cpp` `docs/dev/debt/libSceAgcDriver.md`
PR 10026	`docs/dev/debt/libSceAgcDriver.md`
PR 10049	`core/libs/prx/libSceAgcDriver/tests/Graphics.cpp` `core/shader/recompiler/Recompiler.hpp` `docs/dev/debt/libSceAgcDriver.md`
PR 10193	`core/libs/prx/libSceAgcDriver/Execution/src/Driver/Draw/DrawLookup.cpp` `core/libs/prx/libSceAgcDriver/Graphics/src/Pipeline.cpp` `core/libs/prx/libSceAgcDriver/tests/Graphics.cpp` `core/libs/prx/libSceAgcDriver/tests/ShaderMemory.cpp` `core/shader/recompiler/CacheKey.hpp` `core/shader/recompiler/Recompiler.cpp` `core/shader/recompiler/Recompiler.hpp` `core/shader/recompiler/SpirvBackend/include/SpirvBackend/SpirvEmitterState.hpp` `core/shader/recompiler/SpirvBackend/src/SpirvModuleEmitter.cpp` `docs/dev/debt/libSceAgcDriver.md`
PR 10233	`core/libs/prx/libSceAgcDriver/tests/ShaderMemory.cpp` `docs/dev/debt/libSceAgcDriver.md`
PR 10249	`core/libs/prx/libSceAgcDriver/Graphics/include/State.hpp` `core/libs/prx/libSceAgcDriver/Graphics/src/State.cpp`
PR 10472	`core/libs/prx/libSceAgcDriver/tests/Graphics.cpp` `docs/dev/debt/libSceAgcDriver.md`
PR 10492	`docs/dev/debt/libSceAgcDriver.md`
PR 10497	`docs/dev/debt/libSceAgcDriver.md`
PR 10547	`core/libs/prx/libSceAgcDriver/tests/ShaderMemory.cpp` `core/shader/recompiler/CacheKey.hpp` `core/shader/recompiler/ControlFlow/src/RequestSerializer.cpp` `core/shader/recompiler/Recompiler.cpp` `core/shader/recompiler/Recompiler.hpp` `core/shader/recompiler/tests/RequestSerializerTests.cpp`
PR 10652	`core/shader/recompiler/CacheKey.hpp` `core/shader/recompiler/Recompiler.cpp` `core/shader/recompiler/Recompiler.hpp` `core/shader/recompiler/SpirvBackend/include/SpirvBackend/SpirvEmitterState.hpp`
PR 10757	`core/libs/prx/libSceAgcDriver/tests/ShaderMemory.cpp` `core/shader/recompiler/Recompiler.cpp` `core/shader/recompiler/Recompiler.hpp`
PR 11351	`core/libs/prx/libSceAgcDriver/tests/ShaderMemory.cpp` `core/shader/recompiler/CacheKey.hpp` `core/shader/recompiler/ControlFlow/src/RequestSerializer.cpp` `core/shader/recompiler/Recompiler.cpp` `core/shader/recompiler/Recompiler.hpp` `core/shader/recompiler/SpirvBackend/include/SpirvBackend/SpirvEmitterState.hpp` `core/shader/recompiler/SpirvBackend/src/SpirvModuleEmitter.cpp` `core/shader/recompiler/tests/RequestSerializerTests.cpp`
PR 12120	`docs/dev/debt/libSceAgcDriver.md`
PR 12324	`core/libs/prx/libSceAgcDriver/tests/ShaderMemory.cpp` `core/shader/recompiler/Recompiler.cpp`
PR 12348	`docs/dev/debt/libSceAgcDriver.md`
PR 12548	`core/libs/prx/libSceAgcDriver/tests/ShaderMemory.cpp` `core/shader/recompiler/Recompiler.cpp`
PR 12619	`core/libs/prx/libSceAgcDriver/Graphics/src/Pipeline.cpp` `core/shader/recompiler/Optimization/src/ShaderInfoCollector.cpp` `core/shader/recompiler/SpirvBackend/src/SpirvMemory/SpirvInputOutput.cpp`
PR 12764	`core/shader/recompiler/Recompiler.cpp` `core/shader/recompiler/Recompiler.hpp` `core/shader/recompiler/SpirvBackend/include/SpirvBackend/SpirvEmitterState.hpp` `core/shader/recompiler/SpirvBackend/src/SpirvMemory/SpirvModuleSetup.cpp` `core/shader/recompiler/SpirvBackend/src/SpirvModuleEmitter.cpp` `docs/dev/debt/libSceAgcDriver.md`
PR 12828	`core/libs/prx/libSceAgcDriver/tests/ShaderMemory.cpp` `core/shader/recompiler/Recompiler.cpp` `core/shader/recompiler/Recompiler.hpp`

Tehshortbus · 2026-10-29T21:21:18Z

Checked the overlap report by hand. It matches the Overlap section of the description:

The 13 pull requests this one affects (fix(agc): grow the storage cache budget while it evicts under pressure #6194, perf(agc): reuse image-table elements whose descriptors repeat #6223, fix(shader): preserve raw fragment exports for integer attachments #6235, fix(agc): take a host-coherent memory type for a host import (#6512) #6548, perf(shader): let spanned waves run at any host subgroup size where it can be narrower than 32 #8004, feat(agc): blend into 11_11_10 float color targets in the pixel shader #9020, feat(agc): draw alpha-tested packed-target merges that write depth as a depth pass and an EQUAL merge pass #9335, feat(agc): run mesh-path draws through a compute pass on devices without mesh shaders #9356, feat(agc): emulate mesh-fallback barycentrics through tessellation #9387, feat(agc): blend into 10_11_11 unorm color targets in the pixel shader #9748, feat(agc): test the depth bounds in the pixel program when the device lacks the depthBounds feature #11551, feat(agc): draw partial writes of multisampled 4_4_4_4 targets into a temporary target #11557, feat(agc): apply a T#'s MIN_LOD clamp in the shader without VK_EXT_image_view_min_lod #12662) are on its Depends on line, so the report leaves them out.
64 of the 66 listed here are among the 64 the description already reports as unaffected: merged with this branch and with main, they give the same conflicts in the shared files.
fix(agc): save the guest bytes of results that become pending over a surface that is no longer fully write-watched #7048 and fix(agc): hash an idle unit when its range leaves the write watch instead of at every upload #10026 are new since then. They share only docs/dev/debt/libSceAgcDriver.md, and each merges into it without a conflict, with or without this branch (1b75a2de11 and c037722384). Their conflicts with main are in other files.

AI-assisted: yes (Claude Code)

TechLevelZero · 2026-10-29T21:55:32Z

✅ NVIDIA GPU test: passed

PR head c037722 merged onto main bb04a4f · tested 2026-10-29 21:55 UTC · 21 min


GPU	NVIDIA RTX 2000 Ada Generation, driver 595.99.02, Vulkan 1.4.329
Host	Ubuntu 26.04.1 LTS, kernel 7.0.0-38-generic, container ubuntu:24.04
Logs	build and test logs for this run
Build	✅ passed (Release, `-DANYPS5_ENABLE_SPIRV_TOOLS=ON`)
Tests	✅ 2,240 passed · 6 failed (0 new, 5 also fail on main, 1 unstable on main) · 40 skipped (`ANYPS5_REQUIRE_VULKAN=1`)
Test setup	CPU-only tests without a GPU; GPU tests (and any CPU-only test that failed or skipped without one) once on the GPU with `VK_LAYER_KHRONOS_validation` and `ANYPS5_NO_SHADER_CACHE=1`; main ran the same way
Test selection	2,286 of 2,918 tests run: the ones this change can reach, chosen by `tools/select_tests.py` from main's build graph
Vulkan validation	✅ no new validation IDs vs main (860 tests run on the GPU)
Slow tests (>30 s, GPU tests timed with validation layers)	⚠️ `agc_driver_bda_device` 55.6 s

Unstable on main (1, these also failed in recent main runs on this machine; not counted against this PR)

agc_driver_submit_progress: failed in 4 of the last 6 main runs

Failing on main too (5, not counted against this PR)

agc_compute_pipeline_inspection
agc_submit_progress
agc_submit_progress_no_timeline
guest_mmap_executable
guest_socket_options_multicast

Validation IDs already present on main (4 IDs in 400 tests)

VUID-VkDeviceCreateInfo-pNext-pNext (400 tests): agc_compute_pipeline_replay, agc_driver, agc_driver_address_passes_draw_transitions, agc_driver_address_passes_targets_split, agc_driver_alias_copy_label, agc_driver_array_layer_target, agc_driver_array_layer_target_cpu, agc_driver_bda_anchor_reads, and 392 more
VUID-vkCmdCopyImage-srcImage-01548 (11 tests): agc_driver_depth_clear_hle, agc_driver_depth_plane_buffer_memory, agc_driver_depth_plane_cpu_write_after_memo, agc_driver_depth_plane_cpu_write_after_page_unwatch, agc_driver_depth_plane_cpu_write_after_partial_unwatch, agc_driver_depth_plane_partial_store, agc_driver_depth_plane_partial_store_unimported, agc_driver_depth_storage_dispatch, and 3 more
VUID-vkCmdCopyImage-srcImage-01551 (11 tests): agc_driver_depth_clear_hle, agc_driver_depth_plane_buffer_memory, agc_driver_depth_plane_cpu_write_after_memo, agc_driver_depth_plane_cpu_write_after_page_unwatch, agc_driver_depth_plane_cpu_write_after_partial_unwatch, agc_driver_depth_plane_partial_store, agc_driver_depth_plane_partial_store_unimported, agc_driver_depth_storage_dispatch, and 3 more
VUID-VkPipelineColorBlendStateCreateInfo-pAttachments-00605 (2 tests): agc_driver_prepared_exports, agc_driver_srgb8_color_target

Skipped (40)

agc_driver_bda_reads_shadowed_results: skipped, the device reports watched host memory written when it imports or reads it
agc_driver_bda_reads_shadowed_results_per_build: skipped, the device reports watched host memory written when it imports or reads it
agc_driver_msaa_dcc_compared_import_2: skipped, the device reports host imports that keep write protection, so the write-watch probe compares none of them
agc_driver_msaa_dcc_compared_import_4: skipped, the device reports host imports that keep write protection, so the write-watch probe compares none of them
agc_driver_msaa_dcc_compared_import_8: skipped, the device reports host imports that keep write protection, so the write-watch probe compares none of them
agc_driver_sample_order_grid_reference: skipped, the device does not keep depth drawn at custom sample locations at those locations, even after a transition that names them: the composite of uniform renders holds the guest's depth in 0 of 3005 sloped samples, so no path can be compared with it bit for bit
agc_driver_sparse_dma_buf_import: skipped, host imports are unavailable: /dev/udmabuf is not accessible
agc_driver_stencil_ref_export: skipped, the device lacks VK_EXT_shader_stencil_export
agc_host_import_sparse: skipped, host imports are unavailable: /dev/udmabuf is not accessible
agc_import_settle_initial_dcc: skipped, the device has no deferred settle of host imports: imports under 1 MiB wait for one only on Windows
agc_unit_shadow_completion_store: skipped, the device reports watched host memory written when it imports or reads it
guest_http_uri_locale: 
guest_net_ctl: skipped, the host has no IPv4 address
guest_video_out_color_space: skipped, no display or Vulkan device: SDL_InitSubSystem(VIDEO | GAMECONTROLLER) failed: wayland,x11 not available
guest_video_out_hdr: skipped, no display or Vulkan device: SDL_InitSubSystem(VIDEO | GAMECONTROLLER) failed: wayland,x11 not available
guest_video_out_high_frame_rate: skipped, no display or Vulkan device: SDL_InitSubSystem(VIDEO | GAMECONTROLLER) failed: wayland,x11 not available
guest_video_out_output_mode: skipped, no display or Vulkan device: SDL_InitSubSystem(VIDEO | GAMECONTROLLER) failed: wayland,x11 not available
guest_video_out_resolution_status: skipped, no display or Vulkan device: SDL_InitSubSystem(VIDEO | GAMECONTROLLER) failed: wayland,x11 not available
guest_video_out_wqhd_detection: skipped, no display or Vulkan device: SDL_InitSubSystem(VIDEO | GAMECONTROLLER) failed: wayland,x11 not available
video_out_flip_controls: skipped, no display or Vulkan device: DisplayWindow: Vulkan support is either not configured in SDL or not available in current SDL video driver (offscreen) or platform
video_out_flip_cpuflip: skipped, no display or Vulkan device: DisplayWindow: Vulkan support is either not configured in SDL or not available in current SDL video driver (offscreen) or platform
video_out_flip_cpuflip_in_order: skipped, no display or Vulkan device: DisplayWindow: Vulkan support is either not configured in SDL or not available in current SDL video driver (offscreen) or platform
video_out_flip_eop: skipped, no display or Vulkan device: DisplayWindow: Vulkan support is either not configured in SDL or not available in current SDL video driver (offscreen) or platform
video_out_flip_eop_aftergpu: skipped, no display or Vulkan device: DisplayWindow: Vulkan support is either not configured in SDL or not available in current SDL video driver (offscreen) or platform
video_out_flip_eop_main: skipped, no display or Vulkan device: DisplayWindow: Vulkan support is either not configured in SDL or not available in current SDL video driver (offscreen) or platform
video_out_flip_fresh: skipped, no display or Vulkan device: DisplayWindow: Vulkan support is either not configured in SDL or not available in current SDL video driver (offscreen) or platform
video_out_flip_labels: skipped, no display or Vulkan device: DisplayWindow: Vulkan support is either not configured in SDL or not

Reproduce

git fetch origin c03772238413054855b92c6f1f21473c30014c05 bb04a4f1ea0175e9e4f149f0359a764d62ff3296
git checkout bb04a4f1ea0175e9e4f149f0359a764d62ff3296 && git merge --no-ff c03772238413054855b92c6f1f21473c30014c05
git submodule update --init --recursive
cmake -S . -B build -G Ninja -DCMAKE_BUILD_TYPE=Release -DBUILD_TESTING=ON -DANYPS5_ENABLE_SPIRV_TOOLS=ON
cmake --build build && cmake --build build --target libs
VK_DRIVER_FILES=/nonexistent/icd.json ctest --test-dir build -E '^(agc|shader|recompiler|videoout|video_out|guest_video_out|vulkan)|gpu' --timeout 120 --parallel 32 --output-on-failure
ANYPS5_REQUIRE_VULKAN=1 ANYPS5_NO_SHADER_CACHE=1 VK_INSTANCE_LAYERS=VK_LAYER_KHRONOS_validation \
  ctest --test-dir build -R '^(agc|shader|recompiler|videoout|video_out|guest_video_out|vulkan)|gpu' --timeout 120 --parallel 4 --output-on-failure
# failures are then re-run alone: --timeout 360 --parallel 2; new validation IDs: the test alone, twice, --parallel 1

_{Automated build and test on a contributor's NVIDIA machine; it does not replace upstream CI. A new comment is posted for each new push. Full logs.}

Tehshortbus mentioned this pull request Oct 29, 2026

AGC: Ghost of Yōtei (PPSA26344) stops on a draw that blends unclamped normalized color into two color targets #12796

Closed

github-actions Bot added the unblocks title Names a title ID on the Unblocks line label Oct 29, 2026

oneandonlydean self-assigned this Oct 29, 2026

github-actions Bot added the fix error Closes an issue (Fixes #N) label Oct 29, 2026

GiannisG1 mentioned this pull request Oct 29, 2026

test(agc): skip the unclamped normalized blend test without shaderStorageImageReadWithoutFormat #12968

Merged

12 tasks done

Tehshortbus mentioned this pull request Oct 29, 2026

perf(agc): let the sampled texture cache grow into host memory while a frame's textures don't fit #12684

Closed

12 tasks done

github-actions Bot added shader core/shader docs agc libSceAgc* labels Oct 30, 2026

oneandonlydean merged commit 3d2d264 into boykopovar:main Oct 30, 2026
14 checks passed

MarcZX10 mentioned this pull request Oct 30, 2026

fix(shader): recognize scalarized constant-stride image tables #13149

Merged

12 tasks done

Tehshortbus mentioned this pull request Oct 30, 2026

fix(agc): look up a merged color target again when a recovery evicted its resident image #13178

Merged

12 tasks done

This was referenced Oct 30, 2026

feat(agc): blend into 11_11_10 float color targets in the pixel shader #9020

Open

feat(agc): draw alpha-tested packed-target merges that write depth as a depth pass and an EQUAL merge pass #9335

Open

Navigation Menu

Repository navigation

feat(agc): merge a draw into more than one unclamped normalized color target - #12916

feat(agc): merge a draw into more than one unclamped normalized color target#12916
oneandonlydean merged 1 commit into
boykopovar:mainfrom
Tehshortbus:feat/multi-target-merge

Tehshortbus commented Oct 29, 2026

Uh oh!

github-actions Bot commented Oct 29, 2026 •

edited

Loading

Uh oh!

Tehshortbus commented Oct 29, 2026

Uh oh!

TechLevelZero commented Oct 29, 2026

Uh oh!

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

	`main` `b0465557fd`	this change
tests	7/7 pass	7/7 pass
`VUID-` lines	0	0
`was not found but was requested by env var VK_INSTANCE_LAYERS`	0	0

Conversation

Tehshortbus commented Oct 29, 2026

What

Tested

Checklist

Uh oh!

github-actions Bot commented Oct 29, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Tehshortbus commented Oct 29, 2026

Uh oh!

TechLevelZero commented Oct 29, 2026

✅ NVIDIA GPU test: passed

Uh oh!

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

github-actions Bot commented Oct 29, 2026 •

edited

Loading