[xXPBSXx]

What

agc_storage_pending_after_unwatch takes about 10 s under ASan and UBSan, and agc_driver_recorder, which runs the same checks in its full pass, went from under 10 s to the 30 s limit in the sanitizer runs since #10026 added them. Most of that time is spent building failure messages that are never used: the checks over the watched units pass each word to Require with a message assembled from std::to_string and, for the stored results, the per-unit summary, so every passing word pays for a string. The message is now built only when the check fails; the checks themselves and what they cover are unchanged. CONTRIBUTING asks to shrink a test that gets close to the limit instead of raising its timeout. #15441 raises the timeout of agc_driver_recorder to 120 s in ASan builds; with this change the test takes about 6.7 s instead of 16 to 17 s under ASan on the machine below, so #15441 would remain only as a reserve. Whether it is still needed is for the reviewers.

Tested

Ubuntu 24.04 under WSL2, GCC 13.3.0, configured as sanitizers.yml does (Release, ANYPS5_ENABLE_ASAN=ON, -fsanitize=undefined -fno-sanitize=vptr, same ASAN_OPTIONS and UBSAN_OPTIONS), three runs each on main 1e24993fa3 and on this change on the same main:

	main	this change
agc_storage_pending_after_unwatch	9.91 / 9.83 / 10.53 s	1.07 / 1.04 / 1.01 s
agc_driver_recorder	17.00 / 16.32 / 16.34 s	6.67 / 6.67 / 6.73 s

All runs passed. tools/check_conventions.py --base upstream/main passes.

Overlap: all 1827 open pull requests were checked on 2026-11-05 at 09:30 UTC against main 7ad9749ed0; 56 change tests/Recorder.cpp. 23 merge with and without this change without a conflict, 33 already conflict with main without it, and none conflicts because of this change; none changes pendingAfterLeavingWatchTests.

Checklist

• Based on current main; no other open PR implements the same functions
• One topic per PR; follow-ups go in a new PR
• No comments in code except technical debt
• Unimplemented paths throw (NotImplemented_nid_no_patch); silent stubs are listed under Silent stubs in their library's technical debt file
• No notes, investigation .md files or images added to the repository (attach them to this PR)
• New third-party code is a submodule built from source
• Generic behaviour, not specific to one title
• Depends on: none
• Fixes: none
• Unblocks: none
• GPUs tested: n/a (test-only change to how failure messages are built)
• AI-assisted: yes

AI-assisted (Claude Code).

🤖 Generated with Claude Code

[JJenkx]

Windows check of head 1d899f2b7a, merged onto main aafd88b74b (74993d9a60, clean merge; Recorder.cpp auto-merged).

Setup: Windows 11 Enterprise 22621, WinLibs GCC 15.2 UCRT, Release, Radeon RX 6700 XT, AMD driver 26.10.1, run with ANYPS5_REQUIRE_VULKAN=1 and Vulkan validation layers 1.4.363 enabled. The hosted Windows check does not run these tests on pull requests.

• All 115 tests from agc_driver_recorder_tests pass in 3 of 3 runs, none skipped. The validation layer loaded 336 times, with no VUID- errors.

AI-assisted: yes.

[TechLevelZero]

✅ NVIDIA GPU test: passed

PR head 1d899f2 merged onto main c5bd55a · tested 2026-11-05 11:23 UTC · 12 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	✅ 1,231 passed · 1 failed (0 new, 1 also fail on main) · 4 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	1,236 of 3,972 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 (547 tests run on the GPU)
Slow tests (>30 s, GPU tests timed with validation layers)	⚠️ agc_driver_pipeline_journal 106.2 s

Failing on main too (1, not counted against this PR)

agc_driver_image_table

Validation IDs already present on main (9 IDs in 466 tests)

VUID-VkDeviceCreateInfo-pNext-pNext (466 tests): agc_driver_address_draw_space_reuse, agc_driver_address_passes, agc_driver_alias_copy_label, agc_driver_alpha_to_coverage, agc_driver_array_layer_target, agc_driver_bc_block_copy, agc_driver_bda_anchor_reads, agc_driver_bda_span_reads, and 458 more
VUID-VkShaderModuleCreateInfo-pCode-08737 (14 tests): agc_driver_dpp16_operands, agc_driver_dpp8_pk_fmac, agc_driver_dx10_clamp, agc_driver_f16_fma_denormal_modes, agc_driver_f16_fma_invalid_nan, agc_driver_float16_fma_rounding, agc_driver_float16_modifiers, agc_driver_float16_native, and 6 more
VUID-VkShaderModuleCreateInfo-pCode-08739 (14 tests): agc_driver_dpp16_operands, agc_driver_dpp8_pk_fmac, agc_driver_dx10_clamp, agc_driver_f16_fma_denormal_modes, agc_driver_f16_fma_invalid_nan, agc_driver_float16_fma_rounding, agc_driver_float16_modifiers, agc_driver_float16_native, and 6 more
VUID-VkShaderModuleCreateInfo-pCode-08741 (14 tests): agc_driver_dpp16_operands, agc_driver_dpp8_pk_fmac, agc_driver_dx10_clamp, agc_driver_f16_fma_denormal_modes, agc_driver_f16_fma_invalid_nan, agc_driver_float16_fma_rounding, agc_driver_float16_modifiers, agc_driver_float16_native, and 6 more
VUID-vkCmdCopyImage-srcImage-01548 (7 tests): agc_driver_depth_clear_hle, agc_driver_depth_plane_buffer_memory, agc_driver_depth_plane_partial_store, agc_driver_depth_plane_sampling, agc_driver_depth_storage_dispatch, agc_driver_depth_surface_from_storage, agc_driver_dispatch_move_depth
VUID-vkCmdCopyImage-srcImage-01551 (7 tests): agc_driver_depth_clear_hle, agc_driver_depth_plane_buffer_memory, agc_driver_depth_plane_partial_store, agc_driver_depth_plane_sampling, agc_driver_depth_storage_dispatch, agc_driver_depth_surface_from_storage, agc_driver_dispatch_move_depth
VUID-VkPipelineColorBlendStateCreateInfo-pAttachments-00605 (1 test): agc_driver_srgb8_color_target
VUID-VkPipelineShaderStageCreateInfo-module-parameter (1 test): agc_driver_float16_unused_nan_payloads
VUID-VkPipelineShaderStageCreateInfo-pSpecializationInfo-06849 (1 test): agc_driver_float16_unused_nan_payloads

Skipped (4)

agc_driver_gcn_display_present: skipped, no display: Vulkan support is either not configured in SDL or not available in current SDL video driver (offscreen) or platform
agc_driver_scattered_runs: skipped, host imports are unavailable: /dev/udmabuf is not accessible
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

Reproduce

git fetch origin 1d899f2b7a83c4c20a96388c14fdfa7fca46432e c5bd55a89a07d39840867cc671f622f3d7c4869c
git checkout c5bd55a89a07d39840867cc671f622f3d7c4869c && git merge --no-ff 1d899f2b7a83c4c20a96388c14fdfa7fca46432e
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.}