[zaryabzip]

What

Implements sceSaveDataInitialize2 as sceSaveDataInitialize3 with the same parameter, the way sceSaveDataInitialize already forwards. A title that initializes through the v2 entry point now starts the same save-data session, and sceSaveDataTerminate ends it.

Sources agree that the three initialize entry points are one operation:

• shadPS4 runs the same initialization for Initialize2 and Initialize3;
• fpPS4 takes one parameter pointer for all three;
• SharpEmu also sends Initialize and Initialize2 to one common initializer.

Nothing is measured on a console. The public header gains the parameter, and Initialize2 comes out of the pending-declaration list in the debt file. That list, a single entry of over 6,000 characters, is split into three entries under the length limit, with the same wording and names.

Depends on #10674, which edits the same debt list line. This branch is #10674's head with current main merged in, plus one commit.

The new TestInitialize2 in GuestSaveDataPs4.cpp looks the call up at runtime, because main declares it without parameters. It checks that save-data memory setup is refused before the call, works after it, and is refused again after sceSaveDataTerminate. guest_savedata_ps4_tests now links ${CMAKE_DL_LIBS} for dlsym.

Tested

After the sync with main (8197ab3, a debt-text conflict only), Windows 11 and WSL2 reran it: all 7 guest_savedata* tests pass, with guest_savedata_ps4 3 of 3 on --repeat until-fail:3, and no warnings in the changed files. They also pass merged into main ee9a9a6. The earlier runs follow.

Windows 11 Home 25H2 (i7-11800H, WinLibs MinGW-w64 GCC 15.2.0 UCRT r7, CMake 4.3.1, Ninja 1.13.2, Release, BUILD_TESTING=ON) and Ubuntu 24.04.5 on WSL2 (GCC 13.3.0, CMake 3.28.3), both at bc6a5e4 merged into main 3a2b919 (clean merge):

• No warnings in the changed files on either host.
• ctest -R "^guest_savedata|^savedata_": 100% tests passed, 0 tests failed out of 13 on both hosts. savedata_memory_growth_read_failure is skipped on Linux, as always.
• guest_savedata_ps4 with --repeat until-fail:3: 3 of 3 on both hosts.
• Baseline: on main with only this test file and the ${CMAKE_DL_LIBS} line, guest_savedata_ps4 builds and then aborts with sceSaveDataInitialize2 not implemented on both hosts (Windows exit 0xc0000409).
• Every target that includes libSceSaveData/include/Api.hpp builds on both hosts.
• tools/check_conventions.py passes. Rechecked on reopening against main b9bf246: git merge-tree merges this cleanly with main. It overlaps feat(libSceSaveData): implement sceSaveDataCheckBackupData, sceSaveDataRestoreBackupData and sceSaveDataUmountWithBackup #12748 and feat(libSceSaveData): deliver backup completions to a registered event callback #13551 in the libSceSaveData.native debt line feat(libSceSaveData): declare sceSaveDataGetProgress with its known signature #10674 also edits. It also overlaps fix(libSceSaveData): keep sceSaveDataGetMountedSaveDataCount unimplemented #14150 on one neighbouring include/Api.hpp declaration. fix(libSceSaveData): keep sceSaveDataGetMountedSaveDataCount unimplemented #14150 implements sceSaveDataGetMountedSaveDataCount, not this export. Whichever merges second needs a one-line rebase.

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: feat(libSceSaveData): declare sceSaveDataGetProgress with its known signature #10674
• Fixes: none
• Unblocks: none
• GPUs tested: n/a
• AI-assisted: yes (Claude Code)

[github-actions]

Progress report (7feb37e → c5e2f0e)

📈 System libraries: 26.37% (+0.0027%, +1 implemented functions)

✅ 1 implemented

Library	Name
libSceSaveData.native	sceSaveDataInitialize2

Changed libraries (implemented exports)

Library	Before	After	Change
libSceSaveData.native	51/120 (42.50%)	52/120 (43.33%)	+0.83%

[github-actions]

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 14908	docs/dev/debt/libSceSaveData.native.md L25
PR 15097	docs/dev/debt/libSceSaveData.native.md L24-25

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 13486	core/libs/prx/libSceSaveData.native/Export.cpp core/libs/prx/libSceSaveData/include/Api.hpp docs/dev/debt/libSceSaveData.native.md
PR 14150	core/libs/prx/libSceSaveData.native/CMakeLists.txt core/libs/prx/libSceSaveData.native/Export.cpp core/libs/prx/libSceSaveData/include/Api.hpp docs/dev/debt/libSceSaveData.native.md
PR 14212	docs/dev/debt/libSceSaveData.native.md
PR 15042	core/libs/prx/libSceSaveData.native/Export.cpp

[zaryabzip]

Merged current main into this branch (8197ab3) to clear a conflict in docs/dev/debt/libSceSaveData.native.md. main had dropped sceSaveDataAbort, sceSaveDataClearProgress and sceSaveDataGetMountedSaveDataCount from the pending-declarations line, and the resolution keeps #10674's split entries without those three. No code was changed by hand. #10674 itself now conflicts with main too, so this PR still waits on it.

[zaryabzip]

Closing for now to stay within the 20-open-PR limit. This one can't land before #10674, which conflicts with main; the branch stays as it is and I'll reopen it once #10674 is merged.

[zaryabzip]

Reopened: I now have 20 merged PRs, so the open-PR limit no longer applies to me. This still lands after #10674. Rechecked against main b9bf246: it merges cleanly with main, and the only overlaps are the debt line it shares with #10674, #12748 and #13551, plus one neighbouring declaration in include/Api.hpp with #14150, which implements a different export.

[BunnyFl]

Windows 11, WinLibs GCC 15.2.0 r7, Release. Head 8197ab3 merged into upstream/main 93dbd0f:

• ctest -R '^guest_savedata': 8/8 pass, including guest_savedata_ps4 with the new TestInitialize2 and guest_savedata_dialog.
• Negative control: with main's libSceSaveData.native/Export.cpp and libSceSaveData/include/Api.hpp, guest_savedata_ps4 aborts with sceSaveDataInitialize2 not implemented. So the test reaches the export through the runtime lookup, as intended.
• tools/check_conventions.py --base upstream/main --head 8197ab33fd: passes. The split debt entries clear #10674's debt-length warning.

Forwarding to sceSaveDataInitialize3, as sceSaveDataInitialize already does, matches the three references you list. I approved #10674, which this builds on, separately.

[TechLevelZero]

✅ AMD GPU test: passed

PR head c5e2f0e merged onto main 7feb37e · tested 2026-11-03 03:49 UTC · 14 min

GPU	AMD Radeon RX 6300 (RADV NAVI24), driver Mesa 25.2.8-0ubuntu0.24.04.4, Vulkan 1.4.318
Host	Ubuntu 26.04.1 LTS, kernel 7.0.0-42-generic, container ubuntu:24.04
Logs	build and test logs for this run
Build	✅ passed (Release, -DANYPS5_ENABLE_SPIRV_TOOLS=ON)
Tests	✅ 1,132 passed · 4 failed (0 new, 3 also fail on main, 1 unstable on main) · 7 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	1,143 of 3,743 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 (543 tests run on the GPU)
Slow tests (>30 s, GPU tests timed with validation layers)	⚠️ agc_driver_pipeline_journal 120.2 s, ⚠️ public_headers 65.0 s, ⚠️ agc_driver_image_sample_unnormalized 36.4 s, ⚠️ agc_driver_lazy_resolved_abi 30.5 s

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

agc_driver_pipeline_journal: failed in 5 of the last 7 main runs

Failing on main too (3, not counted against this PR)

agc_driver_image_table
agc_driver_native_depth_bda_clear
agc_driver_native_multisample_depth

Validation IDs already present on main (4 IDs in 466 tests)

VUID-VkDeviceCreateInfo-pNext-pNext (466 tests): 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, agc_driver_bindless_images, and 458 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_storage_dispatch, agc_driver_depth_surface_from_storage, agc_driver_dispatch_move_depth, agc_driver_stencil_null_feedback
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_storage_dispatch, agc_driver_depth_surface_from_storage, agc_driver_dispatch_move_depth, agc_driver_stencil_null_feedback
VUID-VkPipelineColorBlendStateCreateInfo-pAttachments-00605 (1 test): agc_driver_srgb8_color_target

Skipped (7)

agc_driver_copied_writer_label_order: skipped, the device reports a minStorageBufferOffsetAlignment of 4, so the dispatch binds its record at +0x1004 in place and nothing is written back; the test needs an alignment above 4
agc_driver_float16_unused_nan_payloads: skipped, the device has no native f16 conversions with RTE rounding and kept f16 denormals
agc_driver_linear_target_direct_write_completion: skipped, the device does not keep host imports write-watched, so no completion stamp over the target is tracked
agc_driver_ngg_vertex_wave_reduction: skipped, s3 describes a 32-lane wave only on 32-lane subgroups, and the device has 64 lanes
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

Reproduce

git fetch origin c5e2f0ed94d9b15c0f471c0b62792e6951279803 7feb37ee3e30e10139b483fb4df689d5b5b69ab8
git checkout 7feb37ee3e30e10139b483fb4df689d5b5b69ab8 && git merge --no-ff c5e2f0ed94d9b15c0f471c0b62792e6951279803
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 12 --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 16 --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 AMD machine; it does not replace upstream CI. A new comment is posted for each new push. Full logs.}