pr_overlap.py took every added line with APS5_VABI before a name and a
parenthesis as an export, so a declaration such as
extern "C" int* APS5_VABI __error_nid_postfix(); made each pull request
that adds it "implement the same function" as all the others. A line
that ends in a semicolon is a declaration and is skipped; every
definition's first line in the tree ends in a brace, a parenthesis or
a comma.
Registers pr_overlap_exports.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
The Linux job reports "100% tests passed out of 371" with 20 agc_driver execution tests skipped
(llvmpipe reports subgroup size 8, so the tests that need a wider wave exit 77), and the Windows
job reports "100% tests passed, 0 tests failed out of 368" with 143 skipped - the runner has no
Vulkan driver and ANYPS5_REQUIRE_VULKAN is not set there, so 143 of the 147 GPU execution tests
silently do not run. ctest counts a skip as a pass, so neither job notices when the set grows.
Add tools/check_skips.py. It reads the file ctest writes with --output-junit and fails when a test
skips for a reason that is not expected:
- linux: a skip is expected only when the test's own output reports that the device's subgroups
are too narrow for it, so adding a wave32 test does not need the rule edited;
- windows: any agc_driver test, since the runner has no driver for them, while a skip of anything
else fails.
Both jobs now pass --output-junit and run the check after ctest. The JUnit file needs no
redirection, so the console output stays visible and no pipe can hide a failing ctest.
Verified here with the CI command and the CI device (lavapipe, subgroup 8): the real run reports
20 skipped, all expected; a 21st wave32 test skipping for the same reason passes; a skip whose
reason is something else exits 1; a file with 143 agc_driver skips passes with --job windows and
one non-agc skip in it exits 1.
clang assembled every body as wave32, so a --wave64 body that used vcc
or an SGPR pair as a condition failed with "operands are not valid for
this GPU or mode". Pass -mwavefrontsize64 on wave64 runs.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The mesh test was built only on request and never registered, so the
translation of NGG geometry programs to mesh shaders ran in no test
run. On an RTX 4060 it passes. Register it with the other AGC driver
tests.
Three cases would fail or mislead under ctest. Without a Vulkan device
the test failed, and the Windows job has none. A device without
VK_EXT_mesh_shader exited 0, so ctest reported a pass for a test that
drew nothing. On llvmpipe, with the Linux job's Mesa 25.2.8 and Ubuntu
24.04's LLVM 20.1.2, every draw that builds a mesh pipeline stops the
process with "LLVM ERROR: Cannot emit physreg copy instruction"; Mesa
26.2.4 with LLVM 23.1 segfaults in LLVM's code generation instead. The
SPIR-V of all 18 modules the test builds passes spirv-val. Mesa #14376
reports the same error for lavapipe mesh shaders that index shared
memory per invocation; that the cause is the same is not established.
The test now opens its device with OpenVulkanTestDevice and exits 77
without a device, without VK_EXT_mesh_shader and on llvmpipe, saying
why. ANYPS5_REQUIRE_VULKAN still turns a missing device into a
failure. The Linux job therefore skips the test; it runs on hosts with
a GPU.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A converted game fails at launch on the first system function that no
built library exports, one function per run. tools/import_audit.py reads
the registry written by relinker --registry and classifies every imported
NID in one pass:
- implemented: a built library exports it
- stub: exported, but the whole function calls NotImplemented_nid_no_patch
- absent: no built library exports it, so the loader fails
- module: the import names a file the title ships itself
Exports are read from the built .prx files (ELF dynsym and PE export
table), the same view the loader has. Stubs come from the source tree
through tools/progress.py, and NIDs through tools/nid_names.py. The report
also lists needed libraries without a file and imports found only in a
different library than the one they name. --names adds names from an
aerolib.csv file and never downloads one; --json writes the full result
stamped with the repository commit.
The test builds ELF and PE libraries byte by byte, checks that every
reference lands in exactly one class, and pins the NID of
sceMsgDialogProgressBarSetValue (wTpfglkmv34) as an external anchor.
Registered in CTest as import_audit.
guest_formatting_tests runs only on Windows and windows_exception_tests only
exists there, so the stale-entry check failed the Windows job.
Co-Authored-By: Claude <noreply@anthropic.com>
agc_driver_bda_device_tests is registered on main, so its entry hid any
future loss of that registration.
Co-Authored-By: Claude <noreply@anthropic.com>
The Windows job failed the check with two names that are not test
executables. `libhost_thread_local_test_library` is a support library
whose RUNTIME_OUTPUT_DIRECTORY puts it in the test directory; on Linux
its output is a library artifact and never lands there, which is why the
Linux job stayed green. `windows_exception_tests` is a MINGW-only test
executable that no `add_test` registers.
Ninja prints the rule beside each target, and CMake links an executable
with `CXX_EXECUTABLE_LINKER__<target>_<config>` and a shared library with
`CXX_SHARED_LIBRARY_LINKER__...`, on every platform. Requiring EXECUTABLE
in the rule leaves the library out and keeps the Linux count at 362.
`windows_exception_tests` is a real instance of the class this check
looks for, so record it beside the other five rather than fail the job
that found it.
The Windows job failed on the new step with
"no test executables found under build". Ninja prints the target names
with the platform separator, so the Windows build graph lists
tests\name_tests.exe where Linux lists tests/name_tests, and the reader
matched neither the backslash nor the .exe suffix.
Accept both separators and a .exe suffix. Verified by driving the
function with recorded Ninja output for both platforms, which now
return the same names, and by re-running the check against the Linux
tree, which still reports 362 test executables and the same five.
A test executable can be declared with add_test_executable and never
registered with add_test. It is then built and never run, and ctest
reports a full pass over what it knows, so nothing in CI says so.
Measured on main, five test executables are in this state: the driver's
mesh and recorder tests, the guest formatting tests, the video out flip
tests and the BDA device tests.
tools/check_test_registration.py compares the test executables CMake
builds, read from the build graph, against the add_test calls in the
generated CTestTestfile.cmake files, which are written at configure time
and so do not depend on the build having finished. It fails on a test
executable that nothing runs, and records those five by name so the job
stays green until they are dealt with.
Both jobs run it right after configure, before anything is built.
check_conventions.py normalized link targets with os.path.normpath, which gives backslash paths on Windows, so every link added in docs failed the doc-link check and a silent stub listed in TechnicalDebt was not found. Links are repository-relative POSIX paths, so they are normalized with posixpath.
git merge-tree takes trees with --merge-base only from git 2.45. On
older git, such as 2.43 on Ubuntu 24.04, the pairwise merge failed and
merge() read the first line of empty output, so --branch stopped with
an IndexError. The merged trees are now wrapped in commits on top of
the base, which git merge-tree accepts with --merge-base from 2.40, and
a failing git merge-tree raises with git's message.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
IMAGE_LOAD_PCK2/PCK4, IMAGE_LOAD_MIP_PCK2/PCK4, IMAGE_STORE_PCK2/PCK4 and
IMAGE_STORE_MIP_PCK2/PCK4 are separate MIMG opcodes (112/113, 115/116,
118/119, 121/122 on RDNA1 and RDNA2 in AMD's generated tables), not dmask
forms of IMAGE_LOAD_PCK (2), IMAGE_LOAD_MIP_PCK (4), IMAGE_STORE_PCK (10)
and IMAGE_STORE_MIP_PCK (11). The decoder has no entries for them and
throws "unsupported MIMG opcode", so they are no longer credited.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
BREAKING CHANGE: CLI and Python callers must supply all seven float
controls. Add clamp, rounding and FP16 overflow settings so hardware
measurements cannot silently use an unintended mode.
Refs #981
ImageLoadPck, ImageLoadMipPck, ImageStorePck and ImageStoreMipPck are
decoded and implemented for 2- and 4-dword transfers, but the disassembler
and rdna_isa.txt list them under separate _PCK2 and _PCK4 mnemonics.
Existing tests in ImageLoadPacked and ImageStorePacked already cover
these dword counts.
The decoder maps SOPK opcodes 0x17 to 0x1a to SWaitcnt, but no test encoded them and tools/progress.py only counted S_WAITCNT_VSCNT. The test runs the four waits with a null and a live register field and checks that SCC and the register are left unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
v_add_co_u32, v_sub_co_u32 and v_subrev_co_u32 (VOP3 0x30f, 0x310, 0x319)
decode to VAddI32, VSubI32 and VSubrevI32 and had no execution test.
tools/progress.py listed the first and the last as variants of those
opcodes but not v_sub_co_u32, which it reported as missing.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
VOP2 0x2b, 0x2c and 0x2d decode to the opcodes of v_mac_f32, v_madmk_f32
and v_madak_f32 and differ only by the encoding, which keeps them fused.
The progress map derives names from the opcode enum, so it listed the three
fused forms as missing.
The new test decodes the six encodings and checks that the translation
rounds the product separately for 0x1f-0x21, keeps 0x2b-0x2d fused, and
reads the accumulator and the literal from the right operand.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
MIMG opcode 0x33 samples with explicit derivatives, an LOD clamp and a
texel offset. The three address flags already exist; only the opcode was
missing from the decoder table.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
SCbranchCdbg decodes SOPP 0x17-0x1a and STtracedata decodes 0x16 and 0x28,
so S_CBRANCH_CDBGSYS, S_CBRANCH_CDBGUSER, S_CBRANCH_CDBGSYS_OR_USER,
S_CBRANCH_CDBGSYS_AND_USER and S_TTRACEDATA_IMM were reported as missing
although the decoder handles them and existing tests cover them.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
libSceAjm.native routes 13 exports through a static AjmStub helper that
calls NotImplemented_nid_no_patch, so the progress map counted them as
implemented. Calls to such helpers in the same file now count as stubs.
The compare report also lists functions that went back to stubs in its
header.
Co-Authored-By: Claude <noreply@anthropic.com>
progress.py stores the names of implemented and pending items, can
measure another source tree with --root and prints a Markdown diff of
two runs with --compare. The Progress report workflow compares every
pull request with its base, and Progress comment posts the result as a
single comment, updated on every push and removed when progress no
longer changes. Commenting runs as a separate workflow_run so pull
requests from forks never run with write permissions.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
tools/progress.py counts implemented prx functions and RDNA instructions
recognized by the shader recompiler, and renders treemaps, badges and an
index page. The Progress workflow deploys them to GitHub Pages on every
push to main, so nothing generated is committed to the repository.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
PrintOrb
Jan Šlégl
Pedro
darhebkf
Dextroid17
Dean Galvin
Dean
David
Alex
SK Arin
Gerard L.
Daniele Giovane
abderahmane-ai
Adrià Franch
Adrià
Amirtheshwaran Navaneethakrishnan
Hashaam Zafar
popcoder565
Sanjaysrinivas
ebosido444-ai
ploysius
Muhammad Umair
Kikoarl
Nahuel Genchi
Pouare514
DotDebian
SamG-Coder
boykopovar