This project is mirrored from https://review.mlplatform.org/tosa/reference_model.
Pull mirroring updated .
- Jul 14, 2025
-
-
Peng Sun authored
Signed-off-by:
Peng Sun <peng.sun@arm.com> Change-Id: Ie16ccdcdd2d109871975faa0b32589c82979be78
-
Luke Hutton authored
This section adds a comment stating that running operations outside the constraints of level=8k may lead to undefined behaviour. Change-Id: I5f6ea12ecce96ce6d8a1cb7a14de7e0b73c40bb9 Signed-off-by:
Luke Hutton <luke.hutton@arm.com>
-
This commit removes some ambiguity in the README by merging duplicate sections. Signed-off-by:
Luke Hutton <luke.hutton@arm.com> Change-Id: If5eb9e4352b0f98754a8fd64771c807791813ea4
-
Signed-off-by:
Iliyan Georgiev <iliyan.georgiev@arm.com> Change-Id: I5114c3b1655974371101cdb3a61fa9a2cee12900
-
Introduced the following tests: - a sanity test to verify basic model functionality(in memory). - Tests for DType variants to validate behaviour for different data types. - Unit tests generated using tosa_verif_build_tests. Also adding selective test execution for unit tests via pytest markers ('pyb_smoke' and 'pyb_fullsuite'). Sanity and DType tests run by default. To run unit tests, use: pytest -m "pyb_smoke" pytest -m "pyb_fullsuite" Conflict files: .gitignore Signed-off-by:
Deeptanshu Sekhri <deeptanshu.sekhri@arm.com> Change-Id: I7ae3c25bc76362ac0374b07ba8ef79e154c214fc
-
- Jul 11, 2025
-
-
Introduced a "Versioning" section describing the Calendar Versioning (CalVer) scheme used by the project. Change-Id: Ie7b5634c0bb4c39bde8fc9de53ba53501b7cbf0d Signed-off-by:
Peng Sun <peng.sun@arm.com>
-
- Jul 01, 2025
-
-
Change-Id: Ia2c8507bc2559738e8011056b7526f7b3f2e16a6 Signed-off-by:
Yuvaraj Venkatesh <yuvaraj.venkatesh@arm.com>
-
- Jun 30, 2025
-
-
Jeremy Johnson authored
To support BF16 on Windows correctly, add support for floating point fixed data as well as integer. Signed-off-by:
Jeremy Johnson <jeremy.johnson@arm.com> Change-Id: Iad8f16ccff2df56df3f0f67e41e2fba5ecd149e9
-
Jeremy Johnson authored
Enable long path name support by including manifest file when building dlls and exes on Windows. Update tosa_verif_run_tests: * Use replace instead of rename when moving results files, to mimic behaviour of Linux "mv" on Windows. * Also rework the verify results loop to rename the result files first and then verify them. * Remove expected_result_file support as not allowed in desc.json. Change-Id: I9973f22a62371dafc6de030d886349faf18f276b Signed-off-by:
Jeremy Johnson <jeremy.johnson@arm.com>
-
- Jun 27, 2025
-
-
Ian Tayler Lessa authored
This now correctly computes the ULP bound relative to the maximum absolute input, while previously we were computing it relative to the reference output. Extend the unit test coverage for max resize verification to catch regressions and other bugs in bfloat16 and float16 Refactor a calcAbsErrorBound function matching the spec to use across verification code. Signed-off-by:
Ian Tayler Lessa <ian.taylerlessa@arm.com> Change-Id: I04b08d5d160258c560dbfde3b70247d76bf3f225
-
Ian Tayler Lessa authored
RELATIVE allows error relative to some fixed 'max' value, i.e. fixed error for all outputs in an operation. Renaming it to MAX_RELATIVE allows us to add a second mode, OUTPUT_RELATIVE, which will allow error relative to each output in an operator. The OUTPUT_RELATIVE mode will be coming as a separate patch. Signed-off-by:
Ian Tayler Lessa <ian.taylerlessa@arm.com> Change-Id: Ia70549c19ec781b1172cb1a2a7fd0da581489ca5
-
- Jun 26, 2025
-
-
Jeremy Johnson authored
Fix for reading and correctly checking the test requirements versus the chosen profiles/extensions. Add tests for checking all combinations. Signed-off-by:
Jeremy Johnson <jeremy.johnson@arm.com> Change-Id: I8991cdacb34701c156a69c58aa549d3cfec73802
-
Philip742 authored
Signed-off-by:
Philip Wilkinson <philip.wilkinson@arm.com> Change-Id: I9452032305f99e6e90ee920d9bec01b3e95fdf93
-
- Jun 24, 2025
-
-
Philip742 authored
Note does not cover test commands Signed-off-by:
Philip Wilkinson <philip.wilkinson@arm.com> Change-Id: Ib1cb340290cf5c663e00185aaaaad68388bb4477
-
- Jun 20, 2025
-
-
Ian Tayler Lessa authored
Previously we were causing an internal error in allocation instead of signlaling an ERROR in the graph. Also fixes the shape given to the Eigen tensor used to represent shape_t values of rank 0. Signed-off-by:
Ian Tayler Lessa <ian.taylerlessa@arm.com> Signed-off-by:
Peng Sun <peng.sun@arm.com> Change-Id: I682fe0360b002b04af7b08e590d4f1acb4850d5c
-
Peng Sun authored
- Introduced a new typedef: registration_callback_v2_t, which uses const char* instead of std::string for domain and operator names. - Updated MasterRegistry::register_function to include a deprecated overload that forwards std::string parameters to the new const char* version. - Updated loadSharedLibs() to support both v2 and legacy v1 custom operator libraries. It attempts to resolve the v2 interface first, and falls back to v1 if necessary. Change-Id: I9c0d7b768c1b9be5daeaf9617c6fd46000a32758 Signed-off-by:
Peng Sun <peng.sun@arm.com>
-
- Jun 19, 2025
-
-
Philip742 authored
Signed-off-by:
Philip Wilkinson <philip.wilkinson@arm.com> Change-Id: Ia29ed6c2f0cc41061c58997de1f5487c900196e4
-
Jeremy Johnson authored
Use distrubution classes to make random number generation behave the same on Windows and Linux Added generate unit test to check if values match on a given platform Signed-off-by:
Jeremy Johnson <jeremy.johnson@arm.com> Change-Id: Ibf5a452e12326cf4ca5194c7ab83a9e1bdb1ee66
-
Philip742 authored
Signed-off-by:
Philip Wilkinson <philip.wilkinson@arm.com> Change-Id: I3d54dfa61186a00aaee4d548041dbdac1d509788
-
- Jun 17, 2025
-
-
- Implement ewise unary operator tests covering floating-point edge cases - Covers ABS, FLOOR and CEIL - Adjust ABS operator implementation to correctly handle sign of zero inputs. - Register new elementwise unary op test file (ewise_unary_tests.cpp) in CMakeLists. Signed-off-by:
Jerry Zhao <Jerry.Zhao@arm.com> Signed-off-by:
Ian Tayler Lessa <Ian.TaylerLessa@arm.com> Change-Id: I03a74877afd195125dcca03dce6cd0903c8a142e
-
- Jun 13, 2025
-
-
Peng Sun authored
- Apply element-wise `max(abs(x), normal_min)` to handle subnormal values and ensure numerical stability in bounds mode - Updated ops: CONV2D, CONV3D, DEPTHWISE_CONV2D, MATMUL, TRANSPOSE_CONV2D, FFT2D, RFFT2D, AVG_POOL2D Co-authored-by:
Ian Tayler Lessa <ian.taylerlessa@arm.com> Signed-off-by:
Peng Sun <peng.sun@arm.com> Change-Id: I5a2c3a44be647c5acb08d67e045af8250f9380bc
-
- Jun 12, 2025
-
-
Jeremy Johnson authored
* Paths unsupported in load library for python 3.9 * RESCALE min/max calculations not promoted to int64 * Utils product() calculation also not promoted to large enough storage * Fixed color printing output * Made conformance generator build ops in sorted order for consistence * Updated default model file locations and tool help messages * Fixed TransposePermsOutputShapeMismatch ERROR_IF Change-Id: I18cfa49e7016ee5a5453b5669c39e587175a15f5 Signed-off-by:
Jeremy Johnson <jeremy.johnson@arm.com>
-
- Previously, SYSTEM was incorrectly applied to normal includes as well - Adjusted CMake to restrict SYSTEM usage to external dependencies (e.g., FlatBuffers) Signed-off-by:
Yuvaraj Venkatesh <yuvaraj.venkatesh@arm.com> Change-Id: If484b7fb8a708d571b47fb1f17c456d3e6e9849f
-
- Jun 11, 2025
-
-
Jeremy Johnson authored
Only produce any CPP files in lazy data gen mode. Signed-off-by:
Jeremy Johnson <jeremy.johnson@arm.com> Change-Id: I32e25eb078722e16e2d2d3c2b5f7eb84290f695d
-
- Jun 09, 2025
-
-
Jeremy Johnson authored
Due to incorrect tag generation for dynamic tests causing tags to be corrupted on other tests. Signed-off-by:
Jeremy Johnson <jeremy.johnson@arm.com> Change-Id: I00188d7a06f62c41e7fedba24d151dec6479522d
-
Deeptanshu Sekhri authored
- Use find_package to link Python headers and libs for MSVC - Replace VLA with std::vector and copy buffer to py::array_t for safe memory handling - Update setup.py to detect and copy .so or .pyd extensions based on platform Signed-off-by:
Deeptanshu Sekhri <deeptanshu.sekhri@arm.com> Change-Id: I64e8e85eb3664201cd59b277cc605983bc5adf1c
-
Deeptanshu Sekhri authored
Improved clarity and structure for the ModelRunner Python bindings section. Added build instructions with usage example. Signed-off-by:
Deeptanshu Sekhri <deeptanshu.sekhri@arm.com> Change-Id: I8176f83cc8a92260dea9b4cab1b5d5c6fa7ec0f0
-
- Jun 05, 2025
-
-
- Replace all instances of half_float::half with ct::float16 - Remove some special code in fpTrunc for handling bfloat16 as cfloat.h can do that special handling now - Update serialization_lib submodule - Update unit tests to match new serialization_lib version Backport notes: - A change to accomodate Location in unit tests was added because otherwise the serialization lib submodule cannot be updated Signed-off-by:
Eirene Vlassi Pandi <eirene.vlassipandi@arm.com> Signed-off-by:
Ian Tayler Lessa <ian.taylerlessa@arm.com> Change-Id: I6374f2712e943a756322940eea753b7c2f57f943
-
- Jun 03, 2025
-
-
Jeremy Johnson authored
Use a base bound value of 2 instead of 1 for SIN operations. Refactor compliance meta data function in test generator to be less operator specific for ABS_ERROR compliance Signed-off-by:
Jeremy Johnson <jeremy.johnson@arm.com> Change-Id: I90182ce20387d5dda1d0e148f05a2edf830648fa
-
Jeremy Johnson authored
Adds a lower bound of 0.5 ulp relative to 1.0 to allow implementations a larger error for input values close to 1. Signed-off-by:
Jeremy Johnson <jeremy.johnson@arm.com> Change-Id: Id97f5c9802d7896bf10db6fef475b182ffed5c21
-
- May 30, 2025
-
-
Tai Ly authored
This patch changes to use TosaSerializationShape in serialization_lib. Lazy Data Generation mode for tests now embeds CONST_SHAPES in the graph. TOSA_MLIR_TRANSLATOR_REFSPEC: refs/changes/09/684209/28 Signed-off-by:
Tai Ly <tai.ly@arm.com> Signed-off-by:
Jeremy Johnson <jeremy.johnson@arm.com> Change-Id: I754a6085b83bff04f0468e81c3d72f1221dcbde2
-
CID: 3055367, Resource leak Signed-off-by:
Yuvaraj Venkatesh <yuvaraj.venkatesh@arm.com> Change-Id: Ibfcd4146aadc930e3593bcace119543af70a1e43
-
- May 28, 2025
-
-
Jeremy Johnson authored
Allow error bounds ranges, to support CAST ULP range of 1.0 to 0.5 Add unit tests for normal ULP with/without lower setting Signed-off-by:
Jeremy Johnson <jeremy.johnson@arm.com> Change-Id: I5a8eefa39999911b61c8e04f2cde64921c53c69e
-
- May 20, 2025
-
-
Yuvaraj Venkatesh authored
This warnings occurs only in v1.0 branch Change-Id: I84471318fa7ecb452041b5fac9de1e2d77e51eaf Signed-off-by:
Yuvaraj Venkatesh <yuvaraj.venkatesh@arm.com>
-
- May 19, 2025
-
-
- update the version of tools and packages - add high-level featue description - create verif/README.md to introduce detailed CTS information Change-Id: Id80810a700ddf8040954857813cce7464700f45b Signed-off-by:
TatWai Chong <tatwai.chong@arm.com>
-
Jeremy Johnson authored
Includes changes: * 83c3a9c Fix warnings specific to GCC 9.4 * 43fdb52 Allow adding placeholders without data * 3f7f423 Generate strict json by default when serializing Signed-off-by:
Jeremy Johnson <jeremy.johnson@arm.com> Change-Id: Ic3b9615197938a5ae55ff2932f95878d855031c1
-
Jeremy Johnson authored
Enable build tests to run in multi-processing mode Refactor main loop of tosa_verify_conformance_generator Improve logging for multiprocessing jobs Change to use concurrent.futures from multiprocessing Backport v1.0 Conflicts: verif/conformance/tosa_verif_conformance_generator.py Signed-off-by:
Jeremy Johnson <jeremy.johnson@arm.com> Change-Id: I3b556a2d240acce4ff0e9faa8e815eba95489953
-
- May 16, 2025
-
-
Yuvaraj Venkatesh authored
These warnings were not flagged by Clang but triggered by the GCC compiler. GCC 9.4 throws relocation errors when linking with -pie unless all objects are compiled with -fpie. To resolve this, replaced -pie with -fpie to ensure consistent position-independent code generation during both compilation and linking. Change-Id: Iba7ce1d0e84d8040dafe65c41fd553c342507cbc Signed-off-by:
Yuvaraj Venkatesh <yuvaraj.venkatesh@arm.com>
-
- May 15, 2025
-
-
Signed-off-by:
Philip Wilkinson <philip.wilkinson@arm.com> Change-Id: I135a43eb81592ac57f9f137117d003776b256775
-
- Updated dumpTensor to use std::ostringstream instead of dynamic format strings with fprintf. - Parsed g_func_config.fp_format (e.g., "0.5") into width and precision using istringstream. - Applied std::fixed, std::setw, and std::setprecision to format the floating-point value safely and consistently. - Fixed few other openssf warnings. - Refactor dumpTensor function call. Change-Id: I8ccbab5ae4afa38803fc053ba5f423840aba50ce Signed-off-by:
Yuvaraj Venkatesh <yuvaraj.venkatesh@arm.com>
-