8 Commits

Author	SHA1	Message	Date
liang.he	8acce563b4	Merge branch main into dev/instantiate_linking (#4100) ... - fix: when load aot init expr,no type_idx set. (#4094) - Cmake improvements (#4076) - fix(aot_emit_aot_file): prevent buffer emission for zero byte_count (#4095) - fix(unit-test): libc_builtin_test issues (#4073) - feat: add support for EXTERNREF value type and enable AOT validator in fuzz tests (#4083) - [gc] Subtyping fix (#4075) - Add a conditional check for the macro __STDC_VERSION__ (#4080) - Unit test:type matching issue and code redundancy (#4079) - build(deps): Bump github/codeql-action from 3.28.8 to 3.28.9 (#4074) - fix(aot): ensure value_cmp does not exceed br_count in branch table compilation (#4065) - In wasm32, fix potential conversion overflow when enlarging 65536 pages (#4064) - [fuzzing] execute every exported function (#3959) - Update memory allocation functions to use allocator user data (#4043) - Show wasm proposals status during compilation and execution (#3989) - add a validator for aot module (#3995) - Use wasm32-wasip1 instead of wasm32-wasi target for rust code (#4057) - Update Rust target from 'wasm32-wasi' to 'wasm32-wasip1' in CI (#4050) - Fix wasm loader check data segment count (#4039) - Synchronize the GC spec tests to the commit from December 9. 2024. (#4022) - Refine getting const offsets in wasm loader of fast-interp (#4012) - fixes for compiling on windows (#4026) - .github: Add shared lib builds (#3975) - Refine read leb int wasm loader of fast interpreter (#4017) - build(deps): Bump github/codeql-action from 3.28.0 to 3.28.1 (#4020) - build(deps): Bump actions/upload-artifact from 4.5.0 to 4.6.0 (#4021) - Enable shrunk memory by default and add related configurations (#4008) - Add documentation regarding security issues and the status of Wasm proposals (#3972) - Improve stack consistency by ensuring sufficient space for dummy offsets (#4011) - Check whether related table has funcref elem in opcode call_indirect (#3999) - [fuzzing] Use software bound-check during fuzzing (#4003) - Add an example of how to embed WAMR in Zephyr user mode (#3998) - Fix table index calculations in wasm_loader and wasm_mini_loader (#4004) - Ensure __heap_base and __data_end global indices are validated against import count (#3996) - Error message improvement (#4000) - Handle a new scenario where an item is both exported and imported. (#3984) - Optimize memory initialization handling in AOT loader (#3983) - build(deps): Bump actions/upload-artifact from 4.4.3 to 4.5.0 (#3981) - build(deps): Bump github/codeql-action from 3.27.9 to 3.28.0 (#3982) - Add Tianlong into code owners (#3970) - Set thread information earlier in exec_env creation (#3967) - top-level cmake: link llvm libraries to our shared library (#3973) - add reference type support by default for darwin to support WASI-SDK-25 (#3978) - CMakeLists.txt: Do not require C++ (#3956) - [fuzzing] Enable instantiation (#3958) - use a random secret key (#3971) - top-level cmakefile: fix macOS build (#3968) - Only access Zephyr thread stats info when it's available (#3962) - build(deps): Bump github/codeql-action from 3.27.6 to 3.27.9 (#3960) - wasm_export.h: Use "default" visibility for gcc and clang (#3957) - set alignment 4 when loading multi return value (#3955) - Fix aot table instantiate (#3946) - Consume the placeholders that were put when emitting table info (#3940) - Refactor SConscript and add file checks in iwasm.c (#3945) - Improvements for platform thread APIs on Windows and Zephyr (#3941) - Fix incorrect assignment in win_file.c (#3939) - don't return an uninitialized trap if argv_to_results fails (#3935) - support WASM_FUNCREF return type in argv_to_results (#3936) - add thread cpu time for zephyr (#3937) - build(deps): bump github/codeql-action from 3.27.4 to 3.27.5 (#3931) - Update README.md to clarify Windows toolchain support and ESP-IDF reference (#3917) - Enable ref types by default (#3894) - Fix loader small bug (#3928) - add testcases for shared heap and fix POP_MEM_OFFSET of memory64 (#3916) - Use plain assignment rather than bh_memcpy_s (#3924) - Fix WASI Path Mapping Processing (#3923) - Drop declarative elements on module instantiation (#3922) - Check possible integer overflow in aot memory boundary check (#3920) - Fix CI wamr-ide error (#3913) - Support external toolchain on Windows for aot compiler (#3911) - build(deps): bump github/codeql-action from 3.27.1 to 3.27.4 (#3912) - Correct the table index calculation in aot_instantiation (#3903) - Fix a leak in wasm_loader_emit_br_info (#3900) - GlobalValueSet was moved to IRPartitionLayer recently, but we have a local definition anyway (#3899) - build(deps): bump github/codeql-action from 3.27.0 to 3.27.1 (#3902) - Fix linked global initialization in multimodule (#3905) - Wasm loader enhancement: check code size in code entry (#3892) - Refactor AOT loader to support compatible versions (#3891) - Fix out of bounds issue in is_native_addr_in_shared_heap function (#3886) - Fix mmap flags for AOT loader on non-Linux SGX platforms (#3890) - Bump AOT_CURRENT_VERSION for WAMR 2.x (gc, memory64) (#3880) - Refine looking up aot function with index (#3882) - build(deps): bump github/codeql-action from 3.26.13 to 3.27.0 (#3888) - fix(ios): Remove `float-abi` flag (#3889) - Fix out of bounds issues after memory.grow on non-aot non-threads builds (#3872) - Exclude fuzz test python and npm packages in scoreboard scan (#3871) Author: Chris Woods <6069113+woodsmc@users.noreply.github.com> Author: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> Author: Dylan Johnston <18252447+dpjohnst@users.noreply.github.com> Author: edoardo <48774736+xdoardo@users.noreply.github.com> Author: eric <chenliuyang_1989@163.com> Author: Fadumina Barre <bafadumi@amazon.com> Author: Huang Qi <huangqi3@xiaomi.com> Author: James Ring <sjr@jdns.org> Author: Jérôme Vouillon <jerome.vouillon@gmail.com> Author: kk <zhkag@foxmail.com> Author: liang.he@intel.com <liang.he@intel.com> Author: liang.he <liang.he@intel.com> Author: Maks Litskevich <makslit@amazon.com> Author: Marcin Kolny <mkolny@amazon.com> Author: peter-tatrai <peter.tatrai.ext@siemens.com> Author: TianlongLiang <111852609+TianlongLiang@users.noreply.github.com> Author: Tomáš Malý <malytomas@users.noreply.github.com> Author: Viacheslav Palchikov <palchikov@gmail.com> Author: WenLY1 <130950131+WenLY1@users.noreply.github.com> Author: Wenyong Huang <wenyong.huang@intel.com> Author: Xavier Del Campo <90845888+midokura-xavi92@users.noreply.github.com> Author: YAMAMOTO Takashi <yamamoto@midokura.com> Author: yangkun27 <yangkun27@xiaomi.com>	2025-03-03 14:07:14 +08:00
Wenyong Huang	327374cfee	Fix some compile warnings and typos (#3854) ... - Clear some compile warnings - Fix some typos - Fix llvm LICENSE link error - Remove unused aot file and binarydump bin - Add checks when loading AOT exports	2024-10-15 16:04:58 +08:00
Wenyong Huang	9ba36e284c	Implement shared heap for AOT (#3815)	2024-09-29 12:50:59 +08:00
WenLY1	4dacef2d60	shared heap: Fix some issues and add basic unit test case (#3801) ... Fix some issues and add basic unit test case for shared heap feature. Signed-off-by: wenlingyun1 <wenlingyun1@xiaomi.com>	2024-09-20 14:24:38 +08:00
Marcin Kolny	37d7439ef9	[refactoring] Extract read leb to a separate file, share the code between loader and mini loader (#3701) ... There's probably a number of other places where the bh_leb_read could be used (e.g. aot loader) but I'm making the change as small as possible. Further refactoring can be done later.	2024-08-14 08:46:24 +08:00
YAMAMOTO Takashi	4c2af25aff	aot compiler: Use larger alignment for load/store when possible (#3552) ... Consider the following wasm module: ```wast (module (func (export "foo") i32.const 0x104 i32.const 0x12345678 i32.store ) (memory 1 1) ) ``` While the address (0x104) is perfectly aligned for i32.store, as our aot compiler uses 1-byte alignment for load/store LLVM IR instructions, it often produces inefficient machine code, especially for alignment-sensitive targets. For example, the above "foo" function is compiled into the following xtensa machine code. ``` 0000002c <aot_func_internal#0>: 2c: 004136 entry a1, 32 2f: 07a182 movi a8, 0x107 32: 828a add.n a8, a2, a8 34: 291c movi.n a9, 18 36: 004892 s8i a9, a8, 0 39: 06a182 movi a8, 0x106 3c: 828a add.n a8, a2, a8 3e: ffff91 l32r a9, 3c <aot_func_internal#0+0x10> (ff91828a <aot_func_internal#0+0xff91825e>) 3e: R_XTENSA_SLOT0_OP .literal+0x8 41: 004892 s8i a9, a8, 0 44: 05a182 movi a8, 0x105 47: 828a add.n a8, a2, a8 49: ffff91 l32r a9, 48 <aot_func_internal#0+0x1c> (ffff9182 <aot_func_internal#0+0xffff9156>) 49: R_XTENSA_SLOT0_OP .literal+0xc 4c: 41a890 srli a10, a9, 8 4f: 0048a2 s8i a10, a8, 0 52: 04a182 movi a8, 0x104 55: 828a add.n a8, a2, a8 57: 004892 s8i a9, a8, 0 5a: f01d retw.n ``` Note that the each four bytes are stored separately using one-byte-store instruction, s8i. This commit tries to use larger alignments for load/store LLVM IR instructions when possible. with this commit, the above example is compiled into the following machine code, which seems more reasonable. ``` 0000002c <aot_func_internal#0>: 2c: 004136 entry a1, 32 2f: ffff81 l32r a8, 2c <aot_func_internal#0> (81004136 <aot_func_internal#0+0x8100410a>) 2f: R_XTENSA_SLOT0_OP .literal+0x8 32: 416282 s32i a8, a2, 0x104 35: f01d retw.n ``` Note: this doesn't work well for --xip because aot_load_const_from_table() hides the constness of the value. Maybe we need our own mechanism to propagate the constness and the value.	2024-06-22 10:32:52 +08:00
Zhang, Yi	380cd7b0e7	Add unit test suites (#3490)	2024-06-04 11:24:27 +08:00
Marcin Kolny	8c2dc1d011	Add unit tests for the tid allocator (#2519) ... Add simple infrastructure to add more unit tests in the future. At the moment tests are only executed on Linux, but can be extended to other platforms if needed. Use https://github.com/google/googletest/ as a framework.	2023-09-04 13:21:10 +08:00