439 Commits

Author	SHA1	Message	Date
Marcin Kolny	efa8019bdb	Merge dev/simd for fast-interp (#4131) ... * Implement the first few SIMD opcodes for fast interpreter (v128.const, v128.any_true) (#3818) Tested on the following code: ``` (module (import "wasi_snapshot_preview1" "proc_exit" (func $proc_exit (param i32))) (memory (export "memory") 1) ;; WASI entry point (func $main (export "_start") v128.const i8x16 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 v128.any_true if unreachable end v128.const i8x16 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 v128.any_true i32.const 0 i32.eq if unreachable end i32.const 0 call $proc_exit ) ) ``` * implement POP_V128() This is to simplify the simd implementation for fast interpreter * Add all SIMD operations into wasm_interp_fast switch * Add V128 comparison operations Tested using ``` (module (import "wasi_snapshot_preview1" "proc_exit" (func $proc_exit (param i32))) (memory (export "memory") 1) (func $assert_true (param v128) local.get 0 v128.any_true i32.eqz if unreachable end ) (func $main (export "_start") ;; Test v128.not v128.const i8x16 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 v128.not v128.const i8x16 255 255 255 255 255 255 255 255 255 255 255 255 255 255 255 255 i8x16.eq call $assert_true ;; Test v128.and v128.const i8x16 255 255 255 255 0 0 0 0 255 255 255 255 0 0 0 0 v128.const i8x16 255 255 0 0 255 255 0 0 255 255 0 0 255 255 0 0 v128.and v128.const i8x16 255 255 0 0 0 0 0 0 255 255 0 0 0 0 0 0 i8x16.eq call $assert_true ;; Test v128.andnot v128.const i8x16 255 255 255 255 0 0 0 0 255 255 255 255 0 0 0 0 v128.const i8x16 255 255 0 0 255 255 0 0 255 255 0 0 255 255 0 0 v128.andnot v128.const i8x16 0 0 255 255 0 0 0 0 0 0 255 255 0 0 0 0 i8x16.eq call $assert_true ;; Test v128.or v128.const i8x16 255 255 0 0 0 0 255 255 255 255 0 0 0 0 255 0 v128.const i8x16 0 0 255 255 255 255 0 0 0 0 255 255 255 255 0 0 v128.or v128.const i8x16 255 255 255 255 255 255 255 255 255 255 255 255 255 255 255 0 i8x16.eq call $assert_true ;; Test v128.xor v128.const i8x16 255 255 0 0 255 255 0 0 255 255 0 0 255 255 0 0 v128.const i8x16 255 255 255 255 0 0 0 0 255 255 255 255 0 0 0 0 v128.xor v128.const i8x16 0 0 255 255 255 255 0 0 0 0 255 255 255 255 0 0 i8x16.eq call $assert_true i32.const 0 call $proc_exit ) ) ``` * Add first NEON SIMD opcode implementations to fast interpreter (#3859) Add some implementations of SIMD opcodes using NEON instructions. Tested using: ```wast (module (import "wasi_snapshot_preview1" "proc_exit" (func $proc_exit (param i32))) (memory (export "memory") 1) (func $assert_true (param v128) local.get 0 v128.any_true i32.eqz if unreachable end ) (func $main (export "_start") i32.const 0 i32.const 32 memory.grow drop i32.const 0 v128.const i8x16 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 v128.store i32.const 0 v128.load v128.const i8x16 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 i8x16.eq call $assert_true i32.const 16 v128.const i8x16 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 v128.store i32.const 16 v128.load v128.const i8x16 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 i8x16.eq call $assert_true i32.const 0 v128.load v128.const i8x16 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 i8x16.eq call $assert_true drop i32.const 0 i32.const 1 memory.grow drop i32.const 0 i64.const 0x7F80FF017E02FE80 i64.store i32.const 0 v128.load8x8_s v128.const i16x8 127 -128 -1 1 126 2 -2 -128 i16x8.eq call $assert_true i32.const 0 i64.const 0x80FE027E01FF807F i64.store i32.const 0 v128.load8x8_u v128.const i16x8 128 254 2 126 1 255 128 127 i16x8.eq call $assert_true i32.const 0 i64.const 0x8000FFFE7FFF0001 i64.store i32.const 0 v128.load16x4_s v128.const i32x4 -32768 -2 32767 1 i32x4.eq call $assert_true i32.const 0 i64.const 0x8000FFFE7FFF0001 i64.store i32.const 0 v128.load16x4_u v128.const i32x4 32768 65534 32767 1 i32x4.eq call $assert_true i32.const 0 i64.const 0x8000000000000001 i64.store i32.const 0 v128.load32x2_s v128.const i64x2 -2147483648 1 i64x2.eq call $assert_true i32.const 0 i64.const 0x8000000000000001 i64.store i32.const 0 v128.load32x2_u v128.const i64x2 2147483648 1 i64x2.eq call $assert_true call $proc_exit ) ) ``` * Emit imm for lane extract and replace (#3906) * Fix replacement value not being correct (#3919) * Implement load lanes opcodes for wasm (#3942) * Implement final SIMD opcodes: store lane (#4001) * Fix load/store (#4054) * Correctly use unsigned functions (#4055) * implement local and function calls for v128 in the fast interpreter * Fix splat opcodes, add V128 handling in preserve_referenced_local and reserve_block_ret * Fix incorrect memory overflow values + SIMD ifdefs * Fix load/load_splat macros * correct endif wasm loader * Update core/iwasm/interpreter/wasm_opcode.h * Fix spec tests when WASM_CPU_SUPPORTS_UNALIGNED_ADDR_ACCESS is 0 * Resolve merge conflicts arising from main -> dev/simd_for_interp and implement fast interpreter const offset loader support for V128 * Enable SIMDe tests on CI * Document WAMR_BUILD_LIB_SIMDE --------- Co-authored-by: James Marsh <mrshnja@amazon.co.uk> Co-authored-by: jammar1 <108334558+jammar1@users.noreply.github.com> Co-authored-by: Maks Litskevich <makslit@amazon.com> Co-authored-by: Marcin Kolny <marcin.kolny@gmail.com> Co-authored-by: Wenyong Huang <wenyong.huang@intel.com>	2025-03-20 14:23:20 +08:00
liang.he	c6f934434b	Merge pull request #4124 from swankjesse/WAMR_BUILD_GC_HEAP_SIZE_DEFAULT ... Expose WAMR_BUILD_GC_HEAP_SIZE_DEFAULT as a CMake option	2025-03-12 14:55:27 +08:00
liang.he	766f378590	Merge pull request #4033 from g0djan/godjan/iterate_callstack ... Copy callstack API	2025-03-11 10:31:56 +08:00
Huang Qi	412631ac13	fix: correct typos and improve comments across multiple files by codespell (#4116) ... Signed-off-by: Huang Qi <huangqi3@xiaomi.com>	2025-03-07 08:21:54 +08:00
Jesse Wilson	d609acf1ee	Move the default heap size initialization	2025-03-04 18:43:30 -05:00
Jesse Wilson	de82d1946f	Address code review feedback	2025-03-03 20:45:59 -05:00
Georgii Rylov	ffcc1579ea	clang format	2025-03-03 13:41:35 +00:00
Georgii Rylov	bc00b3e492	address comments	2025-03-03 13:36:05 +00:00
Jesse Wilson	beb34c3675	Expose WAMR_BUILD_GC_HEAP_SIZE_DEFAULT as a CMake option ... This is wired through to the GC_HEAP_SIZE_DEFAULT constant. Also honor this value when configuring the engine with the wasm_c_api.	2025-03-02 23:32:04 -05:00
Georgii Rylov	fc3077b74d	address comments	2025-02-27 14:32:17 +00:00
Georgii Rylov	a5d8c0b477	define if not defined	2025-02-26 11:11:17 +00:00
Georgii Rylov	857e6b73c8	formatting	2025-02-26 10:57:50 +00:00
Georgii Rylov	188eb1c941	remove unnecessary includes	2025-02-25 17:42:06 +00:00
TL	f1ffbb5b37	cr suggestions	2025-02-24 21:20:07 +00:00
TL	dfcadc6202	prevent data overflow on 32 bit platform for memory.grow	2025-02-24 21:20:07 +00:00
Georgii Rylov :slightly_smiling_face	cc3f0a096b	Cleaning up	2025-02-24 17:33:14 +00:00
Georgii Rylov :slightly_smiling_face	32338bb7d6	Copy read only API behind a flag instead of using user defined callback	2025-02-24 17:22:05 +00:00
liang.he	964037c9b5	feat: add support for EXTERNREF value type and enable AOT validator in fuzz tests (#4083)	2025-02-17 13:34:18 +08:00
Jérôme Vouillon	159f5890a6	[gc] Subtyping fix (#4075)	2025-02-17 11:55:58 +08:00
TianlongLiang	e6a47d5cee	In wasm32, fix potential conversion overflow when enlarging 65536 pages (#4064) ... fix enlarge 65536 pages conversion overflow in wasm32	2025-02-06 14:48:53 +08:00
liang.he	376385c608	Update memory allocation functions to use allocator user data (#4043)	2025-02-06 13:15:00 +08:00
Georgii Rylov	b9039f986e	Merge branch 'main' into godjan/iterate_callstack	2025-01-28 13:35:47 +00:00
Georgii Rylov	813831de0e	keep devs notes out of public API	2025-01-27 11:39:53 +00:00
Georgii Rylov	1f4d3dd4d4	clang-format	2025-01-27 11:31:02 +00:00
liang.he	e3ddbd58f7	Synchronize the GC spec tests to the commit from December 9. 2024. (#4022) ... - Synchronize the GC spec tests to the commit from December 9. 2024. - Revise the error messages to be consistent with the spec test cases. - bypass gc spec test on the nuttx platform as a workaround	2025-01-20 09:41:47 +08:00
Georgii Rylov	68e4534822	Iterate callstack API	2025-01-17 16:16:45 +00:00
liang.he	7f3e0df21c	Handle a new scenario where an item is both exported and imported. (#3984)	2024-12-29 15:52:12 +08:00
kk	c32a6ceae1	Refactor SConscript and add file checks in iwasm.c (#3945)	2024-12-06 14:54:37 +08:00
James Ring	838dd81e68	don't return an uninitialized trap if argv_to_results fails (#3935) ... Currently, if argv_to_results fails (e.g. because an unsupported type is encountered), an non-null trap with an uninitialized message is returned.	2024-11-28 11:50:44 +08:00
James Ring	e09613c722	support WASM_FUNCREF return type in argv_to_results (#3936)	2024-11-28 11:50:16 +08:00
Dylan Johnston	9d8150efae	Fix WASI Path Mapping Processing (#3923) ... Filesystem paths can be mapped from the host path to a guest path using the format `<guest-path>::<host-path>`. Previously `strtok` was used to find the `::` delimiter. Unfortunately `strtok` processes each delimiter character individually. This meant that the code was ~equivalent to `strtok(mapping_copy, ":")` which breaks with Windows-style paths (E.g. `C:\my_path\`). To fix this `strstr` is used to search for the exact delimiter.	2024-11-24 11:31:55 +08:00
TianlongLiang	0e4dffc479	Fix a leak in wasm_loader_emit_br_info (#3900) ... Reference Info: 377955855 wamr:wasm_mutator_fuzz_loader: Direct-leak in wasm_loader_emit_br_info https://issues.oss-fuzz.com/issues/377955855	2024-11-13 15:16:13 +08:00
liang.he	c7b2683f17	Fix out of bounds issue in is_native_addr_in_shared_heap function (#3886) ... When checking for integer overflow, you may often write tests like p + i < p. This works fine if p and i are unsigned integers, since any overflow in the addition will cause the value to simply "wrap around." However, using this pattern when p is a pointer is problematic because pointer overflow has undefined behavior according to the C and C++ standards. If the addition overflows and has an undefined result, the comparison will likewise be undefined; it may produce an unintended result, or may be deleted entirely by an optimizing compiler.	2024-10-31 12:44:55 +08:00
Wenyong Huang	a3960c834d	Refine looking up aot function with index (#3882) ... * Refine looking up aot function with index * refine the code	2024-10-29 10:58:11 +08:00
Wenyong Huang	48eaa2286a	Refine wasm/aot function instance lookup (#3865) ... Sort the module instance's export functions with the function name, and use binary search to lookup the wasm/aot function.	2024-10-18 15:15:33 +08:00
Wenyong Huang	74d2427fdc	Allow to set native stack boundary to exec_env (#3862) ... Add runtime API wasm_runtime_set_native_stack_boundary. p.s. https://github.com/bytecodealliance/wasm-micro-runtime/issues/3816	2024-10-18 12:49:57 +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	b038f2721b	Merge pull request #3823 from bytecodealliance/dev/shared_heap ... Implement the shared heap feature for interpreter, aot and llvm jit. Add below runtime APIs: ```C wasm_shared_heap_t wasm_runtime_create_shared_heap(SharedHeapInitArgs *init_args); bool wasm_runtime_attach_shared_heap(wasm_module_inst_t module_inst, wasm_shared_heap_t shared_heap); void wasm_runtime_detach_shared_heap(wasm_module_inst_t module_inst); uint64_t wasm_runtime_shared_heap_malloc(wasm_module_inst_t module_inst, uint64_t size, void **p_native_addr); void wasm_runtime_shared_heap_free(wasm_module_inst_t module_inst, uint64_t ptr); ``` And allow wasm app to call API shared_heap_malloc and shared_heap_free: ```C void *shared_heap_malloc(uint32_t size); void shared_heap_free(void *ptr); ```	2024-10-15 14:26:22 +08:00
TianlongLiang	36d438051e	Support table64 extension in classic-interp and AOT running modes (#3811) ... Add table64 extension(in Memory64 proposal) support in classic-interp and AOT running modes, currently still use uint32 to represent table's initial and maximum size to keep AOT ABI unchanged.	2024-10-11 21:14:56 +08:00
WenLY1	19160f0e1d	Fix issues of destroy_shared_heaps (#3847) ... Set shared_heap_list to NULL with lock, and destroy shared_heap_list_lock. Signed-off-by: wenlingyun1 <wenlingyun1@xiaomi.com>	2024-10-10 10:56:36 +08:00
Wenyong Huang	9ba36e284c	Implement shared heap for AOT (#3815)	2024-09-29 12:50:59 +08:00
Wenyong Huang	e87f7a920d	Fix Windows compile error when uvwasi is enabled (#3810) ... No need to compile win_file.c when uvwasi is enabled.	2024-09-25 16:43:00 +08:00
Wenyong Huang	1fd422ae6e	Merge branch main into dev/shared_heap	2024-09-20 14:34:00 +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
Anders Bakken	21330990a8	Add no_resolve to LoadArgs and wasm_runtime_resolve_symbols (#3790) ... Add no_resolve to LoadArgs and wasm_runtime_resolve_symbols so one can delay resolving of symbols. This is useful for inspecting the module between loading and instantiating.	2024-09-20 08:54:09 +08:00
Wenyong Huang	5e20cf383e	Refactor shared heap feature for interpreter mode (#3794) ... To add test cases and samples.	2024-09-18 14:53:41 +08:00
WenLY1	92852f3719	Implement a first version of shared heap feature (#3789) ... ps. https://github.com/bytecodealliance/wasm-micro-runtime/issues/3546	2024-09-14 10:51:42 +08:00
Benbuck Nason	926f662231	Add memory instance support apis (#3786) ... Now that WAMR supports multiple memory instances, this PR adds some APIs to access them in a standard way. This involves moving some existing utility functions out from the `WASM_ENABLE_MULTI_MODULE` blocks they were nested in, but multi-memory and multi-module seem independent as far as I can tell so I assume that's okay. APIs added: ```C wasm_runtime_lookup_memory wasm_runtime_get_default_memory wasm_runtime_get_memory wasm_memory_get_cur_page_count wasm_memory_get_max_page_count wasm_memory_get_bytes_per_page wasm_memory_get_shared wasm_memory_get_base_address wasm_memory_enlarge ```	2024-09-14 10:31:13 +08:00
YAMAMOTO Takashi	f453d9d5ce	Appease GCC strict prototypes warning (#3775)	2024-09-10 09:42:23 +08:00
Wenyong Huang	1329e1d3e1	Add support for multi-memory proposal in classic interpreter (#3742) ... Implement multi-memory for classic-interpreter. Support core spec (and bulk memory) opcodes now, and will support atomic opcodes, and add multi-memory export APIs in the future. PS: Multi-memory spec test patched a lot for linking test to adapt for multi-module implementation.	2024-08-21 12:22:23 +08:00