903a5c1f8c
improve logic of heap_type validation when ref.null ( #4372 )
...
* Follow-up to PR #4300 : prevent potential overflow
PR #4300 introduced the rationale for validating heap_type.
This patch moves the validation before the computation of
type1 to prevent potential overflow.
2025-07-10 08:27:11 +08:00
7d05dbc988
Support extended constant expressions ( #4432 )
...
* implement extended const expr (#4318 )
* add a toggle to enable extended const on wamrc (#4412 )
2025-07-07 13:34:02 +08:00
0127eafbe5
loader: fix a potential overflow issue ( #4427 )
2025-06-30 12:57:57 +08:00
0641dd1506
Fix few shadow warnings ( #4409 )
...
- declaration of ‘memidx’ shadows a previous local
- declaration of ‘count’ shadows a previous local
2025-06-27 11:55:32 +08:00
535004dedc
Fix handling of non-nullable global_type during global import ( #4408 )
2025-06-26 06:59:57 +08:00
8e60feb181
Collective fix for typos and minor bugs ( #4369 )
2025-06-17 09:26:00 +08:00
75bf9797a2
Update type validation in load_table_import() and load_table() ( #4296 )
...
Prevent from value type.
https://webassembly.github.io/spec/core/valid/types.html#table-types
https://webassembly.github.io/gc/core/syntax/types.html#reference-types
2025-06-12 15:01:42 +08:00
928598f1ce
add heap-type check for GC when ref.null ( #4300 )
...
- According to [Link 1](https://webassembly.github.io/gc/core/valid/instructions.html#xref-syntax-instructions-syntax-instr-ref-mathsf-ref-null-mathit-ht ), we must ensure that the heap type is valid when ref.null.
- According to [Link 2](https://webassembly.github.io/gc/core/valid/types.html#heap-types ), a heap type is considered valid if it is either a concrete heap type or an abstract heap type.
However, in this function, the check for abstract heap types (absheaptype) was clearly missing, so this condition needs to be added explicitly in the if statement.
- When GC is disabled, no change is needed.
- When GC is enabled, heap types in WAMR are LEB-encoded values ([Link 3](https://webassembly.github.io/gc/core/appendix/index-types.html )). Therefore, we must use read_leb_int32 to parse the heap type correctly. And we can compute the original type1 using type1 = (uint8)((int32)0x80 + heap_type);.
2025-06-12 09:31:17 +08:00
93ef19b0ca
handle nullable heap reference types in import section ( #4302 )
2025-06-03 13:28:26 +08:00
207da7b22f
updating WASI stdio handle initialization and build options for UVWASI ( #4260 )
2025-05-30 07:42:39 +08:00
7f9e49213e
Enhance type checking for function types in loader and improve error handling ( #4294 )
...
Especially when GC is enabled, a valid item of `module->types` needs additional
checks before casting to WASMFuncType.
Also, avoid overflowing if reftype_map_count is 0.
Additionally, correctly set IN_OSS_FUZZ based on CFLAGS_ENV for sanitizer
configuration. Update ASan and UBSan messages for clarity in non-oss-fuzz
environments.
2025-05-28 20:29:09 +08:00
129dc3a30f
Add select 128 ( #4236 )
...
Add select 128
2025-05-14 12:35:56 +08:00
26aa4830e9
fix buf checking in load_table_section ( #4276 )
...
Signed-off-by: Su Yihan <yihan.su@intel.com >
2025-05-14 06:35:32 +08:00
3cce6fdaac
avoid access null pointer ( #4262 )
2025-05-12 10:21:45 +08:00
f88718d705
JIT: don't join worker threads twice ( #4252 )
...
in case of WASM_ENABLE_LAZY_JIT==0, compile_jit_functions should
have already joined these threads. joining them again here is
an undefined behavior.
2025-05-08 10:24:55 +08:00
216404d7cb
initialize WASI stdio handles to invalid for better error handling ( #4092 )
...
* initialize WASI stdio handles to invalid for better error handling
* implement os_invalid_raw_handle function for consistent invalid handle representation
2025-05-08 08:47:07 +08:00
ea417d7619
Add overflow check for preserved local offset in preserve_referenced_local ( #4211 )
2025-05-07 09:45:49 +08:00
cc1903603d
fix: Remove unused variables in SIMD_v128_const case ( #4197 )
...
Fix compiler warnings about unused variables `high` and `low` in the `SIMD_v128_const` case. These variables are only needed inside the `WASM_ENABLE_FAST_INTERP != 0` conditional block, but were incorrectly declared outside of it, causing unused variable warnings.
Signed-off-by: Huang Qi <huangqi3@xiaomi.com >
2025-04-15 09:47:18 +08:00
9aaf3599ec
prevent frame_offset underflow in wasm_loader ( #4165 )
2025-04-10 12:06:06 +08:00
793135b41c
Fix few integer overflowing ( #4161 )
...
- fix(interpreter): correct offset calculations in wasm_loader_get_const_offset function
- fix(mem-alloc): update offset calculation in gc_migrate for memory migration
- add pointer-overflow sanitizer
2025-04-10 12:04:56 +08:00
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
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
f2e3348305
wasm_loader allocates more spaces for elements ( #4099 )
...
- allocate memory for array initialization based on length
- update reference type mapping for struct initialization
2025-02-26 12:58:45 +08:00
159f5890a6
[gc] Subtyping fix ( #4075 )
2025-02-17 11:55:58 +08:00
b6dea221a6
Fix wasm loader check data segment count ( #4039 )
...
correctly report error when datacount section has non-zero data segment count while the data section is not present
2025-01-21 13:08:09 +08:00
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
831e4bbfd5
Refine getting const offsets in wasm loader of fast-interp ( #4012 )
...
- Refine const offsets in loader for fast-interp
- handle const cell num overflow
- Use const array, remove list
2025-01-20 09:39:32 +08:00
9c3807e124
Refine read leb int wasm loader of fast interpreter ( #4017 )
2025-01-14 17:43:29 +08:00
53da420c41
Enable shrunk memory by default and add related configurations ( #4008 )
...
- Enable shrunk memory by default and add related configurations
- Improve error messages for memory access alignment checks
- Add documentation for WAMR shrunk memory build option
- Update NuttX workflow to disable shrunk memory build option
2025-01-13 07:09:04 +08:00
02683d2eed
Improve stack consistency by ensuring sufficient space for dummy offsets ( #4011 )
...
One more corner case: if the `frame_offset` increases and becomes equal to
the `frame_offset_boundary` after the last assignment within the for loop.
2025-01-09 13:11:25 +08:00
a653746b7b
Check whether related table has funcref elem in opcode call_indirect ( #3999 )
...
* check whether table has funcref elem in call_indirect
* check whether table has funcref elem in call_indirect when gc is enabled
2025-01-06 13:55:43 +08:00
1958808a24
Fix table index calculations in wasm_loader and wasm_mini_loader ( #4004 )
2025-01-05 15:27:40 +08:00
099056b076
Ensure __heap_base and __data_end global indices are validated against import count ( #3996 )
2025-01-03 14:37:09 +08:00
31ff576edf
Error message improvement ( #4000 )
...
Improve error message in the scenario where the runtime was built
with ref types disabled but the module uses reference types feature.
2025-01-03 10:44:25 +08:00
1d111a38d6
Fix loader small bug ( #3928 )
2024-11-26 10:08:51 +08:00
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
bf78863c56
Wasm loader enhancement: check code size in code entry ( #3892 )
...
add wasm loader check: in code entry, the code size should match the size of vec(locals) + expr, and expr should end with opcode end
2024-11-07 13:38:42 +08:00
3ad95303d6
Fix quadratic runtime for duplicate export name detection ( #3861 )
...
Previously, the loader would check the name of a new export against all
existing exports, leading to a quadratic running time.
This change makes the loader parse the entire export section. The
exports are then sorted by name, then adjacent exports are checked for
uniqueness.
2024-10-21 11:02:30 +08:00
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
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
9ba36e284c
Implement shared heap for AOT ( #3815 )
2024-09-29 12:50:59 +08:00
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
c276aca202
Fix compile error when multi-module and tags are enabled ( #3781 )
2024-09-10 09:55:15 +08:00
f453d9d5ce
Appease GCC strict prototypes warning ( #3775 )
2024-09-10 09:42:23 +08:00
cbc2078898
AOT call stack optimizations ( #3773 )
...
- Implement TINY / STANDARD frame modes - tiny mode is only able to keep track on the IP
and func idx, STANDARD mode provides more capabilities (parameters, stack pointer etc.).
- Implement FRAME_PER_FUNCTION / FRAME_PER_CALL modes - frame per function adds
code at the beginning and at the end of each function for allocating / deallocating stack frame,
whereas in per-call mode the frame is allocated before each call. The exception is call to
the imported function, where frame-per-function mode also allocates the stack before the
`call` instruction (as it can't instrument the imported function).
At the moment TINY + FRAME_PER_FUNCTION is automatically enabled in case GC and perf
profiling are disabled and `values` call stack feature is not requested. In all the other cases
STANDARD + FRAME_PER_CALL is used.
STANDARD + FRAME_PER_FUNCTION and TINY + FRAME_PER_CALL are currently not
implemented but possible, and might be enabled in the future.
ps. https://github.com/bytecodealliance/wasm-micro-runtime/issues/3758
2024-09-10 09:05:23 +08:00
6f97822c18
Add wamrc parameter to configure stack frame features ( #3763 )
...
Those parameters can be used to reduce the size of the AOT code.
There's going to be more changes related to AOT code size reduction,
this is just the initial step.
p.s. https://github.com/bytecodealliance/wasm-micro-runtime/issues/3758
2024-09-05 21:44:06 +08:00
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
97c95a2e2f
Fix table idx resolving in op call_indirect/return_call_indirect ( #3726 )
...
The table index in the call_indirect/return_call_indirect opcode should be
one byte 0x00 when ref-types/GC isn't enabled, and should be treated as
leb u32 when ref-types/GC is enabled.
And make aot compiler bail out if ref-types/GC is disabled by command line
argument while ref-types instructions are used.
2024-08-19 10:57:36 +08:00
59f761b58d
aot compiler: Track non-0x00 tableindex as ref types use ( #3695 )
...
Any use of a table index that isn't exactly a null byte (`0x00`) means that
the module makes use of the reference types proposal. This is important
to track because `aot_compiler.c` will blindly assume that all table indices
are a single byte long otherwise.
This fixes a crash in WAMR for modules that contain multi-byte encodings
of table indices in `call_indirect` but make no other use of reference types
features.
2024-08-16 12:37:10 +08:00
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
