- wasi_nn.h: make this compatible with wasi_ephemeral_nn
cf. https://github.com/bytecodealliance/wasm-micro-runtime/issues/4323
- fix WASM_ENABLE_WASI_EPHEMERAL_NN build
this structure is used by host logic as well.
ideally definitions for wasm and host should be separated.
until it happens, check __wasm__ to avoid the breakage.
* this enum is (@witx tag u8) in witx
* it seems that some wasm modules actually use non-zero padding
and cause errors
* it's a bad practice to use C enum for ABI description anyway
sync with a more appropriate version of the definitions.
as we use the "wasi_ephemeral_nn", which is p1-based, it seems
more appropriate to use definitions from witx, not wit.
it's a bit unfortunate p2-based wasi-nn made gratuitous changes
like this from p1.
note: this is an ABI change.
it seems meaningless and quite confusing to access a table with
two aliases ("lookup" and "backends") within a function.
no functional changes are intended.
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.
Previously, if the user sets their own stack boundary, we still compute
the thread stack boundary (which is expensive), then immediately discard
the result. This change makes the expensive call only if we need it for
sure.
* initialize WASI stdio handles to invalid for better error handling
* implement os_invalid_raw_handle function for consistent invalid handle representation
LLVM 16 and later expands cttz intrinsic to a table lookup,
which involves some relocations. (unless ZBB is available,
in which case the native instructions are preferred over
the table-based lowering.)
cf. https://reviews.llvm.org/D128911
the corresponding LLVM intrinsics' return types are same as
their first argument. eg. i64 for llvm.cttz.i64.
cf. https://llvm.org/docs/LangRef.html#llvm-cttz-intrinsic
this commit changes the return types of our versions of the
intrinsics to match llvm versions as our aot compiler,
specifically __call_llvm_intrinsic, assumes.
strictly speaking, this is a potential AOT ABI change.
however, I suppose it isn't a problem for many of 64-bit ABIs
out there, where (lower half of) a 64-bit register is used to
return a 32-bit value anyway. (for such ABIs, this commit
would fix the upper 32-bit value of the register.)
LLVM 19 and later started to use srodata ("small read only data")
sections for RISCV. cf. https://github.com/llvm/llvm-project/pull/82214
this commit makes our aot emitter/loader deal with those sections.
an alternative would be to disable small data sections completely by
setting the "SmallDataLimit" module attribute to zero. however, i feel
this commit is more straightforward and consisitent as we are already
dealing with sdata sections.
while ideally a user should not need to care this kind of
optimization details, in reality i guess it's sometimes useful.
both of clang and GCC expose a similar option. (-fno-jump-tables)
for x86-64, llvm 17 and later sometimes uses "l" prefix
for data sections.
cf. 43249378da
because our aot file emitter/loader doesn't support such
sections, it ends up with load-time errors solving symbols like ".lrodata".
this commit fixes it by avoid placing data in the large data sections.
references:
https://groups.google.com/g/x86-64-abi/c/jnQdJeabxiU1feb00a28c
LLVM 18 and later, instcombine perfoms only one iteration.
it performs extra "verify fixpoint" operation when instcombine
is specified in certain ways, including how we do so here.
a problem is that the verification raises a fatal error when it
finds we didn't reach a fixpoint:
LLVM ERROR: Instruction Combining did not reach a fixpoint
after 1 iterations
while it should be rare, it's quite normal not to reach a fixpoint.
this commit fixes the issue by simply disabing the verification.
cf. 41895843b5
There are two reasons for this optimization:
- The value of dlen can equal 0x1_0000_0000, even in wasm32 mode, because it is derived from (4G-0). This results in a truncation when it is passed to b_memmove_s(). Consequently, s1max becomes 0 and n is greater than s1max. To correct this, a longer type is required.
- The dlen is only used to check if there is enough space in b_memmove_s(). However, from a different angle, after confirming that both src+len and dst+len are within the memory range, we can be assured and there is no need for this explicit check.
* Fix errors on the "i386-windows-msvc" platform
* Refactor symbol name handling for AOT COFF32 binary format
* Fix preprocessor directive placement for Windows compatibility in aot_reloc_x86_32.c
---------
Co-authored-by: liang.he@intel.com <liang.he@intel.com>
