Nathan Froyd
Nathan Froyd
STR: ```rust let name = "test.o"; let file = File::create(Path::new(name)).unwrap(); let mut obj = ArtifactBuilder::new(Triple { architecture: Architecture::X86_64, vendor: Vendor::Unknown, operating_system: OperatingSystem::Unknown, environment: Environment::Unknown, binary_format: BinaryFormat::Elf, }) .name(name.to_owned()) .finish(); obj.declare("zeroinit",...
I ran into this error today: ``` Traceback (most recent call last): File "./bloat.py", line 408, in dump_nm(nmfile, strip_prefix=opts.strip_prefix, cppfilt=opts.cppfilt) File "./bloat.py", line 305, in dump_nm json.dumps(jsonify_tree(dirs, '[everything]'), indent=2)) File...
The patch is relatively easy; what's not clear to me is how something like this really wants to be exported. Should `sysdeps` export a `MappedRegion` that would be backed by...
I was trying to figure out what might be involved in adding Windows support to `lucet-runtime-internals` and came across these two pieces of code: https://github.com/bytecodealliance/lucet/blob/0b51fe7b68a7c75b9d74197096ededeb44703238/lucet-runtime/lucet-runtime-internals/src/region/mmap.rs#L167-L179 https://github.com/bytecodealliance/lucet/blob/0b51fe7b68a7c75b9d74197096ededeb44703238/lucet-runtime/lucet-runtime-internals/src/region/mmap.rs#L199-L216 The two pieces are...
As part of the ongoing work to make `lucetc` work for Windows, cranelift has had some APIs added to enable using `cranelift-module` for some of `lucetc`'s work, rather than `cranelift-faerie`....
This doesn't seem to be supported based on `--help` and a quick browse through the code. It's entirely possible I'm missing something, though! Firefox, at least, would really like this...
### Motivation Pre-work for an eventual storing of arguments inline with `MethodDef` rather than in a separate vector. ### Test plan See included automated tests.
### Motivation People keep asking for this feature at Stripe, and we keep having them to enable experimental features to do it. We've had a small number of people testing...
### Motivation Same reason as #6129, but with a slightly large effect radius. (Also may be worth exploring whether we should hold memory buffers directly, rather than indirecting through `vector`...
### Motivation Fewer dead writes to memory during deserialization, which should theoretically be faster. (The same principle can be applied to the serialization buffers, but that is a more invasive...