When upgrading to the new std lib HashMap API, the process_headers code
regressed because something that was supposed to be a pointer ended up
being a copy of a value. This resulted in the modification of a field
not being picked up.
Also switch from Sha256 to Blake3 while we're at it.
- Use an enum of all field names instead of string literals
- Create a struct type with all fields optional instead of relying on
anonymous struct literals
This should provide better type inference, compile errors, and a
(subjectively) cleaner API.
Avoids a compile error from start.zig:
/home/kivikakk/zig/build/lib/zig/std/start.zig:265:28: error:
expected type 'u8', found 'i8'
return result;
^
/home/kivikakk/zig/build/lib/zig/std/start.zig:265:28: note:
unsigned 8-bit int cannot represent all possible signed 8-bit
values
return result;
- 1MiB objects on the stack doesn't play well with wasmtime.
Reduce these to 512KiB so that the webassembly benchmarks can run.
- Pass expected results to a blackBox() function. Without this, in
release-fast mode, the compiler could detected unused return values,
and would produce results that didn't make sense for siphash.
- Add AEAD constructions to the benchmarks.
- Inline chacha20Core() makes it 4 times faster.
- benchmarkSignatures() -> benchmarkSignature() for consistency.
* Move branch-local register and stack allocation metadata to the
function-local struct. Conditional branches clone this data in order
to restore it after generating machine code for a branch.
Branch-local data is now only the instruction table mapping *ir.Inst
to MCValue.
* Implement conditional branching
- Process operand deaths
- Handle register and stack allocation metadata
* Avoid storing unreferenced or void typed instructions into
the branch-local instruction table.
* Fix integer types reporting the wrong value for hasCodeGenBits.
* Remove the codegen optimization for eliding length-0 jumps. I need to
reexamine how this works because it was causing invalid jumps to be
emitted.
