llparse
llparse copied to clipboard
Generating parsers in LLVM IR
Issue #35 Decide just drop code, interesting thing. Fields in struct same, but by some cases errors on tests. Currently do not know llparse too good for say why :thinking:...
This adds a very rough, unfinished implementation for parsing little- and big-endian integers, based on my question here: https://github.com/nodejs/llparse/issues/31 See also these 2 PRs: * https://github.com/indutny/llparse-builder/pull/1 * https://github.com/indutny/llparse-frontend/pull/1 This only...
I was looking through the code and noticed this project does not (yet) support parsing of binary protocols - is that something you want to add support for later? 🙇
I have started Verilog Preprocessor project using `llparse` here: https://github.com/drom/vpreproc It looks like a good tool for the task. Wouldn't it be nice if `llparse` could generate NAPI bindings, then...
document
Is there any document? Thanks!
Use `_mm_cmpestri` to vectorize ranged checks. This should be generated only with (to be introduced) `sse4: true` option.
[Travis are now recommending removing the __sudo__ tag](https://blog.travis-ci.com/2018-11-19-required-linux-infrastructure-migration). "_If you currently specify __sudo: false__ in your __.travis.yml__, we recommend removing that configuration_"
There's no need to save/restore spans on every `.start()`/`.end()`. The set of active spans is known for each node ahead of time, so we can just store them at `.pause()`.
Should we generate more debug metadata? Would it make any difference in visibility during profiling?
Things like: ``` "target-cpu"="penryn" "target-features"="+cx16,+fxsr,+mmx,+sahf,+sse,+sse2,+sse3,+sse4.1,+ssse3,+x87" ``` Are required to be matched between callee and caller for inlining support.