sse2neon VS daw_json_link

The sse2neon project is a quick way to get C/C++ applications compiling and running on Graviton. The sse2neon header file provides NEON implementations for x64 intrinsics so no source code changes are needed. Each function call (intrinsic) is simply replaced with NEON instructions and will just work on Graviton.

To continue our collaboration I contributed some small changes to KasmVNC on GitHub to use sse2neon for a performance critical part of the application which uses SSE intrinsics and needed to be changed to NEON intrinsics.

I think the talk is very clearly laid out as an incremental journey, and each stepping stone involves contextual decision-making. I don't think Andreas is saying "you must end up with the SSE2 implementation at the end". Using machine-specific intrinsics is another dependency decision very similar to deciding to use a given library. I would have loved the talk and probably still thought of it and posted it, even if it ended before the intrinsics (but I think he does an excellent job at that part too).

And porting SSE2 to Neon is actually pretty easy -- if you use https://github.com/DLTcollab/sse2neon, IME it's very easy to do incrementally (or avoid or postpone indefinitely, depending on your needs).

I have a private cross-platform port, I’m waiting on the resolution of his latest GitHub issue to submit my changes. sse2neon (https://github.com/DLTcollab/sse2neon) was a big help - I also wrote a very primitive sse2scalar for raspbian builds where neon is unavailable. Honestly SIMD doesn’t help much, as you’re usually memory bound under SWGL. The biggest perf win is any amount of asynchronous execution - running off the main thread is good enough and could be applied to your library externally through a command buffer without any changes to your code.

You should go to /include/simd and download sse2neon.h into the folder. Replace appearing in any source files in that directory with "sse2neon.h". You will still encounter errors; remove the lines causing problems, typically containing the phrase ZERO_MODE. ARM processors does not require it.

They might have good luck with https://github.com/beached/daw_json_link it has support for stuff like JSON lines and alike plus other ways that only use as much ram as their underlying data structures do as it parses directly to the user DS. Plus it has an iterator/range interface for things like arrays if needed.

We will add more benchmarks in the future, but for now you can see the comparison of daw_json_link with rapidjson. glaze is faster than daw_json_link, which is over twice as fast as rapidjson.

I am biased, but prefer https://github.com/beached/daw_json_link Super fast and you work with your native data structures without the overhead of DOM parsing/lookup

You can also look into https://github.com/beached/daw_json_link which claims to support nullable values in the readme.

The library I author does this. https://github.com/beached/daw_json_link . It's fast, GB/s too, and provides the mapping mechanism, iteration types, json lines support, event based parser, along with a non-owning json_value for when the mappings don't fit right or if one is querying. Pretty much everything but an owning JSON value as it's not something I've ever needed more than temporarily and brings a lot of complexity that is solved by using the actual C++ data structures one is eventually parsing into anyways.

So DAW JSON Link does have a DOM view, however it does not have a owning view. The json_value(even supports JSON Path in a limited form) type and json_raw mappings can help here. But there is no hard line between parsing view the json_value and the mappings to concrete data structures. One can mix and match.

Is this the repo? Never saw it linked in our convo, and I’d like to give it a whirl.

daw_json_link is what you're looking for : https://github.com/beached/daw_json_link

Check out https://github.com/beached/daw_json_link , it provides a non-typeerased way to parse JSON straight into user-defined data structures.