std-simd VS compiler-explorer

neither proposing nor taking a position on this possible addition)

> ... For completeness we would also like to add that a serious issue is that C still lacks vector operations.

Those are good points. The authors don't take a stance on it, but I do think that syntax for packed structs should be standardized. IMO, so should syntax for inline assembly (both as optional features). These are already common extensions; this is exactly what they should standardize. The additions of "typeof" and #embed are also good examples of this (they had been talking about adding #embed since 1995 [1]).

As for vector instructions, I'm unsure how it could be implemented in a standard way, but I'm not against it. Maybe something like this [2], but with the syntax changed for C instead of C++.

[1]: https://groups.google.com/g/comp.std.c/c/zWFEXDvyTwM

[2]: https://github.com/VcDevel/std-simd

Interesting, thanks for sharing :)

At the time we open-sourced Highway, the standardization process had already started and there were some discussions.

I'm curious why stdlib is the only path you see to default? Compare the activity level of https://github.com/VcDevel/std-simd vs https://github.com/google/highway. As to open-source usage, after years of std::experimental, I see <200 search hits [1], vs >400 for Highway [2], even after excluding several library users.

But that aside, I'm not convinced standardization is the best path for a SIMD library. We and external users extend Highway on a weekly basis as new use cases arise. What if we deferred those changes to 3-monthly meetings, or had to wait for one meeting per WD, CD, (FCD), DIS, (FDIS) stage before it's standardized? Standardization seems more useful for rarely-changing things.

1: https://sourcegraph.com/search?q=context:global+std::experim...

2: https://sourcegraph.com/search?q=context:global+HWY_NAMESPAC...

std-simd - 451 GH stars

Bad news. For SIMD there are not cross-platform intrinsics. Intel intrinsics map directly to SSE/AVX instructions and ARM intrinsics map directly to NEON instructions.

For cross-platform, your best bet is probably https://github.com/VcDevel/std-simd

There's https://eigen.tuxfamily.org/index.php?title=Main_Page But, it's tremendously complicated for anything other than large-scale linear algebra.

And, there's https://github.com/microsoft/DirectXMath But, it has obvious biases :P

But after I worked with auto-vectorization(I wrote about that in part 2), I decided to switch and try myself by adding intrinsics if I was able. You can track my progress here:https://github.com/VcDevel/std-simd/pull/35

STD-SIMD it's almost the same project I was working, but a bit advance https://github.com/VcDevel/std-simd.

If you are worried about Intel vs Arm vs whatever, use https://github.com/VcDevel/std-simd

or https://github.com/KhronosGroup/Vulkan-Hpp which help quite a bit. Or https://github.com/VcDevel/std-simd.

If you want GUIs, same, you have at least (but not only) Qt or WxWidgets.

Want to interface scripting? Pybind11, Boost.Python, WrenBind17 for Wren, Sol2 for Lua... and all things that interface to C work also if you feel brave...

I really think that when it is about getting the job done... C++ goes a long way towards the task.

This is my 20 year experience of C++, almost 13 of those years professionally. Now, back to read the paper. :)

... I just checked your link and wouldn't say that any of these languages have SIMD more than C++ has it currently -

- Java: incubation stage (how is that different from https://github.com/VcDevel/std-simd). Also Java is only getting it soonish for... amd64 and aarch64 ??

You said You won't get "extreme performance" from C++ because it is buried under the weight of decades of compatibility hacks.

Now your whole comment is about vector behavior. You haven't talked about what 'decades of compatibility hacks' are holding back performance. Whatever behavior you want from a vector is not a language limitation.

You could write your own vector and be done with it, although I'm still not sure what you mean, since once you reserve capacity a vector still doubles capacity when you overrun it. The reason this is never a performance obstacle is that if you're going to use more memory anyway, you reserve more up front. This is what any normal programmer does and they move on.

Show what you mean here:

https://godbolt.org/

I've never used ISPC. It's somewhat interesting although since it's Intel focused of course it's not actually portable.

I guess now the goal posts are shifting. First it was that "C++ as a language has performance limitations" now it's "rust has a vector that has a function I want and also I want SIMD stuff that doesn't exist. It does exist? not like that!"

Try to stay on track. You said there were "decades of compatibility hacks" holding back C++ performance then you went down a rabbit hole that has nothing to do with supporting that.

C++ Insights is available online at https://cppinsights.io/

It is also available at a touch of a button within the most excellent https://godbolt.org/

along side the button that takes your code sample to https://quick-bench.com/

Those sites and https://cppreference.com/ are what I'm using constantly while coding.

I recently discovered https://whitebox.systems/ It's a local app with a $69 one-time charge. And, it only really works with "C With Classes" style functions. But, it looks promising as another productivity boost.

[P&H RISC] https://www.google.com/books/edition/_/e8DvDwAAQBAJ

Compiler Explorer by Matt Godbolt [Godbolt] can help better understand what code a compiler generates under different circumstances.

[Godbolt] https://godbolt.org

The official CPU architecture manuals from CPU vendors are surprisingly readable and information-rich. I only read the fragments that I need or that I am interested in and move on. Here is the Intel’s one [Intel]. I use the Combined Volume Set, which is a huge PDF comprising all the ten volumes. It is easier to search in when it’s all in one file. I can open several copies on different pages to make navigation easier.

Intel also has a whole optimization reference manual [Intel] (scroll down, it’s all on the same page). The manual helps understand what exactly the CPU is doing.

[Intel] https://www.intel.com/content/www/us/en/developer/articles/t...

Personally, I believe in automated benchmarks that measure end-to-end what is actually important and notify you when a change impacts performance for the worse.

Let's compile it with https://godbolt.org/, turn on some optimisations and inspect the IR (-O2 -emit-llvm). Copying out the part that corresponds to the while loop:

  4:

resources were extra useful when building deeper intuitions about GPU performance for ML models at work and in graduate school.

- CMU's "Deep Learning Systems" Course is hosted online and has YouTube lectures online. While not generally relevant to software performance, it is especially useful for engineers interested in building strong fundamentals that will serve them well when taking ML models into production environments: https://dlsyscourse.org/

- Compiler Explorer is a tool that allows you easily input some code in and check how the assembly output maps to the source. I think this is exceptionally useful for beginner/intermediate programmers who are familiar with one compiled high-level language and have not been exposed to reading lots of assembly. It is also great for testing how different compiler flags affect assembly output. Many people used to coding in C and C++ probably know about this, but I still run into people who haven't so I share it whenever performance comes up: https://godbolt.org/

To really understand what's going on here we can look at the compiled assembly code. I'm working on a Mac and can do this using the objdump tool. Compiler Explorer is also a handy tool but doesn't seem to support Arm assembly which is what Rust will use when compiling on Apple Silicon.

Play around with it in godbolt if you're really curious: https://godbolt.org/

It sounds like you are very new to programming and C++. If you'll allow me to make a recommendation: trying to set up a C++ in VS Code is quite a difficult task for a beginner. There are a lot of trip ups -- the compiler you're using, how your Code Runner or tasks.json or launch.json are set up, whether you're using Makefiles or Cmake, etc. For beginning with C++, I would really recommend messing around with Compiler Explorer instead (https://godbolt.org/). It was originally designed to turn C++ code into assembly for debugging, but you can use it like a fast scratchpad for learning, and it auto rebuilds as you make changes so you can see errors quickly. Good luck!