VcDevel

std-simd

C++2020–2023

VcDevel / std-simd

std-simd

std::experimental::simd for GCC [ISO/IEC TS 19570:2018]

Systems & Infrastructure

Explore on GitHub Read Hacker News discussions

5Distinct people

7Mentions

1Explicit recommenders

656GitHub stars

REPOSITORY AT A GLANCE

What to know before exploring

What it is

std::experimental::simd for GCC [ISO/IEC TS 19570:2018]

GitHub

Where it is used

No documented product use has been verified yet.

What people are saying

If you need to work on large matrices, Eigen is highly recommended. If you need to do a stream of custom processing in small steps, Eigen is not a good fit. Eigen is a super…

corysama used a recommendation phrase. One indexed comment matches. This is a phrase match, not a verdict.

No classified warning is attached to this page. Read the discussions before treating silence as approval.

Also in these discussions

google/highway · linked in 2 of the same discussions

Mentioned in the same Hacker News threads. That is not proof they are substitutes.

This week on Hacker News

Nov 7–Nov 14 UTC · Direct repository links and separately labeled related project stories. Indexed through 11/14/2026, 1:49:46 PM UTC.

No direct repository links were found in the indexed HN discussions over the past seven days. Older discussions are available below.

Mentions over time

HN mentions by year7 dated mentions

Observed mentions: 2020-08-04 16:48:41 UTC to 2023-10-07 08:05:03 UTC.

Counts come from the repository’s dated source evidence. Index cutoff: 2026-11-02 08:16:59 UTC.

View mention counts and date ranges

UTC ranges include their start and exclude their end. Select a period to filter the source evidence.
Period	Start (UTC)	End (exclusive, UTC)	Mentions	Comments
2020	2020-01-01 00:00:00 UTC	2021-01-01 00:00:00 UTC	2	2
2021	2021-01-01 00:00:00 UTC	2022-01-01 00:00:00 UTC	2	2
2022	2022-01-01 00:00:00 UTC	2023-01-01 00:00:00 UTC	1	1
2023	2023-01-01 00:00:00 UTC	2024-01-01 00:00:00 UTC	2	2

What Hacker News says

7 sources

janwas·Oct 7, 2023

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

Read the rest

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

In SIMD Everywhere Optimization from ARM Neon to RISC-V Vector Extensions

janwas·Jan 9, 2023

Here is my source: https://github.com/VcDevel/std-simd Ah, but this repo mentions that the GCC 11 implementation apparently also works with clang: https://github.com/VcDevel/Vc. Thanks!

In Transcoding Unicode with AVX-512: AMD Zen 4 vs. Intel Ice Lake

corysama·Dec 6, 2022

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

In Optimizing compilers reload vector constants needlessly

germandiago·Jul 16, 2021

I have based my career on top of C++/backend/soft-real time systems. I still have to read the full paper, thanks for the post! Many people rant about C++, but, IMHO, overall, taking into account ecosystem, tools, etc. C++ stands as an almost unbeatable technology when you put everything together. The language has quirks, asymmetries and all of that. C++ lives the curse/blessing of C compatibility. But since C++11 it is nicer to use and all the standards after it have been improving on it: generic lambdas, move semantics, structured bindings, string non-template parameters, constexpr and consteval, decltype, auto, concepts... and now the C/C++ ISO committees seem to be more coordinated: seems they are proposing subsets like attributes, lambdas or nullptr into C, directly, and bool and others. It is amazing what you can do with C++ that is

Read the rest

difficult or almost impossible to do with other languages, and you have basically all C interfaces available, which are the base of so many libraries since C is basically how you interface native code from almost everywhere. On the missing pieces I would mention that you need to use macros to have some kind of reflection for members and pattern matching and networking would be really nice to have. Also, resource embedding could be better (#embed and std::embed seem to be trying to solve this). Also, user-defined attributes for things such as struct MyType { [[serializable]] int myInt; ... } would be really nice and it works well in languages such as C#, Java and D. Modules are still an experiment implementation-wise, but hey, they will improve on the side of hiding implementation details by a big margin and have a lot of potential. As for the ecosystem, nowadays you have CMake (whose language sucks badly) and Meson (which I personally love). Together with Conan package management things have improved a lot since I started coding in around 2001. Pack that with an IDE like CLion or Visual Studio + Resharper or lightweight IDE (Emacs + Lsp and the like) and you have an environment that is very competitive and whose code can be compiled almost anywhere. From ARM to x86, MIPS and even Webassembly. That is why I think C++ is still the way to go if what you want is performance: you also have interfaces such as OpenCL/GL/Vulkan/SIMD libraries (though not C++ standard) where you can access hardware. Also, vendors and open source have things such as https://github.com/cginternals/glbinding or https://github.com/KhronosGroup/Vulkan-Hpp which help quite a bit and always oriented towards C++. It seems that C++ is becoming the natural child of C for inheriting graphics programming and some of the embedded area, though C is still king there. Or look at https://github.com/VcDevel/std-simd to access 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. :)

In Thriving in a Crowded and Changing World: C++ 2006–2020 [pdf]

jcelerier·Apr 8, 2021

... 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 ?? - Rust: those seem to be just the normal intrinsics which are available in every C++ compiler ? - Dart: seems to not go beyond SSE2 atm ? But it looks like the most "officially supported" of the bunch - Javascript: seems to be some intel-specific stuff which isn't available here on any of my JS environments ? * Standardized memory model - Literally false for Rust : https://doc.rust-lang.org/reference/memory-model.html - The C11 one directly comes from C++: https://stackoverflow.com/a/8877562/1495627 - The Go one does not seem to support acquire-release semantics, which ma

Read the rest

kes it quite removed from e.g. ARM and NVidia hardware from what I can read here ? https://golang.org/pkg/sync/atomic/

In All C++20 core language features with examples

nemequ1729·Aug 4, 2020

Apart from the fact that you have to rewrite your code, the big disadvantage to something like this is that it's a bit slower. With SIMDe you're still free to use functions like `_mm_maddubs_epi16` and they'll be really fast on x86, but still work everywhere. With xsimd (and similar libraries) you're generally limited to the lowest common denominator. FWIW, if abstraction layers are your thing you might want to look at std::experimental::simd (https://github.com/VcDevel/std-simd) instead. Google's Highway (https://github.com/google/highway) is also pretty interesting.

In SIMD Everywhere: 0.5.0

corysama·Aug 4, 2020Recommendation phrase

If you need to work on large matrices, Eigen is highly recommended. If you need to do a stream of custom processing in small steps, Eigen is not a good fit. Eigen is a super complicated, template-heavy library that can compile very heavyweight tasks into very efficient code (eventually). For custom work, https://github.com/VcDevel/std-simd is working it's way into the C++ standard.

In SIMD Everywhere: 0.5.0