## Quick Background
-**SIMD** stands for *Single Instruction, Multiple Data*. In other words, SIMD is when the CPU performs a single action on more that one logical pieces of data at the same time. Instead of adding two registers that each contain one `f32` value and getting an `f32` as the result, you might add two registers that each contain `f32x4` (128 bits of data) and then you get an `f32x4` as the output.
+**SIMD** stands for *Single Instruction, Multiple Data*. In other words, SIMD is when the CPU performs a single action on more than one logical piece of data at the same time. Instead of adding two registers that each contain one `f32` value and getting an `f32` as the result, you might add two registers that each contain `f32x4` (128 bits of data) and then you get an `f32x4` as the output.
This might seem a tiny bit weird at first, but there's a good reason for it. Back in the day, as CPUs got faster and faster, eventually they got so fast that the CPU would just melt itself. The heat management (heat sinks, fans, etc) simply couldn't keep up with how much electricity was going through the metal. Two main strategies were developed to help get around the limits of physics.
* One of them you're probably familiar with: Multi-core processors. By giving a processor more than one core, each core can do its own work, and because they're physically distant (at least on the CPU's scale) the heat can still be managed. Unfortunately, not all tasks can just be split up across cores in an efficient way.
* **Vector:** A SIMD value is called a vector. This shouldn't be confused with the `Vec<T>` type. A SIMD vector has a fixed size, known at compile time. All of the elements within the vector are of the same type. This makes vectors *similar to* arrays. One difference is that a vector is generally aligned to its *entire* size (eg: 16 bytes, 32 bytes, etc), not just the size of an individual element. Sometimes vector data is called "packed" data.
-* **Lane:** A single element position within a vector is called a lane. If you have `N` lanes available then they're numbered from `0` to `N-1` when referring to them, again like an array. The biggest difference between an array element and a vector lane is that it is *relatively costly* to access an individual lane value. Generally, the vector has to be pushed out of register onto the stack, then an individual lane is accessed while it's on the stack. For this reason, when working with SIMD you should avoid reading or writing the value of an individual lane during hot loops.
+* **Lane:** A single element position within a vector is called a lane. If you have `N` lanes available then they're numbered from `0` to `N-1` when referring to them, again like an array. The biggest difference between an array element and a vector lane is that in general is *relatively costly* to access an individual lane value. On most architectures, the vector has to be pushed out of the SIMD register onto the stack, then an individual lane is accessed while it's on the stack (and possibly the stack value is read back into a register). For this reason, when working with SIMD you should avoid reading or writing the value of an individual lane during hot loops.
* **Bit Widths:** When talking about SIMD, the bit widths used are the bit size of the vectors involved, *not* the individual elements. So "128-bit SIMD" has 128-bit vectors, and that might be `f32x4`, `i32x4`, `i16x8`, or other variations. While 128-bit SIMD is the most common, there's also 64-bit, 256-bit, and even 512-bit on the newest CPUs.
When using SIMD, you should be familiar with the CPU feature set that you're targeting.
-On `arm` and `aarch64` it's fairly simple. There's just one CPU feature that controls if SIMD is available: `neon` (or "NEON", all caps, as the ARM docs often put it). Neon registers are 128-bit, but they can also operate as 64-bit (the high lanes are just zeroed out).
+On `arm` and `aarch64` it's fairly simple. There's just one CPU feature that controls if SIMD is available: `neon` (or "NEON", all caps, as the ARM docs often put it). Neon registers can be used as 64-bit or 128-bit. When doing 128-bit operations it just uses two 64-bit registers as a single 128-bit register.
> By default, the `aarch64`, `arm`, and `thumb` Rust targets generally do not enable `neon` unless it's in the target string.
* 256-bit (mostly): `avx`, `avx2`, `fma`
* 512-bit (mostly): a *wide* range of `avx512` variations
+The list notes the bit widths available at each feature level, though the operations of the more advanced features can generally be used with the smaller register sizes as well. For example, new operations introduced in `avx` generally have a 128-bit form as well as a 256-bit form. This means that even if you only do 128-bit work you can still benefit from the later feature levels.
+
> By default, the `i686` and `x86_64` Rust targets enable `sse` and `sse2`.
### Selecting Additional Target Features
projects / rust.git / blobdiff