Menu
Does the new RyuJIT compiler ever generate vector (SIMD) CPU instructions, and when?
Side note: The System.Numerics namespace contains types that allow explicit use of Vector operations which may or may not generate SIMD instructions depending on the CPU, CLR version, JITer version, whether compiling directly to native code or not. This question is specifically about when non-vector code (e.g in C# or F#) will produce SIMD instructions.
354
Working of JIT Compiler: The JIT compiler is required to speed up the code execution and provide support for multiple platforms. Its working is given as follows: The JIT compiler converts the Microsoft Intermediate Language(MSIL) or Common Intermediate Language(CIL) into the machine code.
It allows for some run-time optimizations which are not (easily) possible at compile-time: for example, you can take advantage of special features on new CPUs, even if those CPUs didn't exist when you wrote your program - only the JIT compiler needs to know about that.
NET there are three types of JIT (Just-In-Time) compilers which are Explained as Under, Pre-JIT Compiler (Compiles entire code into native code completely) Econo JIT Compiler (Compiles code part by part freeing when required) Normal JIT Compiler (Compiles only that part of code when called and places in cache.
"Pre-jitting" or pre-compiling will improve performance, at start up, because you would be skipping that step. The reason that . NET JITs every time an app and its libraries load is so that it can run on many platforms and architectures with the best possible optimizations without the need for managing your builds.
SIMD code generation in RuyJIT is strictly limited to the types in the System.Numerics.Vectors namespace. Universal SIMD support is going to require a very significant revision of the CLR, such code can only be efficient if SIMD variables are aligned properly. At least to 16 for SSE2, to 32 to be able to use AVX2, to 64 for the upcoming AVX-512.
That's far off right now, the 32-bit CLR can only align to 4, the 64-bit version to 8. The "natural" alignment for resp 32-bit and 64-bit code. The required changes are going to affect about every part of the CLR, the garbage collector and class loader up front. There is no buzz about such a major change being considered. And no sign that it was considered in the CoreCLR project, it would have been the most obvious target version.
If you want to take advantage of SIMD beyond the current support in System.Numerics.Vectors then do so by using the C++ compiler, using the C++/CLI or C++/CX language extensions to interop.
61
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With
