Thanks for using Compiler Explorer
Sponsors
Jakt
C++
Ada
Analysis
Android Java
Android Kotlin
Assembly
C
C3
Carbon
C++ (Circle)
CIRCT
Clean
CMake
CMakeScript
COBOL
C++ for OpenCL
MLIR
Cppx
Cppx-Blue
Cppx-Gold
Cpp2-cppfront
Crystal
C#
CUDA C++
D
Dart
Elixir
Erlang
Fortran
F#
GLSL
Go
Haskell
HLSL
Hook
Hylo
IL
ispc
Java
Julia
Kotlin
LLVM IR
LLVM MIR
Modula-2
Nim
Objective-C
Objective-C++
OCaml
OpenCL C
Pascal
Pony
Python
Racket
Ruby
Rust
Snowball
Scala
Solidity
Spice
SPIR-V
Swift
LLVM TableGen
Toit
TypeScript Native
V
Vala
Visual Basic
WASM
Zig
Javascript
GIMPLE
Ygen
rust source #1
Output
Compile to binary object
Link to binary
Execute the code
Intel asm syntax
Demangle identifiers
Verbose demangling
Filters
Unused labels
Library functions
Directives
Comments
Horizontal whitespace
Debug intrinsics
Compiler
mrustc (master)
rustc 1.0.0
rustc 1.1.0
rustc 1.10.0
rustc 1.11.0
rustc 1.12.0
rustc 1.13.0
rustc 1.14.0
rustc 1.15.1
rustc 1.16.0
rustc 1.17.0
rustc 1.18.0
rustc 1.19.0
rustc 1.2.0
rustc 1.20.0
rustc 1.21.0
rustc 1.22.0
rustc 1.23.0
rustc 1.24.0
rustc 1.25.0
rustc 1.26.0
rustc 1.27.0
rustc 1.27.1
rustc 1.28.0
rustc 1.29.0
rustc 1.3.0
rustc 1.30.0
rustc 1.31.0
rustc 1.32.0
rustc 1.33.0
rustc 1.34.0
rustc 1.35.0
rustc 1.36.0
rustc 1.37.0
rustc 1.38.0
rustc 1.39.0
rustc 1.4.0
rustc 1.40.0
rustc 1.41.0
rustc 1.42.0
rustc 1.43.0
rustc 1.44.0
rustc 1.45.0
rustc 1.45.2
rustc 1.46.0
rustc 1.47.0
rustc 1.48.0
rustc 1.49.0
rustc 1.5.0
rustc 1.50.0
rustc 1.51.0
rustc 1.52.0
rustc 1.53.0
rustc 1.54.0
rustc 1.55.0
rustc 1.56.0
rustc 1.57.0
rustc 1.58.0
rustc 1.59.0
rustc 1.6.0
rustc 1.60.0
rustc 1.61.0
rustc 1.62.0
rustc 1.63.0
rustc 1.64.0
rustc 1.65.0
rustc 1.66.0
rustc 1.67.0
rustc 1.68.0
rustc 1.69.0
rustc 1.7.0
rustc 1.70.0
rustc 1.71.0
rustc 1.72.0
rustc 1.73.0
rustc 1.74.0
rustc 1.75.0
rustc 1.76.0
rustc 1.77.0
rustc 1.78.0
rustc 1.79.0
rustc 1.8.0
rustc 1.80.0
rustc 1.81.0
rustc 1.82.0
rustc 1.9.0
rustc beta
rustc nightly
rustc-cg-gcc (master)
x86-64 GCCRS (GCC master)
x86-64 GCCRS (GCCRS master)
x86-64 GCCRS 14.1 (GCC)
x86-64 GCCRS 14.2 (GCC)
Options
Source code
fn filter_paeth_autovec<const N: usize>(a: [u8; N], b: [u8; N], c: [u8; N]) -> [u8; N] { // This is an optimized version of the paeth filter from the PNG specification, proposed by // Luca Versari for [FPNGE](https://www.lucaversari.it/FJXL_and_FPNGE.pdf). It operates // entirely on unsigned 8-bit quantities, making it more conducive to vectorization. // // p = a + b - c // pa = |p - a| = |a + b - c - a| = |b - c| = max(b, c) - min(b, c) // pb = |p - b| = |a + b - c - b| = |a - c| = max(a, c) - min(a, c) // pc = |p - c| = |a + b - c - c| = |(b - c) + (a - c)| = ... // // Further optimizing the calculation of `pc` a bit tricker. However, notice that: // // a > c && b > c // ==> (a - c) > 0 && (b - c) > 0 // ==> pc > (a - c) && pc > (b - c) // ==> pc > |a - c| && pc > |b - c| // ==> pc > pb && pc > pa // // Meaning that if `c` is smaller than `a` and `b`, the value of `pc` is irrelevant. Similar // reasoning applies if `c` is larger than the other two inputs. Assuming that `c >= b` and // `c <= b` or vice versa: // // pc = ||b - c| - |a - c|| = |pa - pb| = max(pa, pb) - min(pa, pb) let pa = simd_sub(simd_max(b,c), simd_min(c,b)); let pb = simd_sub(simd_max(a,c), simd_min(c,a)); let mut pc = simd_sub(simd_max(pa, pb), simd_min(pa, pb)); for (i, o) in pc.iter_mut().enumerate() { if (a[i] < c[i]) != (c[i] < b[i]) { *o = 255; } } let mut crit_1 = [false; N]; for (i, o) in crit_1.iter_mut().enumerate() { if pa[i] <= pb[i] && pa[i] <= pc[i] { *o = true; } } let mut crit_2 = [false; N]; for (i, o) in crit_1.iter_mut().enumerate() { if pb[i] <= pc[i] { *o = true; } } let mut out = [0; N]; for (i, o) in out.iter_mut().enumerate() { if crit_1[i] { *o = a[i]; } else if crit_2[i] { *o = b[i]; } else { *o = c[i]; } } out } pub fn instantiate(a: [u8; 16], b: [u8; 16], c: [u8; 16]) -> [u8; 16] { filter_paeth_autovec(a, b, c) } #[inline] fn simd_max<const N: usize>(a: [u8; N], b: [u8; N]) -> [u8; N] { let mut out = [0; N]; for (i, o) in out.iter_mut().enumerate() { *o = a[i].max(b[i]) } out } #[inline] fn simd_min<const N: usize>(a: [u8; N], b: [u8; N]) -> [u8; N] { let mut out = [0; N]; for (i, o) in out.iter_mut().enumerate() { *o = a[i].min(b[i]) } out } #[inline] fn simd_sub<const N: usize>(a: [u8; N], b: [u8; N]) -> [u8; N] { let mut out = [0; N]; for (i, o) in out.iter_mut().enumerate() { *o = a[i] - b[i] } out }
Become a Patron
Sponsor on GitHub
Donate via PayPal
Source on GitHub
Mailing list
Installed libraries
Wiki
Report an issue
How it works
Contact the author
CE on Mastodon
About the author
Statistics
Changelog
Version tree