[llvm-dev] Multidimensional array indexing intrinsics

Michael Kruse via llvm-dev llvm-dev at lists.llvm.org
Mon Aug 26 12:35:49 PDT 2019

Hi Prasanth,

sharing your implementation would be awesome! Note that the flang
compiler is currently being reimplemented in a project called f18 [1]
and will use MLIR for code-generation which has an in-built
multi-dimensional types. However, your approach might still be
insightful for lowering MLIR to LLVM-IR (for further optimizations)
and frontends that do not target MLIR.


[1] https://github.com/flang-compiler/f18

Am Mo., 26. Aug. 2019 um 09:45 Uhr schrieb Prashanth N R via llvm-dev
<llvm-dev at lists.llvm.org>:
> Linearized array addresses are an issue in loop nest transformation. To alleviate the same, a multidimensional  array indexing intrinsics have been proposed recently in the llvmp-dev mailing list [1]. From the mailing thread, it looks like there is a consensus on using intrinsics for communicating dimensions [2].
> While working with our own loop transformation framework, we did a similar work on using intrinsics in flang compiler. We are willing to work with the community and open source the implementation in flang compiler. Also we would like to extend the work to the backend to components like SCEV.
> Any feedback/suggestions are most welcome,
> thanks,
> -Prashanth N R
> Compiler Tree technologies
> 1. http://lists.llvm.org/pipermail/llvm-dev/2019-July/134063.html
> 2. http://lists.llvm.org/pipermail/llvm-dev/2019-August/134420.html
> _______________________________________________
> LLVM Developers mailing list
> llvm-dev at lists.llvm.org
> https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev

More information about the llvm-dev mailing list