[Mlir-commits] [mlir] [TOSA] Usage of 32bit integer for 'index to float' in rfft2d (PR #75098)

Georgios Pinitas llvmlistbot at llvm.org
Tue Dec 12 02:04:28 PST 2023


GeorgeARM wrote:

Thanks for this PR. Overall it looks fine to me.
Could you please elaborate in your commit-msg what is the reasoning behind doing this?

If this is due to conversion targets not being able to handle 64bit integer arithmetic (e.g. SPIRV) then I think that this might require more discussion. Not sure how we can handle or if we need a more immediate casting operator from index to floating point that can be lowered later on as per target needs.

https://github.com/llvm/llvm-project/pull/75098


More information about the Mlir-commits mailing list