Stream: git-wasmtime

Topic: wasmtime / Issue #2369 Cranelift AArch64: Various small f...

Wasmtime GitHub notifications bot (Nov 05 2020 at 21:52):

akirilov-arm commented on Issue #2369:

Out of curiosity, is there a performance difference between fmov d0, d1 and mov v0.8b, v1.8b?

According to the Cortex-A55 optimization guide, the MOV latency is 2, while the FMOV one is 1. Strangely, the ORR (which is what MOV aliases to in this case) latency is also 1, but I have decided to be conservative and have made the change. That's the only difference I have found.

Personally, I have been trying to use the term Fpu in VCode names to refer to SIMD & FP scalars, while Vec (e.g. Inst::VecMisc) - to vectors (and yes, there are 128-bit "scalars", though that subtlety pertains mostly to loads, and then to big-endian environments). With that in mind FMOV is better, but ultimately it is the same operation.

Wasmtime GitHub notifications bot (Nov 12 2020 at 01:00):

cfallin commented on Issue #2369:

@akirilov-arm looks like there's a merge conflict on some tests; if you don't mind rebasing, I'm happy to merge this!

Wasmtime GitHub notifications bot (Nov 12 2020 at 22:41):

akirilov-arm commented on Issue #2369:

@cfallin Done.

Last updated: Nov 22 2024 at 17:03 UTC