mirror of
https://github.com/c64scene-ar/llvm-6502.git
synced 2024-07-24 08:29:39 +00:00
4ea3097d08
between splitting a vector into 128-bit lanes and recombining them vs. decomposing things into single-input shuffles and a final blend. This handles a large number of cases in AVX1 where the cross-lane shuffles would be much more expensive to represent even though we end up with a fast blend at the root. Instead, we can do a better job of shuffling in a single lane and then inserting it into the other lanes. This fixes the remaining bits of Halide's regression captured in PR21281 for AVX1. However, the bug persists in AVX2 because I've made this change reasonably conservative. The cases where it makes sense in AVX2 to split into 128-bit lanes are much more rare because we can often do full permutations across all elements of the 256-bit vector. However, the particular test case in PR21281 is an example of one of the rare cases where it is *always* better to work in a single 128-bit lane. I'm going to try to teach the logic to detect and form the good code even in AVX2 next, but it will need to use a separate heuristic. Finally, there is one pesky regression here where we previously would craftily use vpermilps in AVX1 to shuffle both high and low halves at the same time. We no longer pull that off, and not for any really good reason. Ultimately, I think this is just another missing nuance to the selection heuristic that I'll try to add in afterward, but this change already seems strictly worth doing considering the magnitude of the improvements in common matrix math shuffle patterns. As always, please let me know if this causes a surprising regression for you. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221861 91177308-0d34-0410-b5e6-96231b3b80d8 |
||
|---|---|---|
| .. | ||
| AArch64 | ||
| ARM | ||
| CPP | ||
| Generic | ||
| Hexagon | ||
| Inputs | ||
| Mips | ||
| MSP430 | ||
| NVPTX | ||
| PowerPC | ||
| R600 | ||
| SPARC | ||
| SystemZ | ||
| Thumb | ||
| Thumb2 | ||
| X86 | ||
| XCore | ||