Mill Computing, Inc. › Forums › The Mill › Architecture › LCV and Vectorisation
- AuthorPosts
- #3208 |
I’m losing track of what’s been covered in the talks, but the pipelining talk covers loop-carried values (LCV) quite heavily, and I don’t recall it really being mentioned alongside vectorisation.
If writing a vectorised loop for B[i] = A[i] + A[i+2], do we need to shuffle and pick with constant metadata (tuned to the member) to construct a suitably-offset vector for the addition, or is there a better way? And how do we generate those member-tuned metadata values?
I lose track too 🙂
Automatic vectorization doesn’t work yet; the specializer and LLVM are having a hissy fit. For your particular example, the best code is probably just to load twice, with no shuffle involved. The size of the vector loaded is member-dependent, of course.
I’m sorry that LLVM is frustrating your team so much. I’d be happy to try to help with that, on my own time, but the logistics are probably too messy. (I’ve got language designitis, too, with a few languages under my belt.)
I was actually thinking of manual vectorisation in this case, and wondering about the situation in which we can’t afford a load or the loop-carried values are calculated. Though I’m thinking the answer might be the NYF streamers.
Many of my queries seem to boil down to wanting to rearrange vector elements. In this case (and in many others, I think), a shuffle that can use two element sources would do the trick, though two shuffles and a pick can do the same, if they can all run at the same cycle edge.
- AuthorPosts
You must be logged in to reply to this topic.