[llvm] [LV][EVL] Support fixed-order recurrence idiom with EVL tail folding. (PR #124093)

[Luke Lau via llvm-commits]

================
@@ -1635,18 +1635,62 @@ void VPlanTransforms::addActiveLaneMask(
     HeaderMask->replaceAllUsesWith(LaneMask);
 }
 
+/// Adjust the way the resume value is obtained when using tail folding by EVL.
+/// Expanding ExtractFromEnd since the penultimate EVL could not equals to
+/// VFxUF. Expand
+///   %resume = ExtractFromEnd %vec, 1
+/// to
+///   %last.active.idx = sub %EVL, 1
+///   %resume = extractelement %vec, %last.active.idx
+static void adjustResumePhisForEVL(VPlan &Plan, VPValue &EVL) {
----------------
lukel97 wrote:

Do we need to adjust all ExtractFromEnd instructions, not just the ones that use ResumePhi and where the index is zero?

E.g. it looks like `addExitUsersForFirstOrderRecurrences` also creates an ExtractFromEnd with an index of 2. I think that should probably also be adjusted for EVL. 

Would it also maybe be simpler if we added a runtime VF operand to `ExtractFromEnd`, similar to `VPReverseVectorPointerRecipe`? Users would pass in Plan.getVF() when building it and VPInstruction::generate would generate the sub (if needed). Then the EVL transform just becomes a matter of replacing the runtime VF operand with EVL.

https://github.com/llvm/llvm-project/pull/124093