[ARM][MVE] Tail-predication: remove the BTC + 1 overflow checks
This adapts tail-predication to the new semantics of get.active.lane.mask as
defined in D86147. This means that:
- we can remove the BTC + 1 overflow checks because now the loop tripcount is
passed in to the intrinsic,
- we can immediately use that value to setup a counter for the number of
elements processed by the loop and don't need to materialize BTC + 1.
Differential Revision: https://reviews.llvm.org/D86303
diff --git a/llvm/test/CodeGen/Thumb2/mve-vecreduce-loops.ll b/llvm/test/CodeGen/Thumb2/mve-vecreduce-loops.ll
index 539f760..64a76f3 100644
--- a/llvm/test/CodeGen/Thumb2/mve-vecreduce-loops.ll
+++ b/llvm/test/CodeGen/Thumb2/mve-vecreduce-loops.ll
@@ -1730,7 +1730,7 @@
vector.body: ; preds = %vector.body, %vector.ph
%index = phi i32 [ 0, %vector.ph ], [ %index.next, %vector.body ]
%vec.phi = phi i32 [ 0, %vector.ph ], [ %4, %vector.body ]
- %active.lane.mask = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 %index, i32 %trip.count.minus.1)
+ %active.lane.mask = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 %index, i32 %n)
%0 = getelementptr inbounds i32, i32* %x, i32 %index
%1 = bitcast i32* %0 to <4 x i32>*
%wide.masked.load = call <4 x i32> @llvm.masked.load.v4i32.p0v4i32(<4 x i32>* %1, i32 4, <4 x i1> %active.lane.mask, <4 x i32> undef)
@@ -1781,7 +1781,7 @@
vector.body: ; preds = %vector.body, %vector.ph
%index = phi i32 [ 0, %vector.ph ], [ %index.next, %vector.body ]
%vec.phi = phi i32 [ 0, %vector.ph ], [ %7, %vector.body ]
- %active.lane.mask = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 %index, i32 %trip.count.minus.1)
+ %active.lane.mask = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 %index, i32 %n)
%0 = getelementptr inbounds i32, i32* %x, i32 %index
%1 = bitcast i32* %0 to <4 x i32>*
%wide.masked.load = call <4 x i32> @llvm.masked.load.v4i32.p0v4i32(<4 x i32>* %1, i32 4, <4 x i1> %active.lane.mask, <4 x i32> undef)
@@ -1835,7 +1835,7 @@
vector.body: ; preds = %vector.body, %vector.ph
%index = phi i32 [ 0, %vector.ph ], [ %index.next, %vector.body ]
%vec.phi = phi i32 [ 0, %vector.ph ], [ %5, %vector.body ]
- %active.lane.mask = call <8 x i1> @llvm.get.active.lane.mask.v8i1.i32(i32 %index, i32 %trip.count.minus.1)
+ %active.lane.mask = call <8 x i1> @llvm.get.active.lane.mask.v8i1.i32(i32 %index, i32 %n)
%0 = getelementptr inbounds i16, i16* %x, i32 %index
%1 = bitcast i16* %0 to <8 x i16>*
%wide.masked.load = call <8 x i16> @llvm.masked.load.v8i16.p0v8i16(<8 x i16>* %1, i32 2, <8 x i1> %active.lane.mask, <8 x i16> undef)
@@ -1887,7 +1887,7 @@
vector.body: ; preds = %vector.body, %vector.ph
%index = phi i32 [ 0, %vector.ph ], [ %index.next, %vector.body ]
%vec.phi = phi i32 [ 0, %vector.ph ], [ %9, %vector.body ]
- %active.lane.mask = call <8 x i1> @llvm.get.active.lane.mask.v8i1.i32(i32 %index, i32 %trip.count.minus.1)
+ %active.lane.mask = call <8 x i1> @llvm.get.active.lane.mask.v8i1.i32(i32 %index, i32 %n)
%0 = getelementptr inbounds i16, i16* %x, i32 %index
%1 = bitcast i16* %0 to <8 x i16>*
%wide.masked.load = call <8 x i16> @llvm.masked.load.v8i16.p0v8i16(<8 x i16>* %1, i32 2, <8 x i1> %active.lane.mask, <8 x i16> undef)
@@ -1943,7 +1943,7 @@
vector.body: ; preds = %vector.body, %vector.ph
%index = phi i32 [ 0, %vector.ph ], [ %index.next, %vector.body ]
%vec.phi = phi i32 [ 0, %vector.ph ], [ %5, %vector.body ]
- %active.lane.mask = call <16 x i1> @llvm.get.active.lane.mask.v16i1.i32(i32 %index, i32 %trip.count.minus.1)
+ %active.lane.mask = call <16 x i1> @llvm.get.active.lane.mask.v16i1.i32(i32 %index, i32 %n)
%0 = getelementptr inbounds i8, i8* %x, i32 %index
%1 = bitcast i8* %0 to <16 x i8>*
%wide.masked.load = call <16 x i8> @llvm.masked.load.v16i8.p0v16i8(<16 x i8>* %1, i32 1, <16 x i1> %active.lane.mask, <16 x i8> undef)
@@ -1995,7 +1995,7 @@
vector.body: ; preds = %vector.body, %vector.ph
%index = phi i32 [ 0, %vector.ph ], [ %index.next, %vector.body ]
%vec.phi = phi i32 [ 0, %vector.ph ], [ %9, %vector.body ]
- %active.lane.mask = call <16 x i1> @llvm.get.active.lane.mask.v16i1.i32(i32 %index, i32 %trip.count.minus.1)
+ %active.lane.mask = call <16 x i1> @llvm.get.active.lane.mask.v16i1.i32(i32 %index, i32 %n)
%0 = getelementptr inbounds i8, i8* %x, i32 %index
%1 = bitcast i8* %0 to <16 x i8>*
%wide.masked.load = call <16 x i8> @llvm.masked.load.v16i8.p0v16i8(<16 x i8>* %1, i32 1, <16 x i1> %active.lane.mask, <16 x i8> undef)
@@ -2051,7 +2051,7 @@
vector.body: ; preds = %vector.body, %vector.ph
%index = phi i32 [ 0, %vector.ph ], [ %index.next, %vector.body ]
%vec.phi = phi i16 [ 0, %vector.ph ], [ %4, %vector.body ]
- %active.lane.mask = call <8 x i1> @llvm.get.active.lane.mask.v8i1.i32(i32 %index, i32 %trip.count.minus.1)
+ %active.lane.mask = call <8 x i1> @llvm.get.active.lane.mask.v8i1.i32(i32 %index, i32 %n)
%0 = getelementptr inbounds i16, i16* %x, i32 %index
%1 = bitcast i16* %0 to <8 x i16>*
%wide.masked.load = call <8 x i16> @llvm.masked.load.v8i16.p0v8i16(<8 x i16>* %1, i32 2, <8 x i1> %active.lane.mask, <8 x i16> undef)
@@ -2102,7 +2102,7 @@
vector.body: ; preds = %vector.body, %vector.ph
%index = phi i32 [ 0, %vector.ph ], [ %index.next, %vector.body ]
%vec.phi = phi i16 [ 0, %vector.ph ], [ %7, %vector.body ]
- %active.lane.mask = call <8 x i1> @llvm.get.active.lane.mask.v8i1.i32(i32 %index, i32 %trip.count.minus.1)
+ %active.lane.mask = call <8 x i1> @llvm.get.active.lane.mask.v8i1.i32(i32 %index, i32 %n)
%0 = getelementptr inbounds i16, i16* %x, i32 %index
%1 = bitcast i16* %0 to <8 x i16>*
%wide.masked.load = call <8 x i16> @llvm.masked.load.v8i16.p0v8i16(<8 x i16>* %1, i32 2, <8 x i1> %active.lane.mask, <8 x i16> undef)
@@ -2156,7 +2156,7 @@
vector.body: ; preds = %vector.body, %vector.ph
%index = phi i32 [ 0, %vector.ph ], [ %index.next, %vector.body ]
%vec.phi = phi i16 [ 0, %vector.ph ], [ %5, %vector.body ]
- %active.lane.mask = call <16 x i1> @llvm.get.active.lane.mask.v16i1.i32(i32 %index, i32 %trip.count.minus.1)
+ %active.lane.mask = call <16 x i1> @llvm.get.active.lane.mask.v16i1.i32(i32 %index, i32 %n)
%0 = getelementptr inbounds i8, i8* %x, i32 %index
%1 = bitcast i8* %0 to <16 x i8>*
%wide.masked.load = call <16 x i8> @llvm.masked.load.v16i8.p0v16i8(<16 x i8>* %1, i32 1, <16 x i1> %active.lane.mask, <16 x i8> undef)
@@ -2208,7 +2208,7 @@
vector.body: ; preds = %vector.body, %vector.ph
%index = phi i32 [ 0, %vector.ph ], [ %index.next, %vector.body ]
%vec.phi = phi i16 [ 0, %vector.ph ], [ %9, %vector.body ]
- %active.lane.mask = call <16 x i1> @llvm.get.active.lane.mask.v16i1.i32(i32 %index, i32 %trip.count.minus.1)
+ %active.lane.mask = call <16 x i1> @llvm.get.active.lane.mask.v16i1.i32(i32 %index, i32 %n)
%0 = getelementptr inbounds i8, i8* %x, i32 %index
%1 = bitcast i8* %0 to <16 x i8>*
%wide.masked.load = call <16 x i8> @llvm.masked.load.v16i8.p0v16i8(<16 x i8>* %1, i32 1, <16 x i1> %active.lane.mask, <16 x i8> undef)
@@ -2264,7 +2264,7 @@
vector.body: ; preds = %vector.body, %vector.ph
%index = phi i32 [ 0, %vector.ph ], [ %index.next, %vector.body ]
%vec.phi = phi i8 [ 0, %vector.ph ], [ %4, %vector.body ]
- %active.lane.mask = call <16 x i1> @llvm.get.active.lane.mask.v16i1.i32(i32 %index, i32 %trip.count.minus.1)
+ %active.lane.mask = call <16 x i1> @llvm.get.active.lane.mask.v16i1.i32(i32 %index, i32 %n)
%0 = getelementptr inbounds i8, i8* %x, i32 %index
%1 = bitcast i8* %0 to <16 x i8>*
%wide.masked.load = call <16 x i8> @llvm.masked.load.v16i8.p0v16i8(<16 x i8>* %1, i32 1, <16 x i1> %active.lane.mask, <16 x i8> undef)
@@ -2315,7 +2315,7 @@
vector.body: ; preds = %vector.body, %vector.ph
%index = phi i32 [ 0, %vector.ph ], [ %index.next, %vector.body ]
%vec.phi = phi i8 [ 0, %vector.ph ], [ %7, %vector.body ]
- %active.lane.mask = call <16 x i1> @llvm.get.active.lane.mask.v16i1.i32(i32 %index, i32 %trip.count.minus.1)
+ %active.lane.mask = call <16 x i1> @llvm.get.active.lane.mask.v16i1.i32(i32 %index, i32 %n)
%0 = getelementptr inbounds i8, i8* %x, i32 %index
%1 = bitcast i8* %0 to <16 x i8>*
%wide.masked.load = call <16 x i8> @llvm.masked.load.v16i8.p0v16i8(<16 x i8>* %1, i32 1, <16 x i1> %active.lane.mask, <16 x i8> undef)
@@ -2371,7 +2371,7 @@
vector.body: ; preds = %vector.body, %vector.ph
%index = phi i32 [ 0, %vector.ph ], [ %index.next, %vector.body ]
%vec.phi = phi i64 [ 0, %vector.ph ], [ %5, %vector.body ]
- %active.lane.mask = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 %index, i32 %trip.count.minus.1)
+ %active.lane.mask = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 %index, i32 %n)
%0 = getelementptr inbounds i32, i32* %x, i32 %index
%1 = bitcast i32* %0 to <4 x i32>*
%wide.masked.load = call <4 x i32> @llvm.masked.load.v4i32.p0v4i32(<4 x i32>* %1, i32 4, <4 x i1> %active.lane.mask, <4 x i32> undef)
@@ -2425,7 +2425,7 @@
vector.body: ; preds = %vector.body, %vector.ph
%index = phi i32 [ 0, %vector.ph ], [ %index.next, %vector.body ]
%vec.phi = phi i64 [ 0, %vector.ph ], [ %9, %vector.body ]
- %active.lane.mask = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 %index, i32 %trip.count.minus.1)
+ %active.lane.mask = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 %index, i32 %n)
%0 = getelementptr inbounds i32, i32* %x, i32 %index
%1 = bitcast i32* %0 to <4 x i32>*
%wide.masked.load = call <4 x i32> @llvm.masked.load.v4i32.p0v4i32(<4 x i32>* %1, i32 4, <4 x i1> %active.lane.mask, <4 x i32> undef)
@@ -2484,7 +2484,7 @@
vector.body: ; preds = %vector.body, %vector.ph
%index = phi i32 [ 0, %vector.ph ], [ %index.next, %vector.body ]
%vec.phi = phi i64 [ 0, %vector.ph ], [ %9, %vector.body ]
- %active.lane.mask = call <8 x i1> @llvm.get.active.lane.mask.v8i1.i32(i32 %index, i32 %trip.count.minus.1)
+ %active.lane.mask = call <8 x i1> @llvm.get.active.lane.mask.v8i1.i32(i32 %index, i32 %n)
%0 = getelementptr inbounds i16, i16* %x, i32 %index
%1 = bitcast i16* %0 to <8 x i16>*
%wide.masked.load = call <8 x i16> @llvm.masked.load.v8i16.p0v8i16(<8 x i16>* %1, i32 2, <8 x i1> %active.lane.mask, <8 x i16> undef)