[llvm] 7759680 - [SROA] Avoid postponing rewriting load/store by ignoring lifetime intrinsics in partition's promotability checking

[Dmitry Vassiliev via llvm-commits]

Author: Dmitry Vassiliev
Date: 2022-05-17T11:25:59+02:00
New Revision: 7759680e2f88f6d055c37aeb77302874a72f19ce

URL: https://github.com/llvm/llvm-project/commit/7759680e2f88f6d055c37aeb77302874a72f19ce
DIFF: https://github.com/llvm/llvm-project/commit/7759680e2f88f6d055c37aeb77302874a72f19ce.diff

LOG: [SROA] Avoid postponing rewriting load/store by ignoring lifetime intrinsics in partition's promotability checking

This patch fixes a bug that generates unnecessary packing/unpacking structure code because of incorrectly handling lifetime intrinsic.
For example, a partition of an alloca may contain many slices:
```
Partition [0, 4):
  Slice0: [0, 4) used by: load i32 addr;
  Slice1: [0, 4) used by: store i32 v, addr;
  Slice2: [0, 16) used by lifetime.start(16, addr);
```
When SROA determines if the partition can be promoted, lifetime.start is currently treated as a whole alloca load/store, so Slice0 and Slice1 cannot be promoted at this attempt,
but the packing/unpacking code for Slice0 and Slice1 has been generated.
After rewrite lifetime.start/end intrinsic, SROA tries again with Slice0 and Slice1 and finally promotes them, but redundant packing/unpacking code remaining in the IRs.
This patch changes promotability checking to ignore lifetime intrinsic (they will be rewritten to correct sizes later), so we can promote the real users (load/store) at the first attempt with optimal code.

Reviewed By: nikic

Differential Revision: https://reviews.llvm.org/D124967

Added: 
    

Modified: 
    llvm/lib/Transforms/Scalar/SROA.cpp
    llvm/test/Transforms/SROA/lifetime-intrinsic.ll

Removed: 
    


################################################################################
diff  --git a/llvm/lib/Transforms/Scalar/SROA.cpp b/llvm/lib/Transforms/Scalar/SROA.cpp
index a1c335650dffe..ba0d6090a1bc9 100644
--- a/llvm/lib/Transforms/Scalar/SROA.cpp
+++ b/llvm/lib/Transforms/Scalar/SROA.cpp
@@ -1983,13 +1983,22 @@ static bool isIntegerWideningViableForSlice(const Slice &S,
   uint64_t RelBegin = S.beginOffset() - AllocBeginOffset;
   uint64_t RelEnd = S.endOffset() - AllocBeginOffset;
 
+  Use *U = S.getUse();
+
+  // Lifetime intrinsics operate over the whole alloca whose sizes are usually
+  // larger than other load/store slices (RelEnd > Size). But lifetime are
+  // always promotable and should not impact other slices' promotability of the
+  // partition.
+  if (IntrinsicInst *II = dyn_cast<IntrinsicInst>(U->getUser())) {
+    if (II->isLifetimeStartOrEnd() || II->isDroppable())
+      return true;
+  }
+
   // We can't reasonably handle cases where the load or store extends past
   // the end of the alloca's type and into its padding.
   if (RelEnd > Size)
     return false;
 
-  Use *U = S.getUse();
-
   if (LoadInst *LI = dyn_cast<LoadInst>(U->getUser())) {
     if (LI->isVolatile())
       return false;
@@ -2044,9 +2053,6 @@ static bool isIntegerWideningViableForSlice(const Slice &S,
       return false;
     if (!S.isSplittable())
       return false; // Skip any unsplittable intrinsics.
-  } else if (IntrinsicInst *II = dyn_cast<IntrinsicInst>(U->getUser())) {
-    if (!II->isLifetimeStartOrEnd() && !II->isDroppable())
-      return false;
   } else {
     return false;
   }

diff  --git a/llvm/test/Transforms/SROA/lifetime-intrinsic.ll b/llvm/test/Transforms/SROA/lifetime-intrinsic.ll
index 30df748a25359..97137a9d4de47 100644
--- a/llvm/test/Transforms/SROA/lifetime-intrinsic.ll
+++ b/llvm/test/Transforms/SROA/lifetime-intrinsic.ll
@@ -11,11 +11,9 @@
 
 define i16 @with_lifetime(i32 %a, i32 %b) #0 {
 ; CHECK-LABEL: @with_lifetime(
-; CHECK-NEXT:    [[ARR_SROA_0_SROA_0_0_EXTRACT_TRUNC:%.*]] = trunc i32 [[A:%.*]] to i16
-; CHECK-NEXT:    [[ARR_SROA_0_SROA_4_0_EXTRACT_SHIFT:%.*]] = lshr i32 [[A]], 16
-; CHECK-NEXT:    [[ARR_SROA_0_SROA_4_0_EXTRACT_TRUNC:%.*]] = trunc i32 [[ARR_SROA_0_SROA_4_0_EXTRACT_SHIFT]] to i16
+; CHECK-NEXT:    [[ARR_SROA_0_0_EXTRACT_TRUNC:%.*]] = trunc i32 [[A:%.*]] to i16
 ; CHECK-NEXT:    [[ARR_SROA_4_4_EXTRACT_TRUNC:%.*]] = trunc i32 [[B:%.*]] to i16
-; CHECK-NEXT:    [[RET:%.*]] = add i16 [[ARR_SROA_0_SROA_0_0_EXTRACT_TRUNC]], [[ARR_SROA_4_4_EXTRACT_TRUNC]]
+; CHECK-NEXT:    [[RET:%.*]] = add i16 [[ARR_SROA_0_0_EXTRACT_TRUNC]], [[ARR_SROA_4_4_EXTRACT_TRUNC]]
 ; CHECK-NEXT:    ret i16 [[RET]]
 ;
   %arr = alloca %i32x2, align 4