[NFC][AArch64][GlobalISel] Add test coverage for vector load/store legalization #134904
+282
−0
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
tobias-stadler
added
the
skip-precommit-approval
|
@llvm/pr-subscribers-llvm-globalisel @llvm/pr-subscribers-backend-aarch64 Author: Tobias Stadler (tobias-stadler) ChangesPrecommit tests for vector store legalization changes. This exposes a miscompile in LegalizerHelper::reduceLoadStoreWidth for non-byte-sized vector elements, which will be fixed in a follow-up patch. Full diff: https://github.com/llvm/llvm-project/pull/134904.diff 1 Files Affected:
diff --git a/llvm/test/CodeGen/AArch64/GlobalISel/legalize-store-vector.mir b/llvm/test/CodeGen/AArch64/GlobalISel/legalize-store-vector.mir
new file mode 100644
index 0000000000000..b91eef4df0649
--- /dev/null
+++ b/llvm/test/CodeGen/AArch64/GlobalISel/legalize-store-vector.mir
@@ -0,0 +1,149 @@
+# NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py UTC_ARGS: --version 5
+# RUN: llc -O0 -mtriple=aarch64 -verify-machineinstrs -run-pass=legalizer -global-isel-abort=1 %s -o - | FileCheck %s
+
+# FIXME: Scalarized loads for non-byte-sized vector elements overwrite the same address multiple times with partial values
+---
+name: store-narrow-non-byte-sized
+tracksRegLiveness: true
+body: |
+ bb.1:
+ liveins: $x8
+ ; CHECK-LABEL: name: store-narrow-non-byte-sized
+ ; CHECK: liveins: $x8
+ ; CHECK-NEXT: {{ $}}
+ ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(p0) = COPY $x8
+ ; CHECK-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 42
+ ; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(s32) = COPY [[C]](s32)
+ ; CHECK-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 511
+ ; CHECK-NEXT: [[AND:%[0-9]+]]:_(s32) = G_AND [[COPY1]], [[C1]]
+ ; CHECK-NEXT: [[TRUNC:%[0-9]+]]:_(s16) = G_TRUNC [[AND]](s32)
+ ; CHECK-NEXT: G_STORE [[TRUNC]](s16), [[COPY]](p0) :: (store (s16), align 16)
+ ; CHECK-NEXT: [[C2:%[0-9]+]]:_(s64) = G_CONSTANT i64 1
+ ; CHECK-NEXT: [[PTR_ADD:%[0-9]+]]:_(p0) = G_PTR_ADD [[COPY]], [[C2]](s64)
+ ; CHECK-NEXT: [[COPY2:%[0-9]+]]:_(s32) = COPY [[C]](s32)
+ ; CHECK-NEXT: [[COPY3:%[0-9]+]]:_(s32) = COPY [[C1]](s32)
+ ; CHECK-NEXT: [[AND1:%[0-9]+]]:_(s32) = G_AND [[COPY2]], [[COPY3]]
+ ; CHECK-NEXT: [[TRUNC1:%[0-9]+]]:_(s16) = G_TRUNC [[AND1]](s32)
+ ; CHECK-NEXT: G_STORE [[TRUNC1]](s16), [[PTR_ADD]](p0) :: (store (s16) into unknown-address + 1, align 1)
+ ; CHECK-NEXT: [[C3:%[0-9]+]]:_(s64) = G_CONSTANT i64 2
+ ; CHECK-NEXT: [[PTR_ADD1:%[0-9]+]]:_(p0) = G_PTR_ADD [[COPY]], [[C3]](s64)
+ ; CHECK-NEXT: [[COPY4:%[0-9]+]]:_(s32) = COPY [[C]](s32)
+ ; CHECK-NEXT: [[COPY5:%[0-9]+]]:_(s32) = COPY [[C1]](s32)
+ ; CHECK-NEXT: [[AND2:%[0-9]+]]:_(s32) = G_AND [[COPY4]], [[COPY5]]
+ ; CHECK-NEXT: [[TRUNC2:%[0-9]+]]:_(s16) = G_TRUNC [[AND2]](s32)
+ ; CHECK-NEXT: G_STORE [[TRUNC2]](s16), [[PTR_ADD1]](p0) :: (store (s16) into unknown-address + 2)
+ ; CHECK-NEXT: [[C4:%[0-9]+]]:_(s64) = G_CONSTANT i64 3
+ ; CHECK-NEXT: [[PTR_ADD2:%[0-9]+]]:_(p0) = G_PTR_ADD [[COPY]], [[C4]](s64)
+ ; CHECK-NEXT: [[COPY6:%[0-9]+]]:_(s32) = COPY [[C]](s32)
+ ; CHECK-NEXT: [[COPY7:%[0-9]+]]:_(s32) = COPY [[C1]](s32)
+ ; CHECK-NEXT: [[AND3:%[0-9]+]]:_(s32) = G_AND [[COPY6]], [[COPY7]]
+ ; CHECK-NEXT: [[TRUNC3:%[0-9]+]]:_(s16) = G_TRUNC [[AND3]](s32)
+ ; CHECK-NEXT: G_STORE [[TRUNC3]](s16), [[PTR_ADD2]](p0) :: (store (s16) into unknown-address + 3, align 1)
+ ; CHECK-NEXT: [[C5:%[0-9]+]]:_(s64) = G_CONSTANT i64 4
+ ; CHECK-NEXT: [[PTR_ADD3:%[0-9]+]]:_(p0) = G_PTR_ADD [[COPY]], [[C5]](s64)
+ ; CHECK-NEXT: [[COPY8:%[0-9]+]]:_(s32) = COPY [[C]](s32)
+ ; CHECK-NEXT: [[COPY9:%[0-9]+]]:_(s32) = COPY [[C1]](s32)
+ ; CHECK-NEXT: [[AND4:%[0-9]+]]:_(s32) = G_AND [[COPY8]], [[COPY9]]
+ ; CHECK-NEXT: [[TRUNC4:%[0-9]+]]:_(s16) = G_TRUNC [[AND4]](s32)
+ ; CHECK-NEXT: G_STORE [[TRUNC4]](s16), [[PTR_ADD3]](p0) :: (store (s16) into unknown-address + 4, align 4)
+ ; CHECK-NEXT: [[C6:%[0-9]+]]:_(s64) = G_CONSTANT i64 5
+ ; CHECK-NEXT: [[PTR_ADD4:%[0-9]+]]:_(p0) = G_PTR_ADD [[COPY]], [[C6]](s64)
+ ; CHECK-NEXT: [[COPY10:%[0-9]+]]:_(s32) = COPY [[C]](s32)
+ ; CHECK-NEXT: [[COPY11:%[0-9]+]]:_(s32) = COPY [[C1]](s32)
+ ; CHECK-NEXT: [[AND5:%[0-9]+]]:_(s32) = G_AND [[COPY10]], [[COPY11]]
+ ; CHECK-NEXT: [[TRUNC5:%[0-9]+]]:_(s16) = G_TRUNC [[AND5]](s32)
+ ; CHECK-NEXT: G_STORE [[TRUNC5]](s16), [[PTR_ADD4]](p0) :: (store (s16) into unknown-address + 5, align 1)
+ ; CHECK-NEXT: [[C7:%[0-9]+]]:_(s64) = G_CONSTANT i64 6
+ ; CHECK-NEXT: [[PTR_ADD5:%[0-9]+]]:_(p0) = G_PTR_ADD [[COPY]], [[C7]](s64)
+ ; CHECK-NEXT: [[COPY12:%[0-9]+]]:_(s32) = COPY [[C]](s32)
+ ; CHECK-NEXT: [[COPY13:%[0-9]+]]:_(s32) = COPY [[C1]](s32)
+ ; CHECK-NEXT: [[AND6:%[0-9]+]]:_(s32) = G_AND [[COPY12]], [[COPY13]]
+ ; CHECK-NEXT: [[TRUNC6:%[0-9]+]]:_(s16) = G_TRUNC [[AND6]](s32)
+ ; CHECK-NEXT: G_STORE [[TRUNC6]](s16), [[PTR_ADD5]](p0) :: (store (s16) into unknown-address + 6)
+ ; CHECK-NEXT: [[C8:%[0-9]+]]:_(s64) = G_CONSTANT i64 7
+ ; CHECK-NEXT: [[PTR_ADD6:%[0-9]+]]:_(p0) = G_PTR_ADD [[COPY]], [[C8]](s64)
+ ; CHECK-NEXT: [[COPY14:%[0-9]+]]:_(s32) = COPY [[C1]](s32)
+ ; CHECK-NEXT: [[AND7:%[0-9]+]]:_(s32) = G_AND [[C]], [[COPY14]]
+ ; CHECK-NEXT: [[TRUNC7:%[0-9]+]]:_(s16) = G_TRUNC [[AND7]](s32)
+ ; CHECK-NEXT: G_STORE [[TRUNC7]](s16), [[PTR_ADD6]](p0) :: (store (s16) into unknown-address + 7, align 1)
+ ; CHECK-NEXT: RET_ReallyLR
+ %0:_(p0) = COPY $x8
+ %1:_(s9) = G_CONSTANT i9 42
+ %2:_(<8 x s9>) = G_BUILD_VECTOR %1(s9), %1(s9), %1(s9), %1(s9), %1(s9), %1(s9), %1(s9), %1(s9)
+ G_STORE %2(<8 x s9>), %0(p0) :: (store (<8 x s9>), align 16)
+ RET_ReallyLR
+...
+
+# FIXME: Vector stores only sometimes act as per-lane truncating stores, see PR#121169.
+---
+name: store-narrow-per-lane-trunc
+tracksRegLiveness: true
+body: |
+ bb.1:
+ liveins: $x8
+ ; CHECK-LABEL: name: store-narrow-per-lane-trunc
+ ; CHECK: liveins: $x8
+ ; CHECK-NEXT: {{ $}}
+ ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(p0) = COPY $x8
+ ; CHECK-NEXT: [[C:%[0-9]+]]:_(s64) = G_CONSTANT i64 42
+ ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<2 x s64>) = G_BUILD_VECTOR [[C]](s64), [[C]](s64)
+ ; CHECK-NEXT: [[BUILD_VECTOR1:%[0-9]+]]:_(<2 x s64>) = G_BUILD_VECTOR [[C]](s64), [[C]](s64)
+ ; CHECK-NEXT: G_STORE [[BUILD_VECTOR]](<2 x s64>), [[COPY]](p0) :: (store (<2 x s64>))
+ ; CHECK-NEXT: [[C1:%[0-9]+]]:_(s64) = G_CONSTANT i64 16
+ ; CHECK-NEXT: [[PTR_ADD:%[0-9]+]]:_(p0) = G_PTR_ADD [[COPY]], [[C1]](s64)
+ ; CHECK-NEXT: G_STORE [[BUILD_VECTOR1]](<2 x s64>), [[PTR_ADD]](p0) :: (store (<2 x s64>) into unknown-address + 16)
+ ; CHECK-NEXT: RET_ReallyLR
+ %0:_(p0) = COPY $x8
+ %1:_(s64) = G_CONSTANT i64 42
+ %2:_(<4 x s64>) = G_BUILD_VECTOR %1(s64), %1(s64), %1(s64), %1(s64)
+ G_STORE %2(<4 x s64>), %0(p0) :: (store (<4 x s63>), align 16)
+ RET_ReallyLR
+...
+
+# FIXME: Clarify behavior of stores between scalar and vector types in documentation
+---
+name: store-narrow-vector-high-bits
+tracksRegLiveness: true
+body: |
+ bb.1:
+ liveins: $x8
+ ; CHECK-LABEL: name: store-narrow-vector-high-bits
+ ; CHECK: liveins: $x8
+ ; CHECK-NEXT: {{ $}}
+ ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(p0) = COPY $x8
+ ; CHECK-NEXT: [[C:%[0-9]+]]:_(s64) = G_CONSTANT i64 42
+ ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<2 x s64>) = G_BUILD_VECTOR [[C]](s64), [[C]](s64)
+ ; CHECK-NEXT: [[BUILD_VECTOR1:%[0-9]+]]:_(<2 x s64>) = G_BUILD_VECTOR [[C]](s64), [[C]](s64)
+ ; CHECK-NEXT: G_STORE [[BUILD_VECTOR]](<2 x s64>), [[COPY]](p0) :: (store (<2 x s64>))
+ ; CHECK-NEXT: [[C1:%[0-9]+]]:_(s64) = G_CONSTANT i64 16
+ ; CHECK-NEXT: [[PTR_ADD:%[0-9]+]]:_(p0) = G_PTR_ADD [[COPY]], [[C1]](s64)
+ ; CHECK-NEXT: G_STORE [[BUILD_VECTOR1]](<2 x s64>), [[PTR_ADD]](p0) :: (store (<2 x s64>) into unknown-address + 16)
+ ; CHECK-NEXT: RET_ReallyLR
+ %0:_(p0) = COPY $x8
+ %1:_(s64) = G_CONSTANT i64 42
+ %2:_(<4 x s64>) = G_BUILD_VECTOR %1(s64), %1(s64), %1(s64), %1(s64)
+ G_STORE %2(<4 x s64>), %0(p0) :: (store (s252), align 16)
+ RET_ReallyLR
+...
+---
+name: store-narrow-scalar-high-bits
+tracksRegLiveness: true
+body: |
+ bb.1:
+ liveins: $x8
+ ; CHECK-LABEL: name: store-narrow-scalar-high-bits
+ ; CHECK: liveins: $x8
+ ; CHECK-NEXT: {{ $}}
+ ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(p0) = COPY $x8
+ ; CHECK-NEXT: [[C:%[0-9]+]]:_(s64) = G_CONSTANT i64 42
+ ; CHECK-NEXT: [[C1:%[0-9]+]]:_(s64) = G_CONSTANT i64 0
+ ; CHECK-NEXT: G_STORE [[C]](s64), [[COPY]](p0) :: (store (s64), align 16)
+ ; CHECK-NEXT: [[C2:%[0-9]+]]:_(s64) = G_CONSTANT i64 8
+ ; CHECK-NEXT: [[PTR_ADD:%[0-9]+]]:_(p0) = G_PTR_ADD [[COPY]], [[C2]](s64)
+ ; CHECK-NEXT: G_STORE [[C1]](s64), [[PTR_ADD]](p0) :: (store (s64) into unknown-address + 8)
+ ; CHECK-NEXT: RET_ReallyLR
+ %0:_(p0) = COPY $x8
+ %1:_(s128) = G_CONSTANT i128 42
+ G_STORE %1(s128), %0(p0) :: (store (<2 x s63>), align 16)
+ RET_ReallyLR
+...
|
tobias-stadler
force-pushed
the
fix-gisel-vecstore-test
branch
from
25be44e to
89756ea
Compare
tobias-stadler
changed the title
arsenm
approved these changes
Apr 22, 2025
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/18/builds/14920 Here is the relevant piece of the build log for the reference |
IanWood1
pushed a commit
to IanWood1/llvm-project
that referenced
this pull request
May 6, 2025
…galization (llvm#134904) Precommit tests for vector load/store legalization changes. This exposes a miscompile in LegalizerHelper::reduceLoadStoreWidth for non-byte-sized vector elements, which will be fixed in a follow-up patch. The other tests are potential miscompilations due to unclear semantics of vector load/stores, which will be addressed in a follow-up discussion.
IanWood1
pushed a commit
to IanWood1/llvm-project
that referenced
this pull request
May 6, 2025
…-byte-sized types (llvm#136739) LegalizerHelper::reduceLoadStoreWidth does not work for non-byte-sized types, because this would require (un)packing of bits across byte boundaries. Precommit tests: llvm#134904
IanWood1
pushed a commit
to IanWood1/llvm-project
that referenced
this pull request
May 6, 2025
…galization (llvm#134904) Precommit tests for vector load/store legalization changes. This exposes a miscompile in LegalizerHelper::reduceLoadStoreWidth for non-byte-sized vector elements, which will be fixed in a follow-up patch. The other tests are potential miscompilations due to unclear semantics of vector load/stores, which will be addressed in a follow-up discussion.
IanWood1
pushed a commit
to IanWood1/llvm-project
that referenced
this pull request
May 6, 2025
…-byte-sized types (llvm#136739) LegalizerHelper::reduceLoadStoreWidth does not work for non-byte-sized types, because this would require (un)packing of bits across byte boundaries. Precommit tests: llvm#134904
IanWood1
pushed a commit
to IanWood1/llvm-project
that referenced
this pull request
May 6, 2025
…galization (llvm#134904) Precommit tests for vector load/store legalization changes. This exposes a miscompile in LegalizerHelper::reduceLoadStoreWidth for non-byte-sized vector elements, which will be fixed in a follow-up patch. The other tests are potential miscompilations due to unclear semantics of vector load/stores, which will be addressed in a follow-up discussion.
IanWood1
pushed a commit
to IanWood1/llvm-project
that referenced
this pull request
May 6, 2025
…-byte-sized types (llvm#136739) LegalizerHelper::reduceLoadStoreWidth does not work for non-byte-sized types, because this would require (un)packing of bits across byte boundaries. Precommit tests: llvm#134904
GeorgeARM
pushed a commit
to GeorgeARM/llvm-project
that referenced
this pull request
May 7, 2025
…-byte-sized types (llvm#136739) LegalizerHelper::reduceLoadStoreWidth does not work for non-byte-sized types, because this would require (un)packing of bits across byte boundaries. Precommit tests: llvm#134904
Ankur-0429
pushed a commit
to Ankur-0429/llvm-project
that referenced
this pull request
May 9, 2025
…-byte-sized types (llvm#136739) LegalizerHelper::reduceLoadStoreWidth does not work for non-byte-sized types, because this would require (un)packing of bits across byte boundaries. Precommit tests: llvm#134904
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Precommit tests for vector load/store legalization changes. This exposes a miscompile in LegalizerHelper::reduceLoadStoreWidth for non-byte-sized vector elements, which will be fixed in a follow-up patch.
The other tests are potential miscompilations due to unclear semantics of vector load/stores, which will be addressed in a follow-up discussion.