release/21.x: [NVPTX] Disable relative lookup tables (#159748) #160064
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
Member
Author
|
@Artem-B What do you think about merging this PR to the release branch? |
Member
Author
|
@llvm/pr-subscribers-llvm-transforms Author: None (llvmbot) ChangesBackport 1ed1537 Requested by: @nikic Full diff: https://github.com/llvm/llvm-project/pull/160064.diff 2 Files Affected:
diff --git a/llvm/lib/Target/NVPTX/NVPTXTargetTransformInfo.h b/llvm/lib/Target/NVPTX/NVPTXTargetTransformInfo.h
index 9a6e261c811a0..437edfb8ffc75 100644
--- a/llvm/lib/Target/NVPTX/NVPTXTargetTransformInfo.h
+++ b/llvm/lib/Target/NVPTX/NVPTXTargetTransformInfo.h
@@ -183,6 +183,11 @@ class NVPTXTTIImpl final : public BasicTTIImplBase<NVPTXTTIImpl> {
void collectKernelLaunchBounds(
const Function &F,
SmallVectorImpl<std::pair<StringRef, int64_t>> &LB) const override;
+
+ bool shouldBuildRelLookupTables() const override {
+ // Self-referential globals are not supported.
+ return false;
+ }
};
} // end namespace llvm
diff --git a/llvm/test/Transforms/RelLookupTableConverter/nvptx.ll b/llvm/test/Transforms/RelLookupTableConverter/nvptx.ll
new file mode 100644
index 0000000000000..70ebf220c369c
--- /dev/null
+++ b/llvm/test/Transforms/RelLookupTableConverter/nvptx.ll
@@ -0,0 +1,32 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --check-globals all --version 5
+; RUN: opt < %s -passes=rel-lookup-table-converter -relocation-model=pic -S | FileCheck %s
+; REQUIRES: nvptx-registered-target
+target triple = "nvptx64-nvidia-cuda"
+
+; Do not produce relative lookup table for nvptx target.
+
+@a1 = internal constant i32 0, align 4
+@b1 = internal constant i32 0, align 4
+@c1 = internal constant i32 0, align 4
+@d1 = internal constant i32 0, align 4
+
+@switch.table = private unnamed_addr constant [3 x ptr] [ptr @a1, ptr @b1, ptr @c1], align 8
+
+;.
+; CHECK: @a1 = internal constant i32 0, align 4
+; CHECK: @b1 = internal constant i32 0, align 4
+; CHECK: @c1 = internal constant i32 0, align 4
+; CHECK: @d1 = internal constant i32 0, align 4
+; CHECK: @switch.table = private unnamed_addr constant [3 x ptr] [ptr @a1, ptr @b1, ptr @c1], align 8
+;.
+define ptr @internal_linkage(i32 %cond) {
+; CHECK-LABEL: define ptr @internal_linkage(
+; CHECK-SAME: i32 [[COND:%.*]]) {
+; CHECK-NEXT: [[SWITCH_GEP:%.*]] = getelementptr inbounds [3 x ptr], ptr @switch.table, i32 0, i32 [[COND]]
+; CHECK-NEXT: [[RELTABLE_INTRINSIC:%.*]] = load ptr, ptr [[SWITCH_GEP]], align 8
+; CHECK-NEXT: ret ptr [[RELTABLE_INTRINSIC]]
+;
+ %switch.gep = getelementptr inbounds [3 x ptr], ptr @switch.table, i32 0, i32 %cond
+ %switch.load = load ptr, ptr %switch.gep, align 8
+ ret ptr %switch.load
+}
|
Member
Author
|
@llvm/pr-subscribers-backend-nvptx Author: None (llvmbot) ChangesBackport 1ed1537 Requested by: @nikic Full diff: https://github.com/llvm/llvm-project/pull/160064.diff 2 Files Affected:
diff --git a/llvm/lib/Target/NVPTX/NVPTXTargetTransformInfo.h b/llvm/lib/Target/NVPTX/NVPTXTargetTransformInfo.h
index 9a6e261c811a0..437edfb8ffc75 100644
--- a/llvm/lib/Target/NVPTX/NVPTXTargetTransformInfo.h
+++ b/llvm/lib/Target/NVPTX/NVPTXTargetTransformInfo.h
@@ -183,6 +183,11 @@ class NVPTXTTIImpl final : public BasicTTIImplBase<NVPTXTTIImpl> {
void collectKernelLaunchBounds(
const Function &F,
SmallVectorImpl<std::pair<StringRef, int64_t>> &LB) const override;
+
+ bool shouldBuildRelLookupTables() const override {
+ // Self-referential globals are not supported.
+ return false;
+ }
};
} // end namespace llvm
diff --git a/llvm/test/Transforms/RelLookupTableConverter/nvptx.ll b/llvm/test/Transforms/RelLookupTableConverter/nvptx.ll
new file mode 100644
index 0000000000000..70ebf220c369c
--- /dev/null
+++ b/llvm/test/Transforms/RelLookupTableConverter/nvptx.ll
@@ -0,0 +1,32 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --check-globals all --version 5
+; RUN: opt < %s -passes=rel-lookup-table-converter -relocation-model=pic -S | FileCheck %s
+; REQUIRES: nvptx-registered-target
+target triple = "nvptx64-nvidia-cuda"
+
+; Do not produce relative lookup table for nvptx target.
+
+@a1 = internal constant i32 0, align 4
+@b1 = internal constant i32 0, align 4
+@c1 = internal constant i32 0, align 4
+@d1 = internal constant i32 0, align 4
+
+@switch.table = private unnamed_addr constant [3 x ptr] [ptr @a1, ptr @b1, ptr @c1], align 8
+
+;.
+; CHECK: @a1 = internal constant i32 0, align 4
+; CHECK: @b1 = internal constant i32 0, align 4
+; CHECK: @c1 = internal constant i32 0, align 4
+; CHECK: @d1 = internal constant i32 0, align 4
+; CHECK: @switch.table = private unnamed_addr constant [3 x ptr] [ptr @a1, ptr @b1, ptr @c1], align 8
+;.
+define ptr @internal_linkage(i32 %cond) {
+; CHECK-LABEL: define ptr @internal_linkage(
+; CHECK-SAME: i32 [[COND:%.*]]) {
+; CHECK-NEXT: [[SWITCH_GEP:%.*]] = getelementptr inbounds [3 x ptr], ptr @switch.table, i32 0, i32 [[COND]]
+; CHECK-NEXT: [[RELTABLE_INTRINSIC:%.*]] = load ptr, ptr [[SWITCH_GEP]], align 8
+; CHECK-NEXT: ret ptr [[RELTABLE_INTRINSIC]]
+;
+ %switch.gep = getelementptr inbounds [3 x ptr], ptr @switch.table, i32 0, i32 %cond
+ %switch.load = load ptr, ptr %switch.gep, align 8
+ ret ptr %switch.load
+}
|
github-project-automation
bot
moved this from Needs Review
to Needs Merge
in LLVM Release Status
Relative lookup tables result in "LLVM ERROR: Circular dependency found in global variable set", so disable them for this target. (cherry picked from commit 1ed1537)
github-project-automation
bot
moved this from Needs Merge
to Done
in LLVM Release Status
|
@nikic (or anyone else). If you would like to add a note about this fix in the release notes (completely optional). Please reply to this comment with a one or two sentence description of the fix. When you are done, please add the release:note label to this PR. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
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.
Backport 1ed1537
Requested by: @nikic