[emitc] Fix the translation switchop with argument of expressionop #123701
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
jacquesguan
requested review from
marbre,
simon-camp,
aniragil and
mgehre-amd
|
@llvm/pr-subscribers-mlir @llvm/pr-subscribers-mlir-emitc Author: Jianjian Guan (jacquesguan) ChangesFull diff: https://github.com/llvm/llvm-project/pull/123701.diff 2 Files Affected:
diff --git a/mlir/lib/Target/Cpp/TranslateToCpp.cpp b/mlir/lib/Target/Cpp/TranslateToCpp.cpp
index a91f5ab9311401..01de0e41f20353 100644
--- a/mlir/lib/Target/Cpp/TranslateToCpp.cpp
+++ b/mlir/lib/Target/Cpp/TranslateToCpp.cpp
@@ -475,7 +475,10 @@ static LogicalResult printOperation(CppEmitter &emitter,
emitc::SwitchOp switchOp) {
raw_indented_ostream &os = emitter.ostream();
- os << "\nswitch (" << emitter.getOrCreateName(switchOp.getArg()) << ") {";
+ os << "\nswitch (";
+ if (failed(emitter.emitOperand(switchOp.getArg())))
+ return failure();
+ os << ") {";
for (auto pair : llvm::zip(switchOp.getCases(), switchOp.getCaseRegions())) {
os << "\ncase " << std::get<0>(pair) << ": {\n";
diff --git a/mlir/test/Target/Cpp/switch.mlir b/mlir/test/Target/Cpp/switch.mlir
index 1a8f5e2dfd2b61..222562b5c57e2b 100644
--- a/mlir/test/Target/Cpp/switch.mlir
+++ b/mlir/test/Target/Cpp/switch.mlir
@@ -882,3 +882,77 @@ func.func @emitc_switch_ui64() {
}
return
}
+
+// CPP-DEFAULT-LABEL: void emitc_switch_expression() {
+// CPP-DEFAULT: int64_t v1 = 42;
+// CPP-DEFAULT: int64_t v2 = 24;
+// CPP-DEFAULT: switch ((v1 + v2) * v2) {
+// CPP-DEFAULT: case 2: {
+// CPP-DEFAULT: int32_t v3 = func_b();
+// CPP-DEFAULT: break;
+// CPP-DEFAULT: }
+// CPP-DEFAULT: case 5: {
+// CPP-DEFAULT: int32_t v4 = func_a();
+// CPP-DEFAULT: break;
+// CPP-DEFAULT: }
+// CPP-DEFAULT: default: {
+// CPP-DEFAULT: float v5 = 4.200000000e+01f;
+// CPP-DEFAULT: func2(v5);
+// CPP-DEFAULT: break;
+// CPP-DEFAULT: }
+// CPP-DEFAULT: }
+// CPP-DEFAULT: return;
+// CPP-DEFAULT: }
+
+// CPP-DECLTOP-LABEL: void emitc_switch_expression() {
+// CPP-DECLTOP: int64_t v1;
+// CPP-DECLTOP: int64_t v2;
+// CPP-DECLTOP: float v3;
+// CPP-DECLTOP: int32_t v4;
+// CPP-DECLTOP: int32_t v5;
+// CPP-DECLTOP: v1 = 42;
+// CPP-DECLTOP: v2 = 24;
+// CPP-DECLTOP: switch ((v1 + v2) * v2) {
+// CPP-DECLTOP: case 2: {
+// CPP-DECLTOP: v4 = func_b();
+// CPP-DECLTOP: break;
+// CPP-DECLTOP: }
+// CPP-DECLTOP: case 5: {
+// CPP-DECLTOP: v5 = func_a();
+// CPP-DECLTOP: break;
+// CPP-DECLTOP: }
+// CPP-DECLTOP: default: {
+// CPP-DECLTOP: v3 = 4.200000000e+01f;
+// CPP-DECLTOP: func2(v3);
+// CPP-DECLTOP: break;
+// CPP-DECLTOP: }
+// CPP-DECLTOP: }
+// CPP-DECLTOP: return;
+// CPP-DECLTOP: }
+
+func.func @emitc_switch_expression() {
+ %x = "emitc.constant"(){value = 42 : i64} : () -> i64
+ %y = "emitc.constant"(){value = 24 : i64} : () -> i64
+
+ %0 = emitc.expression : i64 {
+ %a = emitc.add %x, %y : (i64, i64) -> i64
+ %b = emitc.mul %a, %y : (i64, i64) -> i64
+ emitc.yield %b : i64
+ }
+
+ emitc.switch %0 : i64
+ case 2 {
+ %1 = emitc.call_opaque "func_b" () : () -> i32
+ emitc.yield
+ }
+ case 5 {
+ %2 = emitc.call_opaque "func_a" () : () -> i32
+ emitc.yield
+ }
+ default {
+ %3 = "emitc.constant"(){value = 42.0 : f32} : () -> f32
+ emitc.call_opaque "func2" (%3) : (f32) -> ()
+ emitc.yield
+ }
+ return
+}
|
mgehre-amd
reviewed
Jan 21, 2025
|
Thanks for the patch! |
jacquesguan
changed the title
Thanks for comment, I retitle this pr and add the description to mark it as a bugfix. |
mgehre-amd
approved these changes
Jan 22, 2025
Thanks, I have it, and merge this pr right now. |
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/94/builds/3992 Here is the relevant piece of the build log for the reference |
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.
Now a
emitc.switchwith argument ofemitc.expressionwouldn't emit its argument to cpp. This patch fix it.