[TableGen] New tblgen Pattern bit to disable DAG pattern imports during GISel

[jofrn]

Added a new DAGISelShouldIgnore property to class Instruction in Target.td, similar to FastISelShouldIgnore. This allows one to avoid a record's DAGISel .td implementation and .inc generation.

[github-actions]

Thank you for submitting a Pull Request (PR) to the LLVM Project!

This PR will be automatically labeled and the relevant teams will be
notified.

If you wish to, you can add reviewers by using the "Reviewers" section on this page.

If this is not working for you, it is probably because you do not have write
permissions for the repository. In which case you can instead tag reviewers by
name in a comment by using @ followed by their GitHub username.

If you have received no comments on your PR for a week, you can request a review
by "ping"ing the PR by adding a comment “Ping”. The common courtesy "ping" rate
is once a week. Please remember that you are asking for valuable time from other developers.

If you have further questions, they may be answered by the LLVM GitHub User Guide.

You can also ask questions in a comment on this PR, on the LLVM Discord or on the forums.

[llvmbot]

@llvm/pr-subscribers-llvm-globalisel

@llvm/pr-subscribers-llvm-selectiondag

Author: None (jofrn)

Changes

Added a new DAGISelShouldIgnore property to class Instruction in Target.td, similar to FastISelShouldIgnore. This allows one to avoid a record's DAGISel .td implementation and .inc generation.

---

Full diff: https://github.com/llvm/llvm-project/pull/88382.diff

4 Files Affected:

• (modified) llvm/include/llvm/Target/Target.td (+4)
• (modified) llvm/utils/TableGen/Common/CodeGenInstruction.cpp (+1)
• (modified) llvm/utils/TableGen/Common/CodeGenInstruction.h (+1)
• (modified) llvm/utils/TableGen/DAGISelEmitter.cpp (+14-1)

diff --git a/llvm/include/llvm/Target/Target.td b/llvm/include/llvm/Target/Target.td
index 1f7dc6922f13e4..61dc04595b62b9 100644
--- a/llvm/include/llvm/Target/Target.td
+++ b/llvm/include/llvm/Target/Target.td
@@ -777,6 +777,10 @@ class Instruction : InstructionEncoding {
   /// SelectionDAG can.
   bit FastISelShouldIgnore = false;
 
+  /// Should DAGISel ignore this instruction. In cases where lowering may
+  /// be done elsewhere or is unneeded, DAGISel may skip over them.
+  bit DAGISelShouldIgnore = false;
+
   /// HasPositionOrder: Indicate tablegen to sort the instructions by record
   /// ID, so that instruction that is defined earlier can be sorted earlier
   /// in the assembly matching table.
diff --git a/llvm/utils/TableGen/Common/CodeGenInstruction.cpp b/llvm/utils/TableGen/Common/CodeGenInstruction.cpp
index 18a4e7b0f18b23..a76efb76bfc319 100644
--- a/llvm/utils/TableGen/Common/CodeGenInstruction.cpp
+++ b/llvm/utils/TableGen/Common/CodeGenInstruction.cpp
@@ -465,6 +465,7 @@ CodeGenInstruction::CodeGenInstruction(Record *R)
   isConvergent = R->getValueAsBit("isConvergent");
   hasNoSchedulingInfo = R->getValueAsBit("hasNoSchedulingInfo");
   FastISelShouldIgnore = R->getValueAsBit("FastISelShouldIgnore");
+  DAGISelShouldIgnore = R->getValueAsBit("DAGISelShouldIgnore");
   variadicOpsAreDefs = R->getValueAsBit("variadicOpsAreDefs");
   isAuthenticated = R->getValueAsBit("isAuthenticated");
 
diff --git a/llvm/utils/TableGen/Common/CodeGenInstruction.h b/llvm/utils/TableGen/Common/CodeGenInstruction.h
index b658259b4892ee..c36bfba245f66b 100644
--- a/llvm/utils/TableGen/Common/CodeGenInstruction.h
+++ b/llvm/utils/TableGen/Common/CodeGenInstruction.h
@@ -282,6 +282,7 @@ class CodeGenInstruction {
   bool isConvergent : 1;
   bool hasNoSchedulingInfo : 1;
   bool FastISelShouldIgnore : 1;
+  bool DAGISelShouldIgnore : 1;
   bool hasChain : 1;
   bool hasChain_Inferred : 1;
   bool variadicOpsAreDefs : 1;
diff --git a/llvm/utils/TableGen/DAGISelEmitter.cpp b/llvm/utils/TableGen/DAGISelEmitter.cpp
index b43a8e659dd9c5..4fbeafdc2355fb 100644
--- a/llvm/utils/TableGen/DAGISelEmitter.cpp
+++ b/llvm/utils/TableGen/DAGISelEmitter.cpp
@@ -165,8 +165,21 @@ void DAGISelEmitter::run(raw_ostream &OS) {
   // Add all the patterns to a temporary list so we can sort them.
   Records.startTimer("Sort patterns");
   std::vector<const PatternToMatch *> Patterns;
-  for (const PatternToMatch &PTM : CGP.ptms())
+  for (const PatternToMatch &PTM : CGP.ptms()) {
+
+    // Disable import of patterns marked as ignore.
+    const TreePatternNode &Dst = PTM.getDstPattern();
+    if (!Dst.isLeaf()) {
+      const Record *Op = Dst.getOperator();
+      const bool shouldIgnore =
+          Op->isSubClassOf("Instruction") &&
+          CGP.getTargetInfo().getInstruction(Op).DAGISelShouldIgnore;
+      if (shouldIgnore)
+        continue;
+    }
+
     Patterns.push_back(&PTM);
+  }
 
   // We want to process the matches in order of minimal cost.  Sort the patterns
   // so the least cost one is at the start.

[shiltian]

could you add a test?

[arsenm]

Should add a test that a warning isn't produced on the use

[github-actions]

✅ With the latest revision this PR passed the C/C++ code formatter.

[shiltian]

LGTM with one nit. I'd wait to see if @arsenm and @Pierre-vh have more comments.

[shiltian]

alignment

[jofrn]

It is aligned around the '=', rather than shifting everyone over.

[Pierre-vh]

One nit but I also have a concern with how the bit is handled

[github-actions]

@jofrn Congratulations on having your first Pull Request (PR) merged into the LLVM Project!

Your changes will be combined with recent changes from other authors, then tested
by our build bots. If there is a problem with a build, you may receive a report in an email or a comment on this PR.

Please check whether problems have been caused by your change specifically, as
the builds can include changes from many authors. It is not uncommon for your
change to be included in a build that fails due to someone else's changes, or
infrastructure issues.

How to do this, and the rest of the post-merge process, is covered in detail here.

If your change does cause a problem, it may be reverted, or you can revert it yourself.
This is a normal part of LLVM development. You can fix your changes and open a new PR to merge them again.

If you don't get any reports, no action is required from you. Your changes are working as expected, well done!