[NVPTX] Add Read/Write/SideEffect attributes to atomic instructions

[LewisCrawford]

Set the mayLoad, mayStore, and hasSideEffects hints for NVPTX atomic instructions. This prevents any optimizations (e.g. rematerialization) from illegally duplicating them and generating broken code.

[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-backend-nvptx

Author: Lewis Crawford (LewisCrawford)

Changes

Set the mayLoad, mayStore, and hasSideEffects hints for NVPTX atomic instructions. This prevents any optimizations (e.g. rematerialization) from illegally duplicating them and generating broken code.

---

Full diff: https://github.com/llvm/llvm-project/pull/109665.diff

1 Files Affected:

• (modified) llvm/lib/Target/NVPTX/NVPTXIntrinsics.td (+46-40)

diff --git a/llvm/lib/Target/NVPTX/NVPTXIntrinsics.td b/llvm/lib/Target/NVPTX/NVPTXIntrinsics.td
index 656fc679a572aa..9b515e6b444c8b 100644
--- a/llvm/lib/Target/NVPTX/NVPTXIntrinsics.td
+++ b/llvm/lib/Target/NVPTX/NVPTXIntrinsics.td
@@ -1633,14 +1633,16 @@ multiclass F_ATOMIC_2_imp<ValueType ptrT, NVPTXRegClass ptrclass,
   ValueType regT, NVPTXRegClass regclass,
   string SpaceStr, string TypeStr, string OpcStr, PatFrag IntOp,
   Operand IMMType, SDNode IMM, list<Predicate> Pred> {
-  def reg : NVPTXInst<(outs regclass:$dst), (ins ptrclass:$addr, regclass:$b),
-    !strconcat("atom", SpaceStr, OpcStr, TypeStr, " \t$dst, [$addr], $b;"),
-    [(set (regT regclass:$dst), (IntOp (ptrT ptrclass:$addr), (regT regclass:$b)))]>,
-  Requires<Pred>;
-  def imm : NVPTXInst<(outs regclass:$dst), (ins ptrclass:$addr, IMMType:$b),
-    !strconcat("atom", SpaceStr, OpcStr, TypeStr, " \t$dst, [$addr], $b;", ""),
-    [(set (regT regclass:$dst), (IntOp (ptrT ptrclass:$addr), IMM:$b))]>,
-  Requires<!if(!or(!eq(TypeStr, ".f16"), !eq(TypeStr, ".bf16")), [Predicate<"false">], Pred)>;
+  let mayLoad = 1, mayStore = 1, hasSideEffects = 1 in {
+    def reg : NVPTXInst<(outs regclass:$dst), (ins ptrclass:$addr, regclass:$b),
+      !strconcat("atom", SpaceStr, OpcStr, TypeStr, " \t$dst, [$addr], $b;"),
+      [(set (regT regclass:$dst), (IntOp (ptrT ptrclass:$addr), (regT regclass:$b)))]>,
+    Requires<Pred>;
+    def imm : NVPTXInst<(outs regclass:$dst), (ins ptrclass:$addr, IMMType:$b),
+      !strconcat("atom", SpaceStr, OpcStr, TypeStr, " \t$dst, [$addr], $b;", ""),
+      [(set (regT regclass:$dst), (IntOp (ptrT ptrclass:$addr), IMM:$b))]>,
+    Requires<!if(!or(!eq(TypeStr, ".f16"), !eq(TypeStr, ".bf16")), [Predicate<"false">], Pred)>;
+  }
 }
 multiclass F_ATOMIC_2<ValueType regT, NVPTXRegClass regclass, string SpaceStr, string TypeStr,
   string OpcStr, PatFrag IntOp, Operand IMMType, SDNode IMM,
@@ -1656,15 +1658,17 @@ multiclass F_ATOMIC_2_NEG_imp<ValueType ptrT, NVPTXRegClass ptrclass,
   ValueType regT, NVPTXRegClass regclass,
   string SpaceStr, string TypeStr, string OpcStr, PatFrag IntOp,
   list<Predicate> Pred> {
-  def reg : NVPTXInst<(outs regclass:$dst), (ins ptrclass:$addr, regclass:$b),
-    !strconcat(
-      "{{ \n\t",
-      ".reg \t.s", TypeStr, " temp; \n\t",
-      "neg.s", TypeStr, " \ttemp, $b; \n\t",
-      "atom", SpaceStr, OpcStr, ".u", TypeStr, " \t$dst, [$addr], temp; \n\t",
-      "}}"),
-    [(set (regT regclass:$dst), (IntOp (ptrT ptrclass:$addr), (regT regclass:$b)))]>,
-  Requires<Pred>;
+  let mayLoad = 1, mayStore = 1, hasSideEffects = 1 in {
+    def reg : NVPTXInst<(outs regclass:$dst), (ins ptrclass:$addr, regclass:$b),
+      !strconcat(
+        "{{ \n\t",
+        ".reg \t.s", TypeStr, " temp; \n\t",
+        "neg.s", TypeStr, " \ttemp, $b; \n\t",
+        "atom", SpaceStr, OpcStr, ".u", TypeStr, " \t$dst, [$addr], temp; \n\t",
+        "}}"),
+      [(set (regT regclass:$dst), (IntOp (ptrT ptrclass:$addr), (regT regclass:$b)))]>,
+    Requires<Pred>;
+  }
 }
 multiclass F_ATOMIC_2_NEG<ValueType regT, NVPTXRegClass regclass, string SpaceStr,
   string TypeStr, string OpcStr, PatFrag IntOp, list<Predicate> Pred = []> {
@@ -1679,29 +1683,31 @@ multiclass F_ATOMIC_3_imp<ValueType ptrT, NVPTXRegClass ptrclass,
   ValueType regT, NVPTXRegClass regclass,
   string SpaceStr, string TypeStr, string OpcStr, PatFrag IntOp,
   Operand IMMType, list<Predicate> Pred> {
-  def reg : NVPTXInst<(outs regclass:$dst),
-    (ins ptrclass:$addr, regclass:$b, regclass:$c),
-    !strconcat("atom", SpaceStr, OpcStr, TypeStr, " \t$dst, [$addr], $b, $c;"),
-    [(set (regT regclass:$dst), (IntOp (ptrT ptrclass:$addr), (regT regclass:$b), (regT regclass:$c)))]>,
-  Requires<Pred>;
-
-  def imm1 : NVPTXInst<(outs regclass:$dst),
-    (ins ptrclass:$addr, IMMType:$b, regclass:$c),
-    !strconcat("atom", SpaceStr, OpcStr, TypeStr, " \t$dst, [$addr], $b, $c;"),
-    [(set (regT regclass:$dst), (IntOp (ptrT ptrclass:$addr), imm:$b, (regT regclass:$c)))]>,
-  Requires<Pred>;
-
-  def imm2 : NVPTXInst<(outs regclass:$dst),
-    (ins ptrclass:$addr, regclass:$b, IMMType:$c),
-    !strconcat("atom", SpaceStr, OpcStr, TypeStr, " \t$dst, [$addr], $b, $c;", ""),
-    [(set (regT regclass:$dst), (IntOp (ptrT ptrclass:$addr), (regT regclass:$b), imm:$c))]>,
-  Requires<Pred>;
-
-  def imm3 : NVPTXInst<(outs regclass:$dst),
-    (ins ptrclass:$addr, IMMType:$b, IMMType:$c),
-    !strconcat("atom", SpaceStr, OpcStr, TypeStr, " \t$dst, [$addr], $b, $c;"),
-    [(set (regT regclass:$dst), (IntOp (ptrT ptrclass:$addr), imm:$b, imm:$c))]>,
-  Requires<Pred>;
+  let mayLoad = 1, mayStore = 1, hasSideEffects = 1 in {
+    def reg : NVPTXInst<(outs regclass:$dst),
+      (ins ptrclass:$addr, regclass:$b, regclass:$c),
+      !strconcat("atom", SpaceStr, OpcStr, TypeStr, " \t$dst, [$addr], $b, $c;"),
+      [(set (regT regclass:$dst), (IntOp (ptrT ptrclass:$addr), (regT regclass:$b), (regT regclass:$c)))]>,
+    Requires<Pred>;
+
+    def imm1 : NVPTXInst<(outs regclass:$dst),
+      (ins ptrclass:$addr, IMMType:$b, regclass:$c),
+      !strconcat("atom", SpaceStr, OpcStr, TypeStr, " \t$dst, [$addr], $b, $c;"),
+      [(set (regT regclass:$dst), (IntOp (ptrT ptrclass:$addr), imm:$b, (regT regclass:$c)))]>,
+    Requires<Pred>;
+
+    def imm2 : NVPTXInst<(outs regclass:$dst),
+      (ins ptrclass:$addr, regclass:$b, IMMType:$c),
+      !strconcat("atom", SpaceStr, OpcStr, TypeStr, " \t$dst, [$addr], $b, $c;", ""),
+      [(set (regT regclass:$dst), (IntOp (ptrT ptrclass:$addr), (regT regclass:$b), imm:$c))]>,
+    Requires<Pred>;
+
+    def imm3 : NVPTXInst<(outs regclass:$dst),
+      (ins ptrclass:$addr, IMMType:$b, IMMType:$c),
+      !strconcat("atom", SpaceStr, OpcStr, TypeStr, " \t$dst, [$addr], $b, $c;"),
+      [(set (regT regclass:$dst), (IntOp (ptrT ptrclass:$addr), imm:$b, imm:$c))]>,
+    Requires<Pred>;
+  }
 }
 multiclass F_ATOMIC_3<ValueType regT, NVPTXRegClass regclass, string SpaceStr, string TypeStr,
   string OpcStr, PatFrag IntOp, Operand IMMType, list<Predicate> Pred = []> {

[Artem-B]

LGTM overall.

Do you have any specific examples demonstrating incorrect behavior we have now?
It may be useful to add a test case for that.

[LewisCrawford]

Sorry, I don't have a good isolated test-case for this unfortunately.

The only example I've seen of this triggering a bug relied on several downstream compiler changes, and I've been struggling to create a reproducer that also fails on this upstream version.

I was hoping this change was trivial enough that it would be ok to commit without an explicit unit test.

[github-actions]

@LewisCrawford 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!