[llvm] [NVPTX] Add family-specific architectures support (PR #141899)

[via llvm-commits]

llvmbot wrote:


<!--LLVM PR SUMMARY COMMENT-->

@llvm/pr-subscribers-backend-nvptx

Author: Rajat Bajpai (rajatbajpai)

<details>
<summary>Changes</summary>

This change adds family-specific architectures support added in [PTX ISA 8.8](https://docs.nvidia.com/cuda/parallel-thread-execution/#ptx-isa-version-8-8). These architectures have "f" suffix. For example, sm_100f.

This change doesn't promote existing features to family-specific architecture.

---
Full diff: https://github.com/llvm/llvm-project/pull/141899.diff


2 Files Affected:

- (modified) llvm/lib/Target/NVPTX/NVPTX.td (+13-6) 
- (modified) llvm/lib/Target/NVPTX/NVPTXSubtarget.h (+23-4) 


``````````diff
diff --git a/llvm/lib/Target/NVPTX/NVPTX.td b/llvm/lib/Target/NVPTX/NVPTX.td
index ff9a187ecf723..3ed2553fa4232 100644
--- a/llvm/lib/Target/NVPTX/NVPTX.td
+++ b/llvm/lib/Target/NVPTX/NVPTX.td
@@ -41,12 +41,14 @@ foreach sm = [20, 21, 30, 32, 35, 37, 50, 52, 53,
 
 // Arch-specific targets. PTX for these is not compatible with any other
 // architectures.
-def SM90a : FeatureSM<"90a", 901>;
-def SM100a: FeatureSM<"100a", 1001>;
-def SM101a: FeatureSM<"101a", 1011>;
-def SM103a: FeatureSM<"103a", 1031>;
-def SM120a: FeatureSM<"120a", 1201>;
-def SM121a: FeatureSM<"121a", 1211>;
+foreach sm = [90, 100, 101, 103, 120, 121] in {
+  def SM#sm#a : FeatureSM<""#sm#"a", !add(!mul(sm, 10), 1)>;
+}
+
+// Family-specific targets. PTX for these is compatible within the same family.
+foreach sm = [100, 101, 103, 120, 121] in {
+  def SM#sm#f : FeatureSM<""#sm#"f", !add(!mul(sm, 10), 2)>;
+}
 
 foreach version = [32, 40, 41, 42, 43, 50, 60, 61, 62, 63, 64, 65,
                    70, 71, 72, 73, 74, 75, 76, 77, 78,
@@ -83,14 +85,19 @@ def : Proc<"sm_90",   [SM90, PTX78]>;
 def : Proc<"sm_90a",  [SM90a, PTX80]>;
 def : Proc<"sm_100",  [SM100, PTX86]>;
 def : Proc<"sm_100a", [SM100a, PTX86]>;
+def : Proc<"sm_100f", [SM100f, PTX88]>;
 def : Proc<"sm_101",  [SM101, PTX86]>;
 def : Proc<"sm_101a", [SM101a, PTX86]>;
+def : Proc<"sm_101f", [SM101f, PTX88]>;
 def : Proc<"sm_103",  [SM103, PTX88]>;
 def : Proc<"sm_103a", [SM103a, PTX88]>;
+def : Proc<"sm_103f", [SM103f, PTX88]>;
 def : Proc<"sm_120",  [SM120, PTX87]>;
 def : Proc<"sm_120a", [SM120a, PTX87]>;
+def : Proc<"sm_120f", [SM120f, PTX88]>;
 def : Proc<"sm_121",  [SM121, PTX88]>;
 def : Proc<"sm_121a", [SM121a, PTX88]>;
+def : Proc<"sm_121f", [SM121f, PTX88]>;
 
 def NVPTXInstrInfo : InstrInfo {
 }
diff --git a/llvm/lib/Target/NVPTX/NVPTXSubtarget.h b/llvm/lib/Target/NVPTX/NVPTXSubtarget.h
index 5136b1ee28502..5e4ab9476cb31 100644
--- a/llvm/lib/Target/NVPTX/NVPTXSubtarget.h
+++ b/llvm/lib/Target/NVPTX/NVPTXSubtarget.h
@@ -132,10 +132,29 @@ class NVPTXSubtarget : public NVPTXGenSubtargetInfo {
   // are supported on the specified architecture only, hence such targets do not
   // follow the onion layer model. hasArchAccelFeatures() allows
   // distinguishing such GPU variants from the base GPU architecture.
-  // - 0 represents base GPU model,
-  // - non-zero value identifies particular architecture-accelerated variant.
-  bool hasArchAccelFeatures() const { return getFullSmVersion() % 10; }
-
+  // - false represents non-accelerated architecture.
+  // - true represents architecture-accelerated variant.
+  bool hasArchAccelFeatures() const {
+    auto FullSMVersionMod = getFullSmVersion() % 10;
+    assert(FullSMVersionMod < 3 && "Invalid architecture!");
+    return FullSMVersionMod == 1;
+  }
+  // GPUs with 'f' suffix have architecture-accelerated features which are
+  // portable across all future architectures under same SM major. For example,
+  // sm_100f features will work for sm_10X future architectures.
+  // - false represents non-family-specific architecture.
+  // - true represents family-specific variant.
+  bool hasFamilySpecificFeatures() const {
+    auto FullSMVersionMod = getFullSmVersion() % 10;
+    assert(FullSMVersionMod < 3 && "Invalid architecture!");
+    return FullSMVersionMod == 2 && PTXVersion >= 88;
+  }
+  // Checks if architecture is accelerated or family-specific.
+  // - false represents neither arch-accelerated nor family-specific arch.
+  // - true represents either arch-accelerated or family-specific arch.
+  bool hasArchAccelOrFamilySpecificFeatures() const {
+    return hasArchAccelFeatures() || hasFamilySpecificFeatures();
+  }
   // If the user did not provide a target we default to the `sm_30` target.
   std::string getTargetName() const {
     return TargetName.empty() ? "sm_30" : TargetName;

``````````

</details>


https://github.com/llvm/llvm-project/pull/141899