[Mlir-commits] [mlir] [MLIR][DLTI] Introduce DLTIQueryInterface and impl for DLTI attrs (PR #104595)

[Rolf Morel]

================
@@ -93,20 +85,63 @@ def DLTI_DataLayoutSpecAttr :
 
     /// Returns the stack alignment identifier.
     StringAttr getStackAlignmentIdentifier(MLIRContext *context) const;
+
+    /// Returns the attribute associated with the key.
+    FailureOr<Attribute> query(DataLayoutEntryKey key) {
+      return llvm::cast<mlir::DataLayoutSpecInterface>(*this).queryHelper(key);
+    }
+  }];
+}
+
+def DLTI_MapAttr : DLTIAttr<"Map", [DLTIQueryInterface]> {
+  let summary = "A mapping of DLTI-information by way of key-value pairs";
+  let description = [{
+    A Data Layout and Target Information map is a list of entries effectively
+    encoding a dictionary, mapping DLTI-related keys to DLTI-related values.
+
+    This attribute's main purpose is to facilate querying IR for arbitrary
+    DLTI-related key-value associations. Note that facility functions exist to
+    perform recursive lookups on nested DLTI map attributes.
+
+    Consider the following shallow usage of a DLTI-map
+    ```
+    #dlti.map<#dlti.dl_entry<"CPU::cache::L1::size_in_bytes", 65536 : i32>>
+    ```
+    versus nested maps, which make it possible to obtain sub-dictionaries of
+    related information (with the following example making use of other
+    attributes that also implement the `DLTIQueryInterface`):
+    ```
+    #dlti.target_system_spec<"CPU":
+      #dlti.target_device_spec<#dlti.dl_entry<"cache",
+        #dlti.map<#dlti.dl_entry<"L1",
+          #dlti.map<#dlti.dl_entry<"size_in_bytes", 65536 : i32>>,
+                  #dlti.dl_entry<"L1d",
+          #dlti.map<#dlti.dl_entry<"size_in_bytes", 32768 : i32>> >>>>
----------------
rolfmorel wrote:

I took it from one for the (passing) test cases: https://github.com/llvm/llvm-project/blob/83b5db6429c68afa073a65d173142d2e56360b3e/mlir/test/Dialect/DLTI/query.mlir#L194

The only difference is that I have inlined the `"size_in_bytes"` aliases.

To aid human parsing of the above, the indentation levels correspond to the level of nesting of the dictionaries and each line contains exactly one key. If this is not helpful (enough) or you have another suggestion, do let me know.

https://github.com/llvm/llvm-project/pull/104595