[flang-commits] [flang] [Flang][OpenMP] Align map clause generation and fix issue with non-shared allocations for assumed shape/size descriptor types (PR #97855)

[via flang-commits]

================
@@ -970,25 +970,21 @@ bool ClauseProcessor::processMap(
                   object.ref(), clauseLocation, asFortran, bounds,
                   treatIndexAsSection);
 
-          auto origSymbol = converter.getSymbolAddress(*object.sym());
-          mlir::Value symAddr = info.addr;
-          if (origSymbol && fir::isTypeWithDescriptor(origSymbol.getType()))
-            symAddr = origSymbol;
-
           // Explicit map captures are captured ByRef by default,
           // optimisation passes may alter this to ByCopy or other capture
           // types to optimise
+          mlir::Value baseOp = info.rawInput;
----------------
agozillon wrote:

Thank you very much @tblah, at least for the moment I don't believe we need the extra information (and testing against the little local map test suite I have made it seems to function as correctly as it did previously), as we tend to package the bounds as a seperate input to Map Info and at least for the moment that's all we care about, it could change in the future but we can likely cross that bridge if/when we come to a situation that requires it. 

https://github.com/llvm/llvm-project/pull/97855