[PATCH] D93659: [AIX][XCOFF] emit vector info of traceback table.

[Digger Lin via Phabricator via llvm-commits]

DiggerLin marked 17 inline comments as done.
DiggerLin added inline comments.


================
Comment at: llvm/lib/Object/XCOFFObjectFile.cpp:998
+    }
+    ParmsType = ParmsTypeOrError.get();
   }
----------------
jasonliu wrote:
> Is llvm-objdump capable of decoding the traceback table? Do we want to have a test case using llvm-objdump? Or are we thinking to put it on another patch?
we have patch which can decode traceback table of object file. https://reviews.llvm.org/D89049


================
Comment at: llvm/lib/Target/PowerPC/PPCMachineFunctionInfo.cpp:138-141
+    if (Elt == FixedType && !hasVectorParms() && Bits > 31)
+      break;
+    if ((Elt != FixedType || hasVectorParms()) && Bits > 30)
+      break;
----------------
jasonliu wrote:
> Could we combine the above two if statement?
can not combine it, for Elt == FixedType && !hasVectorParms()  , it only need one bit, so when Bits ==31, we can still put one bit of FixedType which do not have hasVectorParms().


================
Comment at: llvm/lib/Target/PowerPC/PPCMachineFunctionInfo.h:135
+  /// ParamTypes - Store all the paramters type.
+  SmallVector<ParamType, 32> ParamtersType;
 
----------------
jasonliu wrote:
> Does this store the parameter types when the parameter is on stack? Right now, it doesn't look like so. And this sounds like we are trying to store all paramters including the one on stack.
> Could we add a test case for testing parameters on stack?
we do not store the  parameter types when the parameter is on stack.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D93659/new/

https://reviews.llvm.org/D93659