[memprof] Remove packed qualifier for MemprofRecord::Frame.

[snehasish]

Summary

Now that we use dedicated serialize and deserialize methods in order to
ensure consistency across big and small endian systems. The packed
qualifier on the Frame struct can be removed.

[davidxl]

lgtm

[tejohnson]

What about the other uses of PACKED e.g. in MemProfData.inc? Should/can they be removed?

[snehasish]

In D120147#3332688, @tejohnson wrote:

What about the other uses of PACKED e.g. in MemProfData.inc? Should/can they be removed?

This use of packed is for reading and writing of the indexed profile format. The packed structs in MemProfData.inc are written by the runtime and read by the raw memprof profile reader. The use of packed structs makes the compiler generate alignment aware code, simplifying the implementation in the runtime. Here, within llvm, we have access to support::endian so implementing endian and alignment aware serialization/deserialization is simple. We would have to replicate this in the memprof runtime.

[tejohnson]

In D120147#3332865, @snehasish wrote:

In D120147#3332688, @tejohnson wrote:

What about the other uses of PACKED e.g. in MemProfData.inc? Should/can they be removed?

This use of packed is for reading and writing of the indexed profile format. The packed structs in MemProfData.inc are written by the runtime and read by the raw memprof profile reader. The use of packed structs makes the compiler generate alignment aware code, simplifying the implementation in the runtime. Here, within llvm, we have access to support::endian so implementing endian and alignment aware serialization/deserialization is simple. We would have to replicate this in the memprof runtime.

Oh ok, that makes sense.

lgtm

llvm/include/llvm/ProfileData/MemProf.h

 #include "llvm/ProfileData/MIBEntryDef.inc"
 #undef MIBEntryDef
 };
 
 struct MemProfRecord {
   // Describes a call frame for a dynamic allocation context. The contents of
   // the frame are populated by symbolizing the stack depot call frame from the
   // compiler runtime.
-  PACKED(struct Frame {
+  struct Frame {
     // A uuid (uint64_t) identifying the function. It is obtained by
     // llvm::md5(FunctionName) which returns the lower 64 bits.
     GlobalValue::GUID Function;
     // The source line offset of the call from the beginning of parent function.
     uint32_t LineOffset;
     // The source column number of the call to help distinguish multiple calls
     // on the same line.
     uint32_t Column;
static Frame deserialize(const unsigned char *Ptr) {
                    /*IsInlineFrame=*/I);
     }
 
     // Returns the size of the frame information.
     static constexpr size_t serializedSize() {
       return sizeof(Frame::Function) + sizeof(Frame::LineOffset) +
              sizeof(Frame::Column) + sizeof(Frame::IsInlineFrame);
     }
-  });
+  };
 
   // The dynamic calling context for the allocation.
   std::vector<Frame> CallStack;
   // The statistics obtained from the runtime for the allocation.
   PortableMemInfoBlock Info;
 
   void clear() {
     CallStack.clear();
     Info.clear();
   }
 
   size_t serializedSize() const {
     return sizeof(uint64_t) + // The number of frames to serialize.
-           sizeof(Frame) * CallStack.size() + // The contents of the frames.
+           Frame::serializedSize() *
+               CallStack.size() + // The contents of the frames.
            PortableMemInfoBlock::serializedSize(); // The size of the payload.
   }
 
   // Prints out the contents of the memprof record in YAML.
   void print(llvm::raw_ostream &OS) const {
     OS << "    Callstack:\n";
     // TODO: Print out the frame on one line with to make it easier for deep
     // callstacks once we have a test to check valid YAML is generated.