[LLDB][NFC] Fix a incorrect use of shared_ptr in RenderScriptRuntime.cpp

[xgupta]

Summary

Incorrect use of shared_ptr.

found by PVS-Studio https://pvs-studio.com/en/blog/posts/cpp/1003/, N8 & N9.

[jingham]

I don't think this code is going to compile, is it? You took out the if (!buffer) { but left the RHS } so now the scopes are unbalanced. I think you should just change how the shared pointer is defined - which is presumably the substance of the patch - and leave everything after as is.

[xgupta]

address comments

[xgupta]

In D142309#4075616, @jingham wrote:

I don't think this code is going to compile, is it? You took out the if (!buffer) { but left the RHS } so now the scopes are unbalanced. I think you should just change how the shared pointer is defined - which is presumably the substance of the patch - and leave everything after as is.

Yeah, correct, I didn't notice carefully. Thanks.

[JDevlieghere]

LGTM, but I wonder when we'll be able to get rid of RenderScript completely (CC @labath)

[xgupta]

fix build

[xgupta]

In D142309#4075745, @JDevlieghere wrote:

LGTM, but I wonder when we'll be able to get rid of RenderScript completely (CC @labath)

@JDevlieghere, Shall I commit the changes after running check-lldb?

[labath]

In D142309#4075745, @JDevlieghere wrote:

LGTM, but I wonder when we'll be able to get rid of RenderScript completely (CC @labath)

I am not going to stand in your way. I usually redirect these questions to @srhines, but I understand he's moved on to other things as well.

[xgupta]

In D142309#4102284, @labath wrote:

In D142309#4075745, @JDevlieghere wrote:

LGTM, but I wonder when we'll be able to get rid of RenderScript completely (CC @labath)

I am not going to stand in your way. I usually redirect these questions to @srhines, but I understand he's moved on to other things as well.

Thanks, no problem.

[srhines]

In D142309#4107170, @xgupta wrote:

In D142309#4102284, @labath wrote:

In D142309#4075745, @JDevlieghere wrote:

LGTM, but I wonder when we'll be able to get rid of RenderScript completely (CC @labath)

I am not going to stand in your way. I usually redirect these questions to @srhines, but I understand he's moved on to other things as well.

Thanks, no problem.

RenderScript support can be removed. I don't necessarily have the bandwidth to do it immediately, but if someone wants to clean that up, I'd help with reviewing.

lldb/source/Plugins/LanguageRuntime/RenderScript/RenderScriptRuntime/RenderScriptRuntime.cpp

RenderScriptRuntime::GetAllocationData(AllocationDetails *alloc,
   }
 
   assert(alloc->data_ptr.isValid() && alloc->element.type.isValid() &&
          alloc->element.type_vec_size.isValid() && alloc->size.isValid() &&
          "Allocation information not available");
 
   // Allocate a buffer to copy data into
   const uint32_t size = *alloc->size.get();
-  std::shared_ptr<uint8_t> buffer(new uint8_t[size]);
+  std::shared_ptr<uint8_t []> buffer(new uint8_t[size]);
   if (!buffer) {
     LLDB_LOGF(log, "%s - couldn't allocate a %" PRIu32 " byte buffer",
               __FUNCTION__, size);
     return nullptr;
   }
 
   // Read the inferior memory
   Status err;
bool RenderScriptRuntime::SaveAllocation(Stream &strm, const uint32_t alloc_id,
   Status err = file.get()->Write(&head, num_bytes);
   if (!err.Success()) {
     strm.Printf("Error: '%s' when writing to file '%s'", err.AsCString(), path);
     strm.EOL();
     return false;
   }
 
   // Create the headers describing the element type of the allocation.
-  std::shared_ptr<uint8_t> element_header_buffer(
+  std::shared_ptr<uint8_t []> element_header_buffer(
       new uint8_t[element_header_size]);
   if (element_header_buffer == nullptr) {
     strm.Printf("Internal Error: Couldn't allocate %" PRIu64
                 " bytes on the heap",
                 (uint64_t)element_header_size);
     strm.EOL();
     return false;
   }