[Lldb-commits] [lldb] [lldb] Add accelerator dynamic loader over gdb-remote (PR #217393)

satyanarayana reddy janga via lldb-commits lldb-commits at lists.llvm.org
Sat Sep 26 09:07:00 PDT 2026


================
@@ -0,0 +1,158 @@
+//===----------------------------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "DynamicLoaderAcceleratorGDBRemote.h"
+#include "Plugins/Process/gdb-remote/ProcessGDBRemote.h"
+#include "lldb/Core/Module.h"
+#include "lldb/Core/PluginManager.h"
+#include "lldb/Core/Section.h"
+#include "lldb/Target/Target.h"
+#include "lldb/Utility/LLDBLog.h"
+#include "lldb/Utility/Log.h"
+
+using namespace lldb;
+using namespace lldb_private;
+using namespace lldb_private::process_gdb_remote;
+
+LLDB_PLUGIN_DEFINE(DynamicLoaderAcceleratorGDBRemote)
+
+DynamicLoader *
+DynamicLoaderAcceleratorGDBRemote::CreateInstance(Process *process,
+                                                  bool force) {
+  // The library list comes from the accelerator's GDB server.
+  if (process->GetPluginName() != ProcessGDBRemote::GetPluginNameStatic())
+    return nullptr;
+  if (force)
+    return new DynamicLoaderAcceleratorGDBRemote(process);
+
+  const llvm::Triple &triple =
+      process->GetTarget().GetArchitecture().GetTriple();
+  if (triple.isAMDGPU() || triple.isNVPTX())
----------------
satyajanga wrote:

@clayborg I am confused. 

cpu-process calls the accelerator-plugin-initialize, but gpu-process needs the dyld-plugin-name. 

I think we want gpu-llgs to tell which dyld-plugin to use. so in that case accelerator-initialize response is not the right place.

So I believe your jLLDBSettings is more cleaner approach. I will rename this packet to jLLDBProcessSettings and do the same logic. 

let me know what you think?

https://github.com/llvm/llvm-project/pull/217393


More information about the lldb-commits mailing list