dev-amdgpu: Implement GPU clock MMIOs

The ROCr runtime uses a combination of HSA signal timestamps and
hardware MMIOs to calculate profiling times. At the beginning of an
application a timestamp is read from the GPU using MMIOs. The clock
MMIOs reside in the GFX MMIO region, so a new AMDGPUGfx class is added
to handle these MMIOs.

The timestamp value is expected to be in nanoseconds, so we simply use
the gem5 tick converted to ns.

Change-Id: I7d1cba40d5042a7f7a81fd4d132402dc11b71bd4

src/dev/amdgpu/amdgpu_device.hh

@@ -36,6 +36,7 @@
 
 #include "base/bitunion.hh"
 #include "dev/amdgpu/amdgpu_defines.hh"
+#include "dev/amdgpu/amdgpu_gfx.hh"
 #include "dev/amdgpu/amdgpu_nbio.hh"
 #include "dev/amdgpu/amdgpu_vm.hh"
 #include "dev/amdgpu/memory_manager.hh"
@@ -109,6 +110,7 @@ class AMDGPUDevice : public PciDevice
      * Blocks of the GPU
      */
     AMDGPUNbio nbio;
+    AMDGPUGfx gfx;
     AMDGPUMemoryManager *gpuMemMgr;
     AMDGPUInterruptHandler *deviceIH;
     AMDGPUVM gpuvm;