gpu-compute: Ignore GPU kernel names

ROCm 4 seems to have updated the akc, and the only real issue that has
occured is that we're no longer able to read kernel names in the same
way as we were in ROCm 1.6. This patch removes the prior method of
reading kernel names and gives all kernels a temporary name

Change-Id: I0040e0cf4cd35d6f56ded6a8acfb10c600bcc77a
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/46245
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Matt Sinclair <mattdsinclair@gmail.com>
Reviewed-by: Matthew Poremba <matthew.poremba@amd.com>
Maintainer: Matt Sinclair <mattdsinclair@gmail.com>

src/gpu-compute/gpu_command_processor.cc

@@ -171,7 +171,6 @@ GPUCommandProcessor::submitDispatchPkt(void *raw_pkt, uint32_t queue_id,
     DPRINTF(GPUCommandProc, "Machine code starts at addr: %#x\n",
         machine_code_addr);
 
-    Addr kern_name_addr(0);
     std::string kernel_name;
 
     /**
@@ -184,10 +183,7 @@ GPUCommandProcessor::submitDispatchPkt(void *raw_pkt, uint32_t queue_id,
      * host memory.  I have no idea what BLIT stands for.
      * */
     if (akc.runtime_loader_kernel_symbol) {
-        virt_proxy.readBlob(akc.runtime_loader_kernel_symbol + 0x10,
-            (uint8_t*)&kern_name_addr, 0x8);
-
-        virt_proxy.readString(kernel_name, kern_name_addr);
+        kernel_name = "Some kernel";
     } else {
         kernel_name = "Blit kernel";
     }