gpu-compute: Make headTailMap a std::unordered_map
authorTony Gutierrez <anthony.gutierrez@amd.com>
Tue, 13 Aug 2019 18:15:16 +0000 (14:15 -0400)
committerAnthony Gutierrez <anthony.gutierrez@amd.com>
Mon, 22 Jun 2020 16:14:35 +0000 (16:14 +0000)
There is no reason that the headTailMap needs to be
sorted, so let's use a std::unordered_map.

Change-Id: I18641b893352c18ec86e3775c8947a05a6c6547d
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/29930
Reviewed-by: Anthony Gutierrez <anthony.gutierrez@amd.com>
Maintainer: Anthony Gutierrez <anthony.gutierrez@amd.com>
Tested-by: kokoro <noreply+kokoro@google.com>
src/gpu-compute/compute_unit.hh

index 187cbc9d541e9bdff473cbe2106b0745ac5310d7..110097e0c519c8891398b2d8897f8c56f44705dd 100644 (file)
@@ -981,7 +981,7 @@ class ComputeUnit : public ClockedObject
     // hold the time of the arrival of the first cache block related to
     // a particular GPUDynInst. This is used to calculate the difference
     // between the first and last chace block arrival times.
-    std::map<GPUDynInstPtr, Tick> headTailMap;
+    std::unordered_map<GPUDynInstPtr, Tick> headTailMap;
 };
 
 #endif // __COMPUTE_UNIT_HH__