[Quant] make x86 the default quantization backend (qengine) (#91235)

Xia-Weiwen · pytorchmergebot · commit e1bcbbf18cae · 2023-01-12T02:14:28.000Z
**Summary** Make x86 the default quantization backend (qengine) for X86 CPU platforms. X86 is a unified quantization backend combining goodness of fbgemm and onednn. For more details please see #83888 **Test plan** python test/test_quantization.py Pull Request resolved: #91235 Approved by: https://github.com/jgong5, https://github.com/XiaobingSuper, https://github.com/malfet
diff --git a/aten/src/ATen/Context.cpp b/aten/src/ATen/Context.cpp
@@ -302,7 +302,12 @@ at::QEngine Context::qEngine() const {
 
 #ifdef USE_FBGEMM
     if (fbgemm::fbgemmSupportedCPU()) {
-      qengine = at::kFBGEMM;
+      /* X86 is enabled if and only if fbgemm is available.
+       * It combines goodness of fbgemm and onednn by dispatching.
+       * If onednn not available, always dispatch to fbgemm.
+       * Make it default qengine for X86 CPU platforms.
+      */
+      qengine = at::kX86;
     }
 #endif
     return qengine;
