From: Mirek Klimos <miro@fb.com>
Date: Fri, 4 Nov 2016 18:10:46 +0000 (-0700)
Subject: Trying to improve the efficiency of RequestContext callbacks
X-Git-Tag: v2016.11.07.00~4
X-Git-Url: http://demsky.eecs.uci.edu/git/?a=commitdiff_plain;h=245489e26719816e7ee9939a60c6d359eb7139be;p=folly.git

Trying to improve the efficiency of RequestContext callbacks

Summary: Changed RWSpinLock to a SharedMutex in the RequestContext to see if this would lower the overhead of the onSet/onUnset callbacks.

Reviewed By: djwatson

Differential Revision: D3780396

fbshipit-source-id: 3d91f8194357533072fe053d23e0454a36faecff
---

diff --git a/folly/io/async/Request.h b/folly/io/async/Request.h
index 59d66f94..67f220d4 100644
--- a/folly/io/async/Request.h
+++ b/folly/io/async/Request.h
@@ -23,7 +23,7 @@
 #include <map>
 #include <memory>
 
-#include <folly/RWSpinLock.h>
+#include <folly/SharedMutex.h>
 #include <folly/Synchronized.h>
 
 namespace folly {
@@ -99,7 +99,7 @@ class RequestContext {
   static std::shared_ptr<RequestContext>& getStaticContext();
 
   using Data = std::map<std::string, std::unique_ptr<RequestData>>;
-  folly::Synchronized<Data, folly::RWSpinLock> data_;
+  folly::Synchronized<Data, folly::SharedMutex> data_;
 };
 
 class RequestContextScopeGuard {