Title: [244580] trunk/Source/_javascript_Core
Revision
244580
Author
[email protected]
Date
2019-04-23 21:06:08 -0700 (Tue, 23 Apr 2019)

Log Message

[JSC] Shrink DFG::MinifiedNode
https://bugs.webkit.org/show_bug.cgi?id=197224

Reviewed by Filip Pizlo.

Since it is kept alive with compiled DFG code, we should shrink it to save memory.
If it is effective, we should consider minimizing these OSR exit data more aggressively.

* dfg/DFGMinifiedNode.h:

Modified Paths

Diff

Modified: trunk/Source/_javascript_Core/ChangeLog (244579 => 244580)


--- trunk/Source/_javascript_Core/ChangeLog	2019-04-24 02:14:23 UTC (rev 244579)
+++ trunk/Source/_javascript_Core/ChangeLog	2019-04-24 04:06:08 UTC (rev 244580)
@@ -1,3 +1,15 @@
+2019-04-23  Yusuke Suzuki  <[email protected]>
+
+        [JSC] Shrink DFG::MinifiedNode
+        https://bugs.webkit.org/show_bug.cgi?id=197224
+
+        Reviewed by Filip Pizlo.
+
+        Since it is kept alive with compiled DFG code, we should shrink it to save memory.
+        If it is effective, we should consider minimizing these OSR exit data more aggressively.
+
+        * dfg/DFGMinifiedNode.h:
+
 2019-04-23  Saam Barati  <[email protected]>
 
         LICM incorrectly assumes it'll never insert a node which provably OSR exits

Modified: trunk/Source/_javascript_Core/dfg/DFGMinifiedNode.h (244579 => 244580)


--- trunk/Source/_javascript_Core/dfg/DFGMinifiedNode.h	2019-04-24 02:14:23 UTC (rev 244579)
+++ trunk/Source/_javascript_Core/dfg/DFGMinifiedNode.h	2019-04-24 04:06:08 UTC (rev 244580)
@@ -90,8 +90,8 @@
         return type == PhantomDirectArguments || type == PhantomClonedArguments;
     }
     
+    uint64_t m_info;
     MinifiedID m_id;
-    uint64_t m_info;
     NodeType m_op;
 };
 
_______________________________________________
webkit-changes mailing list
[email protected]
https://lists.webkit.org/mailman/listinfo/webkit-changes

Reply via email to