Make the HTMLDocumentParser yield more aggressively

Source/core/html/parser/BackgroundHTMLParser.cpp

Index: Source/core/html/parser/BackgroundHTMLParser.cpp
diff --git a/Source/core/html/parser/BackgroundHTMLParser.cpp b/Source/core/html/parser/BackgroundHTMLParser.cpp
index 66a62b8d1e77979e627500f19eb0851da2e89e22..0dc1faf108e50c1e3712f53f3d3cab967ecbcc38 100644
--- a/Source/core/html/parser/BackgroundHTMLParser.cpp
+++ b/Source/core/html/parser/BackgroundHTMLParser.cpp
@@ -45,12 +45,12 @@ namespace WebCore {
 // off the main thread's token buffer.
 // At time of writing, this is (10000 + 1000) * 28 bytes = ~308kb of memory.
 // These numbers have not been tuned.

[abarth-chromium, 2014/04/29 18:29:00]

Can you update the math in this comment?

-static const size_t outstandingTokenLimit = 10000;

[eseidel, 2014/04/29 00:23:13]

I didn't mean to leave these in.

[eseidel, 2014/04/29 00:24:12]

These aren't bad.  I think they can just be deferred to a later change with more
mobile testing.

+static const size_t outstandingTokenLimit = 1000;
 
 // We limit our chucks to 1000 tokens, to make sure the main

[abarth-chromium, 2014/04/29 18:29:00]

This comment is now out of date.

 // thread is never waiting on the parser thread for tokens.
 // This was tuned in https://bugs.webkit.org/show_bug.cgi?id=110408.
-static const size_t pendingTokenLimit = 1000;
+static const size_t pendingTokenLimit = 100;
 
 using namespace HTMLNames;