Title: [194409] trunk/Source/_javascript_Core
- Revision
- 194409
- Author
- [email protected]
- Date
- 2015-12-23 18:17:03 -0800 (Wed, 23 Dec 2015)
Log Message
jsc CLI tool crashes on EOF.
<https://webkit.org/b/152522>
Reviewed by Benjamin Poulain.
SourceProvider should treat String() like the empty string for hashing purposes.
This was a subtle behavior change in r194017 due to how zero-length strings are
treated by StringImpl::createSubstringSharingImpl().
I made these SourceProviders store a Ref<StringImpl> internally instead of a
String, to codify the fact that these strings can't be null strings.
I couldn't find a way to cause this crash through the API.
* API/JSScriptRef.cpp:
(OpaqueJSScript::OpaqueJSScript):
* parser/SourceProvider.h:
(JSC::StringSourceProvider::StringSourceProvider):
Modified Paths
Diff
Modified: trunk/Source/_javascript_Core/API/JSScriptRef.cpp (194408 => 194409)
--- trunk/Source/_javascript_Core/API/JSScriptRef.cpp 2015-12-24 01:36:43 UTC (rev 194408)
+++ trunk/Source/_javascript_Core/API/JSScriptRef.cpp 2015-12-24 02:17:03 UTC (rev 194409)
@@ -48,12 +48,12 @@
unsigned hash() const override
{
- return m_source.impl()->hash();
+ return m_source.get().hash();
}
StringView source() const override
{
- return m_source;
+ return m_source.get();
}
VM* vm() const { return m_vm; }
@@ -62,14 +62,14 @@
OpaqueJSScript(VM* vm, const String& url, int startingLineNumber, const String& source)
: SourceProvider(url, TextPosition(OrdinalNumber::fromOneBasedInt(startingLineNumber), OrdinalNumber::first()))
, m_vm(vm)
- , m_source(source)
+ , m_source(source.isNull() ? *StringImpl::empty() : *source.impl())
{
}
virtual ~OpaqueJSScript() { }
VM* m_vm;
- String m_source;
+ Ref<StringImpl> m_source;
};
static bool parseScript(VM* vm, const SourceCode& source, ParserError& error)
Modified: trunk/Source/_javascript_Core/ChangeLog (194408 => 194409)
--- trunk/Source/_javascript_Core/ChangeLog 2015-12-24 01:36:43 UTC (rev 194408)
+++ trunk/Source/_javascript_Core/ChangeLog 2015-12-24 02:17:03 UTC (rev 194409)
@@ -1,3 +1,24 @@
+2015-12-23 Andreas Kling <[email protected]>
+
+ jsc CLI tool crashes on EOF.
+ <https://webkit.org/b/152522>
+
+ Reviewed by Benjamin Poulain.
+
+ SourceProvider should treat String() like the empty string for hashing purposes.
+ This was a subtle behavior change in r194017 due to how zero-length strings are
+ treated by StringImpl::createSubstringSharingImpl().
+
+ I made these SourceProviders store a Ref<StringImpl> internally instead of a
+ String, to codify the fact that these strings can't be null strings.
+
+ I couldn't find a way to cause this crash through the API.
+
+ * API/JSScriptRef.cpp:
+ (OpaqueJSScript::OpaqueJSScript):
+ * parser/SourceProvider.h:
+ (JSC::StringSourceProvider::StringSourceProvider):
+
2015-12-23 Filip Pizlo <[email protected]>
FTL B3 should be able to run crypto-sha1 in eager mode
Modified: trunk/Source/_javascript_Core/parser/SourceProvider.h (194408 => 194409)
--- trunk/Source/_javascript_Core/parser/SourceProvider.h 2015-12-24 01:36:43 UTC (rev 194408)
+++ trunk/Source/_javascript_Core/parser/SourceProvider.h 2015-12-24 02:17:03 UTC (rev 194409)
@@ -90,22 +90,22 @@
unsigned hash() const override
{
- return m_source.impl()->hash();
+ return m_source.get().hash();
}
virtual StringView source() const override
{
- return m_source;
+ return m_source.get();
}
private:
StringSourceProvider(const String& source, const String& url, const TextPosition& startPosition)
: SourceProvider(url, startPosition)
- , m_source(source)
+ , m_source(source.isNull() ? *StringImpl::empty() : *source.impl())
{
}
- String m_source;
+ Ref<StringImpl> m_source;
};
#if ENABLE(WEBASSEMBLY)
_______________________________________________
webkit-changes mailing list
[email protected]
https://lists.webkit.org/mailman/listinfo/webkit-changes