[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[DotGNU]yet another compiler bug (System.Xml related)
From: |
Gopal V |
Subject: |
[DotGNU]yet another compiler bug (System.Xml related) |
Date: |
Fri, 27 Sep 2002 20:21:02 +0530 |
User-agent: |
Mutt/1.2.5i |
Hi All,
Today's "Bug for the day" is due to the hack in System.Xml's
XmlNamespaceManager ... which was introduced to make it faster .
I followed the bug down to pnet/csharp/cs_oper.tc::EqualitySem()
cs_oper.tc:1509
if(ILCanCoerceNode(info, node->expr1,
CSSemGetType(value1),
CSSemGetType(value2)) ||
ILCanCoerceNode(info, node->expr2,
CSSemGetType(value2),
CSSemGetType(value1)))
{
fprintf(stderr,"<%s:%d>
<%d:%d>\n",__FILE__,__LINE__,
ILNode_GetType(node->expr1,info)==ILMachineType_String,
ILNode_GetType(node->expr2,info)==ILMachineType_String
);
was what I used ... and the test file is attached as xml.cs
the output code is also attached as err ...
The error seems to be that ILNode_GetType(node->expr2) seems to
return ILMachineType_String eventhough it is explicitly casted to
(Object)
So the codegen maybe executing
ILRelational_ApplyOp(ILNode_Eq, ILMachineType_String)
{
ILGenCallByName(info,
"bool [.library]System.String::op_Equality"
"(class [.library]System.String, class
[.library]System.String)");
}
Other than these obvious symptoms , I could not pick out the bug in
question ...
Gopal
--
The difference between insanity and genius is measured by success
xml.cs
Description: Text document
err
Description: Text document
- [DotGNU]yet another compiler bug (System.Xml related),
Gopal V <=