According to the specification the characters [#x10000-#xEFFFF] are legal in XML names. However, the W3 validator says that this XML is not well-formed:
<?xml version="1.0"?>
<>value</>
(the name of the attribute is a Unicode character #x10400). Some browsers, like Firefox, also complain about it (Chrome displays XML, IE shows a blank page). Is it an error in tools or the XML is really not well-formed?
Yes, supplementary characters are allowed in XML names.
Your XML is well-formed because the element name uses characters allowed by the Name production in the W3C XML Recommendation.
However:
Online validators that get the file from you over HTTP will have to take care to mind the character encoding. It appears that by the time the W3C Markup Validation Service gets your XML, your character is getting lost in an encoding shuffle:
Try an offline XML parser. My Xerces-J-based validator, for example, correctly identifies your XML as being well-formed.
Be aware that not all characters allowed by NAME are allowed in NCNAMEs. So, although well-formed, XML using such characters cannot be valid according to an XSD where such names are not allowed.