I believe the schema simpleContent.xsd is invalid. (It was introduced to support new tests testing the resolution of bug #4413, that is validateexpr-sc-1 et seq, I believe by Mary Holstege.)

Line 28 has

<xs:complexType name="complexExtendsEmptyMixed">
  <xs:annotation>
    <xs:documentation>
    A complex type that extends a mixed content base with no content model.
    </xs:documentation>
  </xs:annotation>
  <xs:simpleContent>
    <xs:extension base="ss:emptyMixed"/>
  </xs:simpleContent>
</xs:complexType>

where ss:emptyMixed is

<xs:complexType name="emptyMixed" mixed="true">
  <xs:annotation>
    <xs:documentation>
    A complex type that is mixed but with no content model.
    </xs:documentation>
  </xs:annotation>
  <xs:sequence/>
</xs:complexType>

Under XSD 1.0, emptyMixed results in a complexType schema component with {contentType} = (mixed, a particle with (minOccurs=1, maxOccurs=1, term=(compositor=sequence, particles=empty))).

The complexType component derived from complexExtendsEmptyMixed does not satisfy Schema Component Constraint: Derivation Valid (Extension), because none of the conditions in clause 1.4 is true:

1.4.1 is false because the {content type} of the {base type definition} is not a simple type definition

1.4.2 is false because the {content type} of the {base type definition} is not empty (it is a particle whose term is an empty sequence, which is not the same thing)

1.4.3 is false because 1.4.3.1 is false: the {content type} of the complex type definition itself [complexExtendsEmptyMixed] does not specify a particle