Comment 1:

The Go Programming Language Specification
Version of September 4, 2012
http://golang.org/ref/spec
Source code representation
Source code is Unicode text encoded in UTF-8. The text is not canonicalized, so a single
accented code point is distinct from the same character constructed from combining an
accent and a letter; those are treated as two code points. For simplicity, this document
will use the unqualified term character to refer to a Unicode code point in the source
text.
Each code point is distinct; for instance, upper and lower case letters are different
characters.
Implementation restriction: For compatibility with other tools, a compiler may disallow
the NUL character (U+0000) in the source text. 
The Unicode Standard, Version 6.2
Chapter 3 Conformance
http://www.unicode.org/versions/Unicode6.2.0/ch03.pdf
D95
When represented in UTF-8, the byte order mark [U+FEFF] turns into the byte sequence
<EF BB BF>.
D89
In a Unicode encoding form: A Unicode string is said to be in a particular Unicode
encoding form if and only if it consists of a well-formed Unicode code unit sequence of
that Unicode encoding form.
• A Unicode string consisting of a well-formed UTF-8 code unit sequence is said to be
in UTF-8. Such a Unicode string is referred to as a valid UTF-8 string, or a UTF-8
string for short.
D92
• Any UTF-8 byte sequence that does not match the patterns listed in Table 3-7 is
ill-formed.
Table 3-7 lists all of the byte sequences that are well-formed in UTF-8. 
Table 3-7. Well-Formed UTF-8 Byte Sequences [in pertinent part]
Code Points        First Byte Second Byte Third Byte Fourth Byte
U+E000..U+FFFF     EE..EF     80..BF      80..BF 
The Unicode specification defines UTF-8. It looks to me as if the UTF-8 byte sequence
<EF BB BF>, for the BOM U+FEFF code point, is defined by Unicode as a well-formed
sequence of UTF-8 bytes. Therefore, I'm surprised that Go does not accept it. Are there
any other well-formed sequences of UTF-8 bytes does Go not accept, apart from the NUL
character?
Does this break the Go 1 guarantee that "Source code is Unicode text encoded in UTF-8.",
except that "a compiler may disallow the NUL character (U+0000)"?

Comment 2:

Clarification of the spec in https://golang.org/cl/8649043
This isn't really a bug, just an inconsistency, since the property falls under the
'implementation restriction' clause. Retracting the issue.

Status changed to Retracted.