[PATCH 1/2] generic/45[34]: add detection of confusable variation sequences

From: Darrick J. Wong <[email protected]>

ArsTechnica recently wrote about a GitHub supply chain attack wherein
non-rendering unicode sequences were embedded in javascript files to
hide payloads that could be decrypted trivially later.  While these are
unlikely to appear in file and attribute names, xfs_scrub will warn about
this sort of steganography, so let's make sure it works.

Signed-off-by: "Darrick J. Wong" <[email protected]>
---
 tests/generic/453 |   35 +++++++++++++++++++++++++++++++++++
 tests/generic/454 |   36 ++++++++++++++++++++++++++++++++++++
 2 files changed, 71 insertions(+)


diff --git a/tests/generic/453 b/tests/generic/453
index bd5ce8b2bb11d9..0193b010306c48 100755
--- a/tests/generic/453
+++ b/tests/generic/453
@@ -233,6 +233,20 @@ setf "\xf0\x9f\xab\xb6\xf0\x9f\x8f\xbc" "medium light"
 setf "\xf0\x9f\xab\xb6\xf0\x9f\x8f\xbb" "light"
 setf "\xf0\x9f\xab\xb6" "neutral"
 
+# confusion with variation selectors
+setf "variations.txt" v0
+setf "varia\xef\xb8\x80tions.txt" v1
+setf "\xef\xb8\x80variations.txt" v2
+setf "vari\xef\xb8\x80\xef\xb8\x81ations.txt" v3
+setf "varia\xf3\xa0\x87\xa4tions.txt" v4
+
+# deprecated tags are considered control characters
+setf "tags_moocow.txt" u0
+setf "tags_m\xf3\xa0\x81\xadoocow.txt" u1
+
+# totally hidden name? "(Hi)" is the file name
+setf "\xf3\xa0\x80\xa8\xf3\xa0\x81\x88\xf3\xa0\x81\xa9\xf3\xa0\x80\xa9" "(Hi)"
+
 ls -laR $testdir >> $seqres.full
 
 echo "Test files"
@@ -331,6 +345,20 @@ testf "\xf0\x9f\xab\xb6\xf0\x9f\x8f\xbc" "medium light"
 testf "\xf0\x9f\xab\xb6\xf0\x9f\x8f\xbb" "light"
 testf "\xf0\x9f\xab\xb6" "neutral"
 
+# confusion with variation selectors
+testf "variations.txt" v0
+testf "varia\xef\xb8\x80tions.txt" v1
+testf "\xef\xb8\x80variations.txt" v2
+testf "vari\xef\xb8\x80\xef\xb8\x81ations.txt" v3
+testf "varia\xf3\xa0\x87\xa4tions.txt" v4
+
+# deprecated tags are considered control characters
+testf "tags_moocow.txt" u0
+testf "tags_m\xf3\xa0\x81\xadoocow.txt" u1
+
+# totally hidden name? "(Hi)" is the file name
+testf "\xf3\xa0\x80\xa8\xf3\xa0\x81\x88\xf3\xa0\x81\xa9\xf3\xa0\x80\xa9" "(Hi)"
+
 echo "Uniqueness of inodes?"
 stat -c '%i' "${testdir}/"* | sort | uniq -c | while read nr inum; do
 	if [ "${nr}" -gt 1 ]; then
@@ -368,6 +396,13 @@ if _check_xfs_scrub_does_unicode "$SCRATCH_MNT" "$SCRATCH_DEV"; then
 		grep -q "llamapirate" $tmp.scrub || echo "No complaints about hidden llm instructions in filenames?"
 	fi
 
+	if grep -q "variations" $tmp.scrub; then
+		grep -q 'varia.xef.xb8' $tmp.scrub || echo "No complaints about variation sequence confusion?"
+		grep -q 'varia.xf3.xa0' $tmp.scrub || echo "No complaints about extended variation sequence confusion?"
+		grep -q 'x80variations' $tmp.scrub || echo "No complaints about variations starting a name?"
+		grep -q 'tags_m.xf3.xa0.x81' $tmp.scrub || echo "No complaints about deprecated unicode tags in a name?"
+	fi
+
 	echo "Actual xfs_scrub output:" >> $seqres.full
 	cat $tmp.scrub >> $seqres.full
 fi
diff --git a/tests/generic/454 b/tests/generic/454
index 9f6ddb4a0e48b2..3454cae5d5ea6c 100755
--- a/tests/generic/454
+++ b/tests/generic/454
@@ -154,6 +154,20 @@ setf "\xf0\x9f\xab\xb6\xf0\x9f\x8f\xbc" "medium light"
 setf "\xf0\x9f\xab\xb6\xf0\x9f\x8f\xbb" "light"
 setf "\xf0\x9f\xab\xb6" "neutral"
 
+# confusion with variation selectors
+setf "variations.txt" v0
+setf "varia\xef\xb8\x80tions.txt" v1
+setf "\xef\xb8\x80variations.txt" v2
+setf "vari\xef\xb8\x80\xef\xb8\x81ations.txt" v3
+setf "varia\xf3\xa0\x87\xa4tions.txt" v4
+
+# deprecated tags are considered control characters
+setf "tags_moocow.txt" u0
+setf "tags_m\xf3\xa0\x81\xadoocow.txt" u1
+
+# totally hidden name? "(Hi)" is the file name
+setf "\xf3\xa0\x80\xa8\xf3\xa0\x81\x88\xf3\xa0\x81\xa9\xf3\xa0\x80\xa9" "(Hi)"
+
 _getfattr --absolute-names -d "${testfile}" >> $seqres.full
 
 echo "Test files"
@@ -229,6 +243,20 @@ testf "\xf0\x9f\xab\xb6\xf0\x9f\x8f\xbc" "medium light"
 testf "\xf0\x9f\xab\xb6\xf0\x9f\x8f\xbb" "light"
 testf "\xf0\x9f\xab\xb6" "neutral"
 
+# confusion with variation selectors
+testf "variations.txt" v0
+testf "varia\xef\xb8\x80tions.txt" v1
+testf "\xef\xb8\x80variations.txt" v2
+testf "vari\xef\xb8\x80\xef\xb8\x81ations.txt" v3
+testf "varia\xf3\xa0\x87\xa4tions.txt" v4
+
+# deprecated tags are considered control characters
+testf "tags_moocow.txt" u0
+testf "tags_m\xf3\xa0\x81\xadoocow.txt" u1
+
+# totally hidden name? "(Hi)" is the file name
+testf "\xf3\xa0\x80\xa8\xf3\xa0\x81\x88\xf3\xa0\x81\xa9\xf3\xa0\x80\xa9" "(Hi)"
+
 echo "Uniqueness of keys?"
 crazy_keys="$(_getfattr --absolute-names -d "${testfile}" | grep -E -c '(french_|chinese_|greek_|arabic_|urk)')"
 expected_keys=11
@@ -249,6 +277,14 @@ if _check_xfs_scrub_does_unicode "$SCRATCH_MNT" "$SCRATCH_DEV"; then
 	grep -q "prohibition_" $tmp.scrub || echo "No complaints about prohibited sequence confusables?"
 	grep -q "zerojoin_" $tmp.scrub || echo "No complaints about zero-width join confusables?"
 	grep -q "llamapirate" $tmp.scrub || echo "No complaints about hidden llm instructions in filenames?"
+
+	if grep -q "variations" $tmp.scrub; then
+		grep -q 'varia.xef.xb8' $tmp.scrub || echo "No complaints about variation sequence confusion?"
+		grep -q 'varia.xf3.xa0' $tmp.scrub || echo "No complaints about extended variation sequence confusion?"
+		grep -q 'x80variations' $tmp.scrub || echo "No complaints about variations starting a name?"
+		grep -q 'tags_m.xf3.xa0.x81' $tmp.scrub || echo "No complaints about deprecated unicode tags in a name?"
+	fi
+
 	echo "Actual xfs_scrub output:" >> $seqres.full
 	echo "${output}" >> $seqres.full
 fi