diff --git a/.gitignore b/.gitignore
index f0b11aa9..44f269cc 100644
--- a/.gitignore
+++ b/.gitignore
@@ -62,6 +62,7 @@ coverage
 /tests/test_filter_flags
 /tests/test_hardware
 /tests/test_index
+/test/test_lzip_decoder
 /tests/test_memlimit
 /tests/test_stream_flags
 /tests/test_vli
diff --git a/tests/Makefile.am b/tests/Makefile.am
index e612aa68..f84c1749 100644
--- a/tests/Makefile.am
+++ b/tests/Makefile.am
@@ -46,6 +46,7 @@ check_PROGRAMS = \
 	test_index \
 	test_bcj_exact_size \
 	test_memlimit \
+	test_lzip_decoder \
 	test_vli
 
 TESTS = \
@@ -57,6 +58,7 @@ TESTS = \
 	test_index \
 	test_bcj_exact_size \
 	test_memlimit \
+	test_lzip_decoder \
 	test_vli \
 	test_files.sh \
 	test_compress_prepared_bcj_sparc \
diff --git a/tests/test_lzip_decoder.c b/tests/test_lzip_decoder.c
new file mode 100644
index 00000000..a5495bf9
--- /dev/null
+++ b/tests/test_lzip_decoder.c
@@ -0,0 +1,471 @@
+///////////////////////////////////////////////////////////////////////////////
+//
+/// \file       test_lzip_decoder.c
+/// \brief      Tests decoding lzip data
+//
+//  Author:     Jia Tan
+//
+//  This file has been put into the public domain.
+//  You can do whatever you want with this file.
+//
+///////////////////////////////////////////////////////////////////////////////
+
+#include "tests.h"
+
+#ifdef HAVE_LZIP_DECODER
+
+// Memlimit large enough to pass all of the test files
+#define MEMLIMIT (1U << 20)
+#define DECODE_CHUNK_SIZE 1024
+
+
+// Avoiding using data buffers so we don't have to store the data buffers
+// as large hex strings. Instead, store the CRC32 value of the expected data.
+// CRC32 value of "Hello\nWorld\n"
+static const uint32_t hello_world_crc = 0x15A2A343;
+
+// CRC32 value of "Trailing garbage\n"
+static const uint32_t trailing_garbage_crc = 0x87081A60;
+
+
+// Helper function to decode a good file with no flags and plenty high memlimit
+static void
+basic_lzip_decode(const char *src, const uint32_t expected_crc) {
+	size_t file_size;
+	uint8_t *data = tuktest_file_from_srcdir(src, &file_size);
+	uint32_t checksum = 0;
+
+	lzma_stream strm = LZMA_STREAM_INIT;
+	assert_lzma_ret(lzma_lzip_decoder(&strm, MEMLIMIT, 0), LZMA_OK);
+
+	uint8_t *output_buffer = tuktest_malloc(DECODE_CHUNK_SIZE);
+
+	strm.next_in = data;
+	strm.next_out = output_buffer;
+	strm.avail_out = DECODE_CHUNK_SIZE;
+
+	// Feed 1 byte at a time to the decoder to look for any bugs
+	// when switching between decoding sequences
+	lzma_ret ret = LZMA_OK;
+	while (ret == LZMA_OK) {
+		strm.avail_in = 1;
+		ret = lzma_code(&strm, LZMA_RUN);
+		if (strm.avail_out == 0) {
+			checksum = lzma_crc32(output_buffer,
+					strm.next_out - output_buffer,
+					checksum);
+			// No need to free output_buffer because it will
+			// automatically be freed at the end of the test by
+			// tuktest.
+			output_buffer = tuktest_malloc(DECODE_CHUNK_SIZE);
+			strm.next_out = output_buffer;
+			strm.avail_out = DECODE_CHUNK_SIZE;
+		}
+	}
+
+	assert_lzma_ret(ret, LZMA_STREAM_END);
+	assert_uint_eq(strm.total_in, file_size);
+
+	checksum = lzma_crc32(output_buffer, strm.next_out - output_buffer,
+			checksum);
+	assert_uint_eq(checksum, expected_crc);
+
+	lzma_end(&strm);
+}
+
+
+static void
+test_options(void)
+{
+	// Test NULL stream
+	assert_lzma_ret(lzma_lzip_decoder(NULL, MEMLIMIT, 0),
+			LZMA_PROG_ERROR);
+
+	// Test invalid flags
+	lzma_stream strm = LZMA_STREAM_INIT;
+	assert_lzma_ret(lzma_lzip_decoder(&strm, MEMLIMIT, UINT32_MAX),
+			LZMA_OPTIONS_ERROR);
+	// Memlimit tests are done elsewhere
+}
+
+
+static void
+test_v0_decode(void) {
+	// This tests if liblzma can decode lzip version 0 files.
+	// lzip 1.17 and older can decompress this, but lzip 1.18
+	// and newer can no longer decode these files.
+	basic_lzip_decode("files/good-1-v0.lz", hello_world_crc);
+}
+
+
+static void
+test_v1_decode(void) {
+	// This tests decoding a basic lzip v1 file
+	basic_lzip_decode("files/good-1-v1.lz", hello_world_crc);
+}
+
+
+// Helper function to decode a good file with trailing bytes after
+// the lzip stream
+static void
+trailing_helper(const char *src, const uint32_t expected_data_checksum,
+		const uint32_t expected_trailing_checksum) {
+	size_t file_size;
+	uint32_t checksum = 0;
+	uint8_t *data = tuktest_file_from_srcdir(src, &file_size);
+	lzma_stream strm = LZMA_STREAM_INIT;
+	assert_lzma_ret(lzma_lzip_decoder(&strm, MEMLIMIT,
+			LZMA_CONCATENATED), LZMA_OK);
+
+	uint8_t *output_buffer = tuktest_malloc(DECODE_CHUNK_SIZE);
+
+	strm.next_in = data;
+	strm.next_out = output_buffer;
+	strm.avail_in = file_size;
+	strm.avail_out = DECODE_CHUNK_SIZE;
+
+	lzma_ret ret = LZMA_OK;
+	while (ret == LZMA_OK) {
+		ret = lzma_code(&strm, LZMA_RUN);
+		if (strm.avail_out == 0) {
+			checksum = lzma_crc32(output_buffer,
+					strm.next_out - output_buffer,
+					checksum);
+			// No need to free output_buffer because it will
+			// automatically be freed at the end of the test by
+			// tuktest.
+			output_buffer = tuktest_malloc(DECODE_CHUNK_SIZE);
+			strm.next_out = output_buffer;
+			strm.avail_out = DECODE_CHUNK_SIZE;
+		}
+	}
+
+	assert_lzma_ret(ret, LZMA_STREAM_END);
+	assert_uint(strm.total_in, <, file_size);
+
+	checksum = lzma_crc32(output_buffer,
+			strm.next_out - output_buffer,
+			checksum);
+
+	assert_uint_eq(checksum, expected_data_checksum);
+
+	// Trailing data should be readable from strm.next_in
+	checksum = lzma_crc32(strm.next_in, strm.avail_in, 0);
+	assert_uint_eq(checksum, expected_trailing_checksum);
+
+	lzma_end(&strm);
+}
+
+
+// Helper function to decode a bad file and compare to returned error to
+// what the caller expects
+static void
+decode_expect_error(const char *src, lzma_ret expected_error)
+{
+	lzma_stream strm = LZMA_STREAM_INIT;
+	size_t file_size;
+	uint8_t *data = tuktest_file_from_srcdir(src, &file_size);
+
+	assert_lzma_ret(lzma_lzip_decoder(&strm, MEMLIMIT,
+			LZMA_CONCATENATED), LZMA_OK);
+
+	uint8_t output_buffer[DECODE_CHUNK_SIZE];
+
+	strm.avail_in = file_size;
+	strm.next_in = data;
+	strm.avail_out = DECODE_CHUNK_SIZE;
+	strm.next_out = output_buffer;
+
+	lzma_ret ret = LZMA_OK;
+
+	while (ret == LZMA_OK) {
+		// Discard output since we are only looking for errors
+		strm.next_out = output_buffer;
+		strm.avail_out = DECODE_CHUNK_SIZE;
+		if (strm.avail_in == 0)
+			ret = lzma_code(&strm, LZMA_FINISH);
+		else
+			ret = lzma_code(&strm, LZMA_RUN);
+	}
+
+	assert_lzma_ret(ret, expected_error);
+	lzma_end(&strm);
+}
+
+
+static void
+test_v0_trailing(void) {
+	trailing_helper("files/good-1-v0-trailing-1.lz", hello_world_crc,
+			trailing_garbage_crc);
+}
+
+
+static void
+test_v1_trailing(void) {
+	trailing_helper("files/good-1-v1-trailing-1.lz", hello_world_crc,
+			trailing_garbage_crc);
+
+	// The second files/good-1-v1-trailing-2.lz will have the same
+	// expected output and trailing output as
+	// files/good-1-v1-trailing-1.lz, but this tests if the prefix
+	// to the trailing data contains lzip magic bytes.
+	// When this happens, the expected behavior is to silently ignore
+	// the magic byte prefix and consume it from the input file.
+	trailing_helper("files/good-1-v1-trailing-2.lz", hello_world_crc,
+			trailing_garbage_crc);
+
+	// Expect LZMA_BUF error if a file ends with the lzip magic bytes
+	// but does not contain any data after
+	decode_expect_error("files/bad-1-v1-trailing-magic.lz",
+			LZMA_BUF_ERROR);
+}
+
+
+static void
+test_concatentated(void)
+{
+	// First test a file with one v0 member and one v1 member
+	// The first member should contain "Hello\n" and
+	// the second member should contain "World!\n"
+
+	lzma_stream strm = LZMA_STREAM_INIT;
+	size_t file_size;
+	uint8_t *v0_v1 = tuktest_file_from_srcdir("files/good-2-v0-v1.lz",
+		&file_size);
+
+	assert_lzma_ret(lzma_lzip_decoder(&strm, MEMLIMIT,
+			LZMA_CONCATENATED), LZMA_OK);
+
+	uint8_t output_buffer[DECODE_CHUNK_SIZE];
+
+	strm.avail_in = file_size;
+	strm.next_in = v0_v1;
+	strm.avail_out = DECODE_CHUNK_SIZE;
+	strm.next_out = output_buffer;
+
+	assert_lzma_ret(lzma_code(&strm, LZMA_FINISH), LZMA_STREAM_END);
+
+	assert_uint_eq(strm.total_in, file_size);
+
+	uint32_t checksum = lzma_crc32(output_buffer, strm.total_out, 0);
+	assert_uint_eq(checksum, hello_world_crc);
+
+	// The second file contains one v1 member and one v2 member
+	uint8_t *v1_v0 = tuktest_file_from_srcdir("files/good-2-v1-v0.lz",
+		&file_size);
+
+	assert_lzma_ret(lzma_lzip_decoder(&strm, MEMLIMIT,
+			LZMA_CONCATENATED), LZMA_OK);
+
+	strm.avail_in = file_size;
+	strm.next_in = v1_v0;
+	strm.avail_out = DECODE_CHUNK_SIZE;
+	strm.next_out = output_buffer;
+
+	assert_lzma_ret(lzma_code(&strm, LZMA_FINISH), LZMA_STREAM_END);
+
+	assert_uint_eq(strm.total_in, file_size);
+	checksum = lzma_crc32(output_buffer, strm.total_out, 0);
+	assert_uint_eq(checksum, hello_world_crc);
+
+	// The third file contains 2 v1 members
+	uint8_t *v1_v1 = tuktest_file_from_srcdir("files/good-2-v1-v1.lz",
+		&file_size);
+
+	assert_lzma_ret(lzma_lzip_decoder(&strm, MEMLIMIT,
+			LZMA_CONCATENATED), LZMA_OK);
+
+	strm.avail_in = file_size;
+	strm.next_in = v1_v1;
+	strm.avail_out = DECODE_CHUNK_SIZE;
+	strm.next_out = output_buffer;
+
+	assert_lzma_ret(lzma_code(&strm, LZMA_FINISH), LZMA_STREAM_END);
+
+	assert_uint_eq(strm.total_in, file_size);
+	checksum = lzma_crc32(output_buffer, strm.total_out, 0);
+	assert_uint_eq(checksum, hello_world_crc);
+
+	lzma_end(&strm);
+}
+
+
+static void
+test_crc(void) {
+	// Test invalid checksum
+	lzma_stream strm = LZMA_STREAM_INIT;
+	size_t file_size;
+	uint8_t *data = tuktest_file_from_srcdir("files/bad-1-v1-crc32.lz",
+			&file_size);
+
+	assert_lzma_ret(lzma_lzip_decoder(&strm, MEMLIMIT,
+			LZMA_CONCATENATED), LZMA_OK);
+
+	uint8_t output_buffer[DECODE_CHUNK_SIZE];
+
+	strm.avail_in = file_size;
+	strm.next_in = data;
+	strm.avail_out = DECODE_CHUNK_SIZE;
+	strm.next_out = output_buffer;
+
+	assert_lzma_ret(lzma_code(&strm, LZMA_FINISH), LZMA_DATA_ERROR);
+
+	// Test ignoring the checksum value - should decode successfully
+	assert_lzma_ret(lzma_lzip_decoder(&strm, MEMLIMIT,
+			LZMA_CONCATENATED | LZMA_IGNORE_CHECK), LZMA_OK);
+
+	strm.avail_in = file_size;
+	strm.next_in = data;
+	strm.avail_out = DECODE_CHUNK_SIZE;
+	strm.next_out = output_buffer;
+
+	assert_lzma_ret(lzma_code(&strm, LZMA_FINISH), LZMA_STREAM_END);
+	assert_uint_eq(strm.total_in, file_size);
+
+	// Test tell check
+	assert_lzma_ret(lzma_lzip_decoder(&strm, MEMLIMIT,
+			LZMA_CONCATENATED | LZMA_TELL_ANY_CHECK), LZMA_OK);
+
+	strm.avail_in = file_size;
+	strm.next_in = data;
+	strm.avail_out = DECODE_CHUNK_SIZE;
+	strm.next_out = output_buffer;
+
+	assert_lzma_ret(lzma_code(&strm, LZMA_FINISH), LZMA_GET_CHECK);
+	assert_uint_eq(lzma_get_check(&strm), LZMA_CHECK_CRC32);
+	assert_lzma_ret(lzma_code(&strm, LZMA_FINISH), LZMA_DATA_ERROR);
+	lzma_end(&strm);
+}
+
+
+static void
+test_invalid_magic_bytes(void) {
+	uint8_t lzip_id_string[] = { 0x4C, 0x5A, 0x49, 0x50 };
+	lzma_stream strm = LZMA_STREAM_INIT;
+
+	for (uint32_t i = 0; i < ARRAY_SIZE(lzip_id_string); i++) {
+		// Corrupt magic bytes
+		lzip_id_string[i] ^= 1;
+		uint8_t output_buffer[DECODE_CHUNK_SIZE];
+
+		assert_lzma_ret(lzma_lzip_decoder(&strm, MEMLIMIT, 0),
+				LZMA_OK);
+
+		strm.next_in = lzip_id_string;
+		strm.avail_in = sizeof(lzip_id_string);
+		strm.next_out = output_buffer;
+		strm.avail_out = DECODE_CHUNK_SIZE;
+
+		assert_lzma_ret(lzma_code(&strm, LZMA_RUN),
+				LZMA_FORMAT_ERROR);
+
+		// Reset magic bytes
+		lzip_id_string[i] ^= 1;
+	}
+
+	lzma_end(&strm);
+}
+
+
+static void
+test_invalid_version(void)
+{
+	// The file contains a version number that is not 0 or 1,
+	// so it should cause an error
+	decode_expect_error("files/unsupported-1-v234.lz",
+			LZMA_OPTIONS_ERROR);
+}
+
+
+static void
+test_invalid_dictionary_size(void) {
+	// First file has too small dictionary size field
+	decode_expect_error("files/bad-1-v1-dict-1.lz", LZMA_DATA_ERROR);
+	
+	// Second file has too large dictionary size field
+	decode_expect_error("files/bad-1-v1-dict-2.lz", LZMA_DATA_ERROR);
+}
+
+
+static void
+test_invalid_uncomp_size(void) {
+	// Test invalid v0 lzip file uncomp size
+	decode_expect_error("files/bad-1-v0-uncomp-size.lz",
+			LZMA_DATA_ERROR);
+	
+	// Test invalid v1 lzip file uncomp size
+	decode_expect_error("files/bad-1-v1-uncomp-size.lz",
+			LZMA_DATA_ERROR);
+}
+
+
+static void
+test_invalid_member_size(void) {
+	decode_expect_error("files/bad-1-v1-member-size.lz",
+			LZMA_DATA_ERROR);
+}
+
+
+static void
+test_invalid_memlimit(void) {
+	// A very low memlimit should prevent decoding.
+	// Should be able to update the memlimit after failing
+	size_t file_size;
+	uint8_t *data = tuktest_file_from_srcdir("files/good-1-v1.lz",
+			&file_size);
+
+	uint8_t output_buffer[DECODE_CHUNK_SIZE];
+
+	lzma_stream strm = LZMA_STREAM_INIT;
+
+	assert_lzma_ret(lzma_lzip_decoder(&strm, 1, 0), LZMA_OK);
+
+	strm.next_in = data;
+	strm.avail_in = file_size;
+	strm.next_out = output_buffer;
+	strm.avail_out = DECODE_CHUNK_SIZE;
+
+	assert_lzma_ret(lzma_code(&strm, LZMA_FINISH), LZMA_MEMLIMIT_ERROR);
+
+	// Up the memlimit so decoding can continue.
+	// First only increase by a small amount and expect an error
+	assert_lzma_ret(lzma_memlimit_set(&strm, 100), LZMA_MEMLIMIT_ERROR);
+	assert_lzma_ret(lzma_memlimit_set(&strm, MEMLIMIT), LZMA_OK);
+
+	// Finish decoding
+	assert_lzma_ret(lzma_code(&strm, LZMA_FINISH), LZMA_STREAM_END);
+
+	assert_uint_eq(strm.total_in, file_size);
+	uint32_t checksum = lzma_crc32(output_buffer, strm.total_out, 0);
+	assert_uint_eq(checksum, hello_world_crc);
+
+	lzma_end(&strm);
+}
+#endif
+
+
+extern int
+main(int argc, char **argv)
+{
+	tuktest_start(argc, argv);
+
+#ifndef HAVE_LZIP_DECODER
+	tuktest_early_skip("lzip decoder disabled");
+#else
+	tuktest_run(test_options);
+	tuktest_run(test_v0_decode);
+	tuktest_run(test_v1_decode);
+	tuktest_run(test_v0_trailing);
+	tuktest_run(test_v1_trailing);
+	tuktest_run(test_concatentated);
+	tuktest_run(test_crc);
+	tuktest_run(test_invalid_magic_bytes);
+	tuktest_run(test_invalid_version);
+	tuktest_run(test_invalid_dictionary_size);
+	tuktest_run(test_invalid_uncomp_size);
+	tuktest_run(test_invalid_member_size);
+	tuktest_run(test_invalid_memlimit);
+	return tuktest_end();
+#endif
+
+}