def test_from_archive_path():
# delete any existing unpacked directory
fs.rm(tests.data_path("tiny", "corpus", exists=False))
c = clgen.Corpus.from_json(
{"path": tests.data_path("tiny", "corpus.tar.bz2")})
assert TINY_HASH == c.hash
def test_commit(self):
# Create a copy database.
fs.cp(self.db.path, "/tmp/labm8.con.sql")
# Open two connections to database.
c1 = db.Database("/tmp/labm8.con.sql")
c2 = db.Database("/tmp/labm8.con.sql")
cmd = 'SELECT * FROM names WHERE first="Bob" AND last="Marley"'
# Check there's no Bob Marley entry.
self._test(None, c1.execute(cmd).fetchone())
self._test(None, c2.execute(cmd).fetchone())
# Add a Bob Marley entry to one connection.
c1.execute("INSERT INTO names VALUES ('Bob', 'Marley')")
# Create a third database connection.
c3 = db.Database("/tmp/labm8.con.sql")
# Check that the second and third connections can't see this new entry.
self._test(("Bob", "Marley"), c1.execute(cmd).fetchone())
self._test(None, c2.execute(cmd).fetchone())
self._test(None, c3.execute(cmd).fetchone())
# Commit, and repeat. Check that all connections can now see
# Bob Marley.
c1.commit()
self._test(("Bob", "Marley"), c1.execute(cmd).fetchone())
self._test(("Bob", "Marley"), c2.execute(cmd).fetchone())
self._test(("Bob", "Marley"), c3.execute(cmd).fetchone())
# Cool, we're jammin'
fs.rm("/tmp/labm8.con.sql")
def clear(self):
"""
Empty the filesystem cache.
This deletes the entire cache directory.
"""
fs.rm(self.path)
def test_from_archive_path(self):
# delete any existing unpacked directory
fs.rm(tests.data_path("tiny", "corpus"))
c = corpus.Corpus.from_json(
{"path": tests.data_path("tiny", "corpus.tar.bz2")})
self.assertEqual(TINY_HASH, c.hash)
def clear(self):
"""
Empty the filesystem cache.
This deletes the entire cache directory.
"""
fs.rm(self.path)
def drive_testcase(s: db.session_t, testcase: CLgenTestCase,
env: cldrive.OpenCLEnvironment, platform_id: int,
device_id: int, timeout: int = 60) -> return_t:
""" run CLgen program test harness """
harness = clgen_mkharness.mkharness(s, env, testcase)
with NamedTemporaryFile(prefix='cldrive-harness-', delete=False) as tmpfile:
path = tmpfile.name
try:
clgen_mkharness.compile_harness(
harness.src, path, platform_id=platform_id, device_id=device_id)
cmd = ['timeout', '-s9', str(timeout), tmpfile.name]
start_time = time()
proc = subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
stdout, stderr = proc.communicate()
try:
stdout = stdout.decode('utf-8')
except UnicodeError as e:
stdout = '<-- UTF-ERROR -->'
try:
stderr = stderr.decode('utf-8')
except UnicodeError as e:
stderr = '<-- UTF-ERROR -->'
runtime = time() - start_time
return return_t(
runtime=runtime, status=status_t(proc.returncode),
stdout=stdout, stderr=stderr)
finally:
fs.rm(path)
def test_cp(self):
system.echo("Hello, world!", "/tmp/labm8.tmp")
self._test(["Hello, world!"], fs.read("/tmp/labm8.tmp"))
# Cleanup any existing file.
fs.rm("/tmp/labm8.tmp.copy")
self._test(False, fs.exists("/tmp/labm8.tmp.copy"))
fs.cp("/tmp/labm8.tmp", "/tmp/labm8.tmp.copy")
self._test(fs.read("/tmp/labm8.tmp"), fs.read("/tmp/labm8.tmp.copy"))
def _init_error(err: Exception, files_to_rm: List[str]=[]) -> None:
""" tidy up in case of error """
log.error("corpus creation failed. Deleting corpus files")
for path in files_to_rm:
if fs.exists(path):
log.info("removing", path)
fs.rm(path)
raise err
def test_cp():
system.echo("Hello, world!", "/tmp/labm8.tmp")
assert ["Hello, world!"] == fs.read("/tmp/labm8.tmp")
# Cleanup any existing file.
fs.rm("/tmp/labm8.tmp.copy")
assert not fs.exists("/tmp/labm8.tmp.copy")
fs.cp("/tmp/labm8.tmp", "/tmp/labm8.tmp.copy")
assert fs.read("/tmp/labm8.tmp") == fs.read("/tmp/labm8.tmp.copy")
def test_cp_dir(self):
fs.rm("/tmp/labm8")
fs.rm("/tmp/labm8.copy")
fs.mkdir("/tmp/labm8/foo/bar")
self._test(False, fs.exists("/tmp/labm8.copy"))
fs.cp("/tmp/labm8/", "/tmp/labm8.copy")
self._test(True, fs.isdir("/tmp/labm8.copy"))
self._test(True, fs.isdir("/tmp/labm8.copy/foo"))
self._test(True, fs.isdir("/tmp/labm8.copy/foo/bar"))
def test_export_csv_file(self):
tmp = "/tmp/labm8.sql.csv"
self._test(None, self.db.export_csv("names", tmp))
self._test("first,last\n"
"David,Bowie\n"
"David,Brent\n"
"Joe,Bloggs\n",
open(tmp).read())
fs.rm(tmp)
def test_create_db_gh():
db_path = tests.data_path("db", "tmp.db", exists=False)
fs.rm(db_path)
dbutil.create_db(db_path, github=True)
assert fs.exists(db_path)
with pytest.raises(clgen.UserError):
dbutil.create_db(db_path, github=True)
def migrate_1_to_2(old):
"""
SkelCL database migration script.
Arguments:
old (SkelCLDatabase): The database to migrate
"""
# Create temporary database
fs.cp(old.path, "/tmp/omnitune.skelcl.migration.db")
tmp = _db.Database("/tmp/omnitune.skelcl.migration.db")
io.info("Migrating database to version 2.")
backup_path = old.path + ".1"
io.info("Creating backup of old database at '{0}'".format(backup_path))
fs.cp(old.path, backup_path)
# Update database version
tmp.drop_table("version")
tmp.create_table("version",
(("version", "integer"),))
tmp.execute("INSERT INTO version VALUES (2)")
# Rename table "data" to "datasets"
tmp.create_table("datasets",
(("id", "text primary key"),
("width", "integer"),
("height", "integer"),
("tin", "text"),
("tout", "text")))
tmp.execute("INSERT INTO datasets SELECT * FROM data")
tmp.drop_table("data")
# Rename column "scenarios.data" to "scenarios.dataset"
tmp.execute("ALTER TABLE scenarios RENAME TO old_scenarios")
tmp.create_table("scenarios",
(("id", "text primary key"),
("host", "text"),
("device", "text"),
("kernel", "text"),
("dataset", "text")))
tmp.execute("INSERT INTO scenarios SELECT * FROM old_scenarios")
tmp.drop_table("old_scenarios")
tmp.commit()
old_path = old.path
tmp_path = tmp.path
# Copy migrated database over the original one.
fs.cp(tmp_path, old_path)
fs.rm(tmp_path)
old.close()
tmp.close()
io.info("Migration completed.")
def test_cp_dir():
fs.rm("/tmp/labm8")
fs.rm("/tmp/labm8.copy")
fs.mkdir("/tmp/labm8/foo/bar")
assert not fs.exists("/tmp/labm8.copy")
fs.cp("/tmp/labm8/", "/tmp/labm8.copy")
assert fs.isdir("/tmp/labm8.copy")
assert fs.isdir("/tmp/labm8.copy/foo")
assert fs.isdir("/tmp/labm8.copy/foo/bar")
def test_must_exist(self):
system.echo("Hello, world!", "/tmp/labm8.must_exist.txt")
self.assertEqual(fs.must_exist("/tmp/labm8.must_exist.txt"),
"/tmp/labm8.must_exist.txt")
self.assertEqual(fs.must_exist("/tmp", "labm8.must_exist.txt"),
"/tmp/labm8.must_exist.txt")
with self.assertRaises(fs.File404):
fs.must_exist("/not/a/real/path")
fs.rm("/tmp/labm8.must_exist.txt")
def migrate_1_to_2(old):
"""
SkelCL database migration script.
Arguments:
old (SkelCLDatabase): The database to migrate
"""
# Create temporary database
fs.cp(old.path, "/tmp/omnitune.skelcl.migration.db")
tmp = _db.Database("/tmp/omnitune.skelcl.migration.db")
io.info("Migrating database to version 2.")
backup_path = old.path + ".1"
io.info("Creating backup of old database at '{0}'".format(backup_path))
fs.cp(old.path, backup_path)
# Update database version
tmp.drop_table("version")
tmp.create_table("version",
(("version", "integer"),))
tmp.execute("INSERT INTO version VALUES (2)")
# Rename table "data" to "datasets"
tmp.create_table("datasets",
(("id", "text primary key"),
("width", "integer"),
("height", "integer"),
("tin", "text"),
("tout", "text")))
tmp.execute("INSERT INTO datasets SELECT * FROM data")
tmp.drop_table("data")
# Rename column "scenarios.data" to "scenarios.dataset"
tmp.execute("ALTER TABLE scenarios RENAME TO old_scenarios")
tmp.create_table("scenarios",
(("id", "text primary key"),
("host", "text"),
("device", "text"),
("kernel", "text"),
("dataset", "text")))
tmp.execute("INSERT INTO scenarios SELECT * FROM old_scenarios")
tmp.drop_table("old_scenarios")
tmp.commit()
old_path = old.path
tmp_path = tmp.path
# Copy migrated database over the original one.
fs.cp(tmp_path, old_path)
fs.rm(tmp_path)
old.close()
tmp.close()
io.info("Migration completed.")
def migrate_3_to_4(old):
"""
SkelCL database migration script.
Arguments:
old (SkelCLDatabase): The database to migrate
"""
# Create temporary database
fs.rm("/tmp/omnitune.skelcl.migration.db")
tmp = _db.Database("/tmp/omnitune.skelcl.migration.db")
tmp.attach(old.path, "rhs")
io.info("Migrating database to version 4.")
backup_path = old.path + ".3"
io.info("Creating backup of old database at '{0}'".format(backup_path))
fs.cp(old.path, backup_path)
tables = [
"kernels", "kernel_lookup", "kernel_names", "devices", "device_lookup",
"datasets", "dataset_lookup", "scenarios", "params", "runtimes",
"runtime_stats", "oracle_params"
]
for table in tables:
io.info("Copying data from '{}' ...".format(table))
tmp.execute("INSERT INTO {} SELECT * FROM rhs.{}".format(table, table))
tmp_path = tmp.path
old_path = old.path
tmp.execute("VACUUM")
# Sanity checks
bad = False
for table in tables:
old_count = tmp.num_rows("rhs." + table)
tmp_count = tmp.num_rows(table)
if old_count != tmp_count:
io.error("Bad rows count:", old_count, tmp_count)
bad = True
if bad:
io.fatal("Failed sanity check, aborting.")
else:
io.info("Passed sanity check.")
# Copy migrated database over the original one.
fs.cp(tmp_path, old_path)
fs.rm(tmp_path)
old.close()
tmp.close()
io.info("Migration completed.")
def test_scp_user(self):
system.echo("Hello, world!", "/tmp/labm8.tmp")
self._test(["Hello, world!"], fs.read("/tmp/labm8.tmp"))
# Cleanup any existing file.
fs.rm("/tmp/labm8.tmp.copy")
self._test(False, fs.exists("/tmp/labm8.tmp.copy"))
# Perform scp.
system.scp("localhost", "/tmp/labm8.tmp", "/tmp/labm8.tmp.copy",
path="tests/bin", user="******")
self._test(fs.read("/tmp/labm8.tmp"), fs.read("/tmp/labm8.tmp.copy"))
def test_scp_user():
system.echo("Hello, world!", "/tmp/labm8.tmp")
assert ["Hello, world!"] == fs.read("/tmp/labm8.tmp")
# Cleanup any existing file.
fs.rm("/tmp/labm8.tmp.copy")
assert not fs.exists("/tmp/labm8.tmp.copy")
# Perform scp.
system.scp("localhost", "/tmp/labm8.tmp", "/tmp/labm8.tmp.copy",
path="labm8/data/test/bin", user="******")
assert fs.read("/tmp/labm8.tmp") == fs.read("/tmp/labm8.tmp.copy")
def test_create_drop_tables(self):
fs.rm("/tmp/labm8.sql")
_db = db.Database("/tmp/labm8.sql")
_db.drop_table("foo")
self._test(False, "foo" in _db.tables)
_db.create_table("foo", (("id", "integer primary key"),))
self._test(True, "foo" in _db.tables)
_db.drop_table("foo")
self._test(False, "foo" in _db.tables)
def test_empty_table(self):
fs.rm("/tmp/labm8.sql")
_db = db.Database("/tmp/labm8.sql")
_db.create_table("foo", (("id", "integer primary key"),))
_db.execute("INSERT INTO foo VALUES (1)")
_db.execute("INSERT INTO foo VALUES (2)")
_db.execute("INSERT INTO foo VALUES (3)")
self._test(3, _db.num_rows("foo"))
_db.empty_table("foo")
self._test(0, _db.num_rows("foo"))
def test_write_file():
d1 = {"a": "1", "b": "2"}
jsonutil.write_file("/tmp/labm8.write_file.json", d1)
d2 = jsonutil.read_file("/tmp/labm8.write_file.json")
fs.rm("/tmp/labm8.write_file.json")
jsonutil.write_file("/tmp/labm8.write_file2.json", d1)
d3 = jsonutil.read_file("/tmp/labm8.write_file2.json")
fs.rm("/tmp/labm8.write_file2.json")
assert d1 == d2 == d3
def test_GitHubRepo_Index_not_cloned(test_repo: github_repo.GitHubRepo):
"""Indexing a repo which is not cloned does nothing."""
fs.rm(test_repo.clone_dir)
assert not test_repo.IsIndexed()
test_repo.Index([
scrape_repos_pb2.ContentFilesImporterConfig(
source_code_pattern='.*\\.java',
preprocessor=[
"datasets.github.scrape_repos.preprocessors."
"extractors:JavaMethods"
]),
], multiprocessing.Pool(1))
assert not test_repo.IsIndexed()
def test_rmtrash(self):
system.echo("Hello, world!", "/tmp/labm8.tmp")
self.assertTrue(fs.isfile("/tmp/labm8.tmp"))
fs.rmtrash("/tmp/labm8.tmp")
self.assertFalse(fs.isfile("/tmp/labm8.tmp"))
fs.rmtrash("/tmp/labm8.tmp")
fs.rm("/tmp/labm8.tmp")
fs.rm("/tmp/labm8.dir")
fs.mkdir("/tmp/labm8.dir/foo/bar")
system.echo("Hello, world!", "/tmp/labm8.dir/foo/bar/baz")
self.assertTrue(fs.isfile("/tmp/labm8.dir/foo/bar/baz"))
fs.rmtrash("/tmp/labm8.dir")
self.assertFalse(fs.isfile("/tmp/labm8.dir/foo/bar/baz"))
self.assertFalse(fs.isfile("/tmp/labm8.dir/"))
def test_rm():
system.echo("Hello, world!", "/tmp/labm8.tmp")
assert fs.isfile("/tmp/labm8.tmp")
fs.rm("/tmp/labm8.tmp")
assert not fs.isfile("/tmp/labm8.tmp")
fs.rm("/tmp/labm8.tmp")
fs.rm("/tmp/labm8.tmp")
fs.rm("/tmp/labm8.dir")
fs.mkdir("/tmp/labm8.dir/foo/bar")
system.echo("Hello, world!", "/tmp/labm8.dir/foo/bar/baz")
assert fs.isfile("/tmp/labm8.dir/foo/bar/baz")
fs.rm("/tmp/labm8.dir")
assert not fs.isfile("/tmp/labm8.dir/foo/bar/baz")
assert not fs.isfile("/tmp/labm8.dir/")
def __delitem__(self, key: str) -> None:
"""
Delete cached file.
Arguments:
key (str): Key.
Raises:
Cache404: If file not in cache.
"""
assert (isinstance(key, string_types))
path = self._incache(self.keypath(key))
fs.rm(path)
def test_rm(self):
system.echo("Hello, world!", "/tmp/labm8.tmp")
self._test(True, fs.isfile("/tmp/labm8.tmp"))
fs.rm("/tmp/labm8.tmp")
self._test(False, fs.isfile("/tmp/labm8.tmp"))
fs.rm("/tmp/labm8.tmp")
fs.rm("/tmp/labm8.tmp")
fs.rm("/tmp/labm8.dir")
fs.mkdir("/tmp/labm8.dir/foo/bar")
system.echo("Hello, world!", "/tmp/labm8.dir/foo/bar/baz")
self._test(True, fs.isfile("/tmp/labm8.dir/foo/bar/baz"))
fs.rm("/tmp/labm8.dir")
self._test(False, fs.isfile("/tmp/labm8.dir/foo/bar/baz"))
self._test(False, fs.isfile("/tmp/labm8.dir/"))
def _init_error(err: Exception) -> None:
""" tidy up in case of error """
log.error("corpus creation failed. Deleting corpus files")
paths = [
fs.path(self.contentcache.path, "kernels.db"),
fs.path(self.cache.path, "corpus.txt"),
fs.path(self.cache.path, "tensor.npy"),
fs.path(self.cache.path, "atomizer.pkl")
]
for path in paths:
if fs.exists(path):
log.info("removing", path)
fs.rm(path)
raise err
def test_remove_bad_preprocessed(self):
fs.rm("tmp.db")
dbutil.create_db("tmp.db")
db = sqlite3.connect("tmp.db")
c = db.cursor()
# Create some data to test with:
c.execute("DELETE FROM PreprocessedFiles")
c.execute("INSERT INTO PreprocessedFiles VALUES(?,?,?)",
("id1", 0, "good output"))
c.execute("INSERT INTO PreprocessedFiles VALUES(?,?,?)",
("id2", 1, "bad output"))
c.execute("INSERT INTO PreprocessedFiles VALUES(?,?,?)",
("id3", 2, "ugly output"))
db.commit()
c.close()
# Check that data was written properly:
c = db.cursor()
c.execute("SELECT Count(*) FROM PreprocessedFiles")
count = c.fetchone()[0]
self.assertEqual(3, count)
db.close()
preprocess.remove_bad_preprocessed("tmp.db")
# Check that clean worked:
db = sqlite3.connect("tmp.db")
c = db.cursor()
c.execute("SELECT Count(*) FROM PreprocessedFiles")
count = c.fetchone()[0]
self.assertEqual(3, count)
c.execute("SELECT contents FROM PreprocessedFiles WHERE status=1 "
"OR status=2")
rows = c.fetchall()
print(rows)
self.assertTrue(all(not r == "[DELETED]" for r in rows))
# Clean up:
c.execute("DELETE FROM PreprocessedFiles")
db.commit()
c.close()
# Check that clean-up worked:
c = db.cursor()
c.execute("SELECT Count(*) FROM PreprocessedFiles")
count = c.fetchone()[0]
self.assertEqual(0, count)
fs.rm("tmp.db")
def test_write_file(self):
d1 = {
"a": "1",
"b": "2"
}
jsonutil.write_file("/tmp/labm8.write_file.json", d1)
d2 = jsonutil.read_file("/tmp/labm8.write_file.json")
fs.rm("/tmp/labm8.write_file.json")
jsonutil.write_file("/tmp/labm8.write_file2.json", d1)
d3 = jsonutil.read_file("/tmp/labm8.write_file2.json")
fs.rm("/tmp/labm8.write_file2.json")
self.assertEqual(d1, d2)
self.assertEqual(d1, d3)
def __delitem__(self, key):
"""
Delete cached file.
Arguments:
key: Key.
Raises:
KeyError: If file not in cache.
"""
path = self.keypath(key)
if fs.exists(path):
fs.rm(path)
else:
raise KeyError(key)
def __delitem__(self, key):
"""
Delete cached file.
Arguments:
key: Key.
Raises:
KeyError: If file not in cache.
"""
path = self.keypath(key)
if fs.exists(path):
fs.rm(path)
else:
raise KeyError(key)
def test_rmtrash():
with tempfile.NamedTemporaryFile(prefix='labm8_') as f:
assert fs.isfile(f.name)
fs.rmtrash(f.name)
assert not fs.isfile(f.name)
fs.rmtrash(f.name)
fs.rm(f.name)
with tempfile.TemporaryDirectory() as d:
fs.rm(d)
fs.mkdir(d, "foo/bar")
system.echo("Hello, world!", fs.path(d, "foo/bar/baz"))
assert fs.isfile(f, "foo/bar/baz")
fs.rmtrash(d)
assert not fs.isfile(d, "foo/bar/baz")
assert not fs.isdir(d)
def test_remove_bad_preprocessed(self):
fs.rm("tmp.db")
dbutil.create_db("tmp.db")
db = sqlite3.connect("tmp.db")
c = db.cursor()
# Create some data to test with:
c.execute("DELETE FROM PreprocessedFiles")
c.execute("INSERT INTO PreprocessedFiles VALUES(?,?,?)",
("id1", 0, "good output"))
c.execute("INSERT INTO PreprocessedFiles VALUES(?,?,?)",
("id2", 1, "bad output"))
c.execute("INSERT INTO PreprocessedFiles VALUES(?,?,?)",
("id3", 2, "ugly output"))
db.commit()
c.close()
# Check that data was written properly:
c = db.cursor()
c.execute("SELECT Count(*) FROM PreprocessedFiles")
count = c.fetchone()[0]
self.assertEqual(3, count)
db.close()
preprocess.remove_bad_preprocessed("tmp.db")
# Check that clean worked:
db = sqlite3.connect("tmp.db")
c = db.cursor()
c.execute("SELECT Count(*) FROM PreprocessedFiles")
count = c.fetchone()[0]
self.assertEqual(3, count)
c.execute("SELECT contents FROM PreprocessedFiles WHERE status=1 OR status=2")
rows = c.fetchall()
print(rows)
self.assertTrue(all(not r == "[DELETED]" for r in rows))
# Clean up:
c.execute("DELETE FROM PreprocessedFiles")
db.commit()
c.close()
# Check that clean-up worked:
c = db.cursor()
c.execute("SELECT Count(*) FROM PreprocessedFiles")
count = c.fetchone()[0]
self.assertEqual(0, count)
fs.rm("tmp.db")
def test_cp_over_dir():
fs.mkdir("/tmp/labm8.tmp.src")
system.echo("Hello, world!", "/tmp/labm8.tmp.src/foo")
fs.rm("/tmp/labm8.tmp.copy")
fs.mkdir("/tmp/labm8.tmp.copy")
assert fs.isdir("/tmp/labm8.tmp.src")
assert fs.isfile("/tmp/labm8.tmp.src/foo")
assert fs.isdir("/tmp/labm8.tmp.copy")
assert not fs.isfile("/tmp/labm8.tmp.copy/foo")
fs.cp("/tmp/labm8.tmp.src", "/tmp/labm8.tmp.copy/")
assert fs.isdir("/tmp/labm8.tmp.src")
assert fs.isfile("/tmp/labm8.tmp.src/foo")
assert fs.isdir("/tmp/labm8.tmp.copy")
assert fs.isfile("/tmp/labm8.tmp.copy/foo")
assert (fs.read("/tmp/labm8.tmp.src/foo") == fs.read(
"/tmp/labm8.tmp.copy/foo"))
def test_cp_over_dir(self):
fs.mkdir("/tmp/labm8.tmp.src")
system.echo("Hello, world!", "/tmp/labm8.tmp.src/foo")
fs.rm("/tmp/labm8.tmp.copy")
fs.mkdir("/tmp/labm8.tmp.copy")
self._test(True, fs.isdir("/tmp/labm8.tmp.src"))
self._test(True, fs.isfile("/tmp/labm8.tmp.src/foo"))
self._test(True, fs.isdir("/tmp/labm8.tmp.copy"))
self._test(False, fs.isfile("/tmp/labm8.tmp.copy/foo"))
fs.cp("/tmp/labm8.tmp.src", "/tmp/labm8.tmp.copy/")
self._test(True, fs.isdir("/tmp/labm8.tmp.src"))
self._test(True, fs.isfile("/tmp/labm8.tmp.src/foo"))
self._test(True, fs.isdir("/tmp/labm8.tmp.copy"))
self._test(True, fs.isfile("/tmp/labm8.tmp.copy/foo"))
self._test(fs.read("/tmp/labm8.tmp.src/foo"),
fs.read("/tmp/labm8.tmp.copy/foo"))
def RunTestcase(opencl_environment: env.OpenCLEnvironment,
testbed: deepsmith_pb2.Testbed,
testcase: deepsmith_pb2.Testcase,
cflags: typing.List[str]) -> deepsmith_pb2.Result:
"""Run a testcase."""
if testcase.toolchain != 'opencl':
raise ValueError(
f"Unsupported testcase toolchain: '{testcase.toolchain}'")
if testcase.harness.name != 'cldrive':
raise ValueError(
f"Unsupported testcase harness: '{testcase.harness.name}'")
result = deepsmith_pb2.Result()
result.testbed.CopyFrom(testbed)
platform_id, device_id = opencl_environment.ids()
driver = MakeDriver(
testcase, True if testbed.opts['opencl_opt'] == 'enabled' else False)
# MakeDriver() annotates the testcase, so we must only set the testcase field
# of the output result after we have called it.
result.testcase.CopyFrom(testcase)
# Get a temporary file to write and run the driver from.
with tempfile.NamedTemporaryFile(prefix='deepsmith_', delete=False) as f:
path = pathlib.Path(f.name)
try:
CompileDriver(driver, path, platform_id, device_id, cflags=cflags)
timeout = testcase.harness.opts.get('timeout_seconds', '60')
cmd = ['timeout', '-s9', timeout, f.name]
start_time = labdate.GetUtcMillisecondsNow()
proc = opencl_environment.Exec(cmd)
end_time = labdate.GetUtcMillisecondsNow()
# Build result message.
result.returncode = proc.returncode
result.outputs['stdout'] = proc.stdout
result.outputs['stderr'] = proc.stderr
runtime = result.profiling_events.add()
runtime.client = system.HOSTNAME
runtime.type = 'runtime'
runtime.duration_ms = int(
round((end_time - start_time).total_seconds() * 1000))
runtime.event_start_epoch_ms = labdate.MillisecondsTimestamp(
start_time)
result.outcome = GetResultOutcome(result)
except DriverCompilationError as e:
logging.warning('%s', e)
result.outcome = deepsmith_pb2.Result.UNKNOWN
finally:
fs.rm(path)
return result
def test_set_and_get():
fs.rm("/tmp/labm8-cache-set-and-get")
c = cache.FSCache("/tmp/labm8-cache-set-and-get")
# create file
system.echo("Hello, world!", "/tmp/labm8.testfile.txt")
# sanity check
assert fs.read("/tmp/labm8.testfile.txt") == ["Hello, world!"]
# insert file into cache
c['foobar'] = "/tmp/labm8.testfile.txt"
# file must be in cache
assert fs.isfile(c.keypath("foobar"))
# file must have been moved
assert not fs.isfile("/tmp/labm8.testfile.txt")
# check file contents
assert fs.read(c['foobar']) == ["Hello, world!"]
assert fs.read(c['foobar']) == fs.read(c.get('foobar'))
c.clear()
def test_remove_preprocessed(self):
tmpdb = 'test_remove_preprocessed.db'
fs.cp(tests.db_path('10-kernels-preprocessed'), tmpdb)
self.assertEqual(8, dbutil.num_good_kernels(tmpdb))
db = dbutil.connect(tmpdb)
self.assertFalse(dbutil.is_modified(db))
db.close()
dbutil.remove_preprocessed(tmpdb)
self.assertEqual(0, dbutil.num_good_kernels(tmpdb))
db = dbutil.connect(tmpdb)
self.assertTrue(dbutil.is_modified(db))
db.close()
fs.rm(tmpdb)
def drive(command: List[str], src: str) -> return_t:
""" invoke cldrive on source """
start_time = time()
with NamedTemporaryFile() as tmp:
tmp_path = tmp.name
cli = ['timeout', '-s9', '60', './libexec/co.sh', tmp_path] + command
process = Popen(cli, stdin=PIPE, stdout=PIPE, stderr=PIPE)
stdout, stderr = process.communicate(src.encode('utf-8'))
fs.rm(tmp_path)
stdout, stderr = stdout.decode('utf-8'), stderr.decode('utf-8')
runtime = time() - start_time
return return_t(
runtime=runtime, status=status_t(process.returncode),
stdout=stdout, stderr=stderr)
def test_ls_abspaths(self):
fs.cp("tests/data/testdir", "/tmp/testdir")
self._test(["/tmp/testdir/a",
"/tmp/testdir/b",
"/tmp/testdir/c",
"/tmp/testdir/d"],
fs.ls("/tmp/testdir", abspaths=True))
self._test(["/tmp/testdir/a",
"/tmp/testdir/b",
"/tmp/testdir/c",
"/tmp/testdir/c/e",
"/tmp/testdir/c/f",
"/tmp/testdir/c/f/f",
"/tmp/testdir/c/f/f/i",
"/tmp/testdir/c/f/h",
"/tmp/testdir/c/g",
"/tmp/testdir/d"],
fs.ls("/tmp/testdir", recursive=True, abspaths=True))
fs.rm("/tmp/testdir")
def sample_iteration(self, model: Model, quiet: bool = False) -> None:
"""
Run one sample iteration.
Arguments:
model (Model): CLgen model.
"""
assert (isinstance(model, Model))
cache = self.cache(model)
if self.kernel_opts.get("args", None):
start_text = serialize_argspec(self.kernel_opts["args"])
else:
start_text = "__kernel void A("
tmppath = fs.path(cache.path,
"sampler-{pid}.tmp.cl".format(pid=system.PID))
with open(tmppath, "w") as outfile:
opts = {
"output": outfile,
"num_samples": self.batch_size,
"temperature": self.kernel_opts.get("temperature", 1),
"max_length": self.kernel_opts.get("max_length", 10000),
"seed_text": start_text,
"quiet": quiet
}
model.sample(**opts)
sys.stdout.flush()
sys.stderr.flush()
fetch.process_sample_file(cache["kernels.db"],
tmppath,
max_kernel_len=opts["max_length"],
quiet=True)
if self.static_checker:
# TODO: Parse dynamic checker requirement
preprocess.preprocess_db(cache["kernels.db"])
fs.rm(tmppath)
def test_insert():
db_path = tests.data_path("db", "tmp.db", exists=False)
fs.rm(db_path)
dbutil.create_db(db_path)
db = dbutil.connect(db_path)
c = db.cursor()
assert dbutil.num_rows_in(db_path, "ContentFiles") == 0
dbutil.sql_insert_dict(c, "ContentFiles", {"id": "a", "contents": "foo"})
dbutil.sql_insert_dict(c, "PreprocessedFiles", {
"id": "a",
"status": 0,
"contents": "bar"
})
dbutil.sql_insert_dict(c, "PreprocessedFiles", {
"id": "b",
"status": 1,
"contents": "car"
})
db.commit()
c = db.cursor()
assert dbutil.num_rows_in(db_path, "ContentFiles") == 1
assert dbutil.num_rows_in(db_path, "PreprocessedFiles") == 2
assert dbutil.cc(db_path, "ContentFiles", "contents") == 3
assert dbutil.cc(db_path, "ContentFiles", "id") == 1
assert dbutil.lc(db_path, "ContentFiles", "contents") == 1
dbutil.remove_bad_preprocessed(db_path)
assert dbutil.num_rows_in(db_path, "ContentFiles") == 1
# remove_bad_preprocessed doesn't actually delete any rows, just
# replaces contents
assert dbutil.num_rows_in(db_path, "PreprocessedFiles") == 2
dbutil.remove_preprocessed(db_path)
assert dbutil.num_rows_in(db_path, "ContentFiles") == 1
assert dbutil.num_rows_in(db_path, "PreprocessedFiles") == 0
def test_ls_abspaths():
fs.cp("labm8/data/test/testdir", "/tmp/testdir")
assert fs.ls("/tmp/testdir", abspaths=True) == [
"/tmp/testdir/a",
"/tmp/testdir/b",
"/tmp/testdir/c",
"/tmp/testdir/d",
]
assert fs.ls("/tmp/testdir", recursive=True, abspaths=True) == [
"/tmp/testdir/a",
"/tmp/testdir/b",
"/tmp/testdir/c",
"/tmp/testdir/c/e",
"/tmp/testdir/c/f",
"/tmp/testdir/c/f/f",
"/tmp/testdir/c/f/f/i",
"/tmp/testdir/c/f/h",
"/tmp/testdir/c/g",
"/tmp/testdir/d",
]
fs.rm("/tmp/testdir")
def test_set_and_get(self):
fs.rm("/tmp/labm8-cache-set-and-get")
c = cache.FSCache("/tmp/labm8-cache-set-and-get")
# create file
system.echo("Hello, world!", "/tmp/labm8.testfile.txt")
# sanity check
self.assertEqual(fs.read("/tmp/labm8.testfile.txt"),
["Hello, world!"])
# insert file into cache
c['foobar'] = "/tmp/labm8.testfile.txt"
# file must be in cache
self.assertTrue(fs.isfile(c.keypath("foobar")))
# file must have been moved
self.assertFalse(fs.isfile("/tmp/labm8.testfile.txt"))
# check file contents
self.assertTrue(fs.read(c['foobar']), ["Hello, world!"])
self.assertEqual(fs.read(c['foobar']), fs.read(c.get('foobar')))
c.clear()
def build_with_clang(program: Union[CLgenProgram, CLSmithProgram],
clang: str) -> Tuple[int, float, str]:
with NamedTemporaryFile(prefix='buildaclang-', delete=False) as tmpfile:
src_path = tmpfile.name
try:
with open(src_path, "w") as outfile:
print(program.src, file=outfile)
cmd = ['timeout', '-s9', '60s', clang, '-cc1', '-xcl', src_path]
start_time = time()
process = subprocess.Popen(cmd,
universal_newlines=True,
stdout=subprocess.PIPE,
stderr=subprocess.PIPE)
_, stderr = process.communicate()
return process.returncode, time() - start_time, stderr.strip()
finally:
fs.rm(src_path)
def import_clgen_sample(session: session_t,
path: Path,
cl_launchable: bool = False,
harnesses: List[cldriveParams] = [],
delete: bool = False) -> None:
src = fs.read_file(path)
hash_ = crypto.sha1_str(src)
dupe = s.query(CLgenProgram).filter(CLgenProgram.hash == hash_).first()
if dupe:
print(f"warning: ignoring duplicate file {path}")
elif not len(src):
print(f"warning: ignoring empty file {path}")
else:
program = CLgenProgram(hash=hash_,
runtime=len(src) / CLGEN_INFERENCE_CPS,
src=src,
linecount=len(src.split('\n')),
cl_launchable=cl_launchable)
s.add(program)
s.commit()
# Make test harnesses, if required
if harnesses:
env = cldrive.make_env()
for params in harnesses:
testcase = get_or_create(s,
CLgenTestCase,
program_id=program.id,
params_id=params.id)
s.flush()
clgen_mkharness.mkharness(s, env, testcase)
if delete:
fs.rm(path)
def to_dist(self, distpath: str, author: str = None) -> str:
"""
Create a dist file.
Arguments:
distpath (str): Path to dist file.
author (str, optional): Author name.
Returns:
str: Path to generated distfile.
"""
outpath = fs.abspath(distpath) + ".tar.bz2"
if fs.exists(outpath):
raise DistError("file {} exists".format(outpath))
meta = self.meta
if author is not None:
meta["author"] = author
log.debug(clgen.format_json(meta))
try:
tar = tarfile.open(outpath, 'w:bz2')
# write meta
metapath = mktemp(prefix="clgen-", suffix=".json")
clgen.write_file(metapath, clgen.format_json(meta))
log.debug("metafile:", metapath)
# create tarball
tar.add(metapath, arcname="meta.json")
# pack contents:
for path in meta["contents"]:
abspath = fs.path(cache.ROOT, path)
log.verbose("packing", abspath)
tar.add(abspath, arcname=fs.path("contents", path))
# tidy up
fs.rm(metapath)
tar.close()
except Exception as e:
tar.close()
fs.rm(metapath)
fs.rm(outpath)
raise e
return outpath
def test_rm_glob(self):
fs.mkdir("/tmp/labm8.glob")
system.echo("Hello, world!", "/tmp/labm8.glob/1")
system.echo("Hello, world!", "/tmp/labm8.glob/2")
system.echo("Hello, world!", "/tmp/labm8.glob/abc")
fs.rm("/tmp/labm8.glob/a*", glob=False)
self._test(True, fs.isfile("/tmp/labm8.glob/1"))
self._test(True, fs.isfile("/tmp/labm8.glob/2"))
self._test(True, fs.isfile("/tmp/labm8.glob/abc"))
fs.rm("/tmp/labm8.glob/a*")
self._test(True, fs.isfile("/tmp/labm8.glob/1"))
self._test(True, fs.isfile("/tmp/labm8.glob/2"))
self._test(False, fs.isfile("/tmp/labm8.glob/abc"))
fs.rm("/tmp/labm8.glob/*")
self._test(False, fs.isfile("/tmp/labm8.glob/1"))
self._test(False, fs.isfile("/tmp/labm8.glob/2"))
self._test(False, fs.isfile("/tmp/labm8.glob/abc"))
def test_mv_no_dst(self):
system.echo("Hello, world!", "/tmp/labm8.tmp")
with self.assertRaises(IOError):
fs.mv("/tmp/labm8.tmp", "/not/a/real/path")
fs.rm("/tmp/labm8.tmp")
def test_finalise_tight(self):
self._mkplot()
viz.finalise("/tmp/labm8.png", tight=True)
self.assertTrue(fs.exists("/tmp/labm8.png"))
fs.rm("/tmp/labm8.png")
def migrate_0_to_1(old):
"""
SkelCL database migration script.
Arguments:
old (SkelCLDatabase): The database to migrate
"""
def get_source(checksum):
query = old.execute("SELECT source FROM kernels WHERE checksum = ?",
(checksum,))
return query.fetchone()[0]
def get_device_attr(device_id, name, count):
query = old.execute("SELECT * FROM devices WHERE name = ?",
(name,))
attr = query.fetchone()
# Splice into the new
newattr = (device_id, attr[0], count) + attr[2:]
return newattr
def process_row(tmp, row):
# Get column values from row.
host = row[0]
dev_name = row[1]
dev_count = row[2]
kern_checksum = row[3]
north = row[4]
south = row[5]
east = row[6]
west = row[7]
data_width = row[8]
data_height = row[9]
max_wg_size = row[10]
wg_c = row[11]
wg_r = row[12]
runtime = row[13]
type_in = "float"
type_out = "float"
# Lookup source code.
source = get_source(kern_checksum)
user_source = get_user_source(source)
kernel_id = hash_kernel(north, south, east, west, max_wg_size, source)
device_id = hash_device(dev_name, dev_count)
data_id = hash_data(data_width, data_height, type_in, type_out)
scenario_id = hash_scenario(host, device_id, kernel_id, data_id)
params_id = hash_workgroup_size(wg_c, wg_r)
device_attr = get_device_attr(device_id, dev_name, dev_count)
# Add database entries.
tmp.execute("INSERT OR IGNORE INTO kernels VALUES (?,?,?,?,?,?,?)",
(kernel_id,north,south,east,west,max_wg_size,user_source))
placeholders = ",".join(["?"] * len(device_attr))
tmp.execute("INSERT OR IGNORE INTO devices VALUES (" + placeholders + ")",
device_attr)
tmp.execute("INSERT OR IGNORE INTO data VALUES (?,?,?,?,?)",
(data_id, data_width, data_height, type_in, type_out))
tmp.execute("INSERT OR IGNORE INTO params VALUES (?,?,?)",
(params_id, wg_c, wg_r))
tmp.execute("INSERT OR IGNORE INTO scenarios VALUES (?,?,?,?,?)",
(scenario_id, host, device_id, kernel_id, data_id))
tmp.execute("INSERT INTO runtimes VALUES (?,?,?)",
(scenario_id, params_id, runtime))
# Create temporary database
tmp = _db.Database("/tmp/omnitune.skelcl.migration.db")
# Clear anything that's already in the database.
for table in tmp.tables:
tmp.drop_table(table)
io.info("Migrating database to version 1.")
backup_path = old.path + ".0"
io.info("Creating backup of old database at '{0}'".format(backup_path))
fs.cp(old.path, backup_path)
io.debug("Migration: creating tables ...")
# Create table: kernels
tmp.create_table("version",
(("version", "integer"),))
# Set database version
tmp.execute("INSERT INTO version VALUES (1)")
# Create table: kernels
tmp.create_table("kernels",
(("id", "text primary key"),
("north", "integer"),
("south", "integer"),
("east", "integer"),
("west", "integer"),
("max_wg_size", "integer"),
("source", "text")))
# Create table: devices
tmp.create_table("devices",
(("id", "text primary key"),
("name", "text"),
("count", "integer"),
("address_bits", "integer"),
("double_fp_config", "integer"),
("endian_little", "integer"),
("execution_capabilities", "integer"),
("extensions", "text"),
("global_mem_cache_size", "integer"),
("global_mem_cache_type", "integer"),
("global_mem_cacheline_size", "integer"),
("global_mem_size", "integer"),
("host_unified_memory", "integer"),
("image2d_max_height", "integer"),
("image2d_max_width", "integer"),
("image3d_max_depth", "integer"),
("image3d_max_height", "integer"),
("image3d_max_width", "integer"),
("image_support", "integer"),
("local_mem_size", "integer"),
("local_mem_type", "integer"),
("max_clock_frequency", "integer"),
("max_compute_units", "integer"),
("max_constant_args", "integer"),
("max_constant_buffer_size", "integer"),
("max_mem_alloc_size", "integer"),
("max_parameter_size", "integer"),
("max_read_image_args", "integer"),
("max_samplers", "integer"),
("max_work_group_size", "integer"),
("max_work_item_dimensions", "integer"),
("max_work_item_sizes_0", "integer"),
("max_work_item_sizes_1", "integer"),
("max_work_item_sizes_2", "integer"),
("max_write_image_args", "integer"),
("mem_base_addr_align", "integer"),
("min_data_type_align_size", "integer"),
("native_vector_width_char", "integer"),
("native_vector_width_double", "integer"),
("native_vector_width_float", "integer"),
("native_vector_width_half", "integer"),
("native_vector_width_int", "integer"),
("native_vector_width_long", "integer"),
("native_vector_width_short", "integer"),
("preferred_vector_width_char", "integer"),
("preferred_vector_width_double", "integer"),
("preferred_vector_width_float", "integer"),
("preferred_vector_width_half", "integer"),
("preferred_vector_width_int", "integer"),
("preferred_vector_width_long", "integer"),
("preferred_vector_width_short", "integer"),
("queue_properties", "integer"),
("single_fp_config", "integer"),
("type", "integer"),
("vendor", "text"),
("vendor_id", "text"),
("version", "text")))
# Create table: data
tmp.create_table("data",
(("id", "text primary key"),
("width", "integer"),
("height", "integer"),
("tin", "text"),
("tout", "text")))
# Create table: params
tmp.create_table("params",
(("id", "text primary key"),
("wg_c", "integer"),
("wg_r", "integer")))
# Create table: scenarios
tmp.create_table("scenarios",
(("id", "text primary key"),
("host", "text"),
("device", "text"),
("kernel", "text"),
("data", "text")))
# Create table: runtimes
tmp.create_table("runtimes",
(("scenario", "text"),
("params", "text"),
("runtime", "real")))
i = 0
for row in old.execute("SELECT * from runtimes"):
process_row(tmp, row)
i += 1
if not i % 2500:
io.debug("Processed", i, "rows ...")
if not i % 5000:
tmp.commit()
tmp.commit()
old_path = old.path
tmp_path = tmp.path
# Copy migrated database over the original one.
fs.cp(tmp_path, old_path)
fs.rm(tmp_path)
old.close()
tmp.close()
io.info("Migration completed.")
def test_constructor_schema(self):
fs.rm("/tmp/labm8.sql")
_db = db.Database("/tmp/labm8.sql", {
"foo": (("id", "integer"), ("prop", "text"))
})
self._test(["foo"], _db.tables)
def migrate_3_to_4(old):
"""
SkelCL database migration script.
Arguments:
old (SkelCLDatabase): The database to migrate
"""
# Create temporary database
fs.rm("/tmp/omnitune.skelcl.migration.db")
tmp = _db.Database("/tmp/omnitune.skelcl.migration.db")
tmp.attach(old.path, "rhs")
io.info("Migrating database to version 4.")
backup_path = old.path + ".3"
io.info("Creating backup of old database at '{0}'".format(backup_path))
fs.cp(old.path, backup_path)
tables = [
"kernels",
"kernel_lookup",
"kernel_names",
"devices",
"device_lookup",
"datasets",
"dataset_lookup",
"scenarios",
"params",
"runtimes",
"runtime_stats",
"oracle_params"
]
for table in tables:
io.info("Copying data from '{}' ...".format(table))
tmp.execute("INSERT INTO {} SELECT * FROM rhs.{}".format(table, table))
tmp_path = tmp.path
old_path = old.path
tmp.execute("VACUUM")
# Sanity checks
bad = False
for table in tables:
old_count = tmp.num_rows("rhs." + table)
tmp_count = tmp.num_rows(table)
if old_count != tmp_count:
io.error("Bad rows count:", old_count, tmp_count)
bad = True
if bad:
io.fatal("Failed sanity check, aborting.")
else:
io.info("Passed sanity check.")
# Copy migrated database over the original one.
fs.cp(tmp_path, old_path)
fs.rm(tmp_path)
old.close()
tmp.close()
io.info("Migration completed.")
def migrate_2_to_3(old):
"""
SkelCL database migration script.
Arguments:
old (SkelCLDatabase): The database to migrate
"""
def _old_kernel2new(old_id):
kernel = old.execute("SELECT north,south,east,west,max_wg_size,source "
"FROM kernels WHERE id=?",
(old_id,)).fetchone()
if kernel:
return tmp.kernel_id(*kernel)
def _old_scenario2new(old_id):
device, old_kernel, dataset = old.execute("SELECT device,kernel,dataset "
"FROM scenarios WHERE id=?",
(old_id,)).fetchone()
kernel = _old_kernel2new(old_kernel)
return tmp.scenario_id(device, kernel, dataset)
# TODO: Un-comment out code!
# Create temporary database
fs.rm("/tmp/omnitune.skelcl.migration.db")
tmp = _db.Database("/tmp/omnitune.skelcl.migration.db")
tmp.attach(old.path, "rhs")
io.info("Migrating database to version 3.")
backup_path = old.path + ".2"
io.info("Creating backup of old database at '{0}'".format(backup_path))
fs.cp(old.path, backup_path)
tmp_path = tmp.path
old_path = old.path
tmp.run("create_tables")
# Populate feature and lookup tables.
for row in old.execute("SELECT * FROM devices"):
features = row[1:]
id = hash_device(*features)
io.debug("Features extracted for device", id)
row = (id,) + features
tmp.execute("INSERT INTO devices VALUES " +
placeholders(*row), row)
row = (features[0], features[1], id)
tmp.execute("INSERT INTO device_lookup VALUES " +
placeholders(*row), row)
tmp.commit()
for row in old.execute("SELECT * FROM kernels"):
args = row[1:]
tmp.kernel_id(*args)
for row in old.execute("SELECT * FROM datasets"):
features = row[1:]
id = hash_dataset(*features)
io.debug("Features extracted for dataset", id)
row = (id,) + features
tmp.execute("INSERT INTO datasets VALUES " +
placeholders(*row), row)
row = features + (id,)
tmp.execute("INSERT INTO dataset_lookup VALUES " +
placeholders(*row), row)
tmp.commit()
# Populate kernel_names table.
for row in old.execute("SELECT * FROM kernel_names"):
old_id = row[0]
synthetic, name = row[1:]
kernel = _old_kernel2new(old_id)
if kernel:
row = (kernel, synthetic, name)
tmp.execute("INSERT OR IGNORE INTO kernel_names VALUES " +
placeholders(*row), row)
tmp.commit()
# Populate scenarios table.
for row in old.execute("SELECT * FROM scenarios"):
old_id, _, device, old_kernel, dataset = row
kernel = _old_kernel2new(old_kernel)
new_id = hash_scenario(device, kernel, dataset)
row = (new_id, device, kernel, dataset)
tmp.execute("INSERT OR IGNORE INTO scenarios VALUES " +
placeholders(*row), row)
tmp.commit()
# Populate params table.
tmp.execute("INSERT INTO params SELECT * from rhs.params")
tmp.commit()
scenario_replacements = {
row[0]: _old_scenario2new(row[0])
for row in old.execute("SELECT * FROM scenarios")
}
tmp.execute("INSERT INTO runtimes SELECT * from rhs.runtimes")
for old_id, new_id in scenario_replacements.iteritems():
io.info("Runtimes", old_id, "->", new_id)
tmp.execute("UPDATE runtimes SET scenario=? WHERE scenario=?",
(new_id, old_id))
tmp.commit()
# Sanity checks
bad = False
for row in tmp.execute("SELECT DISTINCT scenario FROM runtimes"):
count = tmp.execute("SELECT Count(*) FROM scenarios WHERE id=?",
(row[0],)).fetchone()[0]
if count != 1:
io.error("Bad scenario count:", row[0], count)
bad = True
if bad:
io.fatal("Failed sanity check, aborting.")
else:
io.info("Passed sanity check.")
# Copy migrated database over the original one.
fs.cp(tmp_path, old_path)
fs.rm(tmp_path)
old.close()
tmp.close()
io.info("Migration completed.")
def test_mkdir(self):
fs.rm("/tmp/labm8.dir")
self._test(False, fs.isdir("/tmp/labm8.dir"))
fs.mkdir("/tmp/labm8.dir")
self._test(True, fs.isdir("/tmp/labm8.dir"))
def test_ls_empty_dir(self):
fs.mkdir("/tmp/labm8.empty")
self._test([], fs.ls("/tmp/labm8.empty"))
fs.rm("/tmp/labm8.empty")
def test_mkopen(self):
fs.rm("/tmp/labm8.dir")
self._test(False, fs.isdir("/tmp/labm8.dir/"))
f = fs.mkopen("/tmp/labm8.dir/foo", "w")
self._test(True, fs.isdir("/tmp/labm8.dir/"))
f.close()
def main():
db = _db.Database(experiment.ORACLE_PATH)
ml.start()
# Delete any old stuff.
fs.rm(experiment.IMG_ROOT + "/*")
fs.rm(experiment.TAB_ROOT + "/*")
# Make directories
fs.mkdir(experiment.TAB_ROOT)
fs.mkdir(fs.path(experiment.IMG_ROOT, "scenarios/bars"))
fs.mkdir(fs.path(experiment.IMG_ROOT, "scenarios/heatmap"))
fs.mkdir(fs.path(experiment.IMG_ROOT, "scenarios/trisurf"))
fs.mkdir(fs.path(experiment.IMG_ROOT, "coverage/devices"))
fs.mkdir(fs.path(experiment.IMG_ROOT, "coverage/kernels"))
fs.mkdir(fs.path(experiment.IMG_ROOT, "coverage/datasets"))
fs.mkdir(fs.path(experiment.IMG_ROOT, "safety/devices"))
fs.mkdir(fs.path(experiment.IMG_ROOT, "safety/kernels"))
fs.mkdir(fs.path(experiment.IMG_ROOT, "safety/datasets"))
fs.mkdir(fs.path(experiment.IMG_ROOT, "oracle/devices"))
fs.mkdir(fs.path(experiment.IMG_ROOT, "oracle/kernels"))
fs.mkdir(fs.path(experiment.IMG_ROOT, "oracle/datasets"))
visualise.pie(db.num_scenarios_by_device,
fs.path(experiment.IMG_ROOT, "num_sceanrios_by_device"))
visualise.pie(db.num_runtime_stats_by_device,
fs.path(experiment.IMG_ROOT, "num_runtime_stats_by_device"))
visualise.pie(db.num_scenarios_by_dataset,
fs.path(experiment.IMG_ROOT, "num_sceanrios_by_dataset"))
visualise.pie(db.num_runtime_stats_by_dataset,
fs.path(experiment.IMG_ROOT, "num_runtime_stats_by_dataset"))
visualise.pie(db.num_runtime_stats_by_kernel,
fs.path(experiment.IMG_ROOT, "num_runtime_stats_by_kernel"))
visualise.pie(db.num_runtime_stats_by_kernel,
fs.path(experiment.IMG_ROOT, "num_runtime_stats_by_kernel"))
# Per-scenario plots
for row in db.scenario_properties:
scenario,device,kernel,north,south,east,west,max_wgsize,width,height,tout = row
title = ("{device}: {kernel}[{n},{s},{e},{w}]\n"
"{width} x {height} {type}s"
.format(device=text.truncate(device, 18), kernel=kernel,
n=north, s=south, e=east, w=west,
width=width, height=height, type=tout))
output = fs.path(experiment.IMG_ROOT,
"scenarios/heatmap/{id}.png".format(id=scenario))
space = _space.ParamSpace.from_dict(db.perf_scenario(scenario))
max_c = min(25, len(space.c))
max_r = min(25, len(space.r))
space.reshape(max_c=max_c, max_r=max_r)
# Heatmaps.
mask = _space.ParamSpace(space.c, space.r)
for j in range(len(mask.r)):
for i in range(len(mask.c)):
if space.matrix[j][i] == 0:
r, c = space.r[j], space.c[i]
# TODO: Get values from refused_params table.
if r * c >= max_wgsize:
# Illegal
mask.matrix[j][i] = -1
else:
# Refused
db.execute("INSERT OR IGNORE INTO refused_params VALUES(?,?)",
(scenario, hash_params(c, r)))
space.matrix[j][i] = -1
mask.matrix[j][i] = 1
db.commit()
new_order = list(reversed(range(space.matrix.shape[0])))
data = space.matrix[:][new_order]
figsize=(12,6)
_, ax = plt.subplots(1, 2, figsize=figsize, sharey=True)
sns.heatmap(data, ax=ax[0], vmin=-1, vmax=1,
xticklabels=space.c,
yticklabels=list(reversed(space.r)), square=True)
ax[0].set_title(title)
new_order = list(reversed(range(mask.matrix.shape[0])))
data = mask.matrix[:][new_order]
sns.heatmap(data, ax=ax[1], vmin=-1, vmax=1,
xticklabels=space.c,
yticklabels=list(reversed(space.r)), square=True)
# Set labels.
ax[0].set_ylabel("Rows")
ax[0].set_xlabel("Columns")
ax[1].set_ylabel("Rows")
ax[1].set_xlabel("Columns")
# plt.tight_layout()
# plt.gcf().set_size_inches(*figsize, dpi=300)
viz.finalise(output)
# 3D bars.
output = fs.path(experiment.IMG_ROOT,
"scenarios/bars/{id}.png".format(id=scenario))
space.bar3d(output=output, title=title, zlabel="Performance",
rotation=45)
# Trisurfs.
output = fs.path(experiment.IMG_ROOT,
"scenarios/trisurf/{id}.png".format(id=scenario))
space.trisurf(output=output, title=title, zlabel="Performance",
rotation=45)
#####################
# ML Visualisations #
#####################
#features_tab(db, experiment.TAB_ROOT)
visualise_classification_job(db, "xval")
visualise_classification_job(db, "arch")
visualise_classification_job(db, "xval_real")
visualise_classification_job(db, "synthetic_real")
# Runtime regression accuracy.
visualise_regression_job(db, "xval")
visualise_regression_job(db, "arch")
visualise_regression_job(db, "xval_real")
visualise_regression_job(db, "synthetic_real")
# Whole-dataset plots
visualise.runtimes_variance(db, fs.path(experiment.IMG_ROOT,
"runtime_variance.png"),
min_samples=30)
visualise.num_samples(db, fs.path(experiment.IMG_ROOT,
"num_samples.png"))
visualise.runtimes_range(db, fs.path(experiment.IMG_ROOT,
"runtimes_range.png"))
visualise.max_speedups(db, fs.path(experiment.IMG_ROOT,
"max_speedups.png"))
visualise.kernel_performance(db, fs.path(experiment.IMG_ROOT,
"kernel_performance.png"))
visualise.device_performance(db, fs.path(experiment.IMG_ROOT,
"device_performance.png"))
visualise.dataset_performance(db, fs.path(experiment.IMG_ROOT,
"dataset_performance.png"))
visualise.num_params_vs_accuracy(db, fs.path(experiment.IMG_ROOT,
"num_params_vs_accuracy.png"))
visualise.performance_vs_coverage(db,
fs.path(experiment.IMG_ROOT,
"performance_vs_coverage.png"))
visualise.performance_vs_max_wgsize(
db, fs.path(experiment.IMG_ROOT, "performance_vs_max_wgsize.png")
)
visualise.performance_vs_wgsize(db, fs.path(experiment.IMG_ROOT,
"performance_vs_wgsize.png"))
visualise.performance_vs_wg_c(db, fs.path(experiment.IMG_ROOT,
"performance_vs_wg_c.png"))
visualise.performance_vs_wg_r(db, fs.path(experiment.IMG_ROOT,
"performance_vs_wg_r.png"))
visualise.max_wgsizes(db, fs.path(experiment.IMG_ROOT, "max_wgsizes.png"))
visualise.oracle_speedups(db, fs.path(experiment.IMG_ROOT,
"oracle_speedups.png"))
visualise.coverage(db,
fs.path(experiment.IMG_ROOT, "coverage/coverage.png"))
visualise.safety(db, fs.path(experiment.IMG_ROOT, "safety/safety.png"))
visualise.oracle_wgsizes(db, fs.path(experiment.IMG_ROOT, "oracle/all.png"))
# Per-device plots
for i,device in enumerate(db.devices):
where = ("scenario IN "
"(SELECT id from scenarios WHERE device='{0}')"
.format(device))
output = fs.path(experiment.IMG_ROOT,
"coverage/devices/{0}.png".format(i))
visualise.coverage(db, output=output, where=where, title=device)
output = fs.path(experiment.IMG_ROOT,
"safety/devices/{0}.png".format(i))
visualise.safety(db, output, where=where, title=device)
output = fs.path(experiment.IMG_ROOT,
"oracle/devices/{0}.png".format(i))
visualise.oracle_wgsizes(db, output, where=where, title=device)
where = ("scenario IN (\n"
" SELECT id from scenarios WHERE device='{0}'\n"
") AND scenario IN (\n"
" SELECT id FROM scenarios WHERE kernel IN (\n"
" SELECT id FROM kernel_names WHERE synthetic=0\n"
" )\n"
")"
.format(device))
output = fs.path(experiment.IMG_ROOT,
"coverage/devices/{0}_real.png".format(i))
visualise.coverage(db, output=output, where=where,
title=device + ", real")
output = fs.path(experiment.IMG_ROOT,
"safety/devices/{0}_real.png".format(i))
visualise.safety(db, output, where=where,
title=device + ", real")
output = fs.path(experiment.IMG_ROOT,
"oracle/devices/{0}_real.png".format(i))
visualise.oracle_wgsizes(db, output, where=where,
title=device + ", real")
where = ("scenario IN (\n"
" SELECT id from scenarios WHERE device='{0}'\n"
") AND scenario IN (\n"
" SELECT id FROM scenarios WHERE kernel IN (\n"
" SELECT id FROM kernel_names WHERE synthetic=1\n"
" )\n"
")"
.format(device))
output = fs.path(experiment.IMG_ROOT,
"coverage/devices/{0}_synthetic.png".format(i))
visualise.coverage(db, output=output, where=where,
title=device + ", synthetic")
output = fs.path(experiment.IMG_ROOT,
"safety/devices/{0}_synthetic.png".format(i))
visualise.safety(db, output, where=where,
title=device + ", synthetic")
output = fs.path(experiment.IMG_ROOT,
"oracle/devices/{0}_synthetic.png".format(i))
visualise.oracle_wgsizes(db, output, where=where,
title=device + ", synthetic")
# Per-kernel plots
for kernel,ids in db.lookup_named_kernels().iteritems():
id_wrapped = ['"' + id + '"' for id in ids]
where = ("scenario IN "
"(SELECT id from scenarios WHERE kernel IN ({0}))"
.format(",".join(id_wrapped)))
output = fs.path(experiment.IMG_ROOT,
"coverage/kernels/{0}.png".format(kernel))
visualise.coverage(db, output=output, where=where, title=kernel)
output = fs.path(experiment.IMG_ROOT,
"safety/kernels/{0}.png".format(kernel))
visualise.safety(db, output=output, where=where, title=kernel)
output = fs.path(experiment.IMG_ROOT,
"oracle/kernels/{0}.png".format(kernel))
visualise.safety(db, output=output, where=where, title=kernel)
# Per-dataset plots
for i,dataset in enumerate(db.datasets):
where = ("scenario IN "
"(SELECT id from scenarios WHERE dataset='{0}')"
.format(dataset))
output = fs.path(experiment.IMG_ROOT,
"coverage/datasets/{0}.png".format(i))
visualise.coverage(db, output, where=where, title=dataset)
output = fs.path(experiment.IMG_ROOT,
"safety/datasets/{0}.png".format(i))
visualise.safety(db, output, where=where, title=dataset)
output = fs.path(experiment.IMG_ROOT,
"oracle/datasets/{0}.png".format(i))
visualise.safety(db, output, where=where, title=dataset)
ml.stop()
def test_finalise_figsize(self):
self._mkplot()
viz.finalise("/tmp/labm8.png", figsize=(10, 5))
self.assertTrue(fs.exists("/tmp/labm8.png"))
fs.rm("/tmp/labm8.png")