aboutsummaryrefslogtreecommitdiffstats
path: root/bitbake/lib/bb/siggen.py
blob: 6a903b2fa59056863989d9a1feb88fc1a351c384 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
import hashlib
import logging
import os
import re
import tempfile
import bb.data
from bb.checksum import FileChecksumCache

logger = logging.getLogger('BitBake.SigGen')

try:
    import cPickle as pickle
except ImportError:
    import pickle
    logger.info('Importing cPickle failed.  Falling back to a very slow implementation.')

def init(d):
    siggens = [obj for obj in globals().itervalues()
                      if type(obj) is type and issubclass(obj, SignatureGenerator)]

    desired = d.getVar("BB_SIGNATURE_HANDLER", True) or "noop"
    for sg in siggens:
        if desired == sg.name:
            return sg(d)
            break
    else:
        logger.error("Invalid signature generator '%s', using default 'noop'\n"
                     "Available generators: %s", desired,
                     ', '.join(obj.name for obj in siggens))
        return SignatureGenerator(d)

class SignatureGenerator(object):
    """
    """
    name = "noop"

    def __init__(self, data):
        self.basehash = {}
        self.taskhash = {}
        self.runtaskdeps = {}
        self.file_checksum_values = {}
        self.taints = {}

    def finalise(self, fn, d, varient):
        return

    def get_taskhash(self, fn, task, deps, dataCache):
        return "0"

    def writeout_file_checksum_cache(self):
        """Write/update the file checksum cache onto disk"""
        return

    def stampfile(self, stampbase, file_name, taskname, extrainfo):
        return ("%s.%s.%s" % (stampbase, taskname, extrainfo)).rstrip('.')

    def stampcleanmask(self, stampbase, file_name, taskname, extrainfo):
        return ("%s.%s.%s" % (stampbase, taskname, extrainfo)).rstrip('.')

    def dump_sigtask(self, fn, task, stampbase, runtime):
        return

    def invalidate_task(self, task, d, fn):
        bb.build.del_stamp(task, d, fn)

    def dump_sigs(self, dataCache, options):
        return

    def get_taskdata(self):
        return (self.runtaskdeps, self.taskhash, self.file_checksum_values, self.taints, self.basehash)

    def set_taskdata(self, data):
        self.runtaskdeps, self.taskhash, self.file_checksum_values, self.taints, self.basehash = data

class SignatureGeneratorBasic(SignatureGenerator):
    """
    """
    name = "basic"

    def __init__(self, data):
        self.basehash = {}
        self.taskhash = {}
        self.taskdeps = {}
        self.runtaskdeps = {}
        self.file_checksum_values = {}
        self.taints = {}
        self.gendeps = {}
        self.lookupcache = {}
        self.pkgnameextract = re.compile("(?P<fn>.*)\..*")
        self.basewhitelist = set((data.getVar("BB_HASHBASE_WHITELIST", True) or "").split())
        self.taskwhitelist = None
        self.init_rundepcheck(data)
        checksum_cache_file = data.getVar("BB_HASH_CHECKSUM_CACHE_FILE", True)
        if checksum_cache_file:
            self.checksum_cache = FileChecksumCache()
            self.checksum_cache.init_cache(data, checksum_cache_file)
        else:
            self.checksum_cache = None

    def init_rundepcheck(self, data):
        self.taskwhitelist = data.getVar("BB_HASHTASK_WHITELIST", True) or None
        if self.taskwhitelist:
            self.twl = re.compile(self.taskwhitelist)
        else:
            self.twl = None

    def _build_data(self, fn, d):

        tasklist, gendeps, lookupcache = bb.data.generate_dependencies(d)

        taskdeps = {}
        basehash = {}

        for task in tasklist:
            data = lookupcache[task]

            if data is None:
                bb.error("Task %s from %s seems to be empty?!" % (task, fn))
                data = ''

            gendeps[task] -= self.basewhitelist
            newdeps = gendeps[task]
            seen = set()
            while newdeps:
                nextdeps = newdeps
                seen |= nextdeps
                newdeps = set()
                for dep in nextdeps:
                    if dep in self.basewhitelist:
                        continue
                    gendeps[dep] -= self.basewhitelist
                    newdeps |= gendeps[dep]
                newdeps -= seen

            alldeps = sorted(seen)
            for dep in alldeps:
                data = data + dep
                var = lookupcache[dep]
                if var is not None:
                    data = data + str(var)
            datahash = hashlib.md5(data).hexdigest()
            k = fn + "." + task
            if k in self.basehash and self.basehash[k] != datahash:
                bb.error("When reparsing %s, the basehash value changed from %s to %s. The metadata is not deterministic and this needs to be fixed." % (k, self.basehash[k], datahash))
            self.basehash[k] = datahash
            taskdeps[task] = alldeps

        self.taskdeps[fn] = taskdeps
        self.gendeps[fn] = gendeps
        self.lookupcache[fn] = lookupcache

        return taskdeps

    def finalise(self, fn, d, variant):

        if variant:
            fn = "virtual:" + variant + ":" + fn

        try:
            taskdeps = self._build_data(fn, d)
        except:
            bb.warn("Error during finalise of %s" % fn)
            raise

        #Slow but can be useful for debugging mismatched basehashes
        #for task in self.taskdeps[fn]:
        #    self.dump_sigtask(fn, task, d.getVar("STAMP", True), False)

        for task in taskdeps:
            d.setVar("BB_BASEHASH_task-%s" % task, self.basehash[fn + "." + task])

    def rundep_check(self, fn, recipename, task, dep, depname, dataCache):
        # Return True if we should keep the dependency, False to drop it
        # We only manipulate the dependencies for packages not in the whitelist
        if self.twl and not self.twl.search(recipename):
            # then process the actual dependencies
            if self.twl.search(depname):
                return False
        return True

    def read_taint(self, fn, task, stampbase):
        taint = None
        try:
            with open(stampbase + '.' + task + '.taint', 'r') as taintf:
                taint = taintf.read()
        except IOError:
            pass
        return taint

    def get_taskhash(self, fn, task, deps, dataCache):
        k = fn + "." + task
        data = dataCache.basetaskhash[k]
        self.basehash[k] = data
        self.runtaskdeps[k] = []
        self.file_checksum_values[k] = []
        recipename = dataCache.pkg_fn[fn]

        for dep in sorted(deps, key=clean_basepath):
            depname = dataCache.pkg_fn[self.pkgnameextract.search(dep).group('fn')]
            if not self.rundep_check(fn, recipename, task, dep, depname, dataCache):
                continue
            if dep not in self.taskhash:
                bb.fatal("%s is not in taskhash, caller isn't calling in dependency order?", dep)
            data = data + self.taskhash[dep]
            self.runtaskdeps[k].append(dep)

        if task in dataCache.file_checksums[fn]:
            if self.checksum_cache:
                checksums = self.checksum_cache.get_checksums(dataCache.file_checksums[fn][task], recipename)
            else:
                checksums = bb.fetch2.get_file_checksums(dataCache.file_checksums[fn][task], recipename)
            for (f,cs) in checksums:
                self.file_checksum_values[k].append((f,cs))
                if cs:
                    data = data + cs

        taskdep = dataCache.task_deps[fn]
        if 'nostamp' in taskdep and task in taskdep['nostamp']:
            # Nostamp tasks need an implicit taint so that they force any dependent tasks to run
            import uuid
            taint = str(uuid.uuid4())
            data = data + taint
            self.taints[k] = "nostamp:" + taint

        taint = self.read_taint(fn, task, dataCache.stamp[fn])
        if taint:
            data = data + taint
            self.taints[k] = taint
            logger.warn("%s is tainted from a forced run" % k)

        h = hashlib.md5(data).hexdigest()
        self.taskhash[k] = h
        #d.setVar("BB_TASKHASH_task-%s" % task, taskhash[task])
        return h

    def writeout_file_checksum_cache(self):
        """Write/update the file checksum cache onto disk"""
        if self.checksum_cache:
            self.checksum_cache.save_extras()
            self.checksum_cache.save_merge()
        else:
            bb.fetch2.fetcher_parse_save()
            bb.fetch2.fetcher_parse_done()

    def dump_sigtask(self, fn, task, stampbase, runtime):

        k = fn + "." + task
        referencestamp = stampbase
        if isinstance(runtime, str) and runtime.startswith("customfile"):
            sigfile = stampbase
            referencestamp = runtime[11:]
        elif runtime and k in self.taskhash:
            sigfile = stampbase + "." + task + ".sigdata" + "." + self.taskhash[k]
        else:
            sigfile = stampbase + "." + task + ".sigbasedata" + "." + self.basehash[k]

        bb.utils.mkdirhier(os.path.dirname(sigfile))

        data = {}
        data['task'] = task
        data['basewhitelist'] = self.basewhitelist
        data['taskwhitelist'] = self.taskwhitelist
        data['taskdeps'] = self.taskdeps[fn][task]
        data['basehash'] = self.basehash[k]
        data['gendeps'] = {}
        data['varvals'] = {}
        data['varvals'][task] = self.lookupcache[fn][task]
        for dep in self.taskdeps[fn][task]:
            if dep in self.basewhitelist:
                continue
            data['gendeps'][dep] = self.gendeps[fn][dep]
            data['varvals'][dep] = self.lookupcache[fn][dep]

        if runtime and k in self.taskhash:
            data['runtaskdeps'] = self.runtaskdeps[k]
            data['file_checksum_values'] = [(os.path.basename(f), cs) for f,cs in self.file_checksum_values[k]]
            data['runtaskhashes'] = {}
            for dep in data['runtaskdeps']:
                data['runtaskhashes'][dep] = self.taskhash[dep]
            data['taskhash'] = self.taskhash[k]

        taint = self.read_taint(fn, task, referencestamp)
        if taint:
            data['taint'] = taint

        if runtime and k in self.taints:
            if 'nostamp:' in self.taints[k]:
                data['taint'] = self.taints[k]

        computed_basehash = calc_basehash(data)
        if computed_basehash != self.basehash[k]:
            bb.error("Basehash mismatch %s versus %s for %s" % (computed_basehash, self.basehash[k], k))
        if runtime and k in self.taskhash:
            computed_taskhash = calc_taskhash(data)
            if computed_taskhash != self.taskhash[k]:
                bb.error("Taskhash mismatch %s versus %s for %s" % (computed_taskhash, self.taskhash[k], k))
                sigfile = sigfile.replace(self.taskhash[k], computed_taskhash)

        fd, tmpfile = tempfile.mkstemp(dir=os.path.dirname(sigfile), prefix="sigtask.")
        try:
            with os.fdopen(fd, "wb") as stream:
                p = pickle.dump(data, stream, -1)
                stream.flush()
            os.chmod(tmpfile, 0664)
            os.rename(tmpfile, sigfile)
        except (OSError, IOError) as err:
            try:
                os.unlink(tmpfile)
            except OSError:
                pass
            raise err

    def dump_sigs(self, dataCache, options):
        for fn in self.taskdeps:
            for task in self.taskdeps[fn]:
                k = fn + "." + task
                if k not in self.taskhash:
                    continue
                if dataCache.basetaskhash[k] != self.basehash[k]:
                    bb.error("Bitbake's cached basehash does not match the one we just generated (%s)!" % k)
                    bb.error("The mismatched hashes were %s and %s" % (dataCache.basetaskhash[k], self.basehash[k]))
                self.dump_sigtask(fn, task, dataCache.stamp[fn], True)

class SignatureGeneratorBasicHash(SignatureGeneratorBasic):
    name = "basichash"

    def stampfile(self, stampbase, fn, taskname, extrainfo, clean=False):
        if taskname != "do_setscene" and taskname.endswith("_setscene"):
            k = fn + "." + taskname[:-9]
        else:
            k = fn + "." + taskname
        if clean:
            h = "*"
        elif k in self.taskhash:
            h = self.taskhash[k]
        else:
            # If k is not in basehash, then error
            h = self.basehash[k]
        return ("%s.%s.%s.%s" % (stampbase, taskname, h, extrainfo)).rstrip('.')

    def stampcleanmask(self, stampbase, fn, taskname, extrainfo):
        return self.stampfile(stampbase, fn, taskname, extrainfo, clean=True)
        
    def invalidate_task(self, task, d, fn):
        bb.note("Tainting hash to force rebuild of task %s, %s" % (fn, task))
        bb.build.write_taint(task, d, fn)

def dump_this_task(outfile, d):
    import bb.parse
    fn = d.getVar("BB_FILENAME", True)
    task = "do_" + d.getVar("BB_CURRENTTASK", True)
    referencestamp = bb.build.stamp_internal(task, d, None, True)
    bb.parse.siggen.dump_sigtask(fn, task, outfile, "customfile:" + referencestamp)

def clean_basepath(a):
    b = a.rsplit("/", 2)[1] + a.rsplit("/", 2)[2]
    if a.startswith("virtual:"):
        b = b + ":" + a.rsplit(":", 1)[0]
    return b

def clean_basepaths(a):
    b = {}
    for x in a:
        b[clean_basepath(x)] = a[x]
    return b

def clean_basepaths_list(a):
    b = []
    for x in a:
        b.append(clean_basepath(x))
    return b

def compare_sigfiles(a, b, recursecb = None):
    output = []

    p1 = pickle.Unpickler(open(a, "rb"))
    a_data = p1.load()
    p2 = pickle.Unpickler(open(b, "rb"))
    b_data = p2.load()

    def dict_diff(a, b, whitelist=set()):
        sa = set(a.keys())
        sb = set(b.keys())
        common = sa & sb
        changed = set()
        for i in common:
            if a[i] != b[i] and i not in whitelist:
                changed.add(i)
        added = sb - sa
        removed = sa - sb
        return changed, added, removed

    def file_checksums_diff(a, b):
        from collections import Counter
        # Handle old siginfo format
        if isinstance(a, dict):
            a = [(os.path.basename(f), cs) for f, cs in a.items()]
        if isinstance(b, dict):
            b = [(os.path.basename(f), cs) for f, cs in b.items()]
        # Compare lists, ensuring we can handle duplicate filenames if they exist
        removedcount = Counter(a)
        removedcount.subtract(b)
        addedcount = Counter(b)
        addedcount.subtract(a)
        added = []
        for x in b:
            if addedcount[x] > 0:
                addedcount[x] -= 1
                added.append(x)
        removed = []
        changed = []
        for x in a:
            if removedcount[x] > 0:
                removedcount[x] -= 1
                for y in added:
                    if y[0] == x[0]:
                        changed.append((x[0], x[1], y[1]))
                        added.remove(y)
                        break
                else:
                    removed.append(x)
        added = [x[0] for x in added]
        removed = [x[0] for x in removed]
        return changed, added, removed

    if 'basewhitelist' in a_data and a_data['basewhitelist'] != b_data['basewhitelist']:
        output.append("basewhitelist changed from '%s' to '%s'" % (a_data['basewhitelist'], b_data['basewhitelist']))
        if a_data['basewhitelist'] and b_data['basewhitelist']:
            output.append("changed items: %s" % a_data['basewhitelist'].symmetric_difference(b_data['basewhitelist']))

    if 'taskwhitelist' in a_data and a_data['taskwhitelist'] != b_data['taskwhitelist']:
        output.append("taskwhitelist changed from '%s' to '%s'" % (a_data['taskwhitelist'], b_data['taskwhitelist']))
        if a_data['taskwhitelist'] and b_data['taskwhitelist']:
            output.append("changed items: %s" % a_data['taskwhitelist'].symmetric_difference(b_data['taskwhitelist']))

    if a_data['taskdeps'] != b_data['taskdeps']:
        output.append("Task dependencies changed from:\n%s\nto:\n%s" % (sorted(a_data['taskdeps']), sorted(b_data['taskdeps'])))

    if a_data['basehash'] != b_data['basehash']:
        output.append("basehash changed from %s to %s" % (a_data['basehash'], b_data['basehash']))

    changed, added, removed = dict_diff(a_data['gendeps'], b_data['gendeps'], a_data['basewhitelist'] & b_data['basewhitelist'])
    if changed:
        for dep in changed:
            output.append("List of dependencies for variable %s changed from '%s' to '%s'" % (dep, a_data['gendeps'][dep], b_data['gendeps'][dep]))
            if a_data['gendeps'][dep] and b_data['gendeps'][dep]:
                output.append("changed items: %s" % a_data['gendeps'][dep].symmetric_difference(b_data['gendeps'][dep]))
    if added:
        for dep in added:
            output.append("Dependency on variable %s was added" % (dep))
    if removed:
        for dep in removed:
            output.append("Dependency on Variable %s was removed" % (dep))


    changed, added, removed = dict_diff(a_data['varvals'], b_data['varvals'])
    if changed:
        for dep in changed:
            output.append("Variable %s value changed from '%s' to '%s'" % (dep, a_data['varvals'][dep], b_data['varvals'][dep]))

    changed, added, removed = file_checksums_diff(a_data['file_checksum_values'], b_data['file_checksum_values'])
    if changed:
        for f, old, new in changed:
            output.append("Checksum for file %s changed from %s to %s" % (f, old, new))
    if added:
        for f in added:
            output.append("Dependency on checksum of file %s was added" % (f))
    if removed:
        for f in removed:
            output.append("Dependency on checksum of file %s was removed" % (f))


    if len(a_data['runtaskdeps']) != len(b_data['runtaskdeps']):
        changed = ["Number of task dependencies changed"]
    else:
        changed = []
        for idx, task in enumerate(a_data['runtaskdeps']):
            a = a_data['runtaskdeps'][idx]
            b = b_data['runtaskdeps'][idx]
            if a_data['runtaskhashes'][a] != b_data['runtaskhashes'][b]:
                changed.append("%s with hash %s\n changed to\n%s with hash %s" % (a, a_data['runtaskhashes'][a], b, b_data['runtaskhashes'][b]))

    if changed:
        output.append("runtaskdeps changed from %s to %s" % (clean_basepaths_list(a_data['runtaskdeps']), clean_basepaths_list(b_data['runtaskdeps'])))
        output.append("\n".join(changed))


    if 'runtaskhashes' in a_data and 'runtaskhashes' in b_data:
        a = a_data['runtaskhashes']
        b = b_data['runtaskhashes']
        changed, added, removed = dict_diff(a, b)
        if added:
            for dep in added:
                bdep_found = False
                if removed:
                    for bdep in removed:
                        if b[dep] == a[bdep]:
                            #output.append("Dependency on task %s was replaced by %s with same hash" % (dep, bdep))
                            bdep_found = True
                if not bdep_found:
                    output.append("Dependency on task %s was added with hash %s" % (clean_basepath(dep), b[dep]))
        if removed:
            for dep in removed:
                adep_found = False
                if added:
                    for adep in added:
                        if b[adep] == a[dep]:
                            #output.append("Dependency on task %s was replaced by %s with same hash" % (adep, dep))
                            adep_found = True
                if not adep_found:
                    output.append("Dependency on task %s was removed with hash %s" % (clean_basepath(dep), a[dep]))
        if changed:
            for dep in changed:
                output.append("Hash for dependent task %s changed from %s to %s" % (clean_basepath(dep), a[dep], b[dep]))
                if callable(recursecb):
                    # If a dependent hash changed, might as well print the line above and then defer to the changes in 
                    # that hash since in all likelyhood, they're the same changes this task also saw.
                    recout = recursecb(dep, a[dep], b[dep])
                    if recout:
                        output = [output[-1]] + recout

    a_taint = a_data.get('taint', None)
    b_taint = b_data.get('taint', None)
    if a_taint != b_taint:
        output.append("Taint (by forced/invalidated task) changed from %s to %s" % (a_taint, b_taint))

    return output


def calc_basehash(sigdata):
    task = sigdata['task']
    basedata = sigdata['varvals'][task]

    if basedata is None:
        basedata = ''

    alldeps = sigdata['taskdeps']
    for dep in alldeps:
        basedata = basedata + dep
        val = sigdata['varvals'][dep]
        if val is not None:
            basedata = basedata + str(val)

    return hashlib.md5(basedata).hexdigest()

def calc_taskhash(sigdata):
    data = sigdata['basehash']

    for dep in sigdata['runtaskdeps']:
        data = data + sigdata['runtaskhashes'][dep]

    for c in sigdata['file_checksum_values']:
        data = data + c[1]

    if 'taint' in sigdata:
        if 'nostamp:' in sigdata['taint']:
            data = data + sigdata['taint'][8:]
        else:
            data = data + sigdata['taint']

    return hashlib.md5(data).hexdigest()


def dump_sigfile(a):
    output = []

    p1 = pickle.Unpickler(open(a, "rb"))
    a_data = p1.load()

    output.append("basewhitelist: %s" % (a_data['basewhitelist']))

    output.append("taskwhitelist: %s" % (a_data['taskwhitelist']))

    output.append("Task dependencies: %s" % (sorted(a_data['taskdeps'])))

    output.append("basehash: %s" % (a_data['basehash']))

    for dep in a_data['gendeps']:
        output.append("List of dependencies for variable %s is %s" % (dep, a_data['gendeps'][dep]))

    for dep in a_data['varvals']:
        output.append("Variable %s value is %s" % (dep, a_data['varvals'][dep]))

    if 'runtaskdeps' in a_data:
        output.append("Tasks this task depends on: %s" % (a_data['runtaskdeps']))

    if 'file_checksum_values' in a_data:
        output.append("This task depends on the checksums of files: %s" % (a_data['file_checksum_values']))

    if 'runtaskhashes' in a_data:
        for dep in a_data['runtaskhashes']:
            output.append("Hash for dependent task %s is %s" % (dep, a_data['runtaskhashes'][dep]))

    if 'taint' in a_data:
        output.append("Tainted (by forced/invalidated task): %s" % a_data['taint'])

    if 'task' in a_data:
        computed_basehash = calc_basehash(a_data)
        output.append("Computed base hash is %s and from file %s" % (computed_basehash, a_data['basehash']))
    else:
        output.append("Unable to compute base hash")

    computed_taskhash = calc_taskhash(a_data)
    output.append("Computed task hash is %s" % computed_taskhash)

    return output