Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

fix: separate ZstdDecompressor for each thread. #820

Merged
merged 3 commits into from
Feb 15, 2023

Conversation

jpivarski
Copy link
Member

Does this fix the

>>> fut.compute(scheduler="synchronous")
<Array [[469, 459, 188, 17.8], ..., [414, ...]] type='203939 * var * float32'>
>>> fut.compute(scheduler="threads")
Traceback (most recent call last):
...
zstd.ZstdError: decompression error: Corrupted block detected

issue, @nsmith-?

@jpivarski jpivarski requested a review from agoose77 January 26, 2023 20:06
@jpivarski jpivarski linked an issue Jan 26, 2023 that may be closed by this pull request
@jpivarski jpivarski enabled auto-merge (squash) February 15, 2023 18:14
@jpivarski jpivarski merged commit cb7e291 into main Feb 15, 2023
@jpivarski jpivarski deleted the jpivarski/separate-ZstdDecompressor-by-thread branch February 15, 2023 18:30
@nsmith-
Copy link
Member

nsmith- commented Feb 16, 2023

I should have checked this sooner, but I get now

>>> events.MET.pt.compute(scheduler="synchronous")
<Array [78, 24.6, 5.05, 58.4, ..., 35.7, 29.9, 54] type='203939 * float32'>
>>> events.MET.pt.compute(scheduler="threads")
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/Users/ncsmith/src/coffea/.env/lib/python3.10/site-packages/dask/base.py", line 315, in compute
    (result,) = compute(self, traverse=False, **kwargs)
  File "/Users/ncsmith/src/coffea/.env/lib/python3.10/site-packages/dask/base.py", line 600, in compute
    results = schedule(dsk, keys, **kwargs)
  File "/Users/ncsmith/src/coffea/.env/lib/python3.10/site-packages/dask/threaded.py", line 89, in get
    results = get_async(
  File "/Users/ncsmith/src/coffea/.env/lib/python3.10/site-packages/dask/local.py", line 511, in get_async
    raise_exception(exc, tb)
  File "/Users/ncsmith/src/coffea/.env/lib/python3.10/site-packages/dask/local.py", line 319, in reraise
    raise exc
  File "/Users/ncsmith/src/coffea/.env/lib/python3.10/site-packages/dask/local.py", line 224, in execute_task
    result = _execute_task(task, data)
  File "/Users/ncsmith/src/coffea/.env/lib/python3.10/site-packages/dask/core.py", line 119, in _execute_task
    return func(*(_execute_task(a, cache) for a in args))
  File "/Users/ncsmith/src/coffea/.env/lib/python3.10/site-packages/dask/optimization.py", line 990, in __call__
    return core.get(self.dsk, self.outkey, dict(zip(self.inkeys, args)))
  File "/Users/ncsmith/src/coffea/.env/lib/python3.10/site-packages/dask/core.py", line 149, in get
    result = _execute_task(task, cache)
  File "/Users/ncsmith/src/coffea/.env/lib/python3.10/site-packages/dask/core.py", line 119, in _execute_task
    return func(*(_execute_task(a, cache) for a in args))
  File "/Users/ncsmith/src/coffea/.env/lib/python3.10/site-packages/uproot/_dask.py", line 592, in __call__
    return self.ttrees[i].arrays(
  File "/Users/ncsmith/src/coffea/.env/lib/python3.10/site-packages/uproot/behaviors/TBranch.py", line 834, in arrays
    _ranges_or_baskets_to_arrays(
  File "/Users/ncsmith/src/coffea/.env/lib/python3.10/site-packages/uproot/behaviors/TBranch.py", line 3124, in _ranges_or_baskets_to_arrays
    uproot.source.futures.delayed_raise(*obj)
  File "/Users/ncsmith/src/coffea/.env/lib/python3.10/site-packages/uproot/source/futures.py", line 36, in delayed_raise
    raise exception_value.with_traceback(traceback)
  File "/Users/ncsmith/src/coffea/.env/lib/python3.10/site-packages/uproot/behaviors/TBranch.py", line 3037, in chunk_to_basket
    basket = uproot.models.TBasket.Model_TBasket.read(
  File "/Users/ncsmith/src/coffea/.env/lib/python3.10/site-packages/uproot/model.py", line 867, in read
    self.read_members(chunk, cursor, context, file)
  File "/Users/ncsmith/src/coffea/.env/lib/python3.10/site-packages/uproot/models/TBasket.py", line 291, in read_members
    uncompressed = uproot.compression.decompress(
  File "/Users/ncsmith/src/coffea/.env/lib/python3.10/site-packages/uproot/compression.py", line 415, in decompress
    uncompressed_bytestring = decompressor.decompress(
  File "/Users/ncsmith/src/coffea/.env/lib/python3.10/site-packages/uproot/compression.py", line 249, in decompress
    return self.decompressor.decompress(data)
  File "/Users/ncsmith/src/coffea/.env/lib/python3.10/site-packages/uproot/compression.py", line 243, in decompressor
    if self._decompressor.obj is None:
AttributeError: '_thread._local' object has no attribute 'obj'

The issue is that, apparently, each thread has an independent attribute dictionary:

>>> import threading
>>> x = threading.local()
>>> x.obj = None
>>> print(hasattr(x, "obj"))
True
>>> t = threading.Thread(target=lambda x: print(hasattr(x, "obj")), args=(x,))
>>> t.start()
False

@nsmith-
Copy link
Member

nsmith- commented Feb 16, 2023

I'll prepare a new PR

nsmith- added a commit to nsmith-/uproot5 that referenced this pull request Feb 16, 2023
jpivarski pushed a commit that referenced this pull request Feb 16, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Multithreaded ZSTD decompression is not thread-safe
2 participants