Merge pull request #8572 from matrix-org/rav/cache_hacking/2

Push some deferred wrangling down into DeferredCache
This commit is contained in:
Richard van der Hoff 2020-10-21 17:55:04 +01:00 committed by GitHub
commit 9146a8a691
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
6 changed files with 441 additions and 271 deletions

1
changelog.d/8572.misc Normal file
View File

@ -0,0 +1 @@
Modify `DeferredCache.get()` to return `Deferred`s instead of `ObservableDeferred`s.

View File

@ -57,7 +57,7 @@ class DeferredCache(Generic[KT, VT]):
"""Wraps an LruCache, adding support for Deferred results.
It expects that each entry added with set() will be a Deferred; likewise get()
may return an ObservableDeferred.
will return a Deferred.
"""
__slots__ = (
@ -130,16 +130,22 @@ class DeferredCache(Generic[KT, VT]):
key: KT,
callback: Optional[Callable[[], None]] = None,
update_metrics: bool = True,
) -> Union[ObservableDeferred, VT]:
) -> defer.Deferred:
"""Looks the key up in the caches.
For symmetry with set(), this method does *not* follow the synapse logcontext
rules: the logcontext will not be cleared on return, and the Deferred will run
its callbacks in the sentinel context. In other words: wrap the result with
make_deferred_yieldable() before `await`ing it.
Args:
key(tuple)
callback(fn): Gets called when the entry in the cache is invalidated
key:
callback: Gets called when the entry in the cache is invalidated
update_metrics (bool): whether to update the cache hit rate metrics
Returns:
Either an ObservableDeferred or the result itself
A Deferred which completes with the result. Note that this may later fail
if there is an ongoing set() operation which later completes with a failure.
Raises:
KeyError if the key is not found in the cache
@ -152,7 +158,7 @@ class DeferredCache(Generic[KT, VT]):
m = self.cache.metrics
assert m # we always have a name, so should always have metrics
m.inc_hits()
return val.deferred
return val.deferred.observe()
val2 = self.cache.get(
key, _Sentinel.sentinel, callbacks=callbacks, update_metrics=update_metrics
@ -160,7 +166,7 @@ class DeferredCache(Generic[KT, VT]):
if val2 is _Sentinel.sentinel:
raise KeyError()
else:
return val2
return defer.succeed(val2)
def get_immediate(
self, key: KT, default: T, update_metrics: bool = True
@ -173,7 +179,36 @@ class DeferredCache(Generic[KT, VT]):
key: KT,
value: defer.Deferred,
callback: Optional[Callable[[], None]] = None,
) -> ObservableDeferred:
) -> defer.Deferred:
"""Adds a new entry to the cache (or updates an existing one).
The given `value` *must* be a Deferred.
First any existing entry for the same key is invalidated. Then a new entry
is added to the cache for the given key.
Until the `value` completes, calls to `get()` for the key will also result in an
incomplete Deferred, which will ultimately complete with the same result as
`value`.
If `value` completes successfully, subsequent calls to `get()` will then return
a completed deferred with the same result. If it *fails*, the cache is
invalidated and subequent calls to `get()` will raise a KeyError.
If another call to `set()` happens before `value` completes, then (a) any
invalidation callbacks registered in the interim will be called, (b) any
`get()`s in the interim will continue to complete with the result from the
*original* `value`, (c) any future calls to `get()` will complete with the
result from the *new* `value`.
It is expected that `value` does *not* follow the synapse logcontext rules - ie,
if it is incomplete, it runs its callbacks in the sentinel context.
Args:
key: Key to be set
value: a deferred which will complete with a result to add to the cache
callback: An optional callback to be called when the entry is invalidated
"""
if not isinstance(value, defer.Deferred):
raise TypeError("not a Deferred")
@ -187,6 +222,8 @@ class DeferredCache(Generic[KT, VT]):
if existing_entry:
existing_entry.invalidate()
# XXX: why don't we invalidate the entry in `self.cache` yet?
self._pending_deferred_cache[key] = entry
def compare_and_pop():
@ -230,7 +267,9 @@ class DeferredCache(Generic[KT, VT]):
# _pending_deferred_cache to the real cache.
#
observer.addCallbacks(cb, eb)
return observable
# we return a new Deferred which will be called before any subsequent observers.
return observable.observe()
def prefill(self, key: KT, value: VT, callback: Callable[[], None] = None):
callbacks = [callback] if callback else []

View File

@ -23,7 +23,6 @@ from twisted.internet import defer
from synapse.logging.context import make_deferred_yieldable, preserve_fn
from synapse.util import unwrapFirstError
from synapse.util.async_helpers import ObservableDeferred
from synapse.util.caches.deferred_cache import DeferredCache
logger = logging.getLogger(__name__)
@ -156,7 +155,7 @@ class CacheDescriptor(_CacheDescriptorBase):
keylen=self.num_args,
tree=self.tree,
iterable=self.iterable,
) # type: DeferredCache[Tuple, Any]
) # type: DeferredCache[CacheKey, Any]
def get_cache_key_gen(args, kwargs):
"""Given some args/kwargs return a generator that resolves into
@ -208,26 +207,12 @@ class CacheDescriptor(_CacheDescriptorBase):
kwargs["cache_context"] = _CacheContext.get_instance(cache, cache_key)
try:
cached_result_d = cache.get(cache_key, callback=invalidate_callback)
if isinstance(cached_result_d, ObservableDeferred):
observer = cached_result_d.observe()
else:
observer = defer.succeed(cached_result_d)
ret = cache.get(cache_key, callback=invalidate_callback)
except KeyError:
ret = defer.maybeDeferred(preserve_fn(self.orig), obj, *args, **kwargs)
ret = cache.set(cache_key, ret, callback=invalidate_callback)
def onErr(f):
cache.invalidate(cache_key)
return f
ret.addErrback(onErr)
result_d = cache.set(cache_key, ret, callback=invalidate_callback)
observer = result_d.observe()
return make_deferred_yieldable(observer)
return make_deferred_yieldable(ret)
wrapped = cast(_CachedFunction, _wrapped)
@ -286,7 +271,7 @@ class CacheListDescriptor(_CacheDescriptorBase):
def __get__(self, obj, objtype=None):
cached_method = getattr(obj, self.cached_method_name)
cache = cached_method.cache
cache = cached_method.cache # type: DeferredCache[CacheKey, Any]
num_args = cached_method.num_args
@functools.wraps(self.orig)
@ -326,14 +311,11 @@ class CacheListDescriptor(_CacheDescriptorBase):
for arg in list_args:
try:
res = cache.get(arg_to_cache_key(arg), callback=invalidate_callback)
if not isinstance(res, ObservableDeferred):
results[arg] = res
elif not res.has_succeeded():
res = res.observe()
if not res.called:
res.addCallback(update_results_dict, arg)
cached_defers.append(res)
else:
results[arg] = res.get_result()
results[arg] = res.result
except KeyError:
missing.add(arg)

View File

@ -15,237 +15,9 @@
# limitations under the License.
from mock import Mock
from twisted.internet import defer
from synapse.util.async_helpers import ObservableDeferred
from synapse.util.caches.descriptors import cached
from tests import unittest
class CacheDecoratorTestCase(unittest.HomeserverTestCase):
@defer.inlineCallbacks
def test_passthrough(self):
class A:
@cached()
def func(self, key):
return key
a = A()
self.assertEquals((yield a.func("foo")), "foo")
self.assertEquals((yield a.func("bar")), "bar")
@defer.inlineCallbacks
def test_hit(self):
callcount = [0]
class A:
@cached()
def func(self, key):
callcount[0] += 1
return key
a = A()
yield a.func("foo")
self.assertEquals(callcount[0], 1)
self.assertEquals((yield a.func("foo")), "foo")
self.assertEquals(callcount[0], 1)
@defer.inlineCallbacks
def test_invalidate(self):
callcount = [0]
class A:
@cached()
def func(self, key):
callcount[0] += 1
return key
a = A()
yield a.func("foo")
self.assertEquals(callcount[0], 1)
a.func.invalidate(("foo",))
yield a.func("foo")
self.assertEquals(callcount[0], 2)
def test_invalidate_missing(self):
class A:
@cached()
def func(self, key):
return key
A().func.invalidate(("what",))
@defer.inlineCallbacks
def test_max_entries(self):
callcount = [0]
class A:
@cached(max_entries=10)
def func(self, key):
callcount[0] += 1
return key
a = A()
for k in range(0, 12):
yield a.func(k)
self.assertEquals(callcount[0], 12)
# There must have been at least 2 evictions, meaning if we calculate
# all 12 values again, we must get called at least 2 more times
for k in range(0, 12):
yield a.func(k)
self.assertTrue(
callcount[0] >= 14, msg="Expected callcount >= 14, got %d" % (callcount[0])
)
def test_prefill(self):
callcount = [0]
d = defer.succeed(123)
class A:
@cached()
def func(self, key):
callcount[0] += 1
return d
a = A()
a.func.prefill(("foo",), ObservableDeferred(d))
self.assertEquals(a.func("foo").result, d.result)
self.assertEquals(callcount[0], 0)
@defer.inlineCallbacks
def test_invalidate_context(self):
callcount = [0]
callcount2 = [0]
class A:
@cached()
def func(self, key):
callcount[0] += 1
return key
@cached(cache_context=True)
def func2(self, key, cache_context):
callcount2[0] += 1
return self.func(key, on_invalidate=cache_context.invalidate)
a = A()
yield a.func2("foo")
self.assertEquals(callcount[0], 1)
self.assertEquals(callcount2[0], 1)
a.func.invalidate(("foo",))
yield a.func("foo")
self.assertEquals(callcount[0], 2)
self.assertEquals(callcount2[0], 1)
yield a.func2("foo")
self.assertEquals(callcount[0], 2)
self.assertEquals(callcount2[0], 2)
@defer.inlineCallbacks
def test_eviction_context(self):
callcount = [0]
callcount2 = [0]
class A:
@cached(max_entries=2)
def func(self, key):
callcount[0] += 1
return key
@cached(cache_context=True)
def func2(self, key, cache_context):
callcount2[0] += 1
return self.func(key, on_invalidate=cache_context.invalidate)
a = A()
yield a.func2("foo")
yield a.func2("foo2")
self.assertEquals(callcount[0], 2)
self.assertEquals(callcount2[0], 2)
yield a.func2("foo")
self.assertEquals(callcount[0], 2)
self.assertEquals(callcount2[0], 2)
yield a.func("foo3")
self.assertEquals(callcount[0], 3)
self.assertEquals(callcount2[0], 2)
yield a.func2("foo")
self.assertEquals(callcount[0], 4)
self.assertEquals(callcount2[0], 3)
@defer.inlineCallbacks
def test_double_get(self):
callcount = [0]
callcount2 = [0]
class A:
@cached()
def func(self, key):
callcount[0] += 1
return key
@cached(cache_context=True)
def func2(self, key, cache_context):
callcount2[0] += 1
return self.func(key, on_invalidate=cache_context.invalidate)
a = A()
a.func2.cache.cache = Mock(wraps=a.func2.cache.cache)
yield a.func2("foo")
self.assertEquals(callcount[0], 1)
self.assertEquals(callcount2[0], 1)
a.func2.invalidate(("foo",))
self.assertEquals(a.func2.cache.cache.pop.call_count, 1)
yield a.func2("foo")
a.func2.invalidate(("foo",))
self.assertEquals(a.func2.cache.cache.pop.call_count, 2)
self.assertEquals(callcount[0], 1)
self.assertEquals(callcount2[0], 2)
a.func.invalidate(("foo",))
self.assertEquals(a.func2.cache.cache.pop.call_count, 3)
yield a.func("foo")
self.assertEquals(callcount[0], 2)
self.assertEquals(callcount2[0], 2)
yield a.func2("foo")
self.assertEquals(callcount[0], 2)
self.assertEquals(callcount2[0], 3)
class UpsertManyTests(unittest.HomeserverTestCase):
def prepare(self, reactor, clock, hs):
self.storage = hs.get_datastore()

View File

@ -13,15 +13,16 @@
# See the License for the specific language governing permissions and
# limitations under the License.
import unittest
from functools import partial
from twisted.internet import defer
from synapse.util.caches.deferred_cache import DeferredCache
from tests.unittest import TestCase
class DeferredCacheTestCase(unittest.TestCase):
class DeferredCacheTestCase(TestCase):
def test_empty(self):
cache = DeferredCache("test")
failed = False
@ -36,7 +37,102 @@ class DeferredCacheTestCase(unittest.TestCase):
cache = DeferredCache("test")
cache.prefill("foo", 123)
self.assertEquals(cache.get("foo"), 123)
self.assertEquals(self.successResultOf(cache.get("foo")), 123)
def test_hit_deferred(self):
cache = DeferredCache("test")
origin_d = defer.Deferred()
set_d = cache.set("k1", origin_d)
# get should return an incomplete deferred
get_d = cache.get("k1")
self.assertFalse(get_d.called)
# add a callback that will make sure that the set_d gets called before the get_d
def check1(r):
self.assertTrue(set_d.called)
return r
# TODO: Actually ObservableDeferred *doesn't* run its tests in order on py3.8.
# maybe we should fix that?
# get_d.addCallback(check1)
# now fire off all the deferreds
origin_d.callback(99)
self.assertEqual(self.successResultOf(origin_d), 99)
self.assertEqual(self.successResultOf(set_d), 99)
self.assertEqual(self.successResultOf(get_d), 99)
def test_callbacks(self):
"""Invalidation callbacks are called at the right time"""
cache = DeferredCache("test")
callbacks = set()
# start with an entry, with a callback
cache.prefill("k1", 10, callback=lambda: callbacks.add("prefill"))
# now replace that entry with a pending result
origin_d = defer.Deferred()
set_d = cache.set("k1", origin_d, callback=lambda: callbacks.add("set"))
# ... and also make a get request
get_d = cache.get("k1", callback=lambda: callbacks.add("get"))
# we don't expect the invalidation callback for the original value to have
# been called yet, even though get() will now return a different result.
# I'm not sure if that is by design or not.
self.assertEqual(callbacks, set())
# now fire off all the deferreds
origin_d.callback(20)
self.assertEqual(self.successResultOf(set_d), 20)
self.assertEqual(self.successResultOf(get_d), 20)
# now the original invalidation callback should have been called, but none of
# the others
self.assertEqual(callbacks, {"prefill"})
callbacks.clear()
# another update should invalidate both the previous results
cache.prefill("k1", 30)
self.assertEqual(callbacks, {"set", "get"})
def test_set_fail(self):
cache = DeferredCache("test")
callbacks = set()
# start with an entry, with a callback
cache.prefill("k1", 10, callback=lambda: callbacks.add("prefill"))
# now replace that entry with a pending result
origin_d = defer.Deferred()
set_d = cache.set("k1", origin_d, callback=lambda: callbacks.add("set"))
# ... and also make a get request
get_d = cache.get("k1", callback=lambda: callbacks.add("get"))
# none of the callbacks should have been called yet
self.assertEqual(callbacks, set())
# oh noes! fails!
e = Exception("oops")
origin_d.errback(e)
self.assertIs(self.failureResultOf(set_d, Exception).value, e)
self.assertIs(self.failureResultOf(get_d, Exception).value, e)
# the callbacks for the failed requests should have been called.
# I'm not sure if this is deliberate or not.
self.assertEqual(callbacks, {"get", "set"})
callbacks.clear()
# the old value should still be returned now?
get_d2 = cache.get("k1", callback=lambda: callbacks.add("get2"))
self.assertEqual(self.successResultOf(get_d2), 10)
# replacing the value now should run the callbacks for those requests
# which got the original result
cache.prefill("k1", 30)
self.assertEqual(callbacks, {"prefill", "get2"})
def test_get_immediate(self):
cache = DeferredCache("test")
@ -82,16 +178,15 @@ class DeferredCacheTestCase(unittest.TestCase):
d2 = defer.Deferred()
cache.set("key2", d2, partial(record_callback, 1))
# lookup should return observable deferreds
self.assertFalse(cache.get("key1").has_called())
self.assertFalse(cache.get("key2").has_called())
# lookup should return pending deferreds
self.assertFalse(cache.get("key1").called)
self.assertFalse(cache.get("key2").called)
# let one of the lookups complete
d2.callback("result2")
# for now at least, the cache will return real results rather than an
# observabledeferred
self.assertEqual(cache.get("key2"), "result2")
# now the cache will return a completed deferred
self.assertEqual(self.successResultOf(cache.get("key2")), "result2")
# now do the invalidation
cache.invalidate_all()

View File

@ -14,6 +14,7 @@
# See the License for the specific language governing permissions and
# limitations under the License.
import logging
from typing import Set
import mock
@ -130,6 +131,57 @@ class DescriptorTestCase(unittest.TestCase):
d = obj.fn(1)
self.failureResultOf(d, SynapseError)
def test_cache_with_async_exception(self):
"""The wrapped function returns a failure
"""
class Cls:
result = None
call_count = 0
@cached()
def fn(self, arg1):
self.call_count += 1
return self.result
obj = Cls()
callbacks = set() # type: Set[str]
# set off an asynchronous request
obj.result = origin_d = defer.Deferred()
d1 = obj.fn(1, on_invalidate=lambda: callbacks.add("d1"))
self.assertFalse(d1.called)
# a second request should also return a deferred, but should not call the
# function itself.
d2 = obj.fn(1, on_invalidate=lambda: callbacks.add("d2"))
self.assertFalse(d2.called)
self.assertEqual(obj.call_count, 1)
# no callbacks yet
self.assertEqual(callbacks, set())
# the original request fails
e = Exception("bzz")
origin_d.errback(e)
# ... which should cause the lookups to fail similarly
self.assertIs(self.failureResultOf(d1, Exception).value, e)
self.assertIs(self.failureResultOf(d2, Exception).value, e)
# ... and the callbacks to have been, uh, called.
self.assertEqual(callbacks, {"d1", "d2"})
# ... leaving the cache empty
self.assertEqual(len(obj.fn.cache.cache), 0)
# and a second call should work as normal
obj.result = defer.succeed(100)
d3 = obj.fn(1)
self.assertEqual(self.successResultOf(d3), 100)
self.assertEqual(obj.call_count, 2)
def test_cache_logcontexts(self):
"""Check that logcontexts are set and restored correctly when
using the cache."""
@ -311,6 +363,235 @@ class DescriptorTestCase(unittest.TestCase):
self.failureResultOf(d, SynapseError)
class CacheDecoratorTestCase(unittest.HomeserverTestCase):
"""More tests for @cached
The following is a set of tests that got lost in a different file for a while.
There are probably duplicates of the tests in DescriptorTestCase. Ideally the
duplicates would be removed and the two sets of classes combined.
"""
@defer.inlineCallbacks
def test_passthrough(self):
class A:
@cached()
def func(self, key):
return key
a = A()
self.assertEquals((yield a.func("foo")), "foo")
self.assertEquals((yield a.func("bar")), "bar")
@defer.inlineCallbacks
def test_hit(self):
callcount = [0]
class A:
@cached()
def func(self, key):
callcount[0] += 1
return key
a = A()
yield a.func("foo")
self.assertEquals(callcount[0], 1)
self.assertEquals((yield a.func("foo")), "foo")
self.assertEquals(callcount[0], 1)
@defer.inlineCallbacks
def test_invalidate(self):
callcount = [0]
class A:
@cached()
def func(self, key):
callcount[0] += 1
return key
a = A()
yield a.func("foo")
self.assertEquals(callcount[0], 1)
a.func.invalidate(("foo",))
yield a.func("foo")
self.assertEquals(callcount[0], 2)
def test_invalidate_missing(self):
class A:
@cached()
def func(self, key):
return key
A().func.invalidate(("what",))
@defer.inlineCallbacks
def test_max_entries(self):
callcount = [0]
class A:
@cached(max_entries=10)
def func(self, key):
callcount[0] += 1
return key
a = A()
for k in range(0, 12):
yield a.func(k)
self.assertEquals(callcount[0], 12)
# There must have been at least 2 evictions, meaning if we calculate
# all 12 values again, we must get called at least 2 more times
for k in range(0, 12):
yield a.func(k)
self.assertTrue(
callcount[0] >= 14, msg="Expected callcount >= 14, got %d" % (callcount[0])
)
def test_prefill(self):
callcount = [0]
d = defer.succeed(123)
class A:
@cached()
def func(self, key):
callcount[0] += 1
return d
a = A()
a.func.prefill(("foo",), 456)
self.assertEquals(a.func("foo").result, 456)
self.assertEquals(callcount[0], 0)
@defer.inlineCallbacks
def test_invalidate_context(self):
callcount = [0]
callcount2 = [0]
class A:
@cached()
def func(self, key):
callcount[0] += 1
return key
@cached(cache_context=True)
def func2(self, key, cache_context):
callcount2[0] += 1
return self.func(key, on_invalidate=cache_context.invalidate)
a = A()
yield a.func2("foo")
self.assertEquals(callcount[0], 1)
self.assertEquals(callcount2[0], 1)
a.func.invalidate(("foo",))
yield a.func("foo")
self.assertEquals(callcount[0], 2)
self.assertEquals(callcount2[0], 1)
yield a.func2("foo")
self.assertEquals(callcount[0], 2)
self.assertEquals(callcount2[0], 2)
@defer.inlineCallbacks
def test_eviction_context(self):
callcount = [0]
callcount2 = [0]
class A:
@cached(max_entries=2)
def func(self, key):
callcount[0] += 1
return key
@cached(cache_context=True)
def func2(self, key, cache_context):
callcount2[0] += 1
return self.func(key, on_invalidate=cache_context.invalidate)
a = A()
yield a.func2("foo")
yield a.func2("foo2")
self.assertEquals(callcount[0], 2)
self.assertEquals(callcount2[0], 2)
yield a.func2("foo")
self.assertEquals(callcount[0], 2)
self.assertEquals(callcount2[0], 2)
yield a.func("foo3")
self.assertEquals(callcount[0], 3)
self.assertEquals(callcount2[0], 2)
yield a.func2("foo")
self.assertEquals(callcount[0], 4)
self.assertEquals(callcount2[0], 3)
@defer.inlineCallbacks
def test_double_get(self):
callcount = [0]
callcount2 = [0]
class A:
@cached()
def func(self, key):
callcount[0] += 1
return key
@cached(cache_context=True)
def func2(self, key, cache_context):
callcount2[0] += 1
return self.func(key, on_invalidate=cache_context.invalidate)
a = A()
a.func2.cache.cache = mock.Mock(wraps=a.func2.cache.cache)
yield a.func2("foo")
self.assertEquals(callcount[0], 1)
self.assertEquals(callcount2[0], 1)
a.func2.invalidate(("foo",))
self.assertEquals(a.func2.cache.cache.pop.call_count, 1)
yield a.func2("foo")
a.func2.invalidate(("foo",))
self.assertEquals(a.func2.cache.cache.pop.call_count, 2)
self.assertEquals(callcount[0], 1)
self.assertEquals(callcount2[0], 2)
a.func.invalidate(("foo",))
self.assertEquals(a.func2.cache.cache.pop.call_count, 3)
yield a.func("foo")
self.assertEquals(callcount[0], 2)
self.assertEquals(callcount2[0], 2)
yield a.func2("foo")
self.assertEquals(callcount[0], 2)
self.assertEquals(callcount2[0], 3)
class CachedListDescriptorTestCase(unittest.TestCase):
@defer.inlineCallbacks
def test_cache(self):