Better unsafe perform io #325
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
treeowl
force-pushed
the
better-unsafePerformIO
branch
from
e8b3679
to
fab12cf
Compare
The previous code was struggling to create artificial dependencies to prevent `unsafePerformIO` calls from being floated out of their proper context. Push the context into each `unsafePerformIO` call so the dependency is real. Reading a bit of the resulting Core, I don't expect to see many (if any) performance regressions. Performance will hopefully improve a bit if/when [GHC issue 15127](https://gitlab.haskell.org/ghc/ghc/issues/15127) is fixed.
treeowl
force-pushed
the
better-unsafePerformIO
branch
from
fab12cf
to
9bf7cb5
Compare
|
I believe it should be safe to use |
treeowl
commented
Jul 11, 2019
treeowl
commented
Jul 11, 2019
| unsafeBuildDynamic readV0 v' = Dyn $ unsafeNewIORef x $ UnsafeDyn x | ||
| where x = (readV0, v') | ||
| unsafeBuildDynamic readV0 v' = | ||
| Dyn $ unsafePerformIO $ newIORef $ UnsafeDyn (readV0, v') |
There was a problem hiding this comment.
If this gets called many times, then it might be worth working around GHC 15127 by using runRW# manually to unbox the IORef.
unsafeNewIORef :: a -> IORef a
unsafeNewIORef a = IORef $ STRef ( runRW# $ \s ->
case newMutVar# a (noDuplicate# s) of
(# _s', mv #) -> mv )or
unsafeDupableNewIORef :: a -> IORef a
unsafeDupableNewIORef a = IORef $ STRef ( runRW# $ \s ->
case newMutVar# a s of
(# _s', mv #) -> mv )
There was a problem hiding this comment.
I don't think it does - because unsafeBuildDynamic is not a very frequently used function. Current benchmarks show no significant difference (for any of these changes)
Closed
ali-abrar
added a commit
that referenced
this pull request
Aug 1, 2019
* 'release/0.6.2.1' of github.com:ryantrinkle/reflex: (30 commits) Loosen monoidal-containers version bounds Update CONTRIBUTING.md Update CONTRIBUTING.md Use unsafePerformIO better (#325) Generalize fan following DMap (#318) Distribute more generally Generalize merge Drop *Tag classes Fix deprecation warnings related to updated Data.Some Bump dependent-sum to latest version fan: Rewrite haddock Replace hackage link with badge Use Some for more existential types (which should be efficient now that it is a newtype) Bump upper bounds Typo DRY conditional compilation Bump 'these' upper bound Typo Update changelog for holdDyn fix Make holdDyn lazy in its Event again ...
matthewbauer
added a commit
that referenced
this pull request
Jan 31, 2020
This reverts commit dc3ce15.
ryantrinkle
added a commit
that referenced
this pull request
May 27, 2020
This reverts commit dc3ce15.
ryantrinkle
added a commit
that referenced
this pull request
May 29, 2020
This reverts commit dc3ce15.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
The previous code was struggling to create artificial dependencies
to prevent
unsafePerformIOcalls from being floated out of theirproper context. Push the context into each
unsafePerformIOcallso the dependency is real. Reading a bit of the resulting Core,
I don't expect to see many (if any) performance regressions.
Performance will hopefully improve a bit if/when
GHC issue 15127
is fixed.