Mark some blocking FFI imports as "safe" (fix high cpu usage in xmonad) #78
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
GHC RTS doesn't expect "unsafe" FFI calls to block, so when the parallel garbage collector kicks in, blocking unsafe calls cause all the other GC threads to wait in a busy loop. This can be reproduced by using XMonad.Actions.Submap (which uses XMaskEvent in a blocking fashion) together with +RTS -N. To confirm the parallel GC hypothesis, +RTS -N -qg (disables parallel GC) works fine. (For more details about the reproducer, see the linked xmonad-contrib issues.) The fix is to mark potentially blocking calls "safe", as XNextEvent has been for the last 14 years and as documented in: https://downloads.haskell.org/~ghc/9.0.1/docs/html/users_guide/exts/ffi.html#foreign-imports-and-multi-threading Fixes: xmonad/xmonad-contrib#262 Fixes: xmonad/xmonad-contrib#483 Related: 344d942 ("XNextEvent should be *safe*, for correct behaviour with threads")
This was referenced Aug 9, 2021
|
Can confirm that this works for me so I'm going to merge |
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Outdated suggestions cannot be applied.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.
GHC RTS doesn't expect "unsafe" FFI calls to block, so when the parallel garbage collector kicks in, blocking unsafe calls cause all the other GC threads to wait in a busy loop.
This can be reproduced by using XMonad.Actions.Submap (which uses XMaskEvent in a blocking fashion) together with +RTS -N. To confirm the parallel GC hypothesis, +RTS -N -qg (disables parallel GC) works fine. (For more details about the reproducer, see the linked xmonad-contrib issues.)
The fix is to mark potentially blocking calls "safe", as XNextEvent has been for the last 14 years and as documented in: https://downloads.haskell.org/~ghc/9.0.1/docs/html/users_guide/exts/ffi.html#foreign-imports-and-multi-threading
Fixes: xmonad/xmonad-contrib#262
Fixes: xmonad/xmonad-contrib#483
Related: 344d942 ("XNextEvent should be safe, for correct behaviour with threads")