Skip to content
Snippets Groups Projects
  1. Jul 06, 2021
  2. May 17, 2021
  3. Apr 01, 2021
  4. Mar 24, 2021
  5. Mar 18, 2021
    • Florian Fischer's avatar
      [IoContext] invalidate unsubmitted sqes · 03727b29
      Florian Fischer authored
      io_uring_submit does some inline error checking and consumes less
      cqes than prepared if an error is detected.
      
      Currently we just cancel the Futures, whose prepared sqes were not submitted.
      But this leaves already prepared sqes for those futures in the SQ
      of the io_uring which will be submitted the next time io_uring_submit
      is called.
      
      This results in a violation of the chain guaranty, that dependent operations
      are only executed if all dependencies were successful.
      Additionally this leads to double completions or memory corruption
      because the io_uring will produce cqes for already completed Futures.
      
      To prevent this from happening we track all sqes we prepared to
      invalidate and resubmit those which were not submitted because of a short submit.
      We invalidate sqes by preparing them as NOP instructions and set their
      user data to NULL.
      
      I took this approach instead of rewinding the ring or somethings like
      similar because it seemed safer for me not fiddle with io_uring internals
      and just be less efficient.
      
      Enable previously failing LinkFutureTest test cases.
      03727b29
    • Florian Fischer's avatar
      [IO] mark IO functions returning futures as nodiscard · 79f6eb9f
      Florian Fischer authored
      Fix compilation error in LinkFutureTest where clsoe was used instead
      of closeAndWait.
      79f6eb9f
  6. Mar 16, 2021
  7. Jan 26, 2021
    • Florian Fischer's avatar
      [IO] introduce emper::io a IO subsystem using io_uring · 460c2f05
      Florian Fischer authored
      Empers IO design is based on a proactor pattern where each worker
      can issue IO requests through its exclusive IoContext object which wraps an
      io_uring instance.
      
      IO completions are reaped at 4 places:
      1. After a submit to collect inline completions
      2. Before dispatching a new Fiber
      3. When no new IO can be submitted because the completion queue is full
      4. And by a global completer thread which gets notified about completions
         on worker IoContexts through registered eventfds
      
      All IO requests are modeled as Future objects which can be either
      instantiated and submitted manually, retrieved by POSIX-like non-blocking
      or implicitly used by posix-like blocking functions.
      
      User facing API is exported in the following headers:
      * emper/io.hpp (POSIX-like)
      * emper.h (POSIX-like)
      * emper/io/Future.hpp
      
      Catching short write/reads/sends and resubmitting the request without
      unblocking the Fiber is supported.
      
      Using AlarmFuture objects Fibers have a emper-native way to sleep for
      a given time.
      
      IO request timeouts with TimeoutWrapper class.
      Request Cancellation is supported with Future::cancel() or the
      CancelWrapper() Future class.
      
      A proactor design demands that buffers are committed to the kernel
      as long as the request is active. To guaranty memory safety Futures
      get canceled in their Destructor which will only return after the committed
      memory is free to use.
      
      Linking Futures to chains is supported using the Future::SetDependency()
      method. Future are submitted when their last Future gets submitted.
      A linked Request will start if the previous has finished.
      Error or partial completions will cancel the not started tail of a chain.
      
      TODO: Handle possible situations where the CQ of the global completer is full
      and no more sqe can be submitted to the SQ.
      460c2f05
Loading