Pinned linux_like/sync.rs syscall policy. Fence/load/comparison is one shim action, so no scheduler preemption is introduced between these events.
@[reducible, inline]
Instances For
@[instance_reducible]
@[instance_reducible]
Equations
Equations
- One or more equations did not get rendered due to their size.
Instances For
- takeArgs : Array ArgType → ShimSignatures.Varargs Memory.Provenance → M S (Array (OpTy Memory.Provenance))
- readPointer : OpTy Memory.Provenance → M S (Memory.Pointer (Option Memory.Provenance))
- readI32 : OpTy Memory.Provenance → M S Int32
- readU32 : OpTy Memory.Provenance → M S UInt32
- isNull : Memory.Pointer (Option Memory.Provenance) → M S Bool
- i32Place : Memory.Pointer (Option Memory.Provenance) → M S (MPlaceTy Memory.Provenance)
- timespecPlace : OpTy Memory.Provenance → M S (MPlaceTy Memory.Provenance)
- readTimespec : MPlaceTy Memory.Provenance → M S (Option Clock.Duration)
- timeout : Clock.TimeoutClock → Clock.TimeoutStyle → Clock.Duration → M S Concurrency.ThreadDeadline
- readAcquire : MPlaceTy Memory.Provenance → M S UInt32
- metadata : Concurrency.SyncObject.MetadataHooks S R
- newFutex : M S R
- wait : R → UInt32 → Option Concurrency.ThreadDeadline → MPlaceTy Memory.Provenance → M S Unit
- errnoNegOne : String → MPlaceTy Memory.Provenance → M S Unit
- writeInt : Int → MPlaceTy Memory.Provenance → M S Unit
- writeIsize : Int → MPlaceTy Memory.Provenance → M S Unit
Source scalar from_target_isize uses target width, not the syscall buffer.
- hostUsizeBits : Nat
Instances For
def
Miri.Interpreter.LinuxFutex.syscall
{S R : Type}
(h : Hooks S R)
(args : ShimSignatures.Varargs Memory.Provenance)
(dest : MPlaceTy Memory.Provenance)
:
Equations
- One or more equations did not get rendered due to their size.