Commit 979bd7af authored by Ralf Jung's avatar Ralf Jung

docs: explain why agreement has a chain

parent 6562c4be
......@@ -5,7 +5,7 @@
\subsection{Product}
\label{sec:prodm}
Given a family $(M_i)_{i \in I}$ of CMRAs ($I$ finite), we construct a CMRA for the product $\prod_{i \in I} M_i$.
Given a family $(M_i)_{i \in I}$ of CMRAs ($I$ finite), we construct a CMRA for the product $\prod_{i \in I} M_i$ by lifting everything pointwise.
Frame-preserving updates on the $M_i$ lift to the product:
\begin{mathpar}
......@@ -27,7 +27,7 @@ We obtain the following frame-preserving updates:
\inferH{fpfn-alloc}
{\melt \in \mval}
{\emptyset \mupd \set{[\gname \mapsto \melt]}}
{\emptyset \mupd \setComp{[\gname \mapsto \melt]}{\gname \in K}}
\inferH{fpfn-update}
{\melt \mupd \meltsB}
......@@ -40,7 +40,7 @@ Given some COFE $\cofe$, we define $\agm(\cofe)$ as follows:
\newcommand{\agc}{\mathrm{c}} % the "c" field of an agreement element
\newcommand{\agV}{\mathrm{V}} % the "V" field of an agreement element
\begin{align*}
\monoid \eqdef{}& \recordComp{\agc : \mathbb{N} \to T , \agV : \pset{\mathbb{N}}}{ \All n, m. n \geq m \Ra n \in \agV \Ra m \in \agV } \\
\agm(\cofe) \eqdef{}& \recordComp{\agc : \mathbb{N} \to \cofe , \agV : \pset{\mathbb{N}}}{ \All n, m. n \geq m \Ra n \in \agV \Ra m \in \agV } \\
& \text{quotiented by} \\
\melt \equiv \meltB \eqdef{}& \melt.\agV = \meltB.\agV \land \All n. n \in \melt.\agV \Ra \melt.\agc(n) \nequiv{n} \meltB.\agc(n) \\
\melt \nequiv{n} \meltB \eqdef{}& (\All m \leq n. m \in \melt.\agV \Lra m \in \meltB.\agV) \land (\All m \leq n. m \in \melt.\agV \Ra \melt.\agc(m) \nequiv{m} \meltB.\agc(m)) \\
......@@ -50,8 +50,12 @@ Given some COFE $\cofe$, we define $\agm(\cofe)$ as follows:
\end{align*}
$\agm(-)$ is a locally non-expansive bifunctor from $\COFEs$ to $\CMRAs$.
The reason we store a \emph{chain} $c$ of elements of $T$, rather than a single element, is that $\agm(\cofe)$ needs to be a COFE itself, so we need to be able to give a limit for every chain.
\ralf{Figure out why exactly this is not possible without adding an explicit chain here.}
You can think of the $\agc$ as a \emph{chain} of elements of $\cofe$ that has to converge only for $n \in \agV$ steps.
The reason we store a chain, rather than a single element, is that $\agm(\cofe)$ needs to be a COFE itself, so we need to be able to give a limit for every chain of $\agm(\cofe)$.
However, given such a chain, we cannot constructively define its limit: Clearly, the $\agV$ of the limit is the limit of the $\agV$ of the chain.
But what to pick for the actual data, for the element of $\cofe$?
Only if $\agV = \mathbb{N}$ we have a chain of $\cofe$ that we can take a limit of; if the $\agV$ is smaller, the chain ``cancels'', \ie stops converging as we reach indices $n \notin \agV$.
To mitigate this, we apply the usual construction to close a set; we go from elements of $\cofe$ to chains of $\cofe$.
We define an injection $\ag$ into $\agm(\cofe)$ as follows:
\[ \ag(x) \eqdef \record{\mathrm c \eqdef \Lam \any. x, \mathrm V \eqdef \mathbb{N}} \]
......@@ -69,7 +73,7 @@ There are no interesting frame-preserving updates for $\agm(\cofe)$, but we can
The purpose of the one-shot CMRA is to lazily initialize the state of a ghost location.
Given some CMRA $\monoid$, we define $\oneshotm(\monoid)$ as follows:
\begin{align*}
\monoid \eqdef{}& \ospending + \osshot(\monoid) + \munit + \bot \\
\oneshot(\monoid) \eqdef{}& \ospending + \osshot(\monoid) + \munit + \bot \\
\mval_n \eqdef{}& \set{\ospending, \munit} \cup \setComp{\osshot(\melt)}{\melt \in \mval_n}
\end{align*}
\begin{align*}
......
......@@ -207,8 +207,8 @@ We can derive some specialized forms of the lifting axioms for the operational s
\subsection{Global functor and ghost ownership}
Hereinafter we assume the global CMRA functor (served up as a parameter to Iris) is obtained from a family of functors $(F_i)_{i \in I}$ for some finite $I$ by picking
\[ F(\cofe) \eqdef \prod_{i \in I} \textdom{GhName} \fpfn F_i(\cofe) \]
We don't care so much about what concretely $\textdom{GhName}$ is, as long as it is countable and infinite.
\[ F(\cofe) \eqdef \prod_{i \in I} \textlog{GhName} \fpfn F_i(\cofe) \]
We don't care so much about what concretely $\textlog{GhName}$ is, as long as it is countable and infinite.
With $M_i \eqdef F_i(\iProp)$, we write $\ownGhost{\gname}{\melt : M_i}$ (or just $\ownGhost{\gname}{\melt}$ if $M_i$ is clear from the context) for $\ownGGhost{[i \mapsto [\gname \mapsto \melt]]}$.
In other words, $\ownGhost{\gname}{\melt : M_i}$ asserts that in the current state of monoid $M_i$, the ``ghost location'' $\gname$ is allocated and we own piece $\melt$.
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment