Reduction semantics #109
Reduction semantics #109
Conversation
This patch populates the "Execution" section of the Explainer document with the reduction rules for stack switching.
| @@ -833,6 +833,126 @@ events and only `(on $e switch)` handlers can handle `switch` | |||
| events. The handler search continues past handlers for the wrong kind | |||
| of event, even if they use the correct tag. | |||
|
|
|||
| #### Store extensions | |||
|
|
|||
| * A store component `tags` for allocated tags (from the exception | |||
There was a problem hiding this comment.
Since the exception handling proposal is already merged, I think we can consider taginsts and their store component given.
| - `ref.cont a : [] -> [(ref $ct)]` | ||
| - iff `S.conts[a] = epsilon \/ S.conts[a] = (E : n)` | ||
| - and `$ct ~~ cont $ft` | ||
| - and `$ft ~~ [t1^n] -> [t2*]` |
There was a problem hiding this comment.
This would also need to require that E returns t2*. I think we want to factor both out into a separate typing rule for continuations, which is just invoked here. That rule would be something like
(E : n) : t1^n → t2*
- iff s ⊢ E[val^n] : t2*
- and (s ⊢ val : t1)^n
| - and `$ft ~~ [t1^n] -> [t2*]` | ||
|
|
||
| * `(prompt{<hdl>*} <instr>* end)` represents an active handler | ||
| - `(prompt{((a $l) | (b switch))*}? instr* end) : [t1*] -> [t2*]` |
There was a problem hiding this comment.
This is a bit too cute. A disjunction under an iteration does not really work, since the dimensions of a* and b* wouldn't really make sense then. So I'd suggest
| - `(prompt{((a $l) | (b switch))*}? instr* end) : [t1*] -> [t2*]` | |
| - `(prompt{hdl*} instr* end) : [t1*] -> [t2*]` |
and then have auxiliary rules for typing hdl, like for syntactic handlers.
|
|
||
| The administrative structure `hdl` is defined as. | ||
| ``` | ||
| hdl ::= (<tagaddr> $l) hdl | (<tagaddr> switch) |
There was a problem hiding this comment.
| hdl ::= (<tagaddr> $l) hdl | (<tagaddr> switch) | |
| hdl ::= (<tagaddr> $l) | (<tagaddr> switch) |
| label_n{instr*} H^ea end | ||
| frame_n{F} H^ea end | ||
| catch{...} H^ea end | ||
| prompt{hdl*} H^ea end (iff ea notin ea'*) |
There was a problem hiding this comment.
| prompt{hdl*} H^ea end (iff ea notin ea'*) | |
| prompt{hdl*} H^ea end (iff ea notin tagaddr(hdl*)) |
|
|
||
| * `S; F; v^m (ref.cont ca) (resume_throw $ct $e hdl*) --> S'; F; prompt{hdl*} E[v^m (throw $e)] end` | ||
| - iff `S.conts[ca] = (E : n)` | ||
| - and `(ea = F.tags[$t])*` |
| - and `S.tags[F.tags[$e]].type ~~ [t1^m] -> [t2*]` | ||
| - and `S' = S with conts[ca] = epsilon` | ||
|
|
||
| * `S; F; (prompt{(e $l)*}? v* end) --> S; F; v*` |
There was a problem hiding this comment.
| * `S; F; (prompt{(e $l)*}? v* end) --> S; F; v*` | |
| * `S; F; (prompt{hdl*} v* end) --> S; F; v*` |
|
|
||
| * `S; F; (prompt{(e $l)*}? v* end) --> S; F; v*` | ||
|
|
||
| * `S; F; (prompt H^ea[(suspend $e)] end) --> S; F; trap` |
There was a problem hiding this comment.
It looks like this rule was for barrier, so is obsolete?
| - iff `ea = F.tags[$e]` | ||
|
|
||
| * `S; F; (prompt{hdl1* (ea $l) hdl2*} H^ea[v^n (suspend $e)] end) --> S'; F; v^n (ref.cont |S.conts|) (br $l)` | ||
| - iff `ea notin ea1*` |
There was a problem hiding this comment.
| - iff `ea notin ea1*` | |
| - iff `ea notin tagaddr(hdl1*)` |
| * `S; F; (prompt{hdl1* (ea switch) hdl2*} H^ea[v^n (ref.cont ca) (switch $ct $e)] end) --> S''; F; prompt{hdl1* (ea switch) hdl2*} E[v^n (ref.cont |S.conts|)] end` | ||
| - iff `S.conts[ca] = (E : n')` | ||
| - and `n' = 1 + n` | ||
| - and `ea notin ea1*` |
There was a problem hiding this comment.
| - and `ea notin ea1*` | |
| - and `ea notin tagaddr(hdl1*)` |
This patch populates the "Execution" section of the Explainer document with the reduction rules for stack switching.
Resolves #91.