-
Notifications
You must be signed in to change notification settings - Fork 235
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Start a Tactics V2 #2960
Merged
Merged
Start a Tactics V2 #2960
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
Everest run will be here https://github.com/mtzguido/everest/actions/runs/5238573630 |
Some things to think about:
|
This PR also introduces an assert (loc_disjoint (loc_union l (loc_addresses true r0 (Set.singleton a0))) (loc_of_aloc b))
by (let open FStar.Tactics.V2.Builtins in
let open FStar.Tactics.SMT in
set_rlimit 64;
set_options "--z3cliopt 'smt.qi.eager_threshold=5'";
()) there are also getters/setters for fuel/ifuel. |
Port change to `refl_typing_builtin_wrapper` to V2 as well.
We were getting double seals
For documentation and reducing dependencies. Also more things marked as plugins.
This reverts commit 6d1247b.
With some exceptions, but it's easier if the default is to extract modules here.
This allows to untie some dependency knots in the compiler, mostly in support of using typeclasses/tactics in src/. This way, the users of tactics in src/ (say defining a to_string instance in FStar.Syntax.Print) could only depend on FStar.Stubs.Tactics.Builtins, and not on the *actual* builtins with the implementations, which would be a cycle. This works as long as the dependency is only for tactic execution, as the extracted code will not have such dependency.
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.
This PR introduces a new version of the tactics engine, with differences to (mostly) the representation of terms. The change is heavily disruptive to pretty much everything, so we are staging it by maintaining a
.V1
namespace of Tactics and Reflection, which retains exactly the previous behaviour, and a V2 with new changes. V2, for now, has to be explicitly asked for byopen FStar.Tactics.V2
or any submodule thereof.The two versions can interoperate, the underlying effect is the same, but the relevant view types and primitives can be different. Some files are totally agnostic to the version, such as
FStar.Tactics.SMT
andFStar.Tactics.Visit
(which is now implemented with V2, but just exposesvisit_tm : (term -> Tac term) -> term -> Tac term
),V2 is very explicit about opening/closing names, freshness, etc. This is now all implemented in userspace (
FStar.Tactics.NamedView
), and marked as a plugin so it runs fast.Another significant change is the autogeneration of embeddings for most inductive types, so many more functions can be easily made into plugins. Also typeclasses are moved higher up in the dependency chain and even used within tactic libraries (
FStar.Tactics.MApply.fst
).For now, I think V2 should be considered experimental and subject to change, but I would like to get this into the repo as it's a big change that could diverge.
Karamel needs a tiny patch to ignore a new module,
FStar.Syntax.Syntax
, inulib/
(which is an interface for the compiler'sFStar.Syntax.Syntax
). We may move all of the F* compiler sources into theFStar.Compiler
namespace to make this easier to maintain, but not on this PR.I have a steel branch that works with V2.
For hacl-star I haven't yet ported to V2, but confirmed that it works as before with V1.
Will post an everest green later.