max/lean4-htt
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 2
26741 commits 1 branch 0 tags 5.3 GiB
Commit graph

26741 commits

This branch
This branch
All branches
Author SHA1 Message Date
Leonardo de Moura
c40aed1628 chore: typo 2019-11-16 12:28:52 -08:00
Leonardo de Moura
587e4f199b chore: update stage0 2019-11-16 12:24:19 -08:00
Leonardo de Moura
92a89a07ff chore: Expr.CachedData => Expr.Data 2019-11-16 12:23:39 -08:00
Leonardo de Moura
1d9a76ae45 chore: fix tests 
Remark: weird discrepancies in the new typeclass module.
Ignoring for now since it was not integrated yet.
 2019-11-16 12:19:27 -08:00
Leonardo de Moura
2bf69c539e chore: update stage0 2019-11-16 12:14:34 -08:00
Leonardo de Moura
a3ccbe66cf refactor: Expr fully implemented in Lean 
No hidden fields.
 2019-11-16 12:10:49 -08:00
Leonardo de Moura
3ae5c2f2e4 chore: remove old comment 2019-11-15 14:11:50 -08:00
Leonardo de Moura
052c2c24dc chore: update stage0 2019-11-15 14:08:56 -08:00
Leonardo de Moura
85092412c7 refactor: remove Expr.FVar hack 
@Kha @dselsam:
This hack was preventing us from making `Expr` a "real" Lean type.
This was bad for a few reasons:
- It was hard to extend/modify `Expr` in Lean since we would also have
to modify the C++ code that creates the `Expr` objects with the hidden
fields.
- `Expr.lam` and `Expr.forallE` were not following the Lean layout
standard where we sort fields by size. @Kha: recall we used that to
avoid a UB. The issue with `Expr.lam` and `Expr.forallE` is that they
have a "visible" field (`BinderInfo`), which is smaller than
hidden fields such as hash code.
- `Expr.fvar` had only one field at `Expr.lean,` but four behind the
scenes.

I added a new constructor `Local` that is only accessible from C++.
It is only used in legacy code we inherited from Lean2.
We will eventually delete it.

This refactoring was quite painful since many parts of the codebase
were mixing the new `Expr.fvar` with the old `Expr.local`.
I doubt I would be able to do it without the new staging framework
@Kha built.

BTW, some of the patches are horrible. I didn't care much since we
are going to deleted the super ugly files. That being said,
you should expect new weird bevaior due to `Expr.fvar` vs `Expr.local`.

Next step: use the new `ExprCachedData` to make all `Expr` hidden visibles
accessible from Lean.

checkpoint
 2019-11-15 14:04:26 -08:00
Leonardo de Moura
0926dee83b chore: remove dead code 2019-11-15 11:53:09 -08:00
Leonardo de Moura
aea02eb1f8 chore: fix MVar refactoring issues 2019-11-15 10:04:42 -08:00
Leonardo de Moura
356192aeb9 chore: update stage0 2019-11-15 10:04:42 -08:00
Leonardo de Moura
d9f3b4bf63 refactor: remove Expr.mvar hidden field 2019-11-15 10:04:42 -08:00
Sebastian Ullrich
aa380ff505 chore: update stage0 path in script/pre-push 2019-11-15 18:29:57 +01:00
Sebastian Ullrich
b8d5193ecc chore: update Azure and Nix configs 2019-11-15 18:22:49 +01:00
Sebastian Ullrich
5c286ca6fd chore: bench cross suite: move default.nix and update README 2019-11-15 16:39:14 +01:00
Sebastian Ullrich
425230abc4 chore: fix cross bench suite 2019-11-15 16:39:14 +01:00
Sebastian Ullrich
a961e1314a chore: cross bench suite: fix fatal laziness in rbmap_checkpoint.hs 2019-11-15 16:39:14 +01:00
Sebastian Ullrich
ef9150cacd chore: cross bench suite: add Strict Haskell category (disabled by default) 
No change to slightly faster in general, except for a significant speedup in rbmap_10 that should be fixed in the lazy version
 2019-11-15 16:39:00 +01:00
Leonardo de Moura
6978a88219 chore: update stage0 2019-11-14 22:26:24 -08:00
Leonardo de Moura
beac2a1af5 fix: treat panic as unreachable at ll_infer_type_fn 
Motivation: ensure the correct type `IRType` is inferred for definitions
such as
```
def f (n : UInt32) : UInt32 :=
if n == 0 then panic! "foo"
else n+1
```
 2019-11-14 21:56:09 -08:00
Leonardo de Moura
8697638031 feat: add ExprCachedData 
We are preparing for removing the hack from expr.cpp
 2019-11-14 18:50:53 -08:00
Leonardo de Moura
7775c7816d chore: add UInt* helper functions 2019-11-14 17:26:33 -08:00
Leonardo de Moura
eb52fbca10 chore: skip stdlib tests on Windows 2019-11-14 17:00:04 -08:00
Leonardo de Moura
524872d68a chore: avoid Expr constructors in tests 2019-11-14 16:54:36 -08:00
Leonardo de Moura
c341486e28 refactor: avoid Expr constructors 
Using `mk*` functions instead.
Reason: preparing to remove `src/kernel/expr.cpp` hack.
 2019-11-14 16:32:45 -08:00
Leonardo de Moura
c768a24735 chore: mkApp => mkAppN 
We are goint to use `mkApp` for creating unary applications
 2019-11-14 15:44:17 -08:00
Leonardo de Moura
fd66963e85 chore: remove dead code 2019-11-14 15:24:18 -08:00
Leonardo de Moura
4989a7a7b6 chore: use camelCase, and add shift/casting operations for UInt64 2019-11-14 14:52:52 -08:00
Leonardo de Moura
223a251958 chore: remove legacy constructor 2019-11-14 14:47:56 -08:00
Leonardo de Moura
1f7c27672b feat: add processAssignment 2019-11-14 12:14:46 -08:00
Leonardo de Moura
e7fc90877e feat: add anyRange, allRange, anyRangeM, allRangeM, and modify anyFrom parameter order 2019-11-14 11:25:57 -08:00
Leonardo de Moura
f31fcbba24 chore: remove unnecessary approximation that just complicates the code 2019-11-14 10:57:34 -08:00
Leonardo de Moura
6b94d3fba9 chore: use contains 2019-11-14 09:55:20 -08:00
Leonardo de Moura
238c2b3d17 chore: add helper method 2019-11-14 09:52:41 -08:00
Leonardo de Moura
83e2796e96 chore: valueOpt ==> value? 2019-11-14 09:52:16 -08:00
Leonardo de Moura
79c0833d06 feat: add contains for List and Array 2019-11-14 09:51:50 -08:00
Leonardo de Moura
bbe44a79f2 feat: add tests for compiling stdlib with stage1 
@Kha It seems to work. Not sure how robust it is. CMake is still very
mysterious to me :D
It is a partial test since I don't try to compile the files into .c.
I am not touching the compiler right now. So, it should be fine for now.

The command `ctest -j8 -R stdlib` also works.
 2019-11-13 20:55:08 -08:00
Leonardo de Moura
ed43f4a901 doc: document problems with the variable and variables commands 2019-11-13 20:25:04 -08:00
Leonardo de Moura
a735778ae0 chore: update stage0 2019-11-13 15:46:55 -08:00
Leonardo de Moura
e8cee6c56f chore: update stage0 2019-11-13 15:44:48 -08:00
Leonardo de Moura
af3b465760 feat: apply [specialize] before elaboration 
This is a workaround. See new comment.
We need it for making aa9080e9e more effective.
 2019-11-13 15:41:27 -08:00
Leonardo de Moura
c4d974eb89 feat: allow attributes to be applied before elaboration 
This is useful when the attribute may influence the elaboration of the declaration.
 2019-11-13 15:40:19 -08:00
Leonardo de Moura
92316dff89 chore: remove unnecessary inline 2019-11-13 14:28:56 -08:00
Leonardo de Moura
770e7fc350 chore: update stage0 2019-11-13 13:50:03 -08:00
Leonardo de Moura
aa079f06c6 chore: update stage0 2019-11-13 13:47:53 -08:00
Leonardo de Moura
aa9080e9e7 feat: do not specialize functions inside functions marked with @[specialize] 2019-11-13 13:32:52 -08:00
Leonardo de Moura
53b1128491 chore: minimize amount of inlined code 2019-11-13 12:54:56 -08:00
Leonardo de Moura
fb621af338 chore: move code after relevant comment 2019-11-13 12:46:10 -08:00
Leonardo de Moura
46d817135e feat: save checkAssignment failures as trace messages 2019-11-13 12:42:01 -08:00