mikros/rust
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
3,280 Commits 2 Branches 0 Tags 1.4 GiB
56eb986584
Commit Graph

1276 Commits

Author SHA1 Message Date
Lindsey Kuper
dc9d5e1689 Lots of comments, and some more descriptive names. 2011-05-27 18:20:43 -07:00
Tim Chevalier
b6e0c5829f Check the declaration type in a for loop against the sequence type 
Changed the typechecker to correctly typecheck the declared variable
type in a for or for-each loop against the vector element type (for
a for loop) or the iterator type (for a for-each loop). Added a
test case.
 2011-05-27 17:42:09 -07:00
Tim Chevalier
0c5a55f275 add a function that returns the type of a local decl 2011-05-27 17:34:50 -07:00
Eric Holk
52795049e7 Generating names for wrappers, removed extra logging 2011-05-27 15:37:19 -07:00
Eric Holk
8cfc388d40 Switched calling conventions so that spawn with multiple arguments works. 2011-05-27 15:20:58 -07:00
Eric Holk
d49998f0ed Switching over to wrappers for spawning functions of multiple arguments. Doesn't quite work yet. 2011-05-27 15:20:58 -07:00
Eric Holk
da0e0b6f5e Removing excessively verbous logging. 2011-05-27 15:20:58 -07:00
Eric Holk
a88af906c5 Programs using spawn terminate 2011-05-27 15:20:58 -07:00
Eric Holk
dad426419c Working on 2 argument spawn. The arguments seem to be copied correctly once, but then they get lost. 2011-05-27 15:20:58 -07:00
Eric Holk
bb4c7a9c68 Arguments seem to be correctly passed to child processes now. 2011-05-27 15:20:58 -07:00
Eric Holk
21efd07bc1 New tasks run, but their arguments get lost and something ends up in an infinite loop when the task exits. 2011-05-27 15:20:58 -07:00
Tim Chevalier
0cbcf92749 Use annotations on blocks 
Previously, block_ty returned the type of the terminating
expression of the block (or nil if said expression was absent).
I changed check_expr to write the type of that expression into
the annotation for the block itself, so now block_ty can use the
block's annotation.
 2011-05-27 12:21:37 -07:00
Tim Chevalier
1a96e79fe0 Make _|_ unify with anything 
The typechecker had a number of special cases for unifying types
with _|_ (as with checking if and alt). But, a value of type _|_
should be usable in any context, as such a value always diverges,
and will never be used by its immediate context. Changed unify
accordingly, removed special cases.
 2011-05-26 20:50:25 -07:00
Paul Stansifer
af6b4821c1 Handle circularity in glob imports in a more elegant fashion. 2011-05-27 02:42:33 +00:00
Paul Stansifer
7fe3d821ac Add a map from def_id s to items in resolve::env. 
Use it to provide better error messages in the event of
glob-importing the same name multiple times.
 2011-05-27 02:42:33 +00:00
Paul Stansifer
54ca8565c9 Fix bugs: make sure glob imports show up in the right module, 
and make sure that circular glob imports don't diverge.
 2011-05-27 02:42:33 +00:00
Paul Stansifer
40fe44d23e "import module::*;" now works. 2011-05-27 02:42:33 +00:00
Paul Stansifer
e2bb98001e Part one of adding globbing to resolve.rs: build up glob links. (Part two will follow those links if non-glob resolution fails.) 2011-05-27 02:42:32 +00:00
Graydon Hoare
79c9e13073 More delicious dead code removal from runtime, upcalls. 2011-05-26 18:21:02 -07:00
Tim Chevalier
4f792f2dbb Track spans for each local variable in typestate 
This lets us print better messages in situations with name
shadowing.
 2011-05-26 18:14:32 -07:00
Tim Chevalier
250643c549 Remove unused imports 2011-05-26 18:01:48 -07:00
Graydon Hoare
9f95538462 Remove redisue of unused fields from 'ann'. 2011-05-26 17:26:59 -07:00
Graydon Hoare
2561b01211 Remove residual uses of fold, and fold itself. 2011-05-26 17:16:54 -07:00
Graydon Hoare
8e945dcd81 Remove native glues. All calls to C are direct now. 2011-05-26 16:45:01 -07:00
Michael Sullivan
1b29892b5b Produce a useful error message in trans when attempting to log an unsupported type. 
(Previously, an llvm assert was tripped.)
 2011-05-25 18:54:33 -04:00
Eric Holk
d72d50473f Fixing typechecking for spawn. 2011-05-25 18:01:45 -04:00
Rafael Ávila de Espíndola
94129c4216 dead argument removal part one. 2011-05-25 15:33:57 -04:00
Brian Anderson
9a88ba2b81 rustc: Downcase typeck::demand and typeck::pushdown 2011-05-24 20:26:46 -04:00
Brian Anderson
6473b73174 rustc: Use demand functions instead of open-coded typecheck for spawn 
Un-xfail spawn-non-nil-fn
 2011-05-24 20:23:21 -04:00
Rafael Ávila de Espíndola
de2e84e5b0 Only one gc glue. 2011-05-24 18:52:31 -04:00
Rafael Ávila de Espíndola
ac836dd79c There is only one yield glue. 2011-05-24 18:29:08 -04:00
Rafael Ávila de Espíndola
b1292580b9 Remove dead code for unwind_glue. 2011-05-24 17:28:37 -04:00
Rafael Ávila de Espíndola
0fc91b6ecc There is only one activate function now. 2011-05-24 17:00:45 -04:00
Eric Holk
2e8afc7b47 Merge pull request #404 from lkuper/cleanup 
Just removing some unnecessary imports.
 2011-05-24 13:51:00 -07:00
Marijn Haverbeke
fbc0e840e3 Stop showing a 'cyclic import' message for unresolved imports 
This only happened for single-length import paths, because the import
was being looked up in its own scope.
 2011-05-24 20:58:38 +02:00
Lindsey Kuper
f1d3b88f0a Just removing some unnecessary imports. 2011-05-24 11:22:40 -07:00
Tim Chevalier
bad05e20a3 Improve nonexistent-module error message 
The error message for (for example) "import vec;" without "use std;"
was "cyclic import", which was misleading because there were no
cycles. I changed it to "cyclic import or nonexistent module",
which doesn't break existing tests.
 2011-05-24 10:47:58 -07:00
Tim Chevalier
c9a8bdee2f Improve error message for ambiguous polymorphic types 
If a type still contains free type variables after typechecking (for
example, as with "auto foo = []", the result was an assertion failure
in typeck. Made it a human-readable error message instead.
 2011-05-23 19:46:27 -07:00
Tim Chevalier
b620be9cd2 Typecheck @[] correctly 
Previously, if you wrote
let @vec[int] foo = @[];

that would be a type error. That didn't seem right, so I changed
pushdown to unify the inner type in an unop application with the
argument type of the operator type.
 2011-05-23 17:17:49 -07:00
Tim Chevalier
2a73c4ffab Add line breaks to typestate error messages 2011-05-23 13:32:29 -07:00
Brian Anderson
d82e9c1874 rustc: Add a FIXME to expr_alt type checking 2011-05-21 18:35:21 -04:00
Brian Anderson
4504e23a82 rustc: Remove unnecessary pushdown from check_expr in the expr_if case 2011-05-21 18:18:48 -04:00
Brian Anderson
e24d7ae967 rustc: If all if branches are _|_, then the entire if is _|_ 2011-05-21 18:18:48 -04:00
Brian Anderson
085fb1a2db rustc: Remove outdated comment 2011-05-21 14:43:04 -04:00
Brian Anderson
b62bf8cede rustc: Allow if expressions to fail 2011-05-21 14:22:01 -04:00
Brian Anderson
6824f119fc rustc: Allow alt expressions to fail 2011-05-21 13:22:01 -04:00
Brian Anderson
418b4c4567 rustc: Assign ty_bot to nodes in write::bot_ty. Remove FIXME 2011-05-21 13:02:58 -04:00
Brian Anderson
779dd809d2 rustc: Blocks shouldn't try to return results for ty_bot 2011-05-21 12:56:31 -04:00
Tim Chevalier
755ca8eb73 Correctly check that ! functions always diverge 
Also make _|_ unify with any type.
 2011-05-20 19:54:14 -07:00
Brian Anderson
2fc0bedc83 rustc: Don't write metadata for unexported items 2011-05-20 21:32:57 -04:00
Tim Chevalier
467b938ea8 Merge remote-tracking branch 'graydon/master' into typestate_4 2011-05-20 18:23:03 -07:00
Tim Chevalier
964898eaf1 Error instead of warning for missing returns. Huzzah! 2011-05-20 18:20:54 -07:00
Lindsey Kuper
c3410bf927 More work on anonymous objects. 2011-05-20 17:59:56 -07:00
Tim Chevalier
c8d488b337 Change node_ann hash table to a vector in typestate 2011-05-20 16:57:37 -07:00
Tim Chevalier
9128a1bab8 rm unnecessary logging 2011-05-20 16:57:37 -07:00
Tim Chevalier
7e82f23bc0 rm unused import 2011-05-20 16:57:37 -07:00
Tim Chevalier
551fb13689 Add a "fail" after a while(true) loop in two places 
(in preparation for changing the "function may not return" warning
to an error)
 2011-05-20 16:57:37 -07:00
Tim Chevalier
a08a376d6e Make controlflow part of a function type 
Change ty_fn to have a controlflow field. A 'controlflow' is
essentially a bit of data that says whether or not this function
never returns to the caller (if it never returns, that means it calls
"fail" or another "never-returns" function on every control path).

Also add syntax for annotating functions that never return:

fn foo() -> ! {
   fail;
}

for example. Functions marked with ! implicitly have a result type of
ty_bot, which is a new type that this commit also adds.
 2011-05-20 16:57:37 -07:00
Eric Holk
0de27ce8bd Translated start_task, but it fails on spp == align_down(spp) 2011-05-20 16:51:09 -07:00
Eric Holk
d01948cd07 Called the new_task upcall. There are refcount issues though. 2011-05-20 16:51:08 -07:00
Eric Holk
219351e4b9 A little closure towards translating spawn. We're about ready to do the upcall, except that rustc segfaults. 2011-05-20 16:51:08 -07:00
Eric Holk
d6338a8c15 Added some comments to trans_spawn that sort of explains what needs to happen 2011-05-20 16:51:08 -07:00
Eric Holk
2b124a8140 Started working on translating spawn. It correctly generates the task name in the case that it is not provided. 2011-05-20 16:51:08 -07:00
Graydon Hoare
fa75703700 Twiddle glue inlining heuristics. ~10% win on build time. 2011-05-20 14:57:52 -07:00
Graydon Hoare
67b8501eb8 Elide a bunch of do-nothing take glue. 2011-05-20 14:50:57 -07:00
Graydon Hoare
381ba1d0ba Remove redundant secondary declaration of glue. 2011-05-20 14:48:33 -07:00
Patrick Walton
7e43a31f08 rustc: Move the interner to a new module intended to be used for general data structures 2011-05-20 13:57:53 -07:00
Brian Anderson
b6d2fe4d74 rustc: Translate else if blocks the same as other blocks. Closes #388 2011-05-19 23:32:09 -04:00
Brian Anderson
879a952a37 rustc: Find the correct outer scope in trans_block 2011-05-19 23:32:08 -04:00
Rafael Ávila de Espíndola
40b511a7c5 Revert 106216d673. 
This depends on llvm's 131294.
 2011-05-19 18:30:19 -07:00
Graydon Hoare
494c0c3b61 After all that, give up on "complex" version of type_has_pointers and just manually inspect. Can't calculate property during interning. 2011-05-19 18:28:09 -07:00
Graydon Hoare
7f938f4945 Move the ty_item_table into ty::ctxt.items, move variant_info to ty, and start sketching the contains_pointer property on types. 2011-05-19 17:21:36 -07:00
Graydon Hoare
1d67ee0e88 Remove type_glue.rs, don't think it's going to work. 2011-05-19 17:21:36 -07:00
Patrick Walton
bc87ecf110 rustc: Unify variables using a single type instead of keeping a list and doing it all at the end. This will give us better error messages. 2011-05-19 16:07:12 -07:00
Graydon Hoare
a4dd57705c Move type cache and node type table into type context. 2011-05-19 15:47:15 -07:00
Marijn Haverbeke
8a7b544995 Finish name collision testing. 
And fix a the single screw-up in the source code it caught.
 2011-05-19 21:58:57 +02:00
Paul Stansifer
533d3055f3 No functional change. Largely removed the native module/non-native module distinction from resolve.rs 2011-05-19 12:13:35 -07:00
Tim Chevalier
b74d0d7c3d Typestate checking code no longer uses fold -- only walk. 2011-05-19 11:40:16 -07:00
Tim Chevalier
d60e08c18e Eliminate explicit recursion in tstate.states (walk does that now), 
and kill dead code.
 2011-05-19 11:40:16 -07:00
Tim Chevalier
e1844d2719 Add ann as an argument to visit_fn 2011-05-19 11:40:16 -07:00
Tim Chevalier
6dcd607277 make trans compile 2011-05-19 11:40:16 -07:00
Tim Chevalier
a9bc6715e7 tidy tstate.collect_locals 2011-05-19 11:40:16 -07:00
Tim Chevalier
c0f728712b Rewrite pre_postconditions to use walk instead of fold 2011-05-19 11:40:16 -07:00
Tim Chevalier
2cd769e358 Add ident as an argument to visit_fn (typestate wants it) 2011-05-19 11:40:16 -07:00
Tim Chevalier
ef5473deb1 Trim imports 2011-05-19 11:40:16 -07:00
Tim Chevalier
bb594a03a1 remove now-unused ts field from ann 2011-05-19 11:40:16 -07:00
Tim Chevalier
e16b097599 Rewrite tstate.annotate to use walk instead of fold 
and various other tidying in typestate
 2011-05-19 11:40:16 -07:00
Tim Chevalier
dc83c84662 Add visit_fn_pre and visit_fn_post fields to visitors 
Added visit_fn_pre and visit_fn_post fields that get applied to
both regular functions and methods.
 2011-05-19 11:40:16 -07:00
Tim Chevalier
8cbdaf4f93 Make trans use span_err for the dreaded "ty_var in trans::type_of" error 
This required quite a bit of tiresome plumbing about of spans.
On the bright side, now other errors can be converted to span_err too.

Includes test cases.
 2011-05-19 11:40:16 -07:00
Tim Chevalier
f909dd114d Remove dead file (typestate_check is now a bunch of modules under middle/tstate) 2011-05-19 11:40:16 -07:00
Tim Chevalier
b624b43e47 Beginning of Operation Eliminate Folds 
Started by rewriting collect_locals to use walk instead of fold
 2011-05-19 11:40:16 -07:00
Tim Chevalier
4964023c5e Fix an error message that reversed the expected and actual number of arguments in a pattern 2011-05-19 11:40:16 -07:00
Patrick Walton
6417610788 rustc: Generalize variable bindings so that we can use it for locals too 2011-05-19 11:14:44 -07:00
Patrick Walton
df90934bf7 rustc: Split write_* in typeck into versions that will record fixups 2011-05-18 19:19:20 -07:00
Patrick Walton
d2bdfe36cf rustc: Don't inline free glue 2011-05-18 18:48:57 -07:00
Graydon Hoare
6f8b9e176f Factor free paths out of drop glue, into free glue as in rustboot. 2011-05-18 17:28:08 -07:00
Patrick Walton
0aee8b3bd7 rustc: Thread statement contexts through unification 2011-05-18 15:36:14 -07:00
Rafael Ávila de Espíndola
bb7d4e15a8 Remove dead code. 2011-05-18 16:33:22 -04:00
Marijn Haverbeke
28c4a5586e Start checking name collisions in modules 
(And remove a boatload of duplicate imports from the typestate files.)

Name collision checking in local scopes will follow soon.
 2011-05-18 22:24:20 +02:00
Patrick Walton
13d9f6a264 rustc: Split out type variable fixups from unification 2011-05-18 13:08:54 -07:00
Patrick Walton
37d5dd9471 rustc: Group the variable binding machinery into a separate data structure during unification 2011-05-18 11:54:12 -07:00
Rafael Ávila de Espíndola
a795d0e5b2 And yet more dead code. 2011-05-18 14:28:08 -04:00
Patrick Walton
f4e049c415 rustc: Thread a statement context through the typechecker; use it for generating variable IDs. 40% typechecking speedup. 2011-05-17 22:21:31 -07:00
Patrick Walton
fdf8c49882 rustc: Don't rebuild the AST when typechecking statements 2011-05-17 21:09:34 -07:00
Patrick Walton
ace8058277 rustc: Have typechecking no longer rebuild the AST 2011-05-17 20:13:04 -07:00
Brian Anderson
a98ea4f3ef rustc: Run block cleanups on else if blocks 
With the scheme used to translate 'else if' currently the if expression is
translated in a new (else) scope context. If that if expression wants to
result in a value that requires refcounting then it will need to drop the
refcount in the cleanups of the else block.
 2011-05-17 22:25:24 -04:00
Patrick Walton
e840a37f33 rustc: Remove typeck::check_item_fn(); it's entirely superfluous! 2011-05-17 19:20:24 -07:00
Patrick Walton
87288a186d rustc: Make check_const() not rebuild the AST (although it's not called at the moment!) 2011-05-17 19:13:16 -07:00
Patrick Walton
d21741843c rustc: Make check_block not rebuild the AST 2011-05-17 19:10:24 -07:00
Patrick Walton
10503a4a02 rustc: Don't rebuild the AST when writing back locals 2011-05-17 19:00:29 -07:00
Patrick Walton
8637a4eaf3 rustc: Pull the type out of the correct expression when typechecking channel types. Puts out burning tinderbox. 2011-05-17 18:19:45 -07:00
Patrick Walton
83e1dc63c2 rustc: Fix return value of expr_spawn case in pushdown_expr. Should put out soon-to-be-burning tinderbox. 2011-05-17 17:44:30 -07:00
Patrick Walton
c6d4c31ee4 rustc: Don't rebuild the AST when typechecking expressions 2011-05-17 17:43:06 -07:00
Eric Holk
cff9a3cc2a Fixing compile problems from recent changes. 2011-05-17 14:38:32 -07:00
Eric Holk
cf02c67137 Removing the TODO comment. 2011-05-17 14:38:32 -07:00
Eric Holk
8d10594ed5 Added a few more expr_spawn cases so the spawn.rs test case gets a little further before failing. 2011-05-17 14:38:32 -07:00
Patrick Walton
37768cba92 rustc: Don't rebuild call or bind expressions during typechecking 2011-05-17 14:37:23 -07:00
Patrick Walton
f300de2963 rustc: Make return value checking warnings a little prettier. Also introduce a new "note" diagnostic level. 2011-05-17 14:12:49 -07:00
Patrick Walton
5047ab0b0c rustc: Flatten annotations 2011-05-17 13:55:32 -07:00
Marijn Haverbeke
09d8ef8d51 Finally rename std::_xxx to std::xxx 
Except for _task, which is still a keyword.
 2011-05-17 20:41:41 +02:00
Brian Anderson
fbfd8552ab Return a better result from blocks. Closes issue #377 
Blocks return in a copy of the result of their ending expression, not the
direct result of the ending expression, as that may be a local variable which
gets zeroed by drop_slot.
 2011-05-17 00:46:56 -04:00
Graydon Hoare
4c5e315843 Rename aux.rs to auxiliary.rs since win32 doesn't like files named "aux". Really. 2011-05-16 19:21:29 -07:00
Graydon Hoare
a2e2e78103 Merge remote branch 'origin/master' into HEAD 
Conflicts:
	src/comp/middle/trans.rs
 2011-05-16 19:04:45 -07:00
Graydon Hoare
fbbc1a77d2 Rewrite everything to use [] instead of vec() in value position. 2011-05-16 18:21:22 -07:00
Tim Chevalier
971b5d5151 Started adding support for return checking and non-returning function annotations 
* Reorganized typestate into several modules.

* Made typestate check that any function with a non-nil return type
  returns a value. For now, the check is a warning and not an error
  (see next item).

* Added a "bot" type (prettyprinted as _|_), for constructs like be, ret, break, cont, and
  fail that don't locally return a value that can be inspected. "bot"
  is distinct from "nil". There is no concrete syntax for _|_, while
  the concrete syntax for the nil type is ().

* Added support to the parser for a ! annotation on functions whose
  result type is _|_. Such a function is required to have either a
  fail or a call to another ! function that is reached in all control
  flow paths. The point of this annotation is to mark functions like
  unimpl() and span_err(), so that an alt with a call to err() in one
  case isn't a false positive for the return-value checker. I haven't
  actually annotated anything with it yet.

* Random bugfixes:

* * Fixed bug in trans::trans_binary that was throwing away the
    cleanups for nested subexpressions of an and or or
    (tests: box-inside-if and box-inside-if2).

** In typeck, unify the expected type arguments of a tag with the
   actual specified arguments.
 2011-05-16 16:59:25 -07:00
Tim Chevalier
c75125fcce Noticed that metadata::Encode::enc_sty didn't have a case for ty_task, so I added one (though this wasn't causing the problem I was seeing... 2011-05-16 16:13:11 -07:00
Patrick Walton
f74298215c rustc: Fix indentation in typeck::check_lit() 2011-05-16 15:24:32 -07:00
Patrick Walton
1fc1ad91ce rustc: Don't rebuild the call expression in typeck::replace_expr_type() 2011-05-16 15:20:11 -07:00
Patrick Walton
278d64dafc rustc: Don't unpack annotations in trans::lval_generic_fn() 2011-05-16 15:20:11 -07:00
Patrick Walton
32b8dcb97c rustc: Factor out the code that interns types into an "interner" data structure 2011-05-16 13:58:13 -07:00
Patrick Walton
3760bfab87 rustc: Don't rebuild AST patterns during typechecking 2011-05-16 11:45:00 -07:00
Patrick Walton
ca51b49c0a rustc: Replace the redundant typeck::node_ann_ty_params() with ty::ann_to_type_params() 2011-05-16 11:43:50 -07:00
Patrick Walton
3aaf02b907 rustc: Make the type collection pass no longer rebuild the AST 2011-05-16 10:34:32 -07:00
Patrick Walton
525a5c582f rustc: Fix indentation in ty_param_count_and_ty_for_def() 2011-05-16 10:32:38 -07:00
Patrick Walton
594c70f788 rustc: Add an item_ann() accessor function 2011-05-16 10:32:38 -07:00
Brian Anderson
f77c5ba512 rustc: Wait until generic glue is actually emitted to define it as internal 
The verifier was not liking this when generating unoptimized bitcode. Per LLVM
docs it is not valid for function declarations to be marked internal, but
their implementations may be.
 2011-05-15 19:53:06 -04:00
Brian Anderson
8eaafdaa2d rustc: Rename 'Unify' modules to 'unify' 2011-05-14 21:13:47 -04:00
Patrick Walton
de421afa55 rustc: Use walk instead of fold for the first item collection pass 2011-05-14 11:03:10 -07:00
Patrick Walton
8b3f359442 rustc: Change "Collect" to "collect" 2011-05-14 11:03:10 -07:00
Patrick Walton
85120d9959 rustc: Look types up in the side table 2011-05-13 18:46:44 -07:00
Patrick Walton
ef75860a0a rustc: Write nil types into the node type table wherever plain_ann() is used 2011-05-13 17:55:07 -07:00
Lindsey Kuper
5679f5c55e 'with' no longer a token; whitespace police. 
Plus renaming the anonymous objects test to a more descriptive name,
and XFAILing it because it doesn't work yet.
 2011-05-13 17:37:01 -07:00
Lindsey Kuper
cd31d5d532 More anon obj work; whitespace police in middle::fold 
Passing args to middle::fold::fold_expr_anon_obj by reference to be
consistent with the other folds; adding a dummy fold_expr_anon_obj to
typeck to be filled in later.
 2011-05-13 17:35:13 -07:00
Lindsey Kuper
814b17352c Use new module namespace syntax. 2011-05-13 17:35:13 -07:00
Lindsey Kuper
b6f35c6a4b More work toward anonymous objects. 2011-05-13 17:35:13 -07:00
Lindsey Kuper
d3242b9644 Bug fixes. 
Fixed infinite loop on anonymous objects in parser; added
expr_anon_obj to walk.rs; fixed syntax of test case.
 2011-05-13 17:35:13 -07:00
Lindsey Kuper
6c1c5b3948 More progress on anonymous objects. 
Still segfaulting on the method-overriding.rs test, though.
 2011-05-13 17:35:13 -07:00
Patrick Walton
fac1342513 rustc: Move replace_expr_type() from ty to typeck, as it's only used in the latter 2011-05-13 17:06:30 -07:00
Brian Anderson
843e9435ab rustc: Fix long lines in typeck.rs 2011-05-13 20:06:02 -04:00
Patrick Walton
23d3b0ec57 rustc: Get ann_to_type_params() on board the ann_to_ty_param_substs_opt_and_ty() train 2011-05-13 16:53:35 -07:00
Patrick Walton
0739849e9f rustc: Make all type lookups go through the one ty::ann_to_ty_param_substs_opt_and_ty() function 2011-05-13 16:40:21 -07:00
Patrick Walton
4f2eb31870 rustc: Write types contained in "trivial annotations" to the table 2011-05-13 16:27:55 -07:00
Patrick Walton
e888fe6f9b rustc: Change ty::triv_ann() to take a node ID instead of a full annotation 2011-05-13 14:39:00 -07:00
Brian Anderson
2f440b1452 rustc: lowercase the link and link::write modules 2011-05-13 16:55:37 -04:00
Graydon Hoare
0d32ff723a Slight adjustments to shape, rename to type_glue.rs. 2011-05-13 13:29:50 -07:00
Patrick Walton
c97b05141a rustc: Add write_type() wherever ann_types are written, except for triv_ann() 2011-05-13 13:06:34 -07:00
Patrick Walton
be9aa1cc5e rustc: Fix the type of node_types; stub the write_type function 2011-05-13 12:12:54 -07:00
Patrick Walton
6e34679a65 rustc: Remove a minor rustboot workaround in typeck::resolve_local_types_in_block() 2011-05-13 11:23:00 -07:00
Patrick Walton
8a8f68aa75 rustc: Make typeck::instantiate_path() not return an annotation 2011-05-13 10:53:55 -07:00
Marijn Haverbeke
ae26b775b4 Implement module namespaces 
Module names no longer clash with type and value names. The
tokenizer/parser still needs to be taught to be more careful in
identifying keywords, so that we can use 'str' and 'vec' and so as
module names.
 2011-05-13 17:20:46 +02:00
Marijn Haverbeke
eb419fd8c7 Extend crate format to allow multiple definitions for a single name 
The type/value namespace distinction pretty much works now. Module
namespace is up next.
 2011-05-13 17:20:39 +02:00
Marijn Haverbeke
1a12a7b04b Make module indices hold a list of items 
This way, they can support having both a type and a value
of the same name.
 2011-05-13 13:34:59 +02:00
Marijn Haverbeke
58ec5d1654 Move capture checking into resolve.rs 
Drops capture.rs. The new algorithm also checks for captures function
arguments and obj fields.
 2011-05-13 12:44:05 +02:00
Marijn Haverbeke
89490e416b Change resolve to use walk instead of fold 
Possibly, at some point, we should add a state-passing variant of
walk, or a wrapper that makes it easier to thread state. (See the
repetetive pop_state_for_* functions in this commit.)
 2011-05-13 11:35:25 +02:00
Marijn Haverbeke
7082bf4478 Ensure visit_ty is called on type parameters during walk 2011-05-13 11:35:25 +02:00
Marijn Haverbeke
39a000714a Add visit_arm and visit_method to walk.rs 
The resolver needs to update its state for individual arms and
methods.
 2011-05-13 11:34:51 +02:00
Patrick Walton
5295a4dabd rustc: Flatten recursive vectors in trans::simplify_type(). Prevents an infinite loop. 2011-05-12 18:37:28 -07:00
Patrick Walton
b163247d4e rustc: Beginnings of a "shape" module to convert types to shapes 2011-05-12 18:30:21 -07:00
Patrick Walton
033b38b636 rustc: Make ty::expr_ann() terser 2011-05-12 16:57:08 -07:00
Patrick Walton
0b7dd0d918 rustc: Pass a node type table around, unused as of yet 2011-05-12 16:41:55 -07:00
Graydon Hoare
5ceaf345ed Add stats option and lazily emit glue. 2011-05-12 15:42:51 -07:00
Marijn Haverbeke
3816e57fd2 Downcase std modules again, move to :: for module dereferencing 
This should be a snapshot transition.
 2011-05-12 21:30:44 +02:00
Marijn Haverbeke
6510f1ce7c Change module dereference syntax from . to :: 
This will need to be a snapshot.
 2011-05-12 21:30:35 +02:00
Marijn Haverbeke
358a1aeec9 Keep resolve data in external hash table, rather than embedded defs 
One step closer to removing fold and having a single, immutable AST.
Resolve still uses fold, because it has to detect and transform
expr_field expressions. If we go through on our plan of moving to a
different syntax for module dereferencing, the parser can spit out
expr_field expressions, and resolve can move to walk.

(I am truly sorry for the things I did in typestate_check.rs. I expect
we'll want to change that to walk as well in the near future, at which
point it should probably pass around a context record, which could
hold the def_map.)
 2011-05-12 15:40:21 +02:00
Marijn Haverbeke
2b36e40c58 Ensure ann tags are actually kept around during typechecking 
This way, the tag assigned by the parser stays with the node.
I realize ann replacing is probably going away real soon, but
I needed this now for moving the resolve defs out of the AST.
 2011-05-12 10:51:13 +02:00
Marijn Haverbeke
d719b5cae7 Provide a more useful message when failing to translate a const 2011-05-12 09:17:59 +02:00
Patrick Walton
531515d7ae rustc: Represent types as uints. Cuts typechecking down from 14s to 12s. 2011-05-11 17:05:39 -07:00
Patrick Walton
da09a0320e rustc: Remove a few kludges intended to work around rustboot's lack of structural comparison from equal_type_structures() 2011-05-11 16:18:02 -07:00
Patrick Walton
6953841e25 rustc: Remove magic numbers; they aren't a win 2011-05-11 16:10:17 -07:00
Patrick Walton
df914e8904 rustc: Intern all types. 4s regression. 2011-05-11 15:56:23 -07:00
Patrick Walton
26ab36afb1 rustc: Hoist derived type descriptors to the top of the function 2011-05-11 12:05:02 -07:00
Marijn Haverbeke
14f1fe0e29 Remove mod indices from the AST 
They are now created by the resolve pass, which is the only pass that
needs them, and kept internal to that pass.
 2011-05-11 17:11:44 +02:00
Marijn Haverbeke
fe29d24b6e Get rid of arm indices 2011-05-11 15:26:36 +02:00
Marijn Haverbeke
5405f45274 Get rid of block indices 2011-05-11 15:10:24 +02:00
Marijn Haverbeke
7d086df095 Stop depending on block indices in capture.rs 2011-05-11 13:49:48 +02:00
Marijn Haverbeke
9cb20153a8 Hash only on def_num when storing def_ids that are local 2011-05-11 13:06:43 +02:00
Marijn Haverbeke
7f222423b7 Use a const for crate-num-zero (ast.local_crate) 2011-05-11 12:59:34 +02:00
Marijn Haverbeke
6b46113afc Cache external crate lookups in resolve.rs 
According to --time-passes, resolution went from 2 to 0 seconds. Not
really the bottleneck... but if we want to be crazy fast, just
consider this a future bottleneck that was fixed very timely.
 2011-05-11 12:32:37 +02:00
Marijn Haverbeke
e9c12ab1d0 Rewrite comp/middle/resolve.rs 
* Cleans up the algorithm
 * Move first pass to walk (second still folds)
 * Support part of a type/value namespace split
   (crate metadata and module indices still need to be taught about this)
 * Remove a few blatant inefficiencies (import tables being recreated for
   every lookup, most importantly)
 2011-05-11 12:32:37 +02:00
Graydon Hoare
dc0aab47a7 More alias-ification of trans. 2011-05-11 05:11:59 +00:00
Patrick Walton
f627c2ff21 rustc: Don't interleave allocas with stores when copying arguments to allocas 2011-05-10 18:29:14 -07:00
Graydon Hoare
390ce40165 Begin alias-ifying trans. 2011-05-10 18:07:40 -07:00
Graydon Hoare
cf3e737eee Avoid mangle_name_by_type_only call in tydesc names unless debugging. 2011-05-10 18:07:40 -07:00
Patrick Walton
1448fc1a24 rustc: Lower the size limit on ty_to_short_str() and always call it in favor of direct calls to metadata.Encode.ty_str() in trans 2011-05-10 17:50:39 -07:00
Patrick Walton
10c9266525 rustc: Number everything with an annotation 2011-05-10 17:27:59 -07:00
Graydon Hoare
7e8b57d1bb Only name LLVM values / blocks / types when saving temporaries. 2011-05-10 17:04:41 -07:00
Patrick Walton
17daf1f9bf rustc: Cache intermediate results of metadata.Encode.ty_str. Shaves 30% off translation time. 2011-05-10 14:48:24 -07:00
Graydon Hoare
3bec4c8193 Switch memcpy to memmove since we're willing to compile x = x and such. 2011-05-10 14:22:14 -07:00
Patrick Walton
2849103f24 rustc: Get rid of native Rust glue for all upcalls except the exit task glue 2011-05-10 14:04:34 -07:00
Graydon Hoare
55f0f79a46 Re-enable use of llvm intrinsics for memcpy, memset. Should have aligned stack now. 2011-05-10 11:50:29 -07:00
Brian Anderson
1abb7a868e Fix long lines in fold.rs, typeck.rs 2011-05-09 22:11:56 -04:00
Graydon Hoare
4539f8a03c Alias-ify fold and its users, remove another 85kb. 2011-05-09 17:50:06 -07:00
Graydon Hoare
476442e48c Alias-ify walk, typeck, metadata. Cuts another 80kb off rustc. 2011-05-09 16:17:17 -07:00
Patrick Walton
071ed380fb rustc: Declare upcalls with real signatures; preparation for killing rust_native_rust_* stubs 2011-05-09 15:33:02 -07:00
Patrick Walton
70c759030c rustc: Alias fix part 2 -- Check that the aliasness of function parameters matches. Add a test case. 2011-05-09 14:00:50 -07:00
Graydon Hoare
b2b33de819 Alias-ify a variety of fn signatures in ty. Cuts 180kb off rustc. 2011-05-09 13:09:20 -07:00
Patrick Walton
662e949540 rustc: Alias fix part 1 -- Separate out AST modes from typechecker modes, and introduce an "either value or alias" mode 2011-05-09 12:27:03 -07:00
Patrick Walton
106216d673 rustc: Stop emitting zero-length arrays 2011-05-08 16:26:35 -07:00
Graydon Hoare
db3b9a4992 Stop stringifying integers to get integral constants. 2011-05-07 18:54:23 +00:00
Marijn Haverbeke
a3ec0b1f64 Rename std modules to be camelcased 
(Have fun mergining your stuff with this.)
 2011-05-06 22:51:19 +02:00
Patrick Walton
5e2088f2fa rustc: For stack local derived tydescs, allocate the root tydesc separately from the parameter tydescs. This allows SROA, and consequently inlining, to work. 2011-05-06 12:37:59 -07:00
Patrick Walton
57a5c3ac9f rustc: Factor derived tydesc construction out of get_tydesc() 2011-05-06 12:37:59 -07:00
Patrick Walton
8e7aeed2d5 rustc: Include a representation of the type in question in the symbols generated for tydescs 2011-05-06 12:37:59 -07:00
Patrick Walton
883eb38b28 rustc: Remove some debugging code from decl_native_fn_and_pair() 2011-05-05 17:58:48 -07:00
Patrick Walton
81cda5ad45 rustc: Fix a bunch of argument-passing bugs in decl_native_fn_and_pair() 2011-05-05 17:48:10 -07:00
Lindsey Kuper
9d7fc21967 Consolidating expr_to_str functions. 2011-05-05 16:53:25 -07:00
Rafael Ávila de Espíndola
157f61f8c8 Change the setup so that rust_activate_glue returns to rust_exit_task_glue 
and rust_exit_task_glue calls the rust main.

This is simpler since we only need to setup one frame. It also matches
what ld.so does, so gdb is happy and stops a backtrace at rust_exit_task_glue
instead of continuing past whatever function happened to be before
rust_exit_task_glue is the object file.

This is only the rustc changes and should be merged first.
 2011-05-05 15:14:34 -07:00
Patrick Walton
2d9ccf3ae5 rustc: Add Rust intrinsic support to the crate metadata reader and writer 2011-05-05 14:35:18 -07:00
Patrick Walton
9a1f0977a2 rustc: Fix bug in equal_abi that was preventing Rust intrinsics from working 2011-05-05 14:21:59 -07:00
Tim Chevalier
6b742aec91 Enforce in typechecker that preds return a bool 
as well as a test case
 2011-05-05 11:26:07 -07:00
Tim Chevalier
e3a68e235c Bring back "pred" syntax for writing predicates for check 
This commit reinstates the requirement that the predicate in a
"check" must be a manifest call to a special kind of function
declared with the new "pred" keyword instead of "fn". Preds must
have a boolean return type and can only call other preds; they
can't have any effects (as enforced by the typechecker).
The arguments to a predicate in a check expression must be
slot variables or literals.
 2011-05-05 11:26:07 -07:00
Tim Chevalier
3060eadcba Check well-formedness of constraints 
Check that the operand in a constraint is an explicit name,
and that the operands are all local variables or literals. Still need
to check that the name refers to a pure function.
 2011-05-05 11:26:07 -07:00
Tim Chevalier
4f892dd9d7 Check that the operand in a check is a call 
In addition, fix bug in fold that was turning asserts into checks.

More typechecking still needs to be done.
 2011-05-05 11:26:07 -07:00
Marijn Haverbeke
1f45dda9d2 Remove 'deprecated mutable...' from our code 
This should make compilation a bit less noisy.
 2011-05-05 20:19:43 +02:00
Marijn Haverbeke
3d738e9e06 Return a fresh, unreachable context after ret, break, and cont 
This ensures we don't get compile errors on unreachable code (see
test/run-pass/artificial-block.rs for an example of sane code that
wasn't compiling). In the future, we might want to warn about
non-trivial code appearing in an unreachable context, and/or avoid
generating unreachable code altogether (though I'm sure LLVM will weed
it out as well).
 2011-05-05 20:19:43 +02:00
Marijn Haverbeke
9432626b68 Eradicate fold from capture.rs 
The pass now uses walk.
 2011-05-05 20:19:43 +02:00
Patrick Walton
fea623211f rustc: Add a Link module; move crate writing to it to slim down trans slightly 2011-05-05 10:48:02 -07:00
Marijn Haverbeke
898e089d7b Fix and re-enable destructors 
There was some confusion on whether the destructors took their
argument by pointer or direct value. They now take it directly, just
like other methods. You no longer get a segfault when a constructor
actually does something with its self value.
 2011-05-05 02:16:49 +02:00
Graydon Hoare
cdb6822405 Add a structure for passing option flags around the compiler, put it in session, and use it. 2011-05-04 16:59:21 -07:00
Brian Anderson
bd34770a92 Add missing ret statements to check_mod 
Un-XFAIL compile-fail export tests
 2011-05-04 19:29:27 -04:00
Graydon Hoare
ce9468761c Support new -g flag, only mangle glue names by type when it's passed. 2011-05-04 15:36:42 -07:00
Patrick Walton
ac8eb20224 rustc: Stub support for Rust intrinsics 2011-05-03 18:03:59 -07:00
Patrick Walton
e43729ccf2 rustc: Refactor metadata.Encode.* to not require a type abbreviation table if abbreviation isn't enabled 2011-05-03 14:41:05 -07:00
Patrick Walton
3f317ede41 rustc: Name type glue properly 2011-05-03 14:41:05 -07:00
Brian Anderson
ed40c85af5 Extract ast.is_exported from the resolve module 2011-05-02 22:07:36 -04:00
Patrick Walton
ebc236e91c rustc: Time the Rust translation and LLVM passes separately 2011-05-02 17:50:46 -07:00
Patrick Walton
a833f152ba rustc: Remove the artificial block in ty.rs. Fixes self-hosting. Add a test case for this, XFAIL'd. 2011-05-02 17:50:46 -07:00
Patrick Walton
147a2d655f Un-revert "Use different syntax for checks that matter to typestate", fixing the problem. 
This reverts commit d08b443fff.
 2011-05-02 17:50:46 -07:00
Graydon Hoare
d08b443fff Revert "Use different syntax for checks that matter to typestate" 
This reverts commit aa25f22f19. It broke stage2, not sure why yet.
 2011-05-02 17:35:33 -07:00
Patrick Walton
9d21cf3b61 rustc: Allocate tydescs on the stack when it's safe to do so. 60% compile speed increase. 2011-05-02 15:28:59 -07:00
Tim Chevalier
aa25f22f19 Use different syntax for checks that matter to typestate 
This giant commit changes the syntax of Rust to use "assert" for
"check" expressions that didn't mean anything to the typestate
system, and continue using "check" for checks that are used as
part of typestate checking.

Most of the changes are just replacing "check" with "assert" in test
cases and rustc.
 2011-05-02 12:16:29 -07:00
Patrick Walton
870435caf5 rustc: Add a "fat tydesc" LLVM type to trans 2011-05-02 12:05:21 -07:00
Brian Anderson
1dd63ff42d Remove the search direction from resolve's fold environment 
It's not actually involved in the fold so it can just be passed between the
functions that need it.
 2011-05-01 16:57:36 -04:00
Brian Anderson
a697210234 Hide unexported tag variants 2011-05-01 16:57:36 -04:00
Brian Anderson
459b0ec833 Implement simple module export 2011-05-01 16:57:36 -04:00
Patrick Walton
b101e26d92 rustc: Run scope cleanups in the implicit block created by log statements. Fixes a leak. 2011-04-29 19:19:54 -07:00
Patrick Walton
f6c472d816 rustc: Emit pure native glue; we don't call it yet 2011-04-29 16:40:30 -07:00
Patrick Walton
593fb195a4 rustc: Start threading a purity flag through upcalls 2011-04-29 15:39:48 -07:00
Graydon Hoare
b9babe47da Temporarily disable dtors; they broke self-hosting. 2011-04-29 15:06:53 -07:00
Patrick Walton
1acf404361 rustc: First steps to determine whether tydescs escape 2011-04-29 15:01:07 -07:00