about summary refs log tree commit diff
path: root/src/boot
AgeCommit message (Collapse)AuthorLines
2010-07-28Get slots in trans_tag using Semant tables. Closes #133.Roy Frostig-5/+10
2010-07-28Test the deque a bit. Give it a get-by-index method. Fix two uncovered ↵Roy Frostig-6/+32
state-calculation bugs --- one decently, the other with an ugly hack. Bug on the latter coming right up.
2010-07-27Switch machine-type lexemes to use suffixes. Remove support for foo(bar) as ↵Graydon Hoare-132/+79
a cast notation. Closes #129.
2010-07-27Calm some LLVM indigestion of last change.Graydon Hoare-6/+6
2010-07-27Distill semantics of use-def maps to fewer and more-obvious words.Graydon Hoare-180/+207
- Remove redundant uses of 'resolve' and 'referent' in semant. - Use defn, defn_id, lval, lval_base more consistently. - Make associated query functions more consistent. - Closes #127.
2010-07-26Slight shaving on RA, no more optimizing today. It's fast enough for now.Graydon Hoare-8/+9
2010-07-26Fix numerous non-linearities and O(sizeof(crate)) issues in typestate ↵Graydon Hoare-126/+340
system's dataflow algorithm. No longer substantial in profile.
2010-07-26Note the typestate dataflow timing. The culprit is revealed.Graydon Hoare-1/+2
2010-07-26Add a cumulative timer count to be sure we're getting everything.Graydon Hoare-3/+6
2010-07-26Change ctxt_crate_main to an Ast.name.Graydon Hoare-8/+7
2010-07-26Do some more iflog-guarding.Graydon Hoare-84/+94
2010-07-25Don't write to NULL after calling C natives returning void.Roy Frostig-6/+13
2010-07-23Another quick fix from jmuizelaar for that "some ocaml versions dont't ↵Graydon Hoare-1/+1
notice wrong wildcard count in patterns" bug.
2010-07-23Improve mutability checking. Closes #118.Graydon Hoare-12/+33
2010-07-23Add test for writing-through-uninit bug (reported on IRC by jrmuizel), plus ↵Graydon Hoare-1/+20
fix in typestate system.
2010-07-23Stop using project_lval_ty_from_slot for lval_ty; cover residual un-caught ↵Graydon Hoare-15/+19
cases in type.ml.
2010-07-23Widen write mutability check to cover all writing stmts.Graydon Hoare-10/+22
2010-07-23Include all lval-writing statements in stmt_is_init calculation, not just ↵Graydon Hoare-15/+15
"copy-like". Un-XFAIL generic-tag-alt.rs
2010-07-23Add pretty-printing for alt-tag statements.Graydon Hoare-1/+32
2010-07-23Rename STMT_init_* to STMT_new_*; former name was confusing.Graydon Hoare-120/+108
2010-07-22Notify copy glue of dst-initialization and fix _vec.alloc issues in lib and ↵Roy Frostig-8/+28
runtime. Closes #109.
2010-07-22A certain incomplete quantity of wrestling with "INIT" statements that don't ↵Graydon Hoare-28/+76
actually initialize. Should probably rename them to MAKE. Anyway, WIP, but two steps forward (and one back). More later.
2010-07-22Re-classify some err / bug cases as unimpl. Ideally rustboot should never ↵Graydon Hoare-11/+14
produce a backtrace.
2010-07-22Source FIXME annotations for issue #81 and a correction to STMT_bind fmt.Roy Frostig-2/+5
2010-07-22Add XFAIL'ed test for return-in-iter, call unimpl when we find it. Closes #100.Graydon Hoare-0/+2
2010-07-22Fix mem op= mem bug in trans.ml (via not terribly good fix). Closes #111.Graydon Hoare-1/+8
2010-07-23Fix simple generic type parameters in LLVM.Jeffrey Yasskin-3/+28
2010-07-23Implement tuple access for LLVM.Jeffrey Yasskin-20/+77
This involved adding an Ast.ty return to trans_lval. I also included the code for record and box access, but their tests don't completely pass yet.
2010-07-22Make sure debug locations are propagated to every new block created by makingJeffrey Yasskin-19/+47
them an argument to new_block. gdb should do a good job of single-stepping LLVM-compiled rust programs by rust line number now.
2010-07-20Make bound functions have the right typesPatrick Walton-1/+21
2010-07-20Pick off the last LLVM-crasher from mbx's change.Graydon Hoare-1/+4
2010-07-20Adjust llasm, llabi to be aware of changes to rust_task structure. Still not ↵Graydon Hoare-4/+18
working fully.
2010-07-20Tidy up handling of unimplemented features. These are expected (if ↵Graydon Hoare-54/+90
undesirable) sorts of error, we should handle better than "backtrace and exit 2".
2010-07-19Strip off TY_mutables and TY_constraineds when unboxing values during autoderefPatrick Walton-0/+1
2010-07-19ctxt_auto_deref_lval decides whether to autoderef the entire lval, not its base.Roy Frostig-17/+24
2010-07-19Autoderef objects when passing them as implicit (indirect) arg upon ↵Roy Frostig-2/+11
vtbl-dispatch. Add testcase and XFAIL it on LLVM. Closes #112.
2010-07-19Include the statements on the LHS of a binop when desugaring. Closes #117.Patrick Walton-56/+57
2010-07-19Added a message passing system based on lock free queues for inter-thread ↵Michael Bebenita-1/+1
communication. Channels now buffer on the sending side, and no longer require blocking when sending. Lots of other refactoring and bug fixes.
2010-07-19Add a test for an obvious-seeming (but not actually legal) kind of cast ↵Graydon Hoare-3/+1
attempted in issue #115, downgrade bug to an err in type.ml so you get a better message.
2010-07-19Fix over-optimistic resolution of self-methods within obj scopes. There is ↵Graydon Hoare-5/+1
no such feature in the language at present. Add test to prevent regression. Closes #114.
2010-07-19Missing semicolon in type.ml, plus test to catch regression. Closes #113.Graydon Hoare-1/+1
2010-07-19Change the body-fetching GEP in mark_ty to be dynamic. Our box may be ↵Roy Frostig-1/+1
parametric.
2010-07-20Fix opeq.rs for LLVM by implementing simple augmented-assignment operators.Jeffrey Yasskin-17/+32
2010-07-20Make the LLVM compiler crash when it hits a source construct it doesn't knowJeffrey Yasskin-17/+40
what to do with, rather than silently omitting it from the output.
2010-07-20Teach the LLVM backend to emit line number information.Jeffrey Yasskin-27/+87
This isn't complete: it misses line numbers in certain kinds of block. It's also still tricky to use it on Darwin, where we need to call dsymutil while the .o file is still around in order for gdb to find the debugging symbols.
2010-07-18Stub an interface to the (as-yet-nonexistent) structural comparison glue in ↵Patrick Walton-24/+99
trans
2010-07-18Fix argv.rs under the LLVM compiler.Jeffrey Yasskin-9/+9
The call to rust_start was assuming that all rust main() functions have the same signature, but the compiler doesn't actually canonicalize them. So instead just match the C signature of rust_start, and cast.
2010-07-16Fix IL translation of pattern-alt to allow a value of mutable/constrained ↵Roy Frostig-6/+7
type to be the subject of a pattern-alt.
2010-07-16Ensure that functions that should return a value do; issue 41Patrick Walton-6/+32
2010-07-16Feed the correct return type to the typechecker when typechecking objects, ↵Patrick Walton-14/+24
and add a testcase.