aboutsummaryrefslogtreecommitdiff
path: root/src/boot/me/trans.ml
Commit message (Collapse)AuthorAgeFilesLines
...
* Add XFAIL'ed test for return-in-iter, call unimpl when we find it. Closes #100.Graydon Hoare2010-07-221-0/+2
|
* Fix mem op= mem bug in trans.ml (via not terribly good fix). Closes #111.Graydon Hoare2010-07-221-1/+8
|
* ctxt_auto_deref_lval decides whether to autoderef the entire lval, not its base.Roy Frostig2010-07-191-17/+24
|
* Autoderef objects when passing them as implicit (indirect) arg upon ↵Roy Frostig2010-07-191-2/+11
| | | | vtbl-dispatch. Add testcase and XFAIL it on LLVM. Closes #112.
* Change the body-fetching GEP in mark_ty to be dynamic. Our box may be ↵Roy Frostig2010-07-191-1/+1
| | | | parametric.
* Stub an interface to the (as-yet-nonexistent) structural comparison glue in ↵Patrick Walton2010-07-181-24/+97
| | | | trans
* Fix IL translation of pattern-alt to allow a value of mutable/constrained ↵Roy Frostig2010-07-161-3/+3
| | | | type to be the subject of a pattern-alt.
* More renaming of GEP offsets in abi/trans. This should settle it. Also, ↵Roy Frostig2010-07-161-21/+27
| | | | fix some overenthusiastic renaming in previous commit.
* Clean up nomenclature in the Abi constants brigade. Purge magic GEP numbers ↵Roy Frostig2010-07-161-49/+81
| | | | in trans. Adjust obj_closure_rty (and its consumers) to be more explicit about the fact that it is a box.
* Support nested for-each loops. Closes #79.Jason Orendorff2010-07-161-21/+33
|
* Define (in Abi) and use named constants for indices of elements of an obj IL ↵Roy Frostig2010-07-151-3/+3
| | | | referent type.
* Make mutability no longer a type constructorPatrick Walton2010-07-151-6/+7
|
* Correct the way we get typarams when dropping an obj type in drop_ty by ↵Roy Frostig2010-07-151-5/+3
| | | | using get_tydesc_params.
* Modify copy_fn_args not to bother with dynamic GEP to get the indirect args ↵Roy Frostig2010-07-151-4/+3
| | | | cell in a calltup. It's now possible to obtain it with just regular GEP.
* Explicitly provide an optional closure/obj pointer to trans_call_glue so ↵Roy Frostig2010-07-151-38/+81
| | | | that it can push one in the right position when calling glue, instead of always pushing a null. As far as I can tell this only affects calls to obj drop glue, since only that makes use of an object binding passed as closure/obj, so pass the binding there as needed.
* Adjust trans_call_glue to push/pop an extra null word implicit arg, in ↵Roy Frostig2010-07-151-0/+2
| | | | position of closure/obj pointer.
* Minimize pointless logging during walk.Graydon Hoare2010-07-141-9/+7
|
* Desugar the head stmts all the way out of STMT_for, rather than stashing ↵Graydon Hoare2010-07-131-2/+1
| | | | them in the node. That's only necessary for STMT_while.
* Add more tracing code to trans for refcount events.Graydon Hoare2010-07-121-7/+26
|
* Build fix from jmuizelaar (on ocamlc versions that notice this bug; drat, ↵Graydon Hoare2010-07-081-1/+1
| | | | not all do)
* Use typed slot rather than AST slot in trans_for_loop.Graydon Hoare2010-07-051-3/+4
|
* Fix rec-extend.rs test.Graydon Hoare2010-07-041-7/+11
|
* Comment-out glue interrupt-flag test; unclear why it breaks, but glue ↵Graydon Hoare2010-07-041-1/+4
| | | | call-tree is always finite anyways so not a big deal for now.
* Spill immediates with proper IL type (was making valgrind dislike ↵Graydon Hoare2010-07-041-1/+1
| | | | obj-with-vec.rs).
* Fix bug crashing valgrind in obj-recursion.rs.Graydon Hoare2010-07-041-1/+1
|
* Improve logging in drop paths.Graydon Hoare2010-07-041-10/+20
|
* More corrections to drop/free path to accommodate TY_box.Graydon Hoare2010-07-031-37/+41
|
* Deja vu all over again. Something makes me think this is a usability hazard.Graydon Hoare2010-07-021-3/+6
|
* You may get the feeling that all these changes look eerily the same. You ↵Graydon Hoare2010-07-021-2/+3
| | | | would be correct.
* Simplify type before deciding how to free it. Hint: calling free() on a ↵Graydon Hoare2010-07-021-1/+1
| | | | running task is usually a bad idea.
* Fix bug in clone logic; was ignoring the mutability-strip step in later rule.Graydon Hoare2010-07-021-21/+22
|
* Teach severing logic to handle obj and fn types.Graydon Hoare2010-07-021-15/+28
|
* Trace mark and sever steps in GC.Graydon Hoare2010-07-021-2/+6
|
* Walk (mut/constr stripped) TY_iso edges when boxed. Fault on non-boxed ↵Graydon Hoare2010-07-021-5/+6
| | | | edges; they should not have made it past resolve.
* Correct drop-ty path; many tests now run.Graydon Hoare2010-07-021-11/+5
|
* Improve drop-path logging and simplify box_rc_cell.Graydon Hoare2010-07-021-5/+14
|
* Wrap long lines.Graydon Hoare2010-07-011-2/+6
|
* Yeah. Not even a semantic mix-up: just a damn typo.Graydon Hoare2010-07-011-1/+1
|
* Teach clone_ty to tolerate cloning a mutable (though not state) type.Graydon Hoare2010-07-011-1/+1
|
* Retire defunct copy_ty in favour of trans_copy_ty, which invokes the same ↵Graydon Hoare2010-07-011-71/+60
| | | | logic at the bottom.
* Remove too-optimistic (and not harmless) over application of auto-deref in ↵Graydon Hoare2010-07-011-8/+4
| | | | trans.
* Split out and improve trans_init_box.Graydon Hoare2010-07-011-8/+18
|
* Add logging to trans_init_slot_from_atom.Graydon Hoare2010-07-011-3/+15
|
* Check the simplified type used in check-expr in trans.Graydon Hoare2010-07-011-1/+1
|
* Convey auto-deref judgments made in typechecker to trans layer; control the ↵Graydon Hoare2010-07-011-3/+14
| | | | decision in one place.
* Some work on teaching trans to differentiate between auto-deref and ↵Graydon Hoare2010-07-011-33/+73
| | | | explicit-deref contexts.
* Add more logging to copy/init paths.Graydon Hoare2010-07-011-12/+27
|
* Remove redundant indirection-function.Graydon Hoare2010-07-011-4/+2
|
* Implement STMT_init_box in trans, clean up some of the semant table-accessors.Graydon Hoare2010-07-011-7/+11
|
* Simplify types before analyzing call structure; 2 more tests compile.Graydon Hoare2010-07-011-2/+2
|
generated by cgit v1.2.3 (git 2.25.1) at 2026-03-12 09:14:03 +0000