Eduard-Mihai Burtescu
8c4d5af52b
rustc: remove Ty::layout and move everything to layout_of.
2017-11-19 02:14:29 +02:00
Eduard-Mihai Burtescu
8864668d53
rustc: re-complicate the TyLayout API and use better names.
2017-11-19 02:14:29 +02:00
Eduard-Mihai Burtescu
aa811d728a
rustc: remove source field path from Layout::StructWrappedNullablePointer.
2017-11-19 02:14:29 +02:00
Eduard-Mihai Burtescu
bc8e1f7efa
rustc: use an offset instead of a field path in Layout::StructWrappedNullablePointer.
2017-11-19 02:14:29 +02:00
Eduard-Mihai Burtescu
50a3fd0097
rustc: remove useless 0 prefix from Layout::StructWrappedNullablePointer's discrfield.
2017-11-19 02:14:29 +02:00
Eduard-Mihai Burtescu
44eef7c9ac
rustc: do not inject discriminant fields into Layout::General's variants.
2017-11-19 02:14:28 +02:00
Eduard-Mihai Burtescu
8afa3a01e6
rustc_trans: always insert alignment padding, even before the first field.
2017-11-19 02:14:28 +02:00
Eduard-Mihai Burtescu
0a1fcc32a6
rustc_trans: use *[T; 0] for slice data pointers instead of *T.
2017-11-19 02:14:28 +02:00
Eduard-Mihai Burtescu
84b5a3d84d
rustc_trans: remove the in_memory_type_of distinction.
2017-11-19 02:14:28 +02:00
Eduard-Mihai Burtescu
5b1fdaeb80
rustc_trans: use more of the trans::mir and ty::layout APIs throughout.
2017-11-19 02:14:28 +02:00
Eduard-Mihai Burtescu
260c41b4b8
rustc_trans: do not introspect LLVM aggregate field types.
2017-11-19 02:14:28 +02:00
Eduard-Mihai Burtescu
b8671bef97
rustc_trans: remove obsolete Type methods.
2017-11-19 02:14:28 +02:00
Eduard-Mihai Burtescu
386d59dc89
rustc_trans: use a predictable layout for constant ADTs.
2017-11-19 02:14:28 +02:00
Eduard-Mihai Burtescu
f44b099187
rustc_trans: avoid working with sizes/offsets and alignments as integers.
2017-11-19 02:14:24 +02:00
Garrett Berg
44da4a0656
Add doc for Read
ing from &str
and some related cleanup
2017-11-18 16:45:04 -07:00
Eduardo Pinho
0855ea1832
impl From<T> for RwLock<T>
2017-11-18 21:05:06 +00:00
bors
6160040d85
Auto merge of #46039 - oli-obk:test_suggestions, r=petrochenkov
...
Remove left over dead code from suggestion diagnostic refactoring
More cleanups after #41876 and #45741
2017-11-18 20:52:19 +00:00
Eduard-Mihai Burtescu
9deea47c96
rustc_mir: always downcast enums, even if univariant.
2017-11-18 20:52:38 +02:00
Niko Matsakis
83f5a9605f
fix compile-fail tests
2017-11-18 13:09:19 -05:00
bors
b1409af73d
Auto merge of #46032 - KiChjang:ignore-borrowck-statics, r=nikomatsakis
...
Ignore borrowck for static lvalues and allow assignment to static muts
Fixes #45129 .
Fixes #45641 .
2017-11-18 17:30:26 +00:00
Eduardo Pinho
1bfc6c1296
impl From<T> for Mutex<T>
2017-11-18 16:52:46 +00:00
Niko Matsakis
22c0cbfa86
remove generator_interiors
map
2017-11-18 11:26:46 -05:00
Niko Matsakis
413f07438e
remove generator_sigs
from TypeckTables
2017-11-18 11:16:25 -05:00
bors
1f491e07c9
Auto merge of #46009 - kennytm:fix-38878-again, r=alexcrichton
...
Fix #38878 again — restart linker when seeing SIGBUS in additional to SIGSEGV.
In https://github.com/rust-lang/rust/pull/45985#issuecomment-344586645 we see a linker crashed due to Bus Error (signal 10) on macOS. The error was not caught by #40422 since the PR only handles Segmentation Fault (signal 11). The crash log indicates the problem is the same as #38878 , so we just amend #40422 to include SIGBUS as well.
(Additionally, modified how the crash logs are printed so that irrelevant logs are truly filtered out.)
2017-11-18 15:00:13 +00:00
Guillaume Gomez
6047a03659
Add tooltip for important traits display
2017-11-18 14:37:10 +01:00
Niko Matsakis
d8969815cf
comment wf more
2017-11-18 07:49:52 -05:00
Niko Matsakis
27bedfa36b
give better error messages when a cycle arises
2017-11-18 07:49:52 -05:00
Niko Matsakis
5e0e8ae291
rename issue-21410.rs
to ui/unboxed-closure-no-cyclic-sig.rs
2017-11-18 07:47:39 -05:00
Niko Matsakis
1e9de11e62
add a compile-fail test for cyclic generators being forbidden
2017-11-18 07:47:39 -05:00
Niko Matsakis
5e04c66885
remove the generator_sigs
map, query, and plumbing
2017-11-18 07:47:38 -05:00
Niko Matsakis
7010d8cf51
add Yield and Return type into generator
2017-11-18 07:47:38 -05:00
Niko Matsakis
7c609eb090
TyClosure
no longer needs to use freshen_closure_like
...
All the data is now part of its type.
2017-11-18 07:47:38 -05:00
Niko Matsakis
e6fca1d565
remove the closure_tys
map from TypeckTables
...
The information we need is now part of the closure type.
2017-11-18 07:47:38 -05:00
Niko Matsakis
d0bda669ea
move the signature into the closure type
2017-11-18 07:47:37 -05:00
Niko Matsakis
54f4f396d9
convert the closure_kinds
map to just store the origin information
...
The closure kinds themselves are now completely found in the `ClosureSubsts`.
2017-11-18 07:47:37 -05:00
Niko Matsakis
eb26e30b91
kill the closure_kind
query
2017-11-18 07:47:37 -05:00
Niko Matsakis
2dff9a49e5
stop using the closure_kinds
query / table for anything
...
Closure Kind is now extracted from the closure substs exclusively.
2017-11-18 07:47:37 -05:00
Niko Matsakis
716f75b1b8
traits: prefer ClosureSubsts
to InferCtxt::closure_kind
2017-11-18 07:47:36 -05:00
Niko Matsakis
decf3d33d0
make ty::Predicate
carry a ClosureSubsts
2017-11-18 07:47:36 -05:00
Niko Matsakis
0ac8542abc
make mk_closure
take a ClosureSubsts
2017-11-18 07:47:36 -05:00
Niko Matsakis
3349e7bb45
thread the closure-kind through in the closure substs
...
Similar to how freshen handled things, but "always happening"; we can
thus remove the corresponding code from `freshen`.
2017-11-18 07:47:36 -05:00
Niko Matsakis
1d96819dc0
upvar.rs: rustfmt
2017-11-18 07:47:36 -05:00
Niko Matsakis
eab5b263e8
flatten the arguments to analyze_closure
...
They were oddly tupled.
2017-11-18 07:47:35 -05:00
Scott McMurray
cef45b3baf
Undo the Sized specialization from Iterator::nth
2017-11-18 03:45:51 -08:00
bors
18250b0349
Auto merge of #46073 - GuillaumeGomez:rollup, r=GuillaumeGomez
...
Rollup of 4 pull requests
- Successful merges: #45767 , #46044 , #46066 , #46071
- Failed merges:
2017-11-18 11:38:06 +00:00
Guillaume Gomez
5f1c37a0fa
Rollup merge of #46071 - LooMaclin:fix-46001, r=estebank
...
Remove return_ty from Mir
https://github.com/rust-lang/rust/issues/46001
2017-11-18 12:15:24 +01:00
Guillaume Gomez
1389b6203e
Rollup merge of #46066 - GuillaumeGomez:primitive-search, r=QuietMisdreavus
...
Fix primitive types not showing up
Fixes #46017 .
r? @QuietMisdreavus
2017-11-18 12:15:23 +01:00
Guillaume Gomez
c36228f696
Rollup merge of #46044 - collinanderson:py3, r=kennytm
...
fix some python3 incompatibilities
2017-11-18 12:15:22 +01:00
Guillaume Gomez
cdadb6c725
Rollup merge of #45767 - QuietMisdreavus:who-docs-the-doc-tool, r=frewsxcv
...
rustdoc book: talk about #![doc(test(no_crate_inject))] and #![doc(test(attr(...)))]
While investigating https://github.com/rust-lang/rust/issues/45750 i noticed that `#![doc(test(attr(...)))]` wasn't documented at all. Since this is useful for making your examples follow the same coding guidelines as your code, i wanted to add it to the Rustdoc Book. I also added `#![doc(test(no_crate_inject))]` since it's used in the same place and might be useful for macro-heavy crates. I added mentions for these to "The `doc` attribute" as well as "Documentation tests" since it's useful information in both places.
Technically the step reordering in the second commit is gated on https://github.com/rust-lang/rust/pull/45764 , since before that lands attributes from the doctest come before the ones from `#![doc(test(attr(...)))]`.
2017-11-18 12:15:21 +01:00
bors
130eaaeaba
Auto merge of #46000 - kennytm:fix-45640-extern-type-ice-in-rustdoc, r=QuietMisdreavus
...
Support `extern type` in rustdoc.
Fixes #45640 .
(cc #43467 )
2017-11-18 09:18:13 +00:00