mikros/rust
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
64,269 Commits 2 Branches 0 Tags
Commit Graph

64269 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Ariel Ben-Yehuda
d3476f4b76 cache ADT dtorck results 
This avoids visiting the fields of all structs multiple times, improving
item-bodies checking time by 10% (!).
 2017-04-23 23:09:21 +03:00
Ariel Ben-Yehuda
eadb049799 traits::select: quickly filter out predicates from other traits 
this improves most pre-trans passes's performance by ~1%.
 2017-04-23 23:05:25 +03:00
Tim Neumann
8994277657 FIN: windows-gnu: statically link gcc_s, pthread with llvm 2017-04-23 22:00:03 +02:00
Tim Neumann
f3dda17469 FIN: Compile LLVM with -fno-omit-frame-pointer on 32bit MinGW builds 
to work around an apparently bad optimization.
 2017-04-23 22:00:03 +02:00
Tim Neumann
67560f6fae FIN: switch appveryor to mingw with posix threads 2017-04-23 22:00:03 +02:00
Tim Neumann
6723e69a54 FIN: ignore failing test on emscripten, see #41299 2017-04-23 22:00:02 +02:00
Tim Neumann
5bd4e27db2 FIN: update appveyor mingw description 2017-04-23 22:00:02 +02:00
Tim Neumann
5ae4a58c33 FIN: build comiler-rt wihout Thumb on arm 2017-04-23 22:00:02 +02:00
Tim Neumann
a146431e4c FIN: disable backtrace printing for panic-runtime/abort* on ARM 2017-04-23 22:00:02 +02:00
Tim Neumann
a0ce63be2b FIN: update emscripten builder 2017-04-23 22:00:01 +02:00
Tim Neumann
f622542b64 FIN/CRT: adopt all applicable rust compiler-rt patches 2017-04-23 22:00:01 +02:00
Tim Neumann
aaa307ce47 FIN/LLVM: new rust patches required for 4.0 2017-04-23 22:00:01 +02:00
Tim Neumann
4fcfab16d3 FIN/LLVM: adopt all applicable rust llvm patches 2017-04-23 22:00:01 +02:00
Tim Neumann
167b7eb8b5 FIN: update to upstream (llvm 4.0.1 + fastcomp 1.37.10) & compiler-rt 4.0 2017-04-23 22:00:00 +02:00
Ariel Ben-Yehuda
5412587910 clean-up adt_sized_constraint now that it uses on-demand 2017-04-23 14:37:18 +03:00
bors
23de823e93 Auto merge of #41408 - eddyb:poly-const-eval, r=arielb1 
rustc: generalize monomorphic_const_eval to polymorphic constants.

With the addition of `Substs` to the query key, we can now evaluate *and cache* polymorphic constants.

Fixes #23898 by replacing the crippled explicit-discriminant-only local-crate-only `lookup_variant_by_id` with `ConstVal::Variant` which can describe variants irrespective of their discriminant.

Fixes #41394 by fixing #23898 (for the original testcase) and by not looping past the first discriminant.
 2017-04-23 08:35:00 +00:00
Eduard-Mihai Burtescu
8054377f8f rustc_const_eval: support all unit enum variants. 2017-04-23 11:11:57 +03:00
Eduard-Mihai Burtescu
0ff828baa0 rustc_const_eval: CallOn isn't needed, typeck/const-qualif handle those cases. 2017-04-23 11:11:56 +03:00
Eduard-Mihai Burtescu
e22873d912 rustc: make the const-eval cache polymorphic. 2017-04-23 11:11:54 +03:00
bors
a94124488a Auto merge of #41437 - cuviper:remove-unstable-deprecated, r=alexcrichton 
Remove items that are unstable and deprecated

This removes unstable items that have been deprecated for more than one cycle.

- Since 1.16.0, `#![feature(enumset)]`
    - All of `mod collections::enum_set`
- Since 1.15.0, `#![feature(borrow_state)]`
    - `cell::BorrowState`
    - `RefCell::borrow_state()`
- Since 1.15.0, `#![feature(is_unique)]`
    - `Rc::is_unique()` (made private like `Arc::is_unique()`)
- Since 1.15.0, `#![feature(rc_would_unwrap)]`
    - `Rc::would_wrap()`
- Since 1.13.0, `#![feature(binary_heap_extras)]`
    - `BinaryHeap::push_pop()`
    - `BinaryHeap::replace()`
- Since 1.12.0, `#![feature(as_unsafe_cell)]`
    - `Cell::as_unsafe_cell()`
    - `RefCell::as_unsafe_cell()`
- Since 1.12.0, `#![feature(map_entry_recover_keys)]`
    - `btree_map::OccupiedEntry::remove_pair()`
    - `hash_map::OccupiedEntry::remove_pair()`
- Since 1.11.0, `#![feature(float_extras)]`
    - `Float::nan()`
    - `Float::infinity()`
    - `Float::neg_infinity()`
    - `Float::neg_zero()`
    - `Float::zero()`
    - `Float::one()`
    - `Float::integer_decode()`
    - `f32::integer_decode()`
    - `f32::ldexp()`
    - `f32::frexp()`
    - `f32::next_after()`
    - `f64::integer_decode()`
    - `f64::ldexp()`
    - `f64::frexp()`
    - `f64::next_after()`
- Since 1.11.0, `#![feature(zero_one)]`
    - `num::Zero`
    - `num::One`
 2017-04-23 02:13:55 +00:00
Corey Farwell
13d2534fd3 Remove unused import. 2017-04-22 22:13:22 -04:00
bors
252d3da8a6 Auto merge of #41469 - arielb1:rustc-spring-cleaning, r=eddyb 
Performance audit, Spring 2017

Fix up some quite important performance "surprises" I've found running callgrind on rustc.
 2017-04-22 20:07:01 +00:00
Ariel Ben-Yehuda
a660ad84b3 bail out of selection when there are multiple surviving candidates 
In some cases (e.g. <[int-var] as Add<[int-var]>>), selection can turn up
a large number of candidates. Bailing out early avoids O(n^2) performance.

This improves item-type checking time by quite a bit, resulting in ~2% of total
time-to-typeck.
 2017-04-22 21:02:50 +03:00
Ariel Ben-Yehuda
a0f145ba8c add a cache to impl_polarity 
this is another one of these things that looks *much* worse on valgrind.
 2017-04-22 21:02:41 +03:00
Ariel Ben-Yehuda
81af6fb67c allocate less strings in symbol_names 
this improves trans performance by *another* 10%.
 2017-04-22 21:02:02 +03:00
Ariel Ben-Yehuda
266d36f3b1 weak_lang_items: check for lang attribute before calling value_str 
improves trans performance by *another* 10%.
 2017-04-22 21:00:50 +03:00
Ariel Ben-Yehuda
ece6c8434b cache attributes of items from foreign crates 
this avoids parsing item attributes on each call to `item_attrs`, which takes
off 33% (!) of translation time and 50% (!) of trans-item collection time.
 2017-04-22 21:00:50 +03:00
Ariel Ben-Yehuda
acd0e40b86 short-cut SharedCrateContext::layout_of 
That method is *incredibly* hot, so this ends up saving 10% of trans
time.

BTW, we really should be doing dependency tracking there - and possibly be
taking the respective perf hit (got to find a way to make DTMs fast), but
`layout_cache` is a non-dep-tracking map.
 2017-04-22 21:00:50 +03:00
Ariel Ben-Yehuda
e1377a4f47 avoid calling mk_region unnecessarily 
this improves typeck & trans performance by 1%. This looked hotter on
callgrind than it is on a CPU.
 2017-04-22 21:00:50 +03:00
Ariel Ben-Yehuda
3bf00450cb remove cleanup branches to the resume block 
This improves LLVM performance by 10% lost during the shimmir transition.
 2017-04-22 21:00:50 +03:00
bors
cb4065b9ce Auto merge of #41431 - GuillaumeGomez:hoedown-default, r=aturon 
Re-enable hoedown by default

r? @rust-lang/docs

cc @aturon
 2017-04-22 17:37:04 +00:00
bors
9cc77d7690 Auto merge of #41464 - frewsxcv:rollup, r=frewsxcv 
Rollup of 3 pull requests

- Successful merges: #41077, #41355, #41450
- Failed merges:
 2017-04-22 14:17:36 +00:00
Corey Farwell
b37163c81c Rollup merge of #41450 - rillian:rustc-wrapper, r=alexcrichton 
Update cargo for 1.18.0.

Bump our associated cargo to pick up the RUSTC_WRAPPER feature
for use with build caches.
 2017-04-22 09:02:58 -04:00
Corey Farwell
e282817186 Rollup merge of #41355 - nikomatsakis:incr-comp-refactor-trans-2, r=eddyb 
Refactor trans some more to pave way for incremental compilation

Various refactorings paving the way for the newer approach to incremental compilation (And, in particular, to "query-ifying" trans). My partial goal is to remove `SharedCrateContext`; this PR gets about as far as I can easily get before starting to really want the red/green algorithm.

r? @eddyb
cc @michaelwoerister
 2017-04-22 09:02:57 -04:00
Corey Farwell
5e1351b17b Rollup merge of #41077 - petrochenkov:boundparen, r=nikomatsakis 
syntax: Support parentheses around trait bounds

An implementation for https://github.com/rust-lang/rust/issues/39318#issuecomment-290956826

r? @nikomatsakis
 2017-04-22 09:02:56 -04:00
Guillaume Gomez
9c97882055 Fix line display 2017-04-22 14:56:36 +02:00
bors
6d841da4a0 Auto merge of #39999 - bitshifter:struct_align, r=eddyb 
Implementation of repr struct alignment RFC 1358.

The main changes around rustc::ty::Layout::struct:
* Added abi_align field which stores abi alignment before repr align is applied
* align field contains transitive repr alignment
* Added padding vec which stores padding required after fields

The main user of this information is rustc_trans::adt::struct_llfields
which determines the LLVM fields to be used by LLVM, including padding
fields.

A possible future optimisation would be to put the padding Vec in an Option, since it will be unused unless you are using repr align.
 2017-04-22 11:50:40 +00:00
Guillaume Gomez
91fb6bc1eb Fix tests 2017-04-22 13:25:14 +02:00
Guillaume Gomez
d79b511f5c Fix invalid linkage 2017-04-22 13:25:14 +02:00
Guillaume Gomez
295f25b198 Set hoedown to generate error index 2017-04-22 13:25:14 +02:00
Guillaume Gomez
80a2a94d5a Re-enable hoedown by default 2017-04-22 13:25:14 +02:00
bors
ff13b7c918 Auto merge of #41459 - frewsxcv:rollup, r=frewsxcv 
Rollup of 4 pull requests

- Successful merges: #37658, #41405, #41432, #41435
- Failed merges:
 2017-04-22 03:31:53 +00:00
Corey Farwell
df72158d47 Rollup merge of #41435 - estebank:issue-33884, r=nikomatsakis 
Add test for issue 33884

Fix #33884.

r=nikomatsakis
 2017-04-21 23:29:15 -04:00
Corey Farwell
48a9d5f6d3 Rollup merge of #41432 - abonander:issue_41211, r=jseyfried 
Don't panic if an attribute macro fails to resolve at crate root

Adds temporary regression test; this ideally should work as-is (#41430)

Closes #41211

r? @jseyfried
 2017-04-21 23:29:15 -04:00
Corey Farwell
aff40e2de3 Rollup merge of #41405 - GuillaumeGomez:rustdoc-warning-improvement, r=steveklabnik 
Fix line display for hoedown

Fixes #41401.

r? @rust-lang/docs
 2017-04-21 23:29:14 -04:00
Corey Farwell
5da19f2b69 Rollup merge of #37658 - GuillaumeGomez:ref_suggestion, r=nikomatsakis,eddyb 
Ref suggestion
 2017-04-21 23:29:13 -04:00
Jessica Hamilton
e1afddc29c Haiku: fix initial platform support 2017-04-22 13:47:36 +12:00
Niko Matsakis
6d86f81eb1 move the uses of the trans caches into rustc::traits 
This makes these routines more readily available for other bits of
code. It also will help when refactoring.
 2017-04-21 21:02:14 -04:00
Cameron Hart
946f8e6a59 Use primitive align for tagged enum fill. 
Hopefully will fix assert on ARM where vector types are being used as
the fill type for enums containing repr aligned types greater than the
largest possible native type, thus don't match the Layout's alignment
and triggers an assert.
 2017-04-22 09:38:23 +10:00
Niko Matsakis
85527456e8 use Symbol/InternedString in the cache 2017-04-21 17:26:53 -04:00