Commit Graph

8029 Commits

Author SHA1 Message Date
Jakub Bukaj
cca84e9e21 Remove ty_bot from the type system
We now instead use a fresh variable for expressions that diverge.
2014-10-28 17:54:16 +01:00
Brian Koropoff
e46af8caec Add regression test for issue #16791 2014-10-27 18:51:43 -07:00
Alex Crichton
a33d7617c5 Test fixes and rebase conflicts from rollup 2014-10-27 15:12:48 -07:00
Alex Crichton
20b9897c58 rollup merge of #18362 : kevinmehall/pprint-struct-pat-shorthand 2014-10-27 15:12:47 -07:00
Alex Crichton
c381752b0e rollup merge of #18356 : jakub-/const-pattern-suptype 2014-10-27 15:12:47 -07:00
Alex Crichton
15547e1d72 rollup merge of #18346 : aochagavia/closure-fields 2014-10-27 15:12:47 -07:00
Alex Crichton
1cc938a69a rollup merge of #18337 : bkoropoff/unboxed-imm-upvar-fixes 2014-10-27 15:12:45 -07:00
Alex Crichton
175d6a7435 rollup merge of #18332 : jbcrail/fix-comments 2014-10-27 15:12:30 -07:00
Alex Crichton
f1118cc084 rollup merge of #18324 : jakub-/eneedstest 2014-10-27 15:12:30 -07:00
Alex Crichton
abfc2f3b8d rollup merge of #18315 : arielb1/constexpr-addr 2014-10-27 15:12:29 -07:00
Alex Crichton
e9537735c1 rollup merge of #18250 : dotdash/fix_aliasing 2014-10-27 12:53:01 -07:00
Alex Crichton
0c756de6c5 rollup merge of #18235 : nikomatsakis/issue-18209 2014-10-27 12:53:01 -07:00
Alex Crichton
b3c676ed86 rollup merge of #18229 : bjz/ttdelim 2014-10-27 12:53:01 -07:00
Adolfo Ochagavía
2ce77b33b0 Add test for issue 18343 2014-10-27 16:28:24 +01:00
Kevin Mehall
e5f709079a Preserve struct field pattern shorthand in the prettyprinter.
Use the `is_shorthand` field introduced by #17813 (ead6c4b) to make the
prettyprinter output the shorthand form. Fixes a few places that set
`is_shorthand: true` when the pattern is not a PatIdent with the same
name as the field.
2014-10-27 00:35:35 -07:00
Jakub Bukaj
f16744ce52 Fix a typecheck regression with constant borrowed pointers in patterns
Change the eqtype relationship to be a suptype relationship instead.

Fixes #18350.
Fixes #18352.
2014-10-26 22:35:26 +01:00
bors
f037452447 auto merge of #18143 : mahkoh/rust/reverse_complement, r=alexcrichton
Lots of unsafe code and lots of branches removed. Also multithreaded.

Rust old: 1.208 seconds
Rust new: 0.761 seconds
C: 0.632 seconds
2014-10-26 16:42:33 +00:00
Brian Koropoff
5662bbad07 Add regression test for #18335 2014-10-25 21:51:14 -07:00
Brian Koropoff
6be48ea92e Add regression test for #18238 and #18336 2014-10-25 21:50:08 -07:00
Joseph Crail
30403204d6 Fix spelling mistakes in comments. 2014-10-25 23:11:17 -04:00
Brendan Zabarauskas
dfb4163f83 Use standard capitalisation for TokenTree variants 2014-10-26 09:53:30 +11:00
Brendan Zabarauskas
ec3f0201e7 Rename TokenTree variants for clarity
This should be clearer, and fits in better with the `TTNonterminal` variant.

Renames:

- `TTTok` -> `TTToken`
- `TTDelim` -> `TTDelimited`
- `TTSeq` -> `TTSequence`
2014-10-26 09:53:29 +11:00
Jakub Bukaj
2c744c7f32 Add test cases for E-needstest issues 2014-10-26 00:18:10 +02:00
Ariel Ben-Yehuda
81b7e62918 Prevent pointer -> int casts in constexprs
These cause issues, as addresses aren't fixed at compile-time.

Fixes #18294
2014-10-25 20:18:32 +03:00
bors
a34b8dec69 auto merge of #18171 : jakub-/rust/match-typeck, r=pcwalton
Rather than doing it top-down, with a known expected type, we will now simply establish the appropriate constraints between the pattern and the expression it destructures.

Closes #8783.
Closes #10200.
2014-10-25 07:07:08 +00:00
bors
172b59abe5 auto merge of #18080 : veddan/rust/assume, r=thestinger
Adds an `assume` intrinsic that gets translated to llvm.assume. It is
used on a boolean expression and allows the optimizer to assume that
the expression is true.

This implements #18051.
2014-10-25 00:32:07 +00:00
Jakub Bukaj
1484f9cd46 Update tests with the new error messages 2014-10-24 19:43:47 +02:00
Jakub Bukaj
4380e96c04 Overhaul typechecking of patterns
Instead of checking patterns in a top-down fashion with a known
expected type on entry, this changes makes typeck establish
appropriate constraints between a pattern and the expression
it destructures, and lets inference compute the final types
or produce good error messages if it's impossible.
2014-10-24 19:43:47 +02:00
Jakub Bukaj
3e9ce5afb7 Do not accept functions in enum patterns past resolve 2014-10-24 19:43:47 +02:00
bors
a10917a6a9 auto merge of #16388 : Zoxc/rust/stmesg, r=alexcrichton
This installs signal handlers to print out stack overflow messages on Linux. It also ensures the main thread has a guard page.

This will catch stack overflows in external code. It's done in preparation of switching to stack probes (#16012).

I've done some simple tests with overflowing the main thread, native threads and green threads (with and without UV) on x86-64.
This might work on ARM, MIPS and x86-32.

I've been unable to run the test suite on this because of #16305.
2014-10-24 17:32:04 +00:00
Julian Orth
b3ed61703c ignore-android 2014-10-24 17:19:02 +02:00
Julian Orth
da2152c9ae Improve shootout-reverse-complement 2014-10-24 17:18:10 +02:00
John Kåre Alsaker
70cef9474a Print stack overflow messages for Windows, Linux and OS X
Fixes #17562
2014-10-24 14:36:29 +02:00
P1start
ead6c4b9d4 Add a lint for not using field pattern shorthands
Closes #17792.
2014-10-24 15:44:18 +13:00
Björn Steinbrink
70fe20a698 Fix codegen breaking aliasing rules for functions with sret results
This reverts commit a0ec902e23 "Avoid
unnecessary temporary on assignments".

Leaving out the temporary for the functions return value can lead to a
situation that conflicts with rust's aliasing rules.

Given this:

````rust
fn func(f: &mut Foo) -> Foo { /* ... */ }

fn bar() {
    let mut foo = Foo { /* ... */ };

    foo = func(&mut foo);
}
````

We effectively get two mutable references to the same variable `foo` at
the same time. One for the parameter `f`, and one for the hidden
out-pointer. So we can't just `trans_into` the destination directly, but
must use `trans` to get a new temporary slot from which the result can
be copied.
2014-10-23 11:43:48 +02:00
bors
8a40854660 auto merge of #17868 : nick29581/rust/valgrind, r=alexcrichton
r? @alexcrichton
2014-10-23 05:27:11 +00:00
Nick Cameron
edc5275563 Move some tests to run-pass-valgrind 2014-10-23 13:52:34 +13:00
bors
96991e9335 auto merge of #18224 : nikomatsakis/rust/issue-17594, r=alexcrichton
Add test for issue #17594

Fixes #17594
2014-10-22 23:57:11 +00:00
bors
7d7e409839 auto merge of #18092 : michaelwoerister/rust/lldb-test-versioning, r=alexcrichton
Now that there are build bots with a stable enough LLDB version on OSX we can finally let the tests run on every PR!
😹
2014-10-22 21:32:16 +00:00
Niko Matsakis
de06faf889 Use local cache when there are unbound type variables and where clauses in scope.
Fixes #18209.
2014-10-22 11:35:53 -04:00
Michael Woerister
47e8cf7697 debuginfo: Gate all LLDB debuginfo tests on a minimum LLDB version being available 2014-10-22 10:47:45 +02:00
Niko Matsakis
7b021859a4 Add test for issue #17594 2014-10-21 23:58:09 -04:00
Niko Matsakis
aeba2ccf30 Adjust orphan rules to consider all input types, not just self type.
Fixes #18222.
2014-10-21 23:52:00 -04:00
Niko Matsakis
b066d09be8 Patch up broken error messages 2014-10-21 17:36:15 -04:00
Niko Matsakis
53ede4403b Coherence tests that seemed to be missing. 2014-10-21 12:32:37 -04:00
Niko Matsakis
450263de4a Tests for method resolution in the face of various ambiguities or non-ambiguities. These
are mostly new tests though I also revamped (and renamed) some of the existing tests.
2014-10-21 12:32:36 -04:00
Niko Matsakis
e09fc03701 Various minor cases where errors are reported in slightly different ways. 2014-10-21 12:32:36 -04:00
Niko Matsakis
7f8ca53669 Test where the old infrastructure failed to detect the (applicable) impl of FnMut for
some reason.
2014-10-21 12:32:36 -04:00
Niko Matsakis
df714cfda7 The new method lookup mechanism typechecks calls against the method type declared in the trait, not in the impl. In some cases that results in tighter rules, and in some cases looser. Correct for that. 2014-10-21 12:32:36 -04:00
Victor Berger
dd55c8003c Stability lint checker now handles nested macros.
Closes #17185.
2014-10-20 23:48:29 +02:00