| Age | Commit message (Collapse) | Author | Lines |
|---|
|
`expand_include_str()` in libsyntax seems to have corrupted the CodeMap by always setting the BytePos of any included files to zero. It now uses `CodeMap::new_filemap()` which should set everything properly. This should fix issue #11322 but I don't want to close it before I have confirmation from the reporters that the problem is indeed fixed.
|
|
Turned out to be a 2-line fix, but the compiler fallout was huge.
|
|
|
|
Various documentation changes, change the 'borrowed pointer' terminology to 'reference', fix a problem with 'make dist' on windows.
|
|
A typed arena is a type of arena that can only allocate objects of one
type. It is 3x faster than the existing arena and 13x faster than malloc
on Mac.
r? @brson
|
|
Turns out the pass in resolve was a little too eager to travel back up the
hierarchy chain when looking for trait candidates.
Closes #10465
|
|
|
|
A typed arena is a type of arena that can only allocate objects of one
type. It is 3x faster than the existing arena and 13x faster than malloc
on Mac.
|
|
This will allow capturing of common things like logging messages, stdout prints
(using stdio println), and failure messages (printed to stderr). Any new prints
added to libstd should be funneled through these task handles to allow capture
as well.
Additionally, this commit redirects logging back through a `Logger` trait so the
log level can be usefully consumed by an arbitrary logger.
This commit also introduces methods to set the task-local stdout handles:
* std::io::stdio::set_stdout
* std::io::stdio::set_stderr
* std::io::logging::set_logger
These methods all return the previous logger just in case it needs to be used
for inspection.
I plan on using this infrastructure for extra::test soon, but we don't quite
have the primitives that I'd like to use for it, so it doesn't migrate
extra::test at this time.
Closes #6369
|
|
This ends up causing weird errors like those seen in #11346
Closes #11346
|
|
Updates as mentioned in #11135
|
|
This will allow capturing of common things like logging messages, stdout prints
(using stdio println), and failure messages (printed to stderr). Any new prints
added to libstd should be funneled through these task handles to allow capture
as well.
Additionally, this commit redirects logging back through a `Logger` trait so the
log level can be usefully consumed by an arbitrary logger.
This commit also introduces methods to set the task-local stdout handles:
* std::io::stdio::set_stdout
* std::io::stdio::set_stderr
* std::io::logging::set_logger
These methods all return the previous logger just in case it needs to be used
for inspection.
I plan on using this infrastructure for extra::test soon, but we don't quite
have the primitives that I'd like to use for it, so it doesn't migrate
extra::test at this time.
Closes #6369
|
|
|
|
This ends up causing weird errors like those seen in #11346
Closes #11346
|
|
|
|
There was a scheduling race where a child may not increment the global task
count before the parent exits, and the parent would then think that there are no
more tasks left.
Closes #11039
|
|
There was a scheduling race where a child may not increment the global task
count before the parent exits, and the parent would then think that there are no
more tasks left.
|
|
|
|
pcwalton says this is right, and it looks right to me too.
Closes #4731
|
|
This reverts commit f1b5f59287106fc511d29c425255bd343608065c.
Using a private function of a library is a bad idea: several people (on
Linux) were meeting with linking errors because of it (different/older
versions of glibc).
|
|
If the main closure failed, then the `exit_code` variable would still be `None`,
and the `unwrap()` was failing (triggering a process abort). This changes the
`unwrap()` to an `unwrap_or()` in order to prevent process abort and detect when
the native task failed.
|
|
This removes the feature where newtype structs can be dereferenced like pointers, and likewise where certain enums can be dereferenced (which I imagine nobody realized still existed). This ad-hoc behavior is to be replaced by a more general overloadable dereference trait in the future.
I've been nursing this patch for two months and think it's about rebased up to master.
@nikomatsakis this makes a bunch of your type checking code noticeably uglier.
|
|
|
|
|
|
If there is a lot of data in thread-local storage some implementations
of pthreads (e.g. glibc) fail if you don't request a stack large enough
-- by adjusting for the minimum size we guarantee that our stacks are
always large enough. Issue #6233.
|
|
I'm not sure if this was even intentional at this point.
|
|
r? @nikomatsakis
for the borrow checker changes. Write guards are now eliminated.
|
|
If there is a lot of data in thread-local storage some implementations
of pthreads (e.g. glibc) fail if you don't request a stack large enough
-- by adjusting for the minimum size we guarantee that our stacks are
always large enough. Issue #6233.
|
|
|
|
|
|
|
|
it no longer has boxes in it
|
|
|
|
This ensures that the `crate_type` attribute always contains a value,
and does not contain an invalid value.
Fixes #11256.
|
|
Had to change some stuff in typeck to bootstrap (getting methods in fmt off of Either), but other than that not so painful.
Closes #9157
|
|
|
|
|
|
Closes #11259
|
|
Closes #11085
|
|
|
|
This patch for #9543 throws an `obsolete syntax` error for `extern mod foo (name="bar")` .
I was wondering if [this](https://github.com/fhahn/rust/compare/mozilla:master...fhahn:issue9543-remove-extern-mod-foo?expand=1#diff-da9d34ca1d0f6beee2838cf02e07345cR4444) is the correct place to do this?
I think the wording of the error message could probably be improved as well.
If this approach is OK, I'm going to run the whole test suite tomorrow and update the old syntax to the new one.
|
|
This pull request fixes #11083. The problem was that recursive type definitions were not properly handled for enum types, leading to problems with LLVM's metadata "uniquing". This bug has already been fixed for struct types some time ago (#9658) but I seem to have forgotten about enums back then. I added the offending code from issue #11083 as a test case.
|
|
|
|
Close #11224
|
|
Commit messages are a little more descriptive.
|
|
|
|
This ensures that the `crate_type` attribute always contains a value,
and does not contain an invalid value.
|
|
|
|
Closes #11259
|
|
Closes #11234
|
