aboutsummaryrefslogtreecommitdiff
path: root/src/rt/rust_task.cpp
Commit message (Collapse)AuthorAgeFilesLines
* Fix OS X build.Rafael Ávila de Espíndola2011-05-051-9/+9
|
* Change the setup so that rust_activate_glue returns to rust_exit_task_glueRafael Ávila de Espíndola2011-05-051-38/+19
| | | | | | | | | | | and rust_exit_task_glue calls the rust main. This is simpler since we only need to setup one frame. It also matches what ld.so does, so gdb is happy and stops a backtrace at rust_exit_task_glue instead of continuing past whatever function happened to be before rust_exit_task_glue is the object file. This is the rt part and should be merged after the rust0 part.
* Try to fail nicely, again.Graydon Hoare2011-05-041-0/+2
|
* Remove unused variable.Rafael Ávila de Espíndola2011-05-031-7/+1
|
* Split start into rustc and rustboot versions. This introduces a bit ofRafael Ávila de Espíndola2011-05-031-27/+114
| | | | | | | | | | | duplication, but we will hopefully drop the rustboot one soon. This is also a preparation for changing the rustc one to have the activate glue return to the exit glue which will then call the main function. This (returning to the function that calls main) matches what happens when loader stats a program or a new thread. It lets gdb produce good backtraces and should help with EH too.
* Overhaul logging system in runtimeMarijn Haverbeke2011-04-191-74/+53
| | | | | | | | | | | | | | See https://github.com/graydon/rust/wiki/Logging-vision The runtime logging categories are now treated in the same way as modules in compiled code. Each domain now has a log_lvl that can be used to restrict the logging from that domain (will be used to allow logging to be restricted to a single domain). Features dropped (can be brought back to life if there is interest): - Logger indentation - Multiple categories per log statement - I possibly broke some of the color code -- it confuses me
* rt: Remove half-baked, incomplete, commented-out print_stack_trace() stubPatrick Walton2011-04-141-17/+0
|
* rustc: Remove generalize_ty. Instead, maintain an explicit type parameter ↵Patrick Walton2011-04-141-0/+17
| | | | substitution list.
* add -O2 when compiling rt, fix warnings triggered by optimizerMarijn Haverbeke2011-04-081-2/+2
|
* Move to macro-based logging checks in the C++ codeMarijn Haverbeke2011-04-071-30/+30
| | | | | No functions should be called for log statements that turn out to be inactive.
* Um, that'd be, align the word *before* retpc. Addresses point to the low ↵Graydon Hoare2011-03-301-7/+7
| | | | part of a word. Un-XFAIL nbody on Darwin.
* Ensure task stacks start out with a 16-byte aligned entry frame. Should make ↵Graydon Hoare2011-03-301-22/+48
| | | | OSX behave a bit better.
* Add ABI tagging to crates, adjust rustc output and runtime stack-frame setup ↵Graydon Hoare2011-02-221-18/+34
| | | | so access to argv works.
* Fix typo in rust_task::start. Should be copying 1 fewer initial arg.Graydon Hoare2011-02-211-2/+2
|
* Further corrections to the logging layer in runtime.Graydon Hoare2011-01-141-1/+1
|
* Don't attempt to wake tasks that aren't blocked.Brian Anderson2011-01-031-1/+1
| | | | | | It's possible for a supervised task to kill and wake its supervising task then immediately try to wake it again if the supervising task has joined the supervised. This is the easiest way to prevent that.
* Update frame logic to be compatible with SysV x86 ABI. Improves diagnostics.Graydon Hoare2010-11-161-12/+16
|
* Try to print backtraces on failurePatrick Walton2010-10-111-0/+17
|
* Patchwork of attempted fixes to effect system and gc system; eventually give ↵Graydon Hoare2010-09-291-0/+10
| | | | up and disable it entirely in the runtime. Will need extensive reworking.
* Cleanup, refactoring, and some runtime tests.Michael Bebenita2010-09-101-24/+14
|
* Cache task handles.Michael Bebenita2010-09-081-2/+6
|
* Lots of design changes around proxies and message passing. Made it so that ↵Michael Bebenita2010-09-071-4/+9
| | | | domains can only talk to other domains via handles, and with the help of the rust_kernel.
* Change signature of array_list::pop().Michael Bebenita2010-09-071-1/+2
|
* Disable task-growing logic for now until #151 is fixedPatrick Walton2010-08-201-1/+10
|
* Added labels to blocking conditions.Michael Bebenita2010-08-171-1/+4
|
* Added support for task sleeping in the scheduler.Michael Bebenita2010-08-111-3/+9
|
* Merge commit 'jyasskin/work'Graydon Hoare2010-08-101-19/+24
|\ | | | | | | | | | | Conflicts: src/rt/rust_dom.cpp src/rt/rust_upcall.cpp
| * Add names to tasks and domains. These can either be an explicit literal stringJeffrey Yasskin2010-08-091-19/+24
| | | | | | | | | | after the "spawn" keyword, or implicitly the call expression used to start the spawn.
* | Synthesize a flush_chan upcall right before a channel's ref_count drops to ↵Michael Bebenita2010-08-091-0/+5
| | | | | | | | zero. This should only happen in the Rust code and not in the drop glue, or on the unwind path. This change allows the task owning the channel to block on a flush and delete its own channel. This change also cleans up some code around rust_port and rust_chan.
* | Fixed deadlock in the scheduler caused by condition variables.Michael Bebenita2010-08-091-3/+0
| |
* | Some pretty printing in the runtime.Michael Bebenita2010-08-091-0/+5
|/
* Log join-wait count under rust_log::TASK bits, not ::ALL.Graydon Hoare2010-07-291-1/+1
|
* Move ports out into their own file, add data_message and make communication ↵Michael Bebenita2010-07-281-14/+0
| | | | system use it (and proxies) instead of existing token scheme.
* Move notification-messages out into their own file and unify into ↵Michael Bebenita2010-07-281-0/+18
| | | | notify_message, make them use proxies, cache task proxies in dom.
* Rename rust_proxy_delegate to maybe_proxy, flesh out logic in it. Add ↵Michael Bebenita2010-07-281-1/+1
| | | | strong-ref distinction on rust_proxy.
* Null rust_task::cond on wakeup, add asserts and logging to block/wakeup.Michael Bebenita2010-07-281-5/+11
|
* Fix message formatting in runtime library.Graydon Hoare2010-07-201-6/+6
|
* Added a message passing system based on lock free queues for inter-thread ↵Michael Bebenita2010-07-191-16/+41
| | | | communication. Channels now buffer on the sending side, and no longer require blocking when sending. Lots of other refactoring and bug fixes.
* Adjust call args laid out by fake frame created in rust_task::start to ↵Roy Frostig2010-07-151-0/+3
| | | | reflect new position of closure-or-obj pointer.
* Correct doubly-linked list management logic; bug exposed during conversation ↵Graydon Hoare2010-07-121-0/+4
| | | | on issue #106.
* Fix for compile error when using gcc 4.5.0Chris Double2010-07-111-1/+1
|
* Change from 'spawner' to 'supervisor' in rust_task, and add an unsupervise call.Graydon Hoare2010-07-051-5/+14
|
* Actually link GC allocations onto the gc_alloc_chain, der.Graydon Hoare2010-07-021-0/+1
|
* Move more of the GC logic into the runtime.Graydon Hoare2010-06-281-0/+78
|
* Add rust_task::gc(size_t args) method that drops us back into the GC glue.Graydon Hoare2010-06-251-0/+8
|
* Populate tree.Graydon Hoare2010-06-231-0/+474