Граф коммитов

108 Коммитов

Автор SHA1 Сообщение Дата
Samuel Williams 42130a64f0
Replace copy coroutine with pthread implementation. 2021-07-01 11:23:03 +12:00
Nobuyoshi Nakada 9eae8cdefb
Prefer qualified names under Thread 2021-06-29 11:41:10 +09:00
Nobuyoshi Nakada b7d01b0d1b
Refined define_thread_class
Reduce duplications
* ID caluculations of the same name
* checks against the same name
* registration to the root module hash
2021-06-28 16:52:49 +09:00
Nobuyoshi Nakada e4f891ce8d
Adjust styles [ci skip]
* --braces-after-func-def-line
* --dont-cuddle-else
* --procnames-start-lines
* --space-after-for
* --space-after-if
* --space-after-while
2021-06-17 10:13:40 +09:00
Nobuyoshi Nakada b32e6c168a
Convert initial contents before allocating queue buffer 2021-06-15 15:55:52 +09:00
Nobuyoshi Nakada 2a6bfd2246
Properly convert time_t [Bug #17645] 2021-03-14 20:18:50 +09:00
Nobuyoshi Nakada 8a5a91eead
--dont-cuddle-else [ci skip] 2021-03-14 20:18:30 +09:00
Nobuyoshi Nakada 1f0e0dfb22
Thread::Queue.new should accept an Enumerable [Feature #17327]
Enumerable implements #to_a but not #to_array.
2021-02-12 12:21:49 +09:00
Chris Seaton c3b2bb0969
The Queue constructor should take an initial set of objects
Co-authored-by: Nobuyoshi Nakada <nobu@ruby-lang.org>
2021-02-11 19:14:18 +09:00
Samuel Williams 5f69a7f604
Expose scheduler as public interface & bug fixes. (#3945)
* Rename `rb_scheduler` to `rb_fiber_scheduler`.

* Use public interface if available.

* Use `rb_check_funcall` where possible.

* Don't use `unblock` unless the fiber was non-blocking.
2021-02-09 19:39:56 +13:00
Nobuyoshi Nakada 834b404b9a
[DOC] Fixed indent [ci skip] 2021-01-14 21:35:38 +09:00
Samuel Williams 3b5b309b7b Proposed method for dealing with stack locals which have non-local lifetime. 2020-12-05 11:38:56 +13:00
Nobuyoshi Nakada 0d52dce3a3
strip trailing spaces and adjusted indents [ci skip] 2020-11-12 19:27:20 +09:00
Samuel Williams c39984ec5c Tidy up book keeping for `thread->keeping_mutexes`.
When a scheduler is present, it's entirely possible for
`th->keeping_mutexes` to be updated while enumerating the waitq. Therefore
it must be fetched only during the removal operation.
2020-11-08 20:40:52 +13:00
Samuel Williams f73135233b Don't try to resume blocked fiber on dead thread. 2020-11-08 20:40:52 +13:00
Samuel Williams a08ee8330d Rename to `Fiber#set_scheduler`. 2020-11-07 23:39:50 +13:00
Samuel Williams 70f08f1eed Make `Thread#join` non-blocking. 2020-09-21 11:48:44 +12:00
Benoit Daloze 6987c8997e Remove from waiter in Mutex#lock with ensure when calling rb_scheduler_block()
* Previously this could lead to an invalid waiter entry and then trying
  to wake up that waiter would result in various issues in rb_mutex_unlock_th().
2020-09-20 13:35:39 +02:00
Benoit Daloze af1926e859 Fix copy/paste error from 5bb5e706f1 2020-09-18 11:11:47 +02:00
Benoit Daloze 5bb5e706f1 Only interrupt when there is no scheduler in sync_wakeup()
* When there is a scheduler, the Fiber that would be blocked has already
  been rescheduled and there is no point to interrupt something else.
  That blocked Fiber will be rescheduled as the next call to the scheduler
  (e.g., IO, sleep, other blocking sync).
* See discussion on https://github.com/ruby/ruby/commit/d01954632d
2020-09-18 10:39:27 +02:00
Benoit Daloze d9b943b8e5 Cleanup commented code
* Mutex operations no longer disable the Fiber scheduler.
2020-09-17 17:29:43 +02:00
Benoit Daloze d01954632d Add missing goto found;
* To still remove the lock from the Thread's list of acquired locks.
* Also to not wake up other waiters and preserve blocking behavior.
2020-09-17 17:26:52 +02:00
Benoit Daloze 264889ec3d Fix Mutex#unlock with a scheduler and thread contention
* It would hit "[BUG] unexpected THREAD_STOPPED" before.
2020-09-17 15:15:43 +02:00
Benoit Daloze 9472d16061 Call scheduler.block instead of scheduler.kernel_sleep for blocking Queue/SizedQueue operations
* scheduler.unblock was already already called before but with no corresponding scheduler.block
* add test that Queue#pop makes the scheduler wait until it gets an element.
2020-09-17 14:59:38 +02:00
Benoit Daloze 738a089b3a Rename scheduler.{mutex_lock,mutex_unlock} to {block,unblock}
* Move #kernel_sleep next to #block as it is similar
2020-09-17 14:30:40 +02:00
Samuel Williams 8eea66a0ca Add support for Queue & SizedQueue. 2020-09-14 16:44:09 +12:00
Samuel Williams 0f613cc5f1 Add support for ConditionVariable. 2020-09-14 16:44:09 +12:00
Samuel Williams 1a0cfe2839 Improve handling of urgent notification pipe. 2020-09-14 16:44:09 +12:00
Benoit Daloze 178c1b0922 Make Mutex per-Fiber instead of per-Thread
* Enables Mutex to be used as synchronization between multiple Fibers
  of the same Thread.
* With a Fiber scheduler we can yield to another Fiber on contended
  Mutex#lock instead of blocking the entire thread.
* This also makes the behavior of Mutex consistent across CRuby, JRuby and TruffleRuby.
* [Feature #16792]
2020-09-14 16:44:09 +12:00
Koichi Sasada 79df14c04b Introduce Ractor mechanism for parallel execution
This commit introduces Ractor mechanism to run Ruby program in
parallel. See doc/ractor.md for more details about Ractor.
See ticket [Feature #17100] to see the implementation details
and discussions.

[Feature #17100]

This commit does not complete the implementation. You can find
many bugs on using Ractor. Also the specification will be changed
so that this feature is experimental. You will see a warning when
you make the first Ractor with `Ractor.new`.

I hope this feature can help programmers from thread-safety issues.
2020-09-03 21:11:06 +09:00
卜部昌平 f402dc3557 rb_szqueue_push: do not goto into a branch
I'm not necessarily against every goto in general, but jumping into a
branch is definitely a bad idea.  Better refactor.
2020-06-29 11:05:41 +09:00
Samuel Williams 0e3b0fcdba
Thread scheduler for light weight concurrency. 2020-05-14 22:10:55 +12:00
Nobuyoshi Nakada 5d430c1b34
Added more NORETURN declarations 2020-05-11 00:40:14 +09:00
卜部昌平 115fec062c more on NULL versus functions.
Function pointers are not void*.  See also
ce4ea956d2
8427fca49b
2020-02-07 14:24:19 +09:00
Koichi Sasada d8d581bfc4 add assertion for mutex_lock.
After do_mutex_lock(mutex), the mutex should be owned by the current
thread. Adding an assertion for this assumption.
2019-10-28 12:19:18 +09:00
Nobuyoshi Nakada 0c6f36668a
Adjusted spaces [ci skip] 2019-09-27 10:20:56 +09:00
卜部昌平 69683968a9 fix arity of rb_mutex_synchronize_m
This is just a trivial mistake introduced in
6c56dae4b2.
2019-08-29 18:34:09 +09:00
卜部昌平 703783324c rb_ensure now free from ANYARGS
After 5e86b005c0, I now think ANYARGS is
dangerous and should be extinct.  This commit deletes ANYARGS from
rb_ensure, which also revealed many arity / type mismatches.
2019-08-27 15:52:26 +09:00
Neeraj Bhunwal c1d78a7f0e do_mutex_lock: release mutex before checking for interrupts (fixes issue 15360) 2019-06-11 15:44:11 -07:00
Nobuyoshi Nakada b1aecef873
Use UNALIGNED_MEMBER_PTR
* internal.h (UNALIGNED_MEMBER_ACCESS, UNALIGNED_MEMBER_PTR):
  moved from eval_intern.h.

* compile.c iseq.c, vm.c: use UNALIGNED_MEMBER_PTR for `entries`
  in `struct iseq_catch_table`.

* vm_eval.c, vm_insnhelper.c: use UNALIGNED_MEMBER_PTR for `body`
  in `rb_method_definition_t`.
2019-05-31 16:04:16 +09:00
normal 0fd53f519f thread_sync.c (rb_mutex_t): eliminate fork_gen
The true bug fork_gen was hiding was rb_mutex_abandon_locking_mutex
failing to unconditionally clear the waitq of mutexes it was
waiting on.  So we fix rb_mutex_abandon_locking_mutex, instead,
and eliminate rb_mutex_cleanup_keeping_mutexes.

This commit was tested heavily on a single-core Pentium-M which
was my most reliable reproducer of the "crash.rb" script from
[Bug #15383]

[Bug #14578] [Bug #15383]

Note: [Bug #15430] turned out to be an entirely different
problem: RLIMIT_NPROC limit was hit on the CI VMs.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@66489 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-12-22 01:41:18 +00:00
normal fa5601e7b9 thread_sync.c (rb_mutex_cleanup_keeping_mutexes): update fork_gen
... when clearing waitq. Otherwise, we risk redundantly clearing
valid waiters in future calls to `mutex_ptr`.

Note: I am not sure if this fixes [Bug #15430], and even if it
did, fork_gen is a belt-and-suspenders redundancy for [Bug #15383]
which wastes one word for every Mutex object.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@66477 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-12-21 12:32:52 +00:00
normal 2a742d5f9b thread_sync.c (rb_mutex_abandon_keeping_mutexes): remove unnecessary check
rb_mutex_abandon_all functions fine when passed a NULL value,
so let the compiler deal with the complexity of the branch
instead of the person reading the code.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@66476 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-12-21 12:32:48 +00:00
normal 1df9c2bc1c thread_sync.c (mutex_ptr): only reinitalize waitqueue at fork
Mutexes need to remain locked after forking.

This fixes "[BUG] invalid keeping_mutexes: Attempt to unlock a
mutex which is locked by another thread" and should
fix test_fork_while_parent_locked failures in CI

[ruby-core:90581] [Bug #15424]
[ruby-core:90595] [Bug #15430]

Fixes: r66230 ("handle mutexes held by parent threads in children")

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@66438 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-12-18 09:21:05 +00:00
nobu 98e65d9d92 Prefer rb_check_arity when 0 or 1 arguments
Especially over checking argc then calling rb_scan_args just to
raise an ArgumentError.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@66238 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-12-06 07:49:24 +00:00
normal 818f1c65ab thread_sync.c (mutex_ptr): handle mutexes held by parent threads in children
Mutexes may be held by threads which only exist in the parent
process, so their waitqueues may be populated with references
to other dead threads.  We must reset them at fork.

I am a moron for introducing this bug :<

[ruby-core:90312] [Bug #15383]

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@66230 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-12-05 18:58:45 +00:00
marcandre 18fe495902 Add some missing documentation to Queue#{close|deq} [ci skip]
Patch by Lars Kanis. [Fix GH-1949]

*  Describe the impact of Queue#close to Queue#deq .

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@64756 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-09-16 02:45:16 +00:00
normal 70d6c1a102 thread_sync.c: common wakeup_{one,all} implementation
This let us avoid looping in rb_szqueue_max_set, saves us
some lines of code and reduces binary size slightly
(numbers from 32-bit x86):

         text      data     bss     dec     hex filename
before: 91272       392     156   91820   166ac thread.o
 after: 91200       392     156   91748   16664 thread.o

Inspiration from this taken from the FUTEX_WAKE op
of the Linux futex(2) syscall.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@64542 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-08-26 12:41:16 +00:00
normal b0253a7569 thread.c: use rb_hrtime_t scalar for high-resolution time operations
Relying on "struct timespec" was too annoying API-wise and
used more stack space.  "double" was a bit wacky w.r.t rounding
in the past, so now we'll switch to using a 64-bit type.

Unsigned 64-bit integer is able to give us over nearly 585
years of range with nanoseconds.  This range is good enough
for the Linux kernel internal time representation, so it
ought to be good enough for us.

This reduces the stack usage of functions while GVL is held
(and thus subject to marking) on x86-64 Linux (with ppoll):

	rb_wait_for_single_fd    120 => 104
	do_select                120 => 88

[ruby-core:88582] [Misc #15014]

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@64533 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-08-25 06:58:35 +00:00
normal 79bdb6f9c8 thread*.c: replace GetMutexPtr with mutex_ptr
Following ko1's lead in r59192, this gets rid of non-obvious
assignments which happen inside macros.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@64490 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-08-21 01:01:42 +00:00