Christoph Hertzberg
2c083ace3e
Provide EIGEN_OVERRIDE and EIGEN_FINAL macros to mark virtual function overrides
2018-09-24 18:01:17 +02:00
Gael Guennebaud
c696dbcaa6
Fiw shadowing of last and all
2018-09-21 23:02:33 +02:00
Rasmus Munk Larsen
44d8274383
Cast to longer type.
2018-09-19 13:31:42 -07:00
Rasmus Munk Larsen
d638b62dda
Silence compiler warning.
2018-09-19 13:27:55 -07:00
Rasmus Munk Larsen
febd09dcc0
Silence compiler warnings in ThreadPoolInterface.h.
2018-09-19 11:11:04 -07:00
Ravi Kiran
1f0c941c3d
Collapsed revision
...
* Merged eigen/eigen into default
2018-09-17 18:29:12 -07:00
Christoph Hertzberg
007f165c69
bug #1598 : Let MaxSizeVector respect alignment of objects and add a unit test
...
Also revert 8b3d9ed081
2018-09-14 20:21:56 +02:00
Gael Guennebaud
7f3b17e403
MSVC 2015 supports c++11 thread-local-storage
2018-09-13 18:15:07 +02:00
Rasmus Munk Larsen
8b3d9ed081
Use padding instead of alignment attribute, which MaxSizeVector does not respect. This leads to undefined behavior and hard-to-trace bugs.
2018-09-05 11:20:06 -07:00
Rasmus Munk Larsen
744e2fe0de
Address comments about EIGEN_THREAD_LOCAL.
2018-08-24 10:24:54 -07:00
Rasmus Munk Larsen
8d9bc5cc02
Fix g++ compilation.
2018-08-23 13:06:39 -07:00
Rasmus Munk Larsen
e9f9d70611
Don't rely on __had_feature for g++.
...
Don't use __thread.
Only use thread_local for gcc 4.8 or newer.
2018-08-23 12:59:46 -07:00
Rasmus Munk Larsen
668690978f
Pad PerThread when we emulate thread_local to prevent false sharing.
2018-08-23 12:54:33 -07:00
Rasmus Munk Larsen
6cedc5a9b3
rename mu.
2018-08-23 12:11:58 -07:00
Rasmus Munk Larsen
6e0464004a
Store std::unique_ptr instead of raw pointers in per_thread_map_.
2018-08-23 12:10:08 -07:00
Rasmus Munk Larsen
d35880ed91
merge
2018-08-23 11:36:49 -07:00
Christoph Hertzberg
a709c8efb4
Replace pointers by values or unique_ptr for better leak-safety
2018-08-23 19:41:59 +02:00
Rasmus Munk Larsen
15d4f515e2
Use plain_assert in destructors to avoid throwing in CXX11 tests where main.h owerwrites eigen_assert with a throwing version.
2018-08-14 12:17:46 -07:00
Rasmus Munk Larsen
9bb75d8d31
Add Barrier.h.
2018-08-13 15:34:03 -07:00
Rasmus Munk Larsen
8278ae6313
Add support for thread local support on platforms that do not support it through emulation using a hash map.
2018-08-13 15:31:23 -07:00
Eugene Zhulenev
e204ecdaaf
Remove SimpleThreadPool and always use {NonBlocking}ThreadPool
2018-07-16 15:06:57 -07:00
Gael Guennebaud
b3fd93207b
Fix typos found using codespell
2018-06-07 14:43:02 +02:00
Benoit Steiner
dc524ac716
Fixed compilation warning
2017-07-06 21:11:15 -07:00
Rasmus Munk Larsen
bfd7bf9c5b
Get rid of Init().
2017-03-10 08:48:20 -08:00
Rasmus Munk Larsen
d56ab01094
Use C++11 ctor forwarding to simplify code a bit.
2017-03-10 08:30:22 -08:00
Rasmus Munk Larsen
344c2694a6
Make the non-blocking threadpool more flexible and less wasteful of CPU cycles for high-latency use-cases.
...
* Adds a hint to ThreadPool allowing us to turn off spin waiting. Currently each reader and record yielder op in a graph creates a threadpool with a thread that spins for 1000 iterations through the work stealing loop before yielding. This is wasteful for such ops that process I/O.
* This also changes the number of iterations through the steal loop to be inversely proportional to the number of threads. Since the time of each iteration is proportional to the number of threads, this yields roughly a constant spin time.
* Implement a separate worker loop for the num_threads == 1 case since there is no point in going through the expensive steal loop. Moreover, since Steal() calls PopBack() on the victim queues it might reverse the order in which ops are executed, compared to the order in which they are scheduled, which is usually counter-productive for the types of I/O workloads the single thread pools tend to be used for.
* Store num_threads in a member variable for simplicity and to avoid a data race between the thread creation loop and worker threads calling threads_.size().
2017-03-09 15:41:03 -08:00
Benoit Steiner
3beb180ee5
Don't call EnvThread::OnCancel by default since it doesn't do anything.
2016-12-14 18:33:39 -08:00
Benoit Steiner
8ae68924ed
Made ThreadPoolInterface::Cancel() an optional functionality
2016-12-12 11:58:38 -08:00
Benoit Steiner
2f5b7a199b
Reworked the threadpool cancellation mechanism to not depend on pthread_cancel since it turns out that pthread_cancel doesn't work properly on numerous platforms.
2016-12-09 13:05:14 -08:00
Benoit Steiner
28ee8f42b2
Added a Flush method to the RunQueue
2016-12-08 14:07:56 -08:00
Benoit Steiner
69ef267a77
Added the new threadpool cancel method to the threadpool interface based class.
2016-12-08 14:03:25 -08:00
Benoit Steiner
7bfff85355
Added support for thread cancellation on Linux
2016-12-08 08:12:49 -08:00
Benoit Steiner
eb6ba00cc8
Properly size the list of waiters
2016-09-12 10:31:55 -07:00
Benoit Steiner
13df3441ae
Use MaxSizeVector instead of std::vector: xcode sometimes assumes that std::vector allocates aligned memory and therefore issues aligned instruction to initialize it. This can result in random crashes when compiling with AVX instructions enabled.
2016-09-02 19:25:47 -07:00
Gael Guennebaud
35a8e94577
bug #1167 : simplify installation of header files using cmake's install(DIRECTORY ...) command.
2016-08-29 10:59:37 +02:00
Benoit Steiner
e256acec7c
Avoid unecessary object copies
2016-08-01 17:03:39 -07:00
Rasmus Munk Larsen
a9c1e4d7b7
Return -1 from CurrentThreadId when called by thread outside the pool.
2016-06-23 16:40:07 -07:00
Rasmus Munk Larsen
d39df320d2
Resolve merge.
2016-06-23 15:08:03 -07:00
Benoit Steiner
a29a2cb4ff
Silenced a couple of compilation warnings generated by xcode
2016-06-22 16:43:02 -07:00
Benoit Steiner
f8fcd6b32d
Turned the constructor of the PerThread struct into what is effectively a constant expression to make the code compatible with a wider range of compilers
2016-06-22 16:03:11 -07:00
Benoit Steiner
aedc5be1d6
Avoid generating pseudo random numbers that are multiple of 5: this helps
...
spread the load over multiple cpus without havind to rely on work stealing.
2016-06-14 17:51:47 -07:00
Rasmus Munk Larsen
f1f2ff8208
size_t -> int
2016-06-03 18:06:37 -07:00
Rasmus Munk Larsen
76308e7fd2
Add CurrentThreadId and NumThreads methods to Eigen threadpools and TensorDeviceThreadPool.
2016-06-03 16:28:58 -07:00
Benoit Steiner
c3c8ad8046
Align the first element of the Waiter struct instead of padding it. This reduces its memory footprint a bit while achieving the goal of preventing false sharing
2016-06-02 21:17:41 -07:00
Benoit Steiner
1ae2567861
Fixed some compilation warnings
2016-05-26 15:57:19 -07:00
Benoit Steiner
2a54b70d45
Fixed potential race condition in the non blocking thread pool
2016-05-12 11:45:48 -07:00
Christoph Hertzberg
2150f13d65
fixed some double-promotion and sign-compare warnings
2016-05-11 23:02:26 +02:00
Benoit Steiner
6a5717dc74
Explicitely initialize all the atomic variables.
2016-05-11 10:04:41 -07:00
Benoit Steiner
dc7dbc2df7
Optimized the non blocking thread pool:
...
* Use a pseudo-random permutation of queue indices during random stealing. This ensures that all the queues are considered.
* Directly pop from a non-empty queue when we are waiting for work,
instead of first noticing that there is a non-empty queue and
then doing another round of random stealing to re-discover the non-empty
queue.
* Steal only 1 task from a remote queue instead of half of tasks.
2016-05-09 10:17:17 -07:00
Gael Guennebaud
3dddd34133
Refactor the unsupported CXX11/Core module to internal headers only.
2016-04-26 11:20:25 +02:00
