pika
pika copied to clipboard
pika-org
pika builds on C++ std::execution with fiber, CUDA, HIP, and MPI support.
Fixes #699 - [x] Add a CI config using a different CXX and CUDA standard - [x] Add a GPU stdexec config with different CXX and CUDA standards Note: Looks...
## Expected Behavior build and test successfully ## Actual Behavior - build on x86_64: ```bash 99% tests passed, 2 tests failed out of 190 Label Time Summary: COMPILE_ONLY = 28.38...
## Actual Behavior [log](https://freebsd.org/~yuri/pika-0.21.0-tests.log) ## Steps to Reproduce the Problem Running tests according to the [docs](https://pikacpp.org/usage.html#testing). - pika version (commit/tag): 0.21.0 - Platform (OS, compiler, etc.): FreeBSD 13.2, clang-16
In addition to the current performance benchmarks, it would make sense to monitor the number of instructions using valgrind's cachegrind tool. This should have significantly less variation than timing benchmarks...
#891 adds a special pipeline configuration for pushing performance data to elastic. Once we've reduced duplication in the pipelines and added release configurations to CSCS CI configurations (#862), we should...
The initial number 128 is currently hardcoded. This should be a configurable number.
We currently disable the `schedule_from` customization for `thread_pool_scheduler` with stdexec: https://github.com/pika-org/pika/blob/cb8ebe0667c903cd3155201b2a356f0e15b19855/libs/pika/executors/include/pika/executors/thread_pool_scheduler.hpp#L269-L287. We should make sure the customization is used even with stdexec enabled. In the case of this example, `transfer`...
`any_receiver` should be possible to convert into more or less a reference. The non-type-erased receiver should be possible to store in the operation state, with `any_receiver` simply pointing to the...
While working on #774 there are a lot of use cases of `pika::async` that need to be replaced with something more explicit, but also more verbose. The current `pika::async` could...