Age | Commit message (Collapse) | Author | |
---|---|---|---|
2021-03-16 | Lots of performance tuning of native Linux implementation, but still 5-10% ↵ | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
slower than libdispatch. | |||
2021-03-16 | Simplify non-native-Linux implementation. Appears to make little difference ↵ | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
to benchmarks. | |||
2021-03-16 | Add benchmark results for GCD on Linux, Win32 thread pools on Windows, and ↵ | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
ASIO on both Linux and Windows. My native implementation is overcreating threads currently, so not worth recording those yet. | |||
2021-03-16 | Implement dynamic_thread_pool_group benchmark. | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
2021-03-16 | Begin dynamic_thread_pool_group benchmark program. | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
2021-03-16 | Finally got a fully working native implementation on Linux. I had realised ↵ | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
that attempting to fix this each morning before work was turning into never actually making progress, so I carved out four hours this morning without distraction, and boom here we go. I next need to simplify this implementation, it has become bloated with all the changes, it definitely could be made much leaner. | |||
2021-03-16 | more wip | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
2021-03-16 | Don't leak timer wait objects. | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
2021-03-16 | More wip | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
2021-03-16 | More improvements to dynamic scaling within native Linux implementation of ↵ | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
dynamic_thread_pool_group, but still not working quite right yet. | |||
2021-03-16 | Dynamic scaling within native Linux implementation of ↵ | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
dynamic_thread_pool_group is somewhat working now, but it's not working quite right yet. | |||
2021-03-16 | Implement thread enumeration for Linux native dynamic_thread_pool_group. | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
2021-03-16 | Make the dynamic thread pool work scheduling not quite as strict in the ↵ | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
native Linux implementation, now we only examine higher priority groups once per group cycle. This produces behaviours much more consistent with the Win32 thread pool and Grand Central Dispatch. | |||
2021-03-16 | wip native linux threadpool implementation for dynamic_thread_pool_group. | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
2021-03-16 | Hopefully fix failure to find libdispatch on Mac OS. Also hopefully fix test ↵ | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
failure on Windows. | |||
2021-03-16 | Implemented and debugged a Grand Unified Dispatch backend for ↵ | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
dynamic_thread_pool_group. Works surprisingly nicely on Linux, haven't actually tested it on Mac OS nor FreeBSD, but no reason it shouldn't work just fine. | |||
2021-03-16 | Implement dynamic_thread_pool_group::io_aware_work_item. | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
2021-03-16 | Add unit test to ensure nested dynamic thread pool groups work as expected. ↵ | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
Still Windows only implementation. | |||
2021-03-16 | wip dynamic_thread_pool_group, so far has Windows support only. Note that ↵ | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
LLFIO now hard requires Windows 7 rather than Vista. | |||
2021-03-16 | Add statfs_t::f_iosinprogress and statfs_t::f_ioswaittime which | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
can be used to monitor how congested the underlying hardware device for a handle is. | |||
2021-03-15 | Fix inability to open directory junctions on Windows using `symlink_handle` ↵ | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
(issue #73). | |||
2021-03-15 | Make the presence of `__cpp_modules` not have an effect on its own (#75) | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
2021-03-08 | Suppress warning on Linux. | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
2021-03-06 | Fix lots of bugs due to newer kernel 5.8 in Ubuntu 20.04, and OpenZFS v0.8: | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
1. clone_extents_to() didn't handle partial clones, and now it does. 2. OpenZFS has a bug in SEEK_DATA which caused empty clones, we use a hacky workaround to fix it. 3. mapped_file_handle now fails loudly if it fails to allocate the map due to VMA exhaustion. This closes off a ton of unpleasant surprise in heavily loaded applications. 4. If your Linux kernel was old enough, or you are on ZFS, renaming a mapped_file_handle caused internal state corruption which led to lots of bad things, including `bad_file_descriptor` errors. 5. On Windows, renaming a mapped_file_handle now tears down and restores the maps, as you cannot rename a file with open maps. | |||
2021-02-23 | Fix breakage caused by upstream changes to quickcpplib. De-CRLF cached ↵ | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
filelist files. | |||
2021-02-17 | mapped_file_handle::read() and write() was not detecting when VA space had ↵ | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
been exhausted, and was incorrectly segfaulting the process rather than returning errc::not_enough_memory. | |||
2021-02-16 | Bump _WIN32_WINNT header requirements. | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
2021-02-16 | Fix issue #72 use of NtFlushBuffersEx on Win7, when it is a Win8 API. | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
2021-02-15 | Port LLFIO to latest Outcome, whose TRY operation now uses value semantics ↵ | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
for the unique temporary and therefore all TRY destinations ought to be rvalue refs. | |||
2021-02-03 | Bump _WIN32_WINNT to Win7. | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
2021-02-03 | Suppress warning on older GCCs and clangs. | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
2021-02-03 | Don't install ntkernel-error-category in status-code builds. | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
2021-02-03 | Merge branch 'develop' of https://github.com/ned14/llfio into develop | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
2021-02-03 | Suppress spurious warning on older GCCs. | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
2021-02-02 | Fix lots of warnings on x86 windows. | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
2021-02-02 | Add LLFIO_ENABLE_DEPENDENCY_SMOKE_TEST option for vcpkg. | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
2021-02-01 | Stop warning from to_win32_path() on POSIX. | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
2021-01-29 | Fix last commit on VS2017. | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
2021-01-29 | Port from AFIO v1 `normalise_path()`, whose API I have improved and made ↵ | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
more directed as `to_win32_path()`. | |||
2021-01-27 | Add the beginnings of a to_win32_path(), though it's not finished yet. | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
2021-01-25 | Fix construct<mapped_file_handle> having the wrong parameter order. | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
2021-01-19 | Hopefully fix installability on CI. | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
2020-12-17 | From now on, use Outcome master branch. Also, a few more small fixes to ↵ | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
path_view reference implementation to better match the WG21 proposal. | |||
2020-12-01 | Try more github actions instability workarounds. | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
2020-12-01 | Hopefully fix weird random github actions failure. | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
2020-11-28 | Finally got back onto doing some work on the test io_uring multiplexer. | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
2020-11-27 | Work around build failure in status code config on VS2017. | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
2020-11-26 | Debugged the .ci.cmake script for prebuilt binaries on my own computer, this ↵ | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
time ought to surely work, I hope. | |||
2020-11-25 | Yet again prebuilt binaries. | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |
2020-11-25 | Again prebuilt binaries | Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) | |