The thread flag signaling mechanism is problematic if a separate event
loop thread is not used. The problem is, that one needs to know the
thread id of the loop dispathcer, and that knowledge can be derived
only from the caller thread of the initialization function.
Remove the magic connection from caller thread to dispatcher thread
by using event flags instead of thread flags on signaling. As the
event flags require the massive amount 20 bytes of RAM whereas
thread flags requires none, keep the code behind flag.
The separate eventloop thread may not be necessary on all uses, as one
can use the existing main thread for event dispatching. Add a
conditional nanostack-hal.event-loop-dispatch-from-application, which
disables the thread creation.
Note: the ns_hal_init must be ran from the same thread which will be
used to execute the event loop later.
Few targets need more than just pending IRQ set. They include some flags to be
set that are checked in IRQ handler. This is the case for targets in this
commit.
b2124a5 Fixed multiple deploy steps in Travis
949015a Merge pull request #28 from geky/configurables
67daf9e Added cross-compile targets for testing
a3fd2d4 Added more configurable utils
a0a55fb Added conversion to/from little-endian on disk
4f08424 Added software implementations of bitwise instructions
59ce49f Merge pull request #26 from Sim4n6/master
2f8ae34 Added a git ignore file with .o .d blocks dir and lfs bin
e611cf5 Fix incorrect lookahead population before ack
a25743a Fixed some minor error code differences
6716b55 Fixed error check when truncating files to larger size
809ffde Merge pull request #24 from aldot/silence-shadow-warnings-1
dc513b1 Silenced more of aldot's warnings
aa50e03 Commentary typo fix
6d55755 tests: Silence warnings in template
029361e Silence shadow warnings
fd04ed4 Added autogenerated release notes from commits
3101bc9 Do not print command invocation if QUIET
d82e34c Merge pull request #21 from aldot/doc-tweaks
436707c doc: Editorial tweaks
3457252 doc: Spelling fixes
6d8e0e2 Moved -Werror flag to CI only
88f678f Fixed self-assign warning in tests
3ef4847 Added remove step in tests to force rebuild
f694b14 Merge pull request #16 from geky/versioning
5a38d00 Added deploy step in Travis to push new version as tags
035552a Add version info for software library and on-disk structures
997c2e5 Fixed incorrect reliance on errno in emubd
d88f0ac Added lfs_file_truncate
2ad435e Added files test to littlefs-fuse tests in Travis
1fb6a19 Reduced ctz traverse runtime by 2x
db88727 Added error code LFS_ERR_NOTEMPTY
c2fab8f Added asserts on geometry and updated config documentation
472ccc4 Fixed file truncation without writes
aea3d3d Fixed positive seek bounds checking
be22d34 Updated links to Mbed OS
425aa3c Fixed issue with immediate exhaustion and small unaligned storage
git-subtree-dir: features/filesystem/littlefs/littlefs
git-subtree-split: b2124a5ae54717737f6d96cb24e6f31e02158d1e
Before this patch, when calling lfs_mkdir or lfs_file_open with root
as the target, littlefs wouldn't find the path properly and happily
run into undefined behaviour.
The fix is to populate a directory entry for root in the lfs_dir_find
function. As an added plus, this allowed several special cases around
root to be completely dropped.