| Commit message (Collapse) | Author | Age | Files | Lines |
| |
|
|
|
|
|
| |
Meson is now used to build this package in all of the necessary places.
Change-Id: Ic5155621967fa94ec03eced2193e0d453b174eba
Signed-off-by: William A. Kennington III <wak@google.com>
|
| |
|
|
|
|
|
|
|
| |
Now example binaries and tests don't have to be aware of the nuances for
configuring the build to use the library. This makes the project library
look like any other system dependency.
Change-Id: Ife3bcd3dcbd423f08c513f9a05fc8cabff41e2f6
Signed-off-by: William A. Kennington III <wak@google.com>
|
| |
|
|
|
|
|
|
| |
Arguments including the function are now passed with reference types
forwarded.
Change-Id: I32cf4704737faa6c140d16352528ceb0d738f5b8
Signed-off-by: William A. Kennington III <wak@google.com>
|
| |
|
|
|
|
|
|
|
|
|
|
| |
We don't ever use this constructor and it was effectively package
private. Remove it as we don't ever need to add references to our
sd_event_source*.
Tested:
Built and run through unit test suite.
Change-Id: I1ca9fd4ecfdef1a3fd5aa2ca214be657c1a04b94
Signed-off-by: William A. Kennington III <wak@google.com>
|
| |
|
|
|
|
|
|
|
|
|
| |
We no longer need to roll our own managed handle type now that stdplus
implements a generic one.
Tested:
Built and run through unit tests.
Change-Id: Id34cd8d3ffacf7901d49cac335fa93f744f0e310
Signed-off-by: William A. Kennington III <wak@google.com>
|
| |
|
|
|
|
|
|
|
| |
The code coverage macros from the archive changed in a backward
incompatible way. This adds a workaround to autodetect either version
and do the right thing
Change-Id: Ic010ed59fb332e19e5c76f60e55f864f8f4a43f8
Signed-off-by: William A. Kennington III <wak@google.com>
|
| |
|
|
|
|
|
|
|
|
|
| |
This also cleans up some outstanding issues detected by the
configuration.
Tested:
Ran through unit test scripts.
Change-Id: I3357a2280a681a87532d6d72315260db2ab73e87
Signed-off-by: William A. Kennington III <wak@google.com>
|
| |
|
|
|
|
|
|
| |
For now both autotools and meson will be maintained side by side.
Eventually we should remove the autotools build system.
Change-Id: I04382b17d5267ee218e2658fc163da17853f637a
Signed-off-by: William A. Kennington III <wak@google.com>
|
| |
|
|
|
|
|
|
|
|
|
| |
All of the other classes that take callbacks support having the callback
updated.
Tested:
Builds and passed through the unit test suite.
Change-Id: I0efcd074523fa6ec0cc84c7a930346dd9340f7ec
Signed-off-by: William A. Kennington III <wak@google.com>
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Now that we can update the callbacks of our sources, we can move the
timer object freely by updating the callback when moved.
Tested:
Unit tests pass, and we no longer see any valgrind issues when
moving the timer object.
Change-Id: I15baf97538459ca8b9c48b75dba77d09b7b5075b
Signed-off-by: William A. Kennington III <wak@google.com>
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Sometimes callers want to be able to update the callback used when the
source is acted upon. This is needed for updating references stored in
the callback.
Tested:
Run through the unit test suite.
Change-Id: I78bda32569287964bfc9d49501869d3a2b497f3d
Signed-off-by: William A. Kennington III <wak@google.com>
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Other event sources expose the ability to get th event used for
constructing the source. Since our timer is basically just a time
source, we should be able to trivially get the event object.
Tested:
Built and run through unit tests
Change-Id: I4b4192797652a903cba90d6e73771014df7702d6
Signed-off-by: William A. Kennington III <wak@google.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This change is meant to enable users of the old openbmc timer class to
trivially use the timer if they only want single executions. It also
makes setting up the timer less verbose if you do not already know the
timeout interval.
Tested:
Run through unit tests and did a sample integration with
phosphor-watchdog and phosphor-networkd. Verified that the new
oneshot example works as expected.
Change-Id: I2cd006d1f19fff99bce3f732a16eac9ca9553666
Signed-off-by: William A. Kennington III <wak@google.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
This makes the timer better resemble the other source types, and makes
it easier to access timer properties from the callback.
Since we do not yet have any committed users of this API it is safe to
change it.
Tested:
Unit tests were run and builds work fine.
Change-Id: I9ecb24a20bc822b2ab55900e47a7cae834ba873d
Signed-off-by: William A. Kennington III <wak@google.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
We often need a continually ticking timer for our daemons. This utility
wraps an sd_event time source as a convenience. This is meant to be a
usable replacement for the timer.hpp found in other openbmc projects.
Tested:
New tests pass with full coverage. Changes to the phosphor-watchdog
that rely on this utility work as expected.
Change-Id: Id12aed9e5b018e7eca825c4a7ac7b4f46e2f04c6
Signed-off-by: William A. Kennington III <wak@google.com>
|
| |
|
|
|
|
|
|
|
|
|
|
| |
The current build situation places the test wrappers and logs into the
top level test directory. We can actually build and run these binaries
ink the subdirectories containing the test sources. Do this.
Tested:
Run through the unit test suite and all test still show up and pass.
Change-Id: Ia5c9957242ae4df88dca6433682dc66bbbc3d684
Signed-off-by: William A. Kennington III <wak@google.com>
|
| |
|
|
|
| |
Change-Id: I0efd11f488ceee9b1a03ffaa1522cf0182ae1aca
Signed-off-by: William A. Kennington III <wak@google.com>
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
|
|
| |
These functions were implemented only in newer systemd 237+. We often
deal with older verisons so lets add them back after some time passes.
We could also make them an optional autodetected feature to work around
the issue.
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
| |
The callback mechanism will be used by other callback methods in the
same fashion, so we can template and re-use this functionality
|
| |
|
|
|
| |
This makes callbacks work correctly as the systemd callback functions
need to know where to look for the object storing the real callback.
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
| |
This shouldn't happen in practice but will be needed to supporting
moving.
|
| |
|
|
| |
Also expose a mechanism for retrieving our internal event.
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
| |
This also makes it possible to use in boolean expressions to check
validity.
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|