There are more than enough calls doing string manipulations to deserve
its own files, hence do something about it.
This patch also sorts the #include blocks of all files that needed to be
updated, according to the sorting suggestions from CODING_STYLE. Since
pretty much every file needs our string manipulation functions this
effectively means that most files have sorted #include blocks now.
Also touches a few unrelated include files.
pthread APIs (unlike the rest of libc) return their errors as positive
error codes directly from the functions, rather than using errno. Let's
make sure we always handle things that way.
Currently, we guarantee that if two event-sources with the same priority
fire at the same time, they're always dispatched in the same order. While
this might sound nice in theory, there's is little benefit in providing
stability on that level. We have no control over the order the events are
reported, hence, we cannot guarantee that we get notified about both at
the same time.
By dropping the stability guarantee, we loose roughly 10% Heap swaps in
the prioq on a desktop cold-boot. Krzysztof Kotlenga even reported up to
20% on his tests. This sounds worth optimizing, so drop the stability
guarantee.
We should never access the "signal" part of the event source unless the
event source is actually for a signal. In this case it's a child pid
handler however, hence make sure to use the right signal.
This is a fix for PR #1177, which in turn was a fix for
9da4cb2be2.
RT signals operate in a queue, and we should be careful to never merge
two queued signals into one. Hence, makes sure we only ever dequeue a
single signal at a time and leave the remaining ones queued in the
signalfd. In order to implement correct priorities for the signals
introduce one signalfd per priority, so that we only process the highest
priority signal at a time.
Previously, if the event loop never ran before sd_event_now() would
fail. With this change it will instead fall back to invoking now(). This
way, the function cannot fail anymore, except for programming error when
invoking it with wrong parameters.
This takes into account the fact that many callers did not handle the
error condition correctly, and if the callers did, then they kept simply
invoking now() as fall back on their own. Hence let's shorten the code
using this call, and make things more robust, and let's just fall back
to now() internally.
Whether now() is used or the cache timestamp may still be detected via
the return value of sd_event_now(). If > 0 is returned, then the fall
back to now() was used, if == 0 is returned, then the cached value was
returned.
This patch also simplifies many of the invocations of sd_event_now():
the manual fall back to now() can be removed. Also, in cases where the
call is invoked withing void functions we can now protect the invocation
via assert_se(), acknowledging the fact that the call cannot fail
anymore except for programming errors with the parameters.
This change is inspired by #841.
If we call EPOLL_CTL_DEL, we *REALLY* expect the file-descriptor to be
present in that given epoll-set. We actually track such state via our
s->io.registered flag, so it better be true.
Make sure if that's not true, we treat it similar to assert_return() (ie.,
print a loud warning).
This ports a lot of manual code over to sigprocmask_many() and friends.
Also, we now consistly check for sigprocmask() failures with
assert_se(), since the call cannot realistically fail unless there's a
programming error.
Also encloses a few sd_event_add_signal() calls with (void) when we
ignore the return values for it knowingly.
We protect most of the API from use accross forks, but we still allow both
sd_event and sd_event_source objects to be unref'ed. This would cause
problems as it would unregister sources from the underlying eventfd, hence
also affecting the original instance in the parent process.
This fixes the issue by not touching the fds on unref when done accross a fork,
but still free the memory.
This fixes a regression introduced by
"udevd: move main-loop to sd-event": 693d371d30
where the worker processes were disabling the inotify event source in the
main daemon.
Currently the code will silently blank out events if there are more
then 512 epoll events, causing them never to be handled at all. This
patch removes the cap on the number of events for epoll_wait, thereby
avoiding this issue.
This patch removes includes that are not used. The removals were found with
include-what-you-use which checks if any of the symbols from a header is
in use.
As a followup to 086891e5c1 "log: add an "error" parameter to all
low-level logging calls and intrdouce log_error_errno() as log calls
that take error numbers", use sed to convert the simple cases to use
the new macros:
find . -name '*.[ch]' | xargs sed -r -i -e \
's/log_(debug|info|notice|warning|error|emergency)\("(.*)%s"(.*), strerror\(-([a-zA-Z_]+)\)\);/log_\1_errno(-\4, "\2%m"\3);/'
Multi-line log_*() invocations are not covered.
And we also should add log_unit_*_errno().
When a child event is disabled (in order to be freed) and there is no
SIGCHLD signal event, sd_event_source_set_enabled will disable SIGCHLD
even if there are other child events.
Also remove some unneeded signalfd updates.
https://bugs.freedesktop.org/show_bug.cgi?id=84659
Based-on-a-patch-by: Hristo Venev <mustrumr97@gmail.com>
It is redundant to store 'hash' and 'compare' function pointers in
struct Hashmap separately. The functions always comprise a pair.
Store a single pointer to struct hash_ops instead.
systemd keeps hundreds of hashmaps, so this saves a little bit of
memory.
This will allow sd-event to be integrated into an external event loop, which
in turn will allow (say) glib-based applications to use our various libraries,
without manually integrating each of them (bus, rtnl, dhcp, ...).
The external event-loop should integrate sd-event int he following way:
Every iteration must start with a call to sd_event_prepare(), which will
return 0 if no event sources are ready to be processed, a positive value if
they are and a negative value on error. sd_event_prepare() may only be called
following sd_event_dispatch(); a call to sd_event_wait() indicating that no
sources are ready to be dispatched; or a failed call to sd_event_dispatch() or
sd_event_wait().
A successful call to sd_event_prepare() indicating that no event sources are
ready to be dispatched must be followed by a call to sd_event_wait(),
which will return 0 if it timed out without event sources being ready to
be processed, a negative value on error and a positive value otherwise.
sd_event_wait() may only be called following a successful call to
sd_event_prepare() indicating that no event sources are ready to be dispatched.
If sd_event_wait() indicates that some events sources are ready to be
dispatched, it must be followed by a call to sd_event_dispatch(). This
is the only time sd_event_dispatch() may be called.
This is not certain to be likely.
Lennart says: a frequent usecase is invoking some function regularly in intervals
in such a case every single iteration we'll have to rearm
A call to sd_event_source_set_io_events() skipps calling into the kernel
if the new event-mask matches the old one. This is safe for
level-triggered sources as the kernel moves them onto the ready-list
automatically if events change. However, edge-triggered sources might not
be on the ready-list even though events are present.
A call to sd_event_source_set_io_events() with EPOLLET set might thus be
used to just move the io-source onto the ready-list so the next poll
will return it again. This is very useful to avoid starvation in
priority-based event queues.
Imagine a read() loop on an edge-triggered fd. If we cannot read data fast
enough to drain the receive queue, we might decide to skip reading for now
and schedule it for later. On edge-triggered io-sources we have to make
sure it's put on the ready-list so the next dispatch-round will return it
again if it's still the highest priority task. We could make sd-event
handle edge-triggered sources directly and allow marking them ready again.
However, it's much simpler to let the kernel do that for now via
EPOLL_CTL_MOD.
These are the counterpart of "floating" bus slots, i.e. event sources
that are bound to the lifetime of the event object itself, and thus
don't require an explicit reference to be kept.
safe_close_pair() is more like safe_close(), except that it handles
pairs of fds, and doesn't make and misleading allusion, as it works
similarly well for socketpairs() as for pipe()s...
safe_close() automatically becomes a NOP when a negative fd is passed,
and returns -1 unconditionally. This makes it easy to write lines like
this:
fd = safe_close(fd);
Which will close an fd if it is open, and reset the fd variable
correctly.
By making use of this new scheme we can drop a > 200 lines of code that
was required to test for non-negative fds or to reset the closed fd
variable afterwards.
This new event source is triggered by the dispatching of any non-post
event source. It can thus be used to do clean-up or recheck work,
triggered by any other event source having been executed.
This is different from "defer" event source which are unconditionally
triggered as long as they are enabled. A "defer" event source that does
nothing will result in the event loop busy looping unless it is turned
off eventually. This is different for "post" event sources that will be
only triggered when some other non-post ran, and will thus not keep the
event loop busy on its own.
There was a copy-paste error introduced in commit c2ba3ad660
which causes the following error when using timer units:
Assertion '(x->type == SOURCE_MONOTONIC && y->type == SOURCE_MONOTONIC) || (x->type == SOURCE_REALTIME && y->type == SOURCE_REALTIME)'
failed at src/libsystemd/sd-event/sd-event.c:264, function latest_time_prioq_compare(). Aborting.
Previously the returned object of constructor functions where sometimes
returned as last, sometimes as first and sometimes as second parameter.
Let's clean this up a bit. Here are the new rules:
1. The object the new object is derived from is put first, if there is any
2. The object we are creating will be returned in the next arguments
3. This is followed by any additional arguments
Rationale:
For functions that operate on an object we always put that object first.
Constructors should probably not be too different in this regard. Also,
if the additional parameters might want to use varargs which suggests to
put them last.
Note that this new scheme only applies to constructor functions, not to
all other functions. We do give a lot of freedom for those.
Note that this commit only changes the order of the new functions we
added, for old ones we accept the wrong order and leave it like that.
If -flto is used then gcc will generate a lot more warnings than before,
among them a number of use-without-initialization warnings. Most of them
without are false positives, but let's make them go away, because it
doesn't really matter.
Setting a child-source state to anything but SD_EVENT_OFF currently does
nothing. The condition logic is flawed. Move the state update *below* the
test for the previous state.
Fixes a crash if you call:
sd_event_source_set_enabled(source, SD_EVENT_ON);
sd_event_source_unref(source);
on a child-source in its own callback.
