There's no point in running these transformation for certain files,
mainly anything from src/boot/efi and src/shared/linux, as this code
doesn't have access to our internal utility functions
For example, following transformation:
- isempty(s) ? NULL : s
+ empty_to_null(s)
would get applied to the empty_to_null function itself as well,
causing an infinite recursion, like:
--- src/basic/string-util.h
+++ /tmp/cocci-output-307-9f76e6-string-util.h
@@ -50,11 +50,11 @@ static inline bool isempty(const char *p
}
static inline const char *empty_to_null(const char *p) {
- return isempty(p) ? NULL : p;
+ return empty_to_null(p);
}
Let's avoid that by checking the current match position
For example, the following transformation:
- sizeof(s)-1
+ STRLEN(s)
would replace sizeof by STRLEN even in the STRLEN macro definition
itself, which generates following nonsensical patch:
--- src/basic/macro.h
+++ /tmp/cocci-output-8753-b50773-macro.h
@@ -182,7 +182,7 @@ static inline unsigned long ALIGN_POWER2
* Contrary to strlen(), this is a constant expression.
* @x: a string literal.
*/
-#define STRLEN(x) (sizeof(""x"") - 1)
+#define STRLEN(x) (STRLEN("" x ""))
/*
* container_of - cast a member of a structure out to the containing structure
Let's exclude the macro itself from the transformation to avoid this
The `coccinelle/take-fd.cocci` transformation file attempts to rewrite
r = fd;
fd = -1;
to
r = TAKE_FD(fd);
Unfortunately, using `identifier` or `idexpression` as a metavariable
type in this case wouldn't match more complex location descriptions,
like:
x->fd = fd
fd = -1;
Using 'expression' metavariable type generates false positives,
as you can't specify scope of such expression. The only real example
from the current codebase is the global 'errno' variable, which results
in following patch generated by `spatch`:
--- src/basic/errno-util.h
+++ /tmp/cocci-output-28263-971baa-errno-util.h
@@ -15,8 +15,7 @@ static inline void _reset_errno_(int *sa
#define UNPROTECT_ERRNO \
do { \
- errno = _saved_errno_; \
- _saved_errno_ = -1; \
+ errno = TAKE_FD(_saved_errno_); \
} while (false)
static inline int negative_errno(void) {
Let's explicitly state that the matched expression should not equal
'errno' to avoid this. It's not particularly nice, but it should be
enough, at least for now.
Coccinelle needs a custom isomorphism file with rules (isomorphisms) how
to correctly rewrite conditions with explicit NULL checks (i.e.
if (ptr == NULL)) to their shorter form (i.e. if (!ptr)). Coccinelle
already contains such isomorphisms in its default .iso file, however,
they're in the opposite direction, which results in useless output from
coccinelle/equals-null.cocci.
With this fix, `spatch` should no longer report patches like:
@@ -628,8 +628,9 @@ static int path_deserialize_item(Unit *u
f = path_result_from_string(value);
if (f < 0)
log_unit_debug(u, "Failed to parse result value: %s", value);
- else if (f != PATH_SUCCESS)
- p->result = f;
+ else {if (f != PATH_SUCCESS)
+ p->result = f;
+ }
} else
log_unit_debug(u, "Unknown serialization key: %s", key);
When booting with "udev.log-priority=debug" for example, the output might be
spammed with messages like this:
systemd-udevd[23545]: maximum number (248) of children reached
systemd-udevd[23545]: maximum number (248) of children reached
systemd-udevd[23545]: maximum number (248) of children reached
systemd-udevd[23545]: maximum number (248) of children reached
systemd-udevd[23545]: maximum number (248) of children reached
systemd-udevd[23545]: maximum number (248) of children reached
systemd-udevd[23545]: maximum number (248) of children reached
While the message itself is useful, printing it per batch of events should be
enough.
By default, the available completions are sorted alphabetically, which
is counterproductive in case of syslog priorities. Override the default
behavior using the `nosort` option
When systemd is started, it detects initrd by checking for that file
The usage of that file is not documented anywhere, so mention it early
in the most relevant man-page I could find.
When we determine that a calendar expression cannot elapse anymore,
print a warning but proceed regardless like we normally would.
Quite possibly a remote system has a different understanding of time
(timezone, system clock) than we have, hence we really shouldn't change
behaviour here client side, but log at best, and then leave the decision
what to do to the server side.
Follow-up for #12299
- bridge or bonding master takes a reference of slave links.
- drop link from bridge or bonding master's slave list when slave link
is removed.
- change type of Link::slaves to Set*,
Fixes#12315.
When a uevent is received during the relevant interface is in
LINK_STATE_PENDING, then the interface may be initialized twice.
To prevent that, this introduces LINK_STATE_INITIALIZED.
https://github.com/systemd/systemd/issues/7153#issuecomment-485252308
Apparently this is still confusing for people.
Longer-term, I think we should just make BindMount= automatically "upgrade"
(or "downgrade", depending on how you look at this), any InaccessiblePath=
mountpoints to "tmpfs". I don't see much point in forcing users to remember
this interaction. But let's at least document the status quo, we can always
update the docs if the code changes.
A service might be able to detect errors by itself that may require the
system to take the same action as if the service locked up. Add a
WATCHDOG=trigger state change notification to sd_notify() to let the
service manager know about the self-detected misery and instantly
trigger the configured watchdog behaviour.
The interface provided by those two functions is huge, so this text could
probably be made two or three times as long if all details were described.
But I think it's a good start.
This wraps the call to org.freedesktop.DBus.Introspectable.Introspect.
Using "busctl call" directly is inconvenient because busctl escapes the
string before printing.
Example:
$ busctl introspect --xml org.freedesktop.systemd1 /org/freedesktop/systemd1 | pygmentize -lxml | less -RF
test-bus-introspect is also applied to the tables from test-bus-vtable.c.
test-bus-vtable.c is also used as C++ sources to produce test-bus-vtable-cc,
and our hashmap headers are not C++ compatible. So let's do the introspection
part only in the C version.