Mostly useful for testing purposes. Setting Age to 1s works just as
well, but it is surprising that using 0s (or just 0) does not work.
Also clarify this in the documentation.
d4e9eb91ea changed the behavior for the F and f actions, wrongly sending
them to glob_item(). Restore the old behavior and shortcut straight to
write_one_file().
This resolves problems with filesystems which do not implement the
aio_write file operation. In this case, the kernel will fall back using
a loop writing technique for each pointer in a received iovec. The
result is strange errors in dmesg such as:
[ 31.855871] elevator: type not found
[ 31.856262] elevator: switch to
[ 31.856262] failed
It does not make sense to implement a synchronous aio_write method for
sysfs as this isn't a real filesystem where a reasonable use case for
using writev exists, nor is there an expectation that tmpfiles will be
used to write more data than can be reasonably written in a single write
syscall.
In addition, some sysfs attrs are currently buggy and will NOT reject
the second write with the newline, causing the sysfs value to be zeroed
out. This of course should be fixed in the kernel regardless of any
wrongdoing in userspace, but this simple change makes us immune to such
a bug.
This change means that we do not write a trailing newline by default, as
the expected use case of 'w' is for sysfs and procfs. In exchange, honor
C-style backslash escapes so that if the newline is really needed, the
user can add it.
Break out the write logic into a separate function and simply use it as
a callback to glob_item.
This allows users to consolidate writes to sysfs with multiple similar
pathnames, e.g.
w /sys/class/block/sd[a-z]/queue/read_ahead_kb - - - - 1024
also a number of minor fixups and bug fixes: spelling, oom errors
that didn't print errors, not properly forwarding error codes,
few more consistency issues, et cetera
glibc/glib both use "out of memory" consistantly so maybe we should
consider that instead of this.
Eliminates one string out of a number of binaries. Also fixes extra newline
in udev/scsi_id
https://bugs.freedesktop.org/show_bug.cgi?id=38686
I don't think the usecase case in that bug makes much sense, but all the
other tools do honour /lib in the search path so we probably should do
that here, too.
Allow passing of basename only, instead of the absolute path; letting
systemd-tmpfiles perform a path lookup for the proper fragment path in
the config directories.
This allows distributions to call: systemd-tmpfiles <program.conf> on
upgrade of a package, with respecting the possibly overriden (or even
masked) tmpfile.
Tom Gundersen