No actual code changes, just splitting out of some dev_t handling
related calls from stat-util.[ch], they are quite a number already, and
deserve their own module now I think.
Also, try to settle on the name "devnum" as the name for the concept,
instead of "devno" or "dev" or "devid". "devnum" is the name exported in
udev APIs, hence probably best to stick to that. (this just renames a
few symbols to "devum", local variables are left untouched, to make the
patch not too invasive)
No actual code changes.
This is inspired by this:
https://github.com/systemd/zram-generator/blob/main/src/generator.rs#L29
Given it's easy for PID 1 to pass this to generators, I thin we should.
All generators not written by us likely want to know these things so
that the can adjust to the execution environment.
let's better be safe and use heap allocation for paths which might be
unbounded.
In particular as previously we copied the stack memory to heap anyway,
via a noop path_make_absolute_cwd() call.
Several keyboard devices are erroneously tagged with ID_INPUT_JOYSTICK
because of random buttons they set. For example, the LiteOn Lenovo
Calliope USB Keyboard sets BTN_TRIGGER, BTN_TOP2, BTN_PINKIE and
BTN_BASE, see libinput issue 745 for details.
ID_INPUT_JOYSTICK triggers the uaccess rules, making those keyboards
easily accessible. That's not a problem in the LiteOn example since that
event node doesn't contain the normal keys and eavesdropping on volume
keys is probably not very interesting.
Improve the joystick detection by adding heuristics similar to what
libinput 1.20 uses: check for some specific set of keys that are common
on keyboards but very unlikely on joysticks. If enough of those are
present (or the device has less than 2 axes or joysticks), don't tag it
as joystick.
libinput also checks for > 10 keyboard keys, but this is not done here
to be more conservative.
It turns out that in fa3cd7394c back
in 2013 I got the test reversed: assert_se(strncmp()) should be
assert_se(strncmp==0). So the tests that were using "*" were not entirely
useful ;) The function was refactored a bunch of times since then, and it
seems nobody noticed.
So let's replace this fragile construct by a simple fnmatch, which also
has the advantage that the glob can be inserted in arbitrary places.
Following up for d0aba07f1a: we should have at
least basic tests for all interfaces, even the deprecated ones, so that we
catch obvious errors. This sorts the specifiers the same way that they are
declared in the unit-printf.c, and adds tests for all the specifiers. We
even were setting 'shell', but not using it in a test.
Also, we shouldn't initialize variables in tests. This catches the error fixed
in previous commit.
Simplify generated sysfs paths, since we might get data passed that
includes extra // in the middle.
Also, let's not assume /sys/ prefix without verification.
This doesn't change anything for real uses, because we'd initialize the
variable to NULL for _cleanup_ anyway, but let's follow our general pattern
of always setting the output on "success". (Even if that success is an empty
answer here.)
So here's something we should always keep in mind:
systemd-udevd actually does *two* things with BSD file locks on block
devices:
1. While it probes a device it takes a LOCK_SH lock. Thus everyone else
taking a LOCK_EX lock will temporarily block udev from probing
devices, which is good when making changes to it.
2. Whenever a device is closed after write (detected via inotify), udevd
will issue BLKRRPART (requesting the kernel to reread the partition
table). It does this while holding a LOCK_EX lock on the block
device. Thus anyone else taking LOCK_SH or LOCK_EX will temporarily
block udevd from issuing that ioctl. And that's quite relevant, since
the kernel will temporarily flush out all partitions while re-reading
the partition table and then create them anew. Thus it is smart to
take LOCK_SH when dissecting a block device to ensure that no
BLKRRPART is issued in the background, until we mounted the devices.
This revisits the mess around waiting for partition block devices in
the image dissection code. It implements a nice little trick:
Instead of waiting for the kernel to probe the partition table for us
and generate the block devices from it, we'll just do that ourselves.
How can we do it? Via the BLKPG_ADD_PARTITION ioctl, that the kernel has
supported for a while. This ioctl allows creating partition block
devices off "whole" block devices from userspace, without the partitions
necessarily being present in the partition table at all.
So, whenever we want a partition to be there, we'll just issue
BLKPG_ADD_PARTITION. This can either work, in which case we know the
partition is there, and can use it. Yay. Or it can fail with EBUSY,
which the kernel returns if a partition by the selected partition index
already exists (or if an existing partition overlaps with the new one).
But if that's the case, then that's also OK, because the partition will
already exist.
So, regardless if we win or the kernel wins, for us the outcome is the
same: the partition block device will exist after invoking the ioctl.
Yay.
Net effect: we are not dependent on asynchronous uevent messages to wait
for the devices. Instead we synchronously get what we need. This makes
us independent of the (apparently less than reliable) netlink transport,
and should almost always be quicker.
Hopefully addresses #17469 even on older kernels.
Fixes: #17469
Lennart Poettering
Zbigniew Jędrzejewski-Szmek
Yu Watanabe
Peter Hutterer
amarjargal
Sonali Srivastava
Daan De Meyer
Ben Efros