Commit Graph

24230 Commits

Author SHA1 Message Date
Yu Watanabe
9e2bbf9915 network: make Link and NetDev always have the valid poiter to Manager
c4397d94c3 introduces
link_detach_from_manager() and netdev_detach_from_manager(), and they
set Link::manager or NetDev::manager NULL.
But, at the time e.g. link is removed, hence link_drop() is called,
there may be still some asynchronous netlink call is waiting, and
their callbacks hit assertion.

This make {link,netdev}_detach_from_manager() just drop all references
from manager, but keep the pointer to manager.

Fixes #11411.
2019-01-15 14:48:53 +01:00
Zbigniew Jędrzejewski-Szmek
a6ca3c1921 udev: do logging before setting variables to NULL
gcc-9 diagnoses this as an error.
Reported by Jeff Law.
2019-01-15 13:58:45 +01:00
Topi Miettinen
a1e92eee3e Remove 'inline' attributes from static functions in .c files (#11426)
Let the compiler perform inlining (see #11397).
2019-01-15 08:12:28 +01:00
Yu Watanabe
0eba88dc9f udevadm: fix segfault
Fixes #11416.
2019-01-14 06:34:19 +09:00
Yu Watanabe
8d6c4b310d udev: use ENODATA when 'No entry found from hwdb'
Before:
IMPORT builtin 'hwdb' fails: No such file or directory

After:
IMPORT builtin 'hwdb' fails: No data available

Previous log is confusing and may be understood as hwdb file not exist.
2019-01-13 18:04:10 +09:00
Yu Watanabe
a119328608 udev: use SYNTHETIC_ERRNO() macro in log_device_*_errno() 2019-01-13 17:50:47 +09:00
Topi Miettinen
7ae3561a5a Delete duplicate lines
Found by inspecting results of running this small program:

int main(int argc, const char **argv) {
	for (int i = 1; i < argc; i++) {
		FILE *f;
		char line[1024], prev[1024], *r;
		int lineno;

		prev[0] = '\0';
		lineno = 1;
		f = fopen(argv[i], "r");
		if (!f)
			exit(1);
		do {
			r = fgets(line, sizeof(line), f);
			if (!r)
				break;
			if (strcmp(line, prev) == 0)
				printf("%s:%d: error: dup %s", argv[i], lineno, line);
			lineno++;
			strcpy(prev, line);
		} while (!feof(f));
		fclose(f);
	}
}
2019-01-12 16:02:26 +01:00
Yu Watanabe
e0b7a5d151 udevd: refuse devices which do not have ACTION property 2019-01-12 09:32:20 +09:00
Yu Watanabe
33ad742a84 udevd: drop unnecessary brackets 2019-01-12 09:32:20 +09:00
Yu Watanabe
c0ff3d6cbc udevd: make worker also log ACTION property 2019-01-12 09:32:20 +09:00
Yu Watanabe
25d4f5b071 udevd: reject devices which do not have SEQNUM 2019-01-12 09:32:20 +09:00
Yu Watanabe
956833b417 udevd: provide worker_hash_ops and drop manager_workers_free() 2019-01-12 09:32:20 +09:00
Yu Watanabe
d40534643b udevd: use structured initializer at one more place 2019-01-12 09:32:20 +09:00
Yu Watanabe
1f3f6bd007 udevd: use worker_free() on failure in worker_new()
Otherwise, worker_monitor may not unrefed correctly.
2019-01-12 09:32:20 +09:00
Yu Watanabe
ee0b9e721a sd-device-monitor: fix ordering of setting buffer size
By b1c097af8d (#10239), the receive buffer
size for uevents was set by SO_RCVBUF at first, and fallback to
use SO_RCVBUFFORCE. So, as SO_RCVBUF limits to the buffer size
net.core.rmem_max, which is usually much smaller than 128MB udevd requests,
uevents buffer size was not sufficient.

This fixes the ordering of the request: SO_RCVBUFFORCE first, and
fallback to SO_RCVBUF. Then, udevd's uevent buffer size can be set to
128MB.

This also revert 903893237a.

Fixes #11314 and #10754.
2019-01-11 21:52:23 +01:00
Fabrice Fontaine
5269db454f lockfile-util.c: fix build without F_OFD_SETLK
systemd fails to build on kernel without F_OFD_SETLK since
9714c020fc

So put include missing_fcntl.h

Fixes:
 - http://autobuild.buildroot.org/results/699c078aa078240c6741da4dbd0871450ceeca92

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
2019-01-11 17:37:36 +01:00
Fabrice Fontaine
e2d94d0c16 missing_syscall.h: include errno.h
This include is needed for errno and ENOSYS

Fixes:
 - http://autobuild.buildroot.org/results/699c078aa078240c6741da4dbd0871450ceeca92

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
2019-01-11 17:37:36 +01:00
Fabrice Fontaine
6a9eb11b8a missing_if_link.h: add IFLA_BOND_MODE
systemd fails to build on kernel without IFLA_BOND_MODE (< 3.13) since
9714c020fc

So put back IFLA_BOND_MODE definition

Fixes:
 - http://autobuild.buildroot.org/results/699c078aa078240c6741da4dbd0871450ceeca92

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
2019-01-11 17:37:36 +01:00
Fabrice Fontaine
3ee57870d6 basic/tmpfile-util.c: fix build without O_TMPFILE
systemd fails to build on kernel without O_TMPFILE (< 3.11) since
dea72eda9c

To fix this error, include missing_fcntl.h

Fixes:
 - http://autobuild.buildroot.org/results/699c078aa078240c6741da4dbd0871450ceeca92

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
2019-01-11 17:37:36 +01:00
Fabrice Fontaine
a22692d718 capability: fix build without PR_CAP_AMBIENT
systemd fails to build on kernel without PR_CAP_AMBIENT (< 4.3) since
2a03bb3e65

To fix this error, include missing_prctl.h in all files using
PR_CAP_AMBIENT

Fixes:
 - http://autobuild.buildroot.org/results/699c078aa078240c6741da4dbd0871450ceeca92

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
2019-01-11 17:37:20 +01:00
Zbigniew Jędrzejewski-Szmek
0192cbdb2c Revert "nss: prevent PROTECT_ERRNO from squashing changes to *errnop"
This reverts commit b26c904113.

I don't see anythign wrong, but Ubuntu autopkgtest CI started failing fairly
consistently since this was merged. Let's see if reverting fixes things.
2019-01-10 21:23:14 +01:00
Franck Bui
c0f34168d4 Revert "logind: become the controlling terminal process before restoring VT"
This reverts commit ad96887a12.

Commit adb8688 alone should be enough to fix issue #9754.

Fixes #11269
2019-01-10 15:59:13 +01:00
Zbigniew Jędrzejewski-Szmek
c7e93c4d15 pam_systemd: reword message about not creating a session
The message is changed from
  Cannot create session: Already running in a session...
to
  Not creating session: Already running in a session...

This is more neutral and avoids suggesting a problem.

"Will not create session: ..." was suggested, but it sounds like the action
would have yet to be performed. I think Using present continuous is better.

Fixes #10822 (for good now, I hope).
2019-01-10 15:56:02 +01:00
Sam Morris
b26c904113 nss: prevent PROTECT_ERRNO from squashing changes to *errnop
glibc passes in &errno for errnop, which means PROTECT_ERRNO ends up
squashing our intentional changes to *errnop.

Fixes #11321.
2019-01-10 11:08:42 +01:00
Zbigniew Jędrzejewski-Szmek
a1b939dfc7 Merge pull request #11376 from yuwata/11365-v2
udev: initialize sockets before fork()
2019-01-10 09:08:57 +01:00