- 15 May, 2012 1 commit
-
-
Andrey Vagin authored
to be sure that all declarations consistent. Signed-off-by:
Andrey Vagin <avagin@openvz.org> Signed-off-by:
Pavel Emelyanov <xemul@parallels.com>
-
- 11 May, 2012 6 commits
-
-
Pavel Emelyanov authored
Signed-off-by: -
Pavel Emelyanov authored
What is stored there is the path to mountpoint. The root is the field previously named "parent root". Signed-off-by: -
Pavel Emelyanov authored
Signed-off-by: -
Pavel Emelyanov authored
Just put names right into desc template array. Nobody uses these individually. Signed-off-by: -
Pavel Emelyanov authored
Introduce a helper, that will declare one tmpl entry with one line. Signed-off-by: -
Pavel Emelyanov authored
Opts are now global and flags are just assigned on it. Signed-off-by:
-
- 12 May, 2012 5 commits
-
-
Andrey Vagin authored
A test process should be alive. Signed-off-by:
-
Andrey Vagin authored
It showed old pathes. Dump log : /root/crtools/test/dump/inotify00/15673/dump.log Restore log: /root/crtools/test/dump/inotify00/15673/restore.log Output file: /root/crtools/test/zdtm/live/static/utsname.out make: *** [test] Error 1 Signed-off-by:
-
Cyrill Gorcunov authored
In case if eventpoll or inotify just created but not watchee added we should not fail. Signed-off-by:
Cyrill Gorcunov <gorcunov@openvz.org> Signed-off-by:
-
Andrey Vagin authored
Signed-off-by:
-
Andrey Vagin authored
Signed-off-by:
-
- 11 May, 2012 3 commits
-
-
Cyrill Gorcunov authored
This allows us to detect nonlinear mappings. Signed-off-by:
-
Cyrill Gorcunov authored
While been playing with sockets dumping I found a case where we exit without descriptive error message. Fix it. Signed-off-by:
-
Andrey Vagin authored
I'm a paranoiac. Signed-off-by:
-
- 10 May, 2012 3 commits
-
-
Pavel Emelyanov authored
Only snd and rcv bufs and timeouts. Just as an example, others will be adder sequentially. Signed-off-by: -
Pavel Emelyanov authored
Those sitting on the SOL_SOCKET level are common to different socket families and will be handled in a generic code. Signed-off-by: -
Cyrill Gorcunov authored
Signed-off-by:
-
- 04 May, 2012 22 commits
-
-
Cyrill Gorcunov authored
Signed-off-by:
-
Cyrill Gorcunov authored
Since event polling depends on other files to be opened we split main files list into two parst -- event poll files and all other files, thus defer the creation of eventpoll files in prepare_fds(). Signed-off-by:
-
Andrey Vagin authored
They looks better in this state Signed-off-by:
-
Andrey Vagin authored
Now log messages may be splitted by pid Signed-off-by:
-
Andrey Vagin authored
If the option --log-pid is set, each process will have an own log file. Otherwise PID is added to each log message. A message can't be bigger than one page minus some bytes for pid. Signed-off-by:
-
Andrey Vagin authored
Signed-off-by:
-
Pavel Emelyanov authored
It will fall through all the other checks, but let's do it explicitly for better code understanding. Signed-off-by: -
Pavel Emelyanov authored
This saves some space on dynamic file_desc and makes file_desc_ops looks closer to fdinfo_ops used on dump. Signed-off-by: -
Pavel Emelyanov authored
No functional changes, just a code move. TODO: * need to write personal make_gen_id implementations for special files (device and pos is const for them effectively) * need to merge dump ops with restore ones (file_desc_ops) Signed-off-by: -
Pavel Emelyanov authored
Signed-off-by: -
Pavel Emelyanov authored
Now all the fds dump look very similar and we can reimplement it via indirect calls. Signed-off-by: -
Pavel Emelyanov authored
Reg files are those obtained by open() syscall on restore. Pipes should be checked to belong to pipefs (fifos are not supported yet). Signed-off-by: -
Pavel Emelyanov authored
Funny, but kernel reports some of them with []-s and some (well, ok, inotofy only) without... Signed-off-by: -
Cyrill Gorcunov authored
Signed-off-by:
-
Cyrill Gorcunov authored
v2: - open_mount is cleaned up - byte-stream hex conversion remains untouched since strtol is flipping numbers to LE manner Signed-off-by:
-
Cyrill Gorcunov authored
At moment only open_mnt_root is there, which is needed for inotify restore. Signed-off-by:
-
Cyrill Gorcunov authored
To restore inotify we need to know the mount point device numbers, so this helper parses /proc/pid/mountinfo file for that. Signed-off-by:
-
Cyrill Gorcunov authored
Need it for inotify restore. Signed-off-by:
-
Cyrill Gorcunov authored
Signed-off-by:
-
Cyrill Gorcunov authored
v2: - Move everything into eventpoll.[ch] - Use rst_file_params Signed-off-by:
-
Cyrill Gorcunov authored
Signed-off-by:
-
Cyrill Gorcunov authored
v2: - Pass initial counter value to eventfd call (can't pass flags here since they are obtained with fcntl and must be restored same way or restore will fail) - Use rst_file_params for flags and owner restore - Use eventfd.[ch] instead of eventfs.[ch] - Move show funcs to eventfd.c Signed-off-by:
-
