- 10 Feb, 2012 9 commits
-
-
Kinsbursky Stanislav authored
Otherwise test will sleep in kernel in case of queue is emptry on read or queue is full on write. Signed-off-by:
Stanislav Kinsbursky <skinsbursky@parallels.com> Signed-off-by:
Andrey Vagin <avagin@gmail.com>
-
Andrey Vagin authored
This patch adds new option -d. zdtm.sh -d ... does not do restore and instead of that it wakes up test processes after dump and check results of the test. Signed-off-by:
Pavel Emelyanov <xemul@parallels.com>
-
Kinsbursky Stanislav authored
2 processes - 1 semaphore per each. Both processes checks it's sem migration by id. Child also checks parent semaphore migration my key. Signed-off-by:
-
Kinsbursky Stanislav authored
v2: msgque test removed from zdtm.sh (feature is not supported yet) 1) Added namesapce isolation 2) Added non-empty queue state before migration 3) Added one more backward message send (child send message back after receiving one from parent). Signed-off-by:
-
Kinsbursky Stanislav authored
Signed-off-by:
-
Kinsbursky Stanislav authored
v2: fixed returning zero (errno) value in case of sys_write() didn't fail. Signed-off-by:
-
Kinsbursky Stanislav authored
v2: added missed prints on error pathes Signed-off-by:
-
Kinsbursky Stanislav authored
1) pass cmd line args to function to clone to - required for parametrized tests. 2) handle function result - required for multi-process tests. Signed-off-by:
-
Kinsbursky Stanislav authored
Signed-off-by:
-
- 09 Feb, 2012 5 commits
-
-
Kinsbursky Stanislav authored
Signed-off-by:
Cyrill Gorcunov <gorcunov@openvz.org>
-
Kinsbursky Stanislav authored
Signed-off-by:
-
Kinsbursky Stanislav authored
Signed-off-by:
-
Kinsbursky Stanislav authored
Signed-off-by:
-
Kinsbursky Stanislav authored
Signed-off-by:
-
- 08 Feb, 2012 8 commits
-
-
Kinsbursky Stanislav authored
v2: wrappers names become less obfuscating This patch: 1) Updates function cr_fdset_open() to be suitable for handling fdset creation for dump and show stages. 2) Replaces cr_fdset_open() by new wrapper function cr_fdset_dump(). 3) Replaces prep_cr_fdset_for_restore() by new wrapper function cr_fdset_show(). Signed-off-by:
-
Cyrill Gorcunov authored
No func change Signed-off-by:
-
Cyrill Gorcunov authored
If an error happens during late restore stage, we better exit with error than dive into endless loop. The loop was useful at early prototype times, but not now when we use zdtm test suite. For example, force injected error now reports | Restoring EXE (/home/crtools/test/zdtm/live/static/selfexe00) | 764 | 11308 | Error (cr-restore.c:1280): 11308 exited, status=255 | Error (cr-restore.c:1412): Someone can't be restored which is a way better than silent loop. Reported-by:
-
Kinsbursky Stanislav authored
Signed-off-by:
-
Kinsbursky Stanislav authored
This patch removes collect stage and dumps tunables object right after collect. Signed-off-by:
-
Cyrill Gorcunov authored
In case of critical error is happened during checkpoint procedure, the program should exit immediately. Signed-off-by:
-
Pavel Emelyanov authored
Unix diag and stat report dev_t-s in different formats. Cast one to another when comparing in unix dumping. Signed-off-by:
-
Pavel Emelyanov authored
Signed-off-by:
-
- 07 Feb, 2012 18 commits
-
-
Kir Kolyshkin authored
Way easier this way. Signed-off-by:
Kir Kolyshkin <kir@openvz.org> Signed-off-by:
-
Cyrill Gorcunov authored
It looks ugly but test all cases at once. Signed-off-by: -
Cyrill Gorcunov authored
Signed-off-by:
-
Cyrill Gorcunov authored
Signed-off-by:
-
Cyrill Gorcunov authored
This patch adds ability to checkpoint/restore /proc/pid/exe symlink, so if a process we've just checkpointed has been say /path/to/exe, then at restore time we bring this path back. There some restiction from kernel side: if existing /proc/pid/exe already mapped more than once, the kernel will refuse to change the symlink, so we need to restore it lately when mmaps of crtools itself already unmapped (ie via late call in restorer.c). Signed-off-by:
-
Cyrill Gorcunov authored
Signed-off-by:
-
Cyrill Gorcunov authored
It's temporary action until we move git repo to own server. Signed-off-by: -
https://github.com/avagin/crtoolsCyrill Gorcunov authored
* 'zdtm' of https://github.com/avagin/crtools: zdtm: fix silly goto loop in static/shm
-
Pavel Emelyanov authored
There's only one place with this, but the helper makes the code look almost perfectly. Signed-off-by:
-
Pavel Emelyanov authored
Signed-off-by:
-
Pavel Emelyanov authored
This one reads a name from image, does preparations for bind and calls bind. Signed-off-by:
-
Pavel Emelyanov authored
They fully coincide now, just need to distinguish them by socket type. Thus, add the type member on the unix_sk_listen and merge two hashes. Signed-off-by:
-
Pavel Emelyanov authored
Signed-off-by:
-
Pavel Emelyanov authored
Structure used is the same, logic of connection is the same, and the infrastructure is ready for this. Signed-off-by:
-
Pavel Emelyanov authored
Required for simpler future patching. Signed-off-by:
-
Pavel Emelyanov authored
The unix_dgram_peer coincides with the unix_conn_job. Signed-off-by:
-
Pavel Emelyanov authored
The unix_dgram_bound fully coincides with the unix_sk_listen. Signed-off-by:
-
Pavel Emelyanov authored
By the time we try to resolve datagram socket connection name the other end may not yet be on the hash. Move the address resolution into the conn job. Signed-off-by:
-
