- 02 Mar, 2012 29 commits
-
-
Pavel Emelyanov authored
Just prepare the code for smoother further bootstrap areas allocation. Signed-off-by:
Pavel Emelyanov <xemul@parallels.com> Signed-off-by:
Cyrill Gorcunov <gorcunov@openvz.org>
-
Pavel Emelyanov authored
Signed-off-by:
-
Pavel Emelyanov authored
There are three bugs in this code. 1. self vmas list is released before get-hint is called 2. get-hint code wrongly detects the hole (just bugs, no details) 3. exec hint is mapped without MAP_FIXED, but should Signed-off-by:
-
Pavel Emelyanov authored
It's broken. Stas will fix it soon. Signed-off-by:
-
Pavel Emelyanov authored
Reuse the TCP socket restore, just add connect() and sanity checks for protocol. This is OK, since UDP connect doesn't go to network for connection and (unlike unix sockets) doesn't require peer to be "online". It just puts the peer's creds on socket and returns. Signed-off-by:
-
Pavel Emelyanov authored
These are required for inet sockets, but were not added since listen sockets do not have them. Signed-off-by:
-
Pavel Emelyanov authored
Showers should use pr_msg to avoid loglevel issues. Signed-off-by:
-
Pavel Emelyanov authored
Just a small fixlet in can_dump_inet_sk. Signed-off-by:
-
Pavel Emelyanov authored
Partially reuse the existing TCP code. Signed-off-by:
-
Pavel Emelyanov authored
Create two of them, bind both and connect one end to the other. Then check that send and sendto work and recv(from) receive proper message from proper address. Queues are expected to be dropped while test according to protocol constraints, thus all sends happen after restore. Signed-off-by:
-
Pavel Emelyanov authored
Otherwise logs look poor on failures. Signed-off-by:
-
Pavel Emelyanov authored
Broken by 7aa8e4b6 -- log was not moved to higher fd values. Signed-off-by:
-
Cyrill Gorcunov authored
Signed-off-by: -
Kir Kolyshkin authored
It was never needed in the first place, and not used after commit e315bea3. Reported-by:
Stanislav Kinsburgsky <skinsbursky@openvz.org> Signed-off-by:
Kir Kolyshkin <kir@openvz.org> Signed-off-by:
-
Pavel Emelyanov authored
* /proc/<pid>/map_files * sock diag * ns_last_pid sysctl * SO_PEEK_OFF sockoptions Signed-off-by:
-
Pavel Emelyanov authored
Signed-off-by:
-
Kir Kolyshkin authored
In addition, add target rm if script has failed. Signed-off-by:
-
Kir Kolyshkin authored
No need to hardcode it. Signed-off-by:
-
Kir Kolyshkin authored
Makes invocation way easier. Signed-off-by:
-
Kir Kolyshkin authored
1. We need to regenerate *-blob.h files if generating script changes 2. Do not use hardcoded script name Signed-off-by:
-
Kir Kolyshkin authored
There is no need to make .d files for restorer-blob.h Signed-off-by:
-
Kir Kolyshkin authored
These changes are mostly for better readability and style. The only actual difference (in the output produced) is the last item. (1) Use UPPERCASE variable names (2) Simplify names (get rid of name_ prefix for everything) (3) Rename variables: - name_ifndef to INC_GUARD, - name_prefix_offset to PREFIX, - name_bin to BINARY. (4) Replace sed with tr for simplicity (5) Replace grep with awk condition (6) Don't escape every $ and " inside AWK_CMD, just quote the full string (7) Use cat and 'here document' instead of many echoes (8) Simplify hexdump arguments, use single -e option (9) Insert our name into 'Generated by' comment. Signed-off-by:
-
Cyrill Gorcunov authored
Otherwise I'm getting errors like CR_FD_SK_QUEUES ---------------- Error (./include/util.h:102): Can't read img file: Bad file descriptor ---------------- Signed-off-by: -
Kinsbursky Stanislav authored
Signed-off-by:
-
Kinsbursky Stanislav authored
Signed-off-by:
-
Kinsbursky Stanislav authored
Signed-off-by:
-
Kinsbursky Stanislav authored
Signed-off-by:
-
Kinsbursky Stanislav authored
Signed-off-by:
-
Cyrill Gorcunov authored
Due to code sharing, especially in IPC area, the unbinding is done via helper macros and sysclt engine tuning (new CTL_SHOW action added). Signed-off-by:
-
- 01 Mar, 2012 11 commits
-
-
Cyrill Gorcunov authored
The messages are filtered by their type LOG_MSG - plain messages, they escape any (!) log level filtration and go to stdout LOG_ERROR - error messages LOG_WARN - warning messages LOG_INFO - informative messages LOG_DEBUG - debug messages By default the LOG_WARN log level is used, thus LOG_INFO and LOG_DEBUG messages will not appear in output stream. pr_panic helper was replaced with pr_err, pr_warning shorthanded to pr_warn and old printk if rather pr_msg now. Because we share messages between "show" and "dump" actions, before the "show" action proceed we need to tune up log level and set it to LOG_INFO. Also note that printing of VMA and siginfo now became LOG_INFO messages, it was not that correct to print them regardless the log level. Signed-off-by: -
Cyrill Gorcunov authored
Signed-off-by:
-
Cyrill Gorcunov authored
We have a few helper functions which all do setup logfd but in different program flow context. So rename them for grepability. Signed-off-by:
-
Pavel Emelyanov authored
When seizing a task and waiting for it to stop, we can receive an event from kernel, that the task didn't stop, but caught a signal. Don't treat this as an error, let the victim handle it and proceed. The transition/fork test should work OK now. Signed-off-by:
-
Pavel Emelyanov authored
Empty list on scan error means, that the very top task ran away from us. There's no point in trying to catch one. Signed-off-by:
-
Pavel Emelyanov authored
This can happen while dumping a pid-namespace (we can't do it now), thus put this check not to forget one in the future. Signed-off-by:
-
Pavel Emelyanov authored
Tasks can fork or die while we try to seize them. It's much more simpler and much more reliably to unseize what was seized and walk the tree again in case of some tree check failed. Yes, this makes it impossible to suspend a big tree which constantly forks :( but I'd prefer fixing more urgent issues before. Signed-off-by:
-
Pavel Emelyanov authored
Remove CR_TASK_XXX states, use the TASK_XXX ones (for image). This is required to unseize tasks properly in the next patches. Plus, make sure that pstree_list and the seized set coincide (i.e. handle error in collect_task). Signed-off-by:
-
Pavel Emelyanov authored
When we've seized all the tasks and threads found in /proc check for the /proc contents be the same. Do it one-by-one as we descend the tree. This is OK, since tasks cannot create kids for anyone but themselves or their parents (reparent will be handled later). Signed-off-by:
-
Pavel Emelyanov authored
It's in item already. Signed-off-by:
-
Pavel Emelyanov authored
While we try to seize task it can die and give its pid to somebody else. This can break pstree consistency. Check for parent being valid after task is seized. Signed-off-by:
-
