The return value of 'strtol()' is not checked properly
in _get_vmname_pid() @acrn_vm_ops.c and parse_opt()@acnrd.c,
the return type of 'strtol' is 'long int', but it is assigned
to a variable with type of 'int' and compared to "LONG_MAX"
and "LONG_MIN", which is always false.
This patch is to fix above error case.
Tracked-On: #4110
Signed-off-by: Yonghua Huang <yonghua.huang@intel.com>
Reviewed-by: Yan, Like <like.yan@intel.com>
Acked-by: Yan, Like <like.yan@intel.com>
In release environment, binary files must be stripped in
order to remove debugging code sections and symbol information
that aid attackers in the process of disassembly and reverse
engineering.
Use '-s' linking option to remove symbol table and relocation
information from release binaries.
Tracked-On: #3433
Signed-off-by: Tianhua Sun <tianhuax.s.sun@intel.com>
Reviewed-by: Yonghua Huang <yonghua.huang@intel.com>
When run 'acrnctl add', acrnctl dryruns launch script to get
vmname. Acrnctl assumes there are no more than 64 words in each
line of the script, and creates a buffer to hold 64 pointers to these
words. But we didn't check if the number of words exceed 64 limitation,
which can lead to buffer overflow. This commit fixed this potential
overflow problem.
And also refine some confusing error messages in 'acrnctl add'
Tracked-On: #2734
Acked-by: Yan Like <like.yan@intel.com>
Signed-off-by: Tao Yuhong <yuhong.tao@intel.com>
Reviewed-by: Yonghua Huang <yonghua.huang@intel.com>
The wait_for_stop() first check if all VM are stopped or suspended.
It first check if VMs are stopped, then check if they are suspend.
If the VMs change state from running to stoped, after check VMs
stopped fail, before start check suspend state. wait_for_stop() will
mistake resopnse 'All vms have entered S3 state successfully'
To fix that, wait_for_stop() must only update VMs's state once, and
see if they are stopped, or suspended.
Tracked-On: #2398
Signed-off-by: Tao Yuhong <yuhong.tao@intel.com>
When do IPC with mngr_send_msg(), and ack is required but failed to
get ack, mngr_send_msg() still return 0, that is not correct.
Tracked-On: #2398
Signed-off-by: Tao Yuhong <yuhong.tao@intel.com>
When acrnd_vm_timer_func() launch UOS, it need fork() and run, so
that do not block the main process.
Tracked-On: #2306
Signed-off-by: Tao Yuhong <yuhong.tao@intel.com>
Signed-off-by: Tao Yuhong <yuhong.tao@intel.com>
For a running or suspended VM, its state is updated in 2 steps.
It is first set to VM_CREATED, then set to VM_STARTED/VM_PAUSED. IF
one thread check a running/suspend VM, it may get wrong state VM_CREATED,
while another thread is updating the VMs state.
Tracked-On: #2306
Signed-off-by: Tao Yuhong <yuhong.tao@intel.com>
Signed-off-by: Tao Yuhong <yuhong.tao@intel.com>
Print more information at acrnd_add_work(), query_state()
try_do_works() and handle_acrnd_resume()
Tracked-On: #2306
Signed-off-by: Tao Yuhong <yuhong.tao@intel.com>
Acked-by: Yan, Like <like.yan@intel.com>
When issue happen, we could identify which thread is impacted.
This could help stability issue debugging.
Tracked-On: #2037
Signed-off-by: Yin Fengwei <fengwei.yin@intel.com>
Acked-by: Yu Wang <yu1.wang@intel.com>
The path of vm conf files changed from "/opt/acrn/conf" to "/usr/share/acrn/conf",
since "/opt" is not generally used on clearlinux.
Tracked-On: #2040
Signed-off-by: Yan, Like <like.yan@intel.com>
Acked-by: Yin Fengwei <fengwei.yin@intel.com>
The 'acrnctl reset' is implemented by a stop/start sequence. Reset failure was
observed since UOS is not able to shutdown within default 10s timeout.
This commit increased the default timeout to be 30s, and added an explicit error
message if 'acrnctl reset' due to UOS shutdown timeout.
Tracked-On: #2016
Signed-off-by: Tao Yuhong <yuhong.tao@intel.com>
Reviewed-by: Yan, Like <like.yan@intel.com>
There is a bug to run 'acrnctl add /[path]/script.sh', when the
launch script has an absolute path. Acrnctl will generate wrong path
for temp files and fail to add VM.
And message '/opt/acrn/conf: No such file or directory' always comes
out, until user once successfully run 'acrnctl add' cmd. That is reported
by _scan_added_vm(), because 'opt/acrn' is missing, only 'acrnctl add'
can create it, we should also check it in _scan_added_vm().
Tracked-On: #2013
Acked-by: Yan, Like <like.yan@intel.com>
Signed-off-by: Tao Yuhong <yuhong.tao@intel.com>
When UOS is sleeping, 'acrnctl list' command show it is 'paused', not
'suspended'. That is not correct. Because pausing an UOS means to block
its VCPUs immediately, otherwise suspending an UOS means to put it to
sleep.
Tracked-On: #1910
Acked-by: Yan, Like <like.yan@intel.com>
Signed-off-by: Tao Yuhong <yuhong.tao@intel.com>
I am currently using Yocto build and deploy UOS images, it would be
much more convenient to let the VMs to have the same names with when
they are being built. For instance, a typical image name in Yocto is
like: uos-image-weston, but it can not show the name completely with
'acrnctl list', since the length limit is 16.
Increasing the limit to 32 would fix this problem.
Tracked-On: #1837
Signed-off-by: Ming Liu <liu.ming50@gmail.com>
When SOS shutdown/reboot, systemd will send SIGTERM to acrnd. We should catch up
this signal and stop all vms gracefully.
BTW, this path also fix the following error when stop acrnd service by removing
ExecStop config. Systemd will send SIGTERM signal to process by default.
systemd[9378]: acrnd.service: Failed to execute command: No such file or directory
systemd[9378]: acrnd.service: Failed at step EXEC spawning /usr/bin/killall: No such file or directory
-- Subject: Process /usr/bin/killall could not be executed
-- Defined-By: systemd
-- Support: https://lists.freedesktop.org/mailman/listinfo/systemd-devel
--
-- The process /usr/bin/killall could not be executed and failed.
--
-- The error number returned by this process is 2.
Tracked-On: #1563
Signed-off-by: Kaige Fu <kaige.fu@intel.com>
Acked-by: Yan, Like <like.yan@intel.com>
UOS requires an wakeup reason when resumed from S3 suspend. If user
use "acrnctl resume [VMNAME]" without an reason, 0x0 will be used
as default reason, and this is not work. We can use CBC_WK_RSN_BTN
instead.
Tracked-On: #1550
Signed-off-by: Tao, Yuhong <yuhong.tao@intel.com>
Acked-by: Yan, Like <like.yan@intel.com>
Error msg "Invalid vmname from timer list file" will be print out when we read
a null line of file timer_list. It is a litte bit confusing. So, ignore null line
silently to avoid such confusing msg.
Tracked-On: #1517
Signed-off-by: Kaige Fu <kaige.fu@intel.com>
Acked-by: Yan, Like <like.yan@intel.com>
This patch is mainly for:
1. Check weakup reason first and only load timer list when reason is CBC_WK_RSN_RTC.
2. If failing to load timer list or reason != CBC_WK_RSN_RTC, activate all vms.
3. Add info about when we are about to activate vms reading from timer_list file.
Tracked-On: #1517
Signed-off-by: Kaige Fu <kaige.fu@intel.com>
Acked-by: Yan, Like <like.yan@intel.com>
When guest enter S5, its monitor socket will be closed. If we try to use it to
query the vm state, an error msg will be print out. Acctually, we should not take
it as an error.
So, this patch refine the log msg to show more details when guest trying to enter
S3/S5.
Tracked-On: #1506
Signed-off-by: Kaige Fu <kaige.fu@intel.com>
Acked-by: Yan, Like <like.yan@intel.com>
mngr_send_msg() could take mngr descriptor or real socket fd as argument based on
an assuption that real socket fd is less than 1024, which may be an issue.
This commit takes the real socket fd as descriptor, to avoid the conversion between them.
Tracked-On: #1477
Signed-off-by: Tao, Yuhong <yuhong.tao@intel.com>
Signed-off-by: Yan, Like <like.yan@intel.com>
Update the Makefiel to sync the compiler options with devicemode
and enable options to harden software.
Tracked-On: #1122
Signed-off-by: Yan, Like <like.yan@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
- Print warning if string truncated to avoid the warning generated
by -Wformat-truncation by GCC 7.0 and newer version;
- fixed strncpy size.
Tracked-On: #1122
Signed-off-by: Yan, Like <like.yan@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>
As a system service, acrnd will launch UOS, that must be done
after all required services and conditions are ready, such as
acrnprobe, weston, etc.
Tracked-On: #1278
Signed-off-by: Tao Yuhong <yuhong.tao@intel.com>
Remove the unnecessary "current" field when store/load timer list.
Tracked-On: #927
Signed-off-by: Yan, Like <like.yan@intel.com>
Acked-by: Yin Fengwei <fengwei.yin@intel.com>
This commit reworked the resume flow:
- generate RSN_RTC wakeup reason to resume_vm();
- clear timer_list in memory once it's saved to fs;
- wakeup the suspended VMs only if wakeup by ignition button.
Tracked-On: #927
Signed-off-by: Yan, Like <like.yan@intel.com>
Acked-by: Yin Fengwei <fengwei.yin@intel.com>
There is no need to wakeup SOS in advance since UOS wakeup time is
not required to be that exact.
Tracked-On: #927
Signed-off-by: Yan, Like <like.yan@intel.com>
Reviewed-by: Liu Yuan <yuan1.liu@intel.com>
Acked-by: Yin Fengwei <fengwei.yin@intel.com>
This commit eliminates several compiler warnings via:
- added 'const' quilifier to the 1st argument of send_msg();
- added 'const' quilifier to vm ops functions.
Signed-off-by: Yan, Like <like.yan@intel.com>
get_sos_wakeup_reason() runs into error branch without any error, so
no wakeup reason will be returend.
Acked-by: Yan Like <like.yan@intel.com>
Signed-off-by: Tao Yuhong <yuhong.tao@intel.com>
DM need to know wakeup reason when resume the VM, so
vm_resume(char *name) is updated to vm_resume(char *name, unsigned reason),
in acrn_vm_ops.c
Acked-by: Yan Like <like.yan@intel.com>
Signed-off-by: Tao Yuhong <yuhong.tao@intel.com>
This patch resolves acrnd does not notify vm stop state(suspend/shutdown) to
cbc_lifecycle after SOS enters suspend/shutdown, that causes cbc_lifecycle
cannot trigger SOS suspend/shutdown flow.
Signed-off-by: Yuan Liu <yuan1.liu@intel.com>
Reviewed-by: Like Yan <like.yan@intel.com>
The issue is "acrnctl list" command always returns error with "no vmname
provided" information.
The rootcause is that vmname is an invalid parameter, which does not contain
any vm information.
Signed-off-by: Yuan Liu <yuan1.liu@intel.com>
Reviewed-by: Yuhong Tao <yuhong.tao@intel.com>
Reviewed-by: Like Yan <like.yan@intel.com>
As first step to separate misc folder into github/intel/ioc-cbc-tools we
need let acrn create devel package. This package will let ioc-cbc-tools
use acrn manager interface API to control lifecycle.
The Clear Linux autospec script will *detect* /usr/include/acrn/*.h and
thus help us create the devel package. This devel pacakge will be in the
build_req for ioc-cbc-tools autospec.
Signed-off-by: Alek Du <alek.du@intel.com>
Reviewed-by: Yu Wang <yu1.wang@intel.com>
Reviewed-by: Like Yan <like.yan@intel.com>
Made the titles consistent across the tools (some had capital letters).
Fixed heading levels in acrnctl/acrnd doc (had two H1 headings).
Changed a text-based drawing to use graphviz.
Some general grammar tweaks as well.
Signed-off-by: David B. Kinder <david.b.kinder@intel.com>
acrn-manager message structure has a flexible member
payload[0], that risks to out-of-boundary memory access,
and usage of uninitialized variable.
And the req/ack message pairs has various types, which
extend mngr_msg. mngr_send_msg() requires programmer cast
the type of input messages to struct mngr_msg, that is
inconvenient.
We replace payload[0] with an union, which contains all
payload data. So that type cast for mngr_send_msg() is no
longer needed. And we can avoid potential out-of-boundary
memory accessing and using of uninitialized variable
Acked-by: Yan Like <like.yan@intel.com>
Signed-off-by: Tao Yuhong <yuhong.tao@intel.com>
