| Age | Commit message (Collapse) | Author |
|---|
|
part of the original ISC license that we use in OpenBSD. Done for
files were Henning is the original author.
OK henning@ deraadt@
|
|
don't kill the VM if we are presented with an exit type that is not
something we directly need to handle. Instead, simply re-enter the guest,
possibly injecting an interrupt along the way. Still exit the process
if presented with a triple fault, until we implement reboot support.
ok stefan@
|
|
ok stefan@
|
|
OK mlarkin@
|
|
|
|
On shutdown, there's no need to use kill(2) to kill the child
processes. Just closing the IPC sockets will make the children receive
an EOF, break out from the event loop and then exit.
Tha advantages of this "pipe teardown" are:
* simpler code;
* no need to pledge "proc" in the parent process;
* removal of a (hard to trigger) PID reuse race condition.
ok benno@ claudio@
|
|
On shutdown, there's no need to use kill(2) to kill the child
processes. Just closing the IPC sockets will make the children receive
an EOF, break out from the event loop and then exit.
Tha advantages of this "pipe teardown" are:
* simpler code;
* no need to pledge "proc" in the parent process;
* removal of a (hard to trigger) PID reuse race condition.
ok benno@ claudio@
|
|
On shutdown, there's no need to use kill(2) to kill the child
processes. Just closing the IPC sockets will make the children receive
an EOF, break out from the event loop and then exit.
Tha advantages of this "pipe teardown" are:
* simpler code;
* no need to pledge "proc" in the parent process;
* removal of a (hard to trigger) PID reuse race condition.
ok benno@ claudio@
|
|
Open all fds first before sending messages to the vmm process.
This prevents some intermediate state from being generated that
will confuse vmm if it gets further messages.
Also make sure that vmctl gets the right error reason later by saving
errno upon a failure.
ok reyk@ mlarkin@
|
|
Pointed out by benno@
|
|
move daemon-local functions into new logmsg.c, and reduce
the (mostly whitespace) differences so that log.c's can be diffed easily.
ok claudio@
|
|
|
|
Refactor such that we can share some of the code for beacon printing.
ok sthen@
|
|
Remove these leftovers from the pre-fork+exec era.
Spotted by and ok rzalamena@ ok claudio@
|
|
This registers a handler with libevent that is triggered on incoming
data on the com port.
ok mlarkin@
|
|
This registers a handler with libevent that is called on incoming packets
for the guest. If they cannot be handled immediately (because the virtq is
full), make sure they are handled on VCPU exits.
ok mlarkin@
|
|
* Standardize the logging format of sent and received messages:
"msg-[in|out]: message type: lsr-id A.B.C.D [additional info]";
* Log sent label messages as well, not only the received ones;
* Move the logging of sent notification messages from
send_notification_nbr() to send_notification_full(), this way notification
triggered by the lde process are logged as well;
* Minor clean-up.
|
|
If a neighbor was a nexthop for a given set of prefixes but it's not
anymore, then we should uninstall the associated label mappings (if any)
from the kernel. The same applies for the other way round (neighbor
wasn't a nexthop for a given set of prefixes but now is).
This issue is only evident when we have multiple links between a pair
of LSRs. Generally, when a link is shut down, the whole LDP is torn down
and all label mappings uninstalled automatically.
ok claudio@
|
|
On shutdown, there's no need to use kill(2) to kill the child
processes. Just closing the IPC sockets will make the children receive
an EOF, break out from the event loop and then exit.
Tha advantages of this "pipe teardown" are:
* simpler code;
* no need to pledge "proc" in the parent process;
* removal of a (hard to trigger) PID reuse race condition.
ok claudio@
|
|
|
|
style(9) says:
"Function prototypes for private functions (i.e., functions not used
elsewhere) go at the top of the first source module. In userland,
functions local to one source module should be declared 'static'".
The benefits of doing so include:
* clean up of the eigrpd global namespace;
* improved readability;
* more hints to the compiler/linker to generate more efficient code.
Additional changes:
* Declare all extern variables in header files;
* Clean up the indentation of all function prototypes and global
variables.
ok claudio@ benno@
|
|
Also, there's no need to zero initialize global and static variables,
that's done automatically by the compiler.
ok claudio@ benno@
|
|
This reduces some code duplication and makes it easier to see the
difference between the IPv4 and IPv6 raw sockets API.
tweak + ok claudio@, ok benno@
|
|
During the initialization of the eigrpe process, use inet_pton() to
store the EIGRP multicast addresses in two global variables:
* global.mcast_addr_v4 (in_addr);
* global.mcast_addr_v6 (in6_addr).
This way we don't need to create temporary in_addr/in6_addr variables
everytime we need to use these multicast addresses for something.
"I like this" claudio@ ok benno@
|
|
This is for consistency with the other routing daemons.
ok claudio@ "earlier is better" benno@
|
|
ok claudio@ benno@
|
|
* Add missing close() calls to actually close the pipes, calling just
msgbuf_clean() is not enough;
* Bring back some NOTREACHED lint comments. style(9) says they can be
removed but in some cases they are useful to humans too;
* Add __dead to the shutdown functions;
* Some other minor changes to make eigrpd(8) more similar to the other
routing daemons.
|
|
That allows the main thread to handle the termination
of a VM on reboot/shutdown or normal exit. The way it works
is that VCPUs that terminate for some reason communicate this
to a main thread that takes care of shutdown.
ok mlarkin@
|
|
move daemon-local functions into new logmsg.c, and reduce
the (mostly whitespace) differences so that log.c's can be diffed easily.
ok claudio@
|
|
snmp subsystem until the configuration is done.
OK benno@ claudio@
|
|
On shutdown, there's no need to use kill(2) to kill the child
processes. Just closing the IPC sockets will make the children receive
an EOF, break out from the event loop and then exit.
Tha advantages of this "pipe teardown" are:
* simpler code;
* no need to pledge "proc" in the parent process;
* removal of a (hard to trigger) PID reuse race condition.
"reads good" claudio@
|
|
Pulled from ospfd. Original author: claudio@
ok rzalamena@ benno@ claudio@
|
|
parent and hides the real cause of the termination.
Pulled from ospfd. Original author: claudio@
ok rzalamena@ benno@ claudio@
|
|
relayd_config". This way we can send all the relevant global
configuration to the children, not just the flags and the opts.
With input from and
OK claudio@ benno@
|
|
OK claudio@
|
|
move daemon-local functions into new logmsg.c, and reduce
the (mostly whitespace) differences so that log.c's can be diffed easily.
ok claudio@, feedback from henning@, deraadt@, reyk@
|
|
move daemon-local functions into new logmsg.c, and reduce
the (mostly whitespace) differences so that log.c's can be diffed easily.
ok claudio@, feedback from henning@, deraadt@, reyk@
|
|
move daemon-local functions into new logmsg.c, and reduce
the (mostly whitespace) differences so that log.c's can be diffed easily.
ok claudio@, feedback from henning@, deraadt@, reyk@
|
|
move daemon-local functions into new logmsg.c, and reduce
the (mostly whitespace) differences so that log.c's can be diffed easily.
ok claudio@, feedback from henning@, deraadt@, reyk@
|
|
move daemon-local functions into new logmsg.c, and reduce
the (mostly whitespace) differences so that log.c's can be diffed easily.
ok claudio@, feedback from henning@, deraadt@, reyk@
|
|
let the caller handle EAGAIN.
ok gilles@
|
|
let the caller handle EAGAIN.
ok reyk@ gilles@
|
|
no functional change at this point.
|
|
ok beck
|
|
|
|
OK benno@ rzalamena@
|
|
or killing child processes. - Based on rzalamena@'s diff for httpd.
OK deraadt@ rzalamena@
|
|
receiving side of proc_connect(). Move some code from main into
proc_init(), the function is now called by parent and children, not
just the parent and it is less copy + paste for other daemons.
OK florian@
|
|
ok jsing@
|
|
|
