Donald Sharp [Tue, 16 May 2017 00:55:33 +0000 (20:55 -0400)]
lib: Improve error message for the developer
I keep getting people asking me about what to do
when this error is generated when they are programming
new cli. Maybe this is a bit better bread-crumb?
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
Donald Sharp [Mon, 15 May 2017 23:31:27 +0000 (19:31 -0400)]
*: Consolidate vrf_hooks into vrf_init
We only needed to add/change the vrf callbacks when we initialize
the vrf subsystem. As such it is not necessary to handle the callbacks
in any other way than through the init function.
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
Donald Sharp [Mon, 15 May 2017 23:11:20 +0000 (19:11 -0400)]
*: Remove some unneeded functions
vrf_iflist_create -> By the time this is called in enable, the vrf's iflist
is already created. Additionally this code should be a properly of the vrf
to init/destroy not someone else.
vrf_iflist_terminate -> This function should be a property of vrf deletion
and does not need to be exposed.
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
Donald Sharp [Mon, 15 May 2017 23:02:34 +0000 (19:02 -0400)]
lib, zebra: Cleanup vrf api some
Segregate the vrf enable/disable functionality from other vrf
code. This is to ensure that people are not actually using
the functions when they should not be. Also document the
why of it properly in the new vrf_int.h header.
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
David Lamparter [Sat, 19 Nov 2016 10:57:08 +0000 (11:57 +0100)]
lib: cli: autocomplete variables
Shows known values in the appropriate naming domain when the user hits
<?> or <Tab>. This patch only works in the telnet CLI, the next patch
adds vtysh support.
Fills token->varname based on context. WORD tokens use the WORD - if it
isn't actually "WORD". Other than that, a preceding constant token is
used as name.
Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
David Lamparter [Fri, 18 Nov 2016 11:06:57 +0000 (12:06 +0100)]
lib: parser: add named variables in CLI
struct cmd_token now has a "varname" field which is derived from the
DEFUN's string definition. It can be manually specified with "$name"
after some token, e.g. "foo WORD$var". A later commit adds code to
automatically fill the value if nothing is specified.
Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
* Correct the metric calculation as well as the default metrics;
* Do not show invalid routes in the "show ip eigrp topology".
* Add support to VRFs;
* When downloading a neighbor remove the related routes;
* Fix bugs in the parser of packages they were creating
Invalid default routes;
* Add and remove routes in the zebra;
* Add command "on router eigrp AS";
* Make "delay" and "bandwitch" commands work as well as
Display them in running config;
* Add "no" version of several commands;
* Fix a serious momory leaks;
* Fix segfault when there is no 'successor' route to a
Given prefix;
* Other minor corrections;
Signed-off-by: Renato Westphal <renatowestphal@gmail.com> Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
Daniel Walton [Wed, 10 May 2017 13:42:00 +0000 (09:42 -0400)]
zebra: lsp_install() failed due to ZEBRA_FLAG_SELECTED check
ZEBRA_FLAG_SELECTED hasn't been set yet by the time lsp_install is
called. The call path is:
rib_process -> rib_process_add_fib -> zebra_mpls_lsp_install -> lsp_install
but ZEBRA_FLAG_SELECTED is set in rib_process after it calls
rib_process_add_fib. I can't think of anything that it would hurt to
install the LSP regardless of whether ZEBRA_FLAG_SELECTED is set later.
I also cleaned up some UI (json and display the pretty label names
instead of their numeric values).
Signed-off-by: Daniel Walton <dwalton@cumulusnetworks.com>
Quentin Young [Fri, 5 May 2017 21:22:25 +0000 (21:22 +0000)]
*: update thread_add_* calls
Pass pointer to pointer instead of assigning by return value. See
previous commit message.
To ensure that the behavior stays functionally correct, any assignments
with the result of a thread_add* function have been transformed to set
the pointer to null before passing it. These can be removed wherever the
pointer is known to already be null.
Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
Quentin Young [Fri, 5 May 2017 17:30:21 +0000 (17:30 +0000)]
lib: safely check & set thread pointers
When scheduling a thread, the scheduling function returns a pointer to
the struct thread that was placed on one of the scheduling queues in the
associated thread master. This pointer is used to check whether or not
the thread is scheduled, and is passed to thread_cancel() should the
daemon need to cancel that particular task.
The thread_fetch() function is called to retrieve the next thread to
execute. However, when it returns, the aforementioned pointer is not
updated. As a result, in order for the above use cases to work, every
thread handler function must set the associated pointer to NULL. This is
bug prone, and moreover, not thread safe.
This patch changes the thread scheduling functions to return void. If
the caller needs a reference to the scheduled thread, it must pass in a
pointer to store the pointer to the thread struct in. Subsequent calls
to thread_cancel(), thread_cancel_event() or thread_fetch() will result
in that pointer being nulled before return. These operations occur
within the thread_master critical sections.
Overall this should avoid bugs introduced by thread handler funcs
forgetting to null the associated pointer, double-scheduling caused by
overwriting pointers to currently scheduled threads without performing a
nullity check, and the introduction of true kernel threads causing race
conditions within the userspace threading world.
Also removes the return value for thread_execute since it always returns
null...
Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
Quentin Young [Mon, 24 Apr 2017 22:33:25 +0000 (22:33 +0000)]
*: remove THREAD_ON macros, add nullity check
The way thread.c is written, a caller who wishes to be able to cancel a
thread or avoid scheduling it twice must keep a reference to the thread.
Typically this is done with a long lived pointer whose value is checked
for null in order to know if the thread is currently scheduled. The
check-and-schedule idiom is so common that several wrapper macros in
thread.h existed solely to provide it.
This patch removes those macros and adds a new parameter to all
thread_add_* functions which is a pointer to the struct thread * to
store the result of a scheduling call. If the value passed is non-null,
the thread will only be scheduled if the value is null. This helps with
consistency.
A Coccinelle spatch has been used to transform code of the form:
if (t == NULL)
t = thread_add_* (...)
to the form
thread_add_* (..., &t)
The THREAD_ON macros have also been transformed to the underlying
thread.c calls.
Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
Donald Sharp [Tue, 9 May 2017 20:18:04 +0000 (16:18 -0400)]
*: Remove ability to install frr_sudoers
If the user were to uncomment last line
and allow VTYSH_SHOW to be used as a non-root
account, this would allow arbitrary command completion
inside of vtysh via multiple -c ... -c .... lines
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
projects / mirror / frr.git / log