Bernd Petrovitsch [Thu, 29 Nov 2007 00:49:42 +0000 (00:49 +0000)]
Major Changes:
- renamed "struct local_hna_entry" to "struct ip_prefix_list" since
it is exactly that. Renamed the functions in src/local_hna_set.{c,h}
in the same way.
- each IPv4 and IPv6 function pairs in src/local_hna_set.{c,h} is
called from the same place and have the same signature. So I
condensed each of them.
- Since we have only 3 functions left in src/local_hna_set.{c,h}
and they are used for the configuration, the .h file is now part
of src/olsr_cfg.h and the ,c file of src/cfgparser/olsrd_conf.c.
- replaced "struct ipc_net" with "struct ip_prefix_list" since it
serves the same purpose as the "struct ip_prefix_list"
- replaced "struct ipc_host" with "struct ip_prefix_list" since it
serves the same purpose as the "struct ip_prefix_list" and is
just a special case. No need to duplicate code etc.
- removed "union hna_netmask" from src/olsr_types since we use the
prefix_len everywhere (and that is an olsr_u8_t in several other
struct's).
That implies changes and simplifications in the code handling
them (since the IPv4 is similar to IPv6).
- the config file parser now understands for IPv4 addresses also
"/$prefix".
- On the output side, the patch generates only "/$prefix" which
kills code since it is the same as the IPv6 handling.
- There are some netmask conversions left (mainly in the plugins)
but that be cleaned up afterwards.
- extracted ip{,4,6}{cmp,equal} and formatting functions from
net_olsr.{c,h} into src/ipcalc.{c,h} since net_olsr.h became
IMHO to much of a "put anything in there".
- renamed "sockaddr_to_string()" to "sockaddr4_to_string()"
since it is exactly that (unless I'm missing something).
Minor Changes:
- lib/httpinfo/src/admin_html.h contained just some variable
definitions so it is now integrated in the only user:
admin_interface.c
- olsrd_dot_draw.c got rid of two indicator variables if
a socket is valid or not. Since sockets may use "-1" as the
"not open", "invalid" value, there is no need for two more
ints.
- and the dot_draw plugin is somewhat smaller and easier to read.
- const'ified some functions
Bernd Petrovitsch [Thu, 29 Nov 2007 00:34:22 +0000 (00:34 +0000)]
Patch from Sven-Ola Tuecke <sven-ola@gmx.de> from
http://download-master.berlin.freifunk.net/sven-ola/nylon/packages/olsrd/files/138-olsrd-fixdotdraw.patch: fix the dot_draw plugin
Bernd Petrovitsch [Thu, 29 Nov 2007 00:26:16 +0000 (00:26 +0000)]
Patch from Sven-Ola Tuecke <sven-ola@gmx.de> from
http://download-master.berlin.freifunk.net/sven-ola/nylon/packages/olsrd/files/135-georss-map-for-nameservice.patch: fixes and extensions for the namesaervice plugin
Bernd Petrovitsch [Thu, 29 Nov 2007 00:24:00 +0000 (00:24 +0000)]
Patch from Sven-Ola Tuecke <sven-ola@gmx.de> from
http://download-master.berlin.freifunk.net/sven-ola/nylon/packages/olsrd/files/132-save-the-fish.patch:
really savet he fish and avoid to forward invlaid messages from older olsrd versions
Bernd Petrovitsch [Thu, 29 Nov 2007 00:18:53 +0000 (00:18 +0000)]
Patch from Sven-Ola Tuecke <sven-ola@gmx.de> from
http://download-master.berlin.freifunk.net/sven-ola/nylon/packages/olsrd/files/103-olsrd-rt-exportroute-cleanup.patch:
avoid a malloc()ed list of functions
Bernd Petrovitsch [Thu, 29 Nov 2007 00:10:17 +0000 (00:10 +0000)]
Bug Fixes (of stupid and forgotten bugs) from Sven-Ola Tuecke <sven-ola@gmx.de> from
http://download-master.berlin.freifunk.net/sven-ola/nylon/packages/olsrd/files/102-olsrd-cvs-fixes.patch
Bernd Petrovitsch [Thu, 29 Nov 2007 00:07:37 +0000 (00:07 +0000)]
- a forgotten hangelog entry
- a few more trivial BSD fixes
Bernd Petrovitsch [Wed, 28 Nov 2007 10:09:03 +0000 (10:09 +0000)]
* applied (a somewhat modified) patch by Patrick McCarty <mccartyp@gmail.com>
Bernd Petrovitsch [Tue, 27 Nov 2007 13:39:22 +0000 (13:39 +0000)]
* fixed compile warnings
Bernd Petrovitsch [Tue, 27 Nov 2007 13:35:06 +0000 (13:35 +0000)]
* fixed compile warnings
Bernd Petrovitsch [Sun, 25 Nov 2007 22:08:57 +0000 (22:08 +0000)]
* saved a few bytes
Bernd Petrovitsch [Sun, 25 Nov 2007 22:02:33 +0000 (22:02 +0000)]
* saved a few bytes
Bernd Petrovitsch [Sun, 25 Nov 2007 21:41:05 +0000 (21:41 +0000)]
* fixed name - and commented the unused enum out
Bernd Petrovitsch [Sun, 25 Nov 2007 21:40:35 +0000 (21:40 +0000)]
* oops, fixed typo
Bernd Petrovitsch [Sun, 25 Nov 2007 21:38:13 +0000 (21:38 +0000)]
* rearranged fields a little bit to save space
Bernd Petrovitsch [Sun, 25 Nov 2007 21:23:29 +0000 (21:23 +0000)]
* fixed bug: ip4cmp() was severly broken
Bernd Petrovitsch [Sun, 25 Nov 2007 13:08:54 +0000 (13:08 +0000)]
* added paranoia checks
Bernd Petrovitsch [Fri, 23 Nov 2007 23:40:43 +0000 (23:40 +0000)]
* prefix_len is always an unsigned 8 bit int
Bernd Petrovitsch [Thu, 22 Nov 2007 11:43:36 +0000 (11:43 +0000)]
* bug fix: set "ipsize" and "maxplen" if we set the the "ip_version"
Bernd Petrovitsch [Tue, 20 Nov 2007 23:20:23 +0000 (23:20 +0000)]
* #define MSG_NOSIGNAL only if it's really not defined. Thanks to
G.Fischer@vt.TU-Berlin.DE for reporting this
Bernd Petrovitsch [Tue, 20 Nov 2007 23:19:08 +0000 (23:19 +0000)]
* use "__attribute__((unused))"
* "olsr_argv = argv" is not used in Win32 so compile completely it out
* Fixed a compiler warning reported by G.Fischer@vt.TU-Berlin.DE
* avoid to cluttered source
* sanitize macros
Bernd Petrovitsch [Tue, 20 Nov 2007 23:16:17 +0000 (23:16 +0000)]
* added "#include <limits.h>" as we use some constant from it
Bernd Petrovitsch [Mon, 19 Nov 2007 14:35:10 +0000 (14:35 +0000)]
* added support for EXTRA_CPPFLAGS
Bernd Petrovitsch [Sun, 18 Nov 2007 22:28:35 +0000 (22:28 +0000)]
- olsr_process_message_neighbors is now a static function and called process_message_neighbors
- olsr_linking_this_2_entries is now a static function and called linking_this_2_entries
- lookup_mpr_status() is now a static function
All of them are defined at the to allow automatic inlining.
Bernd Petrovitsch [Sun, 18 Nov 2007 21:58:32 +0000 (21:58 +0000)]
* killed an unused field
Bernd Petrovitsch [Sun, 18 Nov 2007 20:39:09 +0000 (20:39 +0000)]
* and a few lines in the CHANGELOG
Bernd Petrovitsch [Sun, 18 Nov 2007 20:35:59 +0000 (20:35 +0000)]
- use ip{4,}cmp in lq_avl.[ch] to resue the code. inline_avl_comp_ipv4
is superflous now (and replaced by ip4cmp).
- use the global MIN and MAX macro instead of AVLMIN and AVLMAX
Bernd Petrovitsch [Sun, 18 Nov 2007 20:01:08 +0000 (20:01 +0000)]
consolidated ip{4,6,}{cmp,equal}: all are nwo inline functions using only
the minimal necessary operations. Thanks to Hannes Gredler
<hannes@gedler.at> and Sven-Ola Tuecke <mail2news@commando.de> for pointing
that out and suggesting the `inline`, respectively.
Bernd Petrovitsch [Sat, 17 Nov 2007 00:05:54 +0000 (00:05 +0000)]
* patch by Hannes Gredler <hannes@grwedler.at>: fixed a compile warning
Bernd Petrovitsch [Fri, 16 Nov 2007 22:56:54 +0000 (22:56 +0000)]
* added -Wcast-qual: this killed lots of unnecessary type-casts
Bernd Petrovitsch [Fri, 16 Nov 2007 21:43:55 +0000 (21:43 +0000)]
PATCH by Hannes Gredler <hannes@gredler.at>:
- refactoring of TC parsing to kill another pile of malloc()/free()s
saving (again) code and especially run.time performance.
Bernd Petrovitsch [Fri, 16 Nov 2007 19:52:09 +0000 (19:52 +0000)]
* fixed bug: we got garbled output on the line with "shape=box"
* fixed file descriptor leaks ipc_action()
* killed the two indicator variables "ipc_open" and "ipc_socket_up". We
use the filedescriptors as we can easily use -1 as the "socket is down"
indocation
* killed unused return values
* reduced the clutter: we have now a ipc_send_fmt() function which
does the snprintf() and not spread that all over
Bernd Petrovitsch [Fri, 16 Nov 2007 19:34:26 +0000 (19:34 +0000)]
* added -Wbad-function-cast
Bernd Petrovitsch [Fri, 16 Nov 2007 19:14:33 +0000 (19:14 +0000)]
* added -WWendif-labels
Bernd Petrovitsch [Fri, 16 Nov 2007 19:12:56 +0000 (19:12 +0000)]
* added -Wwrite-strings
Bernd Petrovitsch [Thu, 15 Nov 2007 00:35:32 +0000 (00:35 +0000)]
cleanup: merged html.h into the places where it is used since it is much
easier to see what's going on
cleanup: avoid hard-coded values and use sizeof()
fixed the generated html
Bernd Petrovitsch [Wed, 14 Nov 2007 11:08:05 +0000 (11:08 +0000)]
* and somethigng here too
Bernd Petrovitsch [Wed, 14 Nov 2007 11:03:52 +0000 (11:03 +0000)]
Patch by Hannes Gredler <hannes@gredler.at> to avoid setting routes
with an invalid impossible netmask.
Bernd Petrovitsch [Wed, 14 Nov 2007 11:01:03 +0000 (11:01 +0000)]
* fixed a bug in the olsr_netmask_to_prefix() function which produced
invalid netmasks if feeded with 255.255.255.255.
Thnaks to Hannes Gredler <hannes@gredler.at> for pointing to that
function.
* replaced the second "(for()" loop with a simple "switch()" for the
9 legal/valid/supported values of 1 byte netmasks. So we can now detect
illegal ones and report this with returning UINT_MAX as prefix length
(and writing something out).
That really shouldn't happen but God knows what some clowns sends over
the net to annoy us.
Bernd Petrovitsch [Mon, 12 Nov 2007 00:12:15 +0000 (00:12 +0000)]
fixed and improved the `gcc-warnings` script:
* support older gcc's (as on Debian/Sarge) as they have slightly different
error messages
* removed superflous lines
* replaced the call of `sed` with shell built-in features fixing syntax
errors on Debian/Sarge
* set all locale-relevant variables to "C" (just to be sure)
Bernd Petrovitsch [Sun, 11 Nov 2007 23:25:29 +0000 (23:25 +0000)]
* and something for the CHANGELOG
Bernd Petrovitsch [Sun, 11 Nov 2007 23:10:25 +0000 (23:10 +0000)]
* const'ified olsr_ioctl_{add,del}_route{,6}()
There are probably a few "const" missing for win32 and BSD.
* const'ified moer parameters in many more functions in the .h and
.c files
* init-functions with constant or ignored return values are "void"
* created a typedef fro the callback in "struct export_route_entry"
and using it!
* moved the initilaization of the {add,chg,del}_kernel_list variables
into process_routes.c. So they are now "static" there.
* olsr_export_{add,del}_route{,6} are now static in process_routes.c
since they are not used elsewhere.
* olsr_addroute_add_function() and olsr_deldroute_add_function() were
almost identical - made one function out of it.
Same for olsr_addroute_del_function() and olsr_deldroute_del_function().
* factored out common code
* fixed the sort order in avl_comp_ipv4_prefix(): We compare now the IPv4
values in host byte order (and not network byte order). So the route
list in the e.g. httpinfo plugin is now nicely sorted.
Bernd Petrovitsch [Sun, 11 Nov 2007 22:59:29 +0000 (22:59 +0000)]
* const'ified ipc_route_send_rtentry
Bernd Petrovitsch [Sun, 11 Nov 2007 22:56:34 +0000 (22:56 +0000)]
* killed a superflous "return" at the end of a void function
Bernd Petrovitsch [Sun, 11 Nov 2007 22:55:17 +0000 (22:55 +0000)]
* made inline functions from avl_walk_{first,last,next,prev}
* created "const" versions of it
* inline_avl_comp_ipv4() is now also an inline function (and not a macro)
* created a typedef for the avl tree comparison functions and use it!
* cleanup: compare tree->comp to NULL as it is a pointer
* killed a superflous "return"at the end of a void function
Bernd Petrovitsch [Sun, 11 Nov 2007 22:51:54 +0000 (22:51 +0000)]
* moved the #define INLINE to a more appropriate .h file
Bernd Petrovitsch [Sun, 11 Nov 2007 22:00:39 +0000 (22:00 +0000)]
* do not print / in the routeing table
Bernd Petrovitsch [Sun, 11 Nov 2007 21:32:03 +0000 (21:32 +0000)]
* adapted to the one delivered by the Debian package as a result of a recent discussion on olsr-dev@olsr.org
Bernd Petrovitsch [Fri, 9 Nov 2007 00:11:01 +0000 (00:11 +0000)]
* killed gotos
Bernd Petrovitsch [Thu, 8 Nov 2007 23:54:28 +0000 (23:54 +0000)]
* added a comment
Bernd Petrovitsch [Thu, 8 Nov 2007 23:33:54 +0000 (23:33 +0000)]
* win32 doesn't support that feature
Bernd Petrovitsch [Thu, 8 Nov 2007 23:23:13 +0000 (23:23 +0000)]
* and make it compile on Win32
Bernd Petrovitsch [Thu, 8 Nov 2007 22:47:43 +0000 (22:47 +0000)]
Another fat commit:
The main target was:
- Fixed the misleading definition of "v4" in "struct olsr_ip_addr" fom
"olsr_u32_t" (in network-byteorder!) to "struct in_addr". Lots of
temporary variables to call inet_ntoa()/inet_ptoa() vanished .....
- declare "int_addr", "int_netmask" and "int_broadaddr" in "struct interface"
as "struct sockaddr_in" since it is that what we actually want there (and
it is similar to the IPv6 code).
To get that thoroughly via compiler errors, we get:
- We have now ip4_to_string(), ip6_to_string() and olsr_ip_to_string()
to print a "struct in_addr", "struct in6_addr" and "union olsr_ip_addr"
into a string buffer.
Alas, this also annoyed me since ages:
- cleanup: olsr_ip_to_string() and similar non-reentrant functions now must
get a target buffer. To ease that, there is the "struct ipaddr_str"
which is large enough for all of them (read: for an IPv6 address). This
also removes the cyclic buffer there.
All of these function return a "const char *" which can be directly used
for printf(3) and friends.
And some cleanups:
- const'ified more functions
- converted the source to UTF-8.
- "struct sig_msg" uses an olsr_u8_t for a byte array (and not "char")
- force the few inline function to always be inlined.
- #ifdef the body of the olsr_print_hna_set() and olsr_print_neighbor_table()
if nothing is done
- use "inline_avl_comp_ipv4()" in "avl_comp_ipv4()"
- clean up the routes on more signals. Basically we want to do this on all
signals which terminate the program.
- killed a superflous global buffer in src/main.c
This version was breing since weeks and running for severa day in Vienna's
FunkFeuer net without any noticably problem!
Please report anything that broke!
Bernd Petrovitsch [Thu, 8 Nov 2007 22:17:08 +0000 (22:17 +0000)]
* use automatic variables
Bernd Petrovitsch [Thu, 8 Nov 2007 15:34:11 +0000 (15:34 +0000)]
* killed a superflous variable
Bernd Petrovitsch [Thu, 8 Nov 2007 15:28:48 +0000 (15:28 +0000)]
* s/-lpthread/$(OS_LIB_PTHREAD)/
Bernd Petrovitsch [Mon, 5 Nov 2007 15:34:37 +0000 (15:34 +0000)]
and an entry here
Bernd Petrovitsch [Mon, 5 Nov 2007 15:32:55 +0000 (15:32 +0000)]
Cleanup:
* Merged "struct hna4_entry" and "struct hna6_entry" into
"struct local_hna_entry" (as "struct hna_entry" is a different thing)
Both have almost the same data (IP address + netmask/prefix) so we use
the quite new "struct olsr_ip_prefix" to store it.
Also merged the "hna4" and "hna6" pointer in "struct olsr_config" -
look at the global "olsr_cnf->ip_version".
* const'ified here and there
* added a olsr_ip_prefix_to_string() function
Bernd Petrovitsch [Sun, 4 Nov 2007 23:34:57 +0000 (23:34 +0000)]
* removed left-over target
Bernd Petrovitsch [Sun, 4 Nov 2007 19:07:43 +0000 (19:07 +0000)]
* added a TMPFILES variable. Place all files there which should be removed on
a "clean" (which are usually temporary files - thus the name).
* Killed the recursive make of the src/cfgparser directory
- extracted the directory-local dependencies etc. into src/cfgparser/local.mk
src/cfgparser/Makefile retains all of the rest to build in the local directory.
Does actually anyone need that?
- include'ing that in Makefile
- removed all special handling/casing of cfgparser parts in Makefile
* cleanup: plugin rules are now all more similar
Andreas Tonnesen [Sun, 4 Nov 2007 18:57:17 +0000 (18:57 +0000)]
Patch committed in bug
1825587 from the SF bugtracker. This should add functionallity for detecting WLAN interfaces on openBSD. Committing this since the only available devs to assign bugs to are me and Thomas(which is not very active anymore either).
Bernd Petrovitsch [Sun, 4 Nov 2007 18:45:30 +0000 (18:45 +0000)]
* also check -m
Bernd Petrovitsch [Sun, 4 Nov 2007 17:52:13 +0000 (17:52 +0000)]
* cleanup:
- #ifdef more away if NODEBUG is enabled.
- reduce the size of the data and code
Bernd Petrovitsch [Sun, 4 Nov 2007 15:27:50 +0000 (15:27 +0000)]
* do not generate a dependency file for the generated .c file since it makes no sense
Bernd Petrovitsch [Sat, 3 Nov 2007 23:21:28 +0000 (23:21 +0000)]
Killed "struct olsrd_config *cnf" in src/cfgparser/olsrd_conf.h and replaced
it's usage with "olsr_cnf" from src/defs.h.
Serious cleanup: olsrd_get_default_cnf() does no longer initialize the global
"cnf" variable" but uses a local one which is returned as before. And
olsrd_parse_cnf() does no longer return the global variable it is working on.
Bernd Petrovitsch [Sat, 3 Nov 2007 23:11:40 +0000 (23:11 +0000)]
PATCH by John Hay <jhay@meraka.org.za>:
- also printout our own HNAs in the dotdraw plugin.
Bernd Petrovitsch [Sat, 3 Nov 2007 23:03:55 +0000 (23:03 +0000)]
OK, I was too silly to throw it into the correct plugin.
Should be fixed now.
Bernd Petrovitsch [Sat, 3 Nov 2007 22:56:07 +0000 (22:56 +0000)]
added meshcube.org/nylon/utils/olsr-topology-view.pl into
lib/httpinfo/misc as suggested by Jens Nachtigall <nachtigall@web.de>
Bernd Petrovitsch [Fri, 2 Nov 2007 22:00:26 +0000 (22:00 +0000)]
* added a comment
Bernd Petrovitsch [Fri, 2 Nov 2007 20:58:07 +0000 (20:58 +0000)]
Some minor and trivial cleanup: Added two functions to calculate the ETX
value from a struct tc_edge_entry and a struct link_entry, respectively.
And use them everywhere.
Bernd Petrovitsch [Fri, 2 Nov 2007 15:51:46 +0000 (15:51 +0000)]
While working on something else, I stumbled over this. Obviously no one\ 1
ever used the "secure" plugin on 64bit hardware:
- fixed lib/secure/src/md5.h: This was broken as it had
"typedef unsigned long int UINT4;". "unsigned long int" is 8 bytes on
x86_64.
We are using now the standardized types from <inttypes.h>.
- fixed warnings and improved lib/secure/src/md5.c:
* we are using memcpy() and memset() instead of the open-coded loops as
suggested
* const'ified
* added function prototypes for static functions
* moved static functions to the top so that their declaration is before
their use to allow gcc to inline if only used once.
It remains ugly - God knows why there are that so many useless
type-casts.
Bernd Petrovitsch [Fri, 2 Nov 2007 15:37:09 +0000 (15:37 +0000)]
* ooops, hide the dependency generation calls again
Bernd Petrovitsch [Fri, 2 Nov 2007 10:57:30 +0000 (10:57 +0000)]
* completed changelog message
Bernd Petrovitsch [Fri, 2 Nov 2007 10:16:39 +0000 (10:16 +0000)]
With the help of Patrick McCarty <mccartyp@gmail.com>, we got olsr to build
on OpenBSD again:
* OpenBSD'd `sed` is pretty conservative and doesn't understand 's|re|sub|'.
So we fall back to the old 's/re/sub/' and quote all '/' and '.' correctly
(which was the main reason to use '|').
* OpenBSD needs an explicit "#include <sys/types.h>" before
"#include <netinet/in.h>".
Bernd Petrovitsch [Fri, 2 Nov 2007 10:11:43 +0000 (10:11 +0000)]
* as suggested by Jens Nachtigall <nachtigall@web.de>: install the LQ version as default
Bernd Petrovitsch [Fri, 2 Nov 2007 09:38:56 +0000 (09:38 +0000)]
* bug fix by Hannes Gredler <hannes@gredler.at> fixing not deleted tc entry
Bernd Petrovitsch [Tue, 30 Oct 2007 09:19:22 +0000 (09:19 +0000)]
* fix startup breakage
Bernd Petrovitsch [Wed, 24 Oct 2007 13:51:11 +0000 (13:51 +0000)]
* unified typedefs of olsr_{u,}{8,16,32}_t: It builds that way on Linux and Windows and *BSD had it before. That leaves MacOSx for someone to test.
Bernd Petrovitsch [Wed, 24 Oct 2007 13:46:36 +0000 (13:46 +0000)]
* cleanup: use for() instead of while to save lots of lines - and minor other stuff
Bernd Petrovitsch [Wed, 24 Oct 2007 13:45:00 +0000 (13:45 +0000)]
* avoid a warning
Bernd Petrovitsch [Wed, 24 Oct 2007 13:44:26 +0000 (13:44 +0000)]
* cleanup: use ARRAYSIZE, const'ifications and killed an unsed #define
Bernd Petrovitsch [Mon, 22 Oct 2007 21:02:49 +0000 (21:02 +0000)]
* killed unused functions
Aaron Kaplan [Sun, 21 Oct 2007 21:46:43 +0000 (21:46 +0000)]
document NatThreshold (even though it is only a freifunk fw patch at the moment)
Aaron Kaplan [Sun, 21 Oct 2007 21:16:11 +0000 (21:16 +0000)]
updated reality on Mac OS X. Added some extra infos
Bernd Petrovitsch [Sun, 21 Oct 2007 20:55:11 +0000 (20:55 +0000)]
* cont'ified it
Bernd Petrovitsch [Sun, 21 Oct 2007 20:53:13 +0000 (20:53 +0000)]
* avoid useless warnings
Bernd Petrovitsch [Sun, 21 Oct 2007 20:37:58 +0000 (20:37 +0000)]
* cont'ified it
Bernd Petrovitsch [Sun, 21 Oct 2007 20:15:51 +0000 (20:15 +0000)]
* cont'ified it
Bernd Petrovitsch [Sun, 21 Oct 2007 20:09:09 +0000 (20:09 +0000)]
* avoid warnings
Bernd Petrovitsch [Sat, 20 Oct 2007 21:07:24 +0000 (21:07 +0000)]
* added a MIN and MAX macro
Bernd Petrovitsch [Sat, 20 Oct 2007 20:41:04 +0000 (20:41 +0000)]
* fixed a few compile bugs (and simplified it a little bit)
Bernd Petrovitsch [Sat, 20 Oct 2007 20:23:12 +0000 (20:23 +0000)]
Ooops, I overlooked that NODEBUG is also used in defs.h (and not only the
txtinfo plugin):
* added a 4th high level variable: NO_DEBUG_MESSAGES to compile the debug
OLSR_PRINTF completely out.
Bernd Petrovitsch [Sat, 20 Oct 2007 19:27:33 +0000 (19:27 +0000)]
- added "-fomit-frame-pointer" to the normal build
. start a new era on configuration:
We introduce a new higher level configuratino layer to enable or disable
features/configuration options:
Rationale: Take "debug build" as prime example.
To have a real debug build, you want (at least) "-g" in CFLAGS and no
stripping. It is cumbersome and error prone to edit that by hand (or pass
parameters to the "make") and with a new options somewhere (e.g.
"-fomit-frame-pointer" from above) one has to adjust that.
For these (so called) high leveloptions, I started with
--- snip ----
# activate debugging with 1 or deactivate with 0
DEBUG ?= 0
# the optimize option to be set for gcc
OPTIMIZE ?= -O2
# enable mudflap with 1 or deactivate with 0
# you need a recent enough gcc and the libmudflap installed
MUDFLAP ?= 0
--- snip ----
Feel free to propose other useful ones (and submit patches;-).
The lower part, which is basically the former Makefile.inc, gets now "ifeq"
etc. so that a simple "make DEBUG=1" makes a real debug build.
ANd this is available and usefule to the plugin Makeifle's too of course.
Bernd Petrovitsch [Sat, 20 Oct 2007 15:16:32 +0000 (15:16 +0000)]
* fixed more compile bugs if DEBUG is enabled
Bernd Petrovitsch [Sat, 20 Oct 2007 12:59:08 +0000 (12:59 +0000)]
* fixed compile bugs if DEBUG is enabled
Bernd Petrovitsch [Sat, 20 Oct 2007 11:55:23 +0000 (11:55 +0000)]
* set version to 0.5.5pre
Bernd Petrovitsch [Sat, 20 Oct 2007 11:26:25 +0000 (11:26 +0000)]
* set version to 0.5.4
Bernd Petrovitsch [Wed, 17 Oct 2007 10:22:49 +0000 (10:22 +0000)]
* Bug fix from Sven-Ola Tuecke: initialize malloc()ed struct
Bernd Petrovitsch [Wed, 17 Oct 2007 07:30:34 +0000 (07:30 +0000)]
* Bug fix from Sven-Ola Tuecke: initialize malloc()ed struct
Bernd Petrovitsch [Tue, 16 Oct 2007 10:01:29 +0000 (10:01 +0000)]
* applied a nad-edited version of Sven-Ola Tueckes "Save the fish" patch