[Openswan dev] [PATCH 0/2] ESP_NULL support for openswan-2.4.6
mcr at sandelman.ottawa.on.ca
Mon Oct 2 12:50:52 EDT 2006
-----BEGIN PGP SIGNED MESSAGE-----
>>>>> "JuanJo" == JuanJo Ciarlante <jjo-ipsec at mendoza.gov.ar> writes:
>> I can understand this... I will accept your patches, if you'll
>> write the test cases for them.
JuanJo> ughh... I miss that good'ol'times of nonQA devel :-S
JuanJo> I'll dig into testing/ and learn... the path starts there
JuanJo> and under docs/HACKING, right? (tips&tricks for the task
JuanJo> pretty welcome ;)
cp testing/utils/umlsetup-sample.sh umlsetup.sh
vi umlsetup.sh <- get various pieces,
mostly a kernel tree, and a
basicroot, pick UMLPOOL location.
MYBOX=`pwd` export MYBOX
xterm -n EAST -e $UMLPOOL/east/start.sh single &
xterm -n WEST -e $UMLPOOL/west/start.sh single &
(or see http://www.sandelman.ca/tmp/startuml.sh )
login as root/root on each.
east# cd /testing/pluto/basic-pluto-01
east# source eastinit.sh
west# cd /testing/pluto/basic-pluto-01
west# source westinit.sh
west# source westrun.sh
Once that works, then you can run the tests manually:
You'll want to look at the KLIPS tests, which only use one UML.
The patches to pluto to permit it to recognize "NULL" algorithm should
be okay. The problem is that I don't think we want to ship a kernel or
a UML default config that has ESP-NULL present.
So, I think that it has to be a modular load in the test case itself.
I know that we didn't really care if we got all of the modular module
loading working in 2.x, since we expected to use cryptoapi.
] Bear: "Me, I'm just the shape of a bear." | firewalls [
] Michael Richardson, Xelerance Corporation, Ottawa, ON |net architect[
] mcr at xelerance.com http://www.sandelman.ottawa.on.ca/mcr/ |device driver[
] panic("Just another Debian GNU/Linux using, kernel hacking, security guy"); [
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (GNU/Linux)
Comment: Finger me for keys
-----END PGP SIGNATURE-----
More information about the Dev