Here is how to add your new `foo' patch to patch-o-matic:

1) Create a kernel patch which can be applied inside a kernel tree
   with `patch -p1', and call it `foo.patch', or `foo.patch.PROTOCOL'
   if it belongs in linux/net/PROTOCOL/netfilter rather than
   linux/net/ipv4/netfilter (the default).

2) Create a description file called `foo.patch.help'.

3) If your patch creates a new CONFIG option, eliminate that from the
   patch above, and create a file called `foo.patch.config.in'.  The
   format of this file is as follows:

	EXACT LINE TO FOLLOW
	<text to paste in>

   This allows you to specify the entry in net/ipv4/netfilter/Config.in
   that you wish your text to follow.  Note that it must be an exact match.
   You can have more than one of these files, to make multiple entries
   in different places, by calling successive ones foo.patch.config.in*.

   You also need to make an entry in Documentation/Configure.help;
   once again, eliminate this from your patch file, and create a file
   called `foo.patch.configure.help' like so:

	EXACT CONFIG OPTION TO FOLLOW
	<text to paste in>

   Your text will be placed after the config option you indicated
   (with a blank line before and after).  You can have more than one
   of these files, to make multiple entries in different places, by
   calling successive ones foo.patch.configure.help*.

4) If you want to edit the Makefile, eliminate that from the patch
   above, and create a file called `foo.patch.makefile'.  The format
   of this file is as follows:

	EXACT LINE TO MATCH
	<text to paste in>

   You can have more than one of these files, to make multiple entries
   in different places, by calling successive ones foo.patch.makefile*.

Finally, if you want to have a libipt_foo built when the patch is
applied, add a test for your extension in the iptables userspace
distribution, called `userspace/extensions/.foo-test'.  This should be
a small shell script which echos the names of the libraries to be
built if the corresponding include file exists in the kernel tree
(this test may be more complex: figure out some way of reliably
detecting that the patch has been applied to $KERNEL_DIR).  This isn't
strictly part of patch-o-matic, but it means things will just work
normally whether they've used patch-o-matic, or if your patch is
included in future kernel releases.

Enjoy!
Rusty.

