Re: this was answered

[Cloudmaster <sauer@cloudmaster.com>]

On Mon, 15 May 2000, Charles Menzes wrote:
> what is the correct syntax for a patch to the kernel?
> 
> patch p0 < /home/charles/lids-0.9.3/lids-0.9.3-2.2.15.patch
> 
> i think tim mentioned before that there was something wrong with the
> syntax...

Yeah, the way it shoud be done is

cd /usr/src/kernel-patches
wget ftp://nearest.mirror/newest/source.patch.bz2
../linux/scripts/patch-kernel

:)

based on that script, the command to use is 
	patch -p1 -s -N -E -d $sourcedir < $patchfile
because
	-p1 strips the leading slash on filenames if present
	-s  makes it silent (nto neccesarily needed)
	-N  ignores patches already applied or reversed
	-E  forces removal of empty files (which should be done
		automatically anyway
	-d  changes to the specified directory before proceeding
assuming that $sourcedir is your kernel source directory and $patchfile is
your patch file.

However, in the README file in the kernel docs, it is advised that one uses 
	bzip2 -dc patchXX.bz2 | patch -p0
which implies that what you have above will work if you've already got the
patch file uncompressed.  It may be better to use
   patch -p1 -d /usr/src/linux < ~charles/lids-0.9.3/lids-0.9.3-2.2.15.patch
anyway, so you don't have to worry about your source tree being in
/home/kernel/place and the patch generator using /usr/src/linux, however.

I'm partial to just leaving the patches compressed and using the
patch-kernel script to take all the hard work out of typing
"patch -p0 < bzcat patch.bz2".  Mmmm, learning how "patch" works...

--Danny, doing little more than reading man pages and kernel docs


--
To unsubscribe, send email to majordomo@luci.org with
"unsubscribe luci-discuss" in the body.