Skip to contentSkip to conversation

2.6.18-rc6-mm1

DISCUSSION Chuck Ebbert 2 participants 2 messages Sep 11, 2006

lore
Share & tools
lore
#0 lore raw

In-Reply-To: <[email protected]>

Quoted 11 lines · Andrew Morton

wget ftp://ftp.kernel.org/pub/linux/kernel/v2.6/testing/linux-2.6.18-rc6.tar.bz2
wget ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.18-rc6/2.6.18-rc6-mm1/2.6.18-rc6-mm1-broken-out.tar.gz box:/home/akpm> mkdir aa
box:/home/akpm> cd aa
box:/home/akpm/aa> tar xfj ../linux-2.6.18-rc6.tar.bz2
box:/home/akpm/aa> cd linux-2.6.18-rc6
box:/home/akpm/aa/linux-2.6.18-rc6> tar xfz ../../2.6.18-rc6-mm1-broken-out.tar.gz box:/home/akpm/aa/linux-2.6.18-rc6> mv broken-out patches
box:/home/akpm/aa/linux-2.6.18-rc6> quilt push -a > /dev/null
box:/home/akpm/aa/linux-2.6.18-rc6> quilt applied | wc -l
1835

I found the problem:

$ set | fgrep QUILT
QUILT_DIFF_OPTS=-p
QUILT_PATCH_OPTS=--fuzz=0
                 ^^^^^^^^

Your patchset does have conflicts -- you're just ignoring them by accepting fuzz (and patch hunks can even end up being applied at the wrong place.)

Signature · 1 line
Chuck
#1 lore raw
Quoted 18 lines · On Mon, 11 Sep 2006 17:56:26 -0400 Chuck Ebbert <[email protected]>…

In-Reply-To: <[email protected]>

Quoted 11 lines · Andrew Morton

wget ftp://ftp.kernel.org/pub/linux/kernel/v2.6/testing/linux-2.6.18-rc6.tar.bz2
wget ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.18-rc6/2.6.18-rc6-mm1/2.6.18-rc6-mm1-broken-out.tar.gz box:/home/akpm> mkdir aa
box:/home/akpm> cd aa
box:/home/akpm/aa> tar xfj ../linux-2.6.18-rc6.tar.bz2
box:/home/akpm/aa> cd linux-2.6.18-rc6
box:/home/akpm/aa/linux-2.6.18-rc6> tar xfz ../../2.6.18-rc6-mm1-broken-out.tar.gz box:/home/akpm/aa/linux-2.6.18-rc6> mv broken-out patches
box:/home/akpm/aa/linux-2.6.18-rc6> quilt push -a > /dev/null
box:/home/akpm/aa/linux-2.6.18-rc6> quilt applied | wc -l
1835

I found the problem:

Quoted 9 lines
$ set | fgrep QUILT
QUILT_DIFF_OPTS=-p
QUILT_PATCH_OPTS=--fuzz=0
                 ^^^^^^^^

Your patchset does have conflicts -- you're just ignoring them by accepting fuzz (and patch hunks can even end up being applied at the wrong place.)

Sure. The -mm queue always has large amount of fuzz. Lots and lots. I'll occasionally go and rediff the fuzzy patches to clean things up, but that involves pointlessly incrementing the local version number on 200-300 patches, which I prefer to avoid.