Patchwork [1/2] classes/image_types: add IMAGE_TUNE2FS_ext3

login
register
mail settings
Submitter Paul Eggleton
Date April 4, 2012, 3:32 p.m.
Message ID <0d5a88036c74a354a62e13cf83d05f99fca8175b.1333553396.git.paul.eggleton@linux.intel.com>
Download mbox | patch
Permalink /patch/25163/
State New
Headers show

Comments

Paul Eggleton - April 4, 2012, 3:32 p.m.
Add a variable that allows additional options to be passed to tune2fs,
such as -m to control the amount of space reserved for root.

This could potentially be added for other filesystems, but right now we
need it only for ext3 for self-hosted-image.

Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
---
 meta/classes/image_types.bbclass |    4 +++-
 1 files changed, 3 insertions(+), 1 deletions(-)
Otavio Salvador - April 4, 2012, 4:57 p.m.
On Wed, Apr 4, 2012 at 12:32, Paul Eggleton
<paul.eggleton@linux.intel.com> wrote:
> Add a variable that allows additional options to be passed to tune2fs,
> such as -m to control the amount of space reserved for root.

I'd prefer something like

IMG_POSTCMD_<fstype>

So it is generic enough to not be confusing for other filesystems.
Paul Eggleton - April 4, 2012, 5:12 p.m.
On Wednesday 04 April 2012 13:57:04 Otavio Salvador wrote:
> On Wed, Apr 4, 2012 at 12:32, Paul Eggleton
> 
> <paul.eggleton@linux.intel.com> wrote:
> > Add a variable that allows additional options to be passed to tune2fs,
> > such as -m to control the amount of space reserved for root.
> 
> I'd prefer something like
> 
> IMG_POSTCMD_<fstype>
> 
> So it is generic enough to not be confusing for other filesystems.

Are you suggesting this would hold the entire post-command to be run? If not, 
what other post command could we be referring to in the case of other 
filesystems? (tune2fs being valid for ext2, ext3 and ext4).

Cheers,
Paul
Otavio Salvador - April 4, 2012, 5:26 p.m.
On Wed, Apr 4, 2012 at 14:12, Paul Eggleton
<paul.eggleton@linux.intel.com> wrote:
> Are you suggesting this would hold the entire post-command to be run? If not,
> what other post command could we be referring to in the case of other
> filesystems? (tune2fs being valid for ext2, ext3 and ext4).

Yes; it is more generic and for someone using it, it makes no
difference in complexity et all.
Paul Eggleton - April 4, 2012, 7:30 p.m.
On Wednesday 04 April 2012 14:26:26 Otavio Salvador wrote:
> On Wed, Apr 4, 2012 at 14:12, Paul Eggleton
> 
> <paul.eggleton@linux.intel.com> wrote:
> > Are you suggesting this would hold the entire post-command to be run? If
> > not, what other post command could we be referring to in the case of
> > other filesystems? (tune2fs being valid for ext2, ext3 and ext4).
> 
> Yes; it is more generic and for someone using it, it makes no
> difference in complexity et all.

Actually it does, but I can see some value in it. v2 on the way.

Cheers,
Paul

Patch

diff --git a/meta/classes/image_types.bbclass b/meta/classes/image_types.bbclass
index b590ae1..152e5b4 100644
--- a/meta/classes/image_types.bbclass
+++ b/meta/classes/image_types.bbclass
@@ -132,9 +132,11 @@  IMAGE_CMD_ext2 () {
 	rmdir ${DEPLOY_DIR_IMAGE}/tmp.gz-${PN}
 }
 
+IMAGE_TUNE2FS_ext3 ?= ""
+
 IMAGE_CMD_ext3 () {
 	genext2fs -b $ROOTFS_SIZE -d ${IMAGE_ROOTFS} ${EXTRA_IMAGECMD} ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.ext3
-	tune2fs -j ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.ext3
+	tune2fs -j ${IMAGE_TUNE2FS_ext3} ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.ext3
 }
 
 oe_mkext4fs () {