Openembedded Core Discussions
 help / color / mirror / Atom feed
* [PATCH 0/2] rework jffs2 images creation, add jffs2 summary
@ 2011-12-06 23:23 Andrea Adami
  2011-12-06 23:23 ` [PATCH 1/2] image_types.bbclass: implement jffs2 summary images (sum.jffs2) Andrea Adami
                   ` (2 more replies)
  0 siblings, 3 replies; 14+ messages in thread
From: Andrea Adami @ 2011-12-06 23:23 UTC (permalink / raw)
  To: openembedded-core

* The first patch adds support for the creation of
* jffs2 summary images for a faster mount.
* The second patch forces the --no-cleanmarkers option
* to further improve boot time.

Andrea Adami (2):
  image_types.bbclass: implement jffs2 summary images (sum.jffs2)
  image_types.bbclass: disable cleanmarkers for jffs2 images

 meta/classes/image_types.bbclass |    7 +++++--
 1 files changed, 5 insertions(+), 2 deletions(-)

-- 
1.7.3.4




^ permalink raw reply	[flat|nested] 14+ messages in thread
* [PATCH 1/2] image_types.bbclass: implement jffs2 summary images (sum.jffs2)
  2011-12-06 23:23 [PATCH 0/2] rework jffs2 images creation, add jffs2 summary Andrea Adami
@ 2011-12-06 23:23 ` Andrea Adami
  2011-12-08 14:57   ` Tom Rini
  2011-12-06 23:23 ` [PATCH 2/2] image_types.bbclass: disable cleanmarkers for jffs2 images Andrea Adami
  2011-12-08 18:22 ` [PATCH 0/2] rework jffs2 images creation, add jffs2 summary Saul Wold
  2 siblings, 1 reply; 14+ messages in thread
From: Andrea Adami @ 2011-12-06 23:23 UTC (permalink / raw)
  To: openembedded-core

* Building the jffs2 filesystem to include summary information speeds up
* the mount process considerably at the cost of increased size.
* The rate of speedup is generally higher on NAND chips and on the chips
* where the erase block size is large.

Signed-off-by: Andrea Adami <andrea.adami@gmail.com>
---
 meta/classes/image_types.bbclass |    5 ++++-
 1 files changed, 4 insertions(+), 1 deletions(-)

diff --git a/meta/classes/image_types.bbclass b/meta/classes/image_types.bbclass
index 29b6380..bd4b7bc 100644
--- a/meta/classes/image_types.bbclass
+++ b/meta/classes/image_types.bbclass
@@ -35,6 +35,8 @@ XZ_COMPRESSION_LEVEL ?= "-e -9"
 XZ_INTEGRITY_CHECK ?= "crc32"
 
 IMAGE_CMD_jffs2 = "mkfs.jffs2 --root=${IMAGE_ROOTFS} --faketime --output=${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.jffs2 ${EXTRA_IMAGECMD}"
+IMAGE_CMD_sum.jffs2 = "${IMAGE_CMD_jffs2} && sumtool -i ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.jffs2 \
+	-o ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.sum.jffs2 ${EXTRA_IMAGECMD}"
 
 IMAGE_CMD_cramfs = "mkcramfs ${IMAGE_ROOTFS} ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.cramfs ${EXTRA_IMAGECMD}"
 
@@ -138,6 +140,7 @@ EXTRA_IMAGECMD_btrfs ?= ""
 
 IMAGE_DEPENDS = ""
 IMAGE_DEPENDS_jffs2 = "mtd-utils-native"
+IMAGE_DEPENDS_sum.jffs2 = "mtd-utils-native"
 IMAGE_DEPENDS_cramfs = "cramfs-native"
 IMAGE_DEPENDS_ext2 = "genext2fs-native"
 IMAGE_DEPENDS_ext2.gz = "genext2fs-native"
@@ -157,4 +160,4 @@ IMAGE_DEPENDS_ubi = "mtd-utils-native"
 IMAGE_DEPENDS_ubifs = "mtd-utils-native"
 
 # This variable is available to request which values are suitable for IMAGE_FSTYPES
-IMAGE_TYPES = "jffs2 cramfs ext2 ext2.gz ext2.bz2 ext3 ext3.gz ext2.lzma live squashfs squashfs-lzma ubi tar tar.gz tar.bz2 tar.xz cpio cpio.gz cpio.xz cpio.lzma"
+IMAGE_TYPES = "jffs2 sum.jffs2 cramfs ext2 ext2.gz ext2.bz2 ext3 ext3.gz ext2.lzma live squashfs squashfs-lzma ubi tar tar.gz tar.bz2 tar.xz cpio cpio.gz cpio.xz cpio.lzma"
-- 
1.7.3.4




^ permalink raw reply related	[flat|nested] 14+ messages in thread
* [PATCH 2/2] image_types.bbclass: disable cleanmarkers for jffs2 images
  2011-12-06 23:23 [PATCH 0/2] rework jffs2 images creation, add jffs2 summary Andrea Adami
  2011-12-06 23:23 ` [PATCH 1/2] image_types.bbclass: implement jffs2 summary images (sum.jffs2) Andrea Adami
@ 2011-12-06 23:23 ` Andrea Adami
  2011-12-08 14:59   ` Tom Rini
  2011-12-08 18:22 ` [PATCH 0/2] rework jffs2 images creation, add jffs2 summary Saul Wold
  2 siblings, 1 reply; 14+ messages in thread
From: Andrea Adami @ 2011-12-06 23:23 UTC (permalink / raw)
  To: openembedded-core

* Add '-n' to suppress cleanmarker nodes since jffs2
* is usually used for NAND flash and the cleanmarkers are created in
* the OOB area by flash_eraseall -j
*
* From man pages: -n, --no-cleanmarkers
* Do not write cleanmarker nodes to the beginning of each erase block.
* This option can be useful for creating JFFS2 images for use on NAND flash,
* and for creating images which are to be used on a variety of hardware with differing eraseblock sizes.

Signed-off-by: Andrea Adami <andrea.adami@gmail.com>
---
 meta/classes/image_types.bbclass |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/meta/classes/image_types.bbclass b/meta/classes/image_types.bbclass
index bd4b7bc..489fbe1 100644
--- a/meta/classes/image_types.bbclass
+++ b/meta/classes/image_types.bbclass
@@ -34,9 +34,9 @@ runimagecmd () {
 XZ_COMPRESSION_LEVEL ?= "-e -9"
 XZ_INTEGRITY_CHECK ?= "crc32"
 
-IMAGE_CMD_jffs2 = "mkfs.jffs2 --root=${IMAGE_ROOTFS} --faketime --output=${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.jffs2 ${EXTRA_IMAGECMD}"
+IMAGE_CMD_jffs2 = "mkfs.jffs2 --root=${IMAGE_ROOTFS} --faketime --output=${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.jffs2 -n ${EXTRA_IMAGECMD}"
 IMAGE_CMD_sum.jffs2 = "${IMAGE_CMD_jffs2} && sumtool -i ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.jffs2 \
-	-o ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.sum.jffs2 ${EXTRA_IMAGECMD}"
+	-o ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.sum.jffs2 -n ${EXTRA_IMAGECMD}"
 
 IMAGE_CMD_cramfs = "mkcramfs ${IMAGE_ROOTFS} ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.cramfs ${EXTRA_IMAGECMD}"
 
-- 
1.7.3.4




^ permalink raw reply related	[flat|nested] 14+ messages in thread
* Re: [PATCH 1/2] image_types.bbclass: implement jffs2 summary images (sum.jffs2)
  2011-12-06 23:23 ` [PATCH 1/2] image_types.bbclass: implement jffs2 summary images (sum.jffs2) Andrea Adami
@ 2011-12-08 14:57   ` Tom Rini
  2011-12-08 15:10     ` Martin Jansa
  0 siblings, 1 reply; 14+ messages in thread
From: Tom Rini @ 2011-12-08 14:57 UTC (permalink / raw)
  To: Patches and discussions about the oe-core layer

On Tue, Dec 6, 2011 at 4:23 PM, Andrea Adami <andrea.adami@gmail.com> wrote:
> * Building the jffs2 filesystem to include summary information speeds up
> * the mount process considerably at the cost of increased size.
> * The rate of speedup is generally higher on NAND chips and on the chips
> * where the erase block size is large.
>
> Signed-off-by: Andrea Adami <andrea.adami@gmail.com>

Acked-by: Tom Rini <tom.rini@gmail.com>

> ---
>  meta/classes/image_types.bbclass |    5 ++++-
>  1 files changed, 4 insertions(+), 1 deletions(-)
>
> diff --git a/meta/classes/image_types.bbclass b/meta/classes/image_types.bbclass
> index 29b6380..bd4b7bc 100644
> --- a/meta/classes/image_types.bbclass
> +++ b/meta/classes/image_types.bbclass
> @@ -35,6 +35,8 @@ XZ_COMPRESSION_LEVEL ?= "-e -9"
>  XZ_INTEGRITY_CHECK ?= "crc32"
>
>  IMAGE_CMD_jffs2 = "mkfs.jffs2 --root=${IMAGE_ROOTFS} --faketime --output=${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.jffs2 ${EXTRA_IMAGECMD}"
> +IMAGE_CMD_sum.jffs2 = "${IMAGE_CMD_jffs2} && sumtool -i ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.jffs2 \
> +       -o ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.sum.jffs2 ${EXTRA_IMAGECMD}"
>
>  IMAGE_CMD_cramfs = "mkcramfs ${IMAGE_ROOTFS} ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.cramfs ${EXTRA_IMAGECMD}"
>
> @@ -138,6 +140,7 @@ EXTRA_IMAGECMD_btrfs ?= ""
>
>  IMAGE_DEPENDS = ""
>  IMAGE_DEPENDS_jffs2 = "mtd-utils-native"
> +IMAGE_DEPENDS_sum.jffs2 = "mtd-utils-native"
>  IMAGE_DEPENDS_cramfs = "cramfs-native"
>  IMAGE_DEPENDS_ext2 = "genext2fs-native"
>  IMAGE_DEPENDS_ext2.gz = "genext2fs-native"
> @@ -157,4 +160,4 @@ IMAGE_DEPENDS_ubi = "mtd-utils-native"
>  IMAGE_DEPENDS_ubifs = "mtd-utils-native"
>
>  # This variable is available to request which values are suitable for IMAGE_FSTYPES
> -IMAGE_TYPES = "jffs2 cramfs ext2 ext2.gz ext2.bz2 ext3 ext3.gz ext2.lzma live squashfs squashfs-lzma ubi tar tar.gz tar.bz2 tar.xz cpio cpio.gz cpio.xz cpio.lzma"
> +IMAGE_TYPES = "jffs2 sum.jffs2 cramfs ext2 ext2.gz ext2.bz2 ext3 ext3.gz ext2.lzma live squashfs squashfs-lzma ubi tar tar.gz tar.bz2 tar.xz cpio cpio.gz cpio.xz cpio.lzma"
> --
> 1.7.3.4
>
>
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core



-- 
Tom



^ permalink raw reply	[flat|nested] 14+ messages in thread
* Re: [PATCH 2/2] image_types.bbclass: disable cleanmarkers for jffs2 images
  2011-12-06 23:23 ` [PATCH 2/2] image_types.bbclass: disable cleanmarkers for jffs2 images Andrea Adami
@ 2011-12-08 14:59   ` Tom Rini
  0 siblings, 0 replies; 14+ messages in thread
From: Tom Rini @ 2011-12-08 14:59 UTC (permalink / raw)
  To: Patches and discussions about the oe-core layer

On Tue, Dec 6, 2011 at 4:23 PM, Andrea Adami <andrea.adami@gmail.com> wrote:
> * Add '-n' to suppress cleanmarker nodes since jffs2
> * is usually used for NAND flash and the cleanmarkers are created in
> * the OOB area by flash_eraseall -j
> *
> * From man pages: -n, --no-cleanmarkers
> * Do not write cleanmarker nodes to the beginning of each erase block.
> * This option can be useful for creating JFFS2 images for use on NAND flash,
> * and for creating images which are to be used on a variety of hardware with differing eraseblock sizes.
>
> Signed-off-by: Andrea Adami <andrea.adami@gmail.com>

Acked-by: Tom Rini <tom.rini@gmail.com>

> ---
>  meta/classes/image_types.bbclass |    4 ++--
>  1 files changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/meta/classes/image_types.bbclass b/meta/classes/image_types.bbclass
> index bd4b7bc..489fbe1 100644
> --- a/meta/classes/image_types.bbclass
> +++ b/meta/classes/image_types.bbclass
> @@ -34,9 +34,9 @@ runimagecmd () {
>  XZ_COMPRESSION_LEVEL ?= "-e -9"
>  XZ_INTEGRITY_CHECK ?= "crc32"
>
> -IMAGE_CMD_jffs2 = "mkfs.jffs2 --root=${IMAGE_ROOTFS} --faketime --output=${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.jffs2 ${EXTRA_IMAGECMD}"
> +IMAGE_CMD_jffs2 = "mkfs.jffs2 --root=${IMAGE_ROOTFS} --faketime --output=${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.jffs2 -n ${EXTRA_IMAGECMD}"
>  IMAGE_CMD_sum.jffs2 = "${IMAGE_CMD_jffs2} && sumtool -i ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.jffs2 \
> -       -o ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.sum.jffs2 ${EXTRA_IMAGECMD}"
> +       -o ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.sum.jffs2 -n ${EXTRA_IMAGECMD}"
>
>  IMAGE_CMD_cramfs = "mkcramfs ${IMAGE_ROOTFS} ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.cramfs ${EXTRA_IMAGECMD}"
>
> --
> 1.7.3.4
>
>
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core



-- 
Tom



^ permalink raw reply	[flat|nested] 14+ messages in thread
* Re: [PATCH 1/2] image_types.bbclass: implement jffs2 summary images (sum.jffs2)
  2011-12-08 14:57   ` Tom Rini
@ 2011-12-08 15:10     ` Martin Jansa
  2011-12-08 21:30       ` Andrea Adami
  0 siblings, 1 reply; 14+ messages in thread
From: Martin Jansa @ 2011-12-08 15:10 UTC (permalink / raw)
  To: Patches and discussions about the oe-core layer

[-- Attachment #1: Type: text/plain, Size: 3283 bytes --]

On Thu, Dec 08, 2011 at 07:57:52AM -0700, Tom Rini wrote:
> On Tue, Dec 6, 2011 at 4:23 PM, Andrea Adami <andrea.adami@gmail.com> wrote:
> > * Building the jffs2 filesystem to include summary information speeds up
> > * the mount process considerably at the cost of increased size.
> > * The rate of speedup is generally higher on NAND chips and on the chips
> > * where the erase block size is large.
> >
> > Signed-off-by: Andrea Adami <andrea.adami@gmail.com>

I'll repeat my comment from yesterday, but IMHO it would be easier to
see the difference between those 2 images if the IMAGE_CMD_jffs2 output is 
renamed to .jffs2.nosummary and IMAGE_CMD_sum.jffs2 (or renamed IMAGE_CMD_jffs2.summary)
produces .jffs2.summary (or even just .jffs2 if we can agree that .summary is
what enduser wants by default)

But other than that it looks fine and will make my IMAGE_DEPENDS_jffs2 a
bit shorter.

Regards,

> 
> Acked-by: Tom Rini <tom.rini@gmail.com>
> 
> > ---
> >  meta/classes/image_types.bbclass |    5 ++++-
> >  1 files changed, 4 insertions(+), 1 deletions(-)
> >
> > diff --git a/meta/classes/image_types.bbclass b/meta/classes/image_types.bbclass
> > index 29b6380..bd4b7bc 100644
> > --- a/meta/classes/image_types.bbclass
> > +++ b/meta/classes/image_types.bbclass
> > @@ -35,6 +35,8 @@ XZ_COMPRESSION_LEVEL ?= "-e -9"
> >  XZ_INTEGRITY_CHECK ?= "crc32"
> >
> >  IMAGE_CMD_jffs2 = "mkfs.jffs2 --root=${IMAGE_ROOTFS} --faketime --output=${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.jffs2 ${EXTRA_IMAGECMD}"
> > +IMAGE_CMD_sum.jffs2 = "${IMAGE_CMD_jffs2} && sumtool -i ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.jffs2 \
> > +       -o ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.sum.jffs2 ${EXTRA_IMAGECMD}"
> >
> >  IMAGE_CMD_cramfs = "mkcramfs ${IMAGE_ROOTFS} ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.cramfs ${EXTRA_IMAGECMD}"
> >
> > @@ -138,6 +140,7 @@ EXTRA_IMAGECMD_btrfs ?= ""
> >
> >  IMAGE_DEPENDS = ""
> >  IMAGE_DEPENDS_jffs2 = "mtd-utils-native"
> > +IMAGE_DEPENDS_sum.jffs2 = "mtd-utils-native"
> >  IMAGE_DEPENDS_cramfs = "cramfs-native"
> >  IMAGE_DEPENDS_ext2 = "genext2fs-native"
> >  IMAGE_DEPENDS_ext2.gz = "genext2fs-native"
> > @@ -157,4 +160,4 @@ IMAGE_DEPENDS_ubi = "mtd-utils-native"
> >  IMAGE_DEPENDS_ubifs = "mtd-utils-native"
> >
> >  # This variable is available to request which values are suitable for IMAGE_FSTYPES
> > -IMAGE_TYPES = "jffs2 cramfs ext2 ext2.gz ext2.bz2 ext3 ext3.gz ext2.lzma live squashfs squashfs-lzma ubi tar tar.gz tar.bz2 tar.xz cpio cpio.gz cpio.xz cpio.lzma"
> > +IMAGE_TYPES = "jffs2 sum.jffs2 cramfs ext2 ext2.gz ext2.bz2 ext3 ext3.gz ext2.lzma live squashfs squashfs-lzma ubi tar tar.gz tar.bz2 tar.xz cpio cpio.gz cpio.xz cpio.lzma"
> > --
> > 1.7.3.4
> >
> >
> > _______________________________________________
> > Openembedded-core mailing list
> > Openembedded-core@lists.openembedded.org
> > http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
> 
> 
> 
> -- 
> Tom
> 
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core

-- 
Martin 'JaMa' Jansa     jabber: Martin.Jansa@gmail.com

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 205 bytes --]

^ permalink raw reply	[flat|nested] 14+ messages in thread
* Re: [PATCH 0/2] rework jffs2 images creation, add jffs2 summary
  2011-12-06 23:23 [PATCH 0/2] rework jffs2 images creation, add jffs2 summary Andrea Adami
  2011-12-06 23:23 ` [PATCH 1/2] image_types.bbclass: implement jffs2 summary images (sum.jffs2) Andrea Adami
  2011-12-06 23:23 ` [PATCH 2/2] image_types.bbclass: disable cleanmarkers for jffs2 images Andrea Adami
@ 2011-12-08 18:22 ` Saul Wold
  2 siblings, 0 replies; 14+ messages in thread
From: Saul Wold @ 2011-12-08 18:22 UTC (permalink / raw)
  To: Patches and discussions about the oe-core layer

On 12/06/2011 03:23 PM, Andrea Adami wrote:
> * The first patch adds support for the creation of
> * jffs2 summary images for a faster mount.
> * The second patch forces the --no-cleanmarkers option
> * to further improve boot time.
>
> Andrea Adami (2):
>    image_types.bbclass: implement jffs2 summary images (sum.jffs2)
>    image_types.bbclass: disable cleanmarkers for jffs2 images
>
>   meta/classes/image_types.bbclass |    7 +++++--
>   1 files changed, 5 insertions(+), 2 deletions(-)
>
Merged to OE-Core

Thanks
	Sau!



^ permalink raw reply	[flat|nested] 14+ messages in thread
* Re: [PATCH 1/2] image_types.bbclass: implement jffs2 summary images (sum.jffs2)
  2011-12-08 15:10     ` Martin Jansa
@ 2011-12-08 21:30       ` Andrea Adami
  2011-12-08 22:43         ` Ulf Samuelsson
  2011-12-11 10:03         ` Martin Jansa
  0 siblings, 2 replies; 14+ messages in thread
From: Andrea Adami @ 2011-12-08 21:30 UTC (permalink / raw)
  To: Patches and discussions about the oe-core layer

On Thu, Dec 8, 2011 at 4:10 PM, Martin Jansa <martin.jansa@gmail.com> wrote:
> On Thu, Dec 08, 2011 at 07:57:52AM -0700, Tom Rini wrote:
>> On Tue, Dec 6, 2011 at 4:23 PM, Andrea Adami <andrea.adami@gmail.com> wrote:
>> > * Building the jffs2 filesystem to include summary information speeds up
>> > * the mount process considerably at the cost of increased size.
>> > * The rate of speedup is generally higher on NAND chips and on the chips
>> > * where the erase block size is large.
>> >
>> > Signed-off-by: Andrea Adami <andrea.adami@gmail.com>
>
> I'll repeat my comment from yesterday, but IMHO it would be easier to
> see the difference between those 2 images if the IMAGE_CMD_jffs2 output is
> renamed to .jffs2.nosummary and IMAGE_CMD_sum.jffs2 (or renamed IMAGE_CMD_jffs2.summary)
> produces .jffs2.summary (or even just .jffs2 if we can agree that .summary is
> what enduser wants by default)

I tried to keep the patch not too invasive and did not introduce a
special EXTRA string for sumtool.
The most widely used options (-p -l -e) are common, though.

Unfortunately summary means we have a bigger image:
44M Dec  8 16:14 core-image-sato-c7x0-20111208151047.rootfs.jffs2
57M Dec  8 16:14 core-image-sato-c7x0-20111208151047.rootfs.sum.jffs2

Being some devices have strict size requirements I think it is better
to build both kind of images.

Cheers

Andrea

>
> But other than that it looks fine and will make my IMAGE_DEPENDS_jffs2 a
> bit shorter.
>
> Regards,
>
>>
>> Acked-by: Tom Rini <tom.rini@gmail.com>
>>
>> > ---
>> >  meta/classes/image_types.bbclass |    5 ++++-
>> >  1 files changed, 4 insertions(+), 1 deletions(-)
>> >
>> > diff --git a/meta/classes/image_types.bbclass b/meta/classes/image_types.bbclass
>> > index 29b6380..bd4b7bc 100644
>> > --- a/meta/classes/image_types.bbclass
>> > +++ b/meta/classes/image_types.bbclass
>> > @@ -35,6 +35,8 @@ XZ_COMPRESSION_LEVEL ?= "-e -9"
>> >  XZ_INTEGRITY_CHECK ?= "crc32"
>> >
>> >  IMAGE_CMD_jffs2 = "mkfs.jffs2 --root=${IMAGE_ROOTFS} --faketime --output=${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.jffs2 ${EXTRA_IMAGECMD}"
>> > +IMAGE_CMD_sum.jffs2 = "${IMAGE_CMD_jffs2} && sumtool -i ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.jffs2 \
>> > +       -o ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.sum.jffs2 ${EXTRA_IMAGECMD}"
>> >
>> >  IMAGE_CMD_cramfs = "mkcramfs ${IMAGE_ROOTFS} ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.cramfs ${EXTRA_IMAGECMD}"
>> >
>> > @@ -138,6 +140,7 @@ EXTRA_IMAGECMD_btrfs ?= ""
>> >
>> >  IMAGE_DEPENDS = ""
>> >  IMAGE_DEPENDS_jffs2 = "mtd-utils-native"
>> > +IMAGE_DEPENDS_sum.jffs2 = "mtd-utils-native"
>> >  IMAGE_DEPENDS_cramfs = "cramfs-native"
>> >  IMAGE_DEPENDS_ext2 = "genext2fs-native"
>> >  IMAGE_DEPENDS_ext2.gz = "genext2fs-native"
>> > @@ -157,4 +160,4 @@ IMAGE_DEPENDS_ubi = "mtd-utils-native"
>> >  IMAGE_DEPENDS_ubifs = "mtd-utils-native"
>> >
>> >  # This variable is available to request which values are suitable for IMAGE_FSTYPES
>> > -IMAGE_TYPES = "jffs2 cramfs ext2 ext2.gz ext2.bz2 ext3 ext3.gz ext2.lzma live squashfs squashfs-lzma ubi tar tar.gz tar.bz2 tar.xz cpio cpio.gz cpio.xz cpio.lzma"
>> > +IMAGE_TYPES = "jffs2 sum.jffs2 cramfs ext2 ext2.gz ext2.bz2 ext3 ext3.gz ext2.lzma live squashfs squashfs-lzma ubi tar tar.gz tar.bz2 tar.xz cpio cpio.gz cpio.xz cpio.lzma"
>> > --
>> > 1.7.3.4
>> >
>> >
>> > _______________________________________________
>> > Openembedded-core mailing list
>> > Openembedded-core@lists.openembedded.org
>> > http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
>>
>>
>>
>> --
>> Tom
>>
>> _______________________________________________
>> Openembedded-core mailing list
>> Openembedded-core@lists.openembedded.org
>> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
>
> --
> Martin 'JaMa' Jansa     jabber: Martin.Jansa@gmail.com
>
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
>



^ permalink raw reply	[flat|nested] 14+ messages in thread
* Re: [PATCH 1/2] image_types.bbclass: implement jffs2 summary images (sum.jffs2)
  2011-12-08 21:30       ` Andrea Adami
@ 2011-12-08 22:43         ` Ulf Samuelsson
  2011-12-08 22:48           ` Andrea Adami
  2011-12-08 22:56           ` Richard Purdie
  2011-12-11 10:03         ` Martin Jansa
  1 sibling, 2 replies; 14+ messages in thread
From: Ulf Samuelsson @ 2011-12-08 22:43 UTC (permalink / raw)
  To: Patches and discussions about the oe-core layer

On 2011-12-08 22:30, Andrea Adami wrote:
> On Thu, Dec 8, 2011 at 4:10 PM, Martin Jansa<martin.jansa@gmail.com>  wrote:
>> On Thu, Dec 08, 2011 at 07:57:52AM -0700, Tom Rini wrote:
>>> On Tue, Dec 6, 2011 at 4:23 PM, Andrea Adami<andrea.adami@gmail.com>  wrote:
>>>> * Building the jffs2 filesystem to include summary information speeds up
>>>> * the mount process considerably at the cost of increased size.
>>>> * The rate of speedup is generally higher on NAND chips and on the chips
>>>> * where the erase block size is large.
>>>>
>>>> Signed-off-by: Andrea Adami<andrea.adami@gmail.com>
>> I'll repeat my comment from yesterday, but IMHO it would be easier to
>> see the difference between those 2 images if the IMAGE_CMD_jffs2 output is
>> renamed to .jffs2.nosummary and IMAGE_CMD_sum.jffs2 (or renamed IMAGE_CMD_jffs2.summary)
>> produces .jffs2.summary (or even just .jffs2 if we can agree that .summary is
>> what enduser wants by default)
> I tried to keep the patch not too invasive and did not introduce a
> special EXTRA string for sumtool.
> The most widely used options (-p -l -e) are common, though.
>
> Unfortunately summary means we have a bigger image:
> 44M Dec  8 16:14 core-image-sato-c7x0-20111208151047.rootfs.jffs2
> 57M Dec  8 16:14 core-image-sato-c7x0-20111208151047.rootfs.sum.jffs2
>
> Being some devices have strict size requirements I think it is better
> to build both kind of images.
>
Not sure exactly what sumtools does,
but as I understood it, it does things which
the targets need to be doing at first startup otherwise.

Does that not mean that the target rootfs will grow
from 44 MB to 57 MB during that process?

If it does, you never want to have the non-sumtool version.
(I guess if you downloaded the file over a 20 bps channel to
a nuclear submarine, you would disagree)

BR
Ulf




> Cheers
>
> Andrea
>
>> But other than that it looks fine and will make my IMAGE_DEPENDS_jffs2 a
>> bit shorter.
>>
>> Regards,
>>
>>> Acked-by: Tom Rini<tom.rini@gmail.com>
>>>
>>>> ---
>>>>   meta/classes/image_types.bbclass |    5 ++++-
>>>>   1 files changed, 4 insertions(+), 1 deletions(-)
>>>>
>>>> diff --git a/meta/classes/image_types.bbclass b/meta/classes/image_types.bbclass
>>>> index 29b6380..bd4b7bc 100644
>>>> --- a/meta/classes/image_types.bbclass
>>>> +++ b/meta/classes/image_types.bbclass
>>>> @@ -35,6 +35,8 @@ XZ_COMPRESSION_LEVEL ?= "-e -9"
>>>>   XZ_INTEGRITY_CHECK ?= "crc32"
>>>>
>>>>   IMAGE_CMD_jffs2 = "mkfs.jffs2 --root=${IMAGE_ROOTFS} --faketime --output=${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.jffs2 ${EXTRA_IMAGECMD}"
>>>> +IMAGE_CMD_sum.jffs2 = "${IMAGE_CMD_jffs2}&&  sumtool -i ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.jffs2 \
>>>> +       -o ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.sum.jffs2 ${EXTRA_IMAGECMD}"
>>>>
>>>>   IMAGE_CMD_cramfs = "mkcramfs ${IMAGE_ROOTFS} ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.cramfs ${EXTRA_IMAGECMD}"
>>>>
>>>> @@ -138,6 +140,7 @@ EXTRA_IMAGECMD_btrfs ?= ""
>>>>
>>>>   IMAGE_DEPENDS = ""
>>>>   IMAGE_DEPENDS_jffs2 = "mtd-utils-native"
>>>> +IMAGE_DEPENDS_sum.jffs2 = "mtd-utils-native"
>>>>   IMAGE_DEPENDS_cramfs = "cramfs-native"
>>>>   IMAGE_DEPENDS_ext2 = "genext2fs-native"
>>>>   IMAGE_DEPENDS_ext2.gz = "genext2fs-native"
>>>> @@ -157,4 +160,4 @@ IMAGE_DEPENDS_ubi = "mtd-utils-native"
>>>>   IMAGE_DEPENDS_ubifs = "mtd-utils-native"
>>>>
>>>>   # This variable is available to request which values are suitable for IMAGE_FSTYPES
>>>> -IMAGE_TYPES = "jffs2 cramfs ext2 ext2.gz ext2.bz2 ext3 ext3.gz ext2.lzma live squashfs squashfs-lzma ubi tar tar.gz tar.bz2 tar.xz cpio cpio.gz cpio.xz cpio.lzma"
>>>> +IMAGE_TYPES = "jffs2 sum.jffs2 cramfs ext2 ext2.gz ext2.bz2 ext3 ext3.gz ext2.lzma live squashfs squashfs-lzma ubi tar tar.gz tar.bz2 tar.xz cpio cpio.gz cpio.xz cpio.lzma"
>>>> --
>>>> 1.7.3.4
>>>>
>>>>
>>>> _______________________________________________
>>>> Openembedded-core mailing list
>>>> Openembedded-core@lists.openembedded.org
>>>> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
>>>
>>>
>>> --
>>> Tom
>>>
>>> _______________________________________________
>>> Openembedded-core mailing list
>>> Openembedded-core@lists.openembedded.org
>>> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
>> --
>> Martin 'JaMa' Jansa     jabber: Martin.Jansa@gmail.com
>>
>> _______________________________________________
>> Openembedded-core mailing list
>> Openembedded-core@lists.openembedded.org
>> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
>>
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core


-- 
Best Regards
Ulf Samuelsson
eMagii




^ permalink raw reply	[flat|nested] 14+ messages in thread
* Re: [PATCH 1/2] image_types.bbclass: implement jffs2 summary images (sum.jffs2)
  2011-12-08 22:43         ` Ulf Samuelsson
@ 2011-12-08 22:48           ` Andrea Adami
  2011-12-08 22:56           ` Richard Purdie
  1 sibling, 0 replies; 14+ messages in thread
From: Andrea Adami @ 2011-12-08 22:48 UTC (permalink / raw)
  To: ulf, Patches and discussions about the oe-core layer

On Thu, Dec 8, 2011 at 11:43 PM, Ulf Samuelsson
<openembedded-core@emagii.com> wrote:
> On 2011-12-08 22:30, Andrea Adami wrote:
>>
>> On Thu, Dec 8, 2011 at 4:10 PM, Martin Jansa<martin.jansa@gmail.com>
>>  wrote:
>>>
>>> On Thu, Dec 08, 2011 at 07:57:52AM -0700, Tom Rini wrote:
>>>>
>>>> On Tue, Dec 6, 2011 at 4:23 PM, Andrea Adami<andrea.adami@gmail.com>
>>>>  wrote:
>>>>>
>>>>> * Building the jffs2 filesystem to include summary information speeds
>>>>> up
>>>>> * the mount process considerably at the cost of increased size.
>>>>> * The rate of speedup is generally higher on NAND chips and on the
>>>>> chips
>>>>> * where the erase block size is large.
>>>>>
>>>>> Signed-off-by: Andrea Adami<andrea.adami@gmail.com>
>>>
>>> I'll repeat my comment from yesterday, but IMHO it would be easier to
>>> see the difference between those 2 images if the IMAGE_CMD_jffs2 output
>>> is
>>> renamed to .jffs2.nosummary and IMAGE_CMD_sum.jffs2 (or renamed
>>> IMAGE_CMD_jffs2.summary)
>>> produces .jffs2.summary (or even just .jffs2 if we can agree that
>>> .summary is
>>> what enduser wants by default)
>>
>> I tried to keep the patch not too invasive and did not introduce a
>> special EXTRA string for sumtool.
>> The most widely used options (-p -l -e) are common, though.
>>
>> Unfortunately summary means we have a bigger image:
>> 44M Dec  8 16:14 core-image-sato-c7x0-20111208151047.rootfs.jffs2
>> 57M Dec  8 16:14 core-image-sato-c7x0-20111208151047.rootfs.sum.jffs2
>>
>> Being some devices have strict size requirements I think it is better
>> to build both kind of images.
>>
> Not sure exactly what sumtools does,
> but as I understood it, it does things which
> the targets need to be doing at first startup otherwise.

No, this operation cannot be done runtime.

http://www.embedded-linux.co.uk/tutorial/jffs2-summary

"
The tools that does that, mkfs.jffs2, does not do anything about
summary nodes, or to put it another way, by default you will end up
with an image that takes a long time to mount. The secret is that you
have to use a second tool from mtdutils: sumtool.
"

Regards

Andrea

>
> Does that not mean that the target rootfs will grow
> from 44 MB to 57 MB during that process?
>
> If it does, you never want to have the non-sumtool version.
> (I guess if you downloaded the file over a 20 bps channel to
> a nuclear submarine, you would disagree)
>
> BR
> Ulf
>
>
>
>
>
>> Cheers
>>
>> Andrea
>>
>>> But other than that it looks fine and will make my IMAGE_DEPENDS_jffs2 a
>>> bit shorter.
>>>
>>> Regards,
>>>
>>>> Acked-by: Tom Rini<tom.rini@gmail.com>
>>>>
>>>>> ---
>>>>>  meta/classes/image_types.bbclass |    5 ++++-
>>>>>  1 files changed, 4 insertions(+), 1 deletions(-)
>>>>>
>>>>> diff --git a/meta/classes/image_types.bbclass
>>>>> b/meta/classes/image_types.bbclass
>>>>> index 29b6380..bd4b7bc 100644
>>>>> --- a/meta/classes/image_types.bbclass
>>>>> +++ b/meta/classes/image_types.bbclass
>>>>> @@ -35,6 +35,8 @@ XZ_COMPRESSION_LEVEL ?= "-e -9"
>>>>>  XZ_INTEGRITY_CHECK ?= "crc32"
>>>>>
>>>>>  IMAGE_CMD_jffs2 = "mkfs.jffs2 --root=${IMAGE_ROOTFS} --faketime
>>>>> --output=${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.jffs2 ${EXTRA_IMAGECMD}"
>>>>> +IMAGE_CMD_sum.jffs2 = "${IMAGE_CMD_jffs2}&&  sumtool -i
>>>>> ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.jffs2 \
>>>>> +       -o ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.sum.jffs2
>>>>> ${EXTRA_IMAGECMD}"
>>>>>
>>>>>  IMAGE_CMD_cramfs = "mkcramfs ${IMAGE_ROOTFS}
>>>>> ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.cramfs ${EXTRA_IMAGECMD}"
>>>>>
>>>>> @@ -138,6 +140,7 @@ EXTRA_IMAGECMD_btrfs ?= ""
>>>>>
>>>>>  IMAGE_DEPENDS = ""
>>>>>  IMAGE_DEPENDS_jffs2 = "mtd-utils-native"
>>>>> +IMAGE_DEPENDS_sum.jffs2 = "mtd-utils-native"
>>>>>  IMAGE_DEPENDS_cramfs = "cramfs-native"
>>>>>  IMAGE_DEPENDS_ext2 = "genext2fs-native"
>>>>>  IMAGE_DEPENDS_ext2.gz = "genext2fs-native"
>>>>> @@ -157,4 +160,4 @@ IMAGE_DEPENDS_ubi = "mtd-utils-native"
>>>>>  IMAGE_DEPENDS_ubifs = "mtd-utils-native"
>>>>>
>>>>>  # This variable is available to request which values are suitable for
>>>>> IMAGE_FSTYPES
>>>>> -IMAGE_TYPES = "jffs2 cramfs ext2 ext2.gz ext2.bz2 ext3 ext3.gz
>>>>> ext2.lzma live squashfs squashfs-lzma ubi tar tar.gz tar.bz2 tar.xz cpio
>>>>> cpio.gz cpio.xz cpio.lzma"
>>>>> +IMAGE_TYPES = "jffs2 sum.jffs2 cramfs ext2 ext2.gz ext2.bz2 ext3
>>>>> ext3.gz ext2.lzma live squashfs squashfs-lzma ubi tar tar.gz tar.bz2 tar.xz
>>>>> cpio cpio.gz cpio.xz cpio.lzma"
>>>>> --
>>>>> 1.7.3.4
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> Openembedded-core mailing list
>>>>> Openembedded-core@lists.openembedded.org
>>>>> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
>>>>
>>>>
>>>>
>>>> --
>>>> Tom
>>>>
>>>> _______________________________________________
>>>> Openembedded-core mailing list
>>>> Openembedded-core@lists.openembedded.org
>>>> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
>>>
>>> --
>>> Martin 'JaMa' Jansa     jabber: Martin.Jansa@gmail.com
>>>
>>> _______________________________________________
>>> Openembedded-core mailing list
>>> Openembedded-core@lists.openembedded.org
>>> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
>>>
>> _______________________________________________
>> Openembedded-core mailing list
>> Openembedded-core@lists.openembedded.org
>> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
>
>
>
> --
> Best Regards
> Ulf Samuelsson
> eMagii
>
>
>
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core



^ permalink raw reply	[flat|nested] 14+ messages in thread
* Re: [PATCH 1/2] image_types.bbclass: implement jffs2 summary images (sum.jffs2)
  2011-12-08 22:43         ` Ulf Samuelsson
  2011-12-08 22:48           ` Andrea Adami
@ 2011-12-08 22:56           ` Richard Purdie
  1 sibling, 0 replies; 14+ messages in thread
From: Richard Purdie @ 2011-12-08 22:56 UTC (permalink / raw)
  To: ulf, Patches and discussions about the oe-core layer

On Thu, 2011-12-08 at 23:43 +0100, Ulf Samuelsson wrote:
> Not sure exactly what sumtools does,
> but as I understood it, it does things which
> the targets need to be doing at first startup otherwise.
> 
> Does that not mean that the target rootfs will grow
> from 44 MB to 57 MB during that process?
> 
> If it does, you never want to have the non-sumtool version.
> (I guess if you downloaded the file over a 20 bps channel to
> a nuclear submarine, you would disagree)

It doesn't. It will compute all this data into memory at each boot
increasing boot time but it won't write it back to the flash.

The sum images just mean the data can be directly loaded without
computation and reading a lot more of the flash.

Cheers,

Richard






^ permalink raw reply	[flat|nested] 14+ messages in thread
* Re: [PATCH 1/2] image_types.bbclass: implement jffs2 summary images (sum.jffs2)
  2011-12-08 21:30       ` Andrea Adami
  2011-12-08 22:43         ` Ulf Samuelsson
@ 2011-12-11 10:03         ` Martin Jansa
  2011-12-11 22:20           ` Andrea Adami
  1 sibling, 1 reply; 14+ messages in thread
From: Martin Jansa @ 2011-12-11 10:03 UTC (permalink / raw)
  To: Patches and discussions about the oe-core layer

[-- Attachment #1: Type: text/plain, Size: 21364 bytes --]

On Thu, Dec 08, 2011 at 10:30:27PM +0100, Andrea Adami wrote:
> On Thu, Dec 8, 2011 at 4:10 PM, Martin Jansa <martin.jansa@gmail.com> wrote:
> > On Thu, Dec 08, 2011 at 07:57:52AM -0700, Tom Rini wrote:
> >> On Tue, Dec 6, 2011 at 4:23 PM, Andrea Adami <andrea.adami@gmail.com> wrote:
> >> > * Building the jffs2 filesystem to include summary information speeds up
> >> > * the mount process considerably at the cost of increased size.
> >> > * The rate of speedup is generally higher on NAND chips and on the chips
> >> > * where the erase block size is large.
> >> >
> >> > Signed-off-by: Andrea Adami <andrea.adami@gmail.com>
> >
> > I'll repeat my comment from yesterday, but IMHO it would be easier to
> > see the difference between those 2 images if the IMAGE_CMD_jffs2 output is
> > renamed to .jffs2.nosummary and IMAGE_CMD_sum.jffs2 (or renamed IMAGE_CMD_jffs2.summary)
> > produces .jffs2.summary (or even just .jffs2 if we can agree that .summary is
> > what enduser wants by default)
> 
> I tried to keep the patch not too invasive and did not introduce a
> special EXTRA string for sumtool.
> The most widely used options (-p -l -e) are common, though.
> 
> Unfortunately summary means we have a bigger image:
> 44M Dec  8 16:14 core-image-sato-c7x0-20111208151047.rootfs.jffs2
> 57M Dec  8 16:14 core-image-sato-c7x0-20111208151047.rootfs.sum.jffs2
> 
> Being some devices have strict size requirements I think it is better
> to build both kind of images.

Today I've noticed that it doesn't work for me:

I have changed EXTRA_IMAGECMD_jffs2 to use new sum.jffs2:
http://git.shr-project.org/git/?p=meta-smartphone.git;a=blobdiff;f=meta-openmoko/conf/machine/om-gta02.conf;h=ad71d3db386b7d7df6ab484ea9ad80260da30373;hp=7cfb1b2a07b09ac3a35a27d209015ea8533ff632;hb=26d60e47358abe66d7d0fdbe6d5adea8f46bdbe2;hpb=043b5ac84794946e88e880b722fed445d3d985c5

OE om-gta02@shr ~/shr-core $ bitbake -e shr-lite-image | tee -a image.log

OE om-gta02@shr ~/shr-core $ grep ^EXTRA_IMAGECMD image.log | grep jffs2
EXTRA_IMAGECMD_sum.jffs2="--eraseblock=0x20000 --no-cleanmarkers --littleendian --pad"
EXTRA_IMAGECMD_jffs2="--little-endian --eraseblock=0x20000 --pagesize=0x800 --no-cleanmarkers --pad; mv /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-lite-20111211-om-gta02.rootfs.jffs2 /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-lite-20111211-om-gta02.rootfs.jffs2.nosummary;"

This looks right (notice that sumtool is using different --little-endian/--littleendian).

OE om-gta02@shr ~/shr-core $ grep IMAGE_CMD image.log | grep jffs2
# IMAGE_CMD_sum.jffs2=${IMAGE_CMD_jffs2} && sumtool -i ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.jffs2   -o ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.sum.jffs2 -n ${EXTRA_IMAGECMD}
IMAGE_CMD_sum.jffs2="mkfs.jffs2 --root=/OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/shr-lite-image/2.0-r20/rootfs --faketime --output=/OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-lite-20111211-om-gta02.rootfs.jffs2 -n  && sumtool -i /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-lite-20111211-om-gta02.rootfs.jffs2      -o /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-lite-20111211-om-gta02.rootfs.sum.jffs2 -n"
# IMAGE_CMD_jffs2=mkfs.jffs2 --root=${IMAGE_ROOTFS} --faketime --output=${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.jffs2 -n ${EXTRA_IMAGECMD}
IMAGE_CMD_jffs2="mkfs.jffs2 --root=/OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/shr-lite-image/2.0-r20/rootfs --faketime --output=/OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-lite-20111211-om-gta02.rootfs.jffs2 -n"

Here we can see that bitbake -e didn't expand EXTRA_IMAGECMD right, but it seems like do_rootfs have them not empty:

shr-core/tmp/work/om_gta02-oe-linux-gnueabi/shr-image/2.0-r18/temp/run.do_rootfs.688:
        # Image generation code for image type sum.jffs2
        ROOTFS_SIZE=`du -ks /OE/shr-core/tmp/work/om_gta02-oe-linux-gnueabi/shr-image/2.0-r18/rootfs|awk '{size = $1 * 1.3 + 0; OFMT = "%.0f" ; print (size > 65536 ? size : 65536) }'`
        mkfs.jffs2 --root=/OE/shr-core/tmp/work/om_gta02-oe-linux-gnueabi/shr-image/2.0-r18/rootfs --faketime --output=/OE/shr-core/tmp/deploy/images/om-gta02/shr-full-20111210-om-gta02.rootfs.jffs2 -n --eraseblock=0x20000 --no-cleanmarkers --littleendian --pad  && sumtool -i /OE/shr-core/tmp/deploy/images/om-gta02/shr-full-20111210-om-gta02.rootfs.jffs2    -o /OE/shr-core/tmp/deploy/images/om-gta02/shr-full-20111210-om-gta02.rootfs.sum.jffs2 -n --eraseblock=0x20000 --no-cleanmarkers --littleendian --pad
        cd /OE/shr-core/tmp/deploy/images/om-gta02/
        rm -f /OE/shr-core/tmp/deploy/images/om-gta02/full-om-gta02.sum.jffs2
        ln -s shr-full-20111210-om-gta02.rootfs.sum.jffs2 /OE/shr-core/tmp/deploy/images/om-gta02/full-om-gta02.sum.jffs2

Notice that IMAGE_CMD_jffs2 is using EXTRA_IMAGECMD_sum.jffs2 (so params for sumtool not mkfs.jffs2) when it's expanded 
inside IMAGE_CMD_sum.jffs2 variable

openembedded-core/meta/classes/image_types.bbclass:
IMAGE_CMD_jffs2 = "mkfs.jffs2 --root=${IMAGE_ROOTFS} --faketime --output=${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.jffs2 -n ${EXTRA_IMAGECMD}"
IMAGE_CMD_sum.jffs2 = "${IMAGE_CMD_jffs2} && sumtool -i ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.jffs2 \
        -o ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.sum.jffs2 -n ${EXTRA_IMAGECMD}"

And causing
mkfs.jffs2: unrecognized option '--littleendian'
mkfs.jffs2: error!: Usage: mkfs.jffs2 [OPTIONS]

And using EXTRA_IMAGECMD_jffs2 in IMAGE_CMD_jffs2 breaks parsing..

ERROR: Unable to parse /OE/shr-core/openembedded-core/meta/recipes-core/images/core-image-minimal-initramfs.bb                                                                                                               | ETA:  --:--:--
Traceback (most recent call last):
  File "/usr/lib64/python2.7/site-packages/bb/codeparser.py", line 268, in ShellParser.parse_shell(value='\t#set -x\n\trm -rf /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs\n\trm -rf /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/multilib\n\tmkdir -p /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs\n\tmkdir -p /OE/shr-core/tmp-eglibc/deploy/images/om-gta02\n\n\tcp /OE/shr-core/openembedded-core/meta/files/deploydir_readme.txt /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/README_-_DO_NOT_DELETE_FILES_IN_THIS_DIRECTORY.txt\n\n\tif [ "0" != "1" ]; then\n\t\tfor devtable in  /OE/shr-core/meta-handheld/files/device_table-minimal.txt; do\n\t\t\tmakedevs -r /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs -D $devtable\n\t\tdone\n\tfi\n\n\trootfs_ipk_do_rootfs\n\n\tinsert_feed_uris\n\n\tif [ "xldconfig-native:do_populate_sysroot" != "x" ]; then\n\t\t# Run ldconfig on the image to create a valid cache\n\t\t# (new format for cross arch compatibility)\n\t\techo executing: ldconfig -r /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs -c new -v\n\t\tldconfig -r /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs -c new -v\n\tfi\n\n\t# (re)create kernel modules dependencies\n\t# This part is done by kernel-module-* postinstall scripts but if image do\n\t# not contains modules at all there are few moments in boot sequence with\n\t# "unable to open modules.dep" message.\n\tif [ -e /OE/shr-core/tmp-eglibc/sysroots/om-gta02/kernel/kernel-abiversion ]; then\n\t\tKERNEL_VERSION=`cat /OE/shr-core/tmp-eglibc/sysroots/om-gta02/kernel/kernel-abiversion`\n\n\t\tmkdir -p /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs/lib/modules/$KERNEL_VERSION\n\t\tarm-oe-linux-gnueabi-depmod -a -b /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs -F /OE/shr-core/tmp-eglibc/sysroots/om-gta02/kernel/System.map-$KERNEL_VERSION $KERNEL_VERSION\n\tfi\n\n\t${IMAGE_PREPROCESS_COMMAND}\n\n\t\n\t# Image generation code for image type cpio.gz\n\tROOTFS_SIZE=`du -ks /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs|awk \'{size = $1 * 1.3 + 0; OFMT = "%.0f" ; print (size > 8192 ? size : 8192) }\'`\n\tcd /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs && (find . | cpio -o -H newc | gzip -c -9 >/OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.cpio.gz)\n\tcd /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/\n\trm -f /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/core-image-minimal-initramfs-om-gta02.cpio.gz\n\tln -s shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.cpio.gz /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/core-image-minimal-initramfs-om-gta02.cpio.gz\n\t# Image generation code for image type tar.gz\n\tROOTFS_SIZE=`du -ks /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs|awk \'{size = $1 * 1.3 + 0; OFMT = "%.0f" ; print (size > 8192 ? size : 8192) }\'`\n\tcd /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs && tar -zcvf /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.tar.gz .\n\tcd /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/\n\trm -f /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/core-image-minimal-initramfs-om-gta02.tar.gz\n\tln -s shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.tar.gz /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/core-image-minimal-initramfs-om-gta02.tar.gz\n\t# Image generation code for image type sum.jffs2\n\tROOTFS_SIZE=`du -ks /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs|awk \'{size = $1 * 1.3 + 0; OFMT = "%.0f" ; print (size > 8192 ? size : 8192) }\'`\n\tmkfs.jffs2 --root=/OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs --faketime --output=/OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.jffs2 -n --little-endian --eraseblock=0x20000 --pagesize=0x800 --no-cleanmarkers --pad; mv /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.jffs2 /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.jffs2.nosummary; && sumtool -i /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.jffs2 \t-o /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.sum.jffs2 -n --eraseblock=0x20000 --no-cleanmarkers --littleendian --pad \n\tcd /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/\n\trm -f /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/core-image-minimal-initramfs-om-gta02.sum.jffs2\n\tln -s shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.sum.jffs2 /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/core-image-minimal-initramfs-om-gta02.sum.jffs2\n\n\n\t   do_testlab ;\n\n\t\n'):
             try:
    >            tokens, _ = pyshyacc.parse(value, eof=True, debug=False)
             except pyshlex.NeedMore:
  File "/usr/lib64/python2.7/site-packages/bb/pysh/pyshyacc.py", line 673, in parse(input='\t#set -x\n\trm -rf /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs\n\trm -rf /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/multilib\n\tmkdir -p /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs\n\tmkdir -p /OE/shr-core/tmp-eglibc/deploy/images/om-gta02\n\n\tcp /OE/shr-core/openembedded-core/meta/files/deploydir_readme.txt /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/README_-_DO_NOT_DELETE_FILES_IN_THIS_DIRECTORY.txt\n\n\tif [ "0" != "1" ]; then\n\t\tfor devtable in  /OE/shr-core/meta-handheld/files/device_table-minimal.txt; do\n\t\t\tmakedevs -r /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs -D $devtable\n\t\tdone\n\tfi\n\n\trootfs_ipk_do_rootfs\n\n\tinsert_feed_uris\n\n\tif [ "xldconfig-native:do_populate_sysroot" != "x" ]; then\n\t\t# Run ldconfig on the image to create a valid cache\n\t\t# (new format for cross arch compatibility)\n\t\techo executing: ldconfig -r /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs -c new -v\n\t\tldconfig -r /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs -c new -v\n\tfi\n\n\t# (re)create kernel modules dependencies\n\t# This part is done by kernel-module-* postinstall scripts but if image do\n\t# not contains modules at all there are few moments in boot sequence with\n\t# "unable to open modules.dep" message.\n\tif [ -e /OE/shr-core/tmp-eglibc/sysroots/om-gta02/kernel/kernel-abiversion ]; then\n\t\tKERNEL_VERSION=`cat /OE/shr-core/tmp-eglibc/sysroots/om-gta02/kernel/kernel-abiversion`\n\n\t\tmkdir -p /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs/lib/modules/$KERNEL_VERSION\n\t\tarm-oe-linux-gnueabi-depmod -a -b /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs -F /OE/shr-core/tmp-eglibc/sysroots/om-gta02/kernel/System.map-$KERNEL_VERSION $KERNEL_VERSION\n\tfi\n\n\t${IMAGE_PREPROCESS_COMMAND}\n\n\t\n\t# Image generation code for image type cpio.gz\n\tROOTFS_SIZE=`du -ks /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs|awk \'{size = $1 * 1.3 + 0; OFMT = "%.0f" ; print (size > 8192 ? size : 8192) }\'`\n\tcd /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs && (find . | cpio -o -H newc | gzip -c -9 >/OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.cpio.gz)\n\tcd /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/\n\trm -f /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/core-image-minimal-initramfs-om-gta02.cpio.gz\n\tln -s shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.cpio.gz /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/core-image-minimal-initramfs-om-gta02.cpio.gz\n\t# Image generation code for image type tar.gz\n\tROOTFS_SIZE=`du -ks /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs|awk \'{size = $1 * 1.3 + 0; OFMT = "%.0f" ; print (size > 8192 ? size : 8192) }\'`\n\tcd /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs && tar -zcvf /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.tar.gz .\n\tcd /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/\n\trm -f /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/core-image-minimal-initramfs-om-gta02.tar.gz\n\tln -s shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.tar.gz /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/core-image-minimal-initramfs-om-gta02.tar.gz\n\t# Image generation code for image type sum.jffs2\n\tROOTFS_SIZE=`du -ks /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs|awk \'{size = $1 * 1.3 + 0; OFMT = "%.0f" ; print (size > 8192 ? size : 8192) }\'`\n\tmkfs.jffs2 --root=/OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs --faketime --output=/OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.jffs2 -n --little-endian --eraseblock=0x20000 --pagesize=0x800 --no-cleanmarkers --pad; mv /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.jffs2 /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.jffs2.nosummary; && sumtool -i /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.jffs2 \t-o /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.sum.jffs2 -n --eraseblock=0x20000 --no-cleanmarkers --littleendian --pad \n\tcd /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/\n\trm -f /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/core-image-minimal-initramfs-om-gta02.sum.jffs2\n\tln -s shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.sum.jffs2 /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/core-image-minimal-initramfs-om-gta02.sum.jffs2\n\n\n\t   do_testlab ;\n\n\t\n', eof=True, debug=False):
             debug = 2
    >    return yacc.parse(lexer=lexer, debug=debug), remaining

  File "/usr/lib64/python2.7/site-packages/ply/yacc.py", line 265, in LRParser.parse(input=None, lexer=<bb.pysh.pyshlex.PLYLexer instance at 0x9638248>, debug=False, tracking=0, tokenfunc=None):
             else:
    >            return self.parseopt_notrack(input,lexer,debug,tracking,tokenfunc)

  File "/usr/lib64/python2.7/site-packages/ply/yacc.py", line 1047, in LRParser.parseopt_notrack(input=None, lexer=<bb.pysh.pyshlex.PLYLexer instance at 0x9638248>, debug=False, tracking=0, tokenfunc=None):
                                 errtoken.lexer = lexer
    >                        tok = self.errorfunc(errtoken)
                             del errok, token, restart   # Delete special functions
  File "/usr/lib64/python2.7/site-packages/bb/pysh/pyshyacc.py", line 646, in p_error(p=LexToken(AND_IF,'&&',0,0)):
             w('  %r\n' % n)
    >    raise sherrors.ShellSyntaxError(''.join(msg))

ShellSyntaxError: LexToken(AND_IF,'&&',0,0)
followed by:
  LexToken(TOKEN,'sumtool',0,0)
  LexToken(TOKEN,'-i',0,0)
  LexToken(TOKEN,'/OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.jffs2',0,0)
  LexToken(TOKEN,'-o',0,0)
  LexToken(TOKEN,'/OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.sum.jffs2',0,0)



> 
> Cheers
> 
> Andrea
> 
> >
> > But other than that it looks fine and will make my IMAGE_DEPENDS_jffs2 a
> > bit shorter.
> >
> > Regards,
> >
> >>
> >> Acked-by: Tom Rini <tom.rini@gmail.com>
> >>
> >> > ---
> >> >  meta/classes/image_types.bbclass |    5 ++++-
> >> >  1 files changed, 4 insertions(+), 1 deletions(-)
> >> >
> >> > diff --git a/meta/classes/image_types.bbclass b/meta/classes/image_types.bbclass
> >> > index 29b6380..bd4b7bc 100644
> >> > --- a/meta/classes/image_types.bbclass
> >> > +++ b/meta/classes/image_types.bbclass
> >> > @@ -35,6 +35,8 @@ XZ_COMPRESSION_LEVEL ?= "-e -9"
> >> >  XZ_INTEGRITY_CHECK ?= "crc32"
> >> >
> >> >  IMAGE_CMD_jffs2 = "mkfs.jffs2 --root=${IMAGE_ROOTFS} --faketime --output=${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.jffs2 ${EXTRA_IMAGECMD}"
> >> > +IMAGE_CMD_sum.jffs2 = "${IMAGE_CMD_jffs2} && sumtool -i ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.jffs2 \
> >> > +       -o ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.sum.jffs2 ${EXTRA_IMAGECMD}"
> >> >
> >> >  IMAGE_CMD_cramfs = "mkcramfs ${IMAGE_ROOTFS} ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.cramfs ${EXTRA_IMAGECMD}"
> >> >
> >> > @@ -138,6 +140,7 @@ EXTRA_IMAGECMD_btrfs ?= ""
> >> >
> >> >  IMAGE_DEPENDS = ""
> >> >  IMAGE_DEPENDS_jffs2 = "mtd-utils-native"
> >> > +IMAGE_DEPENDS_sum.jffs2 = "mtd-utils-native"
> >> >  IMAGE_DEPENDS_cramfs = "cramfs-native"
> >> >  IMAGE_DEPENDS_ext2 = "genext2fs-native"
> >> >  IMAGE_DEPENDS_ext2.gz = "genext2fs-native"
> >> > @@ -157,4 +160,4 @@ IMAGE_DEPENDS_ubi = "mtd-utils-native"
> >> >  IMAGE_DEPENDS_ubifs = "mtd-utils-native"
> >> >
> >> >  # This variable is available to request which values are suitable for IMAGE_FSTYPES
> >> > -IMAGE_TYPES = "jffs2 cramfs ext2 ext2.gz ext2.bz2 ext3 ext3.gz ext2.lzma live squashfs squashfs-lzma ubi tar tar.gz tar.bz2 tar.xz cpio cpio.gz cpio.xz cpio.lzma"
> >> > +IMAGE_TYPES = "jffs2 sum.jffs2 cramfs ext2 ext2.gz ext2.bz2 ext3 ext3.gz ext2.lzma live squashfs squashfs-lzma ubi tar tar.gz tar.bz2 tar.xz cpio cpio.gz cpio.xz cpio.lzma"
> >> > --
> >> > 1.7.3.4
> >> >
> >> >
> >> > _______________________________________________
> >> > Openembedded-core mailing list
> >> > Openembedded-core@lists.openembedded.org
> >> > http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
> >>
> >>
> >>
> >> --
> >> Tom
> >>
> >> _______________________________________________
> >> Openembedded-core mailing list
> >> Openembedded-core@lists.openembedded.org
> >> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
> >
> > --
> > Martin 'JaMa' Jansa     jabber: Martin.Jansa@gmail.com
> >
> > _______________________________________________
> > Openembedded-core mailing list
> > Openembedded-core@lists.openembedded.org
> > http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
> >
> 
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core

-- 
Martin 'JaMa' Jansa     jabber: Martin.Jansa@gmail.com

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 205 bytes --]

^ permalink raw reply	[flat|nested] 14+ messages in thread
* Re: [PATCH 1/2] image_types.bbclass: implement jffs2 summary images (sum.jffs2)
  2011-12-11 10:03         ` Martin Jansa
@ 2011-12-11 22:20           ` Andrea Adami
  2011-12-11 22:57             ` Martin Jansa
  0 siblings, 1 reply; 14+ messages in thread
From: Andrea Adami @ 2011-12-11 22:20 UTC (permalink / raw)
  To: Patches and discussions about the oe-core layer

On Sun, Dec 11, 2011 at 11:03 AM, Martin Jansa <martin.jansa@gmail.com> wrote:
> On Thu, Dec 08, 2011 at 10:30:27PM +0100, Andrea Adami wrote:
>> On Thu, Dec 8, 2011 at 4:10 PM, Martin Jansa <martin.jansa@gmail.com> wrote:
>> > On Thu, Dec 08, 2011 at 07:57:52AM -0700, Tom Rini wrote:
>> >> On Tue, Dec 6, 2011 at 4:23 PM, Andrea Adami <andrea.adami@gmail.com> wrote:
>> >> > * Building the jffs2 filesystem to include summary information speeds up
>> >> > * the mount process considerably at the cost of increased size.
>> >> > * The rate of speedup is generally higher on NAND chips and on the chips
>> >> > * where the erase block size is large.
>> >> >
>> >> > Signed-off-by: Andrea Adami <andrea.adami@gmail.com>
>> >
>> > I'll repeat my comment from yesterday, but IMHO it would be easier to
>> > see the difference between those 2 images if the IMAGE_CMD_jffs2 output is
>> > renamed to .jffs2.nosummary and IMAGE_CMD_sum.jffs2 (or renamed IMAGE_CMD_jffs2.summary)
>> > produces .jffs2.summary (or even just .jffs2 if we can agree that .summary is
>> > what enduser wants by default)
>>
>> I tried to keep the patch not too invasive and did not introduce a
>> special EXTRA string for sumtool.
>> The most widely used options (-p -l -e) are common, though.
>>
>> Unfortunately summary means we have a bigger image:
>> 44M Dec  8 16:14 core-image-sato-c7x0-20111208151047.rootfs.jffs2
>> 57M Dec  8 16:14 core-image-sato-c7x0-20111208151047.rootfs.sum.jffs2
>>
>> Being some devices have strict size requirements I think it is better
>> to build both kind of images.
>
> Today I've noticed that it doesn't work for me:
>
> I have changed EXTRA_IMAGECMD_jffs2 to use new sum.jffs2:
> http://git.shr-project.org/git/?p=meta-smartphone.git;a=blobdiff;f=meta-openmoko/conf/machine/om-gta02.conf;h=ad71d3db386b7d7df6ab484ea9ad80260da30373;hp=7cfb1b2a07b09ac3a35a27d209015ea8533ff632;hb=26d60e47358abe66d7d0fdbe6d5adea8f46bdbe2;hpb=043b5ac84794946e88e880b722fed445d3d985c5
>

Martin,

the idea of the patch was to keep things simple.

After all most machines only specify endianness, eraseblocksize and padding.
The few which are using summary jffs2 in meta-handhels appear to be
Zaurus and hx4700 (I'll send a patch to fix ipaq's too) and fwiw
beagleboard used '-npl' so I think we cover the most common case when
people are setting

EXTRA_IMAGECMD_jffs2 = "-p -l --eraseblock=${ERASEBLOCKSIZE}"

and for the rare case they add  IMAGE_FSTYPES += "tar.gz jffs2 sum.jffs2":
EXTRA_IMAGECMD_sum.jffs2 = "${EXTRA_IMAGECMD_jffs2}"

This happens to work just as intended, without adding another var for
sumtool args.

        mkfs.jffs2
--root=/oe/oe-core/build/tmp-eglibc/work/c7x0-oe-linux-gnueabi/core-image-sato-1.0-r0/rootfs
--faketime --output=/oe/oe-core/build/tmp-eglibc/deploy/images/core-image-sato-c7x0-20111210181230.rootfs.jffs2
-n -p -l --eraseblock=0x4000
        cd /oe/oe-core/build/tmp-eglibc/deploy/images/
        rm -f /oe/oe-core/build/tmp-eglibc/deploy/images/core-image-sato-c7x0.jffs2
        ln -s core-image-sato-c7x0-20111210181230.rootfs.jffs2
/oe/oe-core/build/tmp-eglibc/deploy/images/core-image-sato-c7x0.jffs2
        # Image generation code for image type sum.jffs2
        ROOTFS_SIZE=`du -ks
/oe/oe-core/build/tmp-eglibc/work/c7x0-oe-linux-gnueabi/core-image-sato-1.0-r0/rootfs|awk
'{size = $1 * 1.3 + 0; OFMT = "%.0f" ; print (size > 65536 ? size :
65536) }'`
        mkfs.jffs2
--root=/oe/oe-core/build/tmp-eglibc/work/c7x0-oe-linux-gnueabi/core-image-sato-1.0-r0/rootfs
--faketime --output=/oe/oe-core/build/tmp-eglibc/deploy/images/core-image-sato-c7x0-20111210181230.rootfs.jffs2
-n -p -l --eraseblock=0x4000 && sumtool -i /oe/
oe-core/build/tmp-eglibc/deploy/images/core-image-sato-c7x0-20111210181230.rootfs.jffs2
        -o /oe/oe-core/build/tmp-eglibc/deploy/images/core-image-sato-c7x0-20111210181230.rootfs.sum.jffs2
-n -p -l --eraseblock=0x4000
        cd /oe/oe-core/build/tmp-eglibc/deploy/images/
        rm -f /oe/oe-core/build/tmp-eglibc/deploy/images/core-image-sato-c7x0.sum.jffs2
        ln -s core-image-sato-c7x0-20111210181230.rootfs.sum.jffs2
/oe/oe-core/build/tmp-eglibc/deploy/images/core-image-sato-c7x0.sum.jffs2



Being that IMAGE_CMD_sum.jffs2 cannot be overriden I adapted the
EXTRA_IMAGECMD_jffs2 and simplified it in the way the string can be
used by sumtool.
To have full customization you'd need a weak assignment in image_types.bbclass.


> OE om-gta02@shr ~/shr-core $ bitbake -e shr-lite-image | tee -a image.log
>
> OE om-gta02@shr ~/shr-core $ grep ^EXTRA_IMAGECMD image.log | grep jffs2
> EXTRA_IMAGECMD_sum.jffs2="--eraseblock=0x20000 --no-cleanmarkers --littleendian --pad"
> EXTRA_IMAGECMD_jffs2="--little-endian --eraseblock=0x20000 --pagesize=0x800 --no-cleanmarkers --pad; mv /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-lite-20111211-om-gta02.rootfs.jffs2 /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-lite-20111211-om-gta02.rootfs.jffs2.nosummary;"
>
> This looks right (notice that sumtool is using different --little-endian/--littleendian).

Pls use -l  ;)

>
> OE om-gta02@shr ~/shr-core $ grep IMAGE_CMD image.log | grep jffs2
> # IMAGE_CMD_sum.jffs2=${IMAGE_CMD_jffs2} && sumtool -i ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.jffs2   -o ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.sum.jffs2 -n ${EXTRA_IMAGECMD}
> IMAGE_CMD_sum.jffs2="mkfs.jffs2 --root=/OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/shr-lite-image/2.0-r20/rootfs --faketime --output=/OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-lite-20111211-om-gta02.rootfs.jffs2 -n  && sumtool -i /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-lite-20111211-om-gta02.rootfs.jffs2      -o /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-lite-20111211-om-gta02.rootfs.sum.jffs2 -n"
> # IMAGE_CMD_jffs2=mkfs.jffs2 --root=${IMAGE_ROOTFS} --faketime --output=${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.jffs2 -n ${EXTRA_IMAGECMD}
> IMAGE_CMD_jffs2="mkfs.jffs2 --root=/OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/shr-lite-image/2.0-r20/rootfs --faketime --output=/OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-lite-20111211-om-gta02.rootfs.jffs2 -n"
>
> Here we can see that bitbake -e didn't expand EXTRA_IMAGECMD right, but it seems like do_rootfs have them not empty:
>
> shr-core/tmp/work/om_gta02-oe-linux-gnueabi/shr-image/2.0-r18/temp/run.do_rootfs.688:
>        # Image generation code for image type sum.jffs2
>        ROOTFS_SIZE=`du -ks /OE/shr-core/tmp/work/om_gta02-oe-linux-gnueabi/shr-image/2.0-r18/rootfs|awk '{size = $1 * 1.3 + 0; OFMT = "%.0f" ; print (size > 65536 ? size : 65536) }'`
>        mkfs.jffs2 --root=/OE/shr-core/tmp/work/om_gta02-oe-linux-gnueabi/shr-image/2.0-r18/rootfs --faketime --output=/OE/shr-core/tmp/deploy/images/om-gta02/shr-full-20111210-om-gta02.rootfs.jffs2 -n --eraseblock=0x20000 --no-cleanmarkers --littleendian --pad  && sumtool -i /OE/shr-core/tmp/deploy/images/om-gta02/shr-full-20111210-om-gta02.rootfs.jffs2    -o /OE/shr-core/tmp/deploy/images/om-gta02/shr-full-20111210-om-gta02.rootfs.sum.jffs2 -n --eraseblock=0x20000 --no-cleanmarkers --littleendian --pad
>        cd /OE/shr-core/tmp/deploy/images/om-gta02/
>        rm -f /OE/shr-core/tmp/deploy/images/om-gta02/full-om-gta02.sum.jffs2
>        ln -s shr-full-20111210-om-gta02.rootfs.sum.jffs2 /OE/shr-core/tmp/deploy/images/om-gta02/full-om-gta02.sum.jffs2
>
> Notice that IMAGE_CMD_jffs2 is using EXTRA_IMAGECMD_sum.jffs2 (so params for sumtool not mkfs.jffs2) when it's expanded
> inside IMAGE_CMD_sum.jffs2 variable
>
> openembedded-core/meta/classes/image_types.bbclass:
> IMAGE_CMD_jffs2 = "mkfs.jffs2 --root=${IMAGE_ROOTFS} --faketime --output=${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.jffs2 -n ${EXTRA_IMAGECMD}"
> IMAGE_CMD_sum.jffs2 = "${IMAGE_CMD_jffs2} && sumtool -i ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.jffs2 \
>        -o ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.sum.jffs2 -n ${EXTRA_IMAGECMD}"
>
> And causing
> mkfs.jffs2: unrecognized option '--littleendian'
> mkfs.jffs2: error!: Usage: mkfs.jffs2 [OPTIONS]
>
> And using EXTRA_IMAGECMD_jffs2 in IMAGE_CMD_jffs2 breaks parsing..
>
> ERROR: Unable to parse /OE/shr-core/openembedded-core/meta/recipes-core/images/core-image-minimal-initramfs.bb                                                                                                               | ETA:  --:--:--
> Traceback (most recent call last):
>  File "/usr/lib64/python2.7/site-packages/bb/codeparser.py", line 268, in ShellParser.parse_shell(value='\t#set -x\n\trm -rf /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs\n\trm -rf /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/multilib\n\tmkdir -p /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs\n\tmkdir -p /OE/shr-core/tmp-eglibc/deploy/images/om-gta02\n\n\tcp /OE/shr-core/openembedded-core/meta/files/deploydir_readme.txt /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/README_-_DO_NOT_DELETE_FILES_IN_THIS_DIRECTORY.txt\n\n\tif [ "0" != "1" ]; then\n\t\tfor devtable in  /OE/shr-core/meta-handheld/files/device_table-minimal.txt; do\n\t\t\tmakedevs -r /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs -D $devtable\n\t\tdone\n\tfi\n\n\trootfs_ipk_do_rootfs\n\n\tinsert_feed_uris\n\n\tif [ "xldconfig-native:do_populate_sysroot" != "x" ]; then\n\t\t# Run ldconfig on the image to create a valid cache\n\t\t# (new format for cross arch compatibility)\n\t\techo executing: ldconfig -r /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs -c new -v\n\t\tldconfig -r /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs -c new -v\n\tfi\n\n\t# (re)create kernel modules dependencies\n\t# This part is done by kernel-module-* postinstall scripts but if image do\n\t# not contains modules at all there are few moments in boot sequence with\n\t# "unable to open modules.dep" message.\n\tif [ -e /OE/shr-core/tmp-eglibc/sysroots/om-gta02/kernel/kernel-abiversion ]; then\n\t\tKERNEL_VERSION=`cat /OE/shr-core/tmp-eglibc/sysroots/om-gta02/kernel/kernel-abiversion`\n\n\t\tmkdir -p /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs/lib/modules/$KERNEL_VERSION\n\t\tarm-oe-linux-gnueabi-depmod -a -b /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs -F /OE/shr-core/tmp-eglibc/sysroots/om-gta02/kernel/System.map-$KERNEL_VERSION $KERNEL_VERSION\n\tfi\n\n\t${IMAGE_PREPROCESS_COMMAND}\n\n\t\n\t# Image generation code for image type cpio.gz\n\tROOTFS_SIZE=`du -ks /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs|awk \'{size = $1 * 1.3 + 0; OFMT = "%.0f" ; print (size > 8192 ? size : 8192) }\'`\n\tcd /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs && (find . | cpio -o -H newc | gzip -c -9 >/OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.cpio.gz)\n\tcd /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/\n\trm -f /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/core-image-minimal-initramfs-om-gta02.cpio.gz\n\tln -s shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.cpio.gz /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/core-image-minimal-initramfs-om-gta02.cpio.gz\n\t# Image generation code for image type tar.gz\n\tROOTFS_SIZE=`du -ks /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs|awk \'{size = $1 * 1.3 + 0; OFMT = "%.0f" ; print (size > 8192 ? size : 8192) }\'`\n\tcd /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs && tar -zcvf /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.tar.gz .\n\tcd /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/\n\trm -f /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/core-image-minimal-initramfs-om-gta02.tar.gz\n\tln -s shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.tar.gz /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/core-image-minimal-initramfs-om-gta02.tar.gz\n\t# Image generation code for image type sum.jffs2\n\tROOTFS_SIZE=`du -ks /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs|awk \'{size = $1 * 1.3 + 0; OFMT = "%.0f" ; print (size > 8192 ? size : 8192) }\'`\n\tmkfs.jffs2 --root=/OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs --faketime --output=/OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.jffs2 -n --little-endian --eraseblock=0x20000 --pagesize=0x800 --no-cleanmarkers --pad; mv /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.jffs2 /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.jffs2.nosummary; && sumtool -i /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.jffs2 \t-o /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.sum.jffs2 -n --eraseblock=0x20000 --no-cleanmarkers --littleendian --pad \n\tcd /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/\n\trm -f /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/core-image-minimal-initramfs-om-gta02.sum.jffs2\n\tln -s shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.sum.jffs2 /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/core-image-minimal-initramfs-om-gta02.sum.jffs2\n\n\n\t   do_testlab ;\n\n\t\n'):
>             try:
>    >            tokens, _ = pyshyacc.parse(value, eof=True, debug=False)
>             except pyshlex.NeedMore:
>  File "/usr/lib64/python2.7/site-packages/bb/pysh/pyshyacc.py", line 673, in parse(input='\t#set -x\n\trm -rf /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs\n\trm -rf /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/multilib\n\tmkdir -p /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs\n\tmkdir -p /OE/shr-core/tmp-eglibc/deploy/images/om-gta02\n\n\tcp /OE/shr-core/openembedded-core/meta/files/deploydir_readme.txt /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/README_-_DO_NOT_DELETE_FILES_IN_THIS_DIRECTORY.txt\n\n\tif [ "0" != "1" ]; then\n\t\tfor devtable in  /OE/shr-core/meta-handheld/files/device_table-minimal.txt; do\n\t\t\tmakedevs -r /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs -D $devtable\n\t\tdone\n\tfi\n\n\trootfs_ipk_do_rootfs\n\n\tinsert_feed_uris\n\n\tif [ "xldconfig-native:do_populate_sysroot" != "x" ]; then\n\t\t# Run ldconfig on the image to create a valid cache\n\t\t# (new format for cross arch compatibility)\n\t\techo executing: ldconfig -r /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs -c new -v\n\t\tldconfig -r /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs -c new -v\n\tfi\n\n\t# (re)create kernel modules dependencies\n\t# This part is done by kernel-module-* postinstall scripts but if image do\n\t# not contains modules at all there are few moments in boot sequence with\n\t# "unable to open modules.dep" message.\n\tif [ -e /OE/shr-core/tmp-eglibc/sysroots/om-gta02/kernel/kernel-abiversion ]; then\n\t\tKERNEL_VERSION=`cat /OE/shr-core/tmp-eglibc/sysroots/om-gta02/kernel/kernel-abiversion`\n\n\t\tmkdir -p /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs/lib/modules/$KERNEL_VERSION\n\t\tarm-oe-linux-gnueabi-depmod -a -b /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs -F /OE/shr-core/tmp-eglibc/sysroots/om-gta02/kernel/System.map-$KERNEL_VERSION $KERNEL_VERSION\n\tfi\n\n\t${IMAGE_PREPROCESS_COMMAND}\n\n\t\n\t# Image generation code for image type cpio.gz\n\tROOTFS_SIZE=`du -ks /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs|awk \'{size = $1 * 1.3 + 0; OFMT = "%.0f" ; print (size > 8192 ? size : 8192) }\'`\n\tcd /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs && (find . | cpio -o -H newc | gzip -c -9 >/OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.cpio.gz)\n\tcd /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/\n\trm -f /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/core-image-minimal-initramfs-om-gta02.cpio.gz\n\tln -s shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.cpio.gz /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/core-image-minimal-initramfs-om-gta02.cpio.gz\n\t# Image generation code for image type tar.gz\n\tROOTFS_SIZE=`du -ks /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs|awk \'{size = $1 * 1.3 + 0; OFMT = "%.0f" ; print (size > 8192 ? size : 8192) }\'`\n\tcd /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs && tar -zcvf /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.tar.gz .\n\tcd /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/\n\trm -f /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/core-image-minimal-initramfs-om-gta02.tar.gz\n\tln -s shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.tar.gz /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/core-image-minimal-initramfs-om-gta02.tar.gz\n\t# Image generation code for image type sum.jffs2\n\tROOTFS_SIZE=`du -ks /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs|awk \'{size = $1 * 1.3 + 0; OFMT = "%.0f" ; print (size > 8192 ? size : 8192) }\'`\n\tmkfs.jffs2 --root=/OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs --faketime --output=/OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.jffs2 -n --little-endian --eraseblock=0x20000 --pagesize=0x800 --no-cleanmarkers --pad; mv /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.jffs2 /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.jffs2.nosummary; && sumtool -i /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.jffs2 \t-o /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.sum.jffs2 -n --eraseblock=0x20000 --no-cleanmarkers --littleendian --pad \n\tcd /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/\n\trm -f /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/core-image-minimal-initramfs-om-gta02.sum.jffs2\n\tln -s shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.sum.jffs2 /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/core-image-minimal-initramfs-om-gta02.sum.jffs2\n\n\n\t   do_testlab ;\n\n\t\n', eof=True, debug=False):
>             debug = 2
>    >    return yacc.parse(lexer=lexer, debug=debug), remaining
>
>  File "/usr/lib64/python2.7/site-packages/ply/yacc.py", line 265, in LRParser.parse(input=None, lexer=<bb.pysh.pyshlex.PLYLexer instance at 0x9638248>, debug=False, tracking=0, tokenfunc=None):
>             else:
>    >            return self.parseopt_notrack(input,lexer,debug,tracking,tokenfunc)
>
>  File "/usr/lib64/python2.7/site-packages/ply/yacc.py", line 1047, in LRParser.parseopt_notrack(input=None, lexer=<bb.pysh.pyshlex.PLYLexer instance at 0x9638248>, debug=False, tracking=0, tokenfunc=None):
>                                 errtoken.lexer = lexer
>    >                        tok = self.errorfunc(errtoken)
>                             del errok, token, restart   # Delete special functions
>  File "/usr/lib64/python2.7/site-packages/bb/pysh/pyshyacc.py", line 646, in p_error(p=LexToken(AND_IF,'&&',0,0)):
>             w('  %r\n' % n)
>    >    raise sherrors.ShellSyntaxError(''.join(msg))
>
> ShellSyntaxError: LexToken(AND_IF,'&&',0,0)
> followed by:
>  LexToken(TOKEN,'sumtool',0,0)
>  LexToken(TOKEN,'-i',0,0)
>  LexToken(TOKEN,'/OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.jffs2',0,0)
>  LexToken(TOKEN,'-o',0,0)
>  LexToken(TOKEN,'/OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.sum.jffs2',0,0)
>
>
>
>>
>> Cheers
>>
>> Andrea
>>
>> >
>> > But other than that it looks fine and will make my IMAGE_DEPENDS_jffs2 a
>> > bit shorter.
>> >
>> > Regards,
>> >
>> >>
>> >> Acked-by: Tom Rini <tom.rini@gmail.com>
>> >>
>> >> > ---
>> >> >  meta/classes/image_types.bbclass |    5 ++++-
>> >> >  1 files changed, 4 insertions(+), 1 deletions(-)
>> >> >
>> >> > diff --git a/meta/classes/image_types.bbclass b/meta/classes/image_types.bbclass
>> >> > index 29b6380..bd4b7bc 100644
>> >> > --- a/meta/classes/image_types.bbclass
>> >> > +++ b/meta/classes/image_types.bbclass
>> >> > @@ -35,6 +35,8 @@ XZ_COMPRESSION_LEVEL ?= "-e -9"
>> >> >  XZ_INTEGRITY_CHECK ?= "crc32"
>> >> >
>> >> >  IMAGE_CMD_jffs2 = "mkfs.jffs2 --root=${IMAGE_ROOTFS} --faketime --output=${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.jffs2 ${EXTRA_IMAGECMD}"
>> >> > +IMAGE_CMD_sum.jffs2 = "${IMAGE_CMD_jffs2} && sumtool -i ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.jffs2 \
>> >> > +       -o ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.sum.jffs2 ${EXTRA_IMAGECMD}"
>> >> >
>> >> >  IMAGE_CMD_cramfs = "mkcramfs ${IMAGE_ROOTFS} ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.cramfs ${EXTRA_IMAGECMD}"
>> >> >
>> >> > @@ -138,6 +140,7 @@ EXTRA_IMAGECMD_btrfs ?= ""
>> >> >
>> >> >  IMAGE_DEPENDS = ""
>> >> >  IMAGE_DEPENDS_jffs2 = "mtd-utils-native"
>> >> > +IMAGE_DEPENDS_sum.jffs2 = "mtd-utils-native"
>> >> >  IMAGE_DEPENDS_cramfs = "cramfs-native"
>> >> >  IMAGE_DEPENDS_ext2 = "genext2fs-native"
>> >> >  IMAGE_DEPENDS_ext2.gz = "genext2fs-native"
>> >> > @@ -157,4 +160,4 @@ IMAGE_DEPENDS_ubi = "mtd-utils-native"
>> >> >  IMAGE_DEPENDS_ubifs = "mtd-utils-native"
>> >> >
>> >> >  # This variable is available to request which values are suitable for IMAGE_FSTYPES
>> >> > -IMAGE_TYPES = "jffs2 cramfs ext2 ext2.gz ext2.bz2 ext3 ext3.gz ext2.lzma live squashfs squashfs-lzma ubi tar tar.gz tar.bz2 tar.xz cpio cpio.gz cpio.xz cpio.lzma"
>> >> > +IMAGE_TYPES = "jffs2 sum.jffs2 cramfs ext2 ext2.gz ext2.bz2 ext3 ext3.gz ext2.lzma live squashfs squashfs-lzma ubi tar tar.gz tar.bz2 tar.xz cpio cpio.gz cpio.xz cpio.lzma"
>> >> > --
>> >> > 1.7.3.4
>> >> >
>> >> >
>> >> > _______________________________________________
>> >> > Openembedded-core mailing list
>> >> > Openembedded-core@lists.openembedded.org
>> >> > http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
>> >>
>> >>
>> >>
>> >> --
>> >> Tom
>> >>
>> >> _______________________________________________
>> >> Openembedded-core mailing list
>> >> Openembedded-core@lists.openembedded.org
>> >> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
>> >
>> > --
>> > Martin 'JaMa' Jansa     jabber: Martin.Jansa@gmail.com
>> >
>> > _______________________________________________
>> > Openembedded-core mailing list
>> > Openembedded-core@lists.openembedded.org
>> > http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
>> >
>>
>> _______________________________________________
>> Openembedded-core mailing list
>> Openembedded-core@lists.openembedded.org
>> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
>
> --
> Martin 'JaMa' Jansa     jabber: Martin.Jansa@gmail.com
>
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
>


Regards
Andrea



^ permalink raw reply	[flat|nested] 14+ messages in thread
* Re: [PATCH 1/2] image_types.bbclass: implement jffs2 summary images (sum.jffs2)
  2011-12-11 22:20           ` Andrea Adami
@ 2011-12-11 22:57             ` Martin Jansa
  0 siblings, 0 replies; 14+ messages in thread
From: Martin Jansa @ 2011-12-11 22:57 UTC (permalink / raw)
  To: Patches and discussions about the oe-core layer

[-- Attachment #1: Type: text/plain, Size: 19162 bytes --]

On Sun, Dec 11, 2011 at 11:20:25PM +0100, Andrea Adami wrote:
> On Sun, Dec 11, 2011 at 11:03 AM, Martin Jansa <martin.jansa@gmail.com> wrote:
> > On Thu, Dec 08, 2011 at 10:30:27PM +0100, Andrea Adami wrote:
> >> On Thu, Dec 8, 2011 at 4:10 PM, Martin Jansa <martin.jansa@gmail.com> wrote:
> >> > On Thu, Dec 08, 2011 at 07:57:52AM -0700, Tom Rini wrote:
> >> >> On Tue, Dec 6, 2011 at 4:23 PM, Andrea Adami <andrea.adami@gmail.com> wrote:
> >> >> > * Building the jffs2 filesystem to include summary information speeds up
> >> >> > * the mount process considerably at the cost of increased size.
> >> >> > * The rate of speedup is generally higher on NAND chips and on the chips
> >> >> > * where the erase block size is large.
> >> >> >
> >> >> > Signed-off-by: Andrea Adami <andrea.adami@gmail.com>
> >> >
> >> > I'll repeat my comment from yesterday, but IMHO it would be easier to
> >> > see the difference between those 2 images if the IMAGE_CMD_jffs2 output is
> >> > renamed to .jffs2.nosummary and IMAGE_CMD_sum.jffs2 (or renamed IMAGE_CMD_jffs2.summary)
> >> > produces .jffs2.summary (or even just .jffs2 if we can agree that .summary is
> >> > what enduser wants by default)
> >>
> >> I tried to keep the patch not too invasive and did not introduce a
> >> special EXTRA string for sumtool.
> >> The most widely used options (-p -l -e) are common, though.
> >>
> >> Unfortunately summary means we have a bigger image:
> >> 44M Dec  8 16:14 core-image-sato-c7x0-20111208151047.rootfs.jffs2
> >> 57M Dec  8 16:14 core-image-sato-c7x0-20111208151047.rootfs.sum.jffs2
> >>
> >> Being some devices have strict size requirements I think it is better
> >> to build both kind of images.
> >
> > Today I've noticed that it doesn't work for me:
> >
> > I have changed EXTRA_IMAGECMD_jffs2 to use new sum.jffs2:
> > http://git.shr-project.org/git/?p=meta-smartphone.git;a=blobdiff;f=meta-openmoko/conf/machine/om-gta02.conf;h=ad71d3db386b7d7df6ab484ea9ad80260da30373;hp=7cfb1b2a07b09ac3a35a27d209015ea8533ff632;hb=26d60e47358abe66d7d0fdbe6d5adea8f46bdbe2;hpb=043b5ac84794946e88e880b722fed445d3d985c5
> >
> 
> Martin,
> 
> the idea of the patch was to keep things simple.
> 
> After all most machines only specify endianness, eraseblocksize and padding.
> The few which are using summary jffs2 in meta-handhels appear to be
> Zaurus and hx4700 (I'll send a patch to fix ipaq's too) and fwiw
> beagleboard used '-npl' so I think we cover the most common case when
> people are setting
> 
> EXTRA_IMAGECMD_jffs2 = "-p -l --eraseblock=${ERASEBLOCKSIZE}"
> 
> and for the rare case they add  IMAGE_FSTYPES += "tar.gz jffs2 sum.jffs2":
> EXTRA_IMAGECMD_sum.jffs2 = "${EXTRA_IMAGECMD_jffs2}"
> 
> This happens to work just as intended, without adding another var for
> sumtool args.
> 
>         mkfs.jffs2
> --root=/oe/oe-core/build/tmp-eglibc/work/c7x0-oe-linux-gnueabi/core-image-sato-1.0-r0/rootfs
> --faketime --output=/oe/oe-core/build/tmp-eglibc/deploy/images/core-image-sato-c7x0-20111210181230.rootfs.jffs2
> -n -p -l --eraseblock=0x4000
>         cd /oe/oe-core/build/tmp-eglibc/deploy/images/
>         rm -f /oe/oe-core/build/tmp-eglibc/deploy/images/core-image-sato-c7x0.jffs2
>         ln -s core-image-sato-c7x0-20111210181230.rootfs.jffs2
> /oe/oe-core/build/tmp-eglibc/deploy/images/core-image-sato-c7x0.jffs2
>         # Image generation code for image type sum.jffs2
>         ROOTFS_SIZE=`du -ks
> /oe/oe-core/build/tmp-eglibc/work/c7x0-oe-linux-gnueabi/core-image-sato-1.0-r0/rootfs|awk
> '{size = $1 * 1.3 + 0; OFMT = "%.0f" ; print (size > 65536 ? size :
> 65536) }'`
>         mkfs.jffs2
> --root=/oe/oe-core/build/tmp-eglibc/work/c7x0-oe-linux-gnueabi/core-image-sato-1.0-r0/rootfs
> --faketime --output=/oe/oe-core/build/tmp-eglibc/deploy/images/core-image-sato-c7x0-20111210181230.rootfs.jffs2
> -n -p -l --eraseblock=0x4000 && sumtool -i /oe/
> oe-core/build/tmp-eglibc/deploy/images/core-image-sato-c7x0-20111210181230.rootfs.jffs2
>         -o /oe/oe-core/build/tmp-eglibc/deploy/images/core-image-sato-c7x0-20111210181230.rootfs.sum.jffs2
> -n -p -l --eraseblock=0x4000
>         cd /oe/oe-core/build/tmp-eglibc/deploy/images/
>         rm -f /oe/oe-core/build/tmp-eglibc/deploy/images/core-image-sato-c7x0.sum.jffs2
>         ln -s core-image-sato-c7x0-20111210181230.rootfs.sum.jffs2
> /oe/oe-core/build/tmp-eglibc/deploy/images/core-image-sato-c7x0.sum.jffs2
> 
> 
> 
> Being that IMAGE_CMD_sum.jffs2 cannot be overriden I adapted the
> EXTRA_IMAGECMD_jffs2 and simplified it in the way the string can be
> used by sumtool.
> To have full customization you'd need a weak assignment in image_types.bbclass.

It can be overriden and that's how I have fixed this (imho wrong) behavior

http://git.shr-project.org/git/?p=meta-smartphone.git;a=blobdiff;f=meta-openmoko/conf/machine/om-gta02.conf;h=662080612add4269955a74db12e31cfe1d0102bc;hp=ad71d3db386b7d7df6ab484ea9ad80260da30373;hb=5cf4e7f98266d5d3543afb7957f4243dca42efd3;hpb=7f8f34488dba6a6f26bea323a50196be2f11a705

because with
IMAGE_CMD_sum.jffs2 = "${IMAGE_CMD_jffs2} && sumtool -i ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.jffs2 \
      -o ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.sum.jffs2 -n ${EXTRA_IMAGECMD}"

I think it's common sense to expect EXTRA_IMAGECMD_sum.jffs2 being
applied to sumtool not mkfs.jffs2

Regards,

> > OE om-gta02@shr ~/shr-core $ bitbake -e shr-lite-image | tee -a image.log
> >
> > OE om-gta02@shr ~/shr-core $ grep ^EXTRA_IMAGECMD image.log | grep jffs2
> > EXTRA_IMAGECMD_sum.jffs2="--eraseblock=0x20000 --no-cleanmarkers --littleendian --pad"
> > EXTRA_IMAGECMD_jffs2="--little-endian --eraseblock=0x20000 --pagesize=0x800 --no-cleanmarkers --pad; mv /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-lite-20111211-om-gta02.rootfs.jffs2 /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-lite-20111211-om-gta02.rootfs.jffs2.nosummary;"
> >
> > This looks right (notice that sumtool is using different --little-endian/--littleendian).
> 
> Pls use -l  ;)
> 
> >
> > OE om-gta02@shr ~/shr-core $ grep IMAGE_CMD image.log | grep jffs2
> > # IMAGE_CMD_sum.jffs2=${IMAGE_CMD_jffs2} && sumtool -i ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.jffs2   -o ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.sum.jffs2 -n ${EXTRA_IMAGECMD}
> > IMAGE_CMD_sum.jffs2="mkfs.jffs2 --root=/OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/shr-lite-image/2.0-r20/rootfs --faketime --output=/OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-lite-20111211-om-gta02.rootfs.jffs2 -n  && sumtool -i /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-lite-20111211-om-gta02.rootfs.jffs2      -o /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-lite-20111211-om-gta02.rootfs.sum.jffs2 -n"
> > # IMAGE_CMD_jffs2=mkfs.jffs2 --root=${IMAGE_ROOTFS} --faketime --output=${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.jffs2 -n ${EXTRA_IMAGECMD}
> > IMAGE_CMD_jffs2="mkfs.jffs2 --root=/OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/shr-lite-image/2.0-r20/rootfs --faketime --output=/OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-lite-20111211-om-gta02.rootfs.jffs2 -n"
> >
> > Here we can see that bitbake -e didn't expand EXTRA_IMAGECMD right, but it seems like do_rootfs have them not empty:
> >
> > shr-core/tmp/work/om_gta02-oe-linux-gnueabi/shr-image/2.0-r18/temp/run.do_rootfs.688:
> >        # Image generation code for image type sum.jffs2
> >        ROOTFS_SIZE=`du -ks /OE/shr-core/tmp/work/om_gta02-oe-linux-gnueabi/shr-image/2.0-r18/rootfs|awk '{size = $1 * 1.3 + 0; OFMT = "%.0f" ; print (size > 65536 ? size : 65536) }'`
> >        mkfs.jffs2 --root=/OE/shr-core/tmp/work/om_gta02-oe-linux-gnueabi/shr-image/2.0-r18/rootfs --faketime --output=/OE/shr-core/tmp/deploy/images/om-gta02/shr-full-20111210-om-gta02.rootfs.jffs2 -n --eraseblock=0x20000 --no-cleanmarkers --littleendian --pad  && sumtool -i /OE/shr-core/tmp/deploy/images/om-gta02/shr-full-20111210-om-gta02.rootfs.jffs2    -o /OE/shr-core/tmp/deploy/images/om-gta02/shr-full-20111210-om-gta02.rootfs.sum.jffs2 -n --eraseblock=0x20000 --no-cleanmarkers --littleendian --pad
> >        cd /OE/shr-core/tmp/deploy/images/om-gta02/
> >        rm -f /OE/shr-core/tmp/deploy/images/om-gta02/full-om-gta02.sum.jffs2
> >        ln -s shr-full-20111210-om-gta02.rootfs.sum.jffs2 /OE/shr-core/tmp/deploy/images/om-gta02/full-om-gta02.sum.jffs2
> >
> > Notice that IMAGE_CMD_jffs2 is using EXTRA_IMAGECMD_sum.jffs2 (so params for sumtool not mkfs.jffs2) when it's expanded
> > inside IMAGE_CMD_sum.jffs2 variable
> >
> > openembedded-core/meta/classes/image_types.bbclass:
> > IMAGE_CMD_jffs2 = "mkfs.jffs2 --root=${IMAGE_ROOTFS} --faketime --output=${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.jffs2 -n ${EXTRA_IMAGECMD}"
> > IMAGE_CMD_sum.jffs2 = "${IMAGE_CMD_jffs2} && sumtool -i ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.jffs2 \
> >        -o ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.sum.jffs2 -n ${EXTRA_IMAGECMD}"
> >
> > And causing
> > mkfs.jffs2: unrecognized option '--littleendian'
> > mkfs.jffs2: error!: Usage: mkfs.jffs2 [OPTIONS]
> >
> > And using EXTRA_IMAGECMD_jffs2 in IMAGE_CMD_jffs2 breaks parsing..
> >
> > ERROR: Unable to parse /OE/shr-core/openembedded-core/meta/recipes-core/images/core-image-minimal-initramfs.bb                                                                                                               | ETA:  --:--:--
> > Traceback (most recent call last):
> >  File "/usr/lib64/python2.7/site-packages/bb/codeparser.py", line 268, in ShellParser.parse_shell(value='\t#set -x\n\trm -rf /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs\n\trm -rf /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/multilib\n\tmkdir -p /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs\n\tmkdir -p /OE/shr-core/tmp-eglibc/deploy/images/om-gta02\n\n\tcp /OE/shr-core/openembedded-core/meta/files/deploydir_readme.txt /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/README_-_DO_NOT_DELETE_FILES_IN_THIS_DIRECTORY.txt\n\n\tif [ "0" != "1" ]; then\n\t\tfor devtable in  /OE/shr-core/meta-handheld/files/device_table-minimal.txt; do\n\t\t\tmakedevs -r /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs -D $devtable\n\t\tdone\n\tfi\n\n\trootfs_ipk_do_rootfs\n\n\tinsert_feed_uris\n\n\tif [ "xldconfig-native:do_populate_sysroot" != "x" ]; then\n\t\t# Run ldconfig on the image to create a valid cache\n\t\t# (new format for cross arch compatibility)\n\t\techo executing: ldconfig -r /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs -c new -v\n\t\tldconfig -r /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs -c new -v\n\tfi\n\n\t# (re)create kernel modules dependencies\n\t# This part is done by kernel-module-* postinstall scripts but if image do\n\t# not contains modules at all there are few moments in boot sequence with\n\t# "unable to open modules.dep" message.\n\tif [ -e /OE/shr-core/tmp-eglibc/sysroots/om-gta02/kernel/kernel-abiversion ]; then\n\t\tKERNEL_VERSION=`cat /OE/shr-core/tmp-eglibc/sysroots/om-gta02/kernel/kernel-abiversion`\n\n\t\tmkdir -p /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs/lib/modules/$KERNEL_VERSION\n\t\tarm-oe-linux-gnueabi-depmod -a -b /OE/shr-core/tmp-
> >             try:
> >    >            tokens, _ = pyshyacc.parse(value, eof=True, debug=False)
> >             except pyshlex.NeedMore:
> >  File "/usr/lib64/python2.7/site-packages/bb/pysh/pyshyacc.py", line 673, in parse(input='\t#set -x\n\trm -rf /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs\n\trm -rf /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/multilib\n\tmkdir -p /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs\n\tmkdir -p /OE/shr-core/tmp-eglibc/deploy/images/om-gta02\n\n\tcp /OE/shr-core/openembedded-core/meta/files/deploydir_readme.txt /OE/shr-core/tmp-eglibc/deploy/images/om-gta02/README_-_DO_NOT_DELETE_FILES_IN_THIS_DIRECTORY.txt\n\n\tif [ "0" != "1" ]; then\n\t\tfor devtable in  /OE/shr-core/meta-handheld/files/device_table-minimal.txt; do\n\t\t\tmakedevs -r /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs -D $devtable\n\t\tdone\n\tfi\n\n\trootfs_ipk_do_rootfs\n\n\tinsert_feed_uris\n\n\tif [ "xldconfig-native:do_populate_sysroot" != "x" ]; then\n\t\t# Run ldconfig on the image to create a valid cache\n\t\t# (new format for cross arch compatibility)\n\t\techo executing: ldconfig -r /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs -c new -v\n\t\tldconfig -r /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs -c new -v\n\tfi\n\n\t# (re)create kernel modules dependencies\n\t# This part is done by kernel-module-* postinstall scripts but if image do\n\t# not contains modules at all there are few moments in boot sequence with\n\t# "unable to open modules.dep" message.\n\tif [ -e /OE/shr-core/tmp-eglibc/sysroots/om-gta02/kernel/kernel-abiversion ]; then\n\t\tKERNEL_VERSION=`cat /OE/shr-core/tmp-eglibc/sysroots/om-gta02/kernel/kernel-abiversion`\n\n\t\tmkdir -p /OE/shr-core/tmp-eglibc/work/om_gta02-oe-linux-gnueabi/core-image-minimal-initramfs/1.0-r0/rootfs/lib/modules/$KERNEL_VERSION\n\t\tarm-oe-linux-gnueabi-depmod -a -b /OE/shr-core/tmp-eglibc/work/om_
> >             debug = 2
> >    >    return yacc.parse(lexer=lexer, debug=debug), remaining
> >
> >  File "/usr/lib64/python2.7/site-packages/ply/yacc.py", line 265, in LRParser.parse(input=None, lexer=<bb.pysh.pyshlex.PLYLexer instance at 0x9638248>, debug=False, tracking=0, tokenfunc=None):
> >             else:
> >    >            return self.parseopt_notrack(input,lexer,debug,tracking,tokenfunc)
> >
> >  File "/usr/lib64/python2.7/site-packages/ply/yacc.py", line 1047, in LRParser.parseopt_notrack(input=None, lexer=<bb.pysh.pyshlex.PLYLexer instance at 0x9638248>, debug=False, tracking=0, tokenfunc=None):
> >                                 errtoken.lexer = lexer
> >    >                        tok = self.errorfunc(errtoken)
> >                             del errok, token, restart   # Delete special functions
> >  File "/usr/lib64/python2.7/site-packages/bb/pysh/pyshyacc.py", line 646, in p_error(p=LexToken(AND_IF,'&&',0,0)):
> >             w('  %r\n' % n)
> >    >    raise sherrors.ShellSyntaxError(''.join(msg))
> >
> > ShellSyntaxError: LexToken(AND_IF,'&&',0,0)
> > followed by:
> >  LexToken(TOKEN,'sumtool',0,0)
> >  LexToken(TOKEN,'-i',0,0)
> >  LexToken(TOKEN,'/OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.jffs2',0,0)
> >  LexToken(TOKEN,'-o',0,0)
> >  LexToken(TOKEN,'/OE/shr-core/tmp-eglibc/deploy/images/om-gta02/shr-core-image-minimal-initramfs-20111211-om-gta02.rootfs.sum.jffs2',0,0)
> >
> >
> >
> >>
> >> Cheers
> >>
> >> Andrea
> >>
> >> >
> >> > But other than that it looks fine and will make my IMAGE_DEPENDS_jffs2 a
> >> > bit shorter.
> >> >
> >> > Regards,
> >> >
> >> >>
> >> >> Acked-by: Tom Rini <tom.rini@gmail.com>
> >> >>
> >> >> > ---
> >> >> >  meta/classes/image_types.bbclass |    5 ++++-
> >> >> >  1 files changed, 4 insertions(+), 1 deletions(-)
> >> >> >
> >> >> > diff --git a/meta/classes/image_types.bbclass b/meta/classes/image_types.bbclass
> >> >> > index 29b6380..bd4b7bc 100644
> >> >> > --- a/meta/classes/image_types.bbclass
> >> >> > +++ b/meta/classes/image_types.bbclass
> >> >> > @@ -35,6 +35,8 @@ XZ_COMPRESSION_LEVEL ?= "-e -9"
> >> >> >  XZ_INTEGRITY_CHECK ?= "crc32"
> >> >> >
> >> >> >  IMAGE_CMD_jffs2 = "mkfs.jffs2 --root=${IMAGE_ROOTFS} --faketime --output=${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.jffs2 ${EXTRA_IMAGECMD}"
> >> >> > +IMAGE_CMD_sum.jffs2 = "${IMAGE_CMD_jffs2} && sumtool -i ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.jffs2 \
> >> >> > +       -o ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.sum.jffs2 ${EXTRA_IMAGECMD}"
> >> >> >
> >> >> >  IMAGE_CMD_cramfs = "mkcramfs ${IMAGE_ROOTFS} ${DEPLOY_DIR_IMAGE}/${IMAGE_NAME}.rootfs.cramfs ${EXTRA_IMAGECMD}"
> >> >> >
> >> >> > @@ -138,6 +140,7 @@ EXTRA_IMAGECMD_btrfs ?= ""
> >> >> >
> >> >> >  IMAGE_DEPENDS = ""
> >> >> >  IMAGE_DEPENDS_jffs2 = "mtd-utils-native"
> >> >> > +IMAGE_DEPENDS_sum.jffs2 = "mtd-utils-native"
> >> >> >  IMAGE_DEPENDS_cramfs = "cramfs-native"
> >> >> >  IMAGE_DEPENDS_ext2 = "genext2fs-native"
> >> >> >  IMAGE_DEPENDS_ext2.gz = "genext2fs-native"
> >> >> > @@ -157,4 +160,4 @@ IMAGE_DEPENDS_ubi = "mtd-utils-native"
> >> >> >  IMAGE_DEPENDS_ubifs = "mtd-utils-native"
> >> >> >
> >> >> >  # This variable is available to request which values are suitable for IMAGE_FSTYPES
> >> >> > -IMAGE_TYPES = "jffs2 cramfs ext2 ext2.gz ext2.bz2 ext3 ext3.gz ext2.lzma live squashfs squashfs-lzma ubi tar tar.gz tar.bz2 tar.xz cpio cpio.gz cpio.xz cpio.lzma"
> >> >> > +IMAGE_TYPES = "jffs2 sum.jffs2 cramfs ext2 ext2.gz ext2.bz2 ext3 ext3.gz ext2.lzma live squashfs squashfs-lzma ubi tar tar.gz tar.bz2 tar.xz cpio cpio.gz cpio.xz cpio.lzma"
> >> >> > --
> >> >> > 1.7.3.4
> >> >> >
> >> >> >
> >> >> > _______________________________________________
> >> >> > Openembedded-core mailing list
> >> >> > Openembedded-core@lists.openembedded.org
> >> >> > http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
> >> >>
> >> >>
> >> >>
> >> >> --
> >> >> Tom
> >> >>
> >> >> _______________________________________________
> >> >> Openembedded-core mailing list
> >> >> Openembedded-core@lists.openembedded.org
> >> >> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
> >> >
> >> > --
> >> > Martin 'JaMa' Jansa     jabber: Martin.Jansa@gmail.com
> >> >
> >> > _______________________________________________
> >> > Openembedded-core mailing list
> >> > Openembedded-core@lists.openembedded.org
> >> > http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
> >> >
> >>
> >> _______________________________________________
> >> Openembedded-core mailing list
> >> Openembedded-core@lists.openembedded.org
> >> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
> >
> > --
> > Martin 'JaMa' Jansa     jabber: Martin.Jansa@gmail.com
> >
> > _______________________________________________
> > Openembedded-core mailing list
> > Openembedded-core@lists.openembedded.org
> > http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
> >
> 
> 
> Regards
> Andrea
> 
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core

-- 
Martin 'JaMa' Jansa     jabber: Martin.Jansa@gmail.com

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 205 bytes --]

^ permalink raw reply	[flat|nested] 14+ messages in thread
end of thread, other threads:[~2011-12-11 23:04 UTC | newest]

Thread overview: 14+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2011-12-06 23:23 [PATCH 0/2] rework jffs2 images creation, add jffs2 summary Andrea Adami
2011-12-06 23:23 ` [PATCH 1/2] image_types.bbclass: implement jffs2 summary images (sum.jffs2) Andrea Adami
2011-12-08 14:57   ` Tom Rini
2011-12-08 15:10     ` Martin Jansa
2011-12-08 21:30       ` Andrea Adami
2011-12-08 22:43         ` Ulf Samuelsson
2011-12-08 22:48           ` Andrea Adami
2011-12-08 22:56           ` Richard Purdie
2011-12-11 10:03         ` Martin Jansa
2011-12-11 22:20           ` Andrea Adami
2011-12-11 22:57             ` Martin Jansa
2011-12-06 23:23 ` [PATCH 2/2] image_types.bbclass: disable cleanmarkers for jffs2 images Andrea Adami
2011-12-08 14:59   ` Tom Rini
2011-12-08 18:22 ` [PATCH 0/2] rework jffs2 images creation, add jffs2 summary Saul Wold

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox