Misc fixes

PATCH Applied Paul Eggleton 5 participants 10 messages Nov 4, 2011 – Nov 18, 2011

lore

Share & tools

lore

Paul Eggleton

author

2011-11-04 17:58 UTC #0 lore raw

cover Misc fixes

The following changes since commit f01fbc17b5d9bf9a227d64fe858014376cd19432:

avahi: fix useradd race condition (2011-11-03 14:48:42 +0000)

are available in the git repository at:
git://git.openembedded.org/openembedded-core-contrib paule/fixes8
http://cgit.openembedded.org/cgit.cgi/openembedded-core-contrib/log/?h=paule/fixes8

Paul Eggleton (3):
allarch.bbclass: disable shlib and debug symbol processing core-image-minimal-initramfs: force IMAGE_FSTYPES
qemugl: switch to new git repo

 meta/classes/allarch.bbclass                       |    5 +++++
 .../images/core-image-minimal-initramfs.bb         |    1 +
 meta/recipes-graphics/mesa/qemugl_git.bb           |    2 +-
 3 files changed, 7 insertions(+), 1 deletions(-)

Signature · 1 line

1.7.5.4

Paul Eggleton

author

2011-11-04 17:58 UTC #1 lore raw

PATCH 1/3 allarch.bbclass: disable shlib and debug symbol processing

all-arch packages should not contain any binaries that need processing, so disable the shared library dependency processing and debug symbol splitting/stripping to save some time during packaging.

Signed-off-by Paul Eggleton <[email protected]>

---
 meta/classes/allarch.bbclass |    5 +++++
 1 files changed, 5 insertions(+), 0 deletions(-)

meta/classes/allarch.bbclass +5 -0

index a2f8179..18c9ae2 100644
@@ -15,3 +15,8 @@ TARGET_CC_ARCH = "none"
 TARGET_LD_ARCH = "none"
 TARGET_AS_ARCH = "none"
 PACKAGE_EXTRA_ARCHS = ""
+
+# No need to do shared library processing or debug symbol handling
+EXCLUDE_FROM_SHLIBS = "1"
+INHIBIT_PACKAGE_DEBUG_SPLIT = "1"
+INHIBIT_PACKAGE_STRIP = "1"

Signature · 1 line

1.7.5.4

Paul Eggleton

author

2011-11-04 17:58 UTC #2 lore raw

PATCH 2/3 core-image-minimal-initramfs: force IMAGE_FSTYPES

In reply to Paul Eggleton

If the user has set their own value for IMAGE_FSTYPES, they may have disabled the cpio.gz image type, preventing the initramfs from being produced in the format that image-live.bbclass expects; so force IMAGE_FSTYPES to cpio.gz within the initramfs image recipe.

Signed-off-by Paul Eggleton <[email protected]>

---
 .../images/core-image-minimal-initramfs.bb         |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

meta/recipes-core/images/core-image-minimal-initramfs.bb +1 -0

index 0bac27a..e4d0e51 100644
@@ -13,3 +13,4 @@ LICENSE = "MIT"
 inherit core-image
 
 IMAGE_ROOTFS_SIZE = "8192"
+IMAGE_FSTYPES = "cpio.gz"

Signature · 1 line

1.7.5.4

Paul Eggleton

author

2011-11-04 17:58 UTC #8 lore raw

PATCH 3/3 qemugl: switch to new git repo

In reply to Paul Eggleton

The qemugl git repository is now on yoctoproject.org.

Signed-off-by Paul Eggleton <[email protected]>

---
 meta/recipes-graphics/mesa/qemugl_git.bb |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

meta/recipes-graphics/mesa/qemugl_git.bb +1 -1

index 1658759..e3552ec 100644
@@ -9,7 +9,7 @@ DEPENDS = "virtual/libx11 xproto glproto libxfixes"
 
 COMPATIBLE_HOST = '(x86_64.*|i.86.*)-(linux|freebsd.*)'
 
-SRC_URI = "git://git.o-hand.com/qemugl.git;protocol=git \
+SRC_URI = "git://git.yoctoproject.org/qemugl;protocol=git \
            file://versionfix.patch \
            file://remove-x11r6-lib-dir.patch \
            file://call_opengl_fix.patch"

Signature · 1 line

1.7.5.4

Koen Kooi

2011-11-04 18:24 UTC #3 lore raw

Re: [PATCH 2/3] core-image-minimal-initramfs: force IMAGE_FSTYPES

In reply to Paul Eggleton

Op 4 nov. 2011, om 18:52 heeft Paul Eggleton het volgende geschreven:

Quoted 17 lines

If the user has set their own value for IMAGE_FSTYPES, they may have disabled the cpio.gz image type, preventing the initramfs from being produced in the format that image-live.bbclass expects; so force IMAGE_FSTYPES to cpio.gz within the initramfs image recipe.
Signed-off-by Paul Eggleton <[email protected]>
---
.../images/core-image-minimal-initramfs.bb | 1 +
1 files changed, 1 insertions(+), 0 deletions(-)
meta/recipes-core/images/core-image-minimal-initramfs.bb +0 -0
index 0bac27a..e4d0e51 100644
@@ -13,3 +13,4 @@ LICENSE = "MIT"
inherit core-image

Quoted 2 lines

IMAGE_ROOTFS_SIZE = "8192"
+IMAGE_FSTYPES = "cpio.gz"

_append or += would give less suprises.

regards,

Koen

Saul Wold

Applied

2011-11-07 17:48 UTC #9 lore raw

In reply to Paul Eggleton

Quoted 13 lines · Paul Eggleton

The following changes since commit f01fbc17b5d9bf9a227d64fe858014376cd19432:
avahi: fix useradd race condition (2011-11-03 14:48:42 +0000)
are available in the git repository at:
git://git.openembedded.org/openembedded-core-contrib paule/fixes8
http://cgit.openembedded.org/cgit.cgi/openembedded-core-contrib/log/?h=paule/fixes8
Paul Eggleton (3):
allarch.bbclass: disable shlib and debug symbol processing core-image-minimal-initramfs: force IMAGE_FSTYPES
qemugl: switch to new git repo

Quoted 5 lines

  meta/classes/allarch.bbclass                       |    5 +++++
  .../images/core-image-minimal-initramfs.bb         |    1 +
  meta/recipes-graphics/mesa/qemugl_git.bb           |    2 +-
  3 files changed, 7 insertions(+), 1 deletions(-)

Merged into OE-Core with tweak to IMAGE_FSTYPES as suggested

Thanks
Sau!

Dmitry Eremin-Solenikov

2011-11-15 08:37 UTC #4 lore raw

Re: [PATCH 2/3] core-image-minimal-initramfs: force IMAGE_FSTYPES

In reply to Koen Kooi

Quoted 20 lines · Koen Kooi

Op 4 nov. 2011, om 18:52 heeft Paul Eggleton het volgende geschreven:
Quoted 17 lines
If the user has set their own value for IMAGE_FSTYPES, they may have disabled the cpio.gz image type, preventing the initramfs from being produced in the format that image-live.bbclass expects; so force IMAGE_FSTYPES to cpio.gz within the initramfs image recipe.
Signed-off-by Paul Eggleton<[email protected]>
---
.../images/core-image-minimal-initramfs.bb | 1 +
1 files changed, 1 insertions(+), 0 deletions(-)
meta/recipes-core/images/core-image-minimal-initramfs.bb +0 -0
index 0bac27a..e4d0e51 100644
@@ -13,3 +13,4 @@ LICENSE = "MIT"
inherit core-image

Quoted 4 lines

Quoted 2 lines
IMAGE_ROOTFS_SIZE = "8192"
+IMAGE_FSTYPES = "cpio.gz"
_append or += would give less suprises.

This was merged as IMAGE_FSTYPES =+ "cpio.gz"
Now this brings problems if I have IMAGE_FSTYPES += "live" in my local.conf / BSP machine.conf.

1) OE tries to generate hddimg for this initramfs image, which is strange idea
2) If OE is trying to generate bootimg (hddimg) when the kernel is not yet deployed, building fails (as bootimg can't find a deployed kernel to put into hddimg).

Please revert this back to original patch as proposed by Paul.

Signature · 2 lines

With best wishes
Dmitry

Paul Eggleton

author

2011-11-18 14:02 UTC #5 lore raw

Re: [PATCH 2/3] core-image-minimal-initramfs: force IMAGE_FSTYPES

Quoted 11 lines · On Tuesday 15 November 2011 12:30:15 Dmitry Eremin-Solenikov wrote:

This was merged as IMAGE_FSTYPES =+ "cpio.gz"
Now this brings problems if I have IMAGE_FSTYPES += "live" in my local.conf / BSP machine.conf.
1) OE tries to generate hddimg for this initramfs image, which is strange idea
2) If OE is trying to generate bootimg (hddimg) when the kernel is not yet deployed, building fails (as bootimg can't find a deployed kernel to put into hddimg).
Please revert this back to original patch as proposed by Paul.

OK, so what are we going to do about this? Do we have to do some filtering of this variable?

Cheers,
Paul

Signature · 2 lines

Paul Eggleton
Intel Open Source Technology Centre

Dmitry Eremin-Solenikov

2011-11-18 14:03 UTC #6 lore raw

Re: [PATCH 2/3] core-image-minimal-initramfs: force IMAGE_FSTYPES

Quoted 11 lines · Paul Eggleton

Quoted 10 lines · On Tuesday 15 November 2011 12:30:15 Dmitry Eremin-Solenikov wrote:
This was merged as IMAGE_FSTYPES =+ "cpio.gz"
Now this brings problems if I have IMAGE_FSTYPES += "live" in my local.conf / BSP machine.conf.
1) OE tries to generate hddimg for this initramfs image, which is strange idea
2) If OE is trying to generate bootimg (hddimg) when the kernel is not yet deployed, building fails (as bootimg can't find a deployed kernel to put into hddimg).

Quoted 4 lines

Quoted 1 line
Please revert this back to original patch as proposed by Paul.
OK, so what are we going to do about this? Do we have to do some filtering of this variable?

I'd propose to just set it via:

IMAGE_FSTYPES = "cpio.gz". It is really no point in generating other types of inititramfs images.

Quoted 4 lines

Cheers,
Paul

Signature · 2 lines

With best wishes
Dmitry

Andrea Adami

2011-11-18 14:13 UTC #7 lore raw

Re: [PATCH 2/3] core-image-minimal-initramfs: force IMAGE_FSTYPES

Quoted 16 lines · Dmitry Eremin-Solenikov

Quoted 15 lines · Paul Eggleton
Quoted 12 lines · On Tuesday 15 November 2011 12:30:15 Dmitry Eremin-Solenikov wrote:
This was merged as IMAGE_FSTYPES =+ "cpio.gz"
Now this brings problems if I have IMAGE_FSTYPES += "live" in my local.conf / BSP machine.conf.
1) OE tries to generate hddimg for this initramfs image, which is strange idea
2) If OE is trying to generate bootimg (hddimg) when the kernel is not yet deployed, building fails (as bootimg can't find a deployed kernel to put into hddimg).
Please revert this back to original patch as proposed by Paul.

Quoted 8 lines

Quoted 3 lines
OK, so what are we going to do about this? Do we have to do some filtering of
this variable?
I'd propose to just set it via:
IMAGE_FSTYPES = "cpio.gz". It is really no point in generating other types of inititramfs images.

In some other initramfs-images "cpio.lzma" is a welcomed option to save size.

Regards

Andrea

Quoted 16 lines

Quoted 5 lines
Cheers,
Paul
--
With best wishes
Dmitry
_______________________________________________
Openembedded-core mailing list
[email protected]
http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core