Skip to contentSkip to conversation

2009.05-rc3 released

Buildroot Peter Korsgaard 7 participants 14 messages May 27, 2009 – May 28, 2009 1 message not in the archive

lore
Share & tools
lore
#0 lore raw

Hi,

I've released RC3. Unless big issues are found, expect this to be the last release candidate before the release - So give it a good test.

Get it from here:

http://buildroot.net/downloads/buildroot-2009.05-rc3.tar.bz2

And report any issues found here or in the bugtracker.

The plan is to release the final 2009.05 release this weekend.

Signature · 1 line
Bye, Peter Korsgaard
#1 lore raw

Does this release address the MPFR/gmp problem I reported a few days ago?

Thanks,
Adam

Quoted 19 lines · Peter Korsgaard

Hi,

I've released RC3. Unless big issues are found, expect this to be the last release candidate before the release - So give it a good test.

Get it from here:

http://buildroot.net/downloads/buildroot-2009.05-rc3.tar.bz2

And report any issues found here or in the bugtracker.

The plan is to release the final 2009.05 release this weekend.

Signature · 5 lines
Bye, Peter Korsgaard
_______________________________________________
buildroot mailing list
buildroot at busybox.net
http://lists.busybox.net/mailman/listinfo/buildroot
#2 lore raw
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

"Adam" == Adam McLaurin <[email protected]> writes:

Adam> Does this release address the MPFR/gmp problem I reported a few days Adam> ago?

No, I unfortunately haven't had time to look into that.

Signature · 1 line
Bye, Peter Korsgaard
#3 lore raw
In reply to Peter Korsgaard
Quoted 5 lines · Peter Korsgaard

Hi,

I've released RC3. Unless big issues are found, expect this to be the last release candidate before the release - So give it a good test.

Quoted 8 lines

Get it from here:

http://buildroot.net/downloads/buildroot-2009.05-rc3.tar.bz2

And report any issues found here or in the bugtracker.

The plan is to release the final 2009.05 release this weekend.

The default linux-header for the tool chain was 2.6.29.4 but the default kernel itself was 2.6.29.2. The attached patch fixes that.

Svein-Erik Skjelbred

Start patch
----------------

(quoted diff) +0 -0

b/target/linux/Config.in.advanced
index de54d03..3a171b5 100644

target/linux/Config.in.advanced +1 -1
@@ -128,7 +128,7 @@ choice
           Select the specific Linux version you want to use
 
  config BR2_LINUX_2_6_STABLE
-       bool "The latest stable Linux kernel (2.6.29.2)"
+       bool "The latest stable Linux kernel (2.6.29.4)"
 
  config BR2_LINUX_2_6_29
         bool "Linux 2.6.29"
(quoted diff) +0 -0

b/target/linux/Config.in.versions
index 467f73c..a37a5ad 100644

target/linux/Config.in.versions +2 -2
@@ -19,13 +19,13 @@ config LINUX26_LATEST_MM_VERSION
 
  config BR2_KERNEL_CURRENT_VERSION
         string
-       default "2.6.29.2"
+       default "2.6.29.4"
         help
           This is the latest stable kernel (including minor version)
 
  config BR2_KERNEL_LATEST_2_6_29
         string
-       default "2"
+       default "4"
 
  config BR2_KERNEL_LATEST_2_6_28
         string

------------------------------
End Patch

#4 lore raw
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

"Svein-Erik" == Svein-Erik Skjelbred <[email protected]> writes:

Hi,

Svein-Erik> The default linux-header for the tool chain was 2.6.29.4 Svein-Erik> but the default kernel itself was 2.6.29.2. The attached Svein-Erik> patch fixes that.

Thanks. Your patch was whitespace mangled - Could you please fix your mailer or use git send-email next time?

I've committed your patch by hand.

Signature · 1 line
Bye, Peter Korsgaard
#5 lore raw
In reply to Peter Korsgaard

On Thu, May 28, 2009 at 5:06 AM, Peter Korsgaard <[email protected]> wrote:
Peter> And report any issues found here or in the bugtracker.
I have been trying to build buildroot with a ubifs image. I haven't quite got there yet, but there are some bugs associated with this version.

target/ubifs/ubifsroot.mk
$(call DOWNLOAD,"$(MKFS_UBIFS_URL)")

This should be a WGET, not a DOWNLOAD.

I have been trying to update the UBIFS support as this is now included in the mtd-utils package under the mkfs.ubifs.
ubifs also depends on e2fsprogs as there is uuid support needed in mkfs.ubifs. I have at least gotten ubifs to compile with a few changes here and there (mainly a change in mtd-utils makefiles to build mkfs.ubifs), but I haven't had time to get an image made yet.

Signature · 4 lines
----
Regards
Paul Archer
ptarcher at gmail.com
#6 lore raw

I Paul,

I tried to fix the ubifs too a while back, but other things got in
the way. I was mainly interested in seeing if it compresses better than jffs2.
Anyway. I think the correct approach would be mimmic jffs2.mk,
don't download anything and let mdtutils build mkfs.ubifs for the host.

Kind Regards,
Thiago A. Correa

Quoted 28 lines · Paul Archer

On Thu, May 28, 2009 at 5:06 AM, Peter Korsgaard <[email protected]> wrote:
Peter> And report any issues found here or in the bugtracker.
I have been trying to build buildroot with a ubifs image. I haven't quite got there yet, but there are some bugs associated with this version.

target/ubifs/ubifsroot.mk
$(call DOWNLOAD,"$(MKFS_UBIFS_URL)")

This should be a WGET, not a DOWNLOAD.

I have been trying to update the UBIFS support as this is now included in the mtd-utils package under the mkfs.ubifs.
ubifs also depends on e2fsprogs as there is uuid support needed in mkfs.ubifs. I have at least gotten ubifs to compile with a few changes here and there (mainly a change in mtd-utils makefiles to build mkfs.ubifs), but I haven't had time to get an image made yet.

--
----
Regards
Paul Archer
ptarcher at gmail.com
_______________________________________________
buildroot mailing list
buildroot at busybox.net
http://lists.busybox.net/mailman/listinfo/buildroot

#7 lore raw

2009/5/28 Thiago A. Corr?a <[email protected]>:

Quoted 3 lines

Anyway. I think the correct approach would be mimmic jffs2.mk,
don't download anything and let mdtutils build mkfs.ubifs for the host.

Thiago,

I agree, and thats the approach I am taking, I have managed to get it to build with these quick patches:

--- buildroot-snapshot-20090525/package/mtd/mtd-utils.git/mtd.mk 2009-05-
25 05:21:21.000000000 +1000
+++ buildroot-snapshot-20090525-modified/package/mtd/mtd-utils.git/mtd.mk 2
009-05-27 13:05:49.000000000 +1000
@@ -3,7 +3,7 @@

 # mtd provides jffs2 utilities
 #
 #############################################################
-MTD_VERSION:=fcb52ccc99679460640386c297023f852b108f68
+MTD_VERSION:=HEAD
 MTD_SOURCE:=mtd-utils-$(MTD_VERSION).tar.gz
 MTD_URL:=http://git.infradead.org/mtd-utils.git?a=snapshot;h=$(MTD_VERSION);sf=
tgz
 MTD_HOST_DIR:= $(TOOL_BUILD_DIR)/mtd-utils-$(MTD_VERSION)
@@ -18,6 +18,7 @@
 #
 #############################################################
 MKFS_JFFS2 := $(MTD_HOST_DIR)/mkfs.jffs2
+MKFS_UBIFS := $(MTD_HOST_DIR)/mkfs.ubifs/mkfs.ubifs
 SUMTOOL := $(MTD_HOST_DIR)/sumtool

$(DL_DIR)/$(MTD_SOURCE):
@@ -27,18 +28,18 @@

        $(MTD_CAT) $(DL_DIR)/$(MTD_SOURCE) | tar -C $(TOOL_BUILD_DIR) $(TAR_OPTI
ONS) -
        rm -rf $(MTD_HOST_DIR)
        mv $(TOOL_BUILD_DIR)/$(MTD_NAME) $(MTD_HOST_DIR)

- toolchain/patch-kernel.sh $(MTD_HOST_DIR) \
- package/mtd/mtd-utils.git mtd-utils-all\*.patch
- toolchain/patch-kernel.sh $(MTD_HOST_DIR) \
- package/mtd/mtd-utils.git mtd-utils-host\*.patch
+# toolchain/patch-kernel.sh $(MTD_HOST_DIR) \
+# package/mtd/mtd-utils.git mtd-utils-all\*.patch
+# toolchain/patch-kernel.sh $(MTD_HOST_DIR) \
+# package/mtd/mtd-utils.git mtd-utils-host\*.patch
touch $@

-$(MKFS_JFFS2): $(MTD_HOST_DIR)/.unpacked +$(MKFS_JFFS2) $(MKFS_UBIFS): $(MTD_HOST_DIR)/.unpacked

        CC="$(HOSTCC)" CROSS= CFLAGS=-I$(LINUX_HEADERS_DIR)/include \
                $(MAKE) \
                BUILDDIR=$(MTD_HOST_DIR) WITHOUT_XATTR=1 \

- -C $(MTD_HOST_DIR) mkfs.jffs2
+ -C $(MTD_HOST_DIR)

$(SUMTOOL): $(MTD_HOST_DIR)/.unpacked

        CC="$(HOSTCC)" CROSS= CFLAGS=-I$(LINUX_HEADERS_DIR)/include \
@@ -46,7 +47,7 @@
                BUILDDIR=$(MTD_HOST_DIR) WITHOUT_XATTR=1 \
                -C $(MTD_HOST_DIR) sumtool

-mtd-host: $(MKFS_JFFS2) $(SUMTOOL)
+mtd-host: $(MKFS_JFFS2) $(MKFS_UBIFS) $(SUMTOOL)

mtd-host-source: $(DL_DIR)/$(MTD_SOURCE)

@@ -65,8 +66,8 @@

        $(MTD_CAT) $(DL_DIR)/$(MTD_SOURCE) | tar -C $(BUILD_DIR) $(TAR_OPTIONS)
-
        rm -rf $(MTD_DIR)
        mv $(BUILD_DIR)/$(MTD_NAME) $(MTD_DIR)

- toolchain/patch-kernel.sh $(MTD_DIR) package/mtd/mtd-utils.git mtd-utils
-all\*.patch
- toolchain/patch-kernel.sh $(MTD_DIR) package/mtd/mtd-utils.git mtd-utils
-target\*.patch
+# toolchain/patch-kernel.sh $(MTD_DIR) package/mtd/mtd-utils.git mtd-utils
-all\*.patch
+# toolchain/patch-kernel.sh $(MTD_DIR) package/mtd/mtd-utils.git mtd-utils
-target\*.patch
touch $@

MTD_TARGETS_n :=

--- buildroot-snapshot-20090525-orig/target/ubifs/ubifsroot.mk

2009-05-25 05:21:21.000000000 +1000
+++ buildroot-snapshot-20090525/target/ubifs/ubifsroot.mk
2009-05-28 15:55:35.000000000 +1000

@@ -1,37 +1,5 @@
#############################################################
#
-# mkfs.ubifs to build to target ubifs filesystems
-#
-#############################################################
-#MKFS_UBIFS_VERSION=2582f128dad78591bc3adcc87c343c690bb82e61 -#MKFS_UBIFS_URL=http://git.infradead.org/users/dedekind/mkfs.ubifs.git?a=snapshot;h=$(MKFS_UBIFS_VERSION);sf=tgz -MKFS_UBIFS_VERSION=v0.4 -MKFS_UBIFS_URL=http://git.infradead.org/users/dedekind/mkfs.ubifs.git?a=snapshot;h=refs/tags/mkfs.ubifs-$(MKFS_UBIFS_VERSION);sf=tgz -MKFS_UBIFS_SOURCE:=mkfs.ubifs-$(MKFS_UBIFS_VERSION).tar.gz
-MKFS_UBIFS_DIR:= $(BUILD_DIR)/mkfs-ubifs-$(MKFS_UBIFS_VERSION)
-MKFS_UBIFS_CAT:=$(ZCAT)
-MKFS_UBIFS_NAME:=mkfs.ubifs
-
-$(DL_DIR)/$(MKFS_UBIFS_SOURCE):
- $(call DOWNLOAD,"$(MKFS_UBIFS_URL)")
-
-$(MKFS_UBIFS_DIR)/.unpacked: $(DL_DIR)/$(MKFS_UBIFS_SOURCE)

-       $(ZCAT) $(DL_DIR)/$(MKFS_UBIFS_SOURCE) | tar -C $(BUILD_DIR) -xvf -
-       mv $(BUILD_DIR)/$(MKFS_UBIFS_NAME) $(MKFS_UBIFS_DIR)
-       toolchain/patch-kernel.sh $(MKFS_UBIFS_DIR) target/ubifs/

mkfs-ubifs-\*.patch
- touch $@
-
-$(MKFS_UBIFS_DIR)/mkfs.ubifs: $(MKFS_UBIFS_DIR)/.unpacked
- $(MAKE) -C $(MKFS_UBIFS_DIR)
- touch -c $@
-
-mkfs.ubifs-dirclean:
- rm -rf $(MKFS_UBIFS_DIR)
-
-mkfs.ubifs: $(MKFS_UBIFS_DIR)/mkfs.ubifs
-
-#############################################################
-#
# Build the ubifs root filesystem image
#
#############################################################
@@ -70,13 +38,19 @@
UBIFS_ROOTFS_COMPRESSOR_PREREQ:= lzma-host
endif

+#
+# mtd-host is a dependency which builds a local copy of mkfs.ubifs if it is needed.
+# the actual build is done from package/mtd/mtd.mk and it sets the +# value of MKFS_UBIFS to either the previously installed copy or the one +# just built.
+#
ifneq ($(UBIFS_ROOTFS_COMPRESSOR),)
UBIFS_TARGET := $(UBIFS_BASE).$(UBIFS_ROOTFS_COMPRESSOR_EXT) else
UBIFS_TARGET := $(UBIFS_BASE)
endif

-$(UBIFS_BASE): host-fakeroot makedevs mkfs.ubifs
+$(UBIFS_TARGET): host-fakeroot makedevs mtd-host

        # Use fakeroot to pretend all target binaries are owned by root
        rm -f $(PROJECT_BUILD_DIR)/_fakeroot.$(notdir $(UBIFS_TARGET))
        touch $(PROJECT_BUILD_DIR)/.fakeroot.00000
@@ -88,7 +62,7 @@
                >> $(PROJECT_BUILD_DIR)/_fakeroot.$(notdir $(UBIFS_TARGET))
 endif
        # Use fakeroot so mkfs.ubifs believes the previous fakery
-       echo "$(MKFS_UBIFS_DIR)/mkfs.ubifs -d $(TARGET_DIR) " \
+       echo "$(MKFS_UBIFS) -d $(TARGET_DIR) " \
                "$(UBIFS_OPTS) -o $(UBIFS_BASE)" >>
$(PROJECT_BUILD_DIR)/_fakeroot.$(notdir $(UBIFS_TARGET))
        chmod a+x $(PROJECT_BUILD_DIR)/_fakeroot.$(notdir $(UBIFS_TARGET))
        $(HOST_DIR)/usr/bin/fakeroot --
$(PROJECT_BUILD_DIR)/_fakeroot.$(notdir $(UBIFS_TARGET))
@@ -108,13 +82,13 @@
        @cp -f $(UBIFS_TARGET) $(UBIFS_COPYTO)

endif

-ubifsroot-source: $(DL_DIR)/$(GENUBIFS_SOURCE)
+ubifsroot-source: mtd-host-source

ubifsroot-clean:
- -$(MAKE) -C $(MKFS_UBIFS_DIR) clean
+ -rm -f $(UBIFS_TARGET)

ubifsroot-dirclean:
- rm -rf $(MKFS_UBIFS_DIR)
+ -rm -rf $(UBIFS_TARGET)

#############################################################
#

With this patch, I successfully get a ubifs image. I think... (not sure if I have touched other stuff unfortunately). Sorry if gmail wraps the patches, I can always send via an attachment if need be.

Regards Paul

Quoted 30 lines · Paul Archer
Quoted 28 lines · Peter Korsgaard

Peter> And report any issues found here or in the bugtracker. I have been trying to build buildroot with a ubifs image. I haven't quite got there yet, but there are some bugs associated with this version.

target/ubifs/ubifsroot.mk
$(call DOWNLOAD,"$(MKFS_UBIFS_URL)")

This should be a WGET, not a DOWNLOAD.

I have been trying to update the UBIFS support as this is now included in the mtd-utils package under the mkfs.ubifs.
ubifs also depends on e2fsprogs as there is uuid support needed in mkfs.ubifs. I have at least gotten ubifs to compile with a few changes here and there (mainly a change in mtd-utils makefiles to build mkfs.ubifs), but I haven't had time to get an image made yet.

--
----
Regards
Paul Archer
ptarcher at gmail.com
_______________________________________________
buildroot mailing list
buildroot at busybox.net
http://lists.busybox.net/mailman/listinfo/buildroot

Signature · 4 lines
----
Regards
Paul Archer
ptarcher at gmail.com
#8 lore raw
In reply to Thiago A. Corrêa

2009/5/28 Thiago A. Corr?a <[email protected]>:

Quoted 3 lines

? I tried to fix the ubifs too a while back, but other things got in the way. I was mainly interested in seeing if it compresses better than jffs2.

I am not 100% sure if this is what you meant by comparing the sizes of the images, but I have made a ubifs image, so I thought I would show the results.

-rw-r--r-- 1 paul paul 4.6M 2009-05-28 16:03 binaries/AT91SAM9G20/at91sam92g20ek-rootfs.arm-20090528.jffs2 -rw-r--r-- 1 paul paul 6.8M 2009-05-28 16:01 binaries/AT91SAM9G20/at91sam92g20ek-rootfs.arm-20090528.ubifs -rw-r--r-- 1 paul paul 4.1M 2009-05-28 16:03 binaries/AT91SAM9G20/at91sam92g20ek-rootfs.arm-20090528.ubifs.bz2 -rw-r--r-- 1 paul paul 4.3M 2009-05-28 16:02 binaries/AT91SAM9G20/at91sam92g20ek-rootfs.arm-20090528.ubifs.gz -rw-r--r-- 1 paul paul 3.6M 2009-05-28 16:04 binaries/AT91SAM9G20/at91sam92g20ek-rootfs.arm-20090528.ubifs.lzma

I can't check what this equates to on a extracted image yet as I need to modify u-boot a fair bit to get this to work, as currently uboot doesn't build with ubifs support for the at91sam9xxx.

If you want the config I used to do further analysis just let me know.

Regards Paul

Quoted 37 lines

? Anyway. I think the correct approach would be mimmic jffs2.mk, don't download anything and let mdtutils build mkfs.ubifs for the host.

Kind Regards,
? ?Thiago A. Correa

Quoted 28 lines · Paul Archer

On Thu, May 28, 2009 at 5:06 AM, Peter Korsgaard <[email protected]> wrote:
Peter> And report any issues found here or in the bugtracker.
I have been trying to build buildroot with a ubifs image. I haven't quite got there yet, but there are some bugs associated with this version.

target/ubifs/ubifsroot.mk
$(call DOWNLOAD,"$(MKFS_UBIFS_URL)")

This should be a WGET, not a DOWNLOAD.

I have been trying to update the UBIFS support as this is now included in the mtd-utils package under the mkfs.ubifs.
ubifs also depends on e2fsprogs as there is uuid support needed in mkfs.ubifs. I have at least gotten ubifs to compile with a few changes here and there (mainly a change in mtd-utils makefiles to build mkfs.ubifs), but I haven't had time to get an image made yet.

--
----
Regards
Paul Archer
ptarcher at gmail.com
_______________________________________________
buildroot mailing list
buildroot at busybox.net
http://lists.busybox.net/mailman/listinfo/buildroot

Signature · 4 lines
----
Regards
Paul Archer
ptarcher at gmail.com

A message that is not in the archive (sent off-list or not yet received): [email protected]

#10 lore raw
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

"Paul" == Paul Archer <[email protected]> writes:

Quoted 1 line · Peter Korsgaard
Quoted 1 line

And report any issues found here or in the bugtracker.

Paul> I have been trying to build buildroot with a ubifs image. I haven't Paul> quite got there yet, but there are some bugs associated with this Paul> version.

Paul> target/ubifs/ubifsroot.mk
Paul> $(call DOWNLOAD,"$(MKFS_UBIFS_URL)")

Paul> This should be a WGET, not a DOWNLOAD.

Thanks - Will fix. I just gave it a test, and it seems like the upstream git tree is gone, so it won't help you much.

Hamish, I believe you were the one adding ubifs support - What is the situation with mkfs.ubifs? Is upstream mtd-utils nowadays like mkfs.jff2?

Seems like we'll have to make ubifsroot as broken for the 2009.05 release :/

Signature · 1 line
Bye, Peter Korsgaard
#11 lore raw

Hi,

Quoted 16 lines · Peter Korsgaard
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

"Paul" == Paul Archer <[email protected]> writes:

Quoted 1 line · Peter Korsgaard
Quoted 1 line

And report any issues found here or in the bugtracker.

Paul> I have been trying to build buildroot with a ubifs image. I haven't Paul> quite got there yet, but there are some bugs associated with this Paul> version.

Paul> target/ubifs/ubifsroot.mk
Paul> $(call DOWNLOAD,"$(MKFS_UBIFS_URL)")

Paul> This should be a WGET, not a DOWNLOAD.

Thanks - Will fix. I just gave it a test, and it seems like the upstream git tree is gone, so it won't help you much.

Quoted 6 lines

Hamish, I believe you were the one adding ubifs support - What is the situation with mkfs.ubifs? Is upstream mtd-utils nowadays like mkfs.jff2?

Seems like we'll have to make ubifsroot as broken for the 2009.05 release :/

Oh, I fixed this in my tree at work I'm pretty sure, but obviously forgot to send in a patch. mkfs.ubifs is included in mtd-utils now, but you have to use the git version (it's not in an mtd-utils release).

I haven't kept up with official buildroot lately, especially since the switch to git. I'll see if I can send a patch to the mailing list tomorrow.. I can't get to my code right now.

Hamish

Signature · 1 line
Hamish Moffatt VK3SB <[email protected]> <[email protected]>
#12 lore raw
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

"Hamish" == Hamish Moffatt <[email protected]> writes:

Hi,

Quoted 2 lines
Quoted 2 lines

Seems like we'll have to make ubifsroot as broken for the 2009.05 release :/

Hamish> Oh, I fixed this in my tree at work I'm pretty sure, but Hamish> obviously forgot to send in a patch. mkfs.ubifs is included Hamish> in mtd-utils now, but you have to use the git version (it's Hamish> not in an mtd-utils release).

Ok, that's a pity. The 1.2.0 release (that we don't even have in BR) is getting a bit old in the tooth (~1 year).

Hamish> I haven't kept up with official buildroot lately, especially Hamish> since the switch to git. I'll see if I can send a patch to Hamish> the mailing list tomorrow.. I can't get to my code right now.

Thanks - Any specific problems with git?

Signature · 1 line
Bye, Peter Korsgaard
#13 lore raw
In reply to Peter Korsgaard

Hi,

Quoted 5 lines · Peter Korsgaard
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

"Paul" == Paul Archer <[email protected]> writes:

Seems like we'll have to make ubifsroot as broken for the 2009.05 release :/

Anything wrong with Paul's patch? It does look better than what I was doing (duplicating the mkfs.jffs2 rules for a mkfs.ubifs).

Basically, what needs fixing is getting mkfs.ubifs for the host, it was previously built by ubifs.mk from a user's tree, and now it's in the main mtdutils tree as part of the distribuition. Unfortunally there is no release containing it yet.

Kind Regards,
Thiago A. Correa

#14 lore raw
In reply to Peter Korsgaard

Hi,

I just tried to compile toolchain for arm926t and when I select "Build/install java compiler and libgcj" I get following error:

checking for exception model to use... configure: error: unable to detect exception model
make[2]: *** [configure-target-libjava] Error 1

Andrzej Wolski