unsatisfied dependencies to libgcc
In the last days we switched our local development from an older oe-dev master to 2011.03-maintenance branch and hit an annoying problem.
The test builds on various developer machines were successful but the build on our continuous integration server (with exactly the same scripting) ended up with a lot of
| * satisfy_dependencies_for: Cannot satisfy the following dependencies for lighttpd: | * libstdc++6 (>= 4.3.3) * libgcc1 (>= 4.3.3) * libgcc1 (>= 4.3.3) * | * opkg_install_cmd: Cannot install package lighttpd.
errors.
After some searching we found that there was no libgcc1_4.3.3-r24.2.6_armv5te.ipk in the deploy area.
I'm not sure which package should produce this but i guess it should come from gcc-cross-4.3.3 (because its recipe version is r24.2.6 in opposite to gcc-4.3.3 r24.1.6).
Looking into the log i hit the fact that the do_rootfs stage for the image was started *before* the do_package_stage of gcc-cross_4.3.3.bb was succeeded.
Any ideas?
PS: Bitbaking gcc-cross explicitly before bitbaking the image wors as a workaround for us at the moment.
Signature · 7 lines
DResearch Fahrzeugelektronik GmbH Otto-Schmirgal-Str. 3, 10319 Berlin, Germany Tel: +49 30 515932-237 mailto:[email protected] Fax: +49 30 515932-299 Geschäftsführer: Dr. Michael Weber, Werner Mögle; Amtsgericht Berlin Charlottenburg; HRB 130120 B; Ust.-IDNr. DE273952058
Quoted 12 lines · Steffen Sledz
In the last days we switched our local development from an older oe-dev master to 2011.03-maintenance branch and hit an annoying problem.
The test builds on various developer machines were successful but the build on our continuous integration server (with exactly the same scripting) ended up with a lot of
| * satisfy_dependencies_for: Cannot satisfy the following dependencies for lighttpd: | * libstdc++6 (>= 4.3.3) * libgcc1 (>= 4.3.3) * libgcc1 (>= 4.3.3) * | * opkg_install_cmd: Cannot install package lighttpd.errors.
After some searching we found that there was no libgcc1_4.3.3-r24.2.6_armv5te.ipk in the deploy area.
Quoted 7 lines
I'm not sure which package should produce this but i guess it should come from gcc-cross-4.3.3 (because its recipe version is r24.2.6 in opposite to gcc-4.3.3 r24.1.6).
Looking into the log i hit the fact that the
do_rootfsstage for the image was started *before* thedo_package_stageof gcc-cross_4.3.3.bb was succeeded.Any ideas?
PS: Bitbaking gcc-cross explicitly before bitbaking the image wors as a workaround for us at the moment.
Another collegue had a similar problem while bitbaking meta-toolchain.
| * satisfy_dependencies_for: Cannot satisfy the following dependencies for task-sdk-bare: | * libstdc++-dev * | * opkg_install_cmd: Cannot install package task-sdk-bare.
So it seems it is not a problem of our image recipe.
Signature · 7 lines
DResearch Fahrzeugelektronik GmbH Otto-Schmirgal-Str. 3, 10319 Berlin, Germany Tel: +49 30 515932-237 mailto:[email protected] Fax: +49 30 515932-299 Geschäftsführer: Dr. Michael Weber, Werner Mögle; Amtsgericht Berlin Charlottenburg; HRB 130120 B; Ust.-IDNr. DE273952058
Quoted 12 lines · Steffen Sledz
In the last days we switched our local development from an older oe-dev master to 2011.03-maintenance branch and hit an annoying problem.
The test builds on various developer machines were successful but the build on our continuous integration server (with exactly the same scripting) ended up with a lot of
| * satisfy_dependencies_for: Cannot satisfy the following dependencies for lighttpd: | * libstdc++6 (>= 4.3.3) * libgcc1 (>= 4.3.3) * libgcc1 (>= 4.3.3) * | * opkg_install_cmd: Cannot install package lighttpd.errors.
After some searching we found that there was no libgcc1_4.3.3-r24.2.6_armv5te.ipk in the deploy area.
Quoted 7 lines
I'm not sure which package should produce this but i guess it should come from gcc-cross-4.3.3 (because its recipe version is r24.2.6 in opposite to gcc-4.3.3 r24.1.6).
Looking into the log i hit the fact that the
do_rootfsstage for the image was started *before* thedo_package_stageof gcc-cross_4.3.3.bb was succeeded.Any ideas?
PS: Bitbaking gcc-cross explicitly before bitbaking the image wors as a workaround for us at the moment.
I made some research in this and there is something i do not understand.
The task-depends.dot generated by bitbake -g (says
"console-image.do_rootfs" [label="console-image do_rootfs\n0:1.0-r0\n/home/sledz/work/angstrom-setup-scripts/sources/openembedded/recipes/images/console-image.bb"] "console-image.do_rootfs" -> "lzo-native.do_populate_sysroot"
"console-image.do_rootfs" -> "bluez4.do_populate_sysroot"
...
So do_rootfs depends only on do_populate_sysroot stages. But the do_rootfs stage itself uses opkg to install the packages into the image.
In my understanding this means that all the ipk files need to be available. But this is not guaranteed by the reported dependencies.
A misunderstanding of mine? Or a bug?
Steffen
Signature · 7 lines
DResearch Fahrzeugelektronik GmbH Otto-Schmirgal-Str. 3, 10319 Berlin, Germany Tel: +49 30 515932-237 mailto:[email protected] Fax: +49 30 515932-299 Geschäftsführer: Dr. Michael Weber, Werner Mögle; Amtsgericht Berlin Charlottenburg; HRB 130120 B; Ust.-IDNr. DE273952058
Am 05.08.2011 09:13, schrieb Steffen Sledz:
Quoted 30 lines · Steffen Sledz
Quoted 12 lines
In the last days we switched our local development from an older oe-dev master to 2011.03-maintenance branch and hit an annoying problem.
The test builds on various developer machines were successful but the build on our continuous integration server (with exactly the same scripting) ended up with a lot of
| * satisfy_dependencies_for: Cannot satisfy the following dependencies for lighttpd: | * libstdc++6 (>= 4.3.3) * libgcc1 (>= 4.3.3) * libgcc1 (>= 4.3.3) * | * opkg_install_cmd: Cannot install package lighttpd.errors.
After some searching we found that there was no libgcc1_4.3.3-r24.2.6_armv5te.ipk in the deploy area.
Quoted 7 lines
I'm not sure which package should produce this but i guess it should come from gcc-cross-4.3.3 (because its recipe version is r24.2.6 in opposite to gcc-4.3.3 r24.1.6).
Looking into the log i hit the fact that the
do_rootfsstage for the image was started *before* thedo_package_stageof gcc-cross_4.3.3.bb was succeeded.Any ideas?
PS: Bitbaking gcc-cross explicitly before bitbaking the image wors as a workaround for us at the moment.
I made some research in this and there is something i do not understand.
The task-depends.dot generated by bitbake -g (says
"console-image.do_rootfs" [label="console-image do_rootfs\n0:1.0-r0\n/home/sledz/work/angstrom-setup-scripts/sources/openembedded/recipes/images/console-image.bb"] "console-image.do_rootfs" -> "lzo-native.do_populate_sysroot"
"console-image.do_rootfs" -> "bluez4.do_populate_sysroot"
...
Quoted 5 lines
So
do_rootfsdepends only ondo_populate_sysrootstages. But thedo_rootfsstage itself uses opkg to install the packages into the image.In my understanding this means that all the ipk files need to be available. But this is not guaranteed by the reported dependencies.
A misunderstanding of mine? Or a bug?
Ping!
Did everyone read my message?
If it really is a misunderstanding of mine, please let me know. But if i'm right this seems to be a critical problem.
Steffen
Signature · 7 lines
DResearch Fahrzeugelektronik GmbH Otto-Schmirgal-Str. 3, 10319 Berlin, Germany Tel: +49 30 515932-237 Fax: +49 30 515932-299 Geschäftsführer: Dr. Michael Weber, Werner Mögle; Amtsgericht Berlin Charlottenburg; HRB 130120 B; Ust.-IDNr. DE273952058
Quoted 35 lines · Steffen Sledz
Am 05.08.2011 09:13, schrieb Steffen Sledz:
Quoted 34 lines · Steffen Sledz
Quoted 12 lines
In the last days we switched our local development from an older oe-dev master to 2011.03-maintenance branch and hit an annoying problem.
The test builds on various developer machines were successful but the build on our continuous integration server (with exactly the same scripting) ended up with a lot of
| * satisfy_dependencies_for: Cannot satisfy the following dependencies for lighttpd: | * libstdc++6 (>= 4.3.3) * libgcc1 (>= 4.3.3) * libgcc1 (>= 4.3.3) * | * opkg_install_cmd: Cannot install package lighttpd.errors.
After some searching we found that there was no libgcc1_4.3.3-r24.2.6_armv5te.ipk in the deploy area.
Quoted 7 lines
I'm not sure which package should produce this but i guess it should come from gcc-cross-4.3.3 (because its recipe version is r24.2.6 in opposite to gcc-4.3.3 r24.1.6).
Looking into the log i hit the fact that the
do_rootfsstage for the image was started *before* thedo_package_stageof gcc-cross_4.3.3.bb was succeeded.Any ideas?
PS: Bitbaking gcc-cross explicitly before bitbaking the image wors as a workaround for us at the moment.
I made some research in this and there is something i do not understand.
The task-depends.dot generated by bitbake -g (says
"console-image.do_rootfs" [label="console-image do_rootfs\n0:1.0-r0\n/home/sledz/work/angstrom-setup-scripts/sources/openembedded/recipes/images/console-image.bb"] "console-image.do_rootfs" -> "lzo-native.do_populate_sysroot"
"console-image.do_rootfs" -> "bluez4.do_populate_sysroot"
...So
do_rootfsdepends only ondo_populate_sysrootstages. But thedo_rootfsstage itself uses opkg to install the packages into the image.In my understanding this means that all the ipk files need to be available. But this is not guaranteed by the reported dependencies.
Quoted 7 lines
Quoted 1 line
A misunderstanding of mine? Or a bug?
Ping!
Did everyone read my message?
If it really is a misunderstanding of mine, please let me know. But if i'm right this seems to be a critical problem.
It sounds both strange and a correct reading of the task lists, iirc. Did we fix this in oe.dev perhaps and just need to pull a change over?
Signature · 1 line
Tom
Quoted 38 lines · Tom Rini
Quoted 37 lines · Steffen Sledz
Am 05.08.2011 09:13, schrieb Steffen Sledz:
Quoted 35 lines · Steffen Sledz
Quoted 12 lines
In the last days we switched our local development from an older oe-dev master to 2011.03-maintenance branch and hit an annoying problem.
The test builds on various developer machines were successful but the build on our continuous integration server (with exactly the same scripting) ended up with a lot of
| * satisfy_dependencies_for: Cannot satisfy the following dependencies for lighttpd: | * libstdc++6 (>= 4.3.3) * libgcc1 (>= 4.3.3) * libgcc1 (>= 4.3.3) * | * opkg_install_cmd: Cannot install package lighttpd.errors.
After some searching we found that there was no libgcc1_4.3.3-r24.2.6_armv5te.ipk in the deploy area.
Quoted 7 lines
I'm not sure which package should produce this but i guess it should come from gcc-cross-4.3.3 (because its recipe version is r24.2.6 in opposite to gcc-4.3.3 r24.1.6).
Looking into the log i hit the fact that the
do_rootfsstage for the image was started *before* thedo_package_stageof gcc-cross_4.3.3.bb was succeeded.Any ideas?
PS: Bitbaking gcc-cross explicitly before bitbaking the image wors as a workaround for us at the moment.
I made some research in this and there is something i do not understand.
The task-depends.dot generated by bitbake -g (says
"console-image.do_rootfs" [label="console-image do_rootfs\n0:1.0-r0\n/home/sledz/work/angstrom-setup-scripts/sources/openembedded/recipes/images/console-image.bb"] "console-image.do_rootfs" -> "lzo-native.do_populate_sysroot"
"console-image.do_rootfs" -> "bluez4.do_populate_sysroot"
...So
do_rootfsdepends only ondo_populate_sysrootstages. But thedo_rootfsstage itself uses opkg to install the packages into the image.In my understanding this means that all the ipk files need to be available. But this is not guaranteed by the reported dependencies.
A misunderstanding of mine? Or a bug?
Quoted 8 lines
Quoted 5 lines
Ping!
Did everyone read my message?
If it really is a misunderstanding of mine, please let me know. But if i'm right this seems to be a critical problem.
It sounds both strange and a correct reading of the task lists, iirc. Did we fix this in oe.dev perhaps and just need to pull a change over?
The mentioned dependency problem was a misinterpretation of the task-depends.dot.
Aside from the do_populate_sysroot there are the following additional dependencies:
"console-image.do_rootfs" -> "console-image.do_package_update_index_ipk"
...
"console-image.do_package_update_index_ipk" -> "gcc-cross.do_package_write_ipk"
...
"gcc-cross.do_package_write_ipk" -> "gcc-cross.do_package"
The problem seems to be located more likely somewhere inside the gcc-4.3.3-r24.1/gcc-cross-4.3.3-r24.2 recipes. There seems to be a confusion between those which leads to undetermined results.
E.g. the content of tmp.6/work/armv5te-angstrom-linux-gnueabi/gcc-cross-4.3.3-r24.2/temp/log.do_package_write_ipk.22477 is:
------------------------>snip<------------------------
Packaged contents of libstdc++-dev into /home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/libstdc++-dev_4.3.3-r24.1.6_armv5te.ipk Packaged contents of libgcc-dev into /home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/libgcc-dev_4.3.3-r24.1.6_armv5te.ipk ------------------------>snip<------------------------
The do_package_write_ipk generates ipk-file with *r24.1.6* (which is the PR of gcc) and not *r24.2.6* (which is the PR of gcc).
BTW: Which of the two recipes (gcc/gcc-cross) should provide the libgcc ipk (in my case none of it did)?
Steffen
Signature · 7 lines
DResearch Fahrzeugelektronik GmbH Otto-Schmirgal-Str. 3, 10319 Berlin, Germany Tel: +49 30 515932-237 mailto:[email protected] Fax: +49 30 515932-299 Geschäftsführer: Dr. Michael Weber, Werner Mögle; Amtsgericht Berlin Charlottenburg; HRB 130120 B; Ust.-IDNr. DE273952058
Quoted 1 line · Steffen Sledz
The
do_package_write_ipkgenerates ipk-file with *r24.1.6* (which is the PR of gcc) and not *r24.2.6* (which is the PR of gcc).
Sorry, little typo. Should be:
The do_package_write_ipk generates ipk-file with *r24.1.6* (which is the PR of gcc) and not *r24.2.6* (which is the PR of gcc-cross).
Quoted 66 lines · Steffen Sledz
Quoted 38 lines · Tom Rini
Quoted 37 lines · Steffen Sledz
Am 05.08.2011 09:13, schrieb Steffen Sledz:
Quoted 35 lines · Steffen Sledz
Quoted 12 lines
In the last days we switched our local development from an older oe-dev master to 2011.03-maintenance branch and hit an annoying problem.
The test builds on various developer machines were successful but the build on our continuous integration server (with exactly the same scripting) ended up with a lot of
| * satisfy_dependencies_for: Cannot satisfy the following dependencies for lighttpd: | * libstdc++6 (>= 4.3.3) * libgcc1 (>= 4.3.3) * libgcc1 (>= 4.3.3) * | * opkg_install_cmd: Cannot install package lighttpd.errors.
After some searching we found that there was no libgcc1_4.3.3-r24.2.6_armv5te.ipk in the deploy area.
Quoted 7 lines
I'm not sure which package should produce this but i guess it should come from gcc-cross-4.3.3 (because its recipe version is r24.2.6 in opposite to gcc-4.3.3 r24.1.6).
Looking into the log i hit the fact that the
do_rootfsstage for the image was started *before* thedo_package_stageof gcc-cross_4.3.3.bb was succeeded.Any ideas?
PS: Bitbaking gcc-cross explicitly before bitbaking the image wors as a workaround for us at the moment.
I made some research in this and there is something i do not understand.
The task-depends.dot generated by bitbake -g (says
"console-image.do_rootfs" [label="console-image do_rootfs\n0:1.0-r0\n/home/sledz/work/angstrom-setup-scripts/sources/openembedded/recipes/images/console-image.bb"] "console-image.do_rootfs" -> "lzo-native.do_populate_sysroot"
"console-image.do_rootfs" -> "bluez4.do_populate_sysroot"
...So
do_rootfsdepends only ondo_populate_sysrootstages. But thedo_rootfsstage itself uses opkg to install the packages into the image.In my understanding this means that all the ipk files need to be available. But this is not guaranteed by the reported dependencies.
A misunderstanding of mine? Or a bug?
Quoted 8 lines
Quoted 5 lines
Ping!
Did everyone read my message?
If it really is a misunderstanding of mine, please let me know. But if i'm right this seems to be a critical problem.
It sounds both strange and a correct reading of the task lists, iirc. Did we fix this in oe.dev perhaps and just need to pull a change over?
The mentioned dependency problem was a misinterpretation of the task-depends.dot.
Aside from the
do_populate_sysrootthere are the following additional dependencies:"console-image.do_rootfs" -> "console-image.do_package_update_index_ipk"
...
"console-image.do_package_update_index_ipk" -> "gcc-cross.do_package_write_ipk"
...
"gcc-cross.do_package_write_ipk" -> "gcc-cross.do_package"The problem seems to be located more likely somewhere inside the gcc-4.3.3-r24.1/gcc-cross-4.3.3-r24.2 recipes. There seems to be a confusion between those which leads to undetermined results.
E.g. the content of tmp.6/work/armv5te-angstrom-linux-gnueabi/gcc-cross-4.3.3-r24.2/temp/log.do_package_write_ipk.22477 is:
------------------------>snip<------------------------
Packaged contents of libstdc++-dev into/home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/libstdc++-dev_4.3.3-r24.1.6_armv5te.ipkPackaged contents of libgcc-dev into/home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/libgcc-dev_4.3.3-r24.1.6_armv5te.ipk------------------------>snip<------------------------
Quoted 3 lines
The
do_package_write_ipkgenerates ipk-file with *r24.1.6* (which is the PR of gcc) and not *r24.2.6* (which is the PR of gcc).BTW: Which of the two recipes (gcc/gcc-cross) should provide the libgcc ipk (in my case none of it did)?
After some bb recipe updates there is a new error of this type.
| * check_data_file_clashes: Package libgcc-s-dev wants to install file /CACHE/hudson/jobs/HydraIP_Linux_release_image/workspace/OE/tmp.6/rootfs/hydraip-hipox-devimage/usr/lib/libgcc_s.so | But that file is already provided by package * libgcc-dev | * opkg_install_cmd: Cannot install package libsqlite3-dev.
:(
Signature · 7 lines
DResearch Fahrzeugelektronik GmbH Otto-Schmirgal-Str. 3, 10319 Berlin, Germany Tel: +49 30 515932-237 mailto:[email protected] Fax: +49 30 515932-299 Geschäftsführer: Dr. Michael Weber, Werner Mögle; Amtsgericht Berlin Charlottenburg; HRB 130120 B; Ust.-IDNr. DE273952058
Quoted 3 lines · Steffen Sledz
Quoted 2 lines · Steffen Sledz
Quoted 1 line · Tom Rini
On Sun, Aug 7, 2011 at 10:37 AM, Steffen Sledz <[email protected]>
Quoted 57 lines · wrote:
Quoted 57 lines
Quoted 57 lines
Quoted 57 lines
Am 05.08.2011 09:13, schrieb Steffen Sledz:
Quoted 46 lines · Steffen Sledz
Quoted 21 lines
In the last days we switched our local development from an older oe-dev master to 2011.03-maintenance branch and hit an annoying problem.
The test builds on various developer machines were successful but the build on our continuous integration server (with exactly the same scripting) ended up with a lot of>>>>>
| * satisfy_dependencies_for: Cannot satisfy the following | dependencies for lighttpd: * libstdc++6 (>= 4.3.3) * | libgcc1 (>= 4.3.3) * libgcc1 (>= 4.3.3) * * | opkg_install_cmd: Cannot install package lighttpd.errors.
After some searching we found that there was no libgcc1_4.3.3-r24.2.6_armv5te.ipk in the deploy area.
I'm not sure which package should produce this but i guess it should come from gcc-cross-4.3.3 (because its recipe version is r24.2.6 in opposite to gcc-4.3.3 r24.1.6).
Quoted 8 lines
Looking into the log i hit the fact that the
do_rootfsstage for the image was started *before* thedo_package_stageof gcc-cross_4.3.3.bb was succeeded.Any ideas?
PS: Bitbaking gcc-cross explicitly before bitbaking the image wors as a workaround for us at the moment.>>>>
I made some research in this and there is something i do not understand.
The task-depends.dot generated by bitbake -g (says
"console-image.do_rootfs" [label="console-image do_rootfs\n0:1.0-r0\n/home/sledz/work/angstrom-setup-scripts/ sources/openembedded/recipes/images/console-image.bb"] "console-image.do_rootfs" -> "lzo-native.do_populate_sysroot" "console-image.do_rootfs" -> "bluez4.do_populate_sysroot"
...So
do_rootfsdepends only ondo_populate_sysrootstages. But thedo_rootfsstage itself uses opkg to install the packages into the image.Quoted 5 lines
In my understanding this means that all the ipk files need to be available. But this is not guaranteed by the reported dependencies.
A misunderstanding of mine? Or a bug?
Ping!
Did everyone read my message?
Quoted 8 lines
Quoted 8 lines
Quoted 4 lines
Quoted 2 lines
If it really is a misunderstanding of mine, please let me know. But if i'm right this seems to be a critical problem.>>
It sounds both strange and a correct reading of the task lists, iirc. Did we fix this in oe.dev perhaps and just need to pull a change over?
The mentioned dependency problem was a misinterpretation of the task-depends.dot.>
Aside from thedo_populate_sysrootthere are the following additional
dependencies:
Quoted 22 lines
Quoted 22 lines
"console-image.do_rootfs" -> "console-image.do_package_update_index_ipk"
...
"console-image.do_package_update_index_ipk" -> "gcc-cross.do_package_write_ipk" ... "gcc-cross.do_package_write_ipk" -> "gcc-cross.do_package"The problem seems to be located more likely somewhere inside the gcc-4.3.3-r24.1/gcc-cross-4.3.3-r24.2 recipes. There seems to be a confusion between those which leads to undetermined results.
E.g. the content of tmp.6/work/armv5te-angstrom-linux-gnueabi/gcc-cross-4.3.3-r24.2/temp/lo g.do_package_write_ipk.22477 is:
------------------------>snip<------------------------ Packaged contents of libstdc++-dev into/home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/libstdc++-dev_4.3.3-r24.1.6_armv5te.ipk Packaged contents of libgcc-dev into/home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/libgcc-dev_4.3.3-r24.1.6_armv5te.ipk ------------------------>snip<------------------------
Quoted 5 lines
Quoted 5 lines
The
do_package_write_ipkgenerates ipk-file with *r24.1.6* (which is the PR of gcc) and not *r24.2.6* (which is the PR of gcc).BTW: Which of the two recipes (gcc/gcc-cross) should provide the libgcc ipk (in my case none of it did)?
It should be probided by both. We provide it in gcc-cross so that people dont have to build target gcc just for libgcc.
Quoted 9 lines
After some bb recipe updates there is a new error of this type.
| * check_data_file_clashes: Package libgcc-s-dev wants to install file | /CACHE/hudson/jobs/HydraIP_Linux_release_image/workspace/OE/tmp.6/roo | tfs/hydraip-hipox-devimage/usr/lib/libgcc_s.so| | But that file is already provided by package * libgcc-dev | | * opkg_install_cmd: Cannot install package libsqlite3-dev.:
what changes did you do ?
libgcc dependencies get encoded into ipks
so you might have to regenerate all ipks for best results.
I dont know yet what the problem you are seeing could be. I do not see it here locally
Quoted 1 line
:(
Quoted 96 lines · Khem Raj
Quoted 3 lines · Steffen Sledz
Quoted 2 lines · Steffen Sledz
Quoted 1 line · Tom Rini
On Sun, Aug 7, 2011 at 10:37 AM, Steffen Sledz <[email protected]>
Quoted 57 lines · wrote:
Quoted 57 lines
Quoted 57 lines
Quoted 57 lines
Am 05.08.2011 09:13, schrieb Steffen Sledz:
Quoted 46 lines · Steffen Sledz
Quoted 21 lines
In the last days we switched our local development from an older oe-dev master to 2011.03-maintenance branch and hit an annoying problem.
The test builds on various developer machines were successful but the build on our continuous integration server (with exactly the same scripting) ended up with a lot of>>>>>
| * satisfy_dependencies_for: Cannot satisfy the following | dependencies for lighttpd: * libstdc++6 (>= 4.3.3) * | libgcc1 (>= 4.3.3) * libgcc1 (>= 4.3.3) * * | opkg_install_cmd: Cannot install package lighttpd.errors.
After some searching we found that there was no libgcc1_4.3.3-r24.2.6_armv5te.ipk in the deploy area.
I'm not sure which package should produce this but i guess it should come from gcc-cross-4.3.3 (because its recipe version is r24.2.6 in opposite to gcc-4.3.3 r24.1.6).
Quoted 8 lines
Looking into the log i hit the fact that the
do_rootfsstage for the image was started *before* thedo_package_stageof gcc-cross_4.3.3.bb was succeeded.Any ideas?
PS: Bitbaking gcc-cross explicitly before bitbaking the image wors as a workaround for us at the moment.>>>>
I made some research in this and there is something i do not understand.
The task-depends.dot generated by bitbake -g (says
"console-image.do_rootfs" [label="console-image do_rootfs\n0:1.0-r0\n/home/sledz/work/angstrom-setup-scripts/ sources/openembedded/recipes/images/console-image.bb"] "console-image.do_rootfs" -> "lzo-native.do_populate_sysroot" "console-image.do_rootfs" -> "bluez4.do_populate_sysroot"
...So
do_rootfsdepends only ondo_populate_sysrootstages. But thedo_rootfsstage itself uses opkg to install the packages into the image.Quoted 5 lines
In my understanding this means that all the ipk files need to be available. But this is not guaranteed by the reported dependencies.
A misunderstanding of mine? Or a bug?
Ping!
Did everyone read my message?
Quoted 8 lines
Quoted 8 lines
Quoted 4 lines
Quoted 2 lines
If it really is a misunderstanding of mine, please let me know. But if i'm right this seems to be a critical problem.>>
It sounds both strange and a correct reading of the task lists, iirc. Did we fix this in oe.dev perhaps and just need to pull a change over?
The mentioned dependency problem was a misinterpretation of the task-depends.dot.>
Aside from thedo_populate_sysrootthere are the following additionaldependencies:
Quoted 25 lines
Quoted 25 lines
"console-image.do_rootfs" -> "console-image.do_package_update_index_ipk"
...
"console-image.do_package_update_index_ipk" -> "gcc-cross.do_package_write_ipk" ... "gcc-cross.do_package_write_ipk" -> "gcc-cross.do_package"The problem seems to be located more likely somewhere inside the gcc-4.3.3-r24.1/gcc-cross-4.3.3-r24.2 recipes. There seems to be a confusion between those which leads to undetermined results.
E.g. the content of tmp.6/work/armv5te-angstrom-linux-gnueabi/gcc-cross-4.3.3-r24.2/temp/lo g.do_package_write_ipk.22477 is:
------------------------>snip<------------------------ Packaged contents of libstdc++-dev into/home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/libstdc++-dev_4.3.3-r24.1.6_armv5te.ipk Packaged contents of libgcc-dev into/home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/libgcc-dev_4.3.3-r24.1.6_armv5te.ipk ------------------------>snip<------------------------The
do_package_write_ipkgenerates ipk-file with *r24.1.6* (which is the PR of gcc) and not *r24.2.6* (which is the PR of gcc).
Quoted 6 lines
Quoted 2 lines
Quoted 2 lines
BTW: Which of the two recipes (gcc/gcc-cross) should provide the libgcc ipk (in my case none of it did)?
It should be probided by both. We provide it in gcc-cross so that people dont have to build target gcc just for libgcc.
But it isn't provided. :(
Here are related log files of both from a clean build (HEAD of 2011.03-maintenance branch, no modification to the gcc recipes):
----------------------->snip<----------------------
Quoted 1 line
cat tmp.6/work/armv5te-angstrom-linux-gnueabi/gcc-4.3.3-r24.1/temp/log.do_package_write_ipk.4930
Packaged contents of gcc into /home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/gcc_4.3.3-r24.1.6_armv5te.ipk Packaged contents of gcc-symlinks into /home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/gcc-symlinks_4.3.3-r24.1.6_armv5te.ipk Packaged contents of g++ into /home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/g++_4.3.3-r24.1.6_armv5te.ipk Packaged contents of g++-symlinks into /home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/g++-symlinks_4.3.3-r24.1.6_armv5te.ipk Packaged contents of cpp into /home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/cpp_4.3.3-r24.1.6_armv5te.ipk Packaged contents of cpp-symlinks into /home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/cpp-symlinks_4.3.3-r24.1.6_armv5te.ipk Packaged contents of g77-symlinks into /home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/g77-symlinks_4.3.3-r24.1.6_armv5te.ipk Packaged contents of gfortran into /home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/gfortran_4.3.3-r24.1.6_armv5te.ipk Packaged contents of gfortran-symlinks into /home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/gfortran-symlinks_4.3.3-r24.1.6_armv5te.ipk Packaged contents of gcov into /home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/gcov_4.3.3-r24.1.6_armv5te.ipk Packaged contents of libgcc-dev into /home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/libgcc-dev_4.3.3-r24.1.6_armv5te.ipk Packaged contents of libstdc++-dev into /home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/libstdc++-dev_4.3.3-r24.1.6_armv5te.ipk Packaged contents of libgfortran-dev into /home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/libgfortran-dev_4.3.3-r24.1.6_armv5te.ipk Packaged contents of gcc-doc into /home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/gcc-doc_4.3.3-r24.1.6_armv5te.ipk
Quoted 1 line
cat tmp.6/work/armv5te-angstrom-linux-gnueabi/gcc-cross-4.3.3-r24.2/temp/log.do_package_write_ipk.22477
Packaged contents of libstdc++-dev into /home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/libstdc++-dev_4.3.3-r24.1.6_armv5te.ipk Packaged contents of libgcc-dev into /home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/libgcc-dev_4.3.3-r24.1.6_armv5te.ipk ----------------------->snip<----------------------
I've frozen this build and can provide additional info if you need more.
Quoted 11 lines
Quoted 9 lines
After some bb recipe updates there is a new error of this type.
| * check_data_file_clashes: Package libgcc-s-dev wants to install file | /CACHE/hudson/jobs/HydraIP_Linux_release_image/workspace/OE/tmp.6/roo | tfs/hydraip-hipox-devimage/usr/lib/libgcc_s.so| | But that file is already provided by package * libgcc-dev | | * opkg_install_cmd: Cannot install package libsqlite3-dev.:
what changes did you do ?
Only updated the version of sqlite3 by removing DP="-1" from the sqlite3_3.7.5.bb recipe.
Quoted 5 lines
libgcc dependencies get encoded into ipks
so you might have to regenerate all ipks for best results.I dont know yet what the problem you are seeing could be. I do not see it here locally
Just a guess. Could it be a problem with tasks running in parallel.
Regards,
Steffen
Signature · 7 lines
DResearch Fahrzeugelektronik GmbH Otto-Schmirgal-Str. 3, 10319 Berlin, Germany Tel: +49 30 515932-237 mailto:[email protected] Fax: +49 30 515932-299 Geschäftsführer: Dr. Michael Weber, Werner Mögle; Amtsgericht Berlin Charlottenburg; HRB 130120 B; Ust.-IDNr. DE273952058
Quoted 125 lines · Steffen Sledz
Quoted 96 lines · Khem Raj
Quoted 3 lines · Steffen Sledz
Quoted 2 lines · Steffen Sledz
Quoted 1 line · Tom Rini
On Sun, Aug 7, 2011 at 10:37 AM, Steffen Sledz <[email protected]>
Quoted 57 lines · wrote:
Quoted 57 lines
Quoted 57 lines
Quoted 57 lines
Am 05.08.2011 09:13, schrieb Steffen Sledz:
Quoted 46 lines · Steffen Sledz
Quoted 21 lines
In the last days we switched our local development from an older oe-dev master to 2011.03-maintenance branch and hit an annoying problem.
The test builds on various developer machines were successful but the build on our continuous integration server (with exactly the same scripting) ended up with a lot of>>>>>
| * satisfy_dependencies_for: Cannot satisfy the following | dependencies for lighttpd: * libstdc++6 (>= 4.3.3) * | libgcc1 (>= 4.3.3) * libgcc1 (>= 4.3.3) * * | opkg_install_cmd: Cannot install package lighttpd.errors.
After some searching we found that there was no libgcc1_4.3.3-r24.2.6_armv5te.ipk in the deploy area.
I'm not sure which package should produce this but i guess it should come from gcc-cross-4.3.3 (because its recipe version is r24.2.6 in opposite to gcc-4.3.3 r24.1.6).
Quoted 8 lines
Looking into the log i hit the fact that the
do_rootfsstage for the image was started *before* thedo_package_stageof gcc-cross_4.3.3.bb was succeeded.Any ideas?
PS: Bitbaking gcc-cross explicitly before bitbaking the image wors as a workaround for us at the moment.>>>>
I made some research in this and there is something i do not understand.
The task-depends.dot generated by bitbake -g (says
"console-image.do_rootfs" [label="console-image do_rootfs\n0:1.0-r0\n/home/sledz/work/angstrom-setup-scripts/ sources/openembedded/recipes/images/console-image.bb"] "console-image.do_rootfs" -> "lzo-native.do_populate_sysroot" "console-image.do_rootfs" -> "bluez4.do_populate_sysroot"
...So
do_rootfsdepends only ondo_populate_sysrootstages. But thedo_rootfsstage itself uses opkg to install the packages into the image.Quoted 5 lines
In my understanding this means that all the ipk files need to be available. But this is not guaranteed by the reported dependencies.
A misunderstanding of mine? Or a bug?
Ping!
Did everyone read my message?
Quoted 8 lines
Quoted 8 lines
Quoted 4 lines
Quoted 2 lines
If it really is a misunderstanding of mine, please let me know. But if i'm right this seems to be a critical problem.>>
It sounds both strange and a correct reading of the task lists, iirc. Did we fix this in oe.dev perhaps and just need to pull a change over?
The mentioned dependency problem was a misinterpretation of the task-depends.dot.>
Aside from thedo_populate_sysrootthere are the following additionaldependencies:
Quoted 25 lines
Quoted 25 lines
"console-image.do_rootfs" -> "console-image.do_package_update_index_ipk"
...
"console-image.do_package_update_index_ipk" -> "gcc-cross.do_package_write_ipk" ... "gcc-cross.do_package_write_ipk" -> "gcc-cross.do_package"The problem seems to be located more likely somewhere inside the gcc-4.3.3-r24.1/gcc-cross-4.3.3-r24.2 recipes. There seems to be a confusion between those which leads to undetermined results.
E.g. the content of tmp.6/work/armv5te-angstrom-linux-gnueabi/gcc-cross-4.3.3-r24.2/temp/lo g.do_package_write_ipk.22477 is:
------------------------>snip<------------------------ Packaged contents of libstdc++-dev into/home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/libstdc++-dev_4.3.3-r24.1.6_armv5te.ipk Packaged contents of libgcc-dev into/home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/libgcc-dev_4.3.3-r24.1.6_armv5te.ipk ------------------------>snip<------------------------The
do_package_write_ipkgenerates ipk-file with *r24.1.6* (which is the PR of gcc) and not *r24.2.6* (which is the PR of gcc).Quoted 6 lines
Quoted 2 lines
Quoted 2 lines
BTW: Which of the two recipes (gcc/gcc-cross) should provide the libgcc ipk (in my case none of it did)?
It should be probided by both. We provide it in gcc-cross so that people dont have to build target gcc just for libgcc.
But it isn't provided. :(
Here are related log files of both from a clean build (HEAD of 2011.03-maintenance branch, no modification to the gcc recipes):
----------------------->snip<----------------------
Quoted 1 line
cat tmp.6/work/armv5te-angstrom-linux-gnueabi/gcc-4.3.3-r24.1/temp/log.do_package_write_ipk.4930
Packaged contents of gcc into
/home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/gcc_4.3.3-r24.1.6_armv5te.ipkPackaged contents of gcc-symlinks into/home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/gcc-symlinks_4.3.3-r24.1.6_armv5te.ipkPackaged contents of g++ into/home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/g++_4.3.3-r24.1.6_armv5te.ipkPackaged contents of g++-symlinks into/home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/g++-symlinks_4.3.3-r24.1.6_armv5te.ipkPackaged contents of cpp into/home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/cpp_4.3.3-r24.1.6_armv5te.ipkPackaged contents of cpp-symlinks into/home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/cpp-symlinks_4.3.3-r24.1.6_armv5te.ipkPackaged contents of g77-symlinks into/home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/g77-symlinks_4.3.3-r24.1.6_armv5te.ipkPackaged contents of gfortran into/home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/gfortran_4.3.3-r24.1.6_armv5te.ipkPackaged contents of gfortran-symlinks into/home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/gfortran-symlinks_4.3.3-r24.1.6_armv5te.ipkPackaged contents of gcov into/home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/gcov_4.3.3-r24.1.6_armv5te.ipkPackaged contents of libgcc-dev into/home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/libgcc-dev_4.3.3-r24.1.6_armv5te.ipkPackaged contents of libstdc++-dev into/home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/libstdc++-dev_4.3.3-r24.1.6_armv5te.ipkPackaged contents of libgfortran-dev into/home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/libgfortran-dev_4.3.3-r24.1.6_armv5te.ipkPackaged contents of gcc-doc into/home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/gcc-doc_4.3.3-r24.1.6_armv5te.ipk
Quoted 6 lines
Quoted 1 line
cat tmp.6/work/armv5te-angstrom-linux-gnueabi/gcc-cross-4.3.3-r24.2/temp/log.do_package_write_ipk.22477
Packaged contents of libstdc++-dev into
/home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/libstdc++-dev_4.3.3-r24.1.6_armv5te.ipkPackaged contents of libgcc-dev into/home/sledz/work/HydraIP/OE/tmp.6/deploy/glibc/ipk/armv5te/libgcc-dev_4.3.3-r24.1.6_armv5te.ipk----------------------->snip<----------------------I've frozen this build and can provide additional info if you need more.
I can give an all clear at least for this problem.
We found that it only occurred in conjunction with using libunwind (which we used to have a working backtrace). After eliminating it none of the mentioned problems was seens again.
Quoted 9 lines
Quoted 9 lines
Quoted 9 lines
After some bb recipe updates there is a new error of this type.
| * check_data_file_clashes: Package libgcc-s-dev wants to install file | /CACHE/hudson/jobs/HydraIP_Linux_release_image/workspace/OE/tmp.6/roo | tfs/hydraip-hipox-devimage/usr/lib/libgcc_s.so| | But that file is already provided by package * libgcc-dev | | * opkg_install_cmd: Cannot install package libsqlite3-dev.:
This problems seems to be different (see Steve Sakomans mails).
Quoted 11 lines
Quoted 1 line
what changes did you do ?
Only updated the version of sqlite3 by removing DP="-1" from the sqlite3_3.7.5.bb recipe.
Quoted 5 lines
libgcc dependencies get encoded into ipks
so you might have to regenerate all ipks for best results.I dont know yet what the problem you are seeing could be. I do not see it here locally
Just a guess. Could it be a problem with tasks running in parallel.
Regards,
Steffen
PS: I'm i bit disappointed in the weak help from the maintainers in solving the problem. :( But i do not bear grudges[1]. ;-)
[1] Hope that's correct english.
Signature · 7 lines
DResearch Fahrzeugelektronik GmbH Otto-Schmirgal-Str. 3, 10319 Berlin, Germany Tel: +49 30 515932-237 mailto:[email protected] Fax: +49 30 515932-299 Geschäftsführer: Dr. Michael Weber, Werner Mögle; Amtsgericht Berlin Charlottenburg; HRB 130120 B; Ust.-IDNr. DE273952058
Quoted 6 lines · Steffen Sledz
Regards,
SteffenPS: I'm i bit disappointed in the weak help from the maintainers in solving the problem. :( But i do not bear grudges[1]. ;-)
[1] Hope that's correct english.
personally I would have liked to help more but was short on time and since I could not reproduce the problem at my end. It was not obvious to me what the problem was.
Quoted 9 lines · Khem Raj
Quoted 6 lines · Steffen Sledz
Regards,
SteffenPS: I'm i bit disappointed in the weak help from the maintainers in solving the problem. :( But i do not bear grudges[1]. ;-)
[1] Hope that's correct english.
personally I would have liked to help more but was short on time and since I could not reproduce the problem at my end. It was not obvious to me what the problem was.
Never mind. It's OK.
Signature · 7 lines
DResearch Fahrzeugelektronik GmbH Otto-Schmirgal-Str. 3, 10319 Berlin, Germany Tel: +49 30 515932-237 mailto:[email protected] Fax: +49 30 515932-299 Geschäftsführer: Dr. Michael Weber, Werner Mögle; Amtsgericht Berlin Charlottenburg; HRB 130120 B; Ust.-IDNr. DE273952058
Quoted 5 lines · Khem Raj
Quoted 4 lines · Steffen Sledz
Regards,
Steffen
Quoted 8 lines
Quoted 4 lines
PS: I'm i bit disappointed in the weak help from the maintainers in solving the problem. :( But i do not bear grudges[1]. ;-)
[1] Hope that's correct english.
personally I would have liked to help more but was short on time and since I could not reproduce the problem at my end. It was not obvious to me what the problem was.
Khem,
Is there an easy way to tell OE to ignore a recommends for either libgcc-dev or libgcc-s-dev? The issue seems to be (at least on my machine) that packages more or less randomly recommend one or the other. Then the image build fails because the two packages supply the same file.
Steve
Quoted 15 lines · Steve Sakoman
Quoted 5 lines · Khem Raj
Quoted 4 lines · Steffen Sledz
Regards,
SteffenQuoted 8 lines
Quoted 4 lines
PS: I'm i bit disappointed in the weak help from the maintainers in solving the problem. :( But i do not bear grudges[1]. ;-)
[1] Hope that's correct english.
personally I would have liked to help more but was short on time and since I could not reproduce the problem at my end. It was not obvious to me what the problem was.
Quoted 7 lines
Khem,
Is there an easy way to tell OE to ignore a recommends for either libgcc-dev or libgcc-s-dev? The issue seems to be (at least on my machine) that packages more or less randomly recommend one or the other. Then the image build fails because the two packages supply the same file.
I dont think there is a way out. I think one of those 2 packages should be stopped from building
Quoted 7 lines
Steve
_______________________________________________
Openembedded-devel mailing list
[email protected]
http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel
Quoted 8 lines · Khem Raj
Quoted 5 lines
Is there an easy way to tell OE to ignore a recommends for either libgcc-dev or libgcc-s-dev? The issue seems to be (at least on my machine) that packages more or less randomly recommend one or the other. Then the image build fails because the two packages supply the same file.
I dont think there is a way out. I think one of those 2 packages should be stopped from building
Which do you think should be disabled?
Steve
Quoted 4 lines · Steve Sakoman
Which do you think should be disabled?
libgcc-dev
Quoted 5 lines · Khem Raj
Quoted 2 lines · Steve Sakoman
Which do you think should be disabled?
libgcc-dev
Thanks! I'll do a test clean build with the generation of this package disabled.
Steve
Quoted 5 lines · Steffen Sledz
After some bb recipe updates there is a new error of this type.
| * check_data_file_clashes: Package libgcc-s-dev wants to install file /CACHE/hudson/jobs/HydraIP_Linux_release_image/workspace/OE/tmp.6/rootfs/hydraip-hipox-devimage/usr/lib/libgcc_s.so | But that file is already provided by package * libgcc-dev | * opkg_install_cmd: Cannot install package libsqlite3-dev.
Others are also running into this issue! I've seen messages from Philip Balister regarding this same problem with his Ettus images:
http://comments.gmane.org/gmane.comp.hardware.usrp.e100/1245
I see it with the GNOME images I build:
| * check_data_file_clashes: Package libgcc-dev wants to install file /media/data/gnome-r13-tmp/rootfs/sakoman-gnome-image/usr/lib/libgcc_s.so
| But that file is already provided by package * libgcc-s-dev
In looking through the packages in my deploy directory I see a split between packages that depend on libgcc-dev and libgcc-s-dev.
Any ideas on how to fix this issue?
Steve
Quoted 2 lines · Steve Sakoman
In looking through the packages in my deploy directory I see a split between packages that depend on libgcc-dev and libgcc-s-dev.
A typo above -- it is a "recommends" for libgcc-dev and libgcc-s-dev in the packages, not a "depends".
Quoted 4 lines
Any ideas on how to fix this issue?
Steve