Bug 2772 - [multilib] no provider for lib64-ossp-uuid/sato-image
Summary: [multilib] no provider for lib64-ossp-uuid/sato-image
Status: VERIFIED INVALID
Alias: None
Product: OE-Core
Classification: Build System, Metadata & Runtime
Component: core (show other bugs)
Version: 1.3
Hardware: x86 Multiple
: Medium normal
Target Milestone: 1.3 M3
Assignee: Radu Moisan
QA Contact:
URL:
Whiteboard:
Depends on:
Blocks:
 
Reported: 2012-07-17 04:26 UTC by Hongna Xu
Modified: 2012-07-31 09:18 UTC (History)
6 users (show)

See Also:
OS type for building Yocto: ---
Type of Regression: ---
Verified:
Documentation change: ---


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Hongna Xu 2012-07-17 04:26:01 UTC
tree/branch: poky/master
commit: eb0cb7e8234f5d2e5623406e9660be91cf52f65e

With latest master, multilib build against sato images could not pass. It reports no provider for ib64-ossp-uuid

the log shows:
**********
Build Configuration:
BB_VERSION        = "1.15.2"
TARGET_ARCH       = "i586"
TARGET_OS         = "linux"
MACHINE           = "qemux86"
DISTRO            = "poky"
DISTRO_VERSION    = "1.2+snapshot-20120717"
TUNE_FEATURES     = "m32 i586"
TARGET_FPU        = ""
meta             
meta-yocto        = "master:eb0cb7e8234f5d2e5623406e9660be91cf52f65e"

NOTE: Resolving any missing task queue dependencies
ERROR: Nothing PROVIDES 'lib64-ossp-uuid' (but virtual:multilib:lib64:/home/build/poky/meta/recipes-devtools/rpm/rpm_5.4.9.bb DEPENDS on or otherwise requires it)
NOTE: Runtime target 'lib64-rpm' is unbuildable, removing...
Missing or unbuildable dependency chain was: ['lib64-rpm', 'lib64-ossp-uuid']
ERROR: Required build target 'lib64-core-image-sato' has no buildable providers.
Missing or unbuildable dependency chain was: ['lib64-core-image-sato', 'lib64-rpm', 'lib64-ossp-uuid']

Summary: There was 1 WARNING message shown.
Summary: There were 2 ERROR messages shown, returning a non-zero exit code.
***************************
Comment 1 Radu Moisan 2012-07-20 09:06:37 UTC
followed the wiki instructions for multilib and then run "MACHINE=qemux86-64 bitbake core-image-sato" - everything went fine, the build finshed succesfully.
Comment 2 Hongna Xu 2012-07-23 01:47:05 UTC
(In reply to comment #1)
> followed the wiki instructions for multilib and then run "MACHINE=qemux86-64
> bitbake core-image-sato" - everything went fine, the build finshed
> succesfully.

which wiki do you follow? and if you ready from my error information, you will find that the recipe i run is not core-image-sato but lib64-core-image-sato, before that you should add the lines below to conf/local.con
*************
MACHINE = "qemux86-64"
require conf/multilib.conf
MULTILIBS = "multilib:lib32"
DEFAULTTUNE_virtclass-multilib-lib32 = "x86"
**************

if you can reproduce it, please reopen it, thanks.
Comment 3 Radu Moisan 2012-07-23 06:10:18 UTC
I followed the instructions found at Yocto Wiki -> Projects -> Multilib
In the future, please add "steps to reproduce", this should be the right way to fill in a bug.
I will reexamine this.
Comment 4 Radu Moisan 2012-07-23 07:41:34 UTC
(In reply to comment #2)
> and if you ready from my error information, you
> will find that the recipe i run is not core-image-sato but
> lib64-core-image-sato, before that you should add the lines below to
> conf/local.con
> *************
> MACHINE = "qemux86-64"
> require conf/multilib.conf
> MULTILIBS = "multilib:lib32"
> DEFAULTTUNE_virtclass-multilib-lib32 = "x86"
> **************
> 
> if you can reproduce it, please reopen it, thanks.

I asked around but could not get a straight answer, what is the difference between "MACHINE=quemux86-64 bitbake core-image-sato" and "MACHINE=quemux86-64 bitbake lib64-core-image-sato". The answer I got so far says they all are the same thing. In my understanding, once you set "MACHINE=quemux86-64" the default would be to build the 64bit libraries.
Comment 5 Hongna Xu 2012-07-23 07:57:54 UTC
hi, Moisan
   you are right, they are the same thing.and i am sorry for the wrong settings in comment 2, my real intention is to build lib64 packages during bitbake qemux86 sato image, so the settings should be:
MACHINE = "qemux86"
require conf/multilib.conf
MULTILIBS = "multilib:lib64"
DEFAULTTUNE_virtclass-multilib-lib64 = "x86-64"

then build lib64-core-image-sato.
Comment 6 Radu Moisan 2012-07-23 10:42:03 UTC
(In reply to comment #5)
> hi, Moisan
>    you are right, they are the same thing.and i am sorry for the wrong
> settings in comment 2, my real intention is to build lib64 packages during
> bitbake qemux86 sato image, so the settings should be:
> MACHINE = "qemux86"
> require conf/multilib.conf
> MULTILIBS = "multilib:lib64"
> DEFAULTTUNE_virtclass-multilib-lib64 = "x86-64"
> 
> then build lib64-core-image-sato.

what is the point of this? what are you tring to achive with this? It doesn't make musch sense to me.
Comment 7 Jiajun Xu 2012-07-26 02:43:44 UTC
(In reply to comment #6)
> (In reply to comment #5)
> hi, Moisan
>    you are right, they are the same
> thing.and i am sorry for the wrong
> settings in comment 2, my real
> intention is to build lib64 packages during
> bitbake qemux86 sato image, so
> the settings should be:
> MACHINE = "qemux86"
> require conf/multilib.conf
>
> MULTILIBS = "multilib:lib64"
> DEFAULTTUNE_virtclass-multilib-lib64 =
> "x86-64"
> 
> then build lib64-core-image-sato.

what is the point of this?
> what are you tring to achive with this? It doesn't make musch sense to me.

The case is to test if lib64 packages building could pass against qemux86 MACHINE. Similarly, we will also test lib32 building against qemux86-64.
Comment 8 Radu Moisan 2012-07-27 08:16:47 UTC
Asked Richard's opinion about this as well as Bogdan's and they agree it does not make much sens what you are tring to achieve with this. Moving this to NEEDINFO for further analisys.
Comment 9 Jiajun Xu 2012-07-29 12:00:13 UTC
(In reply to comment #8)
> Asked Richard's opinion about this as well as Bogdan's and they agree it
> does not make much sens what you are tring to achieve with this. Moving this
> to NEEDINFO for further analisys.

If Richard thinks no need to test such combination, I am OK to close it. 

Laurentiu,
Let's remove all such cases for sato and lsb multilib in future testing.
Comment 10 Radu Moisan 2012-07-31 09:15:46 UTC
(In reply to comment #9)
> (In reply to comment #8)
> > Asked Richard's opinion about this as well as Bogdan's and they agree it
> > does not make much sens what you are tring to achieve with this. Moving this
> > to NEEDINFO for further analisys.
> 
> If Richard thinks no need to test such combination, I am OK to close it. 
> 
> Laurentiu,
> Let's remove all such cases for sato and lsb multilib in future testing.

should be fine, in case there still are doubts, Laurentiu can talk with Richard next week.
Comment 11 Laurentiu Serban 2012-07-31 09:18:15 UTC
Setting this to verified