[OE-core] build logs of all images in or-core

Richard Purdie richard.purdie at linuxfoundation.org
Thu Nov 29 08:35:02 UTC 2012


I had a look through these and we can group them as follows:

Unsupported configuration and sensible error given to user:

  [x11 instead of directfb enabled]
  eglibc.qemuarm.core-image-gtk-directfb.log
  eglibc.qemumips.core-image-gtk-directfb.log
  eglibc.qemuppc.core-image-gtk-directfb.log
  eglibc.qemux86-64.core-image-gtk-directfb.log
  eglibc.qemux86.core-image-gtk-directfb.log

  [live image doesn't work on architecture]
  eglibc.qemuarm.core-image-minimal-initramfs.log
  eglibc.qemumips.core-image-minimal-initramfs.log
  eglibc.qemuppc.core-image-minimal-initramfs.log

  [Built ok but was misconfigured (two kernels)]
  eglibc.qemuarm.core-image-rt-sdk.log
  eglibc.qemuarm.core-image-rt.log
  eglibc.qemux86-64.core-image-rt-sdk.log
  eglibc.qemux86-64.core-image-rt.log
  eglibc.qemux86.core-image-rt-sdk.log
  eglibc.qemux86.core-image-rt.log

  [No linux-yocto-rt]
  eglibc.qemumips.core-image-rt-sdk.log
  eglibc.qemumips.core-image-rt.log
  eglibc.qemuppc.core-image-rt-sdk.log
  eglibc.qemuppc.core-image-rt.log	

In the above list, I would say that the two kernels being built is a bug
we should really fix, the only way to fix it is to make the bitbake
warning about multiple providers an error.

The "interesting" build failures:

  [download issue, enable to fetch sources]
  eglibc.qemuarm.core-image-lsb-sdk.log

  [weird license build failures]
  eglibc.qemuarm.core-image-minimal.log
  eglibc.qemumips.core-image-minimal-dev.log
  eglibc.qemumips.core-image-sato-dev.log
  eglibc.qemuppc.core-image-lsb-dev.log
  eglibc.qemuppc.core-image-minimal-dev.log
  eglibc.qemuppc.core-image-minimal.log
  eglibc.qemux86-64.core-image-basic.log	
  eglibc.qemux86-64.core-image-minimal-dev.log
  eglibc.qemux86.core-image-basic.log
  eglibc.qemux86.core-image-minimal-dev.log
  eglibc.qemux86.core-image-x11.log

Of which this weird license issue is definitely the biggest problem and
the one we need to figure out. The download issue looks to be transient.

Cheers,

Richard








More information about the Openembedded-core mailing list