On Wed, Jun 21, 2017 at 12:03:01PM +0000, Linaro QA wrote:
kernel-describe: 8853995
This looks like it's taken a step backwards?
- test_execve -- failing since build cd171eff5254d3c500f6d3f70b5d76a55667d507, from May 26, 2017, 2:16 p.m.
What are these build identifiers that are being used? If they're kernel commits it'd be good to say "commit" (and ideally with the shorter commit ID as has been repeatedly mentioned) so it's clear. If they are some kind of build identifer then we probably need a direct link to the build so people can decode more easily.
Also these need wrapping within 80 columns, I think just a simple word wrap would do it for this format.
On 21 June 2017 at 15:42, Mark Brown broonie@kernel.org wrote:
On Wed, Jun 21, 2017 at 12:03:01PM +0000, Linaro QA wrote:
kernel-describe: 8853995
This looks like it's taken a step backwards?
I'll check but I think this is how it works on this tree.
- test_execve -- failing since build cd171eff5254d3c500f6d3f70b5d76a55667d507, from May 26, 2017, 2:16 p.m.
What are these build identifiers that are being used? If they're kernel commits it'd be good to say "commit" (and ideally with the shorter commit ID as has been repeatedly mentioned) so it's clear. If they are some kind of build identifer then we probably need a direct link to the build so people can decode more easily.
Also these need wrapping within 80 columns, I think just a simple word wrap would do it for this format.
Arnd mentioned this as well in the other thread. The idea for a fix is to group the failures by 'failed since'. This would make the lines shorter. The 'version' is currently set to full git sha1. As we agreed in the other thread this will change to git describe (or 12 char commit id if there is no git describe).
milosz
On Wed, Jun 21, 2017 at 09:48:41PM +0100, Milosz Wasilewski wrote:
On 21 June 2017 at 15:42, Mark Brown broonie@kernel.org wrote:
On Wed, Jun 21, 2017 at 12:03:01PM +0000, Linaro QA wrote:
kernel-describe: 8853995
This looks like it's taken a step backwards?
I'll check but I think this is how it works on this tree.
Really, no. *All* kernels should end up with something based off a tag.
Arnd mentioned this as well in the other thread. The idea for a fix is to group the failures by 'failed since'. This would make the lines shorter. The 'version' is currently set to full git sha1. As we agreed in the other thread this will change to git describe (or 12 char commit id if there is no git describe).
If you end up with a kernel without a git describe available something has gone wrong with what you're doing.
On 23 June 2017 at 12:22, Mark Brown broonie@kernel.org wrote:
On Wed, Jun 21, 2017 at 09:48:41PM +0100, Milosz Wasilewski wrote:
On 21 June 2017 at 15:42, Mark Brown broonie@kernel.org wrote:
On Wed, Jun 21, 2017 at 12:03:01PM +0000, Linaro QA wrote:
kernel-describe: 8853995
This looks like it's taken a step backwards?
I'll check but I think this is how it works on this tree.
Really, no. *All* kernels should end up with something based off a tag.
milosz@milosz-tp:~/linaro/oe/hikey-linaro$ git remote -v origin https://android.googlesource.com/kernel/hikey-linaro (fetch) origin https://android.googlesource.com/kernel/hikey-linaro (push) milosz@milosz-tp:~/linaro/oe/hikey-linaro$ git branch * android-hikey-linaro-4.4 android-hikey-linaro-4.9 master milosz@milosz-tp:~/linaro/oe/hikey-linaro$ git describe --always 6b116e0f
and 4.9
milosz@milosz-tp:~/linaro/oe/hikey-linaro$ git branch android-hikey-linaro-4.4 * android-hikey-linaro-4.9 master milosz@milosz-tp:~/linaro/oe/hikey-linaro$ git describe --always db724a5
Anything I'm doing wrong?
Arnd mentioned this as well in the other thread. The idea for a fix is to group the failures by 'failed since'. This would make the lines shorter. The 'version' is currently set to full git sha1. As we agreed in the other thread this will change to git describe (or 12 char commit id if there is no git describe).
If you end up with a kernel without a git describe available something has gone wrong with what you're doing.
I've no idea what is wrong. This is what I'm getting after cloning the repository and checking out the branch.
milosz
On Mon, Jul 03, 2017 at 09:08:08AM +0100, Milosz Wasilewski wrote:
On 23 June 2017 at 12:22, Mark Brown broonie@kernel.org wrote:
On Wed, Jun 21, 2017 at 09:48:41PM +0100, Milosz Wasilewski wrote:
On 21 June 2017 at 15:42, Mark Brown broonie@kernel.org wrote:
On Wed, Jun 21, 2017 at 12:03:01PM +0000, Linaro QA wrote:
kernel-describe: 8853995
This looks like it's taken a step backwards?
I'll check but I think this is how it works on this tree.
Really, no. *All* kernels should end up with something based off a tag.
milosz@milosz-tp:~/linaro/oe/hikey-linaro$ git remote -v origin https://android.googlesource.com/kernel/hikey-linaro (fetch) origin https://android.googlesource.com/kernel/hikey-linaro (push) milosz@milosz-tp:~/linaro/oe/hikey-linaro$ git branch
- android-hikey-linaro-4.4 android-hikey-linaro-4.9 master
milosz@milosz-tp:~/linaro/oe/hikey-linaro$ git describe --always 6b116e0f
and 4.9
milosz@milosz-tp:~/linaro/oe/hikey-linaro$ git branch android-hikey-linaro-4.4
- android-hikey-linaro-4.9 master
milosz@milosz-tp:~/linaro/oe/hikey-linaro$ git describe --always db724a5
Anything I'm doing wrong?
Do you not have the upstream branch that contains the git tags for the releases? You will need that.
Also, you might need to use 'git describe --contains', that's what I always use otherwise the output can be "odd" if you don't know exactly what you are looking at...
thanks,
greg k-h
On Mon, Jul 03, 2017 at 10:17:32AM +0200, Greg KH wrote:
On Mon, Jul 03, 2017 at 09:08:08AM +0100, Milosz Wasilewski wrote:
On 23 June 2017 at 12:22, Mark Brown broonie@kernel.org wrote:
On Wed, Jun 21, 2017 at 09:48:41PM +0100, Milosz Wasilewski wrote:
On 21 June 2017 at 15:42, Mark Brown broonie@kernel.org wrote:
On Wed, Jun 21, 2017 at 12:03:01PM +0000, Linaro QA wrote:
kernel-describe: 8853995
This looks like it's taken a step backwards?
I'll check but I think this is how it works on this tree.
Really, no. *All* kernels should end up with something based off a tag.
milosz@milosz-tp:~/linaro/oe/hikey-linaro$ git remote -v origin https://android.googlesource.com/kernel/hikey-linaro (fetch) origin https://android.googlesource.com/kernel/hikey-linaro (push) milosz@milosz-tp:~/linaro/oe/hikey-linaro$ git branch
- android-hikey-linaro-4.4 android-hikey-linaro-4.9 master
milosz@milosz-tp:~/linaro/oe/hikey-linaro$ git describe --always 6b116e0f
and 4.9
milosz@milosz-tp:~/linaro/oe/hikey-linaro$ git branch android-hikey-linaro-4.4
- android-hikey-linaro-4.9 master
milosz@milosz-tp:~/linaro/oe/hikey-linaro$ git describe --always db724a5
Anything I'm doing wrong?
Do you not have the upstream branch that contains the git tags for the releases? You will need that.
And to be clear, the android branches do not contain the upstream stable tags for various odd build-system reasons (i.e. it wrecks havoc with them, and the last time I tried to fix it, it took me a few days, and it still wasn't working properly...)
thanks,
greg k-h
On 3 July 2017 at 09:19, Greg KH gregkh@google.com wrote:
On Mon, Jul 03, 2017 at 10:17:32AM +0200, Greg KH wrote:
On Mon, Jul 03, 2017 at 09:08:08AM +0100, Milosz Wasilewski wrote:
On 23 June 2017 at 12:22, Mark Brown broonie@kernel.org wrote:
On Wed, Jun 21, 2017 at 09:48:41PM +0100, Milosz Wasilewski wrote:
On 21 June 2017 at 15:42, Mark Brown broonie@kernel.org wrote:
On Wed, Jun 21, 2017 at 12:03:01PM +0000, Linaro QA wrote:
> kernel-describe: 8853995
This looks like it's taken a step backwards?
I'll check but I think this is how it works on this tree.
Really, no. *All* kernels should end up with something based off a tag.
milosz@milosz-tp:~/linaro/oe/hikey-linaro$ git remote -v origin https://android.googlesource.com/kernel/hikey-linaro (fetch) origin https://android.googlesource.com/kernel/hikey-linaro (push) milosz@milosz-tp:~/linaro/oe/hikey-linaro$ git branch
- android-hikey-linaro-4.4 android-hikey-linaro-4.9 master
milosz@milosz-tp:~/linaro/oe/hikey-linaro$ git describe --always 6b116e0f
and 4.9
milosz@milosz-tp:~/linaro/oe/hikey-linaro$ git branch android-hikey-linaro-4.4
- android-hikey-linaro-4.9 master
milosz@milosz-tp:~/linaro/oe/hikey-linaro$ git describe --always db724a5
Anything I'm doing wrong?
Do you not have the upstream branch that contains the git tags for the releases? You will need that.
I recently moved the KernelCI jenkins scripts into git, so here's how our build trigger makes sure to fetch Linus tree so we have the tags to create a sensible git describe. https://github.com/kernelci/kernelci-build/blob/master/jenkins/kernel-trigge...
And to be clear, the android branches do not contain the upstream stable tags for various odd build-system reasons (i.e. it wrecks havoc with them, and the last time I tried to fix it, it took me a few days, and it still wasn't working properly...)
thanks,
greg k-h _______________________________________________ Kernel-build-reports mailing list Kernel-build-reports@lists.linaro.org https://lists.linaro.org/mailman/listinfo/kernel-build-reports
On Mon, Jul 03, 2017 at 10:17:32AM +0200, Greg KH wrote:
On Mon, Jul 03, 2017 at 09:08:08AM +0100, Milosz Wasilewski wrote:
Anything I'm doing wrong?
Do you not have the upstream branch that contains the git tags for the releases? You will need that.
Right - the simplest thing is always having Linus' tree as a remote like I suggested before.
On 5 July 2017 at 11:18, Mark Brown broonie@kernel.org wrote:
On Mon, Jul 03, 2017 at 10:17:32AM +0200, Greg KH wrote:
On Mon, Jul 03, 2017 at 09:08:08AM +0100, Milosz Wasilewski wrote:
Anything I'm doing wrong?
Do you not have the upstream branch that contains the git tags for the releases? You will need that.
Right - the simplest thing is always having Linus' tree as a remote like I suggested before.
Fathi made this change couple of days ago. Next builds should have 'proper' git describe. https://git.linaro.org/ci/job/configs.git/commit/?id=2d42fef6599941c332fdd6b...
milosz
kernel-build-reports@lists.linaro.org