lkft/linaro-hikey-stable-rc-4.4-oe: no regressions found in build 4.4.86-hikey-18-gfa7c6cfbb88f
Summary ------------------------------------------------------------------------
kernel-repo: https://git.linaro.org/lkft/arm64-stable-rc.git kernel-branch: 4.4.87-rc1-hikey-20170907 kernel-commit: fa7c6cfbb88f3c4f333dbc9f2c63baf476506b63 kernel-describe: 4.4.86-hikey-18-gfa7c6cfbb88f Test details: https://qa-reports.linaro.org/lkft/linaro-hikey-stable-rc-4.4-oe/build/4.4.8...
No regressions (compared to build 683a6addcc5e)
Documentation - https://collaborate.linaro.org/display/LKFT/Email+Reports
On Fri, Sep 08, 2017 at 08:18:49AM +0000, Linaro QA wrote:
Summary
kernel-repo: https://git.linaro.org/lkft/arm64-stable-rc.git kernel-branch: 4.4.87-rc1-hikey-20170907 kernel-commit: fa7c6cfbb88f3c4f333dbc9f2c63baf476506b63 kernel-describe: 4.4.86-hikey-18-gfa7c6cfbb88f
Can someone explain the difference between "kernel-branch" and "kernel-describe" to me and why they would show two different kernel version numbers?
Why do you have a branch per "rc" anyway? That's going to be crazy...
thanks,
greg k-h
On 8 September 2017 at 12:41, Greg KH gregkh@google.com wrote:
On Fri, Sep 08, 2017 at 08:18:49AM +0000, Linaro QA wrote:
Summary
kernel-repo: https://git.linaro.org/lkft/arm64-stable-rc.git kernel-branch: 4.4.87-rc1-hikey-20170907 kernel-commit: fa7c6cfbb88f3c4f333dbc9f2c63baf476506b63 kernel-describe: 4.4.86-hikey-18-gfa7c6cfbb88f
Can someone explain the difference between "kernel-branch" and "kernel-describe" to me and why they would show two different kernel version numbers?
kernel-branch is supposed to be the kernel branch used. kernel-describe is the output of git describe.
Why do you have a branch per "rc" anyway? That's going to be crazy...
Sumit's tree has a single branch 4.4.y-rc-hikey, which isn't. The kernel-branch value in this case is the tag that he used.
See https://git.linaro.org/lkft/arm64-stable-rc.git
As we use the same template for all the kernels, we end with this bogus value. For this particular tree, we can either remove the value as a whole or set an empty value instead of fallback to the tag name.
thanks,
greg k-h _______________________________________________ Lts-dev mailing list Lts-dev@lists.linaro.org https://lists.linaro.org/mailman/listinfo/lts-dev
On Fri, Sep 08, 2017 at 02:23:50PM +0300, Fathi Boudra wrote:
On 8 September 2017 at 12:41, Greg KH gregkh@google.com wrote:
On Fri, Sep 08, 2017 at 08:18:49AM +0000, Linaro QA wrote:
Summary
kernel-repo: https://git.linaro.org/lkft/arm64-stable-rc.git kernel-branch: 4.4.87-rc1-hikey-20170907 kernel-commit: fa7c6cfbb88f3c4f333dbc9f2c63baf476506b63 kernel-describe: 4.4.86-hikey-18-gfa7c6cfbb88f
Can someone explain the difference between "kernel-branch" and "kernel-describe" to me and why they would show two different kernel version numbers?
kernel-branch is supposed to be the kernel branch used.
But is it?
kernel-describe is the output of git describe.
"git describe --contains"? or just "git describe"?
Why do you have a branch per "rc" anyway? That's going to be crazy...
Sumit's tree has a single branch 4.4.y-rc-hikey, which isn't.
???
The kernel-branch value in this case is the tag that he used.
Ok, so it's wrong?
totally confused...
See https://git.linaro.org/lkft/arm64-stable-rc.git
As we use the same template for all the kernels, we end with this bogus value.
How is it "bogus"? If it is, why even show it?
For this particular tree, we can either remove the value as a whole or set an empty value instead of fallback to the tag name.
still confused as to what is going on...
greg k-h
Hi Greg,
On 8 September 2017 at 18:46, Greg KH gregkh@google.com wrote:
On Fri, Sep 08, 2017 at 02:23:50PM +0300, Fathi Boudra wrote:
On 8 September 2017 at 12:41, Greg KH gregkh@google.com wrote:
On Fri, Sep 08, 2017 at 08:18:49AM +0000, Linaro QA wrote:
Summary
kernel-repo: https://git.linaro.org/lkft/arm64-stable-rc.git kernel-branch: 4.4.87-rc1-hikey-20170907 kernel-commit: fa7c6cfbb88f3c4f333dbc9f2c63baf476506b63 kernel-describe: 4.4.86-hikey-18-gfa7c6cfbb88f
Can someone explain the difference between "kernel-branch" and "kernel-describe" to me and why they would show two different kernel version numbers?
kernel-branch is supposed to be the kernel branch used.
But is it?
kernel-describe is the output of git describe.
"git describe --contains"? or just "git describe"?
Why do you have a branch per "rc" anyway? That's going to be crazy...
Sumit's tree has a single branch 4.4.y-rc-hikey, which isn't.
???
I too didn't understand Fathi's comment here, but a little history for my tree might help, so here it is.
When I created that tree to track stable-rcs, I had originally thought of following your pattern with the stable-rc branches: to have one rebaseable branch per kernel version.
Then I realised that we need this tree only for 4.4-rc for hikey (4.9 onwards, hikey boots off mainline) - so given that stable-rcs have a limited life span, I thought just pushing -rc tags was enough, without requiring update of the branch, since extended testing of -rcs doesn't yield much value: -rc testing is useful only for reporting immediate regressions, if any.
Hope this makes some sense?
The kernel-branch value in this case is the tag that he used.
Ok, so it's wrong?
totally confused...
So, I guess the reporting tool uses 'the branch, or the tag' used to build. In this case, it is a tag.
See https://git.linaro.org/lkft/arm64-stable-rc.git
As we use the same template for all the kernels, we end with this bogus value.
How is it "bogus"? If it is, why even show it?
For this particular tree, we can either remove the value as a whole or set an empty value instead of fallback to the tag name.
still confused as to what is going on...
I think a simple solution will be to change the heading text as 'kernel-branch / tag' to make it less confusing - would you agree?
greg k-h _______________________________________________ Lts-dev mailing list Lts-dev@lists.linaro.org https://lists.linaro.org/mailman/listinfo/lts-dev
Best, Sumit.
On 8 September 2017 at 19:24, Sumit Semwal sumit.semwal@linaro.org wrote:
Hi Greg,
On 8 September 2017 at 18:46, Greg KH gregkh@google.com wrote:
On Fri, Sep 08, 2017 at 02:23:50PM +0300, Fathi Boudra wrote:
On 8 September 2017 at 12:41, Greg KH gregkh@google.com wrote:
On Fri, Sep 08, 2017 at 08:18:49AM +0000, Linaro QA wrote:
Summary
kernel-repo: https://git.linaro.org/lkft/arm64-stable-rc.git kernel-branch: 4.4.87-rc1-hikey-20170907 kernel-commit: fa7c6cfbb88f3c4f333dbc9f2c63baf476506b63 kernel-describe: 4.4.86-hikey-18-gfa7c6cfbb88f
Can someone explain the difference between "kernel-branch" and "kernel-describe" to me and why they would show two different kernel version numbers?
kernel-branch is supposed to be the kernel branch used.
But is it?
kernel-describe is the output of git describe.
"git describe --contains"? or just "git describe"?
Why do you have a branch per "rc" anyway? That's going to be crazy...
Sumit's tree has a single branch 4.4.y-rc-hikey, which isn't.
???
I too didn't understand Fathi's comment here, but a little history for my tree might help, so here it is.
I meant which isn't *used*.
When I created that tree to track stable-rcs, I had originally thought of following your pattern with the stable-rc branches: to have one rebaseable branch per kernel version.
Then I realised that we need this tree only for 4.4-rc for hikey (4.9 onwards, hikey boots off mainline) - so given that stable-rcs have a limited life span, I thought just pushing -rc tags was enough, without requiring update of the branch, since extended testing of -rcs doesn't yield much value: -rc testing is useful only for reporting immediate regressions, if any.
Hope this makes some sense?
The kernel-branch value in this case is the tag that he used.
Ok, so it's wrong?
yes
totally confused...
So, I guess the reporting tool uses 'the branch, or the tag' used to build. In this case, it is a tag.
See https://git.linaro.org/lkft/arm64-stable-rc.git
As we use the same template for all the kernels, we end with this bogus value.
How is it "bogus"? If it is, why even show it?
It's bogus in the sense it's confusing. You would expected a branch name but it isn't for this particular kernel tree. It's the fallback, and the fallback is the tag name.
For this particular tree, we can either remove the value as a whole or set an empty value instead of fallback to the tag name.
still confused as to what is going on...
I think a simple solution will be to change the heading text as 'kernel-branch / tag' to make it less confusing - would you agree?
or remove it to avoid confusion. The point being that you won't know if it's actually the branch name or the fallback tag name.
greg k-h _______________________________________________ Lts-dev mailing list Lts-dev@lists.linaro.org https://lists.linaro.org/mailman/listinfo/lts-dev
Best, Sumit.
On 8 September 2017 at 14:16, Greg KH gregkh@google.com wrote:
On Fri, Sep 08, 2017 at 02:23:50PM +0300, Fathi Boudra wrote:
On 8 September 2017 at 12:41, Greg KH gregkh@google.com wrote:
On Fri, Sep 08, 2017 at 08:18:49AM +0000, Linaro QA wrote:
Summary
kernel-repo: https://git.linaro.org/lkft/arm64-stable-rc.git kernel-branch: 4.4.87-rc1-hikey-20170907 kernel-commit: fa7c6cfbb88f3c4f333dbc9f2c63baf476506b63 kernel-describe: 4.4.86-hikey-18-gfa7c6cfbb88f
Can someone explain the difference between "kernel-branch" and "kernel-describe" to me and why they would show two different kernel version numbers?
kernel-branch is supposed to be the kernel branch used.
But is it?
kernel-describe is the output of git describe.
"git describe --contains"? or just "git describe"?
git describe --always
Why do you have a branch per "rc" anyway? That's going to be crazy...
Sumit's tree has a single branch 4.4.y-rc-hikey, which isn't.
???
The kernel-branch value in this case is the tag that he used.
Ok, so it's wrong?
totally confused...
I changed the email template for Sumit's trees. It will now say 'kernel-tag'. For all the others it's still 'kernel-branch'. I hope this takes the confusion away.
milosz
-
Fathi Boudra -
Greg KH -
Linaro QA -
Milosz Wasilewski -
Sumit Semwal