[+ Guillaume Tucker ]
Quentin Schulz quentin.schulz@bootlin.com writes:
Hi Kevin,
On Thu, Mar 08, 2018 at 04:05:01PM -0800, Kevin Hilman wrote:
kernelci.org bot bot@kernelci.org writes:
stable-rc/linux-4.14.y boot: 148 boots: 4 failed, 133 passed with 7 offline, 3 untried/unknown, 1 conflict (v4.14.24-111-g66060ac1dfa0)
Full Boot Summary: https://kernelci.org/boot/all/job/stable-rc/branch/linux-4.14.y/kernel/v4.14... Full Build Summary: https://kernelci.org/build/stable-rc/branch/linux-4.14.y/kernel/v4.14.24-111...
Tree: stable-rc Branch: linux-4.14.y Git Describe: v4.14.24-111-g66060ac1dfa0 Git Commit: 66060ac1dfa02f02646a55f6ed888c0f2001623e Git URL: http://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git Tested: 76 unique boards, 23 SoC families, 17 builds out of 182
TL;DR; All is well.
Boot Regressions Detected:
arm:
davinci_all_defconfig: da850-evm: lab-baylibre-seattle: failing since 1 day (last pass: v4.14.4-76-gf91a57b206e0 - first fail: v4.14.24)Operator error in my lab.
multi_v7_defconfig: armada-375-db: lab-free-electrons: new failure (last pass: v4.14.24)We're having a hard time with the serial of the armada 375 db since the beginning. We're trying to find solutions to get a more reliable serial.
sun8i-h3-orangepi-pc: lab-free-electrons: failing since 1 day (last pass: v4.14.18-23-g8d861f5b27b0 - first fail: v4.14.24) sunxi_defconfig: sun8i-h3-orangepi-pc: lab-free-electrons: failing since 5 days (last pass: v4.14.18-23-g8d861f5b27b0 - first fail: v4.14.23-116-g0ae11358b0bd)These three are also not kernel issues. Seems the lab having problems connecting to the boards. Cc'd Free Electrons team for closer look.
It seems like the bootloader is corrupted somehow. I'll reflash it.
These issues are definitely not kernel ones as you rightfully stated. IIRC, LAVA in its latest version has updated the returned status of jobs and shouldn't return the same error if it fails before actually starting the kernel. For example, I have a "Job error: bootloader-interrupt timed out after 294 seconds" for the OrangePi.
While we really appreciate you giving us feedback on these issues, I think they shouldn't be mentioned in your overall boot summary (or at least in a different section than kernel errors).
Agreed. This is work in progress.
Or maybe you're waiting on all labs to migrate to v2018.02+ so that all have this enhanced job status report before actually using it?
I added Guillaume for clarifcation, but I thought once the lab is upgraded, it should report these issues as OFFLINE instead of FAIL.
It might be that the kernelCI backend changes haven't been merged into production yet. Hopepfully Guil
Kevin