On Fri, Apr 20, 2018 at 11:16:59PM +0100, Russell King - ARM Linux wrote:
The above are slightly worrying - nothing should have affected those that last passed with v4.16-rc2-29-gf592819553ca. The only change is to change the sed expression to get the kernel BSS size to be a little more lenient with certain toolchains - but if it wasn't working previously the build should have failed.
Any chance of some investigation of this regression before I push the commits upstream to Linus?
These are all from one lab which looks like it was having a bad time internally and dropping connections in the middle of booting. It does look like the boards managed to boot OK while the infrastructure sorted itself out:
https://storage.kernelci.org/rmk/for-next/v4.16-rc2-29-g6dc8c9d147dd/arm/at9...