Control: forcmerge 588675 -1 On Fri, 2016-06-10 at 18:16 -0700, Elliott Mitchell wrote: > Package: src:linux > Version: 3.2.63-2 > Control: found -1 linux/2.6.18 > Control: found -1 linux/3.16.7-ckt25-2 > Control: found -1 linux/3.16.7-ckt20-1+deb8u3 > Control: found -1 linux/3.2.78-1 > Control: found -1 linux/3.16.7-ckt11-1+deb8u6~bpo70+1 > Control: found -1 linux/2.6.32 > > The kernel build scripts are confused by what the SCSI subsystem produces > in /proc/mounts: This is not controlled by the SCSI subsystem. Unlike you, I've read the relevant source code, and I'm the upstream maintainer of initramfs-tools. I've seen problems like this before. When I ask you questions I am not just speculating. > $ awk '$2 == "/" && $1 != "rootfs"' < /proc/mounts > /dev/root / ext3 ro > $ > > A kernel build on such a system will panic on boot unless the root > filesystem is explicitly passed to the kernel by the bootloader. While > in common configurations bootloaders generally default to telling the > kernel what device it should use as root, that has not been documented to > be required. It seems pretty obvious to me that you have to specify the root device somehow. > Since Ben Hutchings thinks #588675 is too insignificant to ever be worthy > of a single line of code to fix, this bug now needs to be fixed (along > with many other utilities that are broken by #588675). I make no judgement about the significance of that bug. But if you refuse to answer a maintainer's reasonable questions about a report, there is no way to progress and it should be closed. Ben. -- Ben Hutchings The program is absolutely right; therefore, the computer must be wrong.
Attachment:
signature.asc
Description: This is a digitally signed message part