[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#991403: mtools: mcopy fails on arm*, breaks d-i builds

Package: mtools
Version: 4.0.33-1
Severity: serious
Tags: d-i
Justification: breaks d-i
X-Debbugs-Cc: debian-boot@lists.debian.org


The latest upload of mtools to unstable breaks d-i builds on arm* (at
least arm64 and armhf), hindering our ability to publish newer releases
of the installer for bullseye:

Example on arm64:

    efi-image -o ./tmp/cdrom_grub/grub_efi -g arm64-efi \
    	-e aa64 -n debian-installer/arm64 \
    	-s y -d ./tmp/cdrom_grub/lib
    util/efi-image: Using pre-signed grub-efi and shim binaries for aa64
    util/efi-image: Running first pass of mkfs.msdos with size 25426 blocks
    mkfs.fat 4.2 (2021-01-31)
    ./tmp/cdrom_grub/grub_efi/efi.img has 4 heads and 32 sectors per track,
    hidden sectors 0x0000;
    logical sector size is 512,
    using 0xf8 media descriptor, with 50848 sectors;
    drive number 0x80;
    filesystem has 2 16-bit FATs and 4 sectors per cluster.
    FAT size is 52 sectors, and provides 12677 clusters.
    There are 4 reserved sectors.
    Root directory contains 512 slots and uses 32 sectors.
    Volume ID is deb00001, no volume label.
    mcopy: No directory slots
    mcopy: No directory slots
    mcopy: No directory slots
    mcopy: No directory slots
    mcopy: No directory slots
    [ repeats a lot ]

Build failures with daily builds:

This can be reproduced with the debian-installer source package in
unstable, after a single change in build/config/common to catch up
with recent ABI change:

    -LINUX_KERNEL_ABI ?= 5.10.0-7
    +LINUX_KERNEL_ABI ?= 5.10.0-8

This is clearly a regression from the previous version (4.0.32-1),
which seemed plausible given the timing of daily build failure
regression (between Jul 21 and Jul 22); but that's also been confirmed
by reproducing the issue in an arm64 sid schroot on amdahl, then by
reverting to the previous version, unpacked in ~/mtools, and PATH
overridden with ~/mtools:$PATH → the issue goes away.

On a side note, I must say I'm quite confused seeing all those new
upstream releases going to unstable instead of experimental during a

Cyril Brulebois (kibi@debian.org)            <https://debamax.com/>
D-I release manager -- Release team member -- Freelance Consultant

Reply to: