Bug #1988418 “Please merge initramfs-tools 0.142 from Debian uns...” : Bugs : initramfs-tools package : Ubuntu

Revision history for this message

Dave Jones (waveform) wrote on 2022-09-01:

#1

1-1988418.debdiff Edit (293.5 KiB, text/plain)

Attaching patch against Debian unstable. For ease of review, relevant commits
and tags have been pushed to the following repository:

https://code.launchpad.net/~waveform/ubuntu/+source/initramfs-tools/+git/initramfs-tools

Specifically:

* logical/0.140ubuntu17 represents our split-out delta on top of old/debian (0.140)
* logical/0.142ubuntu1 represents our rebased delta on top of new/debian (0.142)
* merge/0.142ubuntu1 adds changelog and maintainer changes, and one post-merge change on top of logical/0.142ubuntu1

Hence, the following command may produce output useful to the purposes of review:

git range-diff old/debian..logical/0.140ubuntu17 new/debian..logical/0.142ubuntu1

Dave Jones (waveform) on 2022-09-01

Changed in initramfs-tools (Ubuntu):
status:	New → Confirmed

Revision history for this message

Dave Jones (waveform) wrote on 2022-09-01:

#2

I would caution that there are some substantial changes in this merge. Specifically:

* upstream has now moved to zstd by default (as in Ubuntu) resulting in several bits of the delta changing

* upstream re-wrote a fair amount of logic around the "ip=" init parameter; this is an area we already had a substantial delta which also needed re-writing as a result

* upstream also added configuration for the compression level of the initramfs which required our delta to change

Please do not attempt to review the debdiff; at nearly 300KB it would be pointless. The git range-diff is far more informative, but may need a larger --creation-factor to make the most of it (100 is a reasonable choice).

Revision history for this message

Dave Jones (waveform) wrote on 2022-09-01:

#3

Test packages are building in ppa:waveform/initramfs-tools at:

https://launchpad.net/~waveform/+archive/ubuntu/initramfs-tools

Revision history for this message

Steve Langasek (vorlon) wrote on 2022-11-29:

#4

My standard way of reviewing a merge is to do the merge myself and see what comes up different. Here are my observations:

conf/initramfs.conf: the commented COMPRESSLEVEL setting should match the default, which for us is 1, not 3.

debian/control: upstream has Recommends: ${busybox:Recommends}, zstd, but you only have Recommends: zstd. Why?

debian/rules: you caught that there were build-time tests added by us that should be retained even though upstream dropped their override; good (I missed this).

hook-functions: upstream landed the change to exclude modems from the net usb drivers, but we had cdc-phonet.ko in our list but Debian does not. I do not know why Debian did not include it; I think we should include it in the merge and forward to Debian for consideration.

+ elif [ -n "${DEVICE}" ] || [ -n "${DEVICE6}" ]; then

FWIW I would write this as elif [ -n "$DEVICE$DEVICE6" ]; then

Otherwise, this looks good to me.

Revision history for this message

Dave Jones (waveform) wrote on 2022-11-29:

#5

Download full text (5.6 KiB)

> My standard way of reviewing a merge is to do the merge myself and
> see what comes up different. Here are my observations:
>
> conf/initramfs.conf: the commented COMPRESSLEVEL setting should
> match the default, which for us is 1, not 3.

Good point! Slightly weird that it's set to 3 given that upstream's
default is 9. Still, ours should indeed be consistent with our
defaults -- I'll amend it.

> debian/control: upstream has Recommends: ${busybox:Recommends},
> zstd, but you only have Recommends: zstd. Why?

This is a part where the range-diff comes in handy (sorry this is
probably going to wrap horribly):

1: 896377e ! 1: 1d9229b - Make busybox-initramfs a real runtime dependency, fixing kernel install failures with cryptsetup
    @@ debian/control: Description: generic modular initramfs generator (automation)
      Package: initramfs-tools-core
      Architecture: all
      Multi-Arch: foreign
    --Recommends: ${busybox:Recommends}, pigz
    +-Recommends: ${busybox:Recommends}, zstd
     -Depends: klibc-utils (>= 2.0.4-8~), cpio (>= 2.12), kmod, udev, coreutils (>= 8.24), logsave | e2fsprogs (<< 1.45.3-1~), ${misc:Depends}
    -+Recommends: pigz
    ++Recommends: zstd
     +Depends: ${busybox:Depends}, klibc-utils (>= 2.0.4-8~), cpio (>= 2.12), kmod, udev, coreutils (>= 8.24), logsave | e2fsprogs (<< 1.45.3-1~), ${misc:Depends}
      Suggests: bash-completion
      Breaks: initramfs-tools (<< 0.121~), ${busybox:Breaks}

This is showing that one part of our delta moves ${busybox:Recommends}
to ${busybox:Depends} ("Make busybox-initramfs a real runtime
dependency..."). That part of our delta hasn't changed, but the
context of it has. Re-ordering the lines a bit, the previous diff was:

    --Recommends: ${busybox:Recommends}, pigz
    -+Recommends: pigz
     -Depends: klibc-utils (>= 2.0.4-8~), cpio (>= 2.12), kmod, udev, coreutils (>= 8.24), logsave | e2fsprogs (<< 1.45.3-1~), ${misc:Depends}
     +Depends: ${busybox:Depends}, klibc-utils (>= 2.0.4-8~), cpio (>= 2.12), kmod, udev, coreutils (>= 8.24), logsave | e2fsprogs (<< 1.45.3-1~), ${misc:Depends}

But upstream changed pigz to zstd in Recommends so now the delta is:

    +-Recommends: ${busybox:Recommends}, zstd
    ++Recommends: zstd
     -Depends: klibc-utils (>= 2.0.4-8~), cpio (>= 2.12), kmod, udev, coreutils (>= 8.24), logsave | e2fsprogs (<< 1.45.3-1~), ${misc:Depends}
     +Depends: ${busybox:Depends}, klibc-utils (>= 2.0.4-8~), cpio (>= 2.12), kmod, udev, coreutils (>= 8.24), logsave | e2fsprogs (<< 1.45.3-1~), ${misc:Depends}

In other words, it's a context change but no real change in the delta.

> debian/rules: you caught that there were build-time tests added by
> us that should be retained even though upstream dropped their
> override; good (I missed this).
>
> hook-functions: upstream landed the change to exclude modems from
> the net usb drivers, but we had cdc-phonet.ko in our list but Debian
> does not. I do not know why Debian did not include it; I think we
> should include it in the merge and forward to Debian for
> consideration.

I think Debian does include it, just in a slightly different position?
At least, I can see cdc-phonet.ko added after "ca...

> My standard way of reviewing a merge is to do the merge myself and
> see what comes up different. Here are my observations:
>
> conf/initramfs.conf: the commented COMPRESSLEVEL setting should
> match the default, which for us is 1, not 3.

Good point! Slightly weird that it's set to 3 given that upstream's
default is 9. Still, ours should indeed be consistent with our
defaults -- I'll amend it.

> debian/control: upstream has Recommends: ${busybox:Recommends},
> zstd, but you only have Recommends: zstd. Why?

This is a part where the range-diff comes in handy (sorry this is
probably going to wrap horribly):

1:  896377e !  1:  1d9229b     - Make busybox-initramfs a real runtime dependency, fixing       kernel install failures with cryptsetup
    @@ debian/control: Description: generic modular initramfs generator (automation)
      Package: initramfs-tools-core
      Architecture: all
      Multi-Arch: foreign
    --Recommends: ${busybox:Recommends}, pigz
    +-Recommends: ${busybox:Recommends}, zstd
     -Depends: klibc-utils (>= 2.0.4-8~), cpio (>= 2.12), kmod, udev, coreutils (>= 8.24), logsave | e2fsprogs (<< 1.45.3-1~), ${misc:Depends}
    -+Recommends: pigz
    ++Recommends: zstd
     +Depends: ${busybox:Depends}, klibc-utils (>= 2.0.4-8~), cpio (>= 2.12), kmod, udev, coreutils (>= 8.24), logsave | e2fsprogs (<< 1.45.3-1~), ${misc:Depends}
      Suggests: bash-completion
      Breaks: initramfs-tools (<< 0.121~), ${busybox:Breaks}

This is showing that one part of our delta moves ${busybox:Recommends}
to ${busybox:Depends} ("Make busybox-initramfs a real runtime
dependency..."). That part of our delta hasn't changed, but the
context of it has. Re-ordering the lines a bit, the previous diff was:

--Recommends: ${busybox:Recommends}, pigz
    -+Recommends: pigz
     -Depends: klibc-utils (>= 2.0.4-8~), cpio (>= 2.12), kmod, udev, coreutils (>= 8.24), logsave | e2fsprogs (<< 1.45.3-1~), ${misc:Depends}
     +Depends: ${busybox:Depends}, klibc-utils (>= 2.0.4-8~), cpio (>= 2.12), kmod, udev, coreutils (>= 8.24), logsave | e2fsprogs (<< 1.45.3-1~), ${misc:Depends}

But upstream changed pigz to zstd in Recommends so now the delta is:

+-Recommends: ${busybox:Recommends}, zstd
    ++Recommends: zstd
     -Depends: klibc-utils (>= 2.0.4-8~), cpio (>= 2.12), kmod, udev, coreutils (>= 8.24), logsave | e2fsprogs (<< 1.45.3-1~), ${misc:Depends}
     +Depends: ${busybox:Depends}, klibc-utils (>= 2.0.4-8~), cpio (>= 2.12), kmod, udev, coreutils (>= 8.24), logsave | e2fsprogs (<< 1.45.3-1~), ${misc:Depends}

In other words, it's a context change but no real change in the delta.

> debian/rules: you caught that there were build-time tests added by
> us that should be retained even though upstream dropped their
> override; good (I missed this).
>
> hook-functions: upstream landed the change to exclude modems from
> the net usb drivers, but we had cdc-phonet.ko in our list but Debian
> does not. I do not know why Debian did not include it; I think we
> should include it in the merge and forward to Debian for
> consideration.

I think Debian does include it, just in a slightly different position?
At least, I can see cdc-phonet.ko added after "can" in the following
diff:

$ git diff old/ubuntu..merge/0.142ubuntu1 hook-functions
    ...
    - appletalk arcnet bonding can dummy.ko \
    - hamradio hippi ifb.ko irda macvlan.ko \
    - macvtap.ko pcmcia sb1000.ko team tokenring \
    - tun.ko veth.ko wan wimax wireless \
    - cdc_mbim.ko cdc-phonet.ko hso.ko huawei_cdc_ncm.ko ipheth.ko kalmia.ko \
    - lg-vl600.ko qmi_wwan.ko sierra_net.ko cdc_subset.ko gl620a.ko \
    - net1080.ko plusb.ko cx82310_eth.ko zaurus.ko \
    - xen-netback.ko
    + appletalk arcnet bonding can cdc-phonet.ko \
    + cdc_mbim.ko cdc_subset.ko cx82310_eth.ko \
    + dummy.ko gl620a.ko hamradio hippi hso.ko \
    + huawei_cdc_ncm.ko ifb.ko ipheth.ko irda \
    + kalmia.ko lg-vl600.ko macvlan.ko macvtap.ko \
    + net1080.ko pcmcia plusb.ko qmi_wwan.ko \
    + sb1000.ko sierra_net.ko team tokenring tun.ko \
    + veth.ko wan wimax wireless xen-netback.ko \
    + zaurus.ko
    ...

I was fairly sure I'd verified that all the same drivers were in that
section given it was one of the "merged upstream" bits in the
range-diff, but let's double-check:

>>> a = """- appletalk arcnet bonding can dummy.ko \
    ... - hamradio hippi ifb.ko irda macvlan.ko \
    ... - macvtap.ko pcmcia sb1000.ko team tokenring \
    ... - tun.ko veth.ko wan wimax wireless \
    ... - cdc_mbim.ko cdc-phonet.ko hso.ko huawei_cdc_ncm.ko ipheth.ko kalmia.ko \
    ... - lg-vl600.ko qmi_wwan.ko sierra_net.ko cdc_subset.ko gl620a.ko \
    ... - net1080.ko plusb.ko cx82310_eth.ko zaurus.ko \
    ... - xen-netback.ko
    ... """
    >>> b = """+ appletalk arcnet bonding can cdc-phonet.ko \
    ... + cdc_mbim.ko cdc_subset.ko cx82310_eth.ko \
    ... + dummy.ko gl620a.ko hamradio hippi hso.ko \
    ... + huawei_cdc_ncm.ko ifb.ko ipheth.ko irda \
    ... + kalmia.ko lg-vl600.ko macvlan.ko macvtap.ko \
    ... + net1080.ko pcmcia plusb.ko qmi_wwan.ko \
    ... + sb1000.ko sierra_net.ko team tokenring tun.ko \
    ... + veth.ko wan wimax wireless xen-netback.ko \
    ... + zaurus.ko
    ... """
    >>> a = {word for line in a.splitlines() for word in line.split() if word not in ('-', '+', '\\')}
    >>> b = {word for line in b.splitlines() for word in line.split() if word not in ('-', '+', '\\')}
    >>> a == b
    True

> + elif [ -n "${DEVICE}" ] || [ -n "${DEVICE6}" ]; then
>
> FWIW I would write this as elif [ -n "$DEVICE$DEVICE6" ]; then

Heh, the pedant in me balks slightly at that, but it is indeed more
efficient in shell and god knows initramfs-tools is slow enough as it
is! I'll amend it :)

Revision history for this message

Dave Jones (waveform) wrote on 2022-11-29:

#6

2-1988418.debdiff Edit (293.7 KiB, text/plain)

Attaching amended debdiff addressing review comments

Revision history for this message

Dave Jones (waveform) wrote on 2022-11-29:

#7

3-1988418.debdiff Edit (293.7 KiB, text/plain)

Doh, let's try that again with the right series this time...

Revision history for this message

Dave Jones (waveform) wrote on 2022-11-29:

#8

Test package building in ppa:waveform/initramfs-tools (https://launchpad.net/~waveform/+archive/ubuntu/initramfs-tools)

Revision history for this message

Steve Langasek (vorlon) wrote on 2022-12-08:

#9

> I think Debian does include it, just in a slightly different position?
> At least, I can see cdc-phonet.ko added after "can" in the following
> diff:

Ah, slightly different sorting rules and string-blindness.

We concur now wrt what should be uploaded. I'll look at sponsoring this.

Steve Langasek (vorlon) on 2022-12-08

Changed in initramfs-tools (Ubuntu):
status:	Confirmed → Fix Committed

Revision history for this message

Launchpad Janitor (janitor) wrote on 2022-12-10:

#10

Download full text (13.5 KiB)

This bug was fixed in the package initramfs-tools - 0.142ubuntu1

---------------
initramfs-tools (0.142ubuntu1) lunar; urgency=medium

  * Merge from Debian unstable (LP: #1988418). Remaining changes:
    - Make busybox-initramfs a real runtime dependency, fixing kernel install
      failures with cryptsetup
    - support mounting of loopback devices via loop*= parameters
    - Drop hooks/keymap, scripts/init-top/keymap, and associated code;
      console-setup takes care of this in Ubuntu.
    - Add a new 'fixrtc' script that tries to set the system clock forward
      based on the last mount time of the root disk; without this kludge,
      systems without a working RTC will end up in a perpetual reboot loop.
    - Trim leading whitespaces in dumpe2fs output so date command is happy in
      the fixrtc script
    - scripts/local-premount/fixrtc: ensure that the fixrtc script doesn't
      produce extraneous output from wait-for-root and date
    - Set hostname at boot, for the benefit of mdadm autoassembly.
    - Allow scripts and hooks to specify OPTION=VAR, and unless VAR is set to
      something other than "n", the script will not be included.
    - Restore the framebuffer hook and script, copying KMS and other
      framebuffer drivers into the initramfs, but make them optional; you need
      to set FRAMEBUFFER=y for these to be included.
    - Add hwaddr= alias for BOOTIF= for compatibility.
    - Automatically blacklist vga16fb when vga= or video= specified on kernel
      command-line.
    - add squashfs to list of 'most' modules
    - Add vfat and nls modules to the initramfs.
    - Add Hyper-V paravirtualised device drivers to the initramfs to allow
      booting of stock images in a Hyper-V guest.
    - ppc64el: enable PowerPC NX Crypto Coprocessor
    - Revised mountroot failure support. This is now deprecated and
      effectively removed. However, to prevent breakage in other packages the
      add_mountroot_fail_hook function is still present as an empty stub.
    - resume: only resume when the partition contains a resume image
    - Fix resuming a hibernate session from a swapfile
    - Add new initramfs-tools-bin package containing a binary that uses
      libudev to wait for udev to create the udev device, or wait for udev to
      finish processing if we catch it in the act, and returns the filesystem
      type as already probed by udev.
    - local_device_setup: use wait-for-root if it supports the device alias
      type
    - debian/control: update Vcs-* links for Ubuntu.
    - Add support for uppercase and lowercase uuids.
    - init: Don't interpret "rootdelay" as a static sleep.
    - src/wait-for-root.c, scripts/local: add explicit support for PARTUUID, a
      syntax we may prefer for images that might boot without an initramfs.
    - scripts/local: Let rootdelay determine the wait-for-root timeout.
    - Increase rootdelay to 180s on powerpc/ppc64/ppc64el
    - Support roottimeout= as the authoritative name for rootdelay= since the
      Ubuntu interpretation of rootdelay is *not* a delay and this causes
      inconsistent behavior depending on whether the kernel or an initramfs is
      interpreting it.
   ...

This bug was fixed in the package initramfs-tools - 0.142ubuntu1

---------------
initramfs-tools (0.142ubuntu1) lunar; urgency=medium

* Merge from Debian unstable (LP: #1988418). Remaining changes:
    - Make busybox-initramfs a real runtime dependency, fixing kernel install
      failures with cryptsetup
    - support mounting of loopback devices via loop*= parameters
    - Drop hooks/keymap, scripts/init-top/keymap, and associated code;
      console-setup takes care of this in Ubuntu.
    - Add a new 'fixrtc' script that tries to set the system clock forward
      based on the last mount time of the root disk; without this kludge,
      systems without a working RTC will end up in a perpetual reboot loop.
    - Trim leading whitespaces in dumpe2fs output so date command is happy in
      the fixrtc script
    - scripts/local-premount/fixrtc: ensure that the fixrtc script doesn't
      produce extraneous output from wait-for-root and date
    - Set hostname at boot, for the benefit of mdadm autoassembly.
    - Allow scripts and hooks to specify OPTION=VAR, and unless VAR is set to
      something other than "n", the script will not be included.
    - Restore the framebuffer hook and script, copying KMS and other
      framebuffer drivers into the initramfs, but make them optional; you need
      to set FRAMEBUFFER=y for these to be included.
    - Add hwaddr= alias for BOOTIF= for compatibility.
    - Automatically blacklist vga16fb when vga= or video= specified on kernel
      command-line.
    - add squashfs to list of 'most' modules
    - Add vfat and nls modules to the initramfs.
    - Add Hyper-V paravirtualised device drivers to the initramfs to allow
      booting of stock images in a Hyper-V guest.
    - ppc64el: enable PowerPC NX Crypto Coprocessor
    - Revised mountroot failure support.  This is now deprecated and
      effectively removed.  However, to prevent breakage in other packages the
      add_mountroot_fail_hook function is still present as an empty stub.
    - resume: only resume when the partition contains a resume image
    - Fix resuming a hibernate session from a swapfile
    - Add new initramfs-tools-bin package containing a binary that uses
      libudev to wait for udev to create the udev device, or wait for udev to
      finish processing if we catch it in the act, and returns the filesystem
      type as already probed by udev.
    - local_device_setup: use wait-for-root if it supports the device alias
      type
    - debian/control: update Vcs-* links for Ubuntu.
    - Add support for uppercase and lowercase uuids.
    - init: Don't interpret "rootdelay" as a static sleep.
    - src/wait-for-root.c, scripts/local: add explicit support for PARTUUID, a
      syntax we may prefer for images that might boot without an initramfs.
    - scripts/local: Let rootdelay determine the wait-for-root timeout.
    - Increase rootdelay to 180s on powerpc/ppc64/ppc64el
    - Support roottimeout= as the authoritative name for rootdelay= since the
      Ubuntu interpretation of rootdelay is *not* a delay and this causes
      inconsistent behavior depending on whether the kernel or an initramfs is
      interpreting it.
    - scripts/functions: by default, when setting up networking, use dhclient
      if we're asked to do DHCP explicitly, or if we otherwise are expected to
      do things automatically.
    - scripts/functions: make sure we can try to start all available and
      suitable interfaces if ip= isn't set when setting up the network, and
      exit as soon as we get an IP address.
    - split out IPv6 options in its own cmdline parameter: ip6= ; always use
      dhclient in this case if the value set is anything other than 'off' or
      'none'. Furthermore, parse anything other than 'on', 'dhcp' or 'any' as
      the name of an interface.
    - rework the stop conditions so that we properly handle the ROUNDTTT loop,
      timing out after a short period of time and trying again after a short
      sleep.
    - add a 'done' parameter for both ip= and ip6= so that we can properly
      exit the ROUNDTTT loop once we know that either there is no work to do,
      or that we've achieved what we wanted (that is, to bring up IPv4, IPv6,
      or possibly both).
    - Just let dhclient scan the interfaces if none is specified for IPv6.
      (previously unnoted)
    - Do not handle ip=rarp specially (ipconfig hasn't supported it for some
      time)
    - Call dhclient in simple cases for IPv4 (this makes some small observable
      changes -- for example ip=bootp will now make a DHCP request too -- but
      nothing that seems important).
    - Do not let dhclient processes hang around past the pivot and have them
      respect the shorter timeouts ipconfig was given.
    - Only source ipv4 config in configure_networking() if it exists.
    - hook-functions, mkinitramfs, scripts/functions: support usage of dns
      after configure_networking.
    - Run unittests during autopkgtests.
    - Add some autopkgtests for the network configuration processing.
    - scripts/functions: write netplan config files to /run/netplan for
      network devices configured with configure_networking.
    - Adjust testsuite for correct expected netplan output.
    - Add tests to parse netplan.yaml with netplan generate.
    - Skip setting mac-address, unless unsupported unpredictable netnames are
      used.
    - scripts/functions:_render_netplan(): Make configure_networking() more
      resilient when called by set -e scripts.
    - scripts/functions: include a new option to skip enslaved network
      devices.
    - Auto activate qeth devices on s390x hosts, when using ip= ip6= command
      lines.
    - Add support for creating VLANs via vlan=eth0.1:eth0 on kernel
      commandline, and as VLAN= conf.d snippet.
    - Emit empty dictionary in the ethernets section, for static vlan configs
      without any matches, as otherwise it is invalid config.
    - Skip attempting to chzdev enable VLAN device
    - scripts/functions: Fix the fix for bug #1880735, fixing setting up vlans
      from the kernel commandline.
    - Drop renderer from netplan yaml, such that netplan uses the default
      rendered for the target system, i.e. NetworkManager on the desktop.
    - debian/{control,rules}: Don't run shellcheck on i386 to avoid pulling
      ghc into our minimal supported set for the pruned architecture.
    - Work out the kernel modules required to support ZFS filesystems and add
      them as necessary.
    - Add modules for nvme path components on multipath nvme.
    - debian/initramfs-tools.postinst: remove orphaned old-dkms initrd files
      in /boot.
    - Ensure we have a /usr/sbin directory in initramfs-tools-core
    - Drop usage of klibc-utils' fstype in favor of always using udev blkid.
    - Use set -a/+a around loading conf snippets, to autoexport hook
      configuration variables. This thus allows packaged hooks to configure
      each other.
    - Fix PATH to include /usr/local.
    - Revert upstream commit 0012977; Ubuntu's resume script implementation
      diverges from Debian's and this special-case handling of UUID (which is
      used as the resume target in the default scenario) is required here.
    - Filter out looping or broken symlinks from the initramfs, since they
      cause cpio to explode
    - Always mount loop-mount filesystems read-write; ntfs-3g and vfat can't
      remount yet.
    - Provide a clearer error on mount failure of the Windows host filesystem
    - Fix copying libgcc_s when there is optimized pthreads installed.
    - scripts/local: Re-execute cryptroot local-block script.
    - d/tests: Add explicit call to partprobe on net test, specially in prep-
      image and run-image.
    - scripts/local-premount/resume: Use readlink -f to correctly handle non-
      symlink $resume
    - d/t/control:autopkgtests running qemu need root, add needs-root
      restriction
    - Cherry-pick merge request !26 to support xz compressed firmware.
      Uncompressed firmware is preffered, with a fallback to use compressed
      firmware files.
    - Improve shell scripts to conform to standards in shellcheck > 8.0
    - Increase image file to 2GB in autopkgtest
    - Force copy pthread and libgcc_s libraries via a stub binary. To avoid
      circular shared library dependencies between compiler and libc,
      libpthreads chooses to dlopen libgcc_s, instead of linking against it.
      However, we have no way to parse/know what it needs. And even when we
      hardcode to include libgcc_s1, we don't have a way to know which
      libgcc_s1 is needed on a given system and from which path to include it.
      But there is a hook-function to copy a binary and all of its shared
      library dependencies. Thus create a stub empty executable, with shared
      linking against pthreads and gcc_s1 and always include it in the initrd.
      This way at initrd creation time, ldd is used to correctly resolve these
      shared library dependencies and correctly copy them into the initrd.
      This removes hardcoding paths as to where these libraries must be copied
      from on the host.
    - Lower the compression levels for zstd and lz4 Following the discussion
      on the mailing list, we have reached a conclusion to lower the default
      compression levels:
      - For lz4, the compression level is lowered to 2 from 9
      - For zstd, the compression level is lowered to 1 from 9
    - Include nvdimm modules in block modules, they provide access to block
      pmem devices.
    - Include dax and nfit modules, also needed for pmem devices.
    - Add char/hw_random drivers
  * Removed patches obsoleted/merged by upstream:
    - Switch default initramfs compression to zstd, as it produces the lower
      overall boot time.
    - Don't recommend pigz, which is in universe.
    - Suppress shellcheck SC1010 errors, returning lots of false-positives on
      use of 'done' as a word.
    - hook-functions: when MODULES=list there will be no 'kernel' dir, so
      don't try to find anything there, as it will log an 'error' this is
      needed for amd64-ata-only autopkgtest to pass, or we would need to
      ignore the 'error' with allow-stderr
    - hooks/fsck: always include fsck binaries irrespective of PASSNO value,
      otherwise initrds fail to mount rootfs and have no ability to fix the
      rootfs.
    - Cherry-pick merge request !51 to decompress compressed kernel modules
      for boot speed performance.
    - For net boot include most of net/usb modules, except niche modems.
  * Reinstate build-time shellcheck (but leave shellcheck out of
    autopkgtests as per debian bug #992798)
  * Ensure commented COMPRESSLEVEL is consistent with the default

initramfs-tools (0.142) unstable; urgency=medium

[ Dan Streetman ]
  * [8c9840e] hooks/fsck: always include fsck binary (LP: #1917780)

[ Dimitri John Ledkov ]
  * [d8c5864] mkinitramfs: decompress compressed kernel modules

[ szubersk ]
  * [e447e91] resume: Don't resume from a swap device on a ZVOL

[ Johannes Schauer Marin Rodrigues ]
  * [834d01d] hook-functions: Add pwm-imx27, nwl-dsi, ti-sn65dsi86,
    imx-dcss, mxsfb, mux-mmio and imx8mq-interconnect for MNT Reform 2
    panel/backlight.

[ Rich Ercolani ]
  * [50d945a] Add support for specifying compression level

[ Ben Hutchings ]
  * [5bf479d] hook-functions: Remove support for ieee1394 and i2o
  * [9a4d579] hook-functions: auto_add_modules: Include UFS drivers from
    drivers/ufs/

[ Sven Joachim ]
  * [c064b9e] Copy modules.builtin.modinfo into initramfs (Closes: #1014319)

initramfs-tools (0.141) unstable; urgency=medium

[ Hideki Yamane ]
  * [742c8ee] Use zstd as default compression for initramfs (Closes: #976054)
  * [b6ccfb6] Recommends: s/pigz/zstd/
  * [1063c43] Add check whether zstd support kernel (>= 5.9) and
    CONFIG_RD_ZSTD=y or not
  * [71c29fd] As comparison, compression level 9 is better.
  * [9cee2f0] Consider SOURCE_DATE_EPOCH option for zstd

[ Alper Nebi Yasak ]
  * [db7aec4] hook-functions: Include io-domain in dw_mmc-rockchip hidden
    dependencies

[ Michael Biebl ]
  * [7ec5e26] Setup /dev/fd and /dev/std{in,out,err} after mounting devtmpfs
    (Closes: #981302)

[ Dan Streetman ]
  * [116b532] hook-functions: check for dir before running find on it
    (LP: #1927779)

[ Vagrant Cascadian ]
  * [a5d9b32] hook-functions: Add pwm-rockchip to "fb" modules for Pinebook PRO
    LCD display.
  * [7e15676] hook-functions: Add usb/typec/tcpm modules to support keyboard on
    Pinebook PRO.
  * [5f17c03] hook-functions: Add panel-edp to "fb" modules for Pinebook PRO
    LCD display.

[ Ben Hutchings ]
  * [26f2f00] Only run shellcheck in Salsa CI (Closes: #992798)
  * [bd1519b] Fix shellcheck warning "Avoid x-prefix in comparisons ..."
  * [4937118] hook-functions: Fix shellcheck warning "Expansions inside ${..}
    need to be ..."
  * [883f4ad] mkiniramfs: Log warning about missing compressor to stderr
  * [035190c] mkinitramfs: Generalise kernel compression support check
  * [58e05e4] d/salsa-ci.yml: Disable non-arch:all builds
  * [5679e26] debian/tests/control: Add zstd as dependency to avoid (fatal)
    warnings
  * [e856af4] hook-functions: Support network boot via USB Ethernet adapters
    (Closes: #980788)
  * [ec5c6c7] scripts/functions: Refactor device name logic in
    configure_networking
  * [80eddcf] scripts/functions: Make configure_networking wait for device
    specified any way (Closes: #911727)

[ Ben Love ]
  * [02dc65f] Fully-qualify modprobe calls to avoid busybox

[ Ariel D'Alessandro ]
  * [8a6f73c] hook-functions: Add nvmem-imx-ocotp driver module to network boot

-- Dave Jones <dave.jones@canonical.com>  Thu, 01 Sep 2022 13:43:43 +0100

Changed in initramfs-tools (Ubuntu):
status:	Fix Committed → Fix Released

Ubuntu
initramfs-tools package

Please merge initramfs-tools 0.142 from Debian unstable.

Bug Description

Other bug subscribers

Patches

Remote bug watches

Ubuntuinitramfs-tools package

Please merge initramfs-tools 0.142 from Debian unstable.

Bug Description

Other bug subscribers

Patches

Remote bug watches

Ubuntu
initramfs-tools package