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

Bug#1029804: marked as done (hw-detect: decide whether to automatically install amd64-microcode/intel-microcode)

Your message dated Wed, 22 Mar 2023 19:49:21 +0000
with message-id <E1pf4SX-005HMq-L3@fasolo.debian.org>
and subject line Bug#1029804: fixed in hw-detect 1.155
has caused the Debian Bug report #1029804,
regarding hw-detect: decide whether to automatically install amd64-microcode/intel-microcode
to be marked as done.

This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.

(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact owner@bugs.debian.org
immediately.)


-- 
1029804: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1029804
Debian Bug Tracking System
Contact owner@bugs.debian.org with problems
--- Begin Message --- 
Package: hw-detect
Severity: important
X-Debbugs-Cc: amd64-microcode@packages.debian.org, intel-microcode@packages.debian.org

Hi,

Both of those *-microcode packages are featured in my initial list of
firmware packages that I found interesting to move from non-free to
non-free-firmware:
  https://lists.debian.org/debian-boot/2023/01/msg00150.html

Henrique (in X-D-Cc) is fine with the proposed move, even if that hasn't
happened yet (and I won't consider this a blocker for the upcoming D-I
release):
  https://salsa.debian.org/hmh/amd64-microcode/-/merge_requests/3
  https://salsa.debian.org/hmh/intel-microcode/-/merge_requests/2

That being said, contrary to other firmware packages:
 - we are not going to see log lines in dmesg to tell us that firmware
   loading failed, so we won't try and find firmware packages shipping
   those missing files (via check-missing-firmware);
 - they don't ship appstream metadata which would give us hints about
   hardware that could benefit from firmware packages, even without
   log lines in dmesg (via 50install-firmware, post-base-installer).

We could probably implement specific detection in hw-detect to install
those packages anyway (alternatively, maybe discover but nothing's been
happening there for a while anyway). Should we?

Various topics if we wanted to implement that:
 - Those are not needed during the installation, so installing them
   as soon as possible is not required, contrary to regular firmware
   packages; queueing their installation would be sufficient, along
   with enabling non-free-firmware (apt-setup/non-free-firmware).
 - Those have dependencies, so dpkg -i wouldn't fly and something like
   `in-target apt-get` should be used instead):
     https://salsa.debian.org/installer-team/hw-detect/-/commit/c876e43039
 - intel-microcode requires iucode-tool, currently in contrib, but
   Henrique is fine with moving it to main:
     https://salsa.debian.org/hmh/intel-microcode/-/merge_requests/2
 - What would be the best way to perform detection?

That, along with the move of those 3 packages between sections, is not
something I consider a blocker for the upcoming D-I release, or even for
the first Bookworm release. Just something that seems nice to have, if
we want to.


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

--- End Message ---
--- Begin Message --- 
Source: hw-detect
Source-Version: 1.155
Done: Cyril Brulebois <kibi@debian.org>

We believe that the bug you reported is fixed in the latest version of
hw-detect, which is due to be installed in the Debian FTP archive.

A summary of the changes between this version and the previous one is
attached.

Thank you for reporting the bug, which will now be closed.  If you
have further comments please address them to 1029804@bugs.debian.org,
and the maintainer will reopen the bug report if appropriate.

Debian distribution maintenance software
pp.
Cyril Brulebois <kibi@debian.org> (supplier of updated hw-detect package)

(This message was generated automatically at their request; if you
believe that there is a problem with it please contact the archive
administrators by mailing ftpmaster@ftp-master.debian.org)


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA512

Format: 1.8
Date: Wed, 22 Mar 2023 20:31:11 +0100
Source: hw-detect
Architecture: source
Version: 1.155
Distribution: unstable
Urgency: medium
Maintainer: Debian Install System Team <debian-boot@lists.debian.org>
Changed-By: Cyril Brulebois <kibi@debian.org>
Closes: 1029804 1029849 1031631 1033035
Changes:
 hw-detect (1.155) unstable; urgency=medium
 .
   [ Cyril Brulebois ]
   * Deduplicate the list of requested firmware files, not just the list of
     requesting modules (Closes: #1031631).
   * Implement microcode support when /proc/cpuinfo contains a vendor_id
     field, with one of the following values (Closes: #1029804):
      - Install amd64-microcode on AuthenticAMD.
      - Install intel-microcode on GenuineIntel.
      - Enable non-free-firmware accordingly.
      - Perform installation via finish-install, making sure apt-setup has
        been configured, and using apt-install for dependency resolution.
   * Optimize firmware package installation: process dpkg triggers once,
     after all packages have been installed (i.e. install-firmware only
     triggers a single update-initramfs call).
   * Fix condition around moutmedia calls (See: #1032377).
   * Fix files removal for non-accepted firmware packages (See: #1032377).
   * Add a special case for the mhi module: when the module requesting
     firmware files is “mhi”, use the modules listed as holders (e.g.
     ath11k_pci and qrtr_mhi). This is less precise than the usb special
     case, since /sys/bus/mhi/devices/<device> gives no hints as to which
     network module would be involved (See: #1032140). Thanks to Nicolas
     Dandrimont and Benoît Chauvet for the tests.
   * Adjust dmesg timestamp management: only update the timestamp file if
     there are new lines.
   * Fix package name extraction when removing a firmware package (e.g.
     it failed to install because it was corrupted). Regression in 1.153,
     spotted in #1032970.
   * Build /var/log/firmware-summary as a 3-column summary of firmware (and
     microcode) packages getting installed (Closes: #1029849). Those three
     columns are: package, component, and reason. The reason might be dmesg
     (check-missing-firmware), modalias (install-firmware hook), or cpu
     (install-firmware hook).
 .
   [ Pascal Hambourg ]
   * Fix several glitches (Closes: #1033035):
      - Fix removal of temporary files in /target after installing firmware
        packages (they're in /tmp so the next reboot has been doing the
        trick until now).
      - Determine the package name by using the Package field instead of
        trusting the filename when installing firmware packages.
      - Make sure not to include the possible -n option when setting the
        IFACES variables in check-missing-firmware.
Checksums-Sha1:
 eaa42a8ae1ad674a6d18dd2992fa20b70ecfd296 1990 hw-detect_1.155.dsc
 9918ed7197c5f80945fe5d3cb2ef91e0b4a0298d 195232 hw-detect_1.155.tar.xz
 6e415c8183d2fb5574bf360a5815641cccfaf888 6524 hw-detect_1.155_source.buildinfo
Checksums-Sha256:
 9e55c9507048484b85870d7c89640d91cbd3c3930282ac8f51c5bd5f0eecbe50 1990 hw-detect_1.155.dsc
 b13bb32c6a162a77549dc59584418b191e6cd406e5222357b41bd71651d59b71 195232 hw-detect_1.155.tar.xz
 31a63bad2f86257964f5d545cc361fdb3dfdab1a834c1763b09a91f5f77c6b4e 6524 hw-detect_1.155_source.buildinfo
Files:
 3e9910d3b03016b60ee45f046b8e838c 1990 debian-installer standard hw-detect_1.155.dsc
 4d4062ce48d90a4802b6f8f732bfcc1e 195232 debian-installer standard hw-detect_1.155.tar.xz
 6e5293b429f561996d21cd2f9e58dd61 6524 debian-installer standard hw-detect_1.155_source.buildinfo

-----BEGIN PGP SIGNATURE-----
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=5f86
-----END PGP SIGNATURE-----

--- End Message ---
Reply to: