Bug#1032416:

[Adam Baxter]

IRC has let me know that d-i needs the symlinks in the original ISO are needed, 
rufus breaks these and I missed it on https://wiki.debian.org/DebianInstall.

Thanks anyway, I guess.

Bug#1032416:

[Adam Baxter]

I have retried with 
https://cdimage.debian.org/cdimage/weekly-builds/amd64/iso-dvd/debian-testing-amd64-DVD-1.iso
 and hit the same issue. Image was written with dd and definitely contains 
firmware-iwlwifi_20230210-1_all.deb

I'd also like to note 
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1031696, from the creator of 
Rufus.

Bug#1032416:

[Cyril Brulebois]

Adam Baxter  (2023-03-06):
> IRC has let me know that d-i needs the symlinks in the original ISO
> are needed, rufus breaks these and I missed it on
> https://wiki.debian.org/DebianInstall.

What about a syslog file with D-I Bookworm Alpha 2 and an image that's
written correctly (i.e. non-broken symlinks)?


Cheers,
-- 
Cyril Brulebois (k...@debian.org)
D-I release manager -- Release team member -- Freelance Consultant


signature.asc
Description: PGP signature

Bug#1031738: installation-guide: documentation about limits to kernel boot parameters is outdated

[James Addison]

Source: installation-guide
Followup-For: Bug #1031738
Control: tags -1 patch

Please note: the patch previously offered here isn't suitable; it turns out
that the param limits described in it (32 for Linux, 128 for User Mode Linux)
apply only to the number of argument items that are passed to the 'init'
process from the kernel command line -- not to the number of known kernel
parameters ('ro', 'quiet', and so on) that a kernel will accept at boot-time.

References:

 - 
https://lore.kernel.org/linux-doc/CALDQ5NwGTi3q9B=ezat5h_eltr1cdur9j13utb1-dck-fxo...@mail.gmail.com/T/#t

 - 
https://salsa.debian.org/installer-team/installation-guide/-/merge_requests/24#note_387450

Processed: Re: installation-guide: documentation about limits to kernel boot parameters is outdated

[Debian Bug Tracking System]

Processing control commands:

> tags -1 patch
Bug #1031738 [src:installation-guide] installation-guide: documentation about 
limits to kernel boot parameters is outdated
Ignoring request to alter tags of bug #1031738 to the same tags previously set

-- 
1031738: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1031738
Debian Bug Tracking System
Contact ow...@bugs.debian.org with problems

Bug#1032431: installation-reports: Bullseye installation on Fujitsu LIFEBOOK U9312

[Dennis van Dok]

Package: installation-reports
Severity: normal

Boot method: USB
Image version: 
https://cdimage.debian.org/cdimage/unofficial/non-free/cd-including-firmware/11.6.0+nonfree/amd64/iso-cd/firmware-11.6.0-amd64-netinst.iso
Date: 2023-03-02

Machine: Fujitsu LIFEBOOK U9312
Partitions: 
FilesystemType  1K-blocks  Used  Available Use% Mounted 
on
udev  devtmpfs   16196776 0   16196776   0% /dev
tmpfs tmpfs   3247404  23843245020   1% /run
/dev/mapper/bobo--vg-root ext4 1920358680 374068376 1448667676  21% /
tmpfs tmpfs  16237012183456   16053556   2% /dev/shm
tmpfs tmpfs  5120 8   5112   1% 
/run/lock
/dev/nvme0n1p2ext2 481642171481 285176  38% /boot
/dev/nvme0n1p1vfat 523248  5928 517320   2% 
/boot/efi
tmpfs tmpfs   3247400   1283247272   1% 
/run/user/1000


Model: SAMSUNG MZVL22T0HBLB-00B07 (nvme)
Disk /dev/nvme0n1: 2048GB
Sector size (logical/physical): 512B/512B
Partition Table: gpt
Disk Flags: 

Number  Start   End SizeFile system  Name  Flags
 1  1049kB  538MB   537MB   fat32  boot, esp
 2  538MB   1050MB  512MB   ext2
 3  1050MB  2048GB  2047GB


Base System Installation Checklist:
[O] = OK, [E] = Error (please elaborate below), [ ] = didn't try it

Initial boot:   [O]
Detect network card:[O*]
Configure network:  [O]
Detect media:   [O]
Load installer modules: [O]
Clock/timezone setup:   [O]
User/password setup:[O]
Detect hard drives: [O]
Partition hard drives:  [O]
Install base system:[O]
Install tasks:  [O]
Install boot loader:[O]
Overall install:[O]

Comments/Problems:

Using a netboot install image with non-free firmware; the wireless card
was not detected during the installation so I used the wired network
adapter instead.

The touchpad was not working, but an external mouse could be used.
This later turned out to be a minor issue but a cost me the most time.
Loading the i2c_hid_acpi module got it going.

Since this was a fairly new laptop I expected some things not to work (yet).
The missing firmware files reported by the kernel could be downloaded from
https://git.kernel.org/pub/scm/linux/kernel/git/firmware/linux-firmware.git
which was sufficient to get the wireless card and bluetooth to work.

I always expected to run the system in a mixed stable/testing setup
because I wanted to install pipewire/wireplumber which seems to work
better with Debian 12. I ended up switching almost entirely to testing.

The sound card required the latest firmware from the SOF project. To get
the digital microphone working a different topology file was needed as
outlined on this page:

https://thesofproject.github.io/latest/getting_started/intel_debug/suggestions.html#digital-mic-issues

And this bug report:

https://github.com/thesofproject/linux/issues/4099


The system seems to do great otherwise, as it is my daily workhorse to
replace an earlier Fujitsu model (LIFEBOOK U937). The only non-functioning
element is the Fujitsu PalmSecure biometric sensor.

Also see the hardware report

https://linux-hardware.org/?probe=19a72f502b



-- Package-specific info:

==
Installer lsb-release:
==
DISTRIB_ID=Debian
DISTRIB_DESCRIPTION="Debian GNU/Linux installer"
DISTRIB_RELEASE="11 (bullseye) - installer build 20210731+deb11u7+b1"
X_INSTALLATION_MEDIUM=cdrom

==
Installer hardware-summary:
==
uname -a: Linux bobo 5.10.0-20-amd64 #1 SMP Debian 5.10.158-2 (2022-12-13) 
x86_64 GNU/Linux
lspci -knn: 00:00.0 Host bridge [0600]: Intel Corporation Device [8086:4601] 
(rev 04)
lspci -knn: Subsystem: Fujitsu Client Computing Limited Device [1e26:0087]
lspci -knn: 00:02.0 VGA compatible controller [0300]: Intel Corporation Device 
[8086:46a8] (rev 0c)
lspci -knn: Subsystem: Fujitsu Client Computing Limited Device [1e26:009c]
lspci -knn: 00:04.0 Signal processing controller [1180]: Intel Corporation 
Device [8086:461d] (rev 04)
lspci -knn: Subsystem: Fujitsu Client Computing Limited Device [1e26:0087]
lspci -knn: 00:06.0 PCI bridge [0604]: Intel Corporation Device [8086:464d] 
(rev 04)
lspci -knn: Kernel driver in use: pcieport
lspci -knn: 00:07.0 PCI bridge [0604]: Intel Corporation Device [8086:466e] 
(rev 04)
lspci -knn: Kernel driver in use: pcieport
lspci -knn: 00:07.1 PCI bridge [0604]: Intel Corporation Device [8086:463f] 
(rev 04)
lspci -knn: Kernel driver in use: pcieport
lspci -knn: 00:08.0 System peripheral [0880]: Intel Corporation Device 
[8086:464f] (rev 04)
lspci -knn: Subsystem: Fujitsu Client Computing Limited Device [1e26:0087]
lspci -knn: 00:0a.0 Signal processing controller [1180]: Intel Co

Bug#1032405: installation-reports: missing nvidia nouveau firmware: nvac_fuc084 & nvac_fuc084d

[Sven Joachim]

On 2023-03-06 08:55 +0100, Cyril Brulebois wrote:

> Fab Stz  (2023-03-06):
>> Package: installation-reports
>> Severity: normal
>>
>> I couldn't try on the real system which is an iMac 9.1 (because I don't have
>> physical access to it presently), but there is no package shipping these
>> firmware files which are required by nouveau.
>>
>> firmware: failed to load nouveau/nvac_fuc084 (-2)
>> firmware: failed to load nouveau/nvac_fuc084d (-2)
>>
>> Boot method: DVD
>> Image version: 
>> https://cdimage.debian.org/cdimage/weekly-builds/amd64/iso-dvd/debian-testing-amd64-DVD-1.iso
>> Date: 2023-03-06
>>
>> Machine: imac 9.1
>>
>> Until now, I installed them this way as per
>> https://nouveau.freedesktop.org/VideoAcceleration.html
>>
>> mkdir /tmp/nouveau
>> cd /tmp/nouveau
>> wget https://raw.github.com/envytools/firmware/master/extract_firmware.py
>> wget 
>> http://us.download.nvidia.com/XFree86/Linux-x86/325.15/NVIDIA-Linux-x86-325.15.run
>> sh NVIDIA-Linux-x86-325.15.run --extract-only
>> python2.7 extract_firmware.py  # this script is for python 2 only
>> mkdir /lib/firmware/nouveau
>> cp -d nv* vuc-* /lib/firmware/nouveau/
>>
>>
>> It would be nice if there were a firmware package for it in bookworm
>
> I can't really check license etc. right now, but that looks like
> something that should be filed against src:firmware-nonfree, which ships
> some nvidia-related firmware files in its firmware-misc-nonfree binary
> package.

Already done in #990662.  The firmware in question is not distributable,
so it cannot be shipped in firmware-misc-nonfree.  A downloader package
in contrib could be a solution, though.

Cheers,
   Sven

Bug#1032435: installation-reports: Fails to install GRUB bootloader

[John Talbut]

Package: installation-reports
Severity: important
Tags: d-i
X-Debbugs-Cc: j...@dpets.uk

Boot method: USB
Image version: 
https://gemmei.ftp.acc.umu.se/images/bookworm_di_alpha2/amd64/iso-cd/debian-bookworm-DI-alpha2-amd64-netinst.iso

Date: <6 March 2023>

Machine: Ideapad 5 14
Partitions: 
Disk /dev/nvme1n1: 238.47 GiB, 256060514304 bytes, 500118192 sectors
Disk model: WDC PC SN530 SDBPMPZ-256G-1101
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disklabel type: gpt
Disk identifier: 5E4D836B-80D7-42DA-BB78-BB015B1D879C

Device   Start   End   Sectors   Size Type
/dev/nvme1n1p12048391167389120   190M EFI System
/dev/nvme1n1p2  391168   1368063976896   477M Linux filesystem
/dev/nvme1n1p3 1368064 500117503 498749440 237.8G Linux filesystem


Disk /dev/nvme0n1: 931.51 GiB, 1000204886016 bytes, 1953525168 sectors
Disk model: Samsung SSD 980 1TB
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 16384 bytes / 131072 bytes
Disklabel type: gpt
Disk identifier: C3734DD7-2C43-49F4-BE7D-5F796EA39BFD

Device StartEndSectors   Size Type
/dev/nvme0n1p1  2048 1953523711 1953521664 931.5G Linux filesystem


Disk /dev/sda: 7.21 GiB, 7746879488 bytes, 15130624 sectors
Disk model: USB DISK 2.0
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disklabel type: dos
Disk identifier: 0x28b1675d

Device Boot Start End Sectors  Size Id Type
/dev/sda1  *0 1400831 1400832  684M  0 Empty
/dev/sda24472   23319   18848  9.2M ef EFI (FAT-12/16/32)


Disk /dev/mapper/nvme0n1p1_crypt: 931.5 GiB, 1000186314752 bytes, 
1953488896 sectors

Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 16384 bytes / 131072 bytes


Disk /dev/mapper/nvme1n1p3_crypt: 237.81 GiB, 255342936064 bytes, 
498716672 sectors

Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes


Disk /dev/mapper/dk-rt: 23.28 GiB, 24998051840 bytes, 48824320 sectors
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes


Disk /dev/mapper/dk-sw: 9.31 GiB, 220736 bytes, 19529728 sectors
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes


Base System Installation Checklist:
[O] = OK, [E] = Error (please elaborate below), [ ] = didn't try it

Initial boot:   [O]
Detect network card:[O]
Configure network:  [O]
Detect media:   [O]
Load installer modules: [O]
Clock/timezone setup:   [O]
User/password setup:[O]
Detect hard drives: [O]
Partition hard drives:  [O]
Install base system:[O]
Install tasks:  [O]
Install boot loader:[E]
Overall install:[E]

Comments/Problems:



The system has a 256MB SSD and a 1TB SSD.  The installation is LUKS 
encrypted apart from boot and efi partitions.  The 1TB SSD is for user 
files and is LVM volume group da and logical volume da, the 256MB is 
volume group dk and divided into root, home, var, tmp and swap 
partitions.  The installation proceeded correctly up to installing GRUB. 
 It is not clear where to install it but wherever I try I come up with 
a message "Unable to install GRUB in /dev/sda" (or the equivalent in 
other partitions).  It did seem as if initrd was trying to mount root on 
a partition that had not been decrypted, but trying in rescue mode when 
the partitions are decrypted before attempting to install GRUB comes up 
with the same error.



-- Package-specific info:

==
Installer lsb-release:
==
DISTRIB_ID=Debian
DISTRIB_DESCRIPTION="Debian GNU/Linux installer"
DISTRIB_RELEASE="12 (bookworm) - installer build 20230217"
X_INSTALLATION_MEDIUM=cdrom

==
Installer hardware-summary:
==
Attached

hardware-summary.gz
Description: application/gzip

Bug#1032435: installation-reports: Fails to install GRUB bootloader

[Pascal Hambourg]

Hello John,

On 06/03/2023 at 18:19, John Talbut wrote:


It is not clear where to install it but wherever I try I come up with 
a message "Unable to install GRUB in /dev/sda" (or the equivalent in 
other partitions).


/dev/sda is the USB stick which contains the installer. You cannot 
install GRUB there.


Does the installer boot in EFI or legacy mode ?
grub-installer prompts in which device to install GRUB when installing 
for legacy/BIOS boot. It means either the installer booted in legacy 
mode or booted in EFI mode but you chose to revert to legacy boot after 
being warned (maybe wrongly) about an existing system set up for legacy 
boot. When installing GRUB for EFI boot, it automatically chooses one of 
the selected EFI partitions.


The partition layout on /dev/nvme1n1 is set up for EFI boot. GRUB could 
be installed for legacy boot in /dev/nvme1n1, but it would not be best 
because there is not "BIOS boot" (bios_grub) partition.

Error debian Acer nitro 5

[Diego Santos]

package: installation reports Boot method:  Image version:  Data:  Machine:  External use hdc 1 2 with dual boot windows
I tested it without it as main Output from lspci -knn (or lspci -nn): Basic
system installation checklist: [O] = OK, [E] = Error (describe below), [ ]
= did not try Initial boot: [ ]ok Detect network card: [ ]ok Configure
network: [ ],,ok free kernel mode Detect installation medium: [ ] gave
installation error Load installer modules: [ ] loads davfail Detect hard
drives: [ ] ok Partition hard drives: [ ] ok Install base system: [ ] ok
Clock/time zone setting: [ ]ok Set username/password: [ ]ok Installation
tasks: [ ] ok Install the bootloader: [ ] ok Installation total: [ ] not
installed Accepts Ubuntu 19.04 kernel firmware well

Bug#1032431: installation-reports: Bullseye installation on Fujitsu LIFEBOOK U9312

[Cyril Brulebois]

Hi Dennis,

And thanks for your report.

Dennis van Dok  (2023-03-06):
> Using a netboot install image with non-free firmware; the wireless
> card was not detected during the installation so I used the wired
> network adapter instead.

If you have a few minutes to spare, it would be helpful to know whether
the official D-I Bookworm Alpha 2 image is working better in that regard.
With a newer kernel, and newer firmware packages included (in an official
capacity), I'd expect your card to be detected and working out of the box.
There's no need to go through the whole installation process, just
reaching the network configuration step would be sufficient.

That being said, if you ended up with pulling various bits from testing
anyway, and if those work fine, the installer should be fine as well, so
whether to run that extra test is really up to you. :)

> The touchpad was not working, but an external mouse could be used.
> This later turned out to be a minor issue but a cost me the most time.
> Loading the i2c_hid_acpi module got it going.

Do you need to load it manually in the installed system as well?

The Bullseye installer ships i2c-hid.ko, but no i2c-hid-acpi.ko.

The Bookworm installer ships both.

If the issue you've seen is about the missing touchpad in the installer,
that's unfortunate but expected in Bullseye, and this should work better
with Bookworm. It might just be that the kernel in Bullseye is a tad too
old, seeing that i2c-hid-acpi.ko was introduced in the v5.11 release cycle
(as a side effect of i2c-hid.ko getting reorganized).

If the issue you've seen is about the kernel not loading a required module
automatically, this is likely going to be a problem with Bookworm, both in
the installer environment and in the installed system.

Sometimes, it might be a dependency towards another component that's not
expressed in the Linux kernel code (e.g. via a MODULE_SOFTDEP macro). But
it's a little hard to brainstorm this without knowing more about when you
had to load i2c_hid_acpi manually.

> Since this was a fairly new laptop I expected some things not to work (yet).
> The missing firmware files reported by the kernel could be downloaded from
> https://git.kernel.org/pub/scm/linux/kernel/git/firmware/linux-firmware.git
> which was sufficient to get the wireless card and bluetooth to work.

Hopefully Bookworm would be much better there as there were upgrades to
newer upstream releases, meaning support both in the installer and in the
installed system.

> The sound card required the latest firmware from the SOF project.

That should get installed automatically too.

> To get the digital microphone working a different topology file was
> needed as outlined on this page:
> 
> https://thesofproject.github.io/latest/getting_started/intel_debug/suggestions.html#digital-mic-issues
> 
> And this bug report:
> 
> https://github.com/thesofproject/linux/issues/4099

I suppose this isn't something we can do much about in the installer
context… we focus on getting sound out, not in anyway. ;) (For speech
synthesis.)

> The system seems to do great otherwise, as it is my daily workhorse to
> replace an earlier Fujitsu model (LIFEBOOK U937). The only
> non-functioning element is the Fujitsu PalmSecure biometric sensor.

Nice to hear, and I really hope the Bookworm installer is performing
better with that machine. :)


Cheers,
-- 
Cyril Brulebois (k...@debian.org)
D-I release manager -- Release team member -- Freelance Consultant


signature.asc
Description: PGP signature

Bug#1028547: MT7921e firmware blob exists but is not found

[Jamie Wilkinson]

I am installing on an Asus PN51-E1 which includes the Mediatek MT7921e
wifi device.

I came across this bug while trying to install from
https://cdimage.debian.org/cdimage/bookworm_di_alpha2/ netinst iso on
a USB stick.

I see the same error, but what I was trying to debug when I arrived at
this bug is why we see the error in dmesg:

mt7921e :05:00.0: firmware: failed to load
mediatek/WIFI_MT7961_patch_mcu_1_2_hdr.bin (-2)

because when I break to the console in the installer, I see in
/lib/firmware this very file:

~ # ls /lib/firmware/mediatek/WIFI_*
WIFI_MT7922_patch_mcu_1_1_hdr.bin
WIFI_MT7961_patch_mcu_1_2_hdr.bin
WIFI_RAM_CODE_MT7922_1.bin
WIFI_RAM_CODE_MT7961_1.bin

When I tried again in Expert install mode, I looked before detecting
network devices, and saw that /lib/firmware was empty.  What's
populating this directory?  Is it a chroot and thus the kernel module
can't see it?  Or... is the kernel module looking in a different path?

Bug#1028547: MT7921e firmware blob exists but is not found

[Cyril Brulebois]

Jamie Wilkinson  (2023-03-07):
> I am installing on an Asus PN51-E1 which includes the Mediatek MT7921e
> wifi device.
> 
> I came across this bug while trying to install from
> https://cdimage.debian.org/cdimage/bookworm_di_alpha2/ netinst iso on
> a USB stick.
> 
> I see the same error, but what I was trying to debug when I arrived at
> this bug is why we see the error in dmesg:
> 
> mt7921e :05:00.0: firmware: failed to load
> mediatek/WIFI_MT7961_patch_mcu_1_2_hdr.bin (-2)

That's expected: no firmware is deployed at first, the kernel complains,
d-i notices, and deploys stuff if relevant firmware packages are found,
and reload the relevant modules. This reloading is what breaks the
kernel. See upstream bug report:

https://bugzilla.kernel.org/show_bug.cgi?id=216844

> because when I break to the console in the installer, I see in
> /lib/firmware this very file:
> 
> ~ # ls /lib/firmware/mediatek/WIFI_*
> WIFI_MT7922_patch_mcu_1_1_hdr.bin
> WIFI_MT7961_patch_mcu_1_2_hdr.bin
> WIFI_RAM_CODE_MT7922_1.bin
> WIFI_RAM_CODE_MT7961_1.bin

That's expected after an initial check-missing-firmware run.

> When I tried again in Expert install mode, I looked before detecting
> network devices, and saw that /lib/firmware was empty.  What's
> populating this directory?  Is it a chroot and thus the kernel module
> can't see it?  Or... is the kernel module looking in a different path?

You can check /var/log/syslog, looking for check-missing-firmware lines.
You'll see what it notices, and does.


Cheers,
-- 
Cyril Brulebois (k...@debian.org)
D-I release manager -- Release team member -- Freelance Consultant


signature.asc
Description: PGP signature

Bug#1028547: Firmware is not udpkg -i before hardware detection

[Jamie Wilkinson]

I think I've figured out why we see these errors.

I restarted an Expert mode installation with the latest nightly DI
https://cdimage.debian.org/cdimage/daily-builds/daily/arch-latest/amd64/iso-cd/
dated
2023-03-07 04:13 size 758M

I ran through the installation to "load installer components from
installation media", then opened the terminal on tty2.

I changed into /cdrom/firmware and used `udpkg -i *.deb` to install
everything.  There were many warnings about missing tools which are
now lost off the console.

I manually ran `modprobe mt7921e` and returned without error.  `dmesg`
has the following in the tail:

mt7921e: firmware: direct-loading firmware
mediatek/WIFI_MT7961_patch_mcu_1_2_hdr.bin
mt7921e: HW/SW Version: 0x8a108a10, Build Time: 20230117170855a
mt7921e: firmware: direct-loading firmware mediatek/WIFI_RAM_CODE_MT7961_1.bin
mt7921e: HW/SW Version: 01, Build Time: 20230117170942
mt7921e: firmware: direct-loading firmware mediatek/WIFI_RAM_CODE_MT7961_1.bin
mt7921e: wlp3s0: renamed from wlan0

I then returned to the menu on tty1 and detected network hardware, and
the installer now sees the WiFi interface.

So I think the problem here is that the firmware udebs are not
unpacked before hardware detection starts.

Bug#1028547: Firmware is not udpkg -i before hardware detection

[Cyril Brulebois]

Jamie Wilkinson  (2023-03-07):
> I think I've figured out why we see these errors.

I don't think you did.

> So I think the problem here is that the firmware udebs are not
> unpacked before hardware detection starts.

Those are firmware debs, not firmware udebs. And no, they shouldn't be
unpacked before hardware detection starts.

See my previous answer for the details.

Attaching a full syslog would help figure out what happened *if*
there's indeed something unexpected there.


Cheers,
-- 
Cyril Brulebois (k...@debian.org)
D-I release manager -- Release team member -- Freelance Consultant


signature.asc
Description: PGP signature