Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.debian.maint.boot > #74986 > unrolled thread

Re: No speech in the alpha Installer, and speech freeze in Bookworm

Back to article view | Back to linux.debian.maint.boot

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  Re: No speech in the alpha Installer, and speech freeze in Bookworm Samuel Thibault <sthibault@debian.org> - 2025-04-28 00:40 +0200
    Re: No speech in the alpha Installer, and speech freeze in Bookworm Cleverson Casarin Uliana <clul+debaccess@posteo.com.br> - 2025-04-28 01:40 +0200
    Re: No speech in the alpha Installer, and speech freeze in Bookworm Samuel Thibault <sthibault@debian.org> - 2025-04-28 12:50 +0200
    Re: No speech in the alpha Installer, and speech freeze in Bookworm Steve McIntyre <steve@einval.com> - 2025-05-14 15:20 +0200
      Re: Heads-up: please look at the proposed fix (Was: No speech in the  alpha Installer, and speech freeze in Bookworm) Cyril Brulebois <kibi@debian.org> - 2025-05-23 17:30 +0200
        Re: Lenovo Thinkpad P16s Gen3 espeakup support (Was: Heads-up:  please look at the proposed fix, Was: No speech in the alpha Installer, and  speech freeze in Bookworm) Cyril Brulebois <kibi@debian.org> - 2025-05-24 19:10 +0200
        Re: Lenovo Thinkpad P16s Gen3 espeakup support (Was: Heads-up:  please look at the proposed fix, Was: No speech in the alpha Installer, and  speech freeze in Bookworm) Cyril Brulebois <kibi@debian.org> - 2025-05-25 00:00 +0200
          Re: Lenovo Thinkpad P16s Gen3 espeakup support (Was: Heads-up:  please look at the proposed fix, Was: No speech in the alpha Installer, and  speech freeze in Bookworm) Cyril Brulebois <kibi@debian.org> - 2025-05-25 04:40 +0200
          Re: Test image bug-1092977+intel -> Partial success (Was: Lenovo  Thinkpad P16s Gen3 espeakup support) Cyril Brulebois <kibi@debian.org> - 2025-05-25 18:30 +0200
            Re: Test image bug-1092977+intel -> Partial success Cyril Brulebois <kibi@debian.org> - 2025-05-26 01:10 +0200
      Re: No speech in the alpha Installer, and speech freeze in Bookworm Cyril Brulebois <kibi@debian.org> - 2025-06-29 17:30 +0200

#74986 — Re: No speech in the alpha Installer, and speech freeze in Bookworm

[Multipart message — attachments visible in raw view] — view raw

Hello,

Giving some context to debian-boot & kernel since that's really rather a
question for them: we are not getting sound in the Debian Installer for
at least some sound card.

Cleverson Casarin Uliana, le dim. 27 avril 2025 17:08:33 +0000, a ecrit:
> Hi Samuel, I decided to run the commands on an Arch Linux live image, since
> the Bookworm image has a problem I described previously, where speech
> completely stops randomly and doesn't come back no matter.
> 
> I'm sending all of results in an attached tarball for you.

# cat /proc/asound/cards
 0 [sofhdadsp      ]: sof-hda-dsp - sof-hda-dsp
                      DellInc.-Latitude3550--05YPVY

# lspci [...]
00:1f.3 Multimedia audio controller: Intel Corporation Raptor Lake-P/U/H cAVS (rev 01)
00:1f.3 0401: 8086:51ca (rev 01)

# dmesg | grep firmware
[    4.988349] i915 0000:00:02.0: [drm] Finished loading DMC firmware i915/adlp_dmc.bin (v2.20)
[    5.005209] i915 0000:00:02.0: [drm] GT0: GuC firmware i915/adlp_guc_70.bin version 70.20.0
[    5.005218] i915 0000:00:02.0: [drm] GT0: HuC firmware i915/tgl_huc.bin version 7.9.3
[   48.343863] Bluetooth: hci0: Found device firmware: intel/ibt-0040-0041.sfi

# lsmod attached to this mail

So it looks like it's an "Intel Corporation Raptor Lake-P/U/H cAVS",
numbered 8086:51ca, which according to modules.alias should be getting
driven by snd_soc_avs and snd_sof_pci_intel_tgl, which are present
on the trixie image, and according to the dmesg, on Arch no firmware
was loaded for audio itself. But perhaps there is relation with the
graphical firmware getting loaded for i915... Is your audio normally
going through HDMI?

I don't have more ideas on this.

Samuel

[toc] | [next] | [standalone]

#74987

Hi Samuel, the laptop's audio simply comes through my laptop's speakers, 
as on any normal laptop for personal use. Do you think there can be any 
difference in case I get the commands' results from within a Bookworm 
image? I could try it, even though I'm not sure whether speech will 
survive until the end of my test.

Thanks,
Cleverson

[toc] | [prev] | [next] | [standalone]

#74990

Hello,

Roland Clobus, le lun. 28 avril 2025 12:36:37 +0200, a ecrit:
> On 28/04/2025 00:31, Samuel Thibault wrote:
> > So it looks like it's an "Intel Corporation Raptor Lake-P/U/H cAVS",
> > numbered 8086:51ca
> 
> The firmware detection for the Intel SOF family is hardcoded in the
> installer.
> 
> Recently I updated the firmware detection for the live-images [1], to have
> automatic detection, but I might have missed some other modules in newer
> folders.
> 
> When is the sound missing/broken: only during the installer, or also after
> the installation has completed (or within the live environment)?

We are here talking only about the installer. The question of the
installed system is unknown since installation couldn't be done.

> Could you try with one of the recent Debian live images?
> https://get.debian.org/images/weekly-live-builds/amd64/iso-hybrid/

Trying the live image could also help to check that is missing.

Samuel

[toc] | [prev] | [next] | [standalone]

#75186

On Wed, May 14, 2025 at 03:14:17PM +0200, Cyril Brulebois wrote:
>Hi,
>
>Roland Clobus <rclobus@rclobus.nl> (2025-04-28):
>> The firmware detection for the Intel SOF family is hardcoded in the
>> installer.
>> 
>> Recently I updated the firmware detection for the live-images [1], to
>> have automatic detection, but I might have missed some other modules
>> in newer folders.
>
>Thanks to whoever added it to the wishlist, I've just pushed a commit
>to debian-cd's master branch to refresh the list for the v6.12.y series.
>
>Looping in debian-cd@ to make sure it's getting (cherry-)picked up,
>merged, or whatever, so that it's used for RC 1. Thanks!
>
>  https://salsa.debian.org/images-team/debian-cd/-/commit/513684f3f7ec53bdbd7f177ec53f60c8a31656a5

That's cool - we're still using master for Trixie at this point, so
we're all good. :-)

-- 
Steve McIntyre, Cambridge, UK.                                steve@einval.com
  Getting a SCSI chain working is perfectly simple if you remember that there
  must be exactly three terminations: one on one end of the cable, one on the
  far end, and the goat, terminated over the SCSI chain with a silver-handled
  knife whilst burning *black* candles. --- Anthony DeBoer

[toc] | [prev] | [next] | [standalone]

#75351 — Re: Heads-up: please look at the proposed fix (Was: No speech in the alpha Installer, and speech freeze in Bookworm)

[Multipart message — attachments visible in raw view] — view raw

Hi,

Roland Clobus <rclobus@rclobus.nl> (2025-05-23):
> It turns out that I actually own hardware that needs additional firmware to
> properly work: Lenovo Thinkpad P16s Gen3.
> 
> I found out that one needs to place the firmware files inside the initrd of
> the installer for espeakup to find them, it is not sufficient to generate a
> .patterns file (that will be picked up much later by hw-detect)

Yes, this is similar to the firmware-sof-signed situation.

> I've prepared a patch for the live images, which will be merged soon:
> https://salsa.debian.org/live-team/live-build/-/merge_requests/424
> 
> The ugly thing is that I need the kernel modules 'drm' and 'i915' and their
> dependencies in addition to the firmware to enable sound for espeakup on the
> Lenovo Thinkpad.
> 
> Would it be OK to stuff so much extras in the initrd of the installer?

We already have drm.ko (and a few others, for other reasons). Do you
have a full list of required modules? I can only see a grep call, and at
least drm seems like a HUGE hammer:
 - a current gtk initrd's /kernel/drivers/gpu/drm is 560K
 - a current linux-image's /kernel/drivers/gpu/drm is 9.5M

Also, how does firmware-intel-graphics fit in this picture?


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

[toc] | [prev] | [next] | [standalone]

#75362 — Re: Lenovo Thinkpad P16s Gen3 espeakup support (Was: Heads-up: please look at the proposed fix, Was: No speech in the alpha Installer, and speech freeze in Bookworm)

[Multipart message — attachments visible in raw view] — view raw

Hi,

Roland Clobus <rclobus@rclobus.nl> (2025-05-24):
> With the big hammer I got the espeakup module working on a Lenovo Thinkpad
> P16s Gen3.
> 
> Now I did a few loops with incrementally adding kernel modules to find out
> the minimal requirement:
> * drm needs i915
> * i915 needs cec, drm_display_helper and drm_buddy
> * cec needs rc-core
> 
> > Also, how does firmware-intel-graphics fit in this picture?
> 
> When loading the i915 module, it requests several firmware files from
> firmware-intel-graphics (mtl_dmc.bin, mtl_guc_70.bin, mtl_huc_gsc.bin,
> mtl_gsc_1.bin). It turn out that these firmware files are not required for
> the espeakup module (even though dmesg complains about them)
> 
> So overall we (only) need 5 additional kernel modules:
> i915, drm_buddy, drm_display_helper, cec, rc-core
> 
> With kind regards,
> Roland Clobus
> 
> PS: The filename is rc-core.ko.xz, the module name is rc_core

(Yeah, spotted that one earlier. :))

I'd be tempted to do the .ko thing at d-i build time (we already have a
few other things in a similar-ish way) and the firmware-intel-sound (but
not firmware-intel-graphics) at cd build time. Are you happy to give an
iso a shot?


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

[toc] | [prev] | [next] | [standalone]

#75371 — Re: Lenovo Thinkpad P16s Gen3 espeakup support (Was: Heads-up: please look at the proposed fix, Was: No speech in the alpha Installer, and speech freeze in Bookworm)

[Multipart message — attachments visible in raw view] — view raw

Hi,

Roland Clobus <rclobus@rclobus.nl> (2025-05-24):
> When loading the i915 module, it requests several firmware files from
> firmware-intel-graphics (mtl_dmc.bin, mtl_guc_70.bin, mtl_huc_gsc.bin,
> mtl_gsc_1.bin). It turn out that these firmware files are not required
> for the espeakup module (even though dmesg complains about them)

Great news, thanks! The complaining and/or modalias-based stuff will
have d-i pick them up later on anyway.

(I'm leaving the question around firmware-intel-sound's metadata for
later.)

> So overall we (only) need 5 additional kernel modules: i915,
> drm_buddy, drm_display_helper, cec, rc-core

Re-used/adapted some logic in debian-installer.git, see speech-support
branch (the commit message sucks, and at least that would need some
editing before merging into master).

See kibi/for-master for the debian-cd side (patch 2/3 compared to
current master).

Test image:
  https://people.debian.org/~kibi/bug-1092977+intel/

(also contains partman-crypto fun, unrelated, but one build/stone, two
bugs/birds, right?)

I've at least confirmed I can `modprobe i915` manually…


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

[toc] | [prev] | [next] | [standalone]

#75375 — Re: Lenovo Thinkpad P16s Gen3 espeakup support (Was: Heads-up: please look at the proposed fix, Was: No speech in the alpha Installer, and speech freeze in Bookworm)

[Multipart message — attachments visible in raw view] — view raw

Cyril Brulebois <kibi@debian.org> (2025-05-24):
> Great news, thanks! The complaining and/or modalias-based stuff will
> have d-i pick them up later on anyway.
> 
> (I'm leaving the question around firmware-intel-sound's metadata for
> later.)

There's something but it's not great. That led me to doing the following
but I'm not uploading yet another image just for that. I think what
matters most at the moment is whether we can bootstrap speech synthesis.
If we end up with missing sound on the installed system, it's easy enough
(or should be) to figure out what's going on.

  https://salsa.debian.org/images-team/debian-cd/-/commit/c32217a4f86d18a872d8a2855c78cf74875fce40
  https://salsa.debian.org/images-team/debian-cd/-/commit/7989b4efac6d0514b2095b77ac780cfa88375df2

(The former I didn't link in my previous mail, the latter is new.)

The diff in /firmware/dep11 looks good to me: firmware-sof-signed stuff
unchanged, firmware-intel-sound stuff added (.patterns & .component).


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

[toc] | [prev] | [next] | [standalone]

#75405 — Re: Test image bug-1092977+intel -> Partial success (Was: Lenovo Thinkpad P16s Gen3 espeakup support)

[Multipart message — attachments visible in raw view] — view raw

Roland Clobus <rclobus@rclobus.nl> (2025-05-25):
> Looking at the WWW for this issue: Is a call to `depmod` missing?

Oh, right. The other modules I'm including in this way are used way
later, and the lack of depmod hasn't been spotted until now (the first
module udeb that gets installed papers over that lack). I'll give it
another shot.

> However, I'm a bit surprised by having kernel 6.12.29 in the test
> image.

That's with a local d-i build, built against sid (I've been chasing
quite a number of bugs lately).

> I tested with kernel 6.12.27 (as that is present in the RC1 image),
> because I assumed that 6.12.29 would not migrate to trixie due to
> #1106070 (which would break all live images)
> 
> So in summary:
> * The netinst image needs `depmod` for the added kernel module(s)
> * I didn't test the espeakup yet, but assume it will work

Would you be willing to give espeakup a try with an updated image,
with that extra depmod? I could also switch to building against trixie
if that makes any difference for your testing, but if that's not
stricitly required, I'd love not to toggle back and forth between trixie
and sid (it's easy to leave extra packages — deb and/or udeb — on the
side of the road while doing so).

> * I hope we will have kernel 6.12.30, but definitely not 6.12.29

Sure thing.


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

[toc] | [prev] | [next] | [standalone]

#75426 — Re: Test image bug-1092977+intel -> Partial success

[Multipart message — attachments visible in raw view] — view raw

Hi,

Roland Clobus <rclobus@rclobus.nl> (2025-05-25):
> On 25/05/2025 18:26, Cyril Brulebois wrote:
> > Roland Clobus <rclobus@rclobus.nl> (2025-05-25):
> > > Looking at the WWW for this issue: Is a call to `depmod` missing?
> > 
> > Oh, right. The other modules I'm including in this way are used way
> > later, and the lack of depmod hasn't been spotted until now (the first
> > module udeb that gets installed papers over that lack). I'll give it
> > another shot.
> 
> Thanks.

Here we go [+depmod]:
  https://people.debian.org/~kibi/bug-1092977+intel/debian-13.0.0-amd64-NETINST-1+depmod+efd.iso

This includes your suggestion (on Salsa) to embed both packages at the
same time, otherwise we only see the first or the last filename in
./usr/lib/firmware/.embedded-firmware-debs (depending on the cpio
invocation and/or what happens at runtime) [hence +efd as well].

I've verified that modules.* include information about i915 this time,
while the previous iteration didn't. I'm also able to load this module
by switching to a shell from the language screen, before anything else
happens, so I hope everything is in place for the speech synthesis to
start on your hardware.


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

[toc] | [prev] | [next] | [standalone]

#75770

[Multipart message — attachments visible in raw view] — view raw

Hi,

Roland Clobus <rclobus@rclobus.nl> (2025-05-27):
> I agree, but we would support a few more sound cards now.
> 
> For the cAVS sound card firmware-intel-sound might be required, and no
> additional kernel modules. The firmware-intel-sound.patterns file in the
> netinst image does not contain that identifier (even after the update),
> however the patterns file in the live images does contain that identifier
> [1].
> Interestingly enough, the identifier 8086:51ca (Intel cAVS) is mentioned in
> both patterns files on the live images. I read some kernel code, and
> determined at firmware-intel-sound would be required, but I forgot to write
> down exactly how I came to that conclusion.
> 
> I ended up with [3]:
> for _ALIAS in $(dpkg-query -L ${_LINUX_KERNEL_PACKAGE} | egrep
> 'kernel/sound/soc/intel/.*\.ko(.xz)?$'); do /usr/sbin/modinfo ${_ALIAS} |
> awk '/^alias:/ { print $2 }'; done | sort -u | awk '{ gsub(/[*]/, ".*", $1);
> printf "^%s$\n", $1 }' > ${_FW_DEP11_DIR}/${_PACKAGE}.patterns
> 
> which parses different kernel modules compared to the logic for
> firmware-sof-signed:
> for _ALIAS in $(dpkg-query -L ${_LINUX_KERNEL_PACKAGE} | egrep
> '(kernel/sound/soc/sof/.*\.ko(.xz)?|/snd-soc-.*sof.*\.ko(.xz)?)$'); do
> /usr/sbin/modinfo ${_ALIAS} | awk '/^alias:/ { print $2 }'; done | sort -u |
> awk '{ gsub(/[*]/, ".*", $1); printf "^%s$\n", $1 }' >
> ${_FW_DEP11_DIR}/${_PACKAGE}.patterns
> 						
> On top of this, there are 2 scenarios to consider:
> 1) Have espeakup work during the installation
> 2) Ensure that the installed system has the correct firmware files
> 
> For 1) the additional firmware is required in the initrd
> For 2) the .patterns files are required (the firmware-intel-sound.patterns
> file was not generated

I thought I had followed up to this but it seems that didn't happen…

At the moment, I'm very much not convinced that the initial debian-cd
patches (besides fixing firmware inclusion logic) are correct, so it would
make sense to me to get the firmware-intel-sound addition reverted, so
that the next release's focus is on trying to ensure i915 gets loaded
successfully (hopefully the existing firmware-sof-signed integration gets
improved, with speech synthesis starting right off the bat).

In a follow-up release (possibly after 13.0 depending on the release
team's timing when it comes to the full freeze and the initial trixie
release), we can try and come up with proper firmware-intel-sound support,
which would mean (as you identified) making sure speech synthesis works
within d-i, plus ensuring the relevant package(s) get deployed in the
installed system:
 - Having one without the other would only mean wasting people's time, I
   fear.
 - Plus the huge extra cost for the initrd (size-wise) is clearly not
   worth it at this time.

Also, I suggest turning the firmware-intel-sound topic into a dedicated
bug report, instead of keeping it entangled with firmware-sof-signed in
this overlong thread plus merge requets / commit comments on Salsa…


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

[toc] | [prev] | [standalone]

Back to top | Article view | linux.debian.maint.boot

csiph-web