Privacy LED may not be turned on on OEM Renegade platform with Intel IPU7 camera

[You-Sheng Yang]

[SRU Justification]

[ Impact ]

With CVS firmware >= version 132.39.66.9(0x84.0x27.0x42.0x9), SET_HOST_IDENTIFIER was added. One of the field was not initialized correctly, and it leads to the inability to turn off the privacy LED.

Proposed fix merged as upstream commit https://github.com/intel/vision-drivers/commit/fc4981411e140674b016a3d7dc062434848f5311.

[ Test Plan ]

While different platforms may have different versions of CVS firmware, it's only known reproducible on a few still in-development OEM platforms. To verify:
0. verify CVS firmware version:
   ```
   $ cd /sys/bus/i2c/devices/i2c-INTC*
   $ echo version | sudo tee cmd
   $ cat coredump
   ...
   CVS Firmware Ver: 132.39.66.9 (0x84.0x27.0x42.0x9)
   ...
   ```

1. install linux-oem-6.11 >= 6.11.0-1019,

2. install intel-vision-dkms from -proposed pocket,

3. reboot to the same linux-oem-6.11 kernel, and browse to https://webcamtests.com to activate camera use. The privacy LED should be turned on while using, and off while camera stopped or browser terminated.

4. also covered platforms of previous generations with older CVS fw version:
* Bolan14-DVT2-C3-202410-35910, cvs fw 132.24.50.50 (0x84, 0x18, 0x32, 0x32).
* TRBL-DVT2-C1-202407-34214, cvs fw 1.19.0.0 (0x1, 0x13, 0x0, 0x0).
* TRK4-DVT2-C4-202411-35941, cvs fw 132.24.50.50 (0x84, 0x18, 0x32, 0x32).

[ Where problems could occur ]

No. This is a field in the command data of a SET_HOST_IDENTIFIER command that is always sent in the vision-drivers >= 0~git202412180414.6b1462d8. When a order CVS fw receives such command, it replies a command error and does nothing. For a latter CVS fw, this field affects the privacy LED governing and should be fixed.

[ Other Info ]

The target kernel is linux-oem-6.11, and we're currently compiling visions-drivers/plucky into it. Nominate for Plucky and Questing.

For Oracular, still under discussion.

========== original bug report ==========

[Expects]

1. Launch camera apps like https://webcamtests.com in Firefox/Chrome
2. The privacy LED should be turned on

[Actual]

Privacy LED off all the time.

---
ProblemType: Bug
ApportVersion: 2.28.1-0ubuntu3.5
Architecture: amd64
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC0: ubuntu 1775 F.... pipewire
                      ubuntu 1781 F.... wireplumber
 /dev/snd/seq: ubuntu 1775 F.... pipewire
CasperMD5CheckMismatches: ./casper/initrd ./casper/vmlinuz ./casper/minimal.standard.live.hotfix.manifest ./casper/minimal.standard.live.hotfix.size ./casper/minimal.standard.live.size ./casper/minimal.manifest ./casper/minimal.standard.manifest ./casper/minimal.standard.size ./casper/minimal.hotfix.size ./casper/minimal.standard.live.hotfix.squashfs ./casper/minimal.standard.hotfix.squashfs ./casper/minimal.standard.hotfix.size ./casper/minimal.hotfix.squashfs ./casper/minimal.standard.live.manifest ./casper/minimal.size ./boot/grub/grub.cfg
CasperMD5CheckResult: fail
DistributionChannelDescriptor:
 # This is the distribution channel descriptor for Ubuntu 24.04 for Dell
 # For more information see http://wiki.ubuntu.com/DistributionChannelDescriptor
 canonical-oem-somerville-noble-oem-24.04b-next-20250410-214
DistroRelease: Ubuntu 24.04
InstallationDate: Installed on 2025-04-10 (3 days ago)
InstallationMedia: Ubuntu OEM 24.04.1 LTS "Noble Numbat" - Release amd64 (20250407)
IwConfig:
 lo no wireless extensions.

 wlp0s20f3 no wireless extensions.
MachineType: Dell Inc. Dell Pro Max 14 Premium MA14250
Package: vision-drivers
ProcFB: 0 i915drmfb
ProcKernelCmdLine: BOOT_IMAGE=/boot/vmlinuz-6.11.0-1019-oem root=UUID=3ab86186-55fb-44bc-a01c-67f10fb1e358 ro quiet splash vt.handoff=7
ProcVersionSignature: Ubuntu 6.11.0-1019.19-oem 6.11.11
RelatedPackageVersions:
 linux-restricted-modules-6.11.0-1019-oem N/A
 linux-backports-modules-6.11.0-1019-oem N/A
 linux-firmware 20240318.git3b128b60-0ubuntu2.11+staging.7
Tags: noble
Uname: Linux 6.11.0-1019-oem x86_64
UpgradeStatus: No upgrade log present (probably fresh install)
UserGroups: N/A
_MarkForUpload: True
dmi.bios.date: 04/02/2025
dmi.bios.release: 0.3
dmi.bios.vendor: Dell Inc.
dmi.bios.version: 0.3.37
dmi.board.vendor: Dell Inc.
dmi.chassis.type: 10
dmi.chassis.vendor: Dell Inc.
dmi.ec.firmware.release: 0.0
dmi.modalias: dmi:bvnDellInc.:bvr0.3.37:bd04/02/2025:br0.3:efr0.0:svnDellInc.:pnDellProMax14PremiumMA14250:pvr:rvnDellInc.:rn:rvr:cvnDellInc.:ct10:cvr:sku0D32:
dmi.product.family: Dell Pro Max Laptops
dmi.product.name: Dell Pro Max 14 Premium MA14250
dmi.product.sku: 0D32
dmi.sys.vendor: Dell Inc.

[You-Sheng Yang (vicamo)]

apport information

[You-Sheng Yang (vicamo)]

apport information

[You-Sheng Yang (vicamo)]

apport information

[You-Sheng Yang (vicamo)]

apport information

[You-Sheng Yang (vicamo)]

apport information

[You-Sheng Yang (vicamo)]

apport information

[You-Sheng Yang (vicamo)]

apport information

[You-Sheng Yang (vicamo)]

apport information

[You-Sheng Yang (vicamo)]

apport information

[You-Sheng Yang (vicamo)]

apport information

[You-Sheng Yang (vicamo)]

apport information

[You-Sheng Yang (vicamo)]

apport information

[You-Sheng Yang (vicamo)]

apport information

[You-Sheng Yang (vicamo)]

apport information

[You-Sheng Yang (vicamo)]

apport information

[You-Sheng Yang (vicamo)]

apport information

[You-Sheng Yang (vicamo)]

apport information

[You-Sheng Yang (vicamo)]

Attach proposed debdiff for Plucky and Questing. Also pushed to PPA
https://launchpad.net/~vicamo/+archive/ubuntu/sru/ for review &
verification.

Need sponsor.

[Dave Jones (waveform)]

Had a quick look at this during a patch-pilot shift -- the patch is simple enough, and the test plan looks reasonable, but it only covers the system *after* the proposed version is installed. Ideally, the test plan should re-produce the bug (with the released version of vision-drivers), have the user install the proposed version and re-test to show that this change (and only this change) fixes the issue (as opposed to anything random in the background like other upgrades).

In other words, the test plan is *almost* complete but could do with one additional step at the top, for example, to visit https://webcamtests.com and show that the privacy LED does not light.

The section on where problems could occur needs more detail: it mentions that an error is reported in dmesg. Does that apply to the affect platforms only, or to other platforms? The mere fact an error is reported in dmesg may cause other issues in itself. Furthermore, just because this is fixing an initialization value only, does not make it any less likely that the fix may have knock-on effects. The SRU team are rather likely to pick up on this section as not having enough detail currently.

[You-Sheng Yang (vicamo)]

@waveform, actually we tested released platforms on following platforms:
* Bolan14-DVT2-C3-202410-35910, cvs fw 132.24.50.50 (0x84, 0x18, 0x32, 0x32).
* TRBL-DVT2-C1-202407-34214, cvs fw 1.19.0.0 (0x1, 0x13, 0x0, 0x0).
* TRK4-DVT2-C4-202411-35941, cvs fw 132.24.50.50 (0x84, 0x18, 0x32, 0x32).

These information was replied on JIRA. The test plan was intended for reproduce/verification step, and existing platforms were covered as well.

[You-Sheng Yang (vicamo)]

Updated the two sections in SRU justification to clarify concerns from comments #19.