Your message dated Tue, 1 Sep 2026 21:16:45 +0200 with message-id <4yge5zaiwk3rat76tlyw343wvkri3db65h6cpfa74isqoaljng@dfpdhmcgvl7k> and subject line Re: Bug#1146417: linux: please enable CONFIG_VIDEO_OV5675 and TPS68470 gpio/regulator/clk on amd64 has caused the Debian Bug report #1146417, regarding linux: please enable CONFIG_VIDEO_OV5675 and TPS68470 gpio/regulator/clk on amd64 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 [email protected] immediately.) -- 1146417: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1146417 Debian Bug Tracking System Contact [email protected] with problems
--- Begin Message ---Source: linux Version: 7.1.12-1 Severity: wishlist X-Debbugs-Cc: [email protected] User: [email protected] Usertags: amd64 Dear Maintainer, The built-in camera of the MSI Prestige 13 AI+ Evo A2VMG (Lunar Lake; Intel IPU7 + OmniVision OV5675 sensor behind a TI TPS68470 PMIC) does not work with the Debian amd64 kernel, because four Kconfig symbols the platform needs are not enabled on amd64: # CONFIG_VIDEO_OV5675 is not set # CONFIG_GPIO_TPS68470 is not set # CONFIG_REGULATOR_TPS68470 is not set # CONFIG_COMMON_CLK_TPS68470 is not set CONFIG_VIDEO_OV5675 is already enabled on arm64 (debian/config/arm64/config:1146: CONFIG_VIDEO_OV5675=m), so for that symbol this is only an amd64 gap - and amd64 is where the sensor is actually used, by Intel IPU6/IPU7 laptops. All four are tristate and their dependencies are already met by the existing amd64 config (CONFIG_INTEL_SKL_INT3472=m, CONFIG_I2C=y), so they can be enabled as modules with no further changes: drivers/gpio/Kconfig config GPIO_TPS68470 depends on INTEL_SKL_INT3472 drivers/regulator/Kconfig config REGULATOR_TPS68470 depends on INTEL_SKL_INT3472 || COMPILE_TEST drivers/clk/Kconfig config COMMON_CLK_TPS68470 depends on I2C depends on INTEL_SKL_INT3472 || COMPILE_TEST This is the same class of request as #1109814 (CONFIG_VIDEO_OV02C10) and #1087004 (CONFIG_VIDEO_OV01A10), both of which were accepted. Symptoms with the stock 7.1.8-2 amd64 kernel -------------------------------------------- intel-ipu7 0000:00:05.0: Found supported sensor OVTI5675:00 intel-ipu7 0000:00:05.0: Connected 1 cameras int3472-tps68470 i2c-INT3472:06: TPS68470 REVID: 0x21 int3472-tps68470 i2c-INT3472:06: error -ENODEV: No board-data found for this model The PMIC is found but cannot be set up, its MFD cells (tps68470-clk, tps68470-regulator, tps68470-gpio) have no drivers to bind to, so acpi_gpiochip_add() never runs and never clears the _DEP that OVTI5675:00 has on INT3472:06. As a result no i2c client is created for the sensor and no /dev/v4l-subdev* appears; userspace sees only the bare IPU7 ISYS capture nodes and no camera. Verification ------------ I rebuilt the following out of tree against linux-headers-7.1.8+deb14.1-amd64, from unmodified v7.1.8 sources (no patch in debian/patches/ touches these files): drivers/media/i2c/ov5675.c -> ov5675.ko drivers/gpio/gpio-tps68470.c -> gpio-tps68470.ko drivers/regulator/tps68470-regulator.c -> tps68470-regulator.ko drivers/clk/clk-tps68470.c -> clk-tps68470.ko plus a backport of the upstream DMI entry for this machine (see below). With those five modules in /lib/modules/$(uname -r)/updates/, the camera works after a normal boot: intel-ipu7 0000:00:05.0: CSE authenticate_run done int3472-tps68470 i2c-INT3472:06: TPS68470 REVID: 0x21 intel_ipu7_isys.isys intel_ipu7.isys.40: bind ov5675 1-0036 nlanes is 2 port is 0 intel_ipu7_isys.isys intel_ipu7.isys.40: All sensor registration completed. $ cam -l Available cameras: 1: Internal front camera (\_SB_.LNK0) $ cam -c1 --capture=300 --file=/tmp/cap#.bin ... 2584x1944 at 29.95 fps, 300 frames, real images PipeWire exposes it as node "libcamera_input.__SB_.LNK0" / "Built-in Front Camera". libcamera 0.7 supports IPU7 through the Simple pipeline handler with the (GPU-accelerated) SoftISP, so no out-of-tree userspace such as the Intel camera HAL is needed - the kernel config is the only thing missing. Secondary note: INT3472 board data for this model ------------------------------------------------- Independently of the config question, 7.1 only carries the DMI entry for the "Prestige 14 AI+ Evo C2VMG". The entries for the other two models were added upstream after 7.1 was cut: ae04f36 platform/x86: int3472: Match MSI laptop board name 621e4f7 platform/x86: int3472: Add more MSI AI evo laptops They add "Prestige 13 AI+ Evo A2VMG" (MS-13Q3) and "Prestige 16 AI+ Evo B2VMG" (MS-15A3), both reusing the existing msi_prestige_ai_evo_* data, and they are in 7.2. This resolves itself once forky moves to 7.2. I mention it only because enabling the four config symbols alone is not sufficient while the archive is on 7.1; a cherry-pick of those two commits would make the camera work on 7.1 as well, if the team considers that worthwhile. System information ------------------ Vendor: Micro-Star International Co., Ltd. Product: Prestige 13 AI+ Evo A2VMG Board: MS-13Q3 BIOS: E13Q3IMS.112, 12/22/2025 Kernel: 7.1.8+deb14.1-amd64 (linux-image-amd64 7.1.8-2) Camera: OVTI5675:00 (OV5675) on Intel IPU7, PMIC INT3472:06 (TPS68470) -- This investigation was carried out with the assistance of Claude Code. Every statement above was verified on the machine in question. -- System Information: Debian Release: forky/sid APT prefers stable-updates APT policy: (500, 'stable-updates'), (500, 'stable-security'), (500, 'testing'), (500, 'stable') Architecture: amd64 (x86_64) Kernel: Linux 7.1.8+deb14.1-amd64 (SMP w/8 CPU threads; PREEMPT) Kernel taint flags: TAINT_CRAP, TAINT_OOT_MODULE, TAINT_UNSIGNED_MODULE Locale: LANG=C.UTF-8, LC_CTYPE=ja_JP.utf8 (charmap=UTF-8), LANGUAGE not set Shell: /bin/sh linked to /usr/bin/dash Init: systemd (via /run/systemd/system) LSM: AppArmor: enabled
--- End Message ---
--- Begin Message ---On Tue, Sep 01, 2026 at 09:15:26PM +0200, Bastian Blank wrote: > On Wed, Sep 02, 2026 at 02:22:42AM +0900, Tatsuki Sugiura wrote: > > This investigation was carried out with the assistance of Claude Code. Every > > statement above was verified on the machine in question. > > We don't accept bug reports from slop machines. None of the information > is known to be correct. Actually closing. Bastian -- Schshschshchsch. -- The Gorn, "Arena", stardate 3046.2
--- End Message ---
