Sorry, I've edited the title and description because the framing in my original report was off and I don't want to send anyone down the wrong path. Trying again.
Camera also fails on 7.0.0-14 for me, same way it fails on 7.0.0-15. Same "VIDIOC_STREAMON: Link has been severed" from pipewire, same "int3472-discrete INT3472:05/:06 cannot find GPIO chip INTC1096:00, deferring" floods in dmesg, same "ov2740 ... supply AVDD/DOVDD/DVDD not found, using dummy regulator". So it's not a -14 -> -15 change in the kernel package. Looking at my apt history I also realised the stock 26.04 install on 2026-04-23 already shipped with 7.0.0-14.14, and that's still the .14 deb on disk. linux-firmware hasn't moved either, nor have libcamera, pipewire, or wireplumber. The camera worked on first boot and doesn't now, and I can't find anything I upgraded in between. BIOS is also unchanged, fwupd history is empty. So it's not obviously a package regression. I'm a bit out of my depth past that, but while I was capturing the .14 diagnostics I noticed something I don't really know how to read: In dmesg, the "INT3472:05/:06 cannot find GPIO chip INTC1096:00, deferring" messages happen 32 times in the first second of boot, and then they just stop. There aren't any more INT3472 messages in the rest of the boot log. But on the running system, /sys/class/gpio/gpiochip872 exists and its label is "INTC1096:00", backed by usb_ljca.ljca-gpio.0. So the chip int3472 was looking for does come up - it just comes up after int3472 has stopped retrying. Both INT3472:05 and INT3472:06 have physical_node symlinks (they got instantiated as platform devices) but their driver fields are "(unbound)". I have no idea whether that's the actual cause or just a coincidence I happened to notice - I don't really understand the deferred-probe path well enough to say. Flagging it in case it's a useful signal. If it's nothing, please ignore. Attaching: - diagnostics-7.0.0-14.txt, same shape as the diagnostics.txt I attached originally so it can be diffed against it. And I'll run apport-collect against linux-image-7.0.0-14-generic so the standard apport bundle is there for .14 too. Sorry again for the back-and-forth. I'm dropping the regression-update tag since the "after an apt update" framing isn't right. Happy to grab anything else that would help - I just don't want to keep guessing without direction. ** Attachment added: "diagnostics-7.0.0-14.txt" https://bugs.launchpad.net/ubuntu/+source/linux/+bug/2153096/+attachment/5971434/+files/diagnostics-7.0.0-14.txt -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/2153096 Title: IPU6 webcam (OV2740) on X1 Nano fails on kernel 7.0.0-14/-15: int3472 cannot find GPIO chip INTC1096:00, VSC chain doesn't come up To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/2153096/+subscriptions -- ubuntu-bugs mailing list [email protected] https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
