[Bug 2076263] Re: Failed to build against linux-6.11

[You-Sheng Yang]

> About the packaging changes, why is this being done?
> The d/changelog reference to it has no bug number or any further explanation:
> Also, d/control still has a build-depends on dkms and dh-sequence-dkms.

Per https://bugs.launchpad.net/ubuntu/+source/backport-iwlwifi-
dkms/+bug/2046871/comments/9, a review comment for backport-iwlwifi-dkms
updates in bug 2046871, and an upstream review comment
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1027889, it gives:

> please switch the Build-Depends of your package from `dkms` to `dh-dkms`
> or (preferrably) `dh-sequence-dkms`.
> With the latter you can also drop the `--with dkms` argument to `dh`.

That's exactly what I do in the later updates to dkms packages. We have
already Build-Depends on dh-sequence-dkms, and then the `--with dkms`
argument to `dh` should be removed.

** Bug watch added: Debian Bug tracker #1027889
   https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1027889

-- 
You received this bug notification because you are a member of Ubuntu
Sponsors, which is subscribed to the bug report.
https://bugs.launchpad.net/bugs/2076263

Title:
  Failed to build against linux-6.11

Status in usbio-drivers package in Ubuntu:
  Fix Released
Status in usbio-drivers source package in Noble:
  Fix Committed
Status in usbio-drivers source package in Oracular:
  Fix Released

Bug description:
  [SRU Justification]

  [Impact]

  Upstream commit 0edb555a65d1e ("platform: Make
  platform_driver::remove() return void") fails the build against
  linux-6.11.

  [Fix]

  An upstream usbio release WW3224-USBIO-PV-HFIX1 carries the fix.

  [Test Case]

  a) the dkms module should build, *and* the module should be able to be loaded (include modprobe test)
  b) the package should ALSO build fine in the normal GA kernel, and the module should also load there. In other words, add a test step for the normal GA kernel from noble.

  [Where problems could occur]

  It appears the i2c protocol meant to be governed by this usbio is not
  in well control by the upstream collaborative peers, such that we fell
  into various problems due the the mismatch of firmware version, and
  therefore some ugly patch from Intel was added to work-around that.
  While we don't have too much details about the design and the
  solution, all we can do is to deploy extensive tests as possible.

  [Other Info]

  The SRU collects fixes for a few bugs:
  * bug 2076263: meant for series with kernel >= 6.11, therefore Noble and Oracular,
  * bug 2060101: meant for hwe-6.8 and on, therefore Noble and Oracular,
  * bug 2071821: meant for IPU7 platforms, therefore Oracular only (which will be compiled in linux/oracular and linux-oem-6.11/noble)

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

  /var/lib/dkms/usbio-drivers/0~git202312141918.78ffb706-0ubuntu2/build/drivers/gpio/gpio-usbio.c:486:19: error: initialization of 'void (*)(struct platform_device *)' from incompatible pointer type 'int (*)(struct platform_device *)' [-Wincompatible-pointer-types]
    486 |         .remove = usbio_gpio_remove,
        |                   ^~~~~~~~~~~~~~~~~
  /var/lib/dkms/usbio-drivers/0~git202312141918.78ffb706-0ubuntu2/build/drivers/i2c/busses/i2c-usbio.c:415:19: error: initialization of 'void (*)(struct platform_device *)' from incompatible pointer type 'int (*)(struct platform_device *)' [-Wincompatible-pointer-types]
    415 |         .remove = usbio_i2c_remove,
        |                   ^~~~~~~~~~~~~~~~

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/usbio-drivers/+bug/2076263/+subscriptions