On Sun, Feb 09, 2025 at 07:19:33AM -0800, Guenter Roeck wrote:
> On 2/6/25 08:06, Greg Kroah-Hartman wrote:
> > This is the start of the stable review cycle for the 6.6.76 release.
> > There are 389 patches in this series, all will be posted as a response
> > to this one. If anyone has any issues with these being applied, please
> > let me know.
> >
> > Responses should be made by Sat, 08 Feb 2025 15:51:12 +0000.
> > Anything received after that time might be too late.
> >
> [ ... ]
>
> > Hongbo Li <lihongbo22@xxxxxxxxxx>
> > hostfs: fix the host directory parse when mounting.
>
> This patch results in:
>
> Building um:defconfig ... failed
> --------------
> Error log:
> fs/hostfs/hostfs_kern.c:972:9: error: implicit declaration of function 'fsparam_string_empty'; did you mean 'fsparam_string'? [-Werror=implicit-function-declaration]
> 972 | fsparam_string_empty("hostfs", Opt_hostfs),
>
> because fsparam_string_empty() is not declared globally in v6.6.y.
>
> The patch declaring it is 7b30851a70645 ("fs_parser: move fsparam_string_empty()
> helper into header"). Applying that patch on top of 6.6.76 fixes the problem.
>
> The problem only affects "um" builds since hostfs (CONFIG_HOSTFS) is only available
> and used there. Oddly enough, the patch breaks the build of this file instead of
> fixing the problem it claims to fix, and it looks like no one noticed.
> On top of that, "hostfs: convert hostfs to use the new mount API" was obviously
> not tested. It looks like a substantial change which would definitely warrant
> testing when backported.
>
> That makes me wonder: Should I stop build testing "um" images in older kernels ?
No, it's good for testing and I'm pretty sure that Android still uses it
for their test infrastructure so it matters. I'll go do some reverts
now and push out a new release with this fixed as it's now shown up on
the kernel.ci build reports as well.
thanks for pointing it out.
greg k-h
