[LTP] [PATCH v5 3/7] lib: Add support for max_kver to struct tst_test and tst_fs

[Petr Vorel]

Hi Cyril,

> Hi!
> > diff --git a/include/tst_test.h b/include/tst_test.h
> > index c69362485e..30c32b844e 100644
> > --- a/include/tst_test.h
> > +++ b/include/tst_test.h
> > @@ -280,6 +280,9 @@ struct tst_ulimit_val {
> >   *
> >   * @min_kver: A minimum kernel version supporting the filesystem which has been
> >   *            created with mkfs.
> > + *
> > + * @max_kver: A maximum kernel version supporting the filesystem which has been
> > + *            created with mkfs.
> >   */
> >  struct tst_fs {
> >  	const char *type;
> > @@ -292,6 +295,7 @@ struct tst_fs {
> >  	const void *mnt_data;

> >  	const char *min_kver;
> > +	const char *max_kver;

> Also I would consider not adding this, it may feel like we are
> completing something, but in reality we are adding interface that is
> very unlikely to be ever used.

You're probably right. When I started this there were 3 tests and I thought there
would be more. I also thought this would be useful info in the metadata.  But
in later versions it turn out 2 of them are to be removed. Now I double checked
"tst_kvercmp >" code and it looks like none of them applies to .max_kver usage.
I agree that a single test is not enough as a reason to have it in the metadata.
Maybe there will be more in the future, but really just a few.

=> I should probably send yet another version where I implement the skip in the
setup. Could you please comment these 2 commits (I would not resend them)?

* [v5,4/7] creat07: Avoid timeout on the test failure
https://patchwork.kernel.org/project/ltp/patch/20260810160048.1040517-5-pvorel@suse.cz/

* [v5,5/7] creat07: execve04: Remove version check, add linux-git
https://patchwork.kernel.org/project/ltp/patch/20260810160048.1040517-6-pvorel@suse.cz/

I also find things related to min_kver (will be a separate effort unrelated to
this):
* ioctl10.c: tst_kvercmp detection for old kernels might be solved by
replacing to kconfig check => need to check.
* statx04.c is skipping on btrfs based on tst_kvercmp(). This could define
.filesystems instead of .skip_filesystems (btrfs with .min_kver).
* unlink09.c as well (tmpfs with .min_kver)

Kind regards,
Petr