[v3,0/4] Upgrade kmod 33 -> 34.1

Bring scdoc in from meta-wayland to support generation of the manpages. Convert
to meson from autotools. Clean up usage of kmod in eudev.

Signed-off-by: Alex Kiernan <alex.kiernan@gmail.com>
---
Changes in v3:
- Fix maintainer for scdoc
- Stop using autotools in scdoc (it just has a Makefile)
- Set upstream commit correctly for scdoc 1.11.3
- Fix scdcoc metadata
- Link to v2: https://lore.kernel.org/r/20250313-upgrade-kmod-v2-0-2103ca9a26ad@gmail.com

---
Alex Kiernan (4):
      scdoc: Import from meta-wayland
      scdoc: Upgrade 1.11.2 -> 1.11.3
      kmod: Upgrade 33 -> 34.1
      eudev: Drop --output=... from kmod call in init script

 meta/conf/distro/include/maintainers.inc           |  1 +
 meta/recipes-core/udev/eudev/init                  |  3 +-
 .../scdoc/scdoc/0001-Makefile-drop-static.patch    | 23 +++++++++
 meta/recipes-extended/scdoc/scdoc_1.11.3.bb        | 22 ++++++++
 .../kmod/kmod/avoid_parallel_tests.patch           | 30 -----------
 meta/recipes-kernel/kmod/kmod/gtkdocdir.patch      | 30 -----------
 meta/recipes-kernel/kmod/kmod/run-ptest            |  5 --
 .../kmod/{kmod_33.bb => kmod_34.1.bb}              | 58 +++++++++-------------
 8 files changed, 70 insertions(+), 102 deletions(-)
---
base-commit: cb9bff9eac3670253076d5d09c6893e372be40a8
change-id: 20250313-kmod-ac33fdaf6452

Best regards,

On Thu Mar 13, 2025 at 3:13 PM CET, Alex Kiernan via lists.openembedded.org wrote:
> Bring scdoc in from meta-wayland to support generation of the manpages. Convert
> to meson from autotools. Clean up usage of kmod in eudev.
>
> Signed-off-by: Alex Kiernan <alex.kiernan@gmail.com>
> ---
> Changes in v3:
> - Fix maintainer for scdoc
> - Stop using autotools in scdoc (it just has a Makefile)
> - Set upstream commit correctly for scdoc 1.11.3
> - Fix scdcoc metadata
> - Link to v2: https://lore.kernel.org/r/20250313-upgrade-kmod-v2-0-2103ca9a26ad@gmail.com
>
> ---

Hi,

Thanks for the new version, but it appears to be failing on the
selftests:

AssertionError: Regex didn't match: 'Line \\d+ of \\"/usr/src/debug/kmod/.*/tools/kmod.c\\" starts at address 0x[0-9A-Fa-f]+ <kmod_help>' not found in 'Reading symbols from /tmp/debugfs-xvl6_emr/bin/kmod...\nReading symbols from /tmp/debugfs-xvl6_emr/bin/.debug/kmod...\n0x00007ffff7fe3ac0 in _start () from /tmp/debugfs-xvl6_emr/lib/ld-linux-x86-64.so.2\nLine 118 of "/usr/include/bits/stdio2.h" starts at address 0x55555555edd0 <kmod_help> and ends at 0x55555555edd7 <kmod_help+7>.'

https://autobuilder.yoctoproject.org/valkyrie/api/v2/logs/1759546/raw_inline

Can you have a look at this ?

On Thu, Mar 13, 2025 at 8:04 PM Mathieu Dubois-Briand
<mathieu.dubois-briand@bootlin.com> wrote:
>
> On Thu Mar 13, 2025 at 3:13 PM CET, Alex Kiernan via lists.openembedded.org wrote:
> > Bring scdoc in from meta-wayland to support generation of the manpages. Convert
> > to meson from autotools. Clean up usage of kmod in eudev.
> >
> > Signed-off-by: Alex Kiernan <alex.kiernan@gmail.com>
> > ---
> > Changes in v3:
> > - Fix maintainer for scdoc
> > - Stop using autotools in scdoc (it just has a Makefile)
> > - Set upstream commit correctly for scdoc 1.11.3
> > - Fix scdcoc metadata
> > - Link to v2: https://lore.kernel.org/r/20250313-upgrade-kmod-v2-0-2103ca9a26ad@gmail.com
> >
> > ---
>
> Hi,
>
> Thanks for the new version, but it appears to be failing on the
> selftests:
>
> AssertionError: Regex didn't match: 'Line \\d+ of \\"/usr/src/debug/kmod/.*/tools/kmod.c\\" starts at address 0x[0-9A-Fa-f]+ <kmod_help>' not found in 'Reading symbols from /tmp/debugfs-xvl6_emr/bin/kmod...\nReading symbols from /tmp/debugfs-xvl6_emr/bin/.debug/kmod...\n0x00007ffff7fe3ac0 in _start () from /tmp/debugfs-xvl6_emr/lib/ld-linux-x86-64.so.2\nLine 118 of "/usr/include/bits/stdio2.h" starts at address 0x55555555edd0 <kmod_help> and ends at 0x55555555edd7 <kmod_help+7>.'
>
> https://autobuilder.yoctoproject.org/valkyrie/api/v2/logs/1759546/raw_inline
>

I'm not really sure what the right fix is here... the output from the
gdb command that's being called is now:

# gdb /bin/kmod -ex "info line kmod_help"
GNU gdb (GDB) 16.2
Copyright (C) 2024 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.
Type "show copying" and "show warranty" for details.
This GDB was configured as "aarch64-poky-linux".
Type "show configuration" for configuration details.
For bug reporting instructions, please see:
<https://www.gnu.org/software/gdb/bugs/>.
Find the GDB manual and other documentation resources online at:
    <http://www.gnu.org/software/gdb/documentation/>.

For help, type "help".
Type "apropos word" to search for commands related to "word"...
Reading symbols from /bin/kmod...
Reading symbols from /bin/.debug/kmod...
warning: could not convert 'kmod_help' from the host encoding
(ANSI_X3.4-1968) to UTF-32.
This normally should not happen, please file a bug report.
Line 118 of "/usr/include/bits/stdio2.h" starts at address 0xaac0
<kmod_help> and ends at 0xaacc <kmod_help+12>.

I suspect this is really collateral damage as the test is just using
kmod as a handy target when @Ross Burton rewrote it - does that look
right? And would changing the regex to:

r"Line \d+ of \".*\" starts at address 0x[0-9A-Fa-f]+ <kmod_help>"

be reasonable, since I'm guessing that the debug info is just confused
here, but that's not relevant to the thing which is actually being
tested?