Re: [PATCH v7 0/3] Extended MODVERSIONS Support

[Sami Tolvanen <samitolvanen@google.com>]

On Wed, Oct 23, 2024 at 02:31:27AM +0000, Matthew Maurer wrote:
> This patch series is intended for use alongside the Implement DWARF
> modversions series [1] to enable RUST and MODVERSIONS at the same
> time.
> 
> Elsewhere, we've seen a desire for long symbol name support for LTO
> symbol names [2], and the previous series came up [3] as a possible
> solution rather than hashing, which some have objected [4] to.
> 
> This series adds a MODVERSIONS format which uses a section per column.
> This avoids userspace tools breaking if we need to make a similar change
> to the format in the future - we would do so by adding a new section,
> rather than editing the struct definition. In the new format, the name
> section is formatted as a concatenated sequence of NUL-terminated
> strings, which allows for arbitrary length names.
> 
> Emitting the extended format is guarded by CONFIG_EXTENDED_MODVERSIONS,
> but the kernel always knows how to validate both the original and
> extended formats.
> 
> Selecting RUST and MODVERSIONS is now possible if GENDWARFKSYMS is
> selected, and will implicitly select EXTENDED_MODVERSIONS.
> 
> This series depends upon the module verification refactor patches [5]
> that were split off of v5, and DWARF-based versions [1].
> 
> linuxppc-dev is requested to look at the ppc-specific munging,
> as Luis would like some eyes on there [6].
> 
> [1] https://lore.kernel.org/lkml/20241008183823.36676-21-samitolvanen@google.com/
> [2] https://lore.kernel.org/lkml/20240605032120.3179157-1-song@kernel.org/
> [3] https://lore.kernel.org/lkml/ZoxbEEsK40ASi1cY@bombadil.infradead.org/
> [4] https://lore.kernel.org/lkml/0b2697fd-7ab4-469f-83a6-ec9ebc701ba0@suse.com/
> [5] https://lore.kernel.org/linux-modules/20241015231651.3851138-1-mmaurer@google.com/T/#t
> [6] https://lore.kernel.org/lkml/ZxahDv5ZKdM__0sZ@bombadil.infradead.org/
> 
> Changes in v7:
> - Fix modpost to detect EXTENDED_MODVERSIONS based on a flag
> - Drop patches to fix export_report.pl

Unless Luis prefers to deal with this separately, perhaps you can
just include a patch to delete export_report.pl in this series?

Sami