re: Fixed modular kernel path and different kernels

[Paul Goyette <paul%whooppee.com@localhost>]

On Fri, 20 Jan 2017, matthew green wrote:

> > The 'tarball' variant has other problems, such as "how does the kernel
> > extract a module from the tarball to facilitate auto-load?"  Last time
> > I looked, there was no kernel version of tar or pax.  :)  IMHO, Loss of
> > the autoload capability would severely reduce the useability of modules.
>
> there is "ustarfs" support in the kernel/libsa, that wouldn't
> really be a big issue if this isn't directly usable, it would
> be simple to fix or port some extractor code.

Ah, yes, thanks for reminding me about ustarfs.

Would ustarfs allow for in-place replacement of individual archive
members?  Or would replacement of one member require extracting the
entire archive, replacing a member file, and then rebuilding the
archive?  (I guess one could just append a new member to the archive,
but wouldn't that mean that a modload operation would have to search
the entire archive for the _last_ occurrence of the member you want
to extract?  Feels like a performance concern...)

FWIW, perhaps we should refer to this as the 'archive' variant rather
than 'tarball' variant?   :)


+------------------+--------------------------+------------------------+
| Paul Goyette     | PGP Key fingerprint:     | E-mail addresses:      |
| (Retired)        | FA29 0E3B 35AF E8AE 6651 | paul at whooppee.com   |
| Kernel Developer | 0786 F758 55DE 53BA 7731 | pgoyette at netbsd.org |
+------------------+--------------------------+------------------------+