Re: CVS commit: src/sys/kern

[Michael van Elst <mlelstv%serpens.de@localhost>]

On Mon, Sep 16, 2019 at 03:39:35AM +0200, Emmanuel Dreyfus wrote:
> Thor Lancelot Simon <tls%panix.com@localhost> wrote:
> 
> > I've never been a fan of the wedge: syntax (as was noted at the time, it
> > conflicts with the syntax for specifying network filesystem servers,
> > which actually broke automation I had in place); I would not mind seeing
> > it go.
> 
> IMO, both wedge: and NAME= are ugly, but beside your point I would be in
> favor of NAME= because that is what I used in x86 bootloaders, and
> lazyness commands not to redo it. :-)

A naming scheme like PROTOCOL ":" SPECIFIER isn't that bad. I chose
the NAME= prefix instead for several reasons. Compatibility with
NFS mount paths (that we get told are deprecated whenever we use
them), lookalike with other OSes, and also to not further reduce
path lengths as the string sometimes needs to be stored in small
buffers.


> When you talk about letting it go, you means removing it, or just
> keeping it as is undocumented? Removing could break some setups badly.

We need to keep it for compatibility for some time, and that's how
it should be documented.


> About documentation, wedge: is present in
> share/man/man8/man8.x86/boot.8
> share/man/man9/cpu_rootconf.9
> sys/arch/evbarm/conf/IYONIX
> sys/arch/evbarm/fdt/fdt_machdep.c
> sys/arch/iyonix/conf/GENERIC
> usr.bin/config/config.5

sys/arch/evbarm/fdt/fdt_machdep.c is a different beast. The boot
device selection in it should go completely as is.

There could be a MI version that just allows to feed a bootspec to
the kernel for systems that support FDT but cannot specify boot_args.

We could also adopt its selection of network interfaces per MAC
address for MI code. Just need to come up with a proper encoding
as bootspec.



Greetings,
-- 
                                Michael van Elst
Internet: mlelstv%serpens.de@localhost
                                "A potential Snark may lurk in every tree."