[9fans] Nix/9legacy: compiles without much ado

[9fans] Nix/9legacy: compiles without much ado

[tlaronde]

After fighting with various installations problems, I finally gave Nix
some time.

It happened that compiling Nix under 9legacy was simple enough: the
only thing I had to do was to substitute Ureg.ip to Ureg.pc

The:

https://github.com/tlaronde/nix

is up-to-date and one can download the sources (and install) via:

term% cd some_dir
term% hget http://downloads.kergis.com/nix/nix-9legacy.tgz | tar xz
term% nixdir=some_dir/nix
term% $nixdir/sys/src/nix/nix # setup the namespace
term% cd $nixdir/sys/src/nix/boot
term% objtype=amd64
term% mk
term% cd ../k10
term% mk ../root/nvram
term% mk

The multiboot code is the original one so, since Paul had to adjust
the code in the 9front main variant because the spec has moved, using
the kernel with Grub or some code expecting present multiboot spec
will probably need adjustments.

For my part, since I have a candidate hardware (cheap AM08PRO with 8
hardware AMD64 Ryzen 9 cores), I will go back to the RTL 8125 driver
that is a sine qua none to use it---and I set to myself the limit date of
Monday March the 7th to have it doing its stuff.
-- 
        Thierry Laronde <tlaronde +AT+ kergis +dot+ com>
                     http://www.kergis.com/
                    http://kertex.kergis.com/
Key fingerprint = 0FF7 E906 FBAF FE95 FD89  250D 52B1 AE95 6006 F40C

------------------------------------------
9fans: 9fans
Permalink: https://9fans.topicbox.com/groups/9fans/Tb9ac567b4bcfa86e-M285f8f9a40a6a37c9a0cecbb
Delivery options: https://9fans.topicbox.com/groups/9fans/subscription

Re: [9fans] Nix/9legacy: compiles without much ado

[tlaronde]

On Tue, Mar 25, 2025 at 05:34:42PM +0100, tlaronde@kergis.com wrote:
> After fighting with various installations problems, I finally gave Nix
> some time.
> 
> It happened that compiling Nix under 9legacy was simple enough: the
> only thing I had to do was to substitute Ureg.ip to Ureg.pc
> 
> The:
> 
> https://github.com/tlaronde/nix
> 
> is up-to-date and one can download the sources (and install) via:
> 
> term% cd some_dir
> term% hget http://downloads.kergis.com/nix/nix-9legacy.tgz | tar xz
> term% nixdir=some_dir/nix
> term% $nixdir/sys/src/nix/nix # setup the namespace
> term% cd $nixdir/sys/src/nix/boot
> term% objtype=amd64
> term% mk
> term% cd ../k10
> term% mk ../root/nvram
> term% mk
> 
> The multiboot code is the original one so, since Paul had to adjust
> the code in the 9front main variant because the spec has moved, using
> the kernel with Grub or some code expecting present multiboot spec
> will probably need adjustments.
> 
> For my part, since I have a candidate hardware (cheap AM08PRO with 8
> hardware AMD64 Ryzen 9 cores), I will go back to the RTL 8125 driver
> that is a sine qua none to use it---and I set to myself the limit date of
> Monday March the 7th to have it doing its stuff.
         ^^^^^

April, and of this year 2025 :-^
> --
> Thierry Laronde <tlaronde +AT+ kergis +dot+ com>
>              http://www.kergis.com/
>             http://kertex.kergis.com/
> Key fingerprint = 0FF7 E906 FBAF FE95 FD89  250D 52B1 AE95 6006 F40C

-- 
        Thierry Laronde <tlaronde +AT+ kergis +dot+ com>
                     http://www.kergis.com/
                    http://kertex.kergis.com/
Key fingerprint = 0FF7 E906 FBAF FE95 FD89  250D 52B1 AE95 6006 F40C

------------------------------------------
9fans: 9fans
Permalink: https://9fans.topicbox.com/groups/9fans/Tb9ac567b4bcfa86e-M47e3fad5e369eb873fce2124
Delivery options: https://9fans.topicbox.com/groups/9fans/subscription

Re: [9fans] Nix/9legacy: compiles without much ado

[cinap_lenrek]

I'v taken a look at the linked repository's code:

https://github.com/tlaronde/nix

and wonder where that "new version" Mbi (multiboot info) struct
comes from, as it clearly doesnt comply with this:

https://www.gnu.org/software/grub/manual/multiboot/multiboot.html

even tho, the multiboot header uses the same 0x1BADB002 magic
and the flags field just has bit 0 and 1 set which just negotiate
4K module alignment and memory map passing.

It is also not multiboot2, which uses different magic and
different tag structure.

https://gnu.org/software/grub/manual/multiboot2/multiboot.html

So what is going on here? Is it not possible to negotiate what
struct we get passed? Any pointers?

--
cinap

------------------------------------------
9fans: 9fans
Permalink: https://9fans.topicbox.com/groups/9fans/Tb9ac567b4bcfa86e-Md489ec066454c08f1e847406
Delivery options: https://9fans.topicbox.com/groups/9fans/subscription

Re: [9fans] Nix/9legacy: compiles without much ado

[cinap_lenrek]

waiiit... i miscounted the syms[4] field. my mistake.

so it *IS* 0.6.96 spec.

what old version then are you talking about?

--
cinap

------------------------------------------
9fans: 9fans
Permalink: https://9fans.topicbox.com/groups/9fans/Tb9ac567b4bcfa86e-M1b5656af9bc7dffcb3ff73fe
Delivery options: https://9fans.topicbox.com/groups/9fans/subscription

Re: [9fans] Nix/9legacy: compiles without much ado

[tlaronde]

On Wed, Mar 26, 2025 at 09:55:03PM +0100, cinap_lenrek@felloff.net wrote:
> waiiit... i miscounted the syms[4] field. my mistake.
> 
> so it *IS* 0.6.96 spec.
> 
> what old version then are you talking about?

That the multiboot code was related to an older version to be adjusted
was infered by me because of modifications that Paul A. Lalonde had to
make to the code (k10/l32p.s), and I took "Fix multiboot magic
number passing" and "Fixed the offset of the computed jump around the
multiboot header" for an indication that the multiboot header has
changed (in size and in magic numbers).

The modifications (9front flavor) made were in k10/l32p.s and not in
k10/multiboot.c (except for an added orthogonal function to verify that
there is enough memory).

But it may be that this is related not to the multiboot spec per se
but to implementation details using 9front stub (or 9front vmx to
test nix early).

I'm slowly worming my path in the code, so I may probably make errors.
I'm still to understand how the pieces fit and what is grabbed from
outside the Nix code---in the 9legacy flavor, at the moment, not much
except global headers and compilers; this will be part of the work,
cautiously, to reduce the code to the strict modified or added,
reaching out in 9k/ and in the rest of the code for devices (the
policy about devices regarding the cores---masking, tying, delegating,
sharing...---and the IOMMU will have to be considered at some time).

I'm already surprised that my reverting and extracting of the code
works almost at first. Beginner's luck.
-- 
        Thierry Laronde <tlaronde +AT+ kergis +dot+ com>
                     http://www.kergis.com/
                    http://kertex.kergis.com/
Key fingerprint = 0FF7 E906 FBAF FE95 FD89  250D 52B1 AE95 6006 F40C

------------------------------------------
9fans: 9fans
Permalink: https://9fans.topicbox.com/groups/9fans/Tb9ac567b4bcfa86e-M841999ce9ca0bbbd123ac337
Delivery options: https://9fans.topicbox.com/groups/9fans/subscription