On 26/10/2026 07:13, Gianluca Cannata wrote:
Il giorno lun 26 ott 2026 alle ore 07:44 Michael Kelly
<[email protected]> ha scritto:
Are you working on this ? I can look at doing this if you are not.
Let me know how you want to advance the discussion further: via email,
gmeet, or such.
Cheers,
Gianluca
PS: I also am interested in your work about UEFI so we I think we
should coordinate somehow.
I think that I have already covered most of what is needed to get
hurd-amd64 booting on UEFI. This is the case on my laptop anyway which
now reaches the rumpkernel boot and halts there due to the lack of Nvme
capability. I'm expecting that with that support then it will reach a
console login. I've an ACPI problem which I've bypassed by not loading
the ACPI tables but I'm not expecting that to be unsolvable. It's mostly
a case of transferring prototype code into something acceptable.
I'm mostly guessing here at the moment, and assuming it can be done at
all, but I think there are 2 parts to the Nvme:
1) Supporting Nvme in rumpkernel, perhaps by configuring a new build
component, libpci_nvme or similar, something like the addition of
libpci_piixide via debian/patches.
2) Supporting "nvme" devices within Hurd/rumpdisk.
I'll begin to look at 1) today although the NetBSD build system is
completely unknown to me.
All the best,
Mike.
