Qualcomm CPU / Fedora: AI-maker board project >> need support (paid)!

[Luigi Grax]

my name is Luigi, I am developing TB5 SBC, a new AI-maker board based on Qualcomm CPU

I am looking for a partner a/o paid-consultant in charge of integrating Fedora on the
board

This project is open source and is .... SUPER-cooooool !!   (3\)

contact me on: https://www.linkedin.com/in/luigigrasso/

[Petr Pisar]

V Tue, Mar 22, 2022 at 09:37:14AM -0000, Luigi Grax napsal(a):
>  my name is Luigi, I am developing TB5 SBC, a new AI-maker board based
> on
>  Qualcomm CPU
>   
So it's one of the Aarch64 processors with proprietary peripherals?

>  I am looking for a partner a/o paid-consultant in charge of
> integrating
>  Fedora on the board
>   
Fedora only uses original Linux kernel. No third-party kernel drivers. If you
want Fedora on the board, then you first need to get all drivers into
<https://www.kernel.org/>;.

-- Petr

[Demi Marie Obenour]

On 3/22/22 06:04, Petr Pisar wrote:
>  V Tue, Mar 22, 2022 at 09:37:14AM -0000, Luigi Grax napsal(a):
> > my name is Luigi, I am developing TB5 SBC, a new AI-maker board based on
> > Qualcomm CPU
> >
>  So it's one of the Aarch64 processors with proprietary peripherals?
>  
> > I am looking for a partner a/o paid-consultant in charge of integrating
> > Fedora on the board
> >
>  Fedora only uses original Linux kernel. No third-party kernel drivers. If you
>  want Fedora on the board, then you first need to get all drivers into
>  <https://www.kernel.org/>;. 
All kernel-mode drivers, to be specific.  User-mode drivers are an
underutilized alternative for systems that have an IOMMU/SMMU.  Obviously,
the drivers still need to be free software for Fedora to package them, but
user-mode drivers have the advantage of not running with kernel privilege.
-- 
Sincerely,
Demi Marie Obenour (she/her/hers)

[Petr Pisar]

V Tue, Mar 22, 2022 at 07:30:13AM -0400, Demi Marie Obenour napsal(a):
>  All kernel-mode drivers, to be specific.  User-mode drivers are an
>  underutilized alternative for systems that have an IOMMU/SMMU.  Obviously,
>  the drivers still need to be free software for Fedora to package them, but
>  user-mode drivers have the advantage of not running with kernel privilege. 
That smells like a microkernel :) I did not know it's possible. I can image
one can program IOMMU to direct all reads and writes by a device into a memory
mapped into a user-space process. But how can a user space process receive
interrupts? Is there an in-kernel translation layer which presents interrupts
to the user process somehow? Or does it only work with the modern MSI-like
interrupts which are a ring buffer of messages in a shared memory?

-- Petr

[Demi Marie Obenour]

On 3/22/22 11:19, Petr Pisar wrote:
>  V Tue, Mar 22, 2022 at 07:30:13AM -0400, Demi Marie Obenour
> napsal(a):
> > All kernel-mode drivers, to be specific.  User-mode drivers are an
> > underutilized alternative for systems that have an IOMMU/SMMU.  Obviously,
> > the drivers still need to be free software for Fedora to package them, but
> > user-mode drivers have the advantage of not running with kernel privilege.
>  
>  That smells like a microkernel :) 
That’s because almost all drivers on a microkernel work that way 🙂.
Interrupt controller, IOMMU, and timer drivers are the main exceptions
I know of.

>  I did not know it's possible. I can image
>  one can program IOMMU to direct all reads and writes by a device into a memory
>  mapped into a user-space process. But how can a user space process receive
>  interrupts? Is there an in-kernel translation layer which presents interrupts
>  to the user process somehow? 
There is; see ‘Documentation/driver-api/vfio.rst’ in the kernel
source tree.  To be clear: this is not an excuse for the driver being
proprietary or of low quality, but rather a way to allow the driver to
run as a deprivileged userspace process.

>  Or does it only work with the modern MSI-like
>  interrupts which are a ring buffer of messages in a shared memory?
>  
>  -- Petr 
On x86 I believe it only supports MSI interrupts, but for a different
reason: old-style interrupts cannot be remapped, and so devices that
can create them cannot safely be assigned to an untrusted principal
(which a userspace driver is, from the kernel’s perspective).
Specifically, I believe neither Xen nor Hyper-V support device
passthrough for devices with old-style interrupts, and this is
equivalent (from a security perspective) to VFIO.  I do not know how
this works on ARM.

-- 
Sincerely,
Demi Marie Obenour (she/her/hers)

[Luigi Grax]

hi

thanks for your answers

There is a YOCTO LE Linux project and Android supported on CPU by Qualcomm

peripherals integrated are USB3-HUB from Microchip and MCU

it is a maker-board with very simple architecture