On Wed, Dec 10, 2025 at 10:42 PM Kane Chen <[email protected]> wrote:
>
> Hi Nabih,
>
> Thank you for your review and comments. I will wait a few more days
> to see if there is additional feedback from others. If you encounter
> anything unexpected in your local environment, please let me know. I
> will try to address it in the v4 patch if possible.
>
> Best regards,
> Kane

Hi Kane,

I just discovered that our machines seem to fail to boot halfway
through the stack, but work as expected at the top of the stack. My
guess is that the driver for the AST1700 doesn't quite know how to
deal with half the model existing and half of it missing. Is it
possible to reorganize the patches so the `hw/arm/aspeed: Integrate
AST1700 device into AST27X0` patch comes last? That way the driver
will only know of the existence of the AST1700 when the model fully
exists in QEMU. Mostly to make sure the machine works in any commit
and not just at HEAD.

Thanks!
Nabih


> > -----Original Message-----
> > From: Nabih Estefan <[email protected]>
> > Sent: Thursday, December 11, 2025 7:02 AM
> > To: Kane Chen <[email protected]>
> > Cc: Cédric Le Goater <[email protected]>; Peter Maydell
> > <[email protected]>; Steven Lee <[email protected]>; Troy
> > Lee <[email protected]>; Jamin Lin <[email protected]>; Andrew
> > Jeffery <[email protected]>; Joel Stanley <[email protected]>;
> > open list:ASPEED BMCs <[email protected]>; open list:All patches CC
> > here <[email protected]>; Troy Lee <[email protected]>
> > Subject: Re: [PATCH v3 00/18] hw/arm/aspeed: AST1700 LTPI support and
> > device hookups
> >
> > On Sun, Dec 7, 2025 at 11:45 PM Kane Chen via <[email protected]>
> > wrote:
> > >
> > > From: Kane-Chen-AS <[email protected]>
> > >
> > > Hi all,
> > >
> > > LTPI (LVDS Tunneling Protocol & Interface) is defined in the OCP
> > > DC-SCM
> > > 2.0 specification (see Figure 2):
> > > https://www.opencompute.org/documents/ocp-dc-scm-2-0-ltpi-ver-1-0-pdf
> > >
> > > LTPI provides a protocol and physical interface for tunneling various
> > > low-speed signals between the Host Processor Module (HPM) and the
> > > Satellite Controller Module (SCM). In Figure 2 of the specification,
> > > the AST27x0 SoC (left) integrates two LTPI controllers, allowing it to
> > > connect to up to two AST1700 boards. On the other side, the AST1700
> > > consolidates HPM FPGA functions and multiple peripheral interfaces
> > > (GPIO, UART, I2C, I3C, etc.) onto a single board.
> > >
> > > Because the AST1700 exposes additional I/O interfaces (GPIO, I2C, I3C,
> > > and others), it acts as an I/O expander. Once connected over LTPI, the
> > > AST27x0 can control additional downstream devices through this link.
> > >
> > > This patch series is based on the SGPIO changes:
> > > https://lore.kernel.org/qemu-devel/20251106-aspeed-sgpio-v1-0-b0260937
> > > [email protected]/
> > >
> > > It introduces a basic LTPI controller model and wires it into the
> > > AST27x0 SoC. The series also adds the AST1700-specific LTPI expander
> > > device and incrementally connects common peripherals on the AST1700
> > > model. For the I3C block, which may cause kernel crashes, its MMIO
> > > region is modeled as an unimplemented device to reserve address space
> > > and make the missing functionality explicit, ensuring stable guest
> > > probing.
> > >
> > > In the official release images, the AST1700 functions are not included
> > > by default. To test the AST1700-related functionality, please include
> > > the following DTS files for probing:
> > >
> > https://github.com/AspeedTech-BMC/linux/blob/aspeed-master-v6.6/arch/a
> > > rm64/boot/dts/aspeed/aspeed-ltpi0.dtsi
> > >
> > https://github.com/AspeedTech-BMC/linux/blob/aspeed-master-v6.6/arch/a
> > > rm64/boot/dts/aspeed/aspeed-ltpi1.dtsi
> > >
> > > After including these DTS files in the BMC image, you can verify LTPI
> > > functionality using the following scenarios:
> > >
> > > 1. In U-Boot:
> > >    Run the ltpi command to trigger the LTPI connection and display the
> > >    current connection status.
> > > 2. In BMC Linux:
> > >    Run i2cdetect -y <16-38> to scan and test the I2C buses exposed by
> > >    the AST1700.
> > >
> > > Any feedback or suggestions are appreciated!
> > >
> > > Kane
> > >
> >
> > I left some nit-picky notes on some of the patches, but after fixing patch 
> > 03
> > everything else seems good. FWIW I brought the whole patchset into our
> > branch and not only does everything compile properly, but I can also see and
> > interact with the i2c buses from the AST1700 in our internal platforms. Will
> > wait for v4 to add the reviewed and tested tags, but LGTM!
> >
> > > ---
> > >
> > > ChangeLog
> > > ---------
> > > v3:
> > > - Add PWM model
> > > - Integrate the SGPIO model
> > > - Fix I2C test case failure
> > > - Refine code structure
> > >
> > > v2:
> > > - Separate the AST1700 model into a standalone implementation
> > > - Refine the mechanism for assigning the AST1700 board number
> > >
> > > v1:
> > > - Initial version
> > > ---
> > >
> > > Kane-Chen-AS (18):
> > >   hw/misc: Add LTPI controller
> > >   hw/arm/aspeed: Attach LTPI controller to AST27X0 platform
> > >   hw/misc: Add basic Aspeed PWM model
> > >   hw/arm/aspeed: Add AST1700 LTPI expander device model
> > >   hw/arm/aspeed: Integrate AST1700 device into AST27X0
> > >   hw/arm/aspeed: Integrate interrupt controller for AST1700
> > >   hw/arm/aspeed: Attach LTPI controller to AST1700 model
> > >   hw/arm/aspeed: Attach UART device to AST1700 model
> > >   hw/arm/aspeed: Attach SRAM device to AST1700 model
> > >   hw/arm/aspeed: Attach SPI device to AST1700 model
> > >   hw/arm/aspeed: Attach ADC device to AST1700 model
> > >   hw/arm/aspeed: Attach SCU device to AST1700 model
> > >   hw/arm/aspeed: Attach GPIO device to AST1700 model
> > >   hw/arm/aspeed: attach I2C device to AST1700 model
> > >   hw/arm/aspeed: Attach WDT device to AST1700 model
> > >   hw/arm/aspeed: Attach PWM device to AST1700 model
> > >   hw/arm/aspeed: Attach SGPIOM device to AST1700 model
> > >   hw/arm/aspeed: Model AST1700 I3C block as unimplemented device
> > >
> > >  include/hw/arm/aspeed_ast1700.h |  53 +++++++
> > >  include/hw/arm/aspeed_soc.h     |  25 ++-
> > >  include/hw/i2c/aspeed_i2c.h     |   1 +
> > >  include/hw/intc/aspeed_intc.h   |   2 +
> > >  include/hw/misc/aspeed_ltpi.h   |  32 ++++
> > >  include/hw/misc/aspeed_pwm.h    |  31 ++++
> > >  hw/arm/aspeed_ast1700.c         | 269
> > ++++++++++++++++++++++++++++++++
> > >  hw/arm/aspeed_ast27x0.c         | 163 +++++++++++++++++--
> > >  hw/i2c/aspeed_i2c.c             |  19 ++-
> > >  hw/intc/aspeed_intc.c           |  60 +++++++
> > >  hw/misc/aspeed_ltpi.c           | 194 +++++++++++++++++++++++
> > >  hw/misc/aspeed_pwm.c            | 121 ++++++++++++++
> > >  hw/arm/meson.build              |   1 +
> > >  hw/misc/meson.build             |   2 +
> > >  hw/misc/trace-events            |   4 +
> > >  15 files changed, 957 insertions(+), 20 deletions(-)  create mode
> > > 100644 include/hw/arm/aspeed_ast1700.h  create mode 100644
> > > include/hw/misc/aspeed_ltpi.h  create mode 100644
> > > include/hw/misc/aspeed_pwm.h  create mode 100644
> > > hw/arm/aspeed_ast1700.c  create mode 100644 hw/misc/aspeed_ltpi.c
> > > create mode 100644 hw/misc/aspeed_pwm.c
> > >
> > > --
> > > 2.43.0
> > >
> > >

Reply via email to