Re: [PATCH u-boot v3 2/2] ARM: dts: aspeed: Add dts for ASUS Kommando IPMI card

Anirudh Srinivasan <[email protected]> Mon, 1 Jun 2026 15:21:59 -0500
Newsgroups org.ozlabs.lists.openbmc
Message-ID <CAJ13v3QrD5Q=oMn3bjBr7SdQNXj2zk+N+BmhqhiV8j-Wx0c6Vw@mail.gmail.com>
On Mon, Jun 1, 2026 at 8:33 AM Tan Siewert <[email protected]> wrote:
>
> On 6/1/26 3:22 PM, Anirudh Srinivasan wrote:
> > Hi Tan,
> >
> > On Mon, Jun 1, 2026 at 4:26 AM Tan Siewert <[email protected]> wrote:
> >>
> >>> +
> >>> +&fmc {
> >>> +     status = "okay";
> >>> +
> >>> +     pinctrl-names = "default";
> >>> +     pinctrl-0 = <&pinctrl_fmcquad_default>;
> >>> +
> >>> +     flash@0 {
> >>> +             status = "okay";
> >>> +             spi-max-frequency = <50000000>;
> >>> +             spi-tx-bus-width = <4>;
> >>> +             spi-rx-bus-width = <4>;
> >>> +     };
> >>
> >> Out of curiosity: which revision of the Kommando IPMI card do you have?
> >> Asking because I have two R1.04s and both fail after a power cycle, but
> >> only if the SPI was probed by U-Boot before.
> >
> > My card also says R1.04 on it. My flash chip is a Winbound 25Q512JVFQ.
> >
> > Are you observing this after a graceful reboot from inside an OS, or
> > when you unplug/replug the USB power cable? If you're doing the
> > latter, I've observed that the card doesn't come back up properly when
> > the UART ground is connected, even with the stock fw. I need to unplug
> > power, unplug the UART ground cable and then replug both back in. I
> > see some garbage output on the UART at boot if I don't do this.
>
> I only observe this on re-plugging the USB power cable. Disconnecting
> UART and all other peripherals that have any ground pins makes the issue
> happen less times, but I am still able to trigger it if the power is
> falling at a bad timing (e.g. on init before systemd is starting.)

Given that this happens with the stock fw too, this seems to be a
quirk of the board design. It's likely missing a bleed resistor
somewhere. It doesn't affect the day to day functionality of the card,
only developmental flows where one is constantly reflashing/hard
rebooting the card.

> The R1.01 cards for example don't seem to have this issue at all, as I
> was told by a colleague.
>
>         Tan



-- 
Regards
Anirudh Srinivasan