Re: [PATCH v6 3/3] arm64:?==?utf-8?q? dts: rockchip: Add Radxa CM5 IO Board

[Dragan Simic]

Hello Jimmy,

On Thursday, November 06, 2025 05:53 CET, Jimmy Hon <honyuenkwun at gmail.com> wrote:
> On Wed, Nov 5, 2025 at 9:31 PM Dragan Simic <dsimic at manjaro.org> wrote:
> > With all that in mind, we should specify "no-mmc" here, because
> > we're describing a microSD slot, instead of describing some hybrid
> > MMC/microSD slot.  That also explains why the adapter sold by Radxa
> > is described as not to be used with microSD card slots on SBCs.  I'd
> > also like to hear is this adapter/eMMC chip combo recognized by the
> > kernel when "no-mmc" is specified; it should fail.
> >
> > Actually, not specifying "no-mmc" here may result in some unforeseen
> > issues with some (or perhaps many?) microSD cards, because the MMC
> > drivers will treat them as MMC-capable devices and try to initialize
> > them as such, which may cause all kinds of issues.  In fact, I'm not
> > really sure that the MMC drivers are actually implemented in a way
> > that avoids all possible issues with the storage controllers that
> > are capable of both SD and MMC modes when neither of "no-sd" and
> > "no-mmc" is specified in their DT nodes.
> 
> Hybrid MMC and SD slots are pretty normal on USB card readers. So it's
> normal for the host controller to figure out what kind of card is in
> the slot.
> https://uditagarwal.in/understanding-sd-sdio-and-mmc-interface/
> 
> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=6ae3e537eab9f560b516b001eb89f0cd568bdced
> was the commit that introduced the device tree properties. By the
> wording of the commit message, these device tree properties are used
> to indicate to the driver if the host controller hardware is capable
> of MMC initialization or SD initialization.
> 
> Since the host controller in the RK3588 is capable of all the modes,
> these properties do not need to be specified.
> 
> Since Radxa has the eMMC to uSD adapter, it makes sense Radxa would
> want to configure their microSD card slot on their boards to be a
> hybrid SD/MMC slot.

Thanks for providing further insights!  After thinking a bit more
about it and after remembering that HardKernel also offers a similar
microSD-to-MMC adapter, [1] there should be very few roadblocks that
may actually prevent us from defining physical microSD slots found
on Rockchip boards as hybrid microSD/MMC slots wherever that's
allowed by the slot implementation and tested to work as expected.
Supporting more card standards is always good.

This approach shouldn't be limited to Radxa (or HardKernel) boards
only, because the available microSD-to-MMC adapters aren't designed
specifically to fit some boards only.

The possible roadblocks, as mentioned above, are some unexpected
signal integrity issues that may prevent the MMC mode from working
as expected on some boards, which Jonas pointed out already, [2][3]
and any associated issues in the MMC drivers.

I'll keep checking the code of MMC drivers for the existence of any
associated issues, and I'll possibly turn a few microSD-only slots
into hybrid ones. :)

[1] https://www.hardkernel.com/shop/emmc-module-reader-board-for-os-upgrade/
[2] https://libera.catirclogs.org/linux-rockchip/2025-11-06#38976445;
[3] https://libera.catirclogs.org/linux-rockchip/2025-11-07#38981060;