Re: [regression v6.12.30..v6.12.32] mmc1: mmc_select_hs400 failed, error -110 / boot regression on Lenovo IdeaPad 1 15ADA7

[Salvatore Bonaccorso <carnil@debian.org>]

Hi,

On Tue, Jun 24, 2025 at 11:26:41AM +0300, Adrian Hunter wrote:
> On 23/06/2025 12:13, Shawn Lin wrote:
> > + Jonathan Liu
> > 
> > 在 2025/06/21 星期六 4:21, Salvatore Bonaccorso 写道:
> >> On Fri, Jun 20, 2025 at 09:20:41PM +0200, Salvatore Bonaccorso wrote:
> >>> Hi
> >>>
> >>> In Debian we got a regression report booting on a Lenovo IdeaPad 1
> >>> 15ADA7 dropping finally into the initramfs shell after updating from
> >>> 6.12.30 to 6.12.32 with messages before dropping into the intiramfs
> >>> shell:
> >>>
> >>> mmc1: mmc_select_hs400 failed, error -110
> >>> mmc1: error -110 whilst initialising MMC card
> >>>
> >>> The original report is at https://bugs.debian.org/1107979 and the
> >>> reporter tested as well kernel up to 6.15.3 which still fails to boot.
> >>>
> >>> Another similar report landed with after the same version update as
> >>> https://bugs.debian.org/1107979 .
> >>>
> >>> I only see three commits touching drivers/mmc between
> >>> 6.12.30..6.12.32:
> >>>
> >>> 28306c58daf8 ("mmc: sdhci: Disable SD card clock before changing parameters")
> >>> 38828e0dc771 ("mmc: dw_mmc: add exynos7870 DW MMC support")
> >>> 67bb2175095e ("mmc: host: Wait for Vdd to settle on card power off")
> >>>
> >>> I have found a potential similar issue reported in ArchLinux at
> >>> https://bbs.archlinux.org/viewtopic.php?id=306024
> >>>
> >>> I have asked if we can get more information out of the boot, but maybe
> >>> this regression report already rings  bell for you?
> > 
> > Jonathan reported a similar failure regarding to hs400 on RK3399
> > platform.
> > https://lkml.org/lkml/2025/6/19/145
> > 
> > Maybe you could try to revert :
> > 28306c58daf8 ("mmc: sdhci: Disable SD card clock before changing parameters")
> 
> Given the number of other reports, probably best to revert
> anyway.

FTR, Jeremy Lincicome confirmed that reverting the commit fixes the
issue for him as reported in Debian bug #1108065.

Regards,
Salvatore