Re: [PATCH v5 06/13] mmc: arm_pl180: Power down card supplies at OS handoff
Stephan Gerhold <[email protected]>
| Newsgroups | org.u-boot-project.lists.u-boot |
|---|---|
| Message-ID | <[email protected]> |
On Mon, Aug 24, 2026 at 10:01:15PM +0200, Linus Walleij wrote: > On Mon, Aug 24, 2026 at 3:30 PM Stephan Gerhold > <[email protected]> wrote: > > On Sun, Aug 23, 2026 at 10:39:25PM +0200, Linus Walleij wrote: > > > > PL180 card supplies remain enabled after probing and while U-Boot is > > > running. They are no longer needed once control is handed to the operating > > > system. > > > > > > Mark the driver for removal during OS preparation and disable its VMMC and > > > VQMMC supplies from the remove callback. > > > > > > Signed-off-by: Linus Walleij <[email protected]> > > > > Is it worth briefly toggling the regulator off before Linux turns it on > > again? Linux will typically need to turn them on again later during > > boot, and even if not Linux should be able to deal with unused > > regulators. I'm not sure it's worth adding such cleanup to U-Boot. > > For our specific Ux500 target it is true. > > But this driver is used by a bunch of platforms including some > ARM reference designs, and also booting other operating systems > than Linux. > > Something could EFI boot off one SD card, then have rootfs > on another one and the original boot card undefined in the > hardware description for example. Albeit that sounds a bit > stupid. > > So I thought it'd be more clean to shut off the light on your way > out. > > But it's not like I'm married to the patch. > I don't really have a strong opinion either. We can just leave it up to the maintainers. Can you re-run get_maintainer.pl on this series for v6? Seems like the U-Boot MMC maintainers aren't included in this patch. I think it would also be worth Ccing the "ARM STM STM32MP" maintainers for arm_pl180 changes, they use the STM32_MMCI_ID code path in this driver and reviewed/tested my last changes (Patrice Chotard in particular). Thanks, Stephan