Hi Krzysztof, On Fri, Mar 17, 2023 at 12:55:40PM +0100, Krzysztof Kozlowski wrote: > On 16/03/2023 22:15, Henrik Grimler wrote: > > Previously, the mshc0 alias has been necessary so that > > MMC_CAP_1_8V_DDR | MMC_CAP_8_BIT_DATA are set for mshc_0/mmc_0. > > However, these capabilities should be described in the device tree so > > that we do not have to rely on the alias. > > > > The property mmc-ddr-1_8v replaces MMC_CAP_1_8V_DDR, while bus_width = > > <8>, which is already set for all the mshc0/mmc0 nodes, replaces > > MMC_CAP_8_BIT_DATA. > > > > Also drop other mshc aliases as they are not needed. > > > > Signed-off-by: Henrik Grimler <henrik@xxxxxxxxxx> > > Tested-by: Marek Szyprowski <m.szyprowski@xxxxxxxxxxx> > > > > --- > > > > Changes since v4: > > * Do not set mmc-ddr-1_8v for sdhci_0 on Exynos 4210, > > following Marek's tests > > > > Changes since v3: > > * Drop attempt at node sorting/cleaning > > * Move two mmc alias additions to the other patch > > * Update commit message > > > > Changes since v2: > > * Set mmc-ddr-1_8v in device trees for mshc_0/mmc_0 > > > > > > arch/arm/boot/dts/exynos3250-artik5.dtsi | 1 + > > arch/arm/boot/dts/exynos3250-monk.dts | 1 + > > arch/arm/boot/dts/exynos3250-rinato.dts | 1 + > > Why you do not remove Exynos3250 aliases? Thanks for spotting that, seems I accidentally dropped removal when going from v3 to v4. Will remove them again in next version, and drop Marek's test tag since rinato is one of the devices he tests (or used to test). > Best regards, > Krzysztof Best regards, Henrik Grimler
